<style>
h1 { color: #005e73 !important; }
h2 { color: #005e73 !important; }
@media only screen and (max-width: 600px) {
.logo { display: none; }
</style>
<img class="logo" style="float: right; width: 150px" src="https://www.tuhh.de/typo3conf/ext/tuhhsitepackage2021/Resources/Public/Assets/images/logo-new.svg">
# Annotated DFG Checklist for Data Management Plans (DMP)
*[Markdown-file as template](https://writemd.rz.tuhh.de/VY8ulBWjT3GI9Urs84ft2A) | [Docx-file as template](https://cloud.tuhh.de/index.php/s/sZbm9QsTEtTDAdb) | [German version](https://writemd.rz.tuhh.de/s/9g7h77FGp#)*
:::info
This document provides practical guidance for answering the questions in the [DFG checklist on handling research data](https://www.dfg.de/resource/blob/174736/forschungsdaten-checkliste-en.pdf) (as of 21 Dec 2021). Its aim is to support researchers at TU Hamburg in preparing a Data Management Plan (DMP).
Although this guidance is based on the DFG checklist, it is equally useful for templates from other funding agencies. The questions typically differ only in wording or structure, while the requested content is largely the same.
You may also combine answers from each category into a single narrative.
For additional support in creating your data management plan (DMP), you can use the [TUHH DataPlan](https://tudap.tuhh.de/) tool. Use of the tool is optional but complements the guidance provided in this annotated checklist.
:::
:::warning
If you have questions or suggestions, feel free to contact the TUHH Research Data team: research-data@tuhh.de.
More information on research data management is available at [tub.tuhh.de/en/research-data/](https://www.tub.tuhh.de/en/research-data/).
:::
---
## Category 1: Description of Data
<details> <summary><strong>1.1 How will new data be generated in your project?</strong></summary>
Describe which data will be produced in your project and how they will be generated. Explain the methodology used, including instruments, software, and digital procedures. If multiple datasets are produced using different methods, describe each dataset separately. Document the conditions under which the data will be collected as precisely as possible.
*(Example methods: laboratory and field experiments, measurements, simulations, calculations, analyses, visualisations, photos/video recordings, software development, surveys)*
:::info
<details><summary><strong>Example 1: Civil and Environmental Engineering</strong></summary> “New data will be generated through experimental material testing in the laboratory (e.g., compression and bending tests on concrete and composite samples), continuous sensor recordings (temperature, humidity, vibrations), and numerical simulations (finite element models). The measurement series are planned from March to September 2026.”</details> <details><summary><strong>Example 2: Process Engineering / Chemical Engineering</strong></summary> “New data will be generated through chemical analyses of material samples using gas chromatography–mass spectrometry (GC–MS) and inductively coupled plasma–optical emission spectrometry (ICP–OES). Sampling and evaluation will be carried out in collaboration with the central laboratory, which ensures quality control and calibration. The analyses are planned from May to August 2026.”</details> <details><summary><strong>Example 3: Materials Science / Nanotechnology</strong></summary> “New data will be generated using high-resolution imaging with a transmission electron microscope (TEM) and a scanning electron microscope (SEM) under defined laboratory conditions. The datasets will be used for structural analysis and particle characterisation. Measurements are planned for spring 2026.”</details> <details><summary><strong>Example 4: Computer Engineering / Robotics</strong></summary> “New data will be generated through simulations, laboratory experiments, field tests, and the integration of mobile sensors, drone imagery, and wearable devices (e.g., GPS, environmental and motion sensors). Data collection is planned from March to October 2026 and will be supplemented by automated log data from the control software.”</details>
:::
:::warning
**TU Hamburg:**
Before collecting electron microscopy data, consultation with the [Electron Microscopy Unit (BEEM)](https://www.tuhh.de/beem/en/about-us) is recommended.
For chemical-analytical data, the [Central Laboratory](https://www.tuhh.de/zentrallabor/en/startseite) provides support.
:::
</details>
<details> <summary><strong>1.2 Will existing data be reused?</strong></summary>
Indicate whether existing data will be used in your project. This includes your own prior data as well as data from third parties that are publicly or commercially available. Such data may serve as input for simulations, as reference values, or as comparison datasets.
* Cite datasets as you would scholarly publications (creator, year, dataset title, version, place of publication, and DOI or dataset ID if available).
* If known, state licence or usage conditions.
* If data are not publicly accessible, describe their origin, context, and access conditions (e.g., internal project drives, access restricted to project partners).
* If no suitable datasets exist, a brief note that no relevant data could be identified is sufficient.
:::info
<details><summary><strong>Example 1: Public data (weather data)</strong></summary>
“Deutscher Wetterdienst (2021). Monthly totals of station measurements of precipitation in mm for Germany (Version v21.3) [Dataset].[ urn:x-wmo:md:de.dwd.cdc::OBS_DEU_P1M_RR](https://cdc.dwd.de/sdi/pid/OBS_DEU_P1M_RR/BESCHREIBUNG_OBS_DEU_P1M_RR_de.pdf). Licence: Creative Commons BY 4.0.”
</details> <details><summary><strong>Example 2: Public data (simulation)</strong></summary>
“Knitt, M., Thakkar, M. B., Maroofi, S., Rose, H. W., & Braun, P. M. (2025). Robot Localization Failure Prediction Dataset [Simulation Data]. Hamburg University of Technology. https://doi.org/10.15480/882.15836. Licence: Creative Commons BY 4.0.”
</details> <details><summary><strong>Example 3: Internal data</strong></summary>
“Research Group X (2022). DFT Database: Simulation results on molecular structures and material properties [Internal database]. Stored on the research group’s project server; not publicly accessible; access restricted to project partners.”
</details>
:::
:::success
<details> <summary><strong>Tools for finding suitable datasets</strong></summary>
* **Repositories:** [Re3data — Registry of Research Data Repositories](https://www.re3data.org/) for identifying discipline-specific or general repositories.
* **Search engines:** [DataCite Commons](https://search.datacite.org/) (index of datasets with DOI), [OpenAIRE Explore](https://explore.openaire.eu/search/advanced/research-outcomes), [B2FIND](https://b2find.eudat.eu/), [Google Dataset Search](https://datasetsearch.research.google.com/), [Mendeley Data](https://datasetsearch.research.google.com/).
* **Research data centres:** For sensitive data; [list available via KonsortSWD](https://www.konsortswd.de/en/services/research/all-datacentres/).
* **Data journals:** Publications in [Scientific Data](https://www.nature.com/sdata/), [Data in Brief](https://www.sciencedirect.com/journal/data-in-brief), etc., often include accessible datasets. List at [forschungsdaten.org](https://www.forschungsdaten.org/index.php/Data_Journals).
Additionally, references in related publications or contacting authors directly can be helpful.
