Understanding, interpreting and shaping the technical world
Learner-centred, contextualised teaching of technical artefacts, systems and processes – grounded in empirical classroom research and constructionist didactics.
Technology didactics as an independent scientific discipline
In my research and teaching, I focus on the learner-centred, contextualised teaching of technical artefacts, systems and processes. I am particularly interested in how learners understand, interpret and help shape the technical world – and how classroom teaching can productively guide that engagement.
My work is empirically grounded and combines curriculum development, teacher education and educational research into a coherent research programme that strengthens technology didactics as an independent scientific discipline.
🏗️
Maker Education → more
Constructionist learning through the design of physical-digital artefacts in school and out-of-school contexts.
🧠
Learners' Conceptions → more
Eliciting and didactically using learners' preconceptions and naive theories about technical artefacts, systems and processes.
🧑🎓
Teacher Education → more
Developing professional competences for technology teachers in hybrid teaching and learning environments.
🔭
Empirical Classroom Research → more
Qualitative and design-based research methods for analysing learning processes in technology education.
Working Group · Professorship for Technology & Its Didactics
The Schwäbisch Gmünd University of Education offers two paths into the technology teaching profession: through science-and-technology general studies at primary school, and through technology education at lower secondary level. Both degree programmes combine subject-specific foundations with a practice-oriented, research-based didactics education – made directly tangible through the CreatorLab as a hybrid teaching and learning laboratory.
In the elementary teacher education programme, technology is studied as part of science-and-technology general studies. The aim is to enable children in elementary school to understand, try out and reflect on technical phenomena and artefacts from their everyday world – in the sense of an early foundational technical education with a connectable STEAM orientation.
Didactics of General Studies with a focus on technology didactics: concepts of technical education at primary level, learners' conceptions, action-oriented lesson design
Subject-specific foundations: technology, natural sciences and their didactic interconnection – from simple machines to digital technology
Making & CreatorLab: hands-on modules in the hybrid makerspace – 3D printing, electronics prototyping, constructing as a genuine form of learning
School placements: classroom practice in partner elementary schools across the region – from planning through delivery to reflection
Technology is an independent school subject at lower secondary level in Baden-Württemberg. The degree programme at PH Schwäbisch Gmünd qualifies graduates to teach in the five technology-didactic problem and action fields: work & production, building & housing, transport & traffic, supply & disposal, and information & communication – anchored in the Baden-Württemberg curriculum.
Technology didactics: lesson planning, delivery and reflection across the five action fields; STEAM integration; competence orientation per the BW curriculum
Technology subject science: fundamentals of wood, metal and plastics processing, electrical engineering, computer science/digital technology, construction and building services
CreatorLab & digital fabrication: hands-on training on real production machinery – lathe, laser cutter, 3D printer, soldering station, CNC
Research orientation: empirical classroom research, theses in cooperation with the Professorship for Technology & Its Didactics
🎓 B.Ed. · Lower secondary teaching degree
🎓 M.Ed. · consecutive Master of Education
💡
Why study technology teaching at PH Schwäbisch Gmünd?
Studying at PH Gmünd combines subject science and subject didactics in small, personal study groups. The close ties to partner schools and the network of out-of-school learning sites across the region enable early, intensive practical experience – from lesson planning through to independent delivery.
The CreatorLab is constituted as a hybrid teaching and learning space at the intersection of university research, teacher education and societal practice. In the spirit of the third-mission concept, it conveys academic knowledge through action-oriented making processes.
Workshop manager Dieter Bühler
📡
Twitch Livestream
Synchronous making sessions in which teacher-education students act as learning facilitators and interact live with pupils.
🎬
DIY Tutorial Videos
A persistent OER library on YouTube: asynchronously available how-to formats for self-directed learning after Papert (1980).
🖨️
3D Printing & Laser Cutting
Digital fabrication technologies as didactic tools for producing publicly shareable artefacts.
💬
Discord Community
Asynchronous peer-to-peer support and collective documentation within a community of practice (Lave & Wenger, 1991).
⚡
Electronics Prototyping
Developing functional electronics prototypes as a bridge between abstract theory and concrete practice.
🌐
Hybrid Learning Spaces
Integrating physical presence and virtual participation as complementary dimensions of the learning space.
Showcase · Recent Work from the CreatorLab
Theoretical Foundation
Constructionism as the core didactic principle
The core didactic principle of the CreatorLab follows constructionism after Papert (1980): learning is most effective when learners produce publicly shareable artefacts. Unlike cognitivism, which conceives of knowledge-building as a purely mental process, constructionism emphasises the epistemic function of making: the object – whether a 3D-printed part, an electronic circuit, or a produced tutorial video – is not merely an outcome but a medium of thought itself. Blikstein (2013) explicitly transfers this constructionist principle to digital fabrication technologies, describing FabLabs as sites of the "democratisation of invention" – a direct theoretical bridge to the CreatorLab's equipping with 3D printers and laser cutters.
This approach is empirically grounded by maker-education research: Halverson & Sheridan (2014) define the maker movement as a cultural practice of creative artefact production that creates both physical and digital forums for sharing processes and products – a description that fits precisely with the structure of the CreatorLab (workshop, livestream, Discord). Sheridan et al. (2014) show in a comparative case study of three makerspaces that learning processes there are significantly shaped by the respective community culture and available tools. Martin (2015) argues for the pedagogical potential of the maker movement, while also pointing out that the empirical evidence base for which learning processes it actually fosters is still emerging – a caveat that the CreatorLab, as a research-accompanied format, must also address.
The CreatorLab realises this principle in a hybrid learning space that structurally interweaves physical-analogue and digital-virtual dimensions. Oldenburg (1989) originally developed the concept of the third place for physical, informal meeting points beyond school and home; transferring it to a digitally extended learning space is therefore a deliberate extension of the original concept. A closer fit for the CreatorLab's concrete hybrid structure is the model of connected learning (Ito et al., 2013): it describes precisely that interest-driven, production-centred and peer-supported learning that extends across multiple networked platforms – realised in the CreatorLab through the interplay of livestream (synchronous), Discord (peer-to-peer), and the YouTube library (asynchronous-persistent). The Twitch livestream, the Discord community and the YouTube OER library together form the digital extension of the physical workshop space within the Transfer Hub for Innovation in Society (THIS) at PH Schwäbisch Gmünd.
A core conceptual feature is the explicit role shift of teacher-education students: rather than acting as instructional knowledge transmitters, they act as learning facilitators in the sense of a facilitated learning approach (Rogers, 1969; Schön, 1983). Drawing on the concept of the community of practice (Lave & Wenger, 1991), students position themselves as more experienced peers within a pupil community – without closing off the open, exploratory character of learning. This approach links teacher education and school practice into a cooperative learning format that overcomes spatial and institutional boundaries.
Eco-friendly hydraulic gripper arm at the Children's University: discovering technology through play and exploring sustainable solutions
📗
Papert, S. (1980)
Mindstorms: Children, Computers, and Powerful Ideas. Basic Books. — Foundational work of constructionism: learning through the production of publicly shareable artefacts
🏭
Blikstein, P. (2013)
Digital Fabrication and 'Making' in Education: The Democratization of Invention. In J. Walter-Herrmann & C. Büching (Eds.), FabLabs: Of Machines, Makers and Inventors (pp. 1–21). Bielefeld: transcript. — Transferring constructionism to digital fabrication technologies and FabLabs
🛠️
Halverson, E. R. & Sheridan, K. (2014)
The Maker Movement in Education. Harvard Educational Review, 84(4), 495–504. — Defining the maker movement as a cultural practice of shared physical and digital production
🏫
Sheridan, K. M., Halverson, E. R., Litts, B. K., Brahms, L., Jacobs-Priebe, L. & Owens, T. (2014)
Learning in the Making: A Comparative Case Study of Three Makerspaces. Harvard Educational Review, 84(4), 505–531. — Empirical case study of learning processes in makerspaces
🌱
Martin, L. (2015)
The Promise of the Maker Movement for Education. Journal of Pre-College Engineering Education Research (J-PEER), 5(1), Art. 4. — Pedagogical potential and open evidence questions of the maker movement
📘
Lave, J. & Wenger, E. (1991)
Situated Learning: Legitimate Peripheral Participation. Cambridge University Press. — Community of practice as a framework for cooperative learning in authentic contexts
📙
Rogers, C. R. (1969)
Freedom to Learn. Charles E. Merrill. — Facilitated learning and person-centred pedagogy as the basis of the learning-facilitation concept
📕
Schön, D. A. (1983)
The Reflective Practitioner: How Professionals Think in Action. Basic Books. — Knowing-in-practice and reflective action as a core competence of professional learning facilitation
📒
Oldenburg, R. (1989)
The Great Good Place. Paragon House. — The concept of the third place as an informal (physical) learning space beyond school and home
🌐
Ito, M., Gutiérrez, K., Livingstone, S., Penuel, B., Rhodes, J., Salen, K., Schor, J., Sefton-Green, J. & Watkins, S. C. (2013)
Connected Learning: An Agenda for Research and Design. Irvine, CA: Digital Media and Learning Research Hub. — Interest-driven, production-centred learning across networked platforms
📓
Brown, J. S., Collins, A. & Duguid, P. (1989)
Situated Cognition and the Culture of Learning. Educational Researcher, 18(1), 32–42. — Knowledge as context-bound practice: the basis of situated learning in the CreatorLab
📔
Collins, A., Brown, J. S. & Newman, S. E. (1989)
Cognitive Apprenticeship: Teaching the Crafts of Reading, Writing, and Mathematics. In L. B. Resnick (Ed.), Knowing, Learning, and Instruction (pp. 453–494). Lawrence Erlbaum Associates. — Modelling, coaching and scaffolding as didactic principles of learning facilitation
Synchronous Participation · Twitch
CreatorLab Livestream
The CreatorLab's Twitch channel is the digital heart of the hybrid learning space. In synchronous making sessions, teacher-education students moderate interactive learning sequences.
The CreatorLab combines STEAM as an integrative educational concept with situated learning (Brown, Collins & Duguid, 1989) and the RANG competence dimensions after Irion, Peschel & Schmeinck (2023) — the German initials for Reflexion (reflection), Analyse (analysis), Nutzung (use) and Gestaltung (design) of analogue-digital artefacts, kept here as "RANG" to match the original model. These four dimensions form the competence framework within which STEAM learning is situated in the CreatorLab. The RANG model was originally developed for elementary school; in the CreatorLab it is adapted for higher-education teaching and work with teacher-education students.
Knowledge is inseparable from the context of its creation. In the CreatorLab, machines, materials, fellow learners and the livestream form the authentic situation in which competences are constructed.
🔢
RANG · Irion, Peschel & Schmeinck (2023)
Four competence dimensions of foundational digital education: Reflection · Analysis · Use · Design – as an orientation framework for teachers and learners in digitality.
S
Science
Naturalscience
RANG
Learn more
T
Technology
Digital Tools
RANG
Learn more
E
Engineering
Constructing & Solving
RANG
Learn more
A
Arts
Design & Creativity
RANG
Learn more
M
Mathematics
Modelling & Algorithms
RANG
Learn more
RANG · FOUR COMPETENCE DIMENSIONS OF ANALOGUE-DIGITAL EDUCATION IN THE CREATORLAB
R
Reflexion (Reflection)
A critically reflective approach to digitality: learners question the effects, motives and societal consequences of technical artefacts – in the CreatorLab through discussion of livestream content and maker projects.
A
Analyse (Analysis)
Understanding digital systems and processes: learners identify algorithms, circuits and fundamental computing principles – through electronics prototyping, sensors and CAD modelling in the lab.
N
Nutzung (Use)
Confident use of digital tools: learners operate 3D printers, laser cutters, slicer software and microcontrollers as epistemic tools for problem-solving.
G
Gestaltung (Design)
Actively co-shaping digitality: learners produce livestreams, DIY videos, 3D artefacts and CAD designs – as publicly shareable constructs.
The B@t Detector is an exemplary STEAM learning project: pupils independently construct an electronic device that converts bats' ultrasonic calls (20–100 kHz) into audible, visible and haptic signals. The project combines biology, physics and electronics with hands-on craftsmanship and is didactically enriched with gamification elements (gamepad aesthetics, LED feedback, vibration) – in the spirit of situated learning in authentic, meaningful contexts.
S
Science
Ultrasound physics, echolocation and the ecology of Chiroptera as the scientific core
T
Technology
Digital signal processing and frequency conversion
E
Engineering
Soldering, circuit assembly, enclosure design and iterative prototyping with maker tools
A
Arts
Gamepad design as an aesthetic concept: enclosure shape, colour design, interface design
M
Mathematics
Frequency calculation, speed of sound and the distance formula d = v·t/2 in an applied context
Lohse, T., Mönch, C., Pfeifer, J. & Nepper, H. H. (2026). Technical Education in Focus: Mechanics and Robotics as Future Skills for Teaching. Erasmus+ Blended Intensive Programme.
Interested in a thesis or a research internship in technology didactics or maker education? Feel free to get in touch at any time.
🎓
Technology Didactics · PH Schwäbisch Gmünd
Guide: Technology-Didactic Theses in the Age of AI
Academic work, research design and responsible use of AI for bachelor's and master's theses in technology didactics and maker education.
What is a technology-didactic thesis?
Research at the intersection of technology, didactics and empirical work
A technology-didactic thesis is not a mere literature review – it combines subject-didactic theory with empirical classroom research or constructive development work. The central question is always: How do people learn technical content, and how can classroom teaching productively shape these learning processes? A good thesis contributes to the further development of technology didactics as a discipline – it is not merely an assessment task, but a genuine scientific contribution.
The Professorship for Technology & Its Didactics understands theses as research practice: students enter into a genuine scholarly dialogue with the field as early-career researchers. That means: reading primary sources instead of collecting secondary citations, justifying methods instead of simply applying them, critically discussing results instead of merely describing them.
Typical research areas of the professorship
🧠
Learners' conceptions (conceptual change): What everyday conceptions do pupils have about technical systems (electricity, machines, algorithms)? How can interventions productively change these conceptions? Theoretical basis: Posner et al. (1982), diSessa (1988), Nepper & Windelband (2022).
🔧
Maker education & constructionist learning: How does producing physical artefacts affect technical understanding, motivation and self-efficacy? Theoretical basis: Papert (1980), Hatch (2014), Martinez & Stager (2013).
💻
Digital fabrication technologies: How does the use of 3D printing, laser cutting or electronics development change technology education? Which didactic designs make use of these technologies' potential? Research conducted in the CreatorLab at PH Schwäbisch Gmünd.
🌐
Situated learning & learning site partnerships: How do pupils learn at out-of-school learning sites (Eule, Explorhino, Zukunftsakademie)? What features make learning site partnerships didactically effective? Theoretical basis: Brown et al. (1989), Lave & Wenger (1991).
📊
Effectiveness of technology-didactic interventions: How does a specific teaching unit measurably change competence, motivation or self-efficacy? Pre-post designs, randomised control groups (where feasible in school contexts), effect-size measurement.
🤖
AI in technology education: How do AI tools change technology education and pupils' technical thinking? What prompting competences does the next generation of teachers need? A current research area with substantial development potential.
Fundamental Quality Criteria
🎯
Clear, answerable research question: A good thesis begins with a precise question, not a topic. "How does the use of 3D printing influence the development of technical systems understanding in Year 7 pupils?" is a research question. "3D printing in the classroom" is a topic. The question must be actually answerable within the processing time available – no questions that would require a ten-year longitudinal study.
📚
Theoretical grounding with primary sources: Every empirical thesis needs a theoretical framework. In technology didactics: the conceptual change model (Posner et al., 1982), situated learning (Brown et al., 1989), cognitive apprenticeship (Collins et al., 1989), constructionism (Papert, 1980), RANG (Irion et al., 2023). Mandatory: read primary sources – secondary citations are the exception, not the rule.
🔬
Methodological fit and justification: The method follows the question – not the other way round. "How is X experienced…" → qualitative. "How much does X change…" → quantitative. "How must X be designed so that…" → DBR. Every choice of method must be explicitly justified in the thesis: why this method? What alternatives were considered and why were they rejected?
🔍
Transparent methodology and traceability: Data collection instruments, analysis procedures and coding decisions must be documented such that an independent researcher could follow and replicate the analysis. For qualitative research: document inter-rater reliability (aim for Cohen's κ ≥ .70).
📝
Academic integrity and AI transparency: Complete source citations (APA 7), transparent methodology, critical reflection on limitations. Since winter semester 2023/24: all AI tools used must be fully declared in an appendix (tool, version, date, purpose, extent). Non-declaration counts as an attempt to deceive.
