COST grant to support OpenMod activities

zurich-2018

#1

Dear all,

As discussed during the last workshop in Zurich, we plan to apply for COST funding to support OpenMod conferences and networking activities.

I. COST Call Summary
In a nutshell, the COST program provides on average 137k€/year for the duration of approximately 4 years to fund networking activities among European researchers.

The COST grant would fund the following types of expenses :
• conference facilities
• accomodation
• transport
• grant administration
• …

The COST grant would fund the following meeting types :
• administrative meetings
• work-group meetings
• conferences
• training schools
• short-term scientific missions
• …

The COST grant would NOT fund :
• preparation of conferences and sessions
• research itself

A short summary of how cost works can be found here.

The most relevant documents describing the legal framing are:
COST Grant-Agreement
Cost-Vademecum
Rules for participation and implementation

II. Suggested process
The next deadline for proposing a COST action is 29th November 2018.
By that date, we would need to submit the following documents:
General Features of the application
Technical Annex of the Grant Agreement (see this example for inspiration)
• Cost Policies section (max 1000 words)
• List of institutions that apply for the COST grant

I would thus suggest the following next steps:

Step 1: Collect the following inputs : until 31st August
o workstreams: please indicate any (research, legal or admin) topics that you would be interested to discuss during future conferences so that we can include them in the grant application.
o expressions of interest: please register, if you would like to help us by joining the grant application and management team.
o potential concerns: please highlight any concerns by answering this post, or amending the FAQ below.

Step 2: Draft grant application: until 31st October
Wih the help of Ilija Batas, Ana Estanqueiro and Heidi Heinrichs we have now prepared a draft proposal

Step 3: Review and submit grant application: until 29th November
As the COST program would not pay for research time, we need to make sure, that the proposal is in line with your current and planned research activities. For this purpose, please:

  1. Add (or modify) links to current projects in section 4 of the draft proposal, to give us a feeling how well the work-plan fits with ongoing activities
  2. Fill in this short survey after reading the proposal to indicate in which of the work-groups you would be interested (ideally in the course of this week)

Any other suggestions on the scope, the wording, or references to articles that we should cite are highly appreciated – either directly in the proposal, or in this forum thread (in case of issues that require wider discussion).

Important Notes:
If we do not receive sufficient feedback regarding the intended scope of the proposal until mid-November, we may postpone the application until the next COST call in Spring 2019. This would give us the time to discuss the topic at the Aarhus workshop, but also delay the start of the funding until end 2019/early 2020.

To comply with the COST call criteria, the grant application and management team should come from diverse background (i.e. countries; sectors; research disciplines; career stages; gender), including a minimum share of inclusiveness Target Countries (>50% of researchers); In case there are several people from the same country, we will thus need to select 1-4 researchers per country in a way, which is in line with these criteria.
In order to give an of which countries and back-grounds are still missing, please refer to the following page, which should be updated automatically after each registration.
To provide people with an incentive to get involved early on, we would further take into account the time when you registered your interest (i.e. first-come-first-serve allocation).
Apologies for using a registration form powered by Google – if anyone wants to host& implement a more confidential form, that would of course be welcome. Until then, the form should at least provide more privacy than this discussion forum.

III. FAQ:

  1. What is (and isn’t) the role of the COST grant?

COST will provide funding for conferences (travel, accommodation, meals, room booking etc.); training schools; scientific exchanges etc.
COST will not provide funding to carry out the actual research (this is why they will not hold us accountable if the work plan was not fully achieved)

  1. Is there a problem with Double funding?

To our knowledge no, since COST only covers travel + networking expenses.

  1. Can people be funded that are not part of the “grant application and management team”?

Absolutely yes.

  1. Is the time for preparing workshops funded?

Unfortunately no; the funding only includes the direct travel, accommodation, room booking and meal expenses (?) (tbc: we need to check in more detail in the COST vademecum)

  1. Are countries outside Europe eligible?

mainly EuropeanEU countries. Near-neighbouring countries and a few international countries should be included. There are some special rules for those, but it is necessary to have them on-board + Turkey, Israel…

  1. Do participants have to be from Target inclusiveness countries?

Min 5 COST countries, 3 of which from COST target countries

  1. According to COST rules, 50% of funding should go to Target Inclusiveness Countries. Does this mean, some conference participants may not receive funding?

At least 50% of the actual expense should be spent to the benefit of ITC. In practical term, the COST association considers “at the benefit of ITC” any expense which is related to:

  • Reimbursements of Action participants from ITC
  • All costs of meetings and Training schools organised in an ITC
  • Any stsm which is involves an ITC (as home or host country)
  • All ITC Conference Grants
  1. Do we need to offer summer schools, even though we are not re-imbursed for the preparation time?