</details>
:::
</details>
<details> <summary><strong>1.3 Which data types and file formats will be produced in your project, and how will the data be further processed?</strong></summary>
Describe which (types of) data will be produced (e.g., measurement data, image data, text data, simulation data, software code) and which formats will be used. Also describe the processing steps the data will undergo (analysis, transformation, visualisation) and which tools, scripts, or software will be used.
File formats may differ depending on the project phase (e.g., raw data, processed data, archived versions). Describe these steps as soon as possible. For long-term preservation, converting to an open standard format is recommended.
A table listing data types, formats, and processing steps can be helpful.
:::info
<details><summary><strong>Example 1: Measurement data</strong></summary>
* Raw data: proprietary binary formats (e.g., <code>.spc</code>) from the device.
* Working/analysis format: Excel (<code>.xlsx</code>).
* Archival/long-term format: CSV (<code>.csv</code>).
* Processing/tools: Data cleaning and statistical analysis using Python (Pandas, NumPy); workflow versioning via GitHub.
</details>
<details><summary><strong>Example 2: Electron microscopy images</strong></summary>
* Raw data: device-specific (e.g., <code>.dm4</code>, <code>.ser</code>).
* Working/analysis format: TIFF (<code>.tif</code>).
* Archival/long-term format: TIFF (<code>.tif</code>).
* Processing/tools: Image processing and annotation in MATLAB or ImageJ; versioned workflows.
</details>
<details><summary><strong>Example 3: Text and document data</strong></summary>
* Raw/working versions: Word/LaTeX (<code>.docx</code>, <code>.tex</code>).
* Exchange format: PDF (<code>.pdf</code>).
* Archival/long-term format: PDF/A (<code>.pdf/A</code>) or text (<code>.txt</code>).
* Processing/tools: Document editing, formatting, PDF conversion; version control.
</details>
<details><summary><strong>Example 4: Simulation and modelling data</strong></summary>
* Raw data: MATLAB or solver formats (<code>.mat</code>).
* Working/analysis format: MATLAB/HDF5 (<code>.mat</code>, <code>.h5</code>).
* Archival/long-term format: HDF5 (<code>.h5</code>).
* Processing/tools: Analysis and visualisation in MATLAB; documented workflows.
</details>
<details><summary><strong>Example 5: Software code</strong></summary>
* Raw data: plain text (Python <code>.py</code>, Jupyter <code>.ipynb</code>, MATLAB <code>.m</code>).
* Archival/long-term format: unchanged plain text in repositories.
* Processing/tools: Version control with Git; reproducible workflows.
</details>
:::
:::success
<details><summary><strong>Further resources</strong></summary>
* [forschungsdaten.info: Formate erhalten. Inhalte langfristig sichern](https://forschungsdaten.info/themen/veroeffentlichen-und-archivieren/formate-erhalten/)
* [National Archives: Tables of file formats](https://www.archives.gov/records-mgmt/policy/transfer-guidance-tables.html)
* [LZV.nrw: Interactive overview of common file formats](https://www.lzv.nrw/dateiformate/)
</details>
:::
</details>
<details> <summary><strong>1.4 What is the expected scope or volume of the data?</strong></summary>
Estimate how much data will be produced in your project, stating the volume in GB or TB. Ranges (e.g., <1 GB, 1–100 GB, 1–2 TB, >100 TB) are acceptable. Distinguish between data needed only during the project and those that will be stored long-term. For very large volumes (in the TB range), you may also provide the expected annual rate of data generation to help estimate storage and backup requirements. Make your estimates more concrete by specifying the number of planned measurements, simulations, or surveys, and the expected storage volume per dataset or run.
:::info
<details><summary><strong>Example 1: Interviews and online surveys</strong></summary>
“The project includes 25 semi-structured interviews (approx. 75 minutes each). Audio recordings will be uncompressed (WAV), approx. 400 MB per interview. Additionally, compressed MP3 working copies (approx. 1.9 GB total) and transcripts/annotation files (50–125 MB) will be generated. An online survey with 2,000 participants will produce an additional 10–20 MB of raw data (CSV/SPSS). **Total data volume:** approx. 12–14 GB; long-term storage will include only anonymised transcripts and survey data (<1 GB).”
</details>
<details><summary><strong>Example 2: Laboratory and sensor data</strong></summary>
“Experimental material testing will be carried out on 100 concrete samples. Load and strain measurements will be continuously recorded via sensors (approx. 500 MB of raw data per sample). Additional working copies for analysis will amount to approx. 10–15 GB. Data generation will be staggered over six months. **Total data volume:** approx. 50–65 GB.”
</details>
<details><summary><strong>Example 3: Large-scale simulations / digital twins</strong></summary>
“Extensive simulations of an industrial production system as a digital twin will be performed. Each simulation run covers 48 hours with high temporal resolution (1 second). Each run generates 50–100 GB of raw data; 300 runs are planned. Data generation is staggered over the project duration. **Total data volume:** approx. 15–20 TB.”
</details>
:::
</details>
---
## Category 2: Documentation and Data Quality
<details> <summary><strong>2.1 Which approaches are used to ensure that the data are described in a comprehensible way?</strong></summary>
Describe how you document your research data so that they remain understandable, findable, and reusable in the long term ([FAIR Principles: Findable, Accessible, Interoperable, Reusable](https://www.go-fair.org/fair-principles/)).
Document continuously how and under which conditions your data are generated. Record who created the data, when and where they were collected, which instruments, software, or models were used, which formats and units were applied, and which versions exist.
The goal is that both you and people outside your project can understand, interpret, and reuse the data. Therefore, establish binding standards within the team for naming, structuring, and describing the data. Use recognised discipline-specific metadata standards and controlled vocabularies wherever possible. Ensure that data documentation is available both in human-readable form (e.g., lab notes, codebooks) and machine-readable form (e.g., XML metadata). If helpful, use tools that facilitate the entry of structured metadata, such as electronic lab notebooks or central metadata files. Preparing metadata in machine-readable formats does not necessarily have to be done manually: repositories often provide web forms in which core metadata can be entered and are then automatically converted into the corresponding standard formats.
A good practice is to maintain a ReadMe file for each dataset (e.g., for each top-level folder) throughout the course of the project. These files should contain all the information necessary to understand and correctly use the data during the project, such as folder structure, file naming conventions, access permissions, or backup procedures.
The contents of these various project-internal ReadMe files can later be used to create the final ReadMe for publication. This final version summarizes the key information that external users need to understand, interpret, and reuse the dataset, including context of creation, methods, variables, data structure, access and licensing conditions, versioning, and contact information.
:::info
<details><summary><strong>Example 1: Physical experiments with video recordings</strong></summary> Over a period of twelve months, video data of physical experiments on micron patterns in liquid crystals will be collected. Each video file is systematically named immediately after creation and linked with experimental metadata. The metadata contain information about the respective experimental parameters (e.g., temperature, exposure time, solution concentration, camera angle) and are maintained in a central key file. This file is regularly versioned and additionally backed up automatically.