Formal Requirements & Scope
📄
Bachelor's thesis (B.Ed.): 40–60 pages of body text excluding appendix · 3 months processing time · APA 7th edition · submit an exposé (3–5 pages) before starting · empirical section optional but recommended · 12pt font, 1.5 line spacing, 2.5 cm margins.
📄
Master's thesis (M.Ed. / M.A.): 60–90 pages of body text excluding appendix · 6 months processing time · APA 7th edition · exposé (5–8 pages) mandatory before starting · an empirical component with independent data collection and analysis is expected · colloquium after submission.
⚠️
Follow the structural logic: Introduction (relevance, research question, structure) → theoretical framework → methodology → results → discussion (interpretation, limitations, implications) → conclusion. Every chapter has a distinct function – overlap between results and discussion is a common weakness.
⚠️
Mandatory AI declaration in the appendix: Format: tool · version/date · purpose · extent of use. Full template in the "Using AI" tab. The declaration of authorship also covers the AI declaration – both must be consistent.
"A thesis is not a presentation. It is your first genuine scholarly contribution – a piece of new knowledge about learning with and through technology. Anyone who takes it seriously is entering the research community for the first time."
Phase model for technology-didactic theses
From the first conversation to submission – a structured approach
The path to a finished thesis can be divided into six clearly defined phases. Each phase has its own goals, typical activities and – in modern working methods – specific possibilities for using AI.
1
Orientation & Finding a Topic
Weeks 1–3 (BA) · Weeks 1–4 (MA)
Conversation with the supervisor: narrowing down the research interest, exploring the topic area
Initial literature search in databases (ERIC, FIS Bildung, Google Scholar, ResearchGate)
Formulating a preliminary research question (not yet final)
AI permitted: brainstorming research questions, exploring topic areas, initial literature search with Perplexity or Elicit
2
Exposé & Research Design
Weeks 3–6 (BA) · Weeks 4–8 (MA)
Refining the research question and hypotheses/assumptions
Choosing the methodological framework (qualitative/quantitative/DBR) and justifying it
Producing the exposé (3–5 pages BA, 5–8 pages MA) with a timeline
AI permitted: structuring help for the exposé, feedback on the line of argument, help with phrasing (with disclosure)
AI not permitted: having AI write the exposé without your own intellectual contribution
3
Theoretical Grounding & Literature Review
Weeks 5–10 (BA) · Weeks 7–16 (MA)
Systematic literature search and review: read primary sources, don't just cite them
Building the theoretical framework: definitions, models, state of research
Reference management with Zotero or Citavi
AI permitted: summarising texts for your own orientation, translation help for English-language sources, uncovering research gaps
AI not permitted: taking AI-generated summaries directly into the thesis without reading the original
4
Empirical Data Collection & Analysis
Weeks 8–16 (BA) · Weeks 14–22 (MA)
Data collection: interviews, questionnaires, classroom observation, artefact analysis
Transcription and preparation of the data
Analysis using the chosen method (content analysis after Mayring, grounded theory, statistical procedures)
AI permitted: support with transcription (Whisper, notta.ai), initial category-building as a discussion partner, visualising results
AI not permitted: AI independently interprets data and the results are adopted as-is
5
Writing & Argumentation
Weeks 10–18 (BA) · Weeks 16–24 (MA)
Writing the thesis in your own words: introduction, theory, method, results, discussion, conclusion
Consistent use of the academic terminology of technology didactics
AI permitted: style feedback on your own text, checking logical coherence, synonym suggestions, proofreading
AI not permitted: AI writes sections that are presented as your own intellectual work
Your own chain of argument must be recognisably and demonstrably self-developed
6
Revision, Declaration & Submission
Last 2 weeks
Full review for consistency, formatting and the bibliography
Produce the AI declaration in the appendix: all tools used with version, date, purpose and extent
Sign the declaration of authorship (which includes the AI declaration)
Submit as PDF + DOCX source file at the examinations office
What is an intervention study?
Empirically examining the effectiveness of technology-didactic measures
An intervention study examines whether and how a deliberately applied teaching-and-learning measure influences learning. It is the core of empirical theses in technology didactics. The choice of research design depends on the research question: qualitative designs explore processes, quantitative designs measure effects, and design-based research (DBR) iteratively develops learning environments.
The Three Research Designs
🔍 Qualitative
Qualitative Intervention Study
Understanding how and why
Methods
Guided interviews, content analysis after Mayring, concept mapping, small samples (n=4–20)
BA exampleWhat technical conceptions do pupils develop during the Arduino project? → interviews + content analysis
"The question determines the method – not the other way round. Anyone who chooses the method first and then looks for a matching question isn't doing research design, but method tourism."
Using AI in Academic Work
Tool, not ghostwriter – using AI responsibly
AI language models are powerful tools that can substantially support academic work – when used as thinking tools rather than as a substitute for authorship. The decisive question is not "Am I allowed to use AI?" but "What intellectual contribution am I making myself?"
In technology didactics, the same applies as in Papert's concept of learning: thinking must not be outsourced. Anyone who lets AI do the thinking for them acquires no competence – just like a learner who operates a machine without understanding its process.
Recommended AI Tools by Purpose
🔍
Perplexity AI
Literature search
Getting started in new research areas, an overview of the state of discourse, initial source leads
Suitable for orientation-level research
Always check sources against the original – AI hallucinates references
📄
Elicit
Literature analysis
Systematic literature search, extracting study results from papers, comparing studies
Well suited for systematic reviews
Only use peer-reviewed sources, always read them yourself
🤖
Claude / ChatGPT
Writing & thinking assistance
Feedback on your own text, rephrasing suggestions, discussing structure, checking arguments
Very useful as a thinking partner and proofreader
Do not use as author – your own voice must remain recognisable
🎙️
Whisper / notta.ai
Transcription
Automatic transcription of interviews and classroom recordings
Considerable time savings in qualitative research
Always proofread and correct transcripts manually
📊
Julius AI / ChatGPT (data)
Data analysis
Support with quantitative analysis, visualising results, coding datasets
Helpful for exploratory data analysis
Always verify statistical results with SPSS/R
📚
Zotero + AI Plugins
Reference management
Automatically creating reference entries, APA formatting, summarising PDFs
A must for every thesis
Always check APA output manually
AI Declaration Template
Every AI tool used must be declared in the appendix of the thesis using the following schema. The declaration is part of the declaration of authorship – missing or incomplete information counts as an attempt to deceive under the examination regulations of PH Schwäbisch Gmünd.
Template AI declaration in the appendix (mandatory format)
The following AI-assisted tools were used in the course of this thesis:
1. Claude Sonnet 4 (Anthropic, as of March 2025): Used for language revision of individual paragraphs in Chapter 3 (style feedback, sentence structure) and as a discussion partner for structuring the theoretical section. All content decisions, lines of argument and formulations are the author's own. Extent: approx. 8 interactions over 3 weeks.
2. Whisper v3 (OpenAI, as of February 2025): Used for the automatic initial transcription of the four guided interviews (total duration: 3 h 42 min). All transcripts were subsequently checked in full against the recordings, corrected, and reviewed for anonymisation.
3. Elicit (Elicit Inc., as of January 2025): Used for systematic literature search in phase 1 of the research (search terms: "conceptual change technology education", "maker education empirical"). All 24 identified sources were read in the original and independently evaluated. No AI-generated text was included in the thesis.
The declaration of authorship expressly also covers the correct and complete declaration of all AI tools used, in accordance with § [X] of the examination regulations.
The Three-Zone Model of AI Use
Not all AI use is the same. The following model distinguishes three zones that help you reflect on and situate your own use of AI:
🟢
Zone 1 – Productive use (recommended): AI as a tool that saves time and improves quality without replacing your own intellectual contribution. Transcription, style feedback, literature overviews, translation help, debugging R code, discussing structure. Your own thinking remains central and clearly recognisable.
🟡
Zone 2 – Grey area (proceed with caution): AI generates drafts that are then substantially revised. Permitted if your own revision is substantial and the AI use is declared. Risk: AI language and AI logic creep in unnoticed. Your own academic voice must remain recognisable.
🔴
Zone 3 – Not permitted (deception): AI writes chapters that are presented as your own work. Adopting AI-generated references without checking them. AI interprets research data and the results are adopted unchanged. AI use is not declared, or only partially declared. The research question is fully formulated by AI without your own intellectual contribution.
Citing in Technology Didactics
APA 7th Edition – the professorship's standard
In the Professorship for Technology & Its Didactics, APA 7th Edition is the mandatory citation style. Zotero with the APA 7 style is the recommended reference manager. Basic rule: every statement, model or figure that is adopted needs a citation – whether quoted verbatim or paraphrased.
Examples: Correctly Citing Key Sources
Journal article
Brown, J. S., Collins, A., & Duguid, P. (1989). Situated cognition and the culture of learning. Educational Researcher, 18(1), 32–42. https://doi.org/10.3102/0013189X018001032
Book chapter
Collins, A., Brown, J. S., & Newman, S. E. (1989). Cognitive apprenticeship: Teaching the crafts of reading, writing, and mathematics. In L. B. Resnick (Ed.), Knowing, learning, and instruction: Essays in honor of Robert Glaser (pp. 453–494). Lawrence Erlbaum Associates.
Monograph
Papert, S. (1980). Mindstorms: Children, computers, and powerful ideas. Basic Books.
Citing an AI tool (APA 7)
Anthropic. (2025). Claude (Version claude-sonnet-4-5) [Large language model]. https://www.anthropic.com
In text: (Anthropic, 2025) – with a note in the appendix on the specific purpose of use.
Common Citation Errors and How to Avoid Them
❌
Avoid secondary citations: "Brown et al. (1989), as cited in Müller (2020)" is only acceptable if the original is demonstrably inaccessible. In technology didactics, almost all key texts are available digitally (JSTOR, Google Scholar, ResearchGate, PH library access). Frequent secondary citation signals: "I have not read the primary source."
❌
Wikipedia, Studyflix, Lernhelfer as a source: Usable for initial orientation, but not citable academically. Consequence: find and read the original literature cited there directly. The same applies to teaching materials, textbooks without an academic apparatus, and popular-science articles without peer review.
❌
Adopting AI output without declaring it: Text and phrasing generated or substantially altered by AI models must be declared in the appendix. Non-declaration counts as an attempt to deceive. Paraphrased AI output must also be declared – "rewording" AI text does not remove the duty of disclosure.
❌
Incorrect page numbers: For verbatim quotations, the page number is mandatory in APA 7 (Smith, 2020, p. 45). If the page is missing, the citation is formally incomplete. For e-books without page numbers: use the paragraph number or chapter reference (para. 3 or ch. 2).
❌
Too many verbatim quotations: An academic thesis demonstrates understanding through paraphrasing – not by stringing quotations together. Use verbatim quotations sparingly: when the exact wording is substantively significant (e.g. definitions, terminological choices) or when something would be lost through paraphrasing.
✅
Always give the DOI: Where available, the DOI is mandatory in APA 7. Format: https://doi.org/10.xxxx/yyyy. Zotero does this automatically – still double-check before submission, since Zotero occasionally imports incorrect DOIs.
✅
Manage references from the start: Use Zotero from the very first source – not only at the end. Add every source as you read it for the first time, attach the PDF, enter the abstract and keywords. Zotero BibTeX export and the Word plugin allow seamless integration. Back up the Zotero library regularly.
✅
Use databases systematically: ERIC (Educational Resources Information Center), the FIS Bildung literature database, Google Scholar, JSTOR, Springer Link, Elicit for empirical studies. Combine search terms in English and German. Snowballing: check the reference lists of relevant articles for further sources.
Guidelines for Technology-Didactic Theses with AI
What works – what doesn't
The following overview summarises what counts as productive AI use in the Professorship for Technology & Its Didactics – and what doesn't. The benchmark is always: is the intellectual contribution recognisably the author's own?
✅ Permitted & recommended
Using AI as a brainstorming partner for research questions
Having your own texts checked for style, clarity and coherence
Processing interviews more efficiently with AI-assisted transcription
Making English-language texts easier to access with AI
Discussing and sharpening outline drafts with AI
Debugging statistics software code with AI
Speeding up literature search with Elicit or Perplexity
Declaring AI use fully and transparently
✗ Not permitted
AI writes whole chapters or sections without your own revision
Having the research question formulated entirely by AI
Adopting AI-generated references without checking them
Having the theoretical framework compiled exclusively by AI
Presenting AI's interpretation of research data as your own
Concealing or not declaring AI use
Signing the declaration of authorship without declaring AI use
Using AI as a substitute for reading primary sources
Common Weaknesses in Theses
❌
A topic instead of a research question: Many theses begin with "Topic: maker education" instead of an answerable question. The research question is the backbone of the thesis – everything else (theory, method, results) serves to answer it.
❌
Theory and empirical work disconnected: A common weakness: the theory section discusses concepts that play no role in the empirical section – and vice versa. The theoretical framework must directly inform the data-collection instruments, the analysis categories and the discussion of results.
❌
Results instead of discussion: A frequent problem: the discussion repeats the results instead of embedding them in the theoretical context. Discussion means: interpreting results against the state of research, explaining surprises, naming limitations, and drawing implications for practice and research.
❌
Pages instead of substance: Page counts are a minimum, not a quality guarantee. A 45-page thesis with a clear question, a clean design and substantial insight is preferable to a 65-page thesis with no coherent thread. Padding is recognisable and is assessed accordingly.
❌
Hiding limitations in the last paragraph: The study's limits (small sample, missing control group, self-selection bias) belong transparently in the discussion – not as an excuse at the end, but as scholarly reflection considered from the outset.
The Core Principle
"AI in academic work is like a 3D printer in the CreatorLab: a powerful tool that presupposes competence rather than replacing it. Operating the printer without understanding the process produces plastic parts. Using AI without thinking for yourself produces text. Neither is learning."
Next Steps & Contact
📬
Get in touch: Interested in a thesis? Send an email with a short draft exposé (topic, initial research question, methodological idea, 1–2 pages) to hannes.nepper@ph-gmuend.de – response within 2 weeks.
📖
Read relevant journals:Journal technik·education (tedu) · DiMawe – Die Materialwerkstatt · Journal of Technology Education · International Journal of Technology and Design Education · empirical studies can also be found in ERIC and FIS Bildung.
🔧
Experience the CreatorLab first-hand: Before starting a thesis in maker education, 3D printing or electronics, active participation in a making session in the CreatorLab at PH Gmünd is recommended. Technology didactics must be experienced first-hand before it can be researched.
🎓
Attend the thesis colloquium: Colloquia are held regularly as part of the professorship's seminars, where ongoing theses are presented and discussed. Attending as a listener is possible and highly recommended – you learn an enormous amount from the feedback given to others.
📋
Technology Didactics · PH Schwäbisch Gmünd
Guide: Technology Lesson Design
Structure, formulating learning objectives, lesson-flow planning and hazard assessment following the STOP principle for technology education.
Structure & Quality Requirements
More than a lesson-flow plan – a didactic argument
A lesson design is not a mere description of the sequence of events. It is a didactically justified document that shows you have thought the lesson through, worked through the subject matter analytically, and made a methodologically justified decision for a particular approach. In the technology subject, safety aspects and hazard assessment are additional mandatory components. The design is also a learning tool: writing it carefully means mentally teaching the lesson once before you even step through the door.
The coherent thread is decisive: class profile → subject-matter analysis → didactic justification → learning objectives → methods → lesson flow must build on one another. If a method appears in the lesson-flow plan that was not justified in the didactic analysis, the thread is broken. The document must read as a coherent whole.
1
Cover sheet & general information
½ page
2
Class profile & starting point for learning
1–1.5 pages
3
Subject-matter analysis
1.5–2 pages
4
Didactic Analysis & Justification
1–1.5 pages
5
Learning Objectives (operationalised sub-goals)
½–1 page
6
Hazard Assessment (STOP Principle)
Mandatory for practical work
7
Methodological Planning & Differentiation
1 page
8
Lesson Flow Plan
Table format
9
Appendix (materials, references)
variable
"A lesson design is the proof that teaching is not improvisation."
Learning objectives describe what pupils should be able to do or know by the end of a lesson – not what will be done. A learning objective is not a topic (not: "The pupils cover the electrical circuit"), but an observable competence. Learning objectives have three dimensions: 🧠 Cognitive (knowledge, understanding, applying, analysing, evaluating, creating) · 🔧 Psychomotor (craft skills, technical precision) · ❤️ Affective (attitudes, motivation, dispositions). All three dimensions are relevant in technology education – and psychomotor objectives are often forgotten in other subjects.