It will be up to the grant management team to decide whether / how many summer schools should be supported.
In practice, we do not expect this to be a problem, as some of the open source models are already now organising trainings to build up their user community.
The COST grant could thus be used to support these activities and facilitate further workshops.


#2

Hello @Christian.winzer A few quick thoughts (admittedly without having contacted potentially affected parties or having read the COST documentation properly):

Applicant. The grant holder (GH) needs to be a legal entity (Vademecum:3.1.1). Which gives rise to the idea forming a verein (a voluntary association registered under German law) with the sole purpose of managing such matters. Its charter would necessarily need to be limited to applying for and managing grants to support open energy system modeling and perhaps also taking responsibility for some or all our openmod online services. It is unlikely but not impossible that such a verein could be formed within four months to coincide with the COST application deadline. Moreover the Vademecum wording implies that the GH is an academic institution rather than a verein, but that does not of course rule out a verein. I must say I remain somewhat ambivalent about this route, while noting the potential funding from COST is very substantial.

Funded activities. Some suggestions for discussion:

  • supporting or even adopting the discussion server
  • shifting the mailing list to GNU Mailman
  • running and provisioning a versioned file server running Nextcloud
  • establishing a LibreOffice online (LOOL) server (instead of relying on google docs)
  • improving statutory (GDPR, TMG) compliance on openmod platforms, also article 13 obligations should this provision pass
  • developing (to the extent this is not research) and propagating knowledge on the legal aspects of open code and data, including copyright, database rights, and licensing
  • training related to data, including OKI frictionless data packages
  • training related to software engineering and good practice
  • tackling diversity, including but not limited to gender, EU country inclusiveness (as specified by COST), and possibly extending this to non-aligned countries globally
  • public outreach (various forms) as allowed

Note that COST supports website development, hosting, and maintenance in relation to publicity (Vademecum:10.2) but not translation services (Vademecum:10.5.5). While physical meetings and academic exchanges are prominent in the COST documentation, I have chosen to highlight our virtual meeting places instead. Hope this helps.


#3

Hello @robbie.morrison

Thank you a lot for your suggestions, which are very welcome :slight_smile: !

I have added the ideas regarding funded activities to our collection of research ideas… (which I have now re-labelled as “workstreams”, to indicate that it may contain both research topics, and admin activities).

Personally, my first concern at this point would be to identify the group of people, who would be willing to invest the time to manage the COST grant (write the proposal, process expenses claims, reports etc) by end of August…

> If you have the time, please forward our invitation to join the COST network to suitable contacts (in particular from Target Inclusiveness Countries).

Once enough people have registered, we can decide about how we want to organise ourselves (whether as “Verein”, or just the standard COST approach of a research institution (which is acting as grant holder), supported by an international network of volunteers (which are acting as a management committee).
The amount of admin support we can offer also dependends on the same criterion: the more people are willing to dedicate their time to setting up these services (and maintaining them as webmaster), the more support we can offer. If on the other hand there is just a small group of people, the priority would probably be the core processes of simply managing the COST grant.

I will write a short email in parallel to that to an appropriate COST contact person, in order to verify, whether the registration of a Verein would be possible – in particular with regards to FAQ 7 above, as well as requirements regarding “The financial stability and medium term viability of the chosen Grant Holder Institution.” (Vademecum 3.1.1, point 2). However, if we want to go down that route, someone else would need to take the lead of drafting the statutes of the “Verein”, as I will be busy drafting the main grant application.

Many greetings and speak soon,

Christian


#4

Hi @Christian.winzer - I’ve been recommending the Openmod wiki to my students for a while now so time to contribute. (Didn’t even know you were involved!) Some suggestions and resources:

  1. In the ENSYSTRA project (https://ensystra.eu/) we are committed to open modelling. I’ve been involved in writing our data strategies and protocols, which I can probably share. We will also have a workshop on open modelling at some point.

  2. I organised a workshop on data science for energy back in 2016. Full report here: https://atienergyworkshop.files.wordpress.com/2016/02/c73-post-workshop-report-final-public.pdf
    Many of these questions are still relevant, including

  • Do we need data and modelling standards to more effectively exchange data? These do not generally exist even in industry (e.g., TSOs and DNOs exchange csv files and then spend a lot of time checking and converting them into useful formats). If so, how do these come about?
  • Best practices for recording and using uncertain data. Increasingly we do not just need point estimates but a good quantification of uncertainty. How do we ensure this, especially in an open science framework?
  • How can we use data that is subject to data protection legislation efficiently? In the UK, at least, there is tendency for smart meter data, etc., to increasingly be legally owned by the consumer. It is quite possible that in the future, utilities and the like will therefore have less information about energy demand than they have now despite more data being generated. Which parts of these data are most valuable and how can we ensure that they can be made open without harming consumers?
  • How can we get industry to use open modelling methods? Many big players are hesitant because, e.g., Julia does not come with the support that Matlab does.
  • How do we check the accuracy of long-term planning models? This does not fit well with research projects, which are typically 3-5 years in duration. Nobody ever seems to go back to check, say, ten years down the line. Can open modelling help with this?
  1. The UK is woefully behind Germany when it comes to Open Science so I will contribute to this proposal where I can. I may be able to throw some resources at it. The Alan Turing Institute, the UK’s national institute for data science, will be very interested and can probably at the very least provide meeting space in central London if that is ever required.

