search

agiorguk / gemini

Resources relating to the UK Gemini metadata profile
5 stars 3 forks source link

(an attendee Geospatial Commission Data Discovery project workshop 28 Nov 2020) Suggestion we explicitly default the metadata to CC-0 (or equivalent) #26

Open PeterParslow opened 3 years ago

PeterParslow commented