Learning objectives are the lesson's measuring instrument: at the end of the lesson, the teacher can only judge whether it was successful if the objectives were formulated operationally. Without verifiable objectives, evaluating a lesson is not possible.
Formulation Formula
The pupils [operator] [content], by [method], so that [expected outcome].
Example: The pupils construct a parallel circuit with two LEDs by independently implementing the circuit diagram on a breadboard, so that both LEDs light up and can be switched independently.
Level 1 · Remember
name · state · list
Level 2 · Understand
explain · describe · interpret
Level 3 · Apply
apply · construct · use
Level 4 · Analyse
analyse · examine · break down
Level 5 · Evaluate
judge · evaluate · compare
Level 6 · Create
design · develop · create
Lesson Flow Plan – Phase Model (45 min)
Time
Phase
Teacher Activity
Pupil Activity
Method
0–5 min
Introduction
Present a problem situation, activate prior knowledge
Observe, voice initial ideas
Stimulus, guided discussion
5–10 min
Safety Briefing
Identify hazards, explain protective measures
Listen actively, put on PPE
Direct instruction
10–30 min
Development
Set the task, guide learning, differentiate
Construct and experiment independently
Group/pair work, making
30–38 min
Consolidation
Have results presented, record them
Present results, take notes
Whole class, board notes
38–43 min
Reflection
Relate back to learning objectives, transfer question
Name the learning gain
Exit ticket, quick round
43–45 min
Cleanup
Guide the cleanup routine
Clean tools and put them away
Checklist
⚠️ A safety briefing is mandatory every time a tool is used for the first time. Never forget cleanup time – 5 min minimum, allow more for laser cutting or soldering.
In technology education, pupils work with tools, machinery, electrics and sometimes chemicals. As a teacher you bear a duty of supervision and care towards all pupils – a legal obligation arising from the German Civil Code (§ 832 BGB), the Baden-Württemberg School Act and DGUV regulations. DGUV Regulation 81 (Schools) requires teachers to carry out a documented hazard assessment before using tools and machinery, and to demonstrably instruct pupils in their safe use.
A hazard assessment is not a bureaucratic exercise – it is a systematic thinking tool that prevents harm and legally protects the teacher in the event of an incident. The STOP principle provides a clear, legally grounded order of priority: from the most effective measure (substitution) to the least effective (personal protective equipment). This hierarchy is not optional – it is the statutory standard of occupational safety (§ 4 ArbSchG, German Occupational Safety and Health Act).
Hierarchy of protective measures – always check from top to bottom (§ 4 ArbSchG)
S
Substitution
→
T
Technical measures
→
O
Organisational measures
→
P
Personal (PPE)
⚠️ PPE is always the last measure – not the first. Safety goggles do not replace a missing extraction system. Measures at higher levels work regardless of how people behave – measures at lower levels depend on behaviour and are therefore less reliable.
S – Substitution
Eliminate the hazard at its root
The hazard is eliminated by replacing the dangerous substance, process or tool. Often not fully achievable in a school context, but must be systematically checked. Especially relevant for age groups where certain technologies should not yet be used.
Examples in Technology Education
→Replacing soldering irons with solder-free plug connections (Years 5/6)
→Replacing leaded solder with lead-free solder
→Replacing sawing with pre-cut material if only measurement is being practised
→Replacing laser cutting with milling or hand-cutting for younger classes
→Replacing solvent-based adhesives with water-based ones
T – Technical Measures
Contain the hazard with technical safeguards
Technical safeguards that reduce hazards by spatially or temporally separating people from the danger. They work regardless of behaviour – hence high reliability. Regular maintenance and inspection per DGUV are mandatory.
Examples in Technology Education
→Guards on band saw, circular saw, milling machine
→Extraction system for wood dust (limit: 2 mg/m³) and soldering fumes
→RCD circuit breakers and current limiters for electrical work
→Low voltage (5V/12V) instead of 230V in pupil experiments
→Laser cutter: viewing shield, extraction, laser class 2
→3D printer: enclosed unit, room ventilation against plastic fumes
O – Organisational Measures
Limit the hazard through rules and instructions
Rules, instructions and behavioural requirements that reduce the risk of accidents. Effective only if consistently followed and enforced. Organisational measures are behaviour-dependent – hence less reliable than technical ones.
Examples & Requirements
→Written safety briefing before first use – documented with a signature (DGUV requirement)
→Limiting group size at machines: max. 2 pupils at a time
→No loose hair, no jewellery when using the lathe/mill
→Regular maintenance and inspection of all tools (keep an inspection log)
→Duty of supervision: the teacher never leaves the workshop during practical work
→Posting operating instructions for all machines (DIN format, pictograms)
P – Personal Protective Equipment (PPE)
Last line of defence – never the first
PPE protects the individual when all other measures are not sufficient. PPE is always supplementary – never a substitute for S, T or O. The teacher must ensure PPE is available, functional and in the correct size.
Documentation requirement (DGUV Regulation 81): The safety briefing must be documented in writing. Pupils (or parents/guardians for minors) confirm they have taken note by signing. This list belongs in the appendix of the lesson design and is presented at the teaching assessment.
⚠️
Before every lesson: operational safety check: Check all tools and machines for obvious defects before the lesson (cables, guards, function). Take defective equipment out of service immediately and label it. No repairs by pupils or by unqualified persons.
❌
"They already know that" is not a justification: Even if the class is familiar with a machine, a fresh short briefing is mandatory for new tasks, after long breaks, or under changed conditions. Familiarity does not protect against liability.
❌
PPE as the only measure: "Wearing safety goggles is enough" is not a complete hazard assessment. The STOP principle requires all four levels to be checked systematically – PPE is only to be used where S, T and O are insufficient or not possible.
✅
Use the hazard assessment as a learning opportunity: As part of safety briefings, pupils can identify hazards themselves and propose measures. This fosters risk competence and technical judgement – a core aim of technology education.
"Safety is not a constraint on technology education – it is its ethical foundation. Teaching pupils to work with technology means teaching responsibility at the same time."
Typical Mistakes in Lesson Designs
❌
A topic instead of a learning objective: "Pupils learn about the electrical circuit" is not a learning objective. Learning objectives must begin with an operator verb and describe a verifiable competence.
❌
A method without justification: "I chose group work" is not a justification. Why this method for this topic with this class?
❌
No safety briefing in the lesson flow plan: Mandatory for every practical lesson – as its own phase with a time allocation.
❌
Consolidation dropped under time pressure: Better to include one fewer task – but carry out consolidation and reflection in full.
✅
Coherent thread: Learning objectives, didactic analysis, methods and the lesson flow plan interlock – every decision is justified.
✅
Contact: While writing your lesson design, a draft (3–4 days beforehand) can be submitted for feedback: hannes.nepper@ph-gmuend.de
🧠
Technology Didactics · Conceptions Research
Guide: Responding to Learners' Conceptions
Six didactic response modes for technology-related teaching – from diagnosis through confrontation and elaboration to reflection. With theoretical grounding, classroom examples, diagnostic tools and sequencing guidance.
Key publication: Nepper, H. H., & Windelband, L. (2022). Mögliche Reaktionsmodi zu Lernendenvorstellungen im technikbezogenen Unterricht [Possible response modes to learners' conceptions in technology-related teaching]. In E. Eichelberger et al. (Eds.), Forschendes Lernen und Lehren im Textilen und Technischen Gestalten (pp. 195–206). hep Verlag.
"Learners' conceptions are not errors to be overwritten – they are the starting condition of every learning process. Anyone who ignores them teaches past the learner."
What are learners' conceptions?
Pupils do not enter technology lessons as blank slates. They bring everyday conceptions, subjective theories and intuitive concepts – cognitively coherent patterns of interpretation that have arisen from everyday experience, media, social interaction and encounters with artefacts. These conceptions are not mere errors; in the context in which they arose they often have meaningful explanatory power and allow learners to experience themselves as capable of action.
They are didactically relevant for two reasons: first, powerful misconceptions block the construction of correct technical concepts. Second, they offer points of connection – they show where teaching must begin in order to be cognitively accessible (Ausubel, 1968: "The most important single factor influencing learning is what the learner already knows.")
Types of Learners' Conceptions
Preconceptions
Prior knowledge from everyday experience that exists before instruction. Often functionally correct in everyday contexts, but incomplete from a subject-specific standpoint. Example: "Current flows from the battery to the bulb, then it's used up."
Misconceptions
Technically incorrect concepts that actively impede learning. Resistant to simple instruction. Example: "Heat is a substance that flows from warm to cold."
Fragmentary Knowledge
Isolated, unconnected knowledge fragments without a coherent concept. Learners can reproduce facts but cannot apply them. Example: knows "Ohm" but cannot apply resistance to circuits.
Conceptual Change – Why Conceptions Are So Stable
The central theorem of conceptions research: learners' conceptions are powerful and stable (Duit, 2009). Teaching that ignores them passes them by – pupils give the expected answers in exams but continue to think according to their original logic: resistance to conceptual change.
Posner et al. (1982) – Four Conditions for Conceptual Change
1.Dissatisfaction with the existing conception
2.Intelligibility of the new conception
3.Plausibility – the new conception must be experienced as credible
4.Fruitfulness – the new conception solves new problems
Chi (2008) – Ontological Miscategorisation
Chi distinguishes weak CC (enrichment of knowledge within a category) from radical CC (a shift in ontological category). Understanding electric current as a "substance" rather than a "process" is radical change – particularly persistent, since learners do not readily give up their old category.
Consequence for teaching: Changing conceptions is not a matter of overwriting old information, but a laborious process of cognitive restructuring. It takes time, disruption, elaboration and metacognitive reflection – and must be didactically supported. This is exactly where the six response modes of Nepper & Windelband (2022) come in.
The six response modes do not form a rigid, linear script but a didactic repertoire – depending on the type of conception, the class and the classroom situation, they are combined and weighted differently. All modes build on a prior diagnosis.
1
Mode 1 · Prerequisite for all further modes
Diagnosis
Making conceptions visible – without immediately judging or correcting them
Goal
Making the learners' subjective theory explicit before instruction begins, or at a key juncture
Pupils sketch a circuit before it is built · "Explain how a motor works" as a drawing task
⚠️
Common mistake: Learners give the presumed "correct" answer instead of their actual conception. Counter-strategy: "What do you think happens?" instead of "What do you know?" · anonymous card polling · requiring a preliminary sketch before technical terms are named.
2
Mode 2 · Initiating cognitive dissonance
Confrontation
The existing conception fails to explain a concrete phenomenon – disruption as a learning opportunity
Confrontation deliberately induces cognitive dissonance: the existing conception is confronted with a phenomenon it cannot explain. The goal is Posner's first condition for conceptual change: dissatisfaction with the old conception.
Methods
Contradiction experiment · anomaly task · fix a written prediction → experiment → discrepancy analysis · counter-example · "What would happen if …?" scenario
Circuit Example
Conception: "Current is used up at the bulb." → ammeters before and after the bulb show the same value → dissonance experienced, not merely demonstrated.
⚠️
Critical point: confrontation must be staged as a discovery, not as being caught out. A humiliating confrontation increases resistance. Pupils must observe and name the discrepancy themselves.
3
Mode 3 · Building and connecting the new concept
Elaboration
The new concept is deeply processed, linked to prior knowledge and explored in several contexts
After confrontation, a cognitive vacuum arises: the old conception is destabilised, a new one not yet consolidated. Elaboration fills this vacuum with technically correct, connected knowledge – through active construction, analogy-building and elaborative encoding.
Analogy Work
Water cycle ↔ electric circuit · heat flow ↔ information flow · gear train ↔ gear ratio. Analogies activate existing correct schemas for new domains.
Concept Mapping
Learners create their own concept networks and extend them after teaching sequences. Makes the depth of connections and gaps visible.
Artefact as Medium
In technology education: model building, circuit assembly, prototyping. The physical artefact is a medium of elaboration – the body thinks along (embodied cognition).
4
Mode 4 · Conceptual restructuring
Change
The old conception is explicitly deconstructed – extension, replacement or restructuring
Change can take three forms: extension (the conception was correct in a narrow context and is broadened), replacement (the old conception is abandoned) and restructuring (the ontological category is switched – the most difficult type according to Chi).
Concept Change Journal
In writing: "Before I thought … · Now I understand … · What changed was …" – the comparison makes the shift visible and guards against mere verbal agreement.
Types at a Glance
Extension: current as flow remains, but is now understood as a circuit Replacement: heat as substance → heat as energy Restructuring: force as a property → force as an interaction
⚠️
Critical warning: verbal agreement is not an indicator of conceptual change. Only tasks that require active construction (explaining, drawing, applying to a new context) show whether genuine change has taken place.
5
Mode 5 · Transfer and stabilisation
Application
The new concept is tested in varying contexts – transfer creates stability and prevents regression
A new concept that is not applied in varying contexts remains fragile. Under stress or in new situations, the cognitive system falls back on the old, automated conception (regression). Varied transfer is the most important safeguard against this.
Near Transfer
Same concept, slightly changed context. Build a new circuit variant, find the error in a circuit diagram, swap components.
Far Transfer
Applying the concept across subjects: circuit logic in hydraulic systems or heat-conduction analogies.
Teach-Back
Learners explain to others (peer teaching, explainer video, poster). Gaps in their own understanding become visible in the process.
6
Mode 6 · Metacognition and learner identity
Reflection
The learning process itself becomes the object – where did my conception come from? What moved me?
Metacognitive reflection is the most demanding stage: learners turn their attention to their own thinking process. They ask not only what they have learned, but how and why their conception changed. This mode strengthens epistemic autonomy and contributes to pre-scientific competences.
Methods
Learning portfolio with before-after comparison · concept-change essay · structured reflection card · peer feedback on learning progress · visualising one's own learning path
Guiding Questions for Learners
"Where did my original conception come from?" "What didn't convince me – and why?" "Where might my new conception still have gaps?" "What do I take away for future learning problems?"
Epistemic dimension: reflection turns learners into researchers of themselves. They develop a differentiated picture of how knowledge arises and how scientific concepts differ from everyday conceptions.
Diagnosis is mode 1 and, at the same time, the prerequisite for all other modes. Without a sound diagnosis, the response remains arbitrary. This overview presents proven diagnostic tools for technology education, ordered by effort and depth of insight.
Diagnostic Tools at a Glance
🃏
Concept Cartoons
FASTYEARS 5–12
Picture cards show several characters with different (some correct, some incorrect) statements about a technical phenomenon. Pupils mark who they think is right and justify it in writing. Advantage: low-threshold, non-threatening, activates learners' own conceptions without feeling like a test. Technology example: characters argue about the function of a fuse, a transistor or a gear ratio.
🧪
POE – Predict · Observe · Explain
MEDIUMYEARS 7–10
A structured three-step protocol: (1) pupils write down what they expect. (2) They observe the experiment. (3) They compare their prediction with their observation and explain discrepancies. Important: the prediction must be written down before the experiment – otherwise it gets rationalised after the fact. Directly connects diagnosis with confrontation (mode 2).
✏️
Drawing & Labelling
FASTFROM YEAR 5
Pupils draw and label a technical system without a template: circuit, motor, gearbox, thermal insulation. Drawings externalise mental models in ways that verbal explanations often cannot. Analysis: missing arrows, wrong causal directions, missing parts reveal gaps in conceptions. Comparing drawings from different learning phases → conceptual change becomes directly visible.
🗺️
Concept Mapping (diagnostic)
TIME-INTENSIVEFROM YEAR 8
Pupils arrange terms (on cards) and connect them with labelled arrows. Shows not only which terms are known, but which relations are drawn between them. Especially suited to comparing the start and end of a unit. Software: CmapTools (free), Miro, a Padlet variant. Also possible by hand on large sheets of paper.
🎙️
Clinical Interview
INTENSIVESEMINAR / THESIS
Open-ended questions that elicit conceptions without leading the learner: "Explain to me how this works" · "What happens if …?" · "How do you know that?" Do not evaluate answers, but keep probing deeper (Piaget). In studies: an ideal instrument for seminar and thesis work, case studies on diagnosing conceptions.
Quality criteria: good diagnostic instruments avoid the Hawthorne effect (learners display expected behaviour instead of their own conception). Key points: (1) open rather than closed formats, (2) require justification, (3) phrasing without an implied correct answer, (4) establish psychological safety – diagnosis is not a test.
The six modes are not a checklist to be worked through linearly. Teaching that takes conceptions seriously is an adaptive process: diagnosis informs which modes are needed, and to what depth. The following overview shows typical patterns found in real teaching sequences.