All files are organised in a clearly structured directory system that allows unambiguous assignment of raw data, processed data, metadata, and analysis scripts. A project-wide README file describes the scientific context of the data, the file structure, naming conventions, backup strategy, as well as responsibilities and licensing information.
**Directory structure:**
Projectname_MicronPatterns/
├─ ReadMe.txt
├─ Data/
│ ├─ Raw/
│ │ ├─ Experiment_20240312/
│ │ └─ Experiment_20240405/
│ ├─ Processed/
│ └─ Metadata/
│ ├─ key_file.xlsx
│ ├─ key_file_backup_YYYYMMDD.xlsx
│ └─ data_dictionary.csv
├─ Scripts/
│ ├─ Analysis/
│ ├─ Preprocessing/
│ └─ Visualisation/
└─ Documentation/
│ ├─ metadata_documentation.pdf
│ └─ project_description.txt
**File naming conventions:**
* Videos: `VID_<ExperimentDate>_<RunNumber>.<Format>`, e.g. `VID_20240312_001.mp4`
* Metadata: `key_file_<Version or Date>.xlsx`, e.g. `key_file_20250110.xlsx`
* Scripts: `analysis_<ExperimentSeries>_<Date>.py`, e.g. `analysis_seriesA_20250305.py`
**Contents of core metadata files:**
* `key_file.xlsx`: Assignment of video files to experimental parameters (temperature, concentration, exposure time, camera position).
* `data_dictionary.csv`: Description of all variables used in the key file (name, unit, description, value range).
* `metadata_documentation.pdf`: Information on data structure, versioning, format description, and relationships between datasets.
**Standards used**
The metadata follow the DataCite metadata fields, enabling unambiguous identification, reusability, and publication in the institutional repository.
The README file follows the Cornell University template and is continuously maintained.
</details> <details><summary><strong>Example 2: Interviews and online surveys</strong></summary> Interviews and online surveys are documented continuously throughout the project. Each audio file, transcript, and annotation file receives a unique file name that indicates interview number, date, and version. Files are stored in clearly structured directories organised by project phase, data type, and collection type. A central metadata file is maintained containing information about interview guideline versions, execution, involved persons, times, and methodological specifics. A project-wide README file includes contextual information as well as access and usage conditions.
**Directory structure:**
Projectname/
├─ ReadMe.txt
├─ Interviews/
│ ├─ Raw/Audio/
│ ├─ Transcripts/
│ ├─ Annotations/
│ └─ Metadata/
├─ Online_Surveys/
│ ├─ Raw/
│ ├─ WorkingCopies/
│ └─ Metadata/
**File naming conventions:**
* Interviews: `INT_<Number>_<Date>[_Version].<Format>`, e.g., `INT_001_20251008.wav` (audio), `INT_001_20251008_v1.docx` (transcript)
* Online surveys: `SURV_<Number>_<Date>[_Version].<Format>`, e.g., `SURV_001_20251008.csv` (raw dataset), `SURV_001_20251008_analysis.xlsx` (working copy)
**Contents of core metadata files:**
* `Interviews_metadata.xlsx`: Guideline version, interviewer, date/time, location/context, recording format and version, transcript versions, methodological specifics.
* `Surveys_metadata.xlsx`: Questionnaire version, data collection period, responsible person, variable descriptions, coding/scales, notes on data cleaning.
**Standards used**
The central metadata files follow the [DDI (Data Documentation Initiative)](https://ddialliance.org/) standard. This enables later export to DDI-XML and deposition in repositories.
The human-readable README file follows the [Cornell University template](https://doi.org/10.7298/mhns-zm71) and is continuously updated.
</details>
:::
:::warning
**TU Hamburg:**
When publishing data in [TUHH Open Research (TORE)](https://tore.tuhh.de/home) or software in the [TUHH Zenodo Community](https://zenodo.org/communities/tuhh/records), a README file is mandatory.
:::
:::success
<strong>DFG Guidelines for Good Scientific Practice</strong>
Documentation of research data is a component of Good Scientific Practice. According to [DFG Guideline 12: Documentation](https://wissenschaftliche-integritaet.de/en/code-of-conduct/documentation/), all information relevant to obtaining a research result must be recorded transparently to ensure verifiability and reproducibility.
<strong>Further information</strong>
<details> <summary>Introductory articles</summary>
* [forschungsdaten.info: Datendokumentation – Warum, Was, Wie](https://forschungsdaten.info/themen/beschreiben-und-dokumentieren/datendokumentation/)
* [forschungsdaten.info: FAIRe Daten - Wie die FAIR-Prinzipien umgesetzt werden können](https://forschungsdaten.info/themen/veroeffentlichen-und-archivieren/faire-daten/)
</details>
<details> <summary>Training and educational materials</summary>
* Schmitz, D., Hausen, D. A., Trautwein-Bruns, U., Barodte, W., Dunker, V. S., Grei, P., & Hense, R. (2019). Forschungsdaten und ihre Metadaten [Video]. Medien für die Lehre. https://doi.org/10.18154/RWTH-2018-231101
* Lang, K. (2023, Januar 25). Coffee Lecture Slides: Description of Research Data – The creation of codebooks and README files. Zenodo. [https://doi.org/10.5281/zenodo.7569118](https://doi.org/10.5281/zenodo.7569118)
* Lang, K., Jessica Rex, Annett Schröter, & Nadine Neute. (2021, January 27). Coffee Lecture Slides: 5S Data. Zenodo. https://doi.org/10.5281/zenodo.4454596
* Thuringian Competence Center for Research Data Management. (n.d.). 5S Data: Sort. Retrieved 05.11.2025 from https://forschungsdaten-thueringen.de/steps/articles/5s-data-sort-en.html
* Data Carpentry: File Organization Lesson: https://datacarpentry.github.io/rr-organization1/
</details>
<details> <summary>Templates for README files</summary>
You may use the following templates from Cornell University for creating README files:
* [Writing READMEs for Research Data](https://data.research.cornell.edu/data-management/sharing/readme/)
* [Writing READMEs for Research Code & Software](https://data.research.cornell.edu/data-management/sharing/writing-readmes-for-research-code-software/)
</details>
<details> <summary>Directories for metadata standards</summary>
* [Helmholtz Metadata Collaboraiotn - Metadata Standards Catalog](https://helmholtz-metadaten.de/tools/metadata-standards-catalog)
* [Research Data Alliance - Metadata Standards Catalog](https://rdamsc.bath.ac.uk/)
* [DDC (Digital Curation Center) - Domain-specific Metadata Standards](https://www.dcc.ac.uk/guidance/standards/metadata)
* [FAIRsharing](https://www.fairsharing.org/)
* [TIB Terminology Service](https://terminology.tib.eu/ts)
</details>
:::
</details> <details> <summary><strong>2.2 Which measures are taken to ensure high data quality?</strong></summary>
Describe preventive measures for quality assurance during data collection, data creation, or data processing to systematically avoid errors. Examples include calibration of measurement devices, test and comparative measurements, plausibility checks, standardised procedures, or consistent documentation. Training personnel and using validated software can also support data quality assurance.
