This is a wiki-editable page to develop Frequently Asked Questions (FAQ) for the Open Energy Modelling Initiative to answer the questions people new to open modelling typically have.
The answers should be written in plain English and briefly, so as not to overwhelm people in a hurry.
What is energy modelling?
Energy modelling is the computer modelling of the conversion and flow of energy to meet people’s energy service needs (e.g. electricity, transport, heating, the production of industrial goods). Energy modellers are typically interested in understanding how the energy system worked in the past, how policy and regulation interact to influence it in the short-term, and what future scenarios are possible in the long-term. Energy modelling requires data (e.g. for energy demand, energy sources, costs of infrastructure, time series) and software to process raw data, simulate and/or optimise the energy system, and software to process and visualise the results. Results are then published in reports or scientific journals.
What is open energy modelling?
Open energy modelling means doing energy modelling with open software, open data and open publishing. “Open” means that anybody is free to download the software/data/publications, inspect it, machine process it, share it with others, modify it, and redistribute the changes.
Openness is guaranteed by releasing the model software, data and publications under an open licence, typically one of the standard internationally-recognised open licences already available. A licence tells users exactly what they can and cannot do with the model, thereby providing them with legal certainty. Depending on the licence, users may be obliged to declare where the model came from (“attribution”) or share it under the same terms as the original licence (“share-alike”).
Why is openness important?
Openness is important for many reasons:
- Transparency: Trillions of euros are spent worldwide on energy each year. Decisions on energy policy are often supported by modelling results. There are many potential business interests, political biases, technology prejudices, and similar that can affect modelling choices. Openness improves transparency and increase confidence in the results.
- Quality: Openness encourages the sharing of innovative modelling ideas and can therefore improve the overall quality of modelling in the community.
- Credibility: Allowing other users to inspect and correct your model increases its credibility.
- Avoiding double-work: If every institution develops their own private software and datasets, this is very inefficient. By pooling resources, institutions can focus on improving quality.
- Encouraging cooperation: Openness makes cooperation easier. Non-disclosure agreements (NDA) are not required if everything is open.
- Reproducibility: Users can check your results, and also rerun your model with modified assumptions. Reproducibility is the basis of scientific progress.
- Visibility: Having your model online will increase the visibility of your modelling and research.
- Public funding: Publicly-funded research should be accessible by the public, since the public pays for it.
- Education: Being able to inspect and modify model code helps people to learn how energy models work.
- Public engagement: Openness makes it easier to involve the public to help guide future scenarios and perhaps improve the acceptance of policy outcomes.
- Unknown benefits and synergies: As Rufus Pollock, founder of Open Knowledge International, put it “The best thing to do with your data will be thought of by someone else”.
What open energy data exists already?
Openmod maintains a (not very complete) list of open energy datasets. See also the Open energy system databases Wikipedia page.
A group of open modellers created the OpenEnergyPlatform (OEP), a platform and database for modelling data (only open data). It focuses on model results. It is still under development.
What is a licence?
A licence is a document which tells users what they can and cannot do with your code/data/publication, e.g. inspect it, use it, machine-process it, share it, make derivative works, redistribute it. There are a variety of existing licences written by experts which you can use, see the Openmod guide to choosing a licence, which have different terms and conditions to suit your project.
Which steps are needed to licence my data?
- Clarify the origin of the data and see if there are license requirements (in some cases it may be difficult to license data with originally different (open) licenses)
- Clarify which license should (can) be used - have a look at the Openmod guide to choosing a licence
- Check if all contributors (meaning copyright holders) agree or have previously agreed to the licence (the question who are the copyright holders has to be clarified - that differs in institutions and often depends on formulations in the contract of the employees)
- Anchor the license in the metadata. => You find more about metadata on the openmod wiki page about Metadata
What is the best licence for my software/data?
Openmod maintains a guide to choosing a licence.
Isn’t putting my model online sufficient? Why do I need a licence?
By default certain intellectual property rights automatically apply to your model that prevent other people using it. Examples include copyright and, in some regions, a sui generis database right. A licence makes clear what users can and cannot do with the code/data/publication, thereby providing them with legal certainty.
Isn’t writing “reuse is allowed” sufficient? Why do I need a licence?
Custom wordings and licences are often unclear and/or legally meaningless. For example, “reuse” has many possible interpretations. The advantage of standard licences is that they have been thought through, checked by lawyers and tested in real world situations. They provide users with legal certainty.
But I would never sue anybody! Why do I need a licence?