Suited to persistent misconceptions (Chi: ontological restructuring needed). The confrontation must land noticeably – otherwise elaboration is experienced as abstract. Mode 5 (application) follows in subsequent lessons.
Sequence B – Elaboration-Focused Pattern (when conceptions connect well to correct concepts)
When diagnosis shows that conceptions are only incomplete rather than fundamentally wrong. Confrontation is skipped – instead the existing conception is directly elaborated, tested in tasks and then explicitly extended. Time-saving with a strong class.
Sequence C – Iterative Pattern (multi-layered conceptions · several teaching units)
1→2→3→1 (re)→2 (re)→4→5→6
For conceptions made up of several layers (e.g. energy: substance conception + consumption conception + incorrect conservation concept). Diagnosis and confrontation are run through several times, since each confrontation addresses only one layer. Time-intensive, but unavoidable for radical CC.
Planning note for lesson designs: the choice of sequence pattern must follow from the diagnosis – not from the textbook or the curriculum. Anyone who does not know which conceptions a class brings with them cannot make a well-founded sequencing decision. Diagnosis is therefore always the first planning step – even if it is sometimes only implicitly visible in the lesson design.
Compared to science didactics, technology didactics has a distinctive feature: learning does not happen only through concepts, but also through artefacts, actions and construction processes. The response modes therefore acquire a material dimension – body and tool become part of the learning process.
Typical Misconceptions in Technology Education
CURRENT
"Current is used up by the bulb"
Conception: current flows out of the battery, releases its energy at the bulb – no return flow. Type: substance conception (Chi: ontological restructuring needed). Confrontation: ammeters before and after the bulb → same value. Elaboration: distinguishing energy from current: current is charge flow (is conserved), energy is converted into heat/light.
HEAT
"Metal is colder than wood at the same temperature"
Conception: metal has a lower temperature because it feels colder. Type: confusion of temperature and thermal conductivity. Confrontation: thermometer on metal and wood → same temperature. Elaboration: thermal conductivity as a material property – metal conducts heat away from the skin quickly → feels cold.
GEARS
"A bigger gear turns faster"
Conception: size = speed. Type: preconception from everyday size-force logic. Confrontation: build a gear train, count rotations: the small gear turns faster. Elaboration: gear ratio: number of teeth × rotational speed = constant. Force and speed are inversely related.
FORCE
"Heavy things fall faster"
Conception: falling speed depends on weight (Aristotle). Reinforced by everyday experience (leaf vs. stone). Confrontation: same aerodynamic shape, different mass → fall at the same speed. Elaboration: explicitly address air resistance as the decisive variable.
The Artefact as a Didactic Lever
As a Diagnostic Tool
Learners build a circuit based on their own conception. The circuit externalises the mental conception and makes it analysable. Errors in the circuit reveal errors in thinking.
As a Medium of Confrontation
The incorrectly built circuit doesn't work – that is the experienced confrontation. Not an abstract counter-example, but the concrete failure of one's own model.
As a Medium of Elaboration
The correct concept is built into the artefact – repairing a circuit, optimising a gear train. Embodied cognition: the body understands what the mind has not yet fully grasped.
As a Transfer Test
Mode 5: learners construct a new artefact under changed conditions. Success suggests genuine conceptual understanding – not just rote memorisation.
The response modes of Nepper & Windelband (2022) are embedded in a broad research field: conceptual change research, conceptions research in science and technology, and cognitivist learning psychology.
Nepper, H. H. & Windelband, L. (2022)
Mögliche Reaktionsmodi zu Lernendenvorstellungen im technikbezogenen Unterricht [Possible response modes to learners' conceptions in technology-related teaching]. In E. Eichelberger et al. (Eds.), Forschendes Lernen und Lehren im Textilen und Technischen Gestalten (pp. 195–206). hep Verlag.
Content: primary source · derivation of the six modes for technology-related teaching · empirical grounding from classroom studies
Posner, G. J., Strike, K. A., Hewson, P. W. & Gertzog, W. A. (1982)
Accommodation of a scientific conception: Toward a theory of conceptual change. Science Education, 66(2), 211–227.
Content: the classic four-conditions model for conceptual change · a foundational text of CC research
Chi, M. T. H. (2008)
Three types of conceptual change: Belief revision, mental model transformation, and categorical shift. In S. Vosniadou (Ed.), International Handbook of Research on Conceptual Change (pp. 61–82). Routledge.
Content: distinguishing weak from radical CC · ontological miscategorisation · implications for instructional design
Duit, R. (2009)
Bibliography STCSE – Students' and Teachers' Conceptions and Science Education. IPN Kiel. (8th ed.)
Content: the most comprehensive bibliography on conceptions research · over 8,000 entries · also relevant for technology conceptions
Ausubel, D. P. (1968)
Educational Psychology: A Cognitive View. Holt, Rinehart & Winston.
Content: assimilation theory · "What the learner already knows" · basis for cognitively accessible teaching
Vosniadou, S. (Ed.) (2013)
International Handbook of Research on Conceptual Change. (2nd ed.). Routledge.
Content: the most up-to-date standard reference in CC research · overview of all major theoretical strands
Note for Seminar and Thesis Work
Coursework dealing with learners' conceptions in technology education should cite Nepper & Windelband (2022) as the primary source and situate it within the context of international CC research (Posner, Chi, Duit). A purely phenomenological description of pupils' errors without a theoretical framing is not sufficient for academic work. Duit's (2009) bibliography provides an entry point into domain-specific conceptions research.
🤖
AI Didactics · PH Schwäbisch Gmünd
Guide: Prompt Engineering for Lesson Preparation
How teachers can use AI language models deliberately for planning, differentiation and creating materials – with prompt templates for technology education.
What is prompt engineering – and why is it a teaching competence?
Prompt engineering refers to the systematic design of inputs (prompts) to AI language models in order to obtain high-quality, reliable and task-appropriate outputs. A prompt is not simply a question – it is a didactic instruction to a machine: precise, contextualised and aimed at a defined goal.
For teachers, prompt engineering is a key competence for three reasons: (1) time efficiency – differentiated materials, mark schemes and lesson plans in minutes rather than hours. (2) Didactic control – those who master prompting control level, format, terminology and pedagogical direction. AI without prompt competence delivers mediocre results. (3) Role-model function – teachers who can teach prompt engineering enable pupils to use AI critically in technology education.
The Six Building Blocks of an Effective Prompt
🔬 Sample Prompt · Worksheet on Electrical Circuits
ROLEYou are an experienced technology education specialist for Years 5–8 at a secondary school in Baden-Württemberg.
CONTEXTClass 7b, 24 pupils, 6 of whom need reading-comprehension support. Prior knowledge: conductor/insulator from Year 6.
TASKCreate a two-page worksheet on distinguishing series and parallel circuits, with a cloze text, a drawing task and a transfer question.
FORMATPlain text for a Word document. Numbered tasks. No Markdown formatting.
LEVELLanguage level B1, sentences max. 15 words. Transfer question at Bloom level "Apply".
CONSTRAINTSNo formulas. Everyday reference: Christmas lights. Mark scheme as a separate section.
What Each Building Block Actually Does
The six building blocks are not a rigid template but a thinking framework: the more of these aspects a prompt makes explicit, the less the AI has to "guess" – and the less rework is needed.
ROLE
Sets the subject-specific and linguistic frame. Without a role, the AI answers as a "generic assistant" – with a role, as a subject specialist with a didactic perspective.
CONTEXT
Class, prior knowledge, support needs. The most frequently missing block – and the one with the biggest effect on fit.
TASK
What exactly should be produced – with scope, structure and task types. Vague tasks ("do something about circuits") produce vague results.
FORMAT
Target medium (Word, slide, table), structure, formatting conventions. Prevents having to reformat afterwards.
LEVEL
Language level, sentence length, cognitive demand level (e.g. Bloom). Makes differentiation explicit rather than implicit.
CONSTRAINTS
What to avoid (formulas, technical terms), desired references (everyday context) and extra requirements such as mark schemes.
Ready-Made Prompt Templates for Everyday Teaching
The following templates are ready to copy and paste – simply replace the placeholders in square brackets [Topic] with your own content. The difficulty level indicates how much refinement is typically needed – "Basic" usually produces directly usable results, "Advanced" benefits from one or two rounds of iteration (see the "Workflow & Practice" tab).
📄
Create a Worksheet
Basic
You are a technology education specialist for [school type] in Baden-Württemberg.
Create a worksheet on [topic] for Year [year group].
- Three tasks: revision, application, transfer
- Language level [A2/B1/B2], everyday reference: [example]
- No jargon without explanation
- Mark scheme as a separate section
🔀
Generate Differentiation
Medium
Create three versions of the same material:
VERSION A (basic): cloze texts, B1 level, max. 10-word sentences
VERSION B (standard): as-is
VERSION C (extension): transfer task at Bloom level "Evaluate"
Same page count for all three.
🪞
Have a Lesson Design Reflected On
Advanced
You are an experienced technology subject mentor.
Critically analyse my lesson design:
[insert design]
Answer the following questions specifically:
1. Are the learning objectives operationalised and verifiable?
2. Does the method fit the class and the content?
3. Is the timing realistic (safety briefing,
development phase, consolidation, cleanup)?
4. What differentiation measures are missing?
5. What would a seminar supervisor criticise
first during the teaching assessment?
Be direct and specific. No general praise.
🧠
Concept Cartoon / Diagnosing Conceptions
Advanced
You are a technology education specialist with expertise
in learners' conceptions.
Create a concept cartoon on the topic [topic].
Format: three speech bubbles for Max, Layla and Tom –
all three statements are plausible, but different.
Include the following typical misconceptions:
[misconception 1], [misconception 2].
One statement should be technically correct.
No statement should seem obviously "correct".
Each statement max. 2 short sentences.
Add a teacher's commentary at the end:
which conception belongs to which character?
Why is it didactically valuable?
📊
Develop a Questionnaire / Data-Collection Instrument
Medium
You are a subject didactics specialist with experience
in quantitative classroom research.
Develop a pre-post questionnaire on the topic
[topic, e.g. self-efficacy in soldering].
Requirements:
- 8–10 items on a 5-point Likert scale
- Based on [validated scale, e.g. ASKU]
- Language: B1, max. 15 words per item
- 4 positively and 4 negatively worded items
(for internal consistency checking)
- No jargon without explanation
Add notes on analysis at the end
(Cronbach's alpha, reverse coding).
🔬
Develop Experiment Instructions
Medium
You are a technology education specialist focused
on workshop safety.
Develop experiment instructions on the topic
[topic, e.g. soldering exercise with a terminal block]
for Year [year group], group size [number].
Structure:
1. Learning objective (operationalised)
2. Materials list with quantities
3. Safety notes (DGUV-compliant, as a
separate, highlighted section)
4. Procedure in numbered steps
5. Observation task for pupils
6. Analysis questions (Bloom level "Analyse")
Language B1. Estimate the time required realistically,
including setup and cleanup.
✅
Create Feedback & Assessment Rubrics
Basic
You are a technology education specialist for [school type].
Create an assessment rubric for
[task, e.g. self-built bridge model]
in Year [year group].
- 4 assessment criteria, clearly self-assessable
by pupils
- 3 levels per criterion (e.g. "partially
achieved / achieved / exceeded expectations")
- Concrete, observable wording instead of
vague terms like "good" or "satisfactory"
- Table format: Criterion | Level 1 | Level 2 | Level 3
Add 2 example sentences for individual
feedback per level, phrased informally
and in an appreciative tone.
Prompt Strategies: From Simple to Advanced
Not every task needs the same prompt strategy. The following overview ranks six common strategies by complexity and shows what they are particularly suited to in a school context. Several strategies can also be combined – for example, role prompting with chain-of-thought for a lesson reflection "from the pupil's perspective, step by step".
Strategy
Principle
When to Use
Example
Zero-Shot
Direct task with no example
Simple, clearly defined tasks
"Explain series vs. parallel circuits."
Few-Shot
Task + 2–3 examples
When format or style matters
"Create learning objectives following this pattern: [example 1]…"
Chain-of-Thought
AI reasons step by step
Complex analyses, planning tasks
"Think step by step: analyse this design…"
Role Prompting
AI adopts a perspective
Feedback, proofreading, pupil perspective
"You are a Year 8 pupil. What do you find unclear?"
Iterative Refinement
Improve the result across follow-up prompts
Long materials, several revision rounds
"Good. Make the transfer question more demanding."
Persona Split
Two perspectives at once
Differentiation at a glance
"Write as the teacher AND as a pupil for the same task."
Combining Strategies – an Example
For reflecting on a lesson design, role prompting and chain-of-thought can be combined: the AI adopts the perspective of a subject mentor and works through the analysis in traceable steps. The result reads like a structured reflective conversation rather than a blanket assessment – and can therefore be used directly to prepare for a teaching assessment.
From First Prompt to Finished Material: an Iterative Workflow
The most common mistake when working with AI language models: treating the first output as the finished product. Experience shows that a short, iterative dialogue in 3–4 steps gets to a usable result much faster than a single, overloaded "mega-prompt".
1 📝
Request a Rough Draft
Use the six building blocks (role, context, task, format, level, constraints) to generate a first draft – deliberately without detailed requirements yet.
2 🔍
Check Subject Matter & Didactics
Check the content against the literature/curriculum, assess fit with the class. This is where you decide whether to refine further or start over.
3 🔁
Refine Specifically
Concrete follow-up prompts instead of a global rewrite: "Make task 2 more demanding", "Shorten the introduction to 3 sentences", "Add a differentiation option".
4 ✅
Finalise & Take Responsibility
Final check: verify sources, figures and safety notes, adapt formatting for the target medium, add your own didactic touch.
Do's & Don'ts in Practice
✅ Recommended
Iterate with short, specific follow-up prompts instead of rewording everything
Provide your own examples/materials as a few-shot template
When unsure, specifically ask for sources or justifications
Generate results for different classes in parallel (persona split)
Consciously retain your own final responsibility for subject matter and didactics
❌ Avoid
Adopting generated safety notes or legal references without checking them
Entering personally identifiable pupil data into commercial AI tools
Using a single, very long prompt for several unrelated tasks
Using AI output as a mark scheme 1:1 without checking it yourself
Trusting AI-generated statistics or study references without verification
Transfer to the Classroom: Teaching Pupils to Prompt
The same six-building-block approach can be simplified age-appropriately to teach pupils critical AI use – for example as a three-step model "Who? What? How exactly?" (role/context, task, format/level). This turns prompting itself into a subject of instruction: pupils compare differently phrased prompts and their results, developing a feel for how language steers machine output – a competence that reaches far beyond technology education.
Limits, Risks and Responsibility
⚠️ Systematic limitations: AI hallucinates plausible-sounding facts, sources and legal texts – always check DGUV regulations, curriculum references, standards (DIN/ISO) and safety notes against original sources. AI does not know your class and produces average materials. Data protection: do not enter pupil names or personal data into commercial AI systems. For use with pupils, only use GDPR-compliant tools (e.g. fobizz). The teacher bears the pedagogical, subject-specific and legal responsibility for all materials used – regardless of how they were produced.
Typical Risk Areas at a Glance
🧩
Hallucinations
Invented studies, incorrect standard designations or outdated regulations often sound convincingly phrased. Countermeasure: specifically ask for sources and verify them independently.
🔒
Data Protection
Names, class lists or performance data do not belong in commercial chat tools. For use with pupils, use only GDPR-compliant platforms approved by the school.
⚖️
Responsibility
AI-generated content is a draft, not an approved material. Subject-specific, pedagogical and legal responsibility remains with the teacher – regardless of how it was produced.
📊
Averageness
Without explicit context, AI defaults to statistical averages – the specific class with its particular characteristics must be actively supplied.
Checklist Before Classroom Use
Have all facts, figures and citations been checked against original sources?
Are safety notes (DGUV, workshop regulations) complete and correct?
Does the language level actually fit the class – not just in theory?
Was personal data never entered into the AI tool at any point?
Is a GDPR-compliant tool planned for use with pupils?
Does the material bear a recognisable didactic touch of your own?
"AI doesn't make teachers better – it makes good teachers more efficient."