:::info
<details><summary><strong>Example 1: Laboratory measurements</strong></summary>
TEM and SEM measurements follow standardised workflows (checklists, SOPs, project standards). Measurement devices are calibrated regularly, and critical parameters are verified through repeated measurements.
Data entry follows predefined quality criteria; inputs and outputs are validated. All parameters, procedures, and versions are documented.
Team-internal measures such as expert checks, training, and peer reviews additionally ensure data quality and traceability.
</details>
<details><summary><strong>Example 2: Interviews and online surveys</strong></summary>
Short test recordings are carried out before the interviews begin. Transcriptions follow a unified guideline with spot checks performed by a second team member.
The online survey is implemented using an established survey tool. Input errors are prevented through mandatory responses, filter logic, and plausibility checks. Internal tests and a pretest are conducted beforehand.
All steps (questionnaire versions, test protocols, logbooks) are documented to ensure transparency and traceability.
</details>
:::
:::success
<strong>DFG Guidelines for Good Scientific Practice</strong>
[Guideline 7: Cross-phase quality assurance](https://wissenschaftliche-integritaet.de/en/code-of-conduct/cross-phase-quality-assurance/) recommends continuous quality assurance across all stages of the research process, including compliance with disciplinary standards and documentation of methods, software, and materials. [Guideline 11: Methods and Standards](https://wissenschaftliche-integritaet.de/en/code-of-conduct/methods-and-standards/) emphasises the importance of standardised procedures and comparable methods for valid and reusable results.
<details> <summary><strong>Further information</strong></summary>
[NFDI4Ing Data Quality Metrics:](https://quality.nfdi4ing.de/en/latest/index.html) Guidance on selecting, evaluating, and applying data quality metrics
</details>
:::
</details>
<details><summary><strong>2.3 Are quality controls planned and, if so, in what way?</strong></summary>
Describe measures to verify data quality after collection or processing, in order to detect and correct errors. Examples include regular checks of analysis or processing scripts, versioning (e.g., GitLab), code reviews, parallel checks of critical measurements, professional review by colleagues (peer review), or statistical tests for outliers, missing values, or inconsistencies.
:::info
<details><summary><strong>Example 1: Laboratory measurements</strong></summary>
Datasets are checked by independent repeat measurements; analysis scripts are versioned and regularly tested for consistency. Peer reviews ensure scientific correctness. Outliers are documented and corrected where necessary.
</details>
<details><summary><strong>Example 2: Interviews and online surveys</strong></summary>
Transcripts are checked on a sample basis. Survey data are inspected for completeness, plausibility, and inconsistencies. Statistical analyses for outliers further ensure data quality. Review logs and dataset versions are documented.
</details>
:::
:::success
<strong>DFG Guidelines for Good Scientific Practice</strong>
[Guideline 7: Cross-phase quality assurance](https://wissenschaftliche-integritaet.de/en/code-of-conduct/cross-phase-quality-assurance/) recommends continuous, research-accompanying quality assurance throughout the research process, including checking and correcting data. [Guideline 11: Methods and Standards](https://wissenschaftliche-integritaet.de/en/code-of-conduct/methods-and-standards/) highlights that standardised procedures and comparable methods ensure the validity and reusability of results.
<details><summary><strong>Further information</strong></summary>
[NFDI4Ing Data Quality Metrics:](https://quality.nfdi4ing.de/en/latest/index.html) Guidance on selecting, evaluating, and applying data quality metrics
</details>
:::
</details>
<details> <summary><strong>2.4 Which digital methods and tools (e.g., software) are required to use the data?</strong></summary>
Indicate which software, tools, or devices are necessary in order to correctly use, analyse, or reproduce the data. This includes programs, scripts, plugins, or measurement instruments:
* Specify versions of all software used.
* Document custom scripts or tools used for analysis.
* Prefer standardised, open, and widely used software solutions.
:::info
<details><summary><strong>Example 1: MATLAB and Python Scripts</strong></summary>
“MATLAB Version X is used for data analysis, complemented by custom Python scripts.”
</details>
<details><summary><strong>Example 2: Jupyter Notebooks</strong></summary>
“Jupyter Notebooks are used for visualisation and analysis; the documentation of the notebooks is provided together with the data.”
:::
</details>
</details>
---
## Category 3: Storage and Technical Backup
<details> <summary><strong>3.1 How are the data stored and backed up during the project?</strong></summary>
During the project, data may be stored in different locations. Since not all storage locations are automatically backed up, a regular backup strategy is mandatory. Create copies and use multiple storage locations according to the [3-2-1 backup rule](https://nfdi4chem.de/3-2-1-rule/). Describe:
* Storage locations: e.g., TUHH file services (File Service, long-term storage) and web services (Nextcloud, GitLab), institute servers, external drives, public cloud services, or local computers.
* Backup frequency: e.g., daily, weekly, or after each measurement.
* Tools and procedures: e.g., automatic backup software or manual backup.
* Responsibilities: who in the project team is responsible for the backups.
* Access and retention: where backups are stored and who can access them.
* Backup measures while traveling: e.g., daily backups and separate storage in a hotel safe.
:::info
<details><summary><strong>Example 1: 3-2-1 Backup Strategy</strong></summary>
Raw data are transferred after each measurement from the instrument PC to the TUHH File Service (team area), where they are automatically mirrored and saved via snapshot.
A second copy is created weekly on an external hard drive kept offline at the institute.
A third copy is archived monthly in the TUHH long-term storage (LZS), which is located in a separate data center and retains data for up to 10 years.
The backup process is partly automated and partly manual by project staff; the project management checks completeness monthly.
During business trips, new data are backed up daily on an external hard drive and transferred to the TUHH infrastructure upon return.
</details>
:::
:::warning
**TU Hamburg**
Please contact the [TU Hamburg Data Center](https://www.tuhh.de/rzt/startseite/kontakt) if you need dedicated storage space for your research data.
**TUHH Storage Services and Platforms**
TUHH provides several services for storing and managing research data:
* [File Service](https://www.tuhh.de/rzt/services/dateidienste/fileservice): Home areas (20 GB, expandable) and team areas (up to 20 TB). All data are mirrored hourly and snapshots are kept for up to 8 weeks.
* [Long-Term Storage (LZS)](https://www.tuhh.de/rzt/services/dateidienste/langzeitspeicher): Storage of research and project data for up to 10 years, with dedicated areas per organizational unit; annual folders are automatically created, read-only, and deleted after 10 years.
* [Nextcloud](https://www.tuhh.de/rzt/services/dateidienste/cloud): Platform for secure file sharing and collaboration within TUHH; external collaborators can join via share links. Files are encrypted and stored on servers within the TU network. Not intended for archiving or backups; accounts are deleted after expiration.