Users need legal certainty, which is provided by a licence. You might not sue, but your institution or legal department might feel differently.
How can I track who is using my model?
You can ask users to attribute your code or data and/or cite the publication describing your model. For example, the renewables.ninja asks users to cite the papers on wind and solar conversion.
If your model is being regularly updated, you might want to create a mailing list for users to learn about updates to your project; this is another way to understand who the active users of your project are.
What is wrong with forcing users to register to download my model?
Openness includes the right for third parties to redistribute the code/data, so it doesn’t make sense to try to control access. Barriers to access run the risk of reducing the number of potential users and deterring casual users. If access is truly open, many people can use and cite your resource; if access is restricted, only the most dedicated users will end up being able to access it and cite it. If you want to be able to track users, see the answer to the above question.
What is metadata and why is it important?
Metadata means “data about data”. Metadata summarizes basic information about data, which can make finding and working with particular instances of data easier. It is important because there you get information that helps to reuse the data, to get clues of the quality of the data and to provide references to scientific work. You can find more information on the openmod wiki page about Metadata.
I like the idea of open energy modelling, but my boss is resisting. How can I convince them?
See the Openmod guide to convincing your boss.
What incentive do I have to open my model?
It will lead to increased visibility of your project. Other users may contribute towards and improve your model. If you write a paper describing your project, other people may cite it. You can continue to work on your project after you’ve left your current institution.
Our model is terribly documented, why would anyone use it if we open it?
There may be any number of parts of your project which are useful to third parties, even without documentation. Users may help to document your model. It is also a good idea to document your model for your colleagues (and yourself) using the model in the future.
We’ve sunk 1000 person-years into our model, why should we give it away?
See “Why is openness important?”. Because it’s good for transparency, it was probably funded by the public, it will save 1000 more person-years if we all cooperate more.
Our code is a mess, don’t we need to tidy it up first?
Your code many be useful to others in ways it is hard to foresee. Publish now, then other experts can help clean up your code; you’re going to want to have clean code anyway, it’s better for maintenance and finding errors.
Aren’t we obliged to provide support for our project?
No. You can choose to provide support if you want to build up a user base. A user base can be advantageous, since they will find bugs, improve the project and cite it.
Won’t private companies profit from our work?
Possibly. Next time you do a funding application, you can justify your work by the boost to the economy it has given. You can also choose a licence which prevents its use in commercial projects. (Comment: commercial is not defined in legal terms. A NGO would classify as commercial as well.)
If we make it open, how will we get funding for future projects if everyone else can just steal our model?
Funding bodies will always prefer to give money to the main originators of a software tool or dataset; having open projects means people are more likely to be aware of you and want to form consortia with you (TODO: back this up with examples).
Our model contains industry secrets, how can we publish it?
You can selectively open up the model and keep parts closed. Industry is increasingly interested in open tools (TODO: back this up with examples).
What do I do if I don’t know who has contributed to the model in the past?
In some countries, such as Germany, this is irrelevant, because the institution/employer owns the copyright of your work - you need to convince therefore the institution/employer. In other countries where the user has copyright, you may need to track down the previous contributors. Check the local copyright law in your country.
What is the Open Energy Modelling Initiative (Openmod)?
The Open Energy Modelling Initiative (Openmod) is a grass roots community of open energy modellers from universities, research institutions and the interested public that promotes open code, open data and open publishing in energy modelling. It consists of a mailing list, a wiki, a forum, a website and a regular workshop (every six months at a different location). The infrastructure is run by volunteers. There is no official membership and no membership fee.
Openmod was founded in September 2014 in Berlin, Germany. As of August 2018, there were 475 people signed up to the mailing list, mainly from Europe, but also from Africa, the Americas, Asia and Australasia.
See also the Open Energy Modelling Initiative Wikipedia page.
What does Openmod do?
Openmod is what participants make of it. The following issues are currently being tackled by openmod members:
- good practice in open source projects
- reducing barriers to developing open source projects
- helping data owners understand the merits of openness
- energy data and metadata standards
- energy model classification and cataloguing
- open/free software and data licensing
- open access to research results and journal articles
- improved research computing skills
Who is in charge of Openmod?
Nobody is in charge of Openmod. It is what its participants make of it.
Who maintains the Openmod websites?
A list of people who voluntarily maintain the website and other internet services is available here.
How is the workshop organised?
The previous workshop host solicits volunteers to host the next workshop, then organises voting to choose the next host from among the volunteers. The host is benevolent dictator for the event.