§
Legal Notice
Information pursuant to § 5 TMG (German Telemedia Act) · § 55 RStV (Interstate Broadcasting Treaty)
Responsible for Content
Prof. Dr. Hannes Helmut Nepper
Holder of the Professorship for Technology and Its Didactics
Pädagogische Hochschule Schwäbisch Gmünd (University of Education)
Oberbettringer Straße 200
73525 Schwäbisch Gmünd
Baden-Württemberg, Germany
Pädagogische Hochschule Schwäbisch Gmünd
Public-law corporation (Körperschaft des öffentlichen Rechts)
Rectorate: Oberbettringer Straße 200, 73525 Schwäbisch Gmünd
Supervisory authority: Ministry of Science, Research and the Arts of Baden-Württemberg
VAT Identification Number
Pursuant to § 27a of the German VAT Act (UStG): as a public-law corporation, Pädagogische Hochschule Schwäbisch Gmünd is exempt from VAT (§ 4 No. 22 UStG).
Responsible for Content Pursuant to § 55 (2) RStV
Prof. Dr. Hannes Helmut Nepper (address as above)
Disclaimer
Liability for content: the content of this site has been created with the greatest possible care. No guarantee is given for the accuracy, completeness or timeliness of the content (§ 7 (1) TMG). As a service provider, we are responsible for our own content in accordance with general law; there is no obligation to monitor transmitted or stored third-party information (§§ 8–10 TMG).
Liability for links: this website contains links to external third-party websites over whose content we have no influence. The respective provider is always responsible for the content of linked pages. Permanent monitoring of linked pages without concrete evidence of a legal infringement is not reasonably expected.
Technical Implementation
This website was created through a prompt engineering process using Claude.ai (Anthropic). Concept, content and iterative development were carried out by Prof. Dr. Hannes Helmut Nepper; the technical implementation was produced through AI-assisted generation of HTML, CSS and JavaScript.
Copyright
The content and works created by the site operator are subject to German copyright law (UrhG). Reproduction, editing, distribution and any kind of use beyond the scope of copyright law require the written consent of the respective author. Downloads and copies of this site are permitted only for private, non-commercial use.
Privacy policy pursuant to GDPR · Art. 13, 14 GDPR
1. Data Controller
The data controller within the meaning of the GDPR is: Prof. Dr. Hannes Helmut Nepper
Pädagogische Hochschule Schwäbisch Gmünd
Oberbettringer Straße 200, 73525 Schwäbisch Gmünd
Email: hannes.nepper@ph-gmuend.de
2. Data Protection Officer
The official data protection officer of Pädagogische Hochschule Schwäbisch Gmünd can be reached at: www.ph-gmuend.de → Data Protection
3. Collection and Processing of Personal Data
This website collects and processes personal data only to the extent technically necessary to provide the site. When you access this website, the following data is automatically stored temporarily in the server log:
IP address (anonymised) · date and time of access
Page accessed · referrer URL · browser and operating system
Legal basis: Art. 6 (1) (f) GDPR (legitimate interest in technical provision of the site). Retention period: 7 days.
4. Embedded Third-Party Services
Twitch (Amazon Web Services, Inc.): this site embeds content from twitch.tv. When the embed code loads, data (IP address, usage behaviour) may be transmitted to Twitch. Privacy policy: twitch.tv/legal/privacy-notice. Legal basis: Art. 6 (1) (f) GDPR.
YouTube (Google LLC): embedded YouTube videos. Data is transmitted to Google upon activation. Privacy policy: policies.google.com/privacy. Legal basis: Art. 6 (1) (a) GDPR (consent through activation).
Google Fonts: fonts are loaded from Google servers, in the course of which the IP address is transmitted. Legal basis: Art. 6 (1) (f) GDPR.
5. Cookies
This website does not set its own cookies. Third-party providers (Twitch, YouTube) may set their own cookies; these are subject to the respective provider's privacy policy.
6. Your Rights (Art. 15–22 GDPR)
Art. 15 GDPR – right of access to stored personal data
Art. 16 GDPR – right to rectification of inaccurate data
Art. 17 GDPR – right to erasure ("right to be forgotten")
Art. 18 GDPR – right to restriction of processing
Art. 20 GDPR – right to data portability
Art. 21 GDPR – right to object
Complaints can be lodged with the competent supervisory authority: the State Commissioner for Data Protection and Freedom of Information Baden-Württemberg, www.baden-wuerttemberg.datenschutz.de.
7. Currency of This Policy
Last updated: June 2026. This privacy policy is updated as needed.
📋 Mission & Concept · CreatorLab
The CreatorLab Manifesto: From Novice to Expert
An educational-theory foundation for gradual competence development in the hybrid makerspace at PH Schwäbisch Gmünd – grounded in the guiding principle of the Baden-Württemberg curriculum.
"To find one's way in a highly technologised world, interdisciplinary patterns of thought and action are necessary. Technical systems must therefore always be considered together with economic, ecological and social contexts. At the centre stands the relationship between human beings, technology and the environment."
— Baden-Württemberg Curriculum · Technology subject, guiding principle for competence development
Mission · People · Technology · Environment
The Triadic Basic Relationship as a Starting Point
The CreatorLab sees itself as a learning space in which the relationship between people, technology and the environment formulated in the Baden-Württemberg curriculum is not only reflected on theoretically but experienced and shaped practically. Making is not an end in itself – it is the epistemic path through which learners grasp through action the interplay of technical, ecological and social systems.
👤
People
Creativity, action and reflection are at the centre of every making process. People are designers, not mere users of technical systems.
⚙️
Technology
Artefacts, systems and processes – from 3D printing to electronic circuits to software – are both the object and the medium of learning.
🌿
Environment
Ecology, society and economics form the framework within which technical decisions must always be reflected on and taken responsibly.
This triad is not a static scheme but a dynamic field of tension: people shape technology, technology shapes people, and both are embedded in an environment which in turn has repercussions for design and use decisions. This mutual interpenetration – formulated in the Baden-Württemberg curriculum as interdisciplinary competence for action – is the normative compass by which every learning setting in the CreatorLab is oriented: from a first soldering exercise to the design of one's own OER tutorial.
"Whoever understands only the technology has understood half of it. Whoever thinks of technology in relation to people and the environment begins to design."
The Learner at the Centre
Learner-Orientation Instead of Content Delivery
The CreatorLab's mission takes its starting point not from the subject matter but from the learning person. Every trainee teacher, every pupil, every workshop participant brings their own prior experience, interests and conceptions (preconceptions) – these are not disruptive factors but the starting point of every learning process. The CreatorLab therefore explicitly understands itself as a learner-oriented space: the choice of project, the pace of work and the depth of engagement lie, as far as possible, in the hands of the learners themselves.
This stance has three immediate consequences for how the learning space is designed. First: differentiation is the norm, not the exception. Since learners with completely different levels of prior knowledge come together – from first contact with a soldering iron to confident mastery of microcontroller programming – offerings must be designed so that they connect at every competence level. Second: mistakes are recognised as a legitimate part of the learning path (see Thesis V, "Mistakes are data") – the psychological safety to be allowed to fail is a prerequisite for exploratory learning. Third: self-efficacy is actively fostered. Anyone who completes their own artefact in the CreatorLab – however small – experiences themselves as capable of action in a technologised world. This experience of self-efficacy is, especially for trainee teachers, a central building block of professional identity: only those who have experienced for themselves what building competence feels like can later credibly guide that process for pupils.
🧩
Taking Prior Knowledge Seriously
Learners' conceptions are not bypassed but used as the starting point for connecting offers – learning builds on what's there rather than replacing it.
🛤️
Individual Learning Paths
Project choice, pace and depth of study are flexible – the CreatorLab offers corridors rather than prescribed paths.
💪
Self-Efficacy
Independently completing artefacts strengthens confidence in one's own capacity to act – reaching beyond the CreatorLab itself.
Concretely, this means for practice in the CreatorLab: every project – whether a bat detector, a 3D-printing workflow or a Twitch tutorial – is examined for which resources it consumes, which societal needs it addresses, and which competences it builds in those involved. In this way, a technical hobby room becomes a place of socially responsible education.
Social Responsibility Through Competence
From Individual Skill to Societal Impact
The CreatorLab never understands the acquisition of technical competences as an end in itself. Anyone who learns to design a circuit, construct a 3D model or produce a learning tutorial simultaneously acquires the ability to recognise the societal significance of technical decisions. This connection between individual competence and collective responsibility is explicitly embedded in the Baden-Württemberg curriculum's people-technology-environment guiding principle: technical education should enable people to participate, informed and capable of judgement, in societal negotiation processes about technology – for example concerning resource consumption, sustainability or digital participation.
Concretely, this responsibility unfolds in the CreatorLab on several levels. At the level of the artefact, every project involves reflecting on where materials come from, how durable a product is, and what happens to it after use – repairability, choice of material and energy consumption are fixed components of project planning, not afterthoughts. At the level of teacher education, students acquire not only subject-specific but also facilitative competence: they learn how to prepare technical matters so that pupils become capable of independent judgement – a key competence for informed participation in a democratic, technologised society. At the level of the public, finally, visibility via Twitch livestream and OER materials makes the knowledge acquired in the CreatorLab available beyond the university – technical education becomes a public good in which people outside PH Schwäbisch Gmünd can also share.
♻️
Responsible Design
Choice of materials, energy consumption and durability of artefacts are considered from the outset – sustainability is a design criterion, not an afterthought.
🧑🏫
Fostering Informed Judgement
Trainee teachers practise preparing technical matters so that pupils become capable of independent judgement and action.
🌍
Knowledge as a Public Good
Livestreams and OER materials carry the competences developed in the CreatorLab beyond the university into society.
Read this way, the mission comes full circle: the individual learning person – with their prior experience, their pace, their motivation – stands at the centre of every learning arrangement. Yet the competences acquired in the CreatorLab are never purely private possessions; they feed back into the relationship between people, technology and the environment: into one's own way of life, into one's later teaching as a teacher, and into societal debates about shaping a technologised world. The CreatorLab is thus at once a place of personal development and a place where social responsibility is practised as lived practice – not as an abstract demand, but as the direct consequence of what one has learned through doing.
Honesty About Our Own Aspirations
Limits and Open Questions
A mission statement that describes only its own strengths would not be an academic document but advertising. The CreatorLab therefore deliberately names three areas of tension:
⚠️
Supervision Amid Heterogeneity
When novices and advanced participants work on dangerous machines at the same time, the supervisory burden increases disproportionately. Differentiation must not come at the expense of workplace safety – an unresolved organisational problem given limited staff resources.
📊
Evidence Base Still Developing
As Martin (2015) notes for the maker movement as a whole, it remains empirically open which learning processes are reliably fostered by makerspaces. The CreatorLab addresses this through its own research, but cannot claim this aspiration as already fulfilled.
⏳
Resource Limits
Individual learning paths and guided support are staff- and time-intensive. The ideal of completely free project choice runs up against the limits of finite supervisory capacity – a structural tension, not a temporary problem.
Competence Progression · Cognitive Learning
From Novice to Expert – a Gradual Learning Path
The CreatorLab follows the model of cognitive apprenticeship (Collins, Brown & Newman, 1989) as well as the Dreyfus model of skill acquisition (Dreyfus & Dreyfus, 1980; Dreyfus & Dreyfus, 1986). Learning does not proceed as linear knowledge acquisition but as a situated, gradual process of transformation – from rule-bound novice status to situationally appropriate expertise. Learners pass through five distinct stages, each requiring specific learning environments, social roles and epistemic practices. Important for understanding this: the Dreyfus model was originally described for development within a single, clearly bounded skill domain, and tended to be conceived linearly and sequentially. The CreatorLab deliberately adapts it for a context with many parallel, changing project domains – the non-linearity described below is thus an explicit extension of the original model, not an implicit departure from it.
1 🔍
Novice
Following rules, observing, imitating
2 🔧
Advanced Beginner
Recognising situations, scaffolding
3 ⚡
Competent
Planning, prioritising, peer teaching
4 🎯
Proficient
Intuition, guiding learning, mentoring
5 🏆
Expert
Holistic, situational, expert teaching
1
Stage 1 · Novice
Rule-Guided Exploration 🔍
The novice enters the CreatorLab as an observer. First encounters with machines, materials and making processes are shaped by rule-orientation and imitation. Learners follow step-by-step instructions in DIY tutorial videos, take part receptively in the Twitch livestream, and experience tools as objects that are both foreign and familiar. The Baden-Württemberg curriculum's guiding principle – the relationship between people, technology and the environment – only gradually fills with meaning.
ImitationTutorialsObservationFollowing rules
2
Stage 2 · Advanced Beginner
Situated Action & Scaffolding 🔧
The first independent production takes centre stage: a simple 3D model, a basic circuit, a short tutorial video. The learner recognises situational features (Brown, Collins & Duguid, 1989) and distinguishes between contexts. Scaffolding by students acting as learning guides enables productive failure in a safe environment. The Discord community offers low-threshold support.
ScaffoldingFirst artefactsPeer supportDiscord
3
Stage 3 · Competent
Purposeful Design & Peer Teaching ⚡
The competent learner plans projects independently, prioritises sub-goals and chooses tools appropriately for the situation. Design decisions in the bat detector or in the 3D-printing workflow are deliberately reflected on. The learner begins to support novices – and discovers that teaching deepens learning. The RANG competence model is lived: reflection, analysis, use and design interlock.
RANGIndependent projectsPeer teachingReflection
4
Stage 4 · Proficient
Intuitive Competence & Guiding Learning 🎯
The proficient maker increasingly acts intuitively and holistically: problems are no longer analysed step by step but recognised and solved situationally. The role shifts – the learner becomes a learning guide in the sense of Rogers (1969) and Schön (1983). Trainee teachers at this stage moderate Twitch livestreams, produce OER videos and co-construct the CreatorLab curriculum. The people-technology-environment relationship is understood as a field of tension that can be actively shaped.
IntuitionGuiding learningTwitch moderationOER production
5
Stage 5 · Expert
Holistic Expertise & Systemic Thinking 🏆
The expert acts holistically, situationally and with context sensitivity – without explicit rule application. In the CreatorLab, expertise means not only technical mastery but the ability to think about and shape technical systems within their economic, ecological and social contexts – precisely in the sense of the Baden-Württemberg curriculum. The expert produces knowledge through research (design-based research), reflects on technology-didactic concepts, and contributes to the further development of the CreatorLab ecosystem. Expert teaching: they convey not rules but embody a stance.
DBRSystemic thinkingExpert teachingInnovation
Importantly: the transition between stages is not linear-chronological but project-dependent and reversible. A person may act at stage 4 in a familiar area (e.g. 3D printing), while experiencing themselves as a novice again in a new field (e.g. microcontroller programming). The CreatorLab deliberately accommodates this simultaneity of different competence levels – it is an expression of a living, heterogeneous learning space.
Integrative Concept
STEAM × RANG × People-Technology-Environment
The CreatorLab integrates three conceptual frameworks into a coherent educational concept: the STEAM perspective opens up disciplinary boundaries, the RANG model describes competence dimensions of basic digital literacy, and the Baden-Württemberg guiding principle grounds everything in the societal mandate of technical education.
🔬
STEAM
Science · Technology · Engineering · Arts · Mathematics. Disciplines are interwoven within the project rather than taught in isolation – a bat detector combines acoustics, electronics, programming, design and mathematics in a single artefact.
🧭
RANG
Reflexion (Reflection) · Analyse (Analysis) · Nutzung (Use) · Gestaltung (Design) (Irion, Peschel & Schmeinck, 2023). These four dimensions of basic digital literacy interlock in the CreatorLab: whoever designs must simultaneously be able to analyse, use and reflect.
🌱
People · Technology · Environment
The guiding principle of the Baden-Württemberg curriculum forms the normative framework within which STEAM and RANG competences are acquired and applied – technical education as a societal mandate.
The three frameworks are not additive but mutually constitutive: STEAM provides the disciplinary and methodological breadth, RANG structures competence development in working with digital tools, and the Baden-Württemberg guiding principle ensures that technical education never remains l'art pour l'art but is always examined for its societal effect. In practice, this integration means that every CreatorLab project is planned along all three axes from the outset.
Core Principles of the Concept
🧱
Constructionism
Learning happens through building publicly shareable artefacts (Papert, 1980) – the object is a medium for thinking, not just an outcome.
📍
Situated Learning
Knowledge arises inseparably from the context of its application (Lave & Wenger, 1991) – not abstractly, prior to practice.
🤝
Community of Practice
Novices, advanced learners and experts learn together in a community of practice with fluid roles.
🎓
Cognitive Apprenticeship
Modelling, coaching, scaffolding and fading accompany the transition from guided to independent action.
🔀
Hybrid Learning Spaces
The physical workshop, the Twitch livestream and the Discord community merge into a continuous learning ecosystem.
📐
Dreyfus Progression
Gradual competence development from novice to expert structures learning paths individually and project-dependently.