* [GitLab](https://www.tuhh.de/rzt/services/webserver/gitlab): Platform for version control and collaborative software projects; TUHH members can create projects, external users can only contribute; roles define access rights; functional accounts are not recommended. All data remain within TUHH.
**TUHH Guidelines:**
* [Storage of data in cloud or collaboration services according to TUHH regulations](https://www.tuhh.de/t3resources/tuhh/download/universitaet/it_strategie/Dokumente_TUHH_intern/TUHH_Cloud-Richtlinie_31-10-2014.pdf)
:::
:::success
<strong>Further Information</strong></summary>
<details><summary>Guides</summary>
* [NFDI4Chem: 3-2-1 Rule](hhttps://nfdi4chem.de/3-2-1-rule/)
* Heber et al.: Guideline zur digitalen Datensparsamkeit. In: obib, 2024/2. DOI: [10.5282/o-bib/6036](https://doi.org/10.5282/o-bib/6036)
</details>
<details><summary>Storage Media</summary>
* [forschungsdaten.info: Datenspeicherung und die Lebensdauer von Datenträgern](https://forschungsdaten.info/themen/speichern-und-rechnen/datenspeicherung-und-die-lebensdauer-von-datentraegern/)
* [Katarzyna Biernacka: AB Speicherung und Backup. Vergleich von Speichermedien](https://padlet.com/biernack/ab-speicherung-und-backup-c3izgipkcyfgoxke)
</details>
:::
</details>
<details><summary><strong>3.2 How is the security of sensitive data ensured during the project (access and usage control)?</strong></summary>
Describe how sensitive data are protected from unauthorized access during the project. Sensitive data include, e.g., personal data, confidential information from industry collaborations, or other protected research data. Possible protection measures include:
* Access restrictions: passwords, role- and rights-based access, encrypted storage, secure data transfer, physical protection of external drives.
* Anonymization/pseudonymization of sensitive data.
* Clear role assignment: access based on the need-to-know principle, documentation, and regular review of access rights.
:::info
<details><summary><strong>Example 1: Confidential Data</strong></summary>
“Confidential data from the industry collaboration are stored exclusively on TUHH servers with restricted access. Access is limited to project management and selected staff according to the need-to-know principle. Sharing occurs only after prior agreement and under the terms of the Non-Disclosure Agreements (NDAs).”
</details>
<details><summary><strong>Example 2: Personal Data</strong></summary>
“Personal data are pseudonymized and stored encrypted; access is restricted to authorized team members.”
</details>
:::
:::success
<details><summary><strong>Further Information</strong></summary>
* [forschungsdaten.info: Hilfreiche Tools. Daten anonymisieren](https://forschungsdaten.info/praxis-kompakt/tools/#c518401)
* Möhrstedt, J., Dähne, J., Zschiegner, M.-A., & Schuckmann, K. (2025, Januar 9). Verschlüsselung im Forschungsdatenmanagement || Vorstellung von Advanced Encryption Standard. Zenodo. DOI: [10.5281/zenodo.14620073](https://doi.org/10.5281/zenodo.14620073)
</details>
:::
</details>
---
## Category 4: Legal Obligations and Framework Conditions
<details> <summary><strong>4.1 What legal particularities exist regarding the handling of research data in your project?</strong> </summary>
Describe which legal, contractual, or ethical requirements must be observed when handling research data in your project and how you comply with them. Possible aspects include:
* Legal requirements in different countries
* Data protection requirements when processing personal research data (e.g., legal basis under GDPR, consent forms, information obligations, technical and organisational measures)
* Specific ethical requirements (e.g., research involving vulnerable groups, medical–technical areas)
* Restrictions arising from contracts or confidentiality agreements in industry collaborations
* Intellectual property rights (e.g., copyright, patents)
* Risks arising from publishing sensitive data that could endanger individuals, groups, institutions, or objects
* Data made available only under restricted conditions
:::info
<details> <summary><strong>Example 1: Clinical data</strong> </summary>
“The use of clinical imaging data takes place only after a positive ethics approval by the [Committee for Ethical Issues of TUHH](https://www.tuhh.de/tuhh/tu-hamburg/struktur/verwaltung-und-zentrale-einrichtungen/ra-rechtsreferat/ordnungen-richtlinien/satzung-des-ausschusses-fuer-ethische-fragestellung-des-as), which reviews compliance with ethical standards and the protection of participants. All data are anonymised to meet the requirements of the GDPR and other data protection regulations. Access to the data is restricted to authorised project members on a need-to-know basis; access is logged and regularly reviewed. The [Ethics and Data Protection Decision Tree of the European Union](https://ec.europa.eu/assets/rtd/ethics-data-protection-decision-tree/index.html) is used as guidance for complying with legal and ethical requirements.”
</details>
:::
:::warning
**Consultation at TU Hamburg**
For legal issues related to research data, the following offices provide support:
* [Legal Affairs](https://www.tuhh.de/tuhh/tu-hamburg/struktur/verwaltung-und-zentrale-einrichtungen/ra-rechtsreferat) (Email: recht@tuhh.de): Advice on data protection, patent and trademark law, copyright, and related rights.
* [Data Protection Coordination](https://www.tuhh.de/tuhh/tu-hamburg/struktur/verwaltung-und-zentrale-einrichtungen/ra-stabsstelle-recht/datenschutz) (Email: datenschutz@tuhh.de): Information and advice on data protection.
* [Office for Basic Affairs / Cooperation Contracts](https://www.tuhh.de/tuhh/tu-hamburg/struktur/verwaltung-und-zentrale-einrichtungen/abteilung-7-finanzen): Advice on confidentiality agreements in industry collaborations.
:::
:::success
<strong>Further information</strong>
<details><summary>Introductory articles</summary>
* [forschungsdaten.info: Rechte und Pflichten](https://forschungsdaten.info/themen/rechte-und-pflichten/)
* [forschungsdaten.info: Umgang mit Unternehmensdaten planen](https://forschungsdaten.info/themen/informieren-und-planen/umgang-mit-unternehmensdaten-planen/)
</details> <details><summary>Recommendations and decision aids</summary>
* [Statements and recommendations of the NFDI Section Ethical, Legal and Social Aspects](https://www.nfdi.de/section-elsa/?lang=en)
* [Ethics and Data Protection Decision Tree of the European Union](https://ec.europa.eu/assets/rtd/ethics-data-protection-decision-tree/index.html)
</details> <details><summary>Anonymisation of research data</summary>
* Ebel, T. & Meyermann, A. (2015): Hinweise zur Anonymisierung von quantitativen Daten. Forschungsdatenzentrum (FDZ) Bildung am DIPF. Frankfurt am Main (forschungsdaten bildung informiert, 3).
https://www.forschungsdaten-bildung.de/files/fdb-informiert-nr-3.pdf
* Meyermann, Alexia; Porzelt, Maike (2014): Hinweise zur Anonymisierung von qualitativen Daten. Forschungsdatenzentrum (FDZ) Bildung am DIPF. Frankfurt am Main (forschungsdaten bildung informiert, 1). https://www.forschungsdaten-bildung.de/files/fdb-informiert-nr-1.pdf
</details>
:::
</details>
<details> <summary><strong>4.2 Are any impacts or restrictions regarding later publication or accessibility expected?</strong></summary>
Describe whether legal, contractual, or ethical requirements may restrict access to or publication of your research data. Typical reasons include the protection of personal data, confidentiality agreements, commercial interests, planned patent applications, or embargo periods in cooperation contracts. After embargo or blocking periods expire, publication is often possible; therefore, prepare for later data publication early.