#5

Hello @hvanderweijde Interesting. Some responses and recent references follow.

Unfortunately I cannot say much but there will be a significant report on open electricity data made public in early September 2018 which should cover a number of the points you raise.

Smart metered data from households almost certainly classifies as personal information under the GDPR. In which case, the Right to Data Portability (RtDP) (GDPR article.20) also applies. Spatially aggregated smart meter data should not present privacy issues but, as you indicate, may well be politically difficult to obtain. For many modeling purposes, distribution transformer logs should suffice together with suitable downscaling techniques.

Data standards can be specific or general. See here for an emerging energy sector metadata standard. There is also work underway to develop an energy sector ontology (systematized knowledge) being led by @ludwig.huelk

Not sure I agree with your impressions regarding proprietary versus open source support. My experiences in relation to open source support are favorable (for instance, I recently got an unsolicited personal introduction to the lead designer of Julia).

Regarding model longevity, I think open source has much to offer here. Besides some long established closed models are going open incrementally, with one example being MESSAGEix.

In terms of meeting in London, I am sure the openmod community would like to develop better contacts with UK researchers (currently mostly limited to ICL). You might well consider hosting an openmod workshop in due course?

References

European Commission (December 2017). Legal opinion: legal aspects of European energy data — Output 2 of the “Study on the quality of electricity market data”. Brussels, Belgium: European Commission.

Hirth, Lion, Jonathan Mühlenpfordt, and Marisa Bulkeley (1 September 2018). “The ENTSO-E Transparency Platform: a review of Europe’s most ambitious electricity data platform”. Applied Energy. 225: 1054–1067. ISSN 0306-2619. doi:10.1016/j.apenergy.2018.04.048. Creative Commons CC-BY-4.0 license.

Postscript

Just to comment on your “increasingly be legally owned by the consumer” remark for household metered data. As far as I am aware, there is no legal ownership in such datasets. The GDPR is very intentionally formulated under human rights and not property rights doctrines. Nor is such data is covered by any intellectual property right. It is not authored, so copyright does not apply. It is a spin-off from provision and billing, so a database right does not apply. It has no competitive value to the householder (who is not engaged in commerce in any case) and therefore does not class as a trade secret. I am not aware of any special legislation in the United Kingdom that grants property rights to household electricity data. Or am I missing something? Google suggests not.


Aarhus-2019 workshop
#6

Dear all, as announced through the mailing list, we now have a draft COST grant proposal. However, as the COST program would not pay for research time, we need to make sure, that the proposal is in line with your current and planned research activities. For this purpose, please:

  1. Add (or modify) links to current projects in section 4 of the draft proposal, to give us a feeling how well the work-plan fits with ongoing activities
  2. Fill in this short survey after reading the proposal to indicate in which of the work-groups you would be interested (ideally in the course of this week)

Any other suggestions on the scope, the wording, or references to articles that we should cite are highly appreciated – either directly in the proposal, or in this forum thread (in case of issues that require wider discussion).

Please note that if we do not receive sufficient feedback regarding the intended scope of the proposal until mid-November, we may postpone the application until the next COST call in Spring 2019. This would give us the time to discuss the topic at the Aarhus workshop, but also delay the start of the funding until end 2019/early 2020.


#7

Dear all,

in order to increase the chances of obtaining COST funding, we have eventually decided to postpone the submission of our proposal until the next collection date, which will be on 5th September 2019.

We want to use the time until then to further grow the consortium with regards to researchers from Inclusiveness Target Countries, as well as other institutions and initiatives that could help increase the impact of our network. You can find an overview of who registered so far here. If you know anyone, for whom this could be of interest, feel free to forward our short-summary, and invite them to register.

Within this context, we are investigating potential ideas for coordination with other modeling initiatives, such as the EMP-E, the IEA-ETSAP, or IRENA, e.g. by organizing a joint model evaluation exercise, where we could re-calculate some of their outlooks using different open source models. If you would like to contribute to this discussion, please feel free to post your ideas in this forum thread, and or join our next call on 9th January, 12:00-14:00 CET.

Looking forward to your replies and many greetings,

Christian