🏛️
Third Mission
The CreatorLab connects university research, teacher education and societal practice – explained in detail in Thesis VII, "The CreatorLab is a Third Place", in the "Seven Theses" tab.
Manifesto · Core Theses
Seven Theses of the CreatorLab
The following seven theses condense the CreatorLab's educational-theory foundation into guiding principles for action. They are less rules than stances – points of orientation against which decisions about projects, spaces and roles in the CreatorLab can be measured.
🔨
I. Learning Is Making
Cognitive construction happens through producing publicly shareable artefacts. The object is not an outcome but a medium of thought (Papert, 1980). In the CreatorLab, you think with your hands.
🌐
II. Context Is Curriculum
Knowledge is inseparable from the context in which it arises (Brown, Collins & Duguid, 1989). The Twitch livestream, the Discord community, the workshop machines – all of these are learning material.
🔄
III. Roles Are Fluid
Novices become experts. Experts become learning guides. Trainee teachers are not instructors but more experienced peers in a community of practice (Lave & Wenger, 1991).
🌿
IV. Technology Is Embedded
Technical systems are never considered in isolation, but always within their economic, ecological and social contexts – in the sense of the Baden-Württemberg curriculum and its people-technology-environment guiding principle.
🔬
V. Mistakes Are Data
Failure in the CreatorLab is not a shortcoming but epistemic insight. Troubleshooting an electronic circuit, a failed 3D print – both are productive learning in situated contexts.
📡
VI. Visibility Is Motivation
Publicly shareable artefacts – livestreams, OER videos, 3D objects – generate authentic motivation. The public nature of the Twitch stream is not an add-on but a constitutive principle of the CreatorLab learning format.
🏛️
VII. The CreatorLab Is a Third Place
Beyond school and university, the CreatorLab forms an informal, open learning space (Oldenburg, 1989) that overcomes institutional boundaries. As a third-mission institution of PH Schwäbisch Gmünd, it connects university research, teacher education and societal practice into a coherent ecosystem of lifelong learning.
"The CreatorLab is not a space where you learn technology. It is a space where you learn, through technology, to understand and shape the world."
References & Theoretical Foundations
Sources of the CreatorLab Manifesto
The CreatorLab manifesto draws on established theories of learning and educational research, on international maker-education research, and on current work in technology didactics. The following list documents the central reference points, organised by thematic focus.
Educational-Policy Framework
Ministry of Education, Youth and Sport Baden-Württemberg (2016). Bildungsplan Baden-Württemberg, Fach Technik [Baden-Württemberg curriculum, Technology subject]. — Guiding principle for competence development: people, technology, environment as the subject-specific normative framework. (Note: this is the subject-specific guiding principle of the Technology subject, not identical with one of the six overarching guiding perspectives of the Baden-Württemberg 2016 curriculum.)
Situated Learning & Skill Acquisition
Collins, A., Brown, J. S. & Newman, S. E. (1989). Cognitive Apprenticeship. In Resnick (Ed.), Knowing, Learning, and Instruction (pp. 453–494). LEA.
Dreyfus, S. E. & Dreyfus, H. L. (1980). A Five-Stage Model of the Mental Activities Involved in Directed Skill Acquisition. Unpublished report, University of California, Berkeley. — The original, domain-specific five-stage model
Dreyfus, H. L. & Dreyfus, S. E. (1986). Mind over Machine. Free Press. — Book-length version of the stage model; deliberately adapted in the CreatorLab as a project-dependent, non-linear model rather than a rigid sequence
Lave, J. & Wenger, E. (1991). Situated Learning: Legitimate Peripheral Participation. Cambridge University Press.
Brown, J. S., Collins, A. & Duguid, P. (1989). Situated Cognition and the Culture of Learning. Educational Researcher, 18(1), 32–42.
Irion, T., Peschel, M. & Schmeinck, D. (2023). Grundlegende Bildung in der Digitalität [Foundational education in digitality]. Was müssen Kinder heute angesichts des digitalen Wandels lernen? In Grundschulverband (Ed.), Grundschule und Digitalität (pp. 18–42). Grundschulverband. — Source of the RANG dimensions (Reflexion/Reflection, Analyse/Analysis, Nutzung/Use, Gestaltung/Design); originally developed for primary school, adapted in the CreatorLab for higher education
Papert, S. (1980). Mindstorms: Children, Computers, and Powerful Ideas. Basic Books.
Maker Education & Digital Fabrication
Dougherty, D. (2012). The Maker Movement. Innovations: Technology, Governance, Globalization, 7(3), 11–14. — A programmatic founding text of the maker movement
Halverson, E. R. & Sheridan, K. (2014). The Maker Movement in Education. Harvard Educational Review, 84(4), 495–504. — Defines the maker movement as a shared physical-digital production culture
Sheridan, K. M., Halverson, E. R., Litts, B. K., Brahms, L., Jacobs-Priebe, L. & Owens, T. (2014). Learning in the Making: A Comparative Case Study of Three Makerspaces. Harvard Educational Review, 84(4), 505–531.
Blikstein, P. (2013). Digital Fabrication and 'Making' in Education: The Democratization of Invention. In Walter-Herrmann & Büching (Eds.), FabLabs: Of Machines, Makers and Inventors (pp. 1–21). transcript. — A bridge between constructionism and FabLab/makerspace practice
Martinez, S. L. & Stager, G. (2013). Invent to Learn: Making, Tinkering, and Engineering in the Classroom. Constructing Modern Knowledge Press.
Martin, L. (2015). The Promise of the Maker Movement for Education. Journal of Pre-College Engineering Education Research (J-PEER), 5(1), Art. 4. — Openly notes that the empirical evidence base of the maker movement is still developing
Ito, M., Gutiérrez, K., Livingstone, S., Penuel, B., Rhodes, J., Salen, K., Schor, J., Sefton-Green, J. & Watkins, S. C. (2013). Connected Learning: An Agenda for Research and Design. Digital Media and Learning Research Hub. — Interest-driven, production-centred learning across connected platforms (livestream, Discord, OER)
Learning Space & Guiding Learning
Rogers, C. R. (1969). Freedom to Learn. Charles E. Merrill.
Schön, D. A. (1983). The Reflective Practitioner: How Professionals Think in Action. Basic Books.
Oldenburg, R. (1989). The Great Good Place. Paragon House. — The concept of the third place, originally for physical spaces; extended in the CreatorLab to a hybrid space
🌤️
Weather & Excursion Tips
Live weather for Schwäbisch Gmünd · 7-day forecast · hourly view · excursion tips & a maker project
–°
–
Feels like –°C
🦄
Schwäbisch Gmünd · "Gold City" in the Ostalbkreis
~1162
Germany's oldest Hohenstaufen town
63,000
Residents · district seat
320 m
Elevation above sea level · Remstal
🦄 1277
The city's first unicorn coat of arms
Schwäbisch Gmünd is Europe's jewellery capital – for centuries a centre of goldsmithing and the gemstone trade. The Gothic Holy Cross Minster (1351–1521), designed by Heinrich Parler, is regarded as one of the most important hall churches in southern Germany. The Stadtgarten on the 2014 Landesgartenschau grounds and the three Hohenstaufen hills — Hohenstaufen, Rechberg and Stuifen — shape the townscape.
The University of Education with its CreatorLab brings together maker education, technology didactics and research in one place. The Ostalbkreis offers unique natural areas: the Swabian Jura, the Welzheim Forest, the Härtsfeld plateau and the Nördlingen Ries (a UNESCO Global Geopark).
🦄 The unicorn coat of arms: since a seal impression from 1277, Schwäbisch Gmünd has borne a rearing silver (white) unicorn on a red field – one of the oldest municipal coats of arms in Baden-Württemberg, celebrated in 2025 as "750 years of the unicorn". Why a unicorn was chosen specifically is not clearly documented; one interpretation points to the Hohenstaufens, another to the founding legend of St John's Church. As a Free Imperial City, Gmünd's guilds also used the unicorn as a hallmark and quality seal for export goods. This same heraldic animal also marks the CreatorLab as 🦄 on the mountain-biking and hiking maps.
⏳ Loading weather data…
–%
💧 Humidity
–
🌬️ km/h wind
–
🔵 hPa pressure
–
👁️ km visibility
–
☀️ UV index
–
☔ Rain chance
–
🌫️ Air quality
–
Direction
UV Index
LowModerateHighVery HighExtreme
–
Today's Sun Path
Temperature Trend – Next 24 Hours
7-Day Forecast
Excursion Tips
🏰
Hohenstaufen
A hike up the historic Hohenstaufen hill (684 m) – once the site of the ancestral castle of the Hohenstaufen imperial dynasty. A grand panoramic view over the Hohenstaufen hills as far as the Alps. An easy climb (45 min), also well suited for families.
Seasonal (late November until shortly before Christmas): a historic market square around the Holy Cross Minster, mulled wine, arts and crafts – one of the region's most beautiful Christmas markets.
A popular sledging slope at the foot of the Hohenstaufen when it snows – open gradient, easily accessible, suitable for families. Bring your own sledge.
A snow-covered Hohenstaufen-hills backdrop with clear long-distance views – often empty and especially peaceful in winter. Sturdy, non-slip shoes recommended.
Arduino Weather Station with OLED Display & Pressure Trend
A fully functional weather station that measures temperature, humidity and air pressure, displays them on an OLED screen, and calculates a simple weather forecast from the pressure trend. Ideal as project work for Years 7–9 or as a CreatorLab seminar project. All materials are available in the CreatorLab.
Build the circuit: DHT22 on pin D2, BMP280 via I2C (SDA=A4, SCL=A5), OLED also on the I2C bus (address 0x3C). Connect all components on the breadboard and use wire colours systematically (red=5V, black=GND, yellow=data).
2
Install the libraries: in the Arduino IDE via the Library Manager: "Adafruit DHT sensor library" (+ Adafruit Unified Sensor), "Adafruit BMP280 Library", "Adafruit SSD1306" and "Adafruit GFX Library". Confirm all dependencies.
3
Read the sensors: the DHT22 provides temperature (°C) and relative humidity (%), the BMP280 provides absolute air pressure (hPa) and optionally altitude (m above sea level). Open the serial monitor and check the values – typical errors: wrong I2C address, missing pull-up resistors.
4
Program the OLED display: initialise the Adafruit_SSD1306 library (128×64 pixels). Output three lines: temperature, humidity, air pressure. display.clearDisplay() at the start of each update cycle, display.display() at the end. Font size 1–2 depending on the information.
5
Calculate the pressure trend: every 10 min, write the current pressure into an array (ring buffer, 6 values = 1 hour). If the average falls >4 hPa/3h → rain likely → umbrella icon on the OLED. If it rises → fine-weather icon. Move this logic into its own function.
6
3D-print the enclosure: use the FDM printer in the CreatorLab: an outer housing with ventilation slots for the sensors (to ensure air circulation), a viewing window for the OLED, an opening for the USB cable. PETG filament recommended (more heat-resistant than PLA). Print time approx. 3 h.
7
Extension (optional): use an ESP8266 instead of an Arduino Uno → Wi-Fi connectivity, send data to an MQTT broker or Thingspeak, log historical data to an SD card, visualise it in a web dashboard with Chart.js. Ideal for a master's project seminar.
☀️
CreatorLab Project
UV Index Traffic Light with Sun-Protection Warning System
A compact traffic light that measures current UV radiation and uses an RGB LED to show whether and what kind of sun protection is needed. Raises awareness of skin protection and makes invisible radiation visible – a rewarding introductory project for Years 5–7.
Arduino / ESP32GUVA-S12SD UV sensorRGB LED (WS2812)0.96" OLED I2C3D-printed enclosure with diffuserDuration: 2–3 h
1
Connect the sensor: the GUVA-S12SD provides an analogue voltage proportional to UV intensity. Signal pin to an analogue input (e.g. A0), supply 3.3–5 V depending on the module. Orient the sensor as unobstructed as possible facing upward (no housing in the light path).
2
Convert the voltage to a UV index: first convert the analogue value (0–1023) to volts, then translate it into the UV index (0–11+) using a linear approximation formula from the datasheet. Output the values over the serial interface and compare them with an online UV index service for calibration.
3
Program the traffic-light logic: divide the UV index into three to five levels (e.g. 0–2 green, 3–5 yellow, 6–7 orange, 8+ red). Drive the WS2812 LED accordingly (FastLED or Adafruit_NeoPixel library). Fade transitions smoothly instead of switching abruptly.
4
Add an OLED display: show the current UV index as a number, plus a plain-text recommendation ("apply sunscreen", "avoid midday sun"). Store recommendation texts as an array and select the one matching the traffic-light level.
5
Print the enclosure with a diffuser: a 3D-printed enclosure with a translucent diffuser element over the UV sensor (e.g. a thin-walled section or an inset piece of natural filament) for more even light coupling. A sturdy base for outdoor use in the schoolyard.
6
Classroom connection: log the UV index over the course of the day and correlate it with cloud cover, season and time of day. Discussion: why does UV radiation matter even under a cloudy sky? Link to skin-cancer prevention and the ozone layer.
🌧️
CreatorLab Project
Precipitation Gauge with Tipping-Bucket Sensor
A digital rain gauge based on the tipping-bucket principle, which counts every precipitation pulse and uses it to calculate the amount of rainfall in mm. Provides genuine local precipitation data over weeks – a classic of citizen-science weather observation.
Arduino / ESP32Tipping-bucket sensorReed switchmicroSD module3D-printed funnel & tipping bucketDuration: 5–7 h
1
Build the mechanism: a funnel collects rainwater and channels it onto a small, balanced tipping bucket. Once one side fills, it tips, empties, and the other side fills – each tip corresponds to a defined amount of water (e.g. 0.2 mm).
2
Install the reed switch: attach a small magnet to the tipping-bucket axle, and fix a reed switch beneath it. Every tip triggers a brief switching pulse, which is counted via interrupt on the microcontroller – no polling needed, saving energy.
3
Calibrate: slowly pour in a defined amount of water (e.g. 100 ml) and count the number of tips. Use this to calculate the mm value per pulse and store it as a constant in the code. Confirm calibration with two to three runs.
4
Set up data logging: write every pulse with a timestamp to a microSD card (CSV: date, time, pulse count). Calculate hourly and daily totals and store them separately – the basis for later analysis in Excel or Python.
5
3D-printed enclosure: a large-diameter funnel (for accurate values), a tipping-bucket chamber with PETG bearing mounts, a base with mounting holes. Design the water drainage so that no back-up occurs.
6
Classroom connection: compare your own daily totals with nearby German Weather Service (DWD) station data – how much do microlocations differ? Use a school year's worth of precipitation data as the basis for a small climate analysis.
💨
CreatorLab Project
Wind Measuring Station: Anemometer & Wind Direction
A mechanical wind measuring station with a rotating cup anemometer for wind speed and a wind vane with a magnetic sensor for wind direction. A more demanding project with a large 3D-printing and mechanical component.
Arduino / ESP32Hall sensorMagnetoresistive sensor (direction)Ball bearings3D-printed cup rotor & wind vaneDuration: 8–10 h
1
Print & mount the cup rotor: 3D-print three or four half-cups with arms and mount them on a rotating axle with a ball bearing. The cups make the rotor turn proportionally to wind speed.
2
Measure rotational speed with a Hall sensor: attach a small magnet to the rotating axle, and place a stationary Hall sensor beneath it. Each rotation generates a pulse; count these over a fixed time window (e.g. 5 s) and convert to rotations per second.
3
Calibrate the wind speed: convert rotation count to m/s – ideally through a comparison measurement with a reference anemometer, or via known circumferential speed at a defined drive rotation (e.g. calibrating with a cordless drill).
4
Capture wind direction: mount the wind vane on a second axle, with several magnets positioned at different points on its underside. Arrange magnetoresistive or Hall sensors in 8 compass directions and determine which sensor is active.
5
Data output & logging: display wind speed (m/s, maximum gust) and direction (8-point wind rose) on the OLED, and log via microSD or Wi-Fi (ESP32). Calculate a 10-minute average as the standard value used in weather reports.
6
Classroom connection: build a wind rose over several days and compare it with the prevailing wind direction at your location. Link to wind energy: how much energy is in the measured wind (P ∝ v³)? Discuss site selection for wind turbines.
🔋
CreatorLab Project
Solar-Powered Mini Climate Station with Data Logger
A self-sufficient, solar-powered climate station for outdoor use over weeks or months – with no cable connection. Combines the sensor setup of the weather station with power management, deep-sleep mode and long-term data logging.
Set up the power supply: connect the solar panel (5–6 V) to a LiPo battery (3.7 V) via a TP4056 charging module. Power the ESP32 from the battery. Connect a voltage divider for battery-voltage measurement to an analogue input to monitor the state of charge.