Source: P. Brettschneider / [forschungsdaten.info](https://forschungsdaten.info/themen/rechte-und-pflichten/forschungsdaten-veroeffentlichen/) / CC-BY 4.0
:::info
<details> <summary><strong>Example 1: Embargo</strong></summary>
“The data will be published after the 12-month embargo period has expired in order not to interfere with the patent application.”
</details>
<details> <summary><strong>Example 2: No restrictions</strong></summary>
“There are no legal, contractual, or ethical restrictions on the publication of the data.”
</details>
:::
:::success
<details><summary><strong>Further information</strong></summary>
* forschungsdaten.info: Rights and obligations – Publishing research data
</details>
:::
</details>
<details> <summary><strong>4.3 How are usage rights, copyright considerations, and ownership issues taken into account?</strong></summary>
This question relates to the use of third-party data as well as the planned publication of your own data. Consider both copyright and ownership aspects.
For data generated in the project: Check whether legal protection exists (e.g., copyright, database rights, contract law) and, where appropriate, assign open licences (e.g., [Creative Commons](https://creativecommons.org/share-your-work/)) to facilitate reuse. Measurement and experimental data are usually not protected by copyright because they lack the necessary level of creativity; in such cases, placing a free-use notice (e.g., [Public Domain Mark](https://creativecommons.org/publicdomain/mark/1.0/)) may be useful.
For reused data: Use or publish them only when all rights are clearly clarified. Cite sources correctly and comply with licence terms or user agreements. Publication without proper rights is not permitted.
:::warning
**Consultation at TU Hamburg**
For legal questions concerning research data, the following offices are available:
* [Legal Affairs](https://www.tuhh.de/tuhh/tu-hamburg/struktur/verwaltung-und-zentrale-einrichtungen/ra-rechtsreferat) (Email: recht@tuhh.de): Advice on data protection, patent and trademark law, copyright, and related rights.
* [Office for Basic Affairs / Cooperation Contracts](https://www.tuhh.de/tuhh/tu-hamburg/struktur/verwaltung-und-zentrale-einrichtungen/abteilung-7-finanzen): Advice on confidentiality agreements in industry collaborations.
**Licensing in TORE**
If you wish to publish your data in [TUHH Open Research (TORE)](https://tore.tuhh.de/), you can choose from various standard licences in the [“Licence” field](https://media.tuhh.de/tub/fis/hilfe/en/docs/03_forschungsdaten/forschungsdaten_hochladen/#license) for data and software. If you cannot find a suitable licence, feel free to contact us.
Data submission is carried out through a deposit or publication licence: by accepting it, you grant the University Library permission to archive your data in TORE in the long term and to make it publicly or restrictedly available based on your specifications.
:::
:::success
<strong>Further information</strong>
<details><summary>Overview articles and guides</summary>
* [forschungsdaten.info: Urheberrecht](https://forschungsdaten.info/themen/rechte-und-pflichten/urheberrecht/)
* [forschungsdaten.info: Rechtssichere Nachnutzung von Forschungsdaten(-banken)](https://forschungsdaten.info/themen/rechte-und-pflichten/rechtssichere-nachnutzung-von-forschungsdaten-banken/)
* Kreutzer, T., & Lahmann, H. (2021). Rechtsfragen bei Open Science: Ein Leitfaden. Hamburg University Press. https://doi.org/10.15460/HUP.211
</details> <details><summary>Training materials</summary>
* Brettschneider, P., Biernacka, K., Böker, E., Danker, S. A., Jacob, J., Perry, A., Wiljes, C., & Wuttke, U. (2021, August 24). Urheberrecht und Lizenzierung bei Forschungsdaten. Zenodo. https://doi.org/10.5281/zenodo.5243232
</details> <details><summary>Licences</summary>
* Peter Brettschneider, Alexandra Axtmann, Elisabeth Böker, Dirk von Suchodoletz: „Offene Lizenzen für Forschungsdaten. Rechtliche Bewertung und Praxistauglichkeit verbreiteter Lizenzmodelle“. o-bib. Das offene Bibliotheksjournal /Herausgeber VDB, Bd. 8 Nr. 3, 2021, 1-22. DOI: https://doi.org/10.5282/o-bib/5749
* [“Forschungslizenzen” website](https://forschungslizenzen.de/)
* [Choose a License (for software publications)](https://choosealicense.com/)
</details>
:::
</details>
<details> <summary><strong>4.4 Are there any important scientific codes or disciplinary standards that should be taken into account?</strong></summary>
Many disciplines, research institutions, and funding bodies provide guidelines and policies for handling research data. For your project, these requirements should be taken into account to ensure responsible research data management.
In addition to the general requirements of the DFG, discipline-specific rules as well as guidelines from funders, projects, or journals may be relevant. Large collaborative projects typically develop specific policies tailored to the project context.
:::warning
**TU Hamburg**
In addition to the [DFG Guidelines for Safeguarding Good Scientific Practice](https://doi.org/10.5281/zenodo.3923601), you may refer to the [Statutes for Safeguarding Good Scientific Practice and Handling Suspected Scientific Misconduct](https://www.tuhh.de/tuhh/tu-hamburg/struktur/verwaltung-und-zentrale-einrichtungen/ra-rechtsreferat/ordnungen-richtlinien/gwp-satzung) as well as the [Guidelines for Handling Research Data at TU Hamburg](https://www.tuhh.de/tuhh/en/research-and-transfer/translate-to-alternative-gute-wissenschaftliche-praxis/translate-to-alternative-leitlinien-forschungsdaten).
