HeatpumpMonitor.org open data license?

We don’t currently specify a data license or terms of use associated with data available through heatpumpmonitor.org.

Do we need a data license? what do we want people to do with the data - or - what might we not want people to do with the data?

My general view is that it would be great if a wide range of individuals, organisations, installation companies, manufacturers can make use of the data in order to improve the quality and performance of heat pump installations.

If other organisations are downloading, processing and analysing data made freely available on heatpumpmonitor.org it would be good to see output & analysis shared under the same open terms in order to give back to participants and the wider community. HeatpumpMonitor is a peer to peer learning tool and we would ideally like people using the data to engage under the same principle.

In my mind it would be unfair if an organisation downloaded the data from heatpumpmonitor.org and then created a product based largely on this data but did not give back in any way to the community.

In addition to open engagement in the community it would be great to find ways to enable organisations with funding available who wish to make use of the data to be able to pay participants for the use of their data ( E.g grant proposals for research programs at universities often include significant costs to cover monitoring). Heatpumpmonitor.org participants in have often gone to a substantial expense to install the monitoring hardware required and have also often invested a lot of personal time making sense of the data and participating here helping others.

Perhaps as a starting point we could adopt an open data licence such as:

or:

Any thoughts on this topic would be very welcome!

We are participating in the HeatGeek hackathon this weekend EarthAI x HeatGeek: Hackathon at Sustainable Ventures Tickets, Sat 15 Jun 2024 at 11:00 | Eventbrite and it’s been on my mind that this question of the data licence or data use terms is something we should really be clear about.

1 Like

My thoughts:

  • It’s always good to be explicit about which license applies to any freely-shared resource, so I’d definitely vote for adding one
  • In my view, we should encourage use of the data rather than discouraging it - however I agree it would be inappropriate for an organisation to benefit from the data without an obligation to share the results of their analysis with the wider community
  • Of the two licenses you mentioned, the Creative Commons one is likely to be more familiar to most people, which probably gives it a slight advantage, but I’d be happy with either of those
1 Like

I can’t quite see how it might be possible to charge a commercial firm using the data to obtain a commercial or technical advantage for themselves, after you’ve attached either license to it. It would surely only be possible by attaching a ‘personal use only’ license, and even then, you’d need to police it somehow and enforce the terms if they were violated.

You could invite contributions - but I see it as unlikely this would get past the bean-counter in the business.

Would be good to have something in there so any poor performing examples on there can’t be misconstrued by the media to push an anti HP agenda.

“A sample of heat pump data taken from heatpumpmonitor.org showed that heating your property could cost £10000 a year” or something stupid

I mean, they probably would anyway

Thanks all.

Agreed that it’s best to encourage use of the data rather than discouraging it. It would be interesting to hear from any organisations who might like to use the data if Creative Commons Attribution-ShareAlike 4.0 would be too restrictive.

Nothing to prevent multiple licenses?

  • Public domain have at it for any purpose (why not…)

  • Creative Commons sharealike (essentially making it useless for commercial firms; also making it difficult for privateers to share the analysis undertaken?

  • Paywalled. Detailed view, or historic data, only accessible/visible to those who pay to support the heat pump monitor service. No consent to continually scrape. Paywall fee set based on ability to pay - academic £zero to minimal just to get a good compliant data sharing agreement in place; £heatpumpmegaco several; MCS or Government appointer consultant£many?

2 Likes

I like the paywall option because it gives the option to support the service/project; and the option to restrict access to granular and potentially personally sensitive data such as occupancy patterns to only those who have declared how they will be processing this personal data under the GDPRs.

The fee can reflect the value of the dataset to the organisation, the value of their contribution to the cause, and their ability to pay.

You might also choose zero just to force the signing of a DPA in no order to get more than weekly cumulative energy figures for heat/elec.

2 Likes

Definitely keen that we don’t restrict use too much. I’m going towards the idea of a simple attribution only open licence. I just noticed this open data set for LCOE data which is a good example GitHub - iain-staffell/GNESTE: The Global and National Energy Systems Techno-Economic (GNESTE) Database It uses the creative commons licence that only requires attribution https://creativecommons.org/licenses/by/4.0/.

It’s an interesting idea. Perhaps the first step is to implement token based access with certain limits in terms of request rate, granularity and then have the option to get in contact with us to discuss additional requirements.

There’s also more that would be good to do on the data privacy front and user control over what level of data access is allowed. E.g the ability to restrict access to detailed data but allow the summary level stats to be public.

So shall we go ahead with Creative commons CC-BY-4.0 for now?

1 Like

That would seem sensible to me. Certainly better than having no license applied to the data (which is effectively what we have at the moment) where we currently don’t even require Attribution.

Enforcement of the license terms is a whole other question but no different for Data versus Software or Hardware designs.

2 Likes