2
Connect the sensors as in Project 1: wire up the DHT22 and BMP280 as in the basic weather station. Also add a light sensor (LDR or BH1750) to roughly estimate sunshine duration.
3
Program deep-sleep mode: put the ESP32 into deep sleep between measurements (e.g. wake every 15 minutes, measure, log, sleep again). This drastically reduces power consumption – essential for weeks of maintenance-free solar operation.
4
Log data to microSD: save each measurement with a timestamp (e.g. via NTP with a brief Wi-Fi connection, or an internal RTC) as a CSV line. Keep an eye on file size – for hourly measurements, one card lasts for years.
5
Print a weatherproof enclosure: a housing with radiation shielding (louvres or a double wall, as in a Stevenson screen) for the sensors, so direct sunlight does not distort the temperature reading. Mount the solar panel separately on top, ideally facing south.
6
Classroom connection: choose a location with little infrastructure (e.g. a school garden, the CreatorLab outdoor area) and collect data over a school half-year. Discuss the energy balance: how many days does the battery last without sun? Link to self-sufficient sensor networks in remote regions.
🌍 Climate Education · From Sensor to Climate Perspective
🌡️
Education for Sustainable Development · ESD
Five Sensors, One Question: How Is Our Climate Changing?
The five maker projects in this section are more than individual craft projects – together they form a small sensor network of your own. Anyone who collects their own data on temperature, UV radiation, precipitation, wind and air pressure intuitively understands the difference between weather (day-to-day, local) and climate (long-term average, global) – the basis for every societal climate debate.
Live Comparison · Weather Today vs. Climate Average (2015–2024)
⏳ Loading climate data…
Source: Open-Meteo Historical Weather API · 10-year average for this calendar date, location Schwäbisch Gmünd
🌡️ Microclimate & Urban Climate
Run the weather station (Project 1) and the solar climate station (Project 5) in parallel at different locations: schoolyard vs. Stadtgarten vs. forest edge. Where is it 3°C warmer, and why? → Make the urban heat island tangible with your own data.
📊 Data Literacy
Analyse the CSV logs of all five devices with Python (pandas) or Excel: averages, extreme values, trend lines, correlations between air pressure, precipitation and wind. Visualise with Chart.js or Matplotlib – climate change becomes readable in numbers, directly on-site in Schwäbisch Gmünd.
☀️ Radiation, Wind & Energy
The UV traffic light (Project 2) and the wind measuring station (Project 4) supply data on solar radiation and wind speed. Make renewable energy tangible: how many hours of sunshine, how much wind energy is actually available over the year at your own location?
🌱 Action Orientation
The precipitation gauge (Project 3) provides real rainfall data for the school garden – from measurement to impact: learners present their data to the town council or on the school website. Citizen science – feeding your own readings into the DWD network (Netatmo API, OpenWeatherMap Community). Technology as societal participation.
💡
Curricular connection: the ESD guiding perspective (B) in the Baden-Württemberg curriculum – competence area "Understanding and shaping the world". The five projects connect technology (sensors, programming, power supply), science (atmosphere, thermodynamics, radiation) and society (climate justice, political action) in a coherent maker curriculum – from a beginner project to a self-sufficient long-term station.
🚵
Mountain Bike Trails
Trails & routes in the Ostalbkreis – from family-friendly forest tracks to technical singletrack + a CreatorLab maker project
The Ostalbkreis offers an exceptionally varied mountain-biking landscape: from the gentle Remstal to the technical trails of the Härtsfeld plateau to the flowing forest tracks of the Welzheim Forest, there is something for every riding style. The region is topographically varied – elevation differences from 200 to over 1,000 metres per tour, natural forest tracks and historic Hohenstaufen hills as a backdrop.
⚠️ Legal notice: the "2-metre rule" in Baden-Württemberg
Baden-Württemberg is the only German state where cycling in forests is, in principle, only permitted on paths at least 2 m wide under § 37 (3) of the state forestry act – violations are a regulatory offence. Narrower paths (singletrack) may be affected; the exact path width is often not obvious on-site. This particularly concerns teachers and activity leaders travelling with school or club groups. Please inform yourself independently before every tour about the current status of the relevant paths (e.g. with the responsible forestry authority) – this page cannot definitively clarify this for every individual route. In addition, note wildlife refuge zones and temporary closures during hunting or breeding/nesting seasons, which are signposted.
A classic loop around Gmünd over the three Hohenstaufen hills — Hohenstaufen (684 m), Rechberg — and the Welzheim Forest. A combination of wide forest tracks and flowing trail sections. Well signposted, several places to stop for refreshments. Especially beautiful in autumn with the changing leaves.
The most technically demanding tour in the Ostalbkreis – rocky sections across the Härtsfeld plateau, tight bends through juniper heathland, short technical drops. The Härtsfeld (650–700 m) offers magnificent views across the Ries. For experienced riders with technical skill only. Good equipment mandatory (helmet, protective gear).
A family-friendly loop around Lorch with a wonderful view of the Romanesque Lorch Abbey (12th century). Mostly wide forest tracks with a few lovely trail sections. Lorch itself is worth a short visit – you can stop for refreshments at the abbey inn. Ideal as a first mountain-biking adventure.
From the Albuch near Dischingen via Heidenheim towards the Nördlingen Ries – wonderful meadow trails along the edge of the Jura, panoramic views over the Ries and the foothills. Bike transport is partly available. Good for bikepacking beginners – moderate terrain, a pleasant mix of plateau and escarpment.
Right on the western edge of the Ostalbkreis: fast, flowing forest trails through the dense Welzheim Forest. Few technical obstacles, but plenty of speed and a lovely forest atmosphere. Popular with families and casual riders. The forest offers pleasantly cool air even on hot days.
Shady sections in the Welzheim Forest provide cool relief; on the Jura escarpment there's usually a pleasant breeze.
🍂 Autumn
Stauferland Trail (1), Härtsfeld Singletrack (2)
Dry, grippy ground and autumn foliage make the Hohenstaufen hills and the Härtsfeld plateau especially appealing.
❄️ Winter
Stauferland Trail (1, only in dry conditions & daylight)
Natural trails suffer badly from frost/wet conditions; the forest tracks on the Stauferland Trail are more robust, but take care with snow/ice. If there's snow, switch to the winter excursion tips in the 🌤️ Weather popup instead (Christmas market, ice rink, winter hike).
🔧 CreatorLab Maker Projects · Mountain Biking
📷
CreatorLab Project
Automatic Trail Camera with Motion Detection & IP65 Enclosure
A weatherproof camera that automatically takes photos or videos when it detects motion – ideal for wildlife observation or as a trail camera on mountain-bike routes. Raspberry Pi Zero 2W at its core, a PIR sensor for motion detection, a weatherproof 3D-printed enclosure (IP65) for outdoor use. Perfect as project work for Years 8–10 or as a seminar project.
Raspberry Pi Zero 2WPi Camera v2 (8 MP)PIR sensor HC-SR501LiPo battery 3,000 mAhmicroSD 64 GB3D-printed PETG (IP65)Duration: 6–8 h
1
Set up the Raspberry Pi: flash Raspberry Pi OS Lite (64-bit) to microSD using Raspberry Pi Imager. Enable SSH (create an ssh file in the boot folder), enter Wi-Fi credentials in wpa_supplicant.conf. Enable the camera interface in raspi-config: "Interface Options → Camera → Enable".
2
Connect the PIR sensor: HC-SR501: VCC to 5V (pin 2), GND to GND (pin 6), OUT to GPIO17 (pin 11). Set sensitivity (left potentiometer: 3–7 m range) and delay time (right potentiometer: 5–300 s) with a screwdriver. Test: read the GPIO output using Python.
3
Python recording script: when the PIR is triggered → the camera starts a 10-second H264 video (picamera2 library). Filename with a timestamp (datetime.now().strftime). Store files in /home/pi/clips/. Set up the script as a systemd service so it starts automatically at boot.
4
Power-saving mode: throttle CPU clock speed between recordings (cpufreq-set -g powersave), disable Wi-Fi and HDMI (script at startup). Standby consumption: ~0.1 W → battery life 48–72 h. Remove the LED on the PIR sensor or cover it with tape (camouflage in the field).
5
Build the IP65 enclosure in the CreatorLab: print an outer housing from PETG (weatherproof to -20°C) on the FDM printer: a camera opening with acrylic glass (3 mm, laser cutter), cable feed-throughs sealed with rubber grommets. Print a mounting bracket for a tree strap as well. Total print time: approx. 4 h.
6
Extension – AI classification: install TensorFlow Lite on the Pi, load a pretrained MobileNet model for animal/human/bicycle classification. Automatically label each video clip and save only relevant clips → significantly reduced storage requirements. Ideal for project work with an AI focus.
🚨
CreatorLab Project
Crash & Emergency Tracker with Accelerometer
A compact device on the handlebar or helmet that detects a crash based on sudden changes in acceleration and orientation and raises an alarm via LED/buzzer – with optional GPS location reporting. Raises awareness of trail safety and is an instructive project on sensors and threshold algorithms.
Connect the MPU6050: connect the MPU6050 to the ESP32 via I2C (SDA/SCL), supply via 3.3 V. Install a library (e.g. Adafruit_MPU6050) and check raw acceleration (g) and rotation rate (°/s) data on all three axes via the serial monitor.
2
Record a normal ride: during a test ride, continuously log total acceleration (the vector magnitude of x/y/z). Note typical values during normal riding, braking and small jumps – this data forms the basis for sensible thresholds.
3
Crash-detection logic: a crash usually manifests as (a) a very high, brief acceleration spike (impact) followed by (b) an unusual, sustained tilt (the bike lying on the ground). Combine both conditions to avoid false alarms from ordinary potholes.
4
Program the alarm chain: when a crash is detected: trigger the buzzer and LED alarm immediately, then a 15-second wait with a cancel button ("I'm okay"). If there is no confirmation, capture the GPS position and prepare it for later transmission (e.g. SMS via SIM800L or Wi-Fi notification).
5
Integrate the GPS module: connect the NEO-6M via UART, use the TinyGPS++ library to parse latitude and longitude. Format the position as a Google Maps link (https://maps.google.com/?q=lat,lng) so it can be shared directly.
6
Print the enclosure & mount: 3D-print a compact, shock-resistant enclosure with a rubber seal for handlebar or helmet mounting. Keep the centre of gravity low so the device doesn't distort the sensor readings. Classroom discussion: the limits of automatic crash detection and data privacy for location data.
🛞
CreatorLab Project
Tyre-Pressure Early-Warning System (TPMS) for Mountain Bikes
A simple tyre-pressure system that monitors air pressure in the front and rear tyres and warns when pressure is too low – important for trail safety (pinch flats, tyre damage) and efficiency. Combines pressure sensing with wireless transmission between two small transmitter units and a handlebar display.
2× ESP32-C3 (transmitter)1× ESP32 (receiver/display)2× pressure sensor (0–6 bar)0.96" OLED I2CCoin-cell batteries3D-printed valve adapters & mountsDuration: 7–9 h
1
Adapt the pressure sensor to the valve: connect a small pressure sensor (analogue or I2C, depending on the model available) to the Schrader valve via a 3D-printed adapter. Use an O-ring seal to prevent air loss – a pressure test before use is mandatory.
2
Write the transmitter firmware: the ESP32-C3 reads the pressure sensor at regular intervals (e.g. every 30 s), converts the signal to bar/PSI and sends the value via ESP-NOW (connectionless, very energy-efficient) to the receiver unit. Include a device ID to distinguish front and rear wheels.
3
Receiver & display: the receiver ESP32 accepts both transmitter packets and shows front and rear tyre pressure simultaneously on the OLED. Store a target pressure range (e.g. 1.8–2.2 bar) and highlight it with colour or a warning symbol if it is exceeded or undershot.
4
Power and enclosure concept: keep the transmitter units as small and light as possible (coin cells, deep sleep between measurements). Print 3D, vibration-resistant enclosures for the transmitters on the rim/spoke, and a robust handlebar mount for the display.
5
Calibration & testing: compare sensor readings against a mechanical floor pump with a pressure gauge, correct the offset in the code. Carry out a test ride with deliberately slightly low pressure to check the warning function.
6
Classroom connection: discussion of tyre pressure, rolling resistance and puncture risk – linked to physics (pressure, volume, temperature dependence per Gay-Lussac's law). Use data series across several rides as an introduction to sensor-data analysis.
A bike-mounted data logger that records gradient angle, vibration and speed along a route and automatically creates a difficulty profile from it – e.g. for a CreatorLab-made trail map or to document new sections of trail.
Set up the sensors: MPU6050 for gradient (pitch/roll) and vibration (acceleration variance), a GPS module for position and speed, all driven via the ESP32. Mount the components centrally on the frame (e.g. down tube) to minimise flex.
2
Define the data structure: write one CSV line per second: timestamp, GPS position, speed, gradient angle, vibration metric (e.g. the standard deviation of acceleration over 1 s). This structure forms the basis for later analysis.
3
Recording & logging: write data continuously to microSD (one file per tour, filename with date/time). Make sure the card has sufficient write speed (Class 10 or better) to avoid data loss at high recording rates.
4
Calculate a difficulty index: after the tour, divide the route into short segments (e.g. 20 m), calculate average gradient and vibration metric per segment, and combine them into a difficulty index (e.g. a weighted sum). Translate the index into categories (easy/medium/hard).
5
Visualisation: display the GPX track with the difficulty index per segment colour-coded in a mapping app (e.g. with Python/Folium or QGIS) – green/yellow/red along the route. Use the result as the basis for a CreatorLab trail map.
6
Classroom connection: comparing subjective assessment ("felt hard") with objective measurements. Discussion: what factors make a trail "hard" – gradient, surface, corner radius? Data-driven vs. experience-based trail rating.
☀️
CreatorLab Project
Solar Charging Station for Trail Rest Stops
A robust, freestanding solar station for rest stops along mountain-bike routes that charges smartphones, GPS devices or lights via USB – useful for long tours and a contribution to trailside infrastructure. Combines power management with outdoor-rated enclosure building.
Solar panel 10–20 WCharge controller (MPPT/PWM)LiFePO4 battery 12 VUSB power-bank module (5V/12V)Weatherproof 3D-printed/wood enclosureDuration: 8–10 h
1
Plan the power circuit: solar panel → charge controller → battery → loads (USB outputs). The charge controller prevents overcharging and deep discharge of the battery. Calculate sizing together: how much energy does the panel deliver on an average day, how many charging cycles should be possible?
2
Wire the components: solar panel to the charge-controller input, battery to the controller's battery output, USB power-bank module (with its own deep-discharge protection) to the load output. Plan fuses between the main components, choose cable cross-sections appropriate to the current load.
3
Measurement & display unit: add a small display or LED indicator for battery level and current solar output (e.g. with a simple voltage/current measurement module). Helps users see whether the station is currently charging or the battery is running low.
4
Build the outdoor enclosure: build a weatherproof enclosure from 3D-printed PETG or combined with laser-cut plywood: mount the solar panel firmly, tilted to face south, protect USB sockets from rain with a hinged lid/seal, and house the battery well ventilated but protected.
5
Site selection & installation: together with forestry or trail authorities, choose a suitable, sun-exposed location at an existing rest stop. Mount the station securely against theft and stably (e.g. on a post), and obtain permission.
6
Classroom connection: collect real measurement data on hours of sunshine and charging output over several weeks and compare it with the calculated energy demand. Discuss decentralised, renewable energy supply in public space and maker contributions to local infrastructure.
The Trail as a Measuring Course – Mountain Biking with a Researcher's Eye
The five maker projects in this section turn the mountain bike from a pure sports device into a mobile measurement system. Forests in the Ostalbkreis are under pressure from heat, drought and bark beetles – the trail camera, on-bike sensors and solar technology make this change visible, measurable and open to discussion. A citizen-science approach: learners become researchers in their own environment.
📷 Observing the Forest Over Time
The trail camera (Project 1) over several months at the same location in the Welzheim Forest or the Hohenstaufen woodlands. A time-lapse series documents leaf emergence, dry periods and crown damage. Measure phenology yourself – is leaf emergence shifting compared to historical reference values?
🦊 Biodiversity & Habitat
Which other animal species use the trail? Use the trail camera to map wildlife corridors and analyse activity patterns (day/night, season). Understand biodiversity loss as a local phenomenon – link camera data with iNaturalist observations.