:::
:::success
<strong>Further information</strong>
<details><summary>Overview articles</summary>
* [TUHH University Library: Which guidelines provide orientation?](https://www.tub.tuhh.de/en/publishing/research-data/guidelines-and-basics/)
* [forschungsdaten.info: Leitlinien und Policies. Grundregeln für den Umgang mit Forschungsdaten (Website)](https://forschungsdaten.info/themen/ethik-und-gute-wissenschaftliche-praxis/leitlinien-und-policies/)
</details> <details><summary>Guidelines in detail</summary>
* [forschungsdaten.info: Gute Wissenschaftliche Praxis und FDM. Ein Überblick über die Leitlinien der DFG (Website)](https://forschungsdaten.info/themen/ethik-und-gute-wissenschaftliche-praxis/gute-wissenschaftliche-praxis-und-fdm/)
* [DFG Guidelines on the Handling of Research Data (PDF)](https://www.dfg.de/resource/blob/172098/4ababf7a149da4247d018931587d76d6/guidelines-research-data-data.pdf)
* [Subject-specific DFG Recommendations on Handling Research Data](https://www.dfg.de/en/research-funding/funding-initiative/research-data/recommendations)
</details>
:::
</details>
---
## Category 5: Data Sharing and Long-Term Accessibility
<details> <summary><strong>5.1 Are these data suitable for reuse in other contexts?</strong></summary>
Describe whether the research data generated in the project may be useful not only for the specific project, but also for other research questions, disciplines, or application areas. Consider potential reuse in research, teaching, development, or societal contexts.
:::success
**Example 1:**
“The measurement data can be reused in follow-up projects on energy efficiency or for validating other models.”
**Example 2:**
“The image and sensor data can also be used as training data for machine learning methods.”
**Example 3:**
“The datasets are suitable for teaching purposes, for example to demonstrate data analysis techniques.”
**Example 4:**
“The material sample data can be reused in materials science and industrial applications.”
:::
</details>
<details> <summary><strong>5.2 According to which criteria are research data selected for reuse by others?</strong></summary>
Describe the criteria you use to decide which research data will be made available for reuse by third parties. Clarify whether the data are unique and non-reproducible—thus having a high priority for archiving—or whether they are reproducible data that could be collected again. Define how scientific added value, data quality, legal conditions (e.g., copyrights, data protection), as well as file format and long-term readability are taken into account. Explain how costs and benefits are weighed, and how you ensure that only valid and reusable data are provided.
:::info
<details><summary><strong>Example 1: Experimental data and measurement data</strong></summary>
“Data from complex or one-time experimental investigations are prioritised for archiving because repeating them is expensive or not feasible. Reproducible measurement or sensor data are selected when they provide significant added value for follow-up projects, model validation, or teaching. For all datasets, proper documentation, open formats, and traceability are ensured to support reuse.”
</details>
<details><summary><strong>Example 2: Survey data</strong></summary>
“The data from the nationwide online student survey in summer semester 2024 on the impact of the energy crisis on study behaviour are archived because they represent a unique dataset. To ensure high data quality, only fully completed interviews (n = 3,246) are published.
To protect participants, personal information (name, email, IP address) is removed before archiving. Anonymisation of remaining variables is carried out, e.g., by aggregating sensitive information (such as income in categories rather than exact values). For reuse, data are provided in open formats (CSV), with proprietary formats (SPSS .sav) additionally offered.
Pretest data, incomplete interviews, and internal project documentation are not made available because they do not provide scientific added value.”
</details>
:::
:::success
**DFG Guidelines for Good Scientific Practice**
Note that research data underlying a publication, or supporting the reproducibility of research results, must generally be retained for 10 years according to [DFG Guideline 17: Archiving](https://wissenschaftliche-integritaet.de/kodex/archivierung/).
<details> <summary><strong>Further information</strong></summary>
* [Digital Curation Center: Five steps to decide what data to keep](https://www.dcc.ac.uk/guidance/how-guides/five-steps-decide-what-data-keep)
</details>
:::
</details>
<details> <summary><strong>5.3/5.4 Do you plan to archive your data in an appropriate infrastructure? If yes, how and where?</strong></summary>
First, check how and where your data can be preserved long term. Does your research field have a recognised repository (e.g., [Chemotion](https://chemotion.net/) for chemistry, [NOMAD](https://nomad-lab.eu/nomad-lab/) for materials science, [PANGEA](https://www.pangaea.de/) for climate data, [Repo4Cat](https://repository.nfdi4cat.org/) for catalysis data etc.)? To identify domain-specific repositories, you can use directories such as [re3data.org](https://www.re3data.org/). If such an offering exists and the repository accepts your data, they can be archived and published there. If your data are not suitable for specialised repositories, you can use our [institutional repository TORE](https://tore.tuhh.de) for datasets or [Zenodo](https://zenodo.org/communities/tuhh/records) for software.
:::warning
**TU Hamburg – Institutional Archiving**
For data that cannot be shared externally, TUHH provides infrastructure for archiving. Currently, a [long-term storage service (LSZ)](https://www.tuhh.de/rzt/services/dateidienste/langzeitspeicher) is available for data preservation for up to ten years. If needed, coordinate early with the [Computer Center](https://www.tuhh.de/rzt/startseite/kontakt).
**TORE – Publishing Research Data**
Would you like to archive and publish your [research data on TORE](https://tore.tuhh.de/search?configuration=tuhh_datasets)? Our [step-by-step guide](https://media.tuhh.de/tub/fis/hilfe/en/docs/03_forschungsdaten/forschungsdaten_hochladen/) shows you how.
If you are uploading data to TORE for the first time, we recommend a personal consultation. Together we will check your data and the necessary preparation. This helps avoid duplicate uploads and makes the publication process much easier. Please contact us at: forschungsdaten@tuhh.de.
**Zenodo – Making Software Citable**
Software plays an important role in many TUHH projects. It is often developed in GitLab or GitHub, but should remain permanently accessible and citable. Zenodo enables archiving, DOI assignment, and linking to the [TUHH community](https://zenodo.org/communities/tuhh/records) so that your software appears alongside other TUHH contributions.
How this works is explained in our [step-by-step guide](https://www.tub.tuhh.de/en/2024/03/03/making-your-code-citable/).
:::
:::success
**DFG Guidelines for Good Scientific Practice**
According to [Guideline 17: Archiving](https://wissenschaftliche-integritaet.de/kodex/archivierung/), research data must be preserved for at least 10 years. This includes raw data, underlying essential materials, and—if applicable—research software used. Shorter retention periods are only permissible with clear justification.
:::
</details>
<details> <summary><strong>5.5/5.6 Are there embargo periods? When will the research data be usable by third parties?</strong></summary>
Many repositories, including TORE, allow you to set embargo or access restrictions. Your data do not have to be available immediately after the project ends—you may first want to complete further publications or wait for the release of relevant articles. The main priority is ensuring that the data will eventually be available and findable. Provide your research data as openly as possible and as restricted as necessary. However, the argument “these are my data” without valid reasons is not acceptable under Good Scientific Practice.
:::warning
**Restricted publication in TORE**
In [TUHH Open Research (TORE)](https://tore.tuhh.de/), you can [restrict access to individual files or set an embargo](https://media.tuhh.de/tub/fis/hilfe/en/docs/03_forschungsdaten/forschungsdaten_hochladen/#access-restriction-and-embargo) if legal restrictions apply (closed access).
With an access restriction, only authorised individuals may view the file; other users receive no access. An embargo sets a temporary period during which the file is not publicly accessible. After the embargo expires, the file is automatically released.
:::
:::success
**DFG Guidelines for Good Scientific Practice**
The importance of publishing research data is emphasised again in[ Guideline 13: Providing public access to research results](https://wissenschaftliche-integritaet.de/en/code-of-conduct/providing-public-access-to-research-results/). Accordingly, research data should be made available as promptly as possible and ideally follow the FAIR Principles.
:::
</details>
---
## Category 6: Responsibilities and Resources
<details> <summary><strong>6.1 Who is responsible for the appropriate handling of research data (description of roles and responsibilities within the project)?</strong> </summary>
Please name the person or persons responsible for research data management within the project and describe their tasks. Responsibility may be distributed across several roles (e.g. project lead, data manager, IT staff). Typical tasks may include:
* Monitoring compliance with quality and documentation standards
* Introducing new project members to data management workflows
* Ensuring regular data backups
* Maintaining and providing documentation (e.g. file naming conventions, directory structures, versioning)
* Acting as the central point of contact for questions related to research data management
**Recommended format:**
Title; first name last name; ORCID (if available); professional position; institute; contact details; role in the project.
:::info
<details> <summary><strong>Example 1: Small projects</strong></summary>
“The project lead [name, ORCID, institute] assumes overall responsibility; the data manager responsible within the research project [name, ORCID, institute] is in charge of backups and documentation. New team/project members are introduced to the project’s research data management standards by [name, ORCID, institute].”
</details> <details> <summary><strong>Example 2: Collaborative projects</strong></summary>
“Responsibility for research data management is organised at two levels: At project level, a central data manager (coordinating institution, NAME) coordinates the implementation of RDM policies, organises training sessions and provides templates and documentation standards. In each of the three sub-projects, a responsible person is appointed who is in charge of compliance with standards, quality control, data management plans and data publication.
Partner 1: Name1, ORCID, position, institute, contact details, role in the project
Partner 2: Name2, ORCID, position, institute, contact details, role in the project
Partner 3: Name3, ORCID, position, institute, contact details, role in the project
New project members are informed about workflows by the sub-project coordinators. In addition, the TUHH University Library supports the project by offering RDM training for doctoral researchers.”
</details>
:::
:::success
**DFG Guidelines for Good Scientific Practice**
According to [Guideline 8](https://wissenschaftliche-integritaet.de/en/code-of-conduct/stakeholders-responsibilities-and-roles/), the roles and responsibilities of all persons involved must be clearly defined at all stages of the research process. This also applies to the handling of research data.
:::
</details>
<details> <summary><strong>6.2 What resources (costs, time or other) are required to ensure appropriate handling of research data within the project?</strong></summary>
Please plan sufficient human, technical and financial resources for research data management and consider it an integral part of your project. Personnel resources may include tasks related to data management, quality assurance or IT support. In large collaborative projects, it may be advisable to establish a dedicated position for IT and data management. Technical resources include storage space, servers, software, compute clusters or discipline-specific repositories. Ongoing operating costs as well as efforts required to maintain websites, databases or applications should also be taken into account. Use of the institutional repository of TU Hamburg (TORE) is free of charge.
:::warning
**Hamburg University of Technology**
If additional storage capacities are required for your project, early coordination with the [TUHH Data Center](https://www.tuhh.de/rzt/startseite/kontakt) is recommended.
:::
:::success
<strong>Further information</strong>
<details><summary>DFG funding for research data</summary>
In particular for very large collaborative projects, establishing a dedicated position for IT and data management may be appropriate and can also be funded by the DFG.
* [More information on eligible funding can be found on the DFG website.](https://www.dfg.de/en/research-funding/funding-initiative/research-data/resources-available)
</details>
<details><summary>Guidance on RDM budget planning</summary>
* [forschungsdaten.info: RDM budget planning – anticipating and planning costs](https://forschungsdaten.info/themen/informieren-und-planen/fdm-budgetplanung/)
* [LUH/TIB Research Data Service Team: Estimating the costs of research data management](https://forschungsdaten.info/themen/informieren-und-planen/fdm-budgetplanung/)
</details>
:::
</details>
<details> <summary><strong>6.3 Who is responsible for curating the data after the end of the project?</strong></summary>
Data curation refers to the ongoing intellectual maintenance of data, such as adding or correcting data in a database or ensuring long-term usability and interpretability.
If you archive and publish your data in TUHH Open Research (TORE) or other repositories, no content-related curation is carried out. Changes or corrections to files can only be made by creating new versions, in accordance with DOI assignment policies. Active migration to new file formats is not provided; the repository ensures the physical preservation of files through bitstream preservation.
Please clarify at an early stage who will be responsible for long-term archiving and maintenance of the data after the project has ended. Ideally, this person should be employed at TU Hamburg on a long-term basis, for example in an institute leadership role, as project lead or senior engineer.
**Recommended format:**
Title; first name last name; ORCID (if available); professional position; institute; contact details; role in the project.
:::info
<details> <summary><strong>Example 1: Data curation after the end of the project (short)</strong></summary>
After the end of the project, no content-related curation takes place; the data are archived in a recognised repository with a DOI and bitstream preservation.
</details>
<details> <summary><strong>Example 2: Data curation after the end of the project (long)</strong></summary>
After the end of the project duration, no content-related curation of the data takes place. The research data are published via the institutional repository TORE (TU Hamburg Open Research) with a DOI and made permanently discoverable.
Corrections or additions after the end of the project are only possible through the publication of new versions of the datasets. Ongoing content-related maintenance or migration to new file formats is not envisaged.
Long-term archiving of the data is ensured through the bitstream preservation provided by the TORE repository. Responsibility for long-term oversight and any necessary administrative updates lies with the project lead: [first name last name]; [ORCID]; [professional position]; [institute]; [contact details].
</details>
:::
</details>
---
[](https://creativecommons.org/publicdomain/zero/1.0/)
All texts in this guide, including the text modules, are **freely reusable (CC0 1.0)**.
If you have any questions, please feel free to contact the **Research Data Team** at the University Library of TU Hamburg:
📧 [research-data@tuhh.de](mailto:forschungsdaten@tuhh.de)
📞 +49 (0)40 30601-3311
For more specialised subject-specific support, please contact the [Helpdesks of the National Research Data Infrastructures (NFDI)](https://www.nfdi.de/helpdesks/?lang=en).
You are welcome to use the [TUHH Data Plan](https://tudap.tuhh.de/) tool to create your DMP:
<a href="https://tudap.tuhh.de/">
<img src="https://writemd.rz.tuhh.de/uploads/94e0410b-6660-4152-9e4f-b3350d84ea67.svg" alt="TUHH DataPlan" width="150">
</a>
*Last updated on 10 April 2026.*