🚵 Trail Profiles & Use Conflicts
The trail difficulty logger (Project 4) shows how gradient and surface affect ground stress – combined with trail-camera data on usage times, this produces a fact-based foundation for path management concepts, rather than opinion against opinion in nature-conservation disputes.
🔋 Energy & Material Cycles on the Trail
The solar charging station (Project 5) makes energy generation in the forest tangible; the crash tracker (Project 2) and tyre-pressure TPMS (Project 3) raise awareness of resource conservation – fewer tyre failures, less puncture-repair waste, safer tours. Discussion: which technology is really needed on the trail, and which is just comfort?
💡
Curricular connection: the ESD guiding perspective (B), the interdisciplinary subject Human–Nature–Culture (MNK), and Technology Years 7–10. The five projects combine ecology (forestry, biodiversity), technology (embedded systems, sensors, power supply) and society (nature-conservation law, climate adaptation, road/trail safety) into a holistic maker curriculum built around trail and forest.
🧗
Hiking Routes in the Ostalbkreis
The most beautiful trails from cosy valleys to the Jura escarpment + a CreatorLab maker project for GPS-loving hikers
The Ostalbkreis is one of the most hiking-rich regions in Baden-Württemberg. Three natural areas overlap: the Swabian Jura with its striking escarpment edges, the Welzheim Forest with dense beech woods and quiet streams, and the Nördlingen Ries – a UNESCO Global Geopark in a 25 km-wide meteorite crater. Add to that the historically significant Hohenstaufen hills around Schwäbisch Gmünd as rewarding nearby destinations.
Filter Routes by Suitability
5 Top Hiking Routes in the Ostalbkreis
1
Three Emperors' Hills Trail
📏 22 km↑ 550 m elevation⏱ 6 h🟡 Medium
Crossing all three Hohenstaufen cones — Hohenstaufen (684 m), Rechberg (707 m) and Stuifen (757 m) — one of the most beautiful panoramic circular routes in the Ostalbkreis. In good weather, views reach as far as the Alps. Historically: the ancestral hills of the Hohenstaufen imperial dynasty. Refreshment options on all three summits (seasonal). Very well signposted.
A relaxed wine-country hike through the Remstal between Gmünd and Lorch – through orchards, vineyards and idyllic villages. Especially attractive in autumn: seasonal wine taverns (Besenwirtschaften) along the way invite you to stop. Family-friendly, pushchair-suitable on long stretches. Lorch Abbey (12th century) is a cultural highlight at the end of the tour.
Along the dramatic Jura escarpment with views across the entire foothills. Juniper heathland, bizarre rock formations (Rosenstein, 684 m) and medieval castle ruins shape the scenery. Especially recommended in spring (orchid bloom) and autumn. The Rosenstein offers one of the best views in the whole Swabian Jura.
A relaxed forest walk through the natural Welzheim Forest to the historic Teufelsmühle (Devil's Mill). Several streams to cross, climbing rocks for children, shady forest tracks. Good to walk all year round. Perfect for families with younger children or as an after-work outing from the PH. The Teufelsmühle (18th century) is a restored mill building with a small café.
On the eastern edge of the Ostalbkreis: a hike along the rim of the Nördlingen Ries – one of the best-preserved meteorite craters on Earth (a UNESCO Global Geopark, 15 million years old, 25 km in diameter). The Wallerstein viewing tower (383 m) offers a 360° panorama over the Ries. Geo-information boards along the way explain how it formed.
No route currently matches this selection – try a different filter.
📅 Seasonal Recommendations
Season
Recommended Route(s)
Why
🌱 Spring
Jura Escarpment Trail (3), Remstal Wine Trail (2)
Orchid bloom on the juniper heathland, blossoming fruit trees in the Remstal – both routes show the landscape awakening.
☀️ Summer
Teufelsmühle (4), Ries Rim Panoramic Trail (5)
Shady forest tracks in the Welzheim Forest offer cool relief; clear long-distance visibility on the Ries rim gives the best views.
🍂 Autumn
Remstal Wine Trail (2), Three Emperors' Hills Trail (1)
Seasonal wine taverns in the Remstal, autumn foliage and especially clear long-distance views from the Hohenstaufen hills.
❄️ Winter
Three Emperors' Hills Trail (1, with crampons if snowy), Teufelsmühle (4)
The Hohenstaufen hills offer long-distance views to the Alps in clear air; the Teufelsmühle loop is walkable all year round.
🔧 CreatorLab Maker Projects · Hiking
🗺️
CreatorLab Project
GPS Hiking Tracker with E-Paper Display & GPX Export
A portable GPS tracker that displays position, altitude, speed and distance covered on a power-saving e-paper display, saves all waypoints as a GPX file on microSD, and optionally measures air pressure and temperature. Compatible with Komoot, AllTrails and OpenStreetMap. Perfect for Years 8–10 and as a seminar project at the PH.
Set up the TTGO T-Beam: in the Arduino IDE, add the ESP32 boards URL (Espressif). Select the "TTGO T-Beam" board, install the driver (CP2102). Libraries via the Library Manager: TinyGPS++, GxEPD2 (e-paper), Adafruit BME280, SPI, Wire. Upload an initial sketch to test the serial GPS output.
2
Read and process GPS data: TinyGPS++ parses NMEA strings from the integrated Ublox NEO-6M module. Read: gps.location (lat/lng), gps.altitude (m), gps.speed (km/h), gps.satellites (count). Show the satellite count on the display – a fix is only reliable from ≥4 satellites. Fix wait time outdoors: 30–90 seconds.
3
Drive the e-paper display: initialise the GxEPD2 library (SPI connection). A dashboard with four fields: position (coordinates), altitude, speed, distance. Update every 30 s – e-paper retains the image without power, hence ideal energy consumption. Use partial updates for fields (faster, no full-screen flicker).
4
Log the GPX track to microSD: include the SD library, open a GPX file with an ISO 8601 timestamp. Write a waypoint (trkpt) every 10 s: lat, lon, ele, time. Open and close the GPX file correctly (trkType, trkseg). The finished GPX file is directly compatible with Komoot, AllTrails, OSM and Google My Maps.
5
Integrate weather sensing: connect the BME280 via I2C (address 0x76 or 0x77). Measure temperature, humidity and absolute air pressure. Show a pressure-trend arrow on the display: compare with the value from 3 h ago → a pressure drop >4 hPa → rain warning. Alternatively, calculate altitude from air pressure (the barometric altitude formula).
6
3D-print an ergonomic enclosure: print a wrist mount with a quick-release system in the CreatorLab from PETG (weatherproof). A separate compartment for the 18650 battery with a charging board (TP4056). A push-button for display updates. The overall housing is splash-proof (IP54) through a tight fit. Print time approx. 3.5 h.
🔍
CreatorLab Project
Digital Scavenger Hunt: Geocaching Boxes with NFC/QR
A series of 3D-printed, weatherproof cache boxes with NFC tags or QR codes, hidden along a hiking route. Scanning one opens a digital puzzle, extra information or an OER learning resource – ideal as a class or CreatorLab project connecting hiking and digital learning.
NFC tags (NTAG215)Smartphone (scan app)Weatherproof 3D-printed boxesStatic webpage / QR generatorLogbook (paper)Duration: 4–6 h (+ route planning)
1
Define the concept & route: together, choose an existing hiking route (e.g. from the CreatorLab hiking catalogue) and identify 4–6 sensible stations – at distinctive points such as viewpoints, information boards or junctions. Give each station a theme (geology, history, natural history, puzzle).
2
Create content for each station: design a short digital page for each station (text, image, optionally a short video or audio) – e.g. as a simple HTML page or via a free link service. Content should suit the location and include a small learning point or task.
3
Write NFC tags & generate QR codes: write the respective URL to NTAG215 tags with a smartphone app (NFC Tools or similar); also generate QR codes as a fallback (for devices without NFC). Apply both to the cache container – waterproof laminated or under a transparent lid.
4
3D-print the cache boxes: print small, unobtrusive boxes with a lid and sealing lip (PETG recommended). Allow space for the NFC tag/QR code, a small logbook and optionally a stamp or sticker. Match the colour to the terrain (camouflage), but clearly mark them as a "CreatorLab cache".
5
Place the boxes & document them: attach the boxes securely at the chosen stations (e.g. with a strap on a post, not within a nature reserve). Document the coordinates of all stations and combine them into an overview map – e.g. linked with the existing route map.
6
Classroom connection: test the scavenger hunt with a class: how does the digital layer change the hiking experience? Reflect on media literacy, the location-based nature of learning content, and the sustainable use of outdoor technology (maintenance, removal, weatherproofing).
📋
CreatorLab Project
DIY Info Board with a QR Code to OER Material
A modular, weatherproof info board for the trailside that connects short printed basic information with a QR code to in-depth OER material – for example on the formation of the Nördlingen Ries, the history of the Hohenstaufen hills, or the ecology of the Welzheim Forest. Combines laser cutting, layout work and OER creation.
Laser cutter (wood/acrylic)Weatherproof print filmQR code generatorOER platform (e.g. your own page)3D-printed frame & post mountDuration: 5–7 h
1
Define the topic & location: choose a specific spot along a route that tells a story (e.g. a geo-stop at the Ries rim, a castle ruin in the Hohenstaufen hills, a stream in the Welzheim Forest). Reduce content to the essentials: one central statement plus one or two supporting facts.
2
Design the layout: create the board layout in a vector program: a title, a brief explanation, a simple illustration or map, and a clearly visible QR code area. Test font sizes and colour contrasts for outdoor readability (even in direct sunlight).
3
Create the OER content: prepare in-depth material as an openly licensed (OER) web document – e.g. a short page with additional explanations, a quiz, or an audio-guide snippet. Publish the content under an open licence (e.g. CC BY) and add the licence attribution to the board.
4
Generate & place the QR code: generate a QR code linking to the OER page, sized so it can be reliably scanned from a normal reading distance (approx. 30–50 cm). Use high-contrast colours (e.g. black on a light background) for reliable scanning even in poor light.
5
Laser-cut & laminate the board: cut and engrave/print a backing panel from weatherproof material (e.g. coated plywood or acrylic) on the laser cutter. Laminate the print film weatherproof or engrave directly into the material to minimise UV fading.
6
Print & mount the frame: design a 3D-printed frame with a post mount that secures the board safely and against theft. Obtain permission from the responsible authorities (forestry, municipality) before installation – a link to legal and planning aspects of outdoor projects.
⏱️
CreatorLab Project
Educational Altimeter Watch with Barometer & E-Ink Display
A compact wristwatch that uses an air-pressure sensor to display the current altitude above sea level, records the elevation profile of a hike, and displays it on a power-saving e-ink screen. A good introductory project on sensors, calibration and wearable electronics – no GPS, with a focus on air-pressure physics.
Understand the basic principle: air pressure decreases with altitude – the barometric altitude formula lets you calculate approximate altitude from the measured pressure (and a reference pressure at sea level). In class: derive/discuss the formula before using it in code.
2
Connect & calibrate the sensor: connect the BMP280/BME280 to the ESP32-C3 via I2C. Before the hike, enter the known altitude at the starting point (e.g. from a map) – the sensor uses this to calculate the current reference air pressure (QNH) for more accurate relative altitude readings.
3
Calculate & display altitude: calculate the altitude difference from the current and reference pressure and show it on the display – both absolute altitude (m above sea level) and elevation gain/loss since the start. Update every 5–10 s to save battery.
4
Record the elevation profile: store altitude values with timestamps in an internal array or on a small SPI-flash/SD solution. At the end of the tour, calculate and display total ascent, total descent, and the highest/lowest point.
5
Design a wearable enclosure: 3D-print a light, wrist-worn housing or combine it with a fabric wristband. Position the display for easy reading, integrate buttons for calibration and mode switching (current altitude / profile summary).
6
Classroom connection: compare your own altitude measurements with map data or GPS altitude readings – where do discrepancies arise, and why (weather conditions, pressure changes over the day)? Link to physics (gas pressure, atmospheric layers) and off-GPS orientation in the field.
🌿
CreatorLab Project
Plant & Bird Identification Helper with Camera & AI
A handy device with a camera that takes photos of plants, leaves or birds and, using a simple on-device AI model, suggests an initial identification – including brief additional information. Connects nature observation on hikes with practical AI education and citizen-science platforms like iNaturalist.
Raspberry Pi Zero 2W / ESP32-CAMCamera moduleSmall display (OLED/TFT)TensorFlow Lite modelLiPo battery3D-printed handheld caseDuration: 8–10 h
1
Prepare the hardware: connect a Raspberry Pi Zero 2W (more computing power for TensorFlow Lite) or an ESP32-CAM (simpler, but more limited) to a camera module and a small display. Design the power supply around a LiPo battery with a charging module for several hours of outdoor use.
2
Choose a pretrained model: select an existing, openly licensed TensorFlow Lite model for plant or bird recognition (e.g. based on MobileNet, trained on a plant/bird dataset) and transfer it to the device. Weigh model size against speed.
3
Program the recognition: when a shutter button is pressed, take a photo, scale it to the size expected by the model, and run it through the model. Show the top 2/3 suggestions with probability scores on the display – deliberately presented as a "suggestion", not a certain identification.
4
Store additional information: store short local info texts for the most commonly recognised species in the region (e.g. typical plants/birds in the Ostalbkreis) – habitat, notable features, protection status. These are shown alongside the recognition result.
5
Citizen-science connection: provide an option to save the photo + GPS position (if available) + recognition result, so finds can later be uploaded to, e.g., iNaturalist – where the community then provides an expert-verified identification.
6
Enclosure & ethical reflection: 3D-print a robust, handheld case with a camera opening and a shutter button. Reflect in class: the limits of automatic identification, handling misclassifications, data privacy for location data, and the importance of citizen-science communities for nature conservation.
🌍 Climate Education · Hiking as Climate Exploration
🥾
Education for Sustainable Development · ESD
Every Step a Data Point – Learning to Read the Landscape
The five maker projects in this section turn a classic hike into a scientific field expedition. The Swabian Jura is a hotspot of climate change: dry grasslands, juniper heathland and mixed beech forests respond sensitively to changes in temperature and precipitation. With self-built technology – from GPS to geocaching to AI-assisted species identification – these changes become measurable and communicable.
🗺️ Mapping Landscape Change
Enrich GPX tracks from the hiking tracker (Project 1) with geotagged photos and observations. Document vegetation along the route: what grows where? Comparing across years: are species shifting uphill? Use the Three Emperors' Hills Trail as a long-term transect.
📈 Elevation & Climate Zones
The altimeter watch (Project 4) makes the temperature drop with altitude (adiabatic rate ~0.6°C/100m) tangible right on your own wrist. How does the temperature change between the valley town of Gmünd (320 m) and the Stuifen summit (757 m)? Analyse the elevation profile and temperature trend together.
🌿 Biodiversity & the Water Cycle
The plant & bird identification helper (Project 5) makes citizen science possible right at the trailside – upload finds to iNaturalist and observe shifts over the years. On-site observation: have springs that used to flow dried up? Experience the hydrogeology and biodiversity of the Jura together.
📋 Sharing Knowledge & Explaining the Landscape
DIY info boards with an OER QR code (Project 3) and geocaching stations (Project 2) make collected climate data and landscape knowledge accessible to other hikers. Food security and land use as a global question – experienced locally between grain fields and traditional orchards along the route.
💡
Curricular connection: the ESD guiding perspective (B), Geography Years 5–10, and the interdisciplinary subject World–Time–Society (WZG). The five projects combine geography (physical geography, climate geography), technology (GPS, sensors, AI, data visualisation) and society (climate justice, global food systems, nature conservation, knowledge transfer) into a multi-day project unit in the field.
Source: t1p.de/kixma
🎓 KPH Wien/Niederösterreich
Source: t1p.de/u0fpz
🎓 Leuphana Universität Lüneburg
Source: t1p.de/tud6f
🎓 Norwegian University of Science and Technology
Source: t1p.de/kdqd2
🎓 Ludwigsburg University of Education
Source: t1p.de/du640
🎓 St.Gallen University of Teacher Education
Source: t1p.de/u591x
🎓 University College of Teacher Education Tyrol
Source: t1p.de/xbtpz
🎓 Rhodes University
Source: t1p.de/l4jep
🎓 RWTH Aachen University
Source: t1p.de/agrxn
🎓 Technical University of Liberec
Source: t1p.de/0orcu
🎓 Thomas More
Source: t1p.de/v3yz6
🎓 University of Helsinki
Source: t1p.de/8mtlm
🎓 Leipzig University
Source: t1p.de/lncxz
🎓 Paderborn University
Source: t1p.de/ctgb1
🎓 University of Queensland
Source: t1p.de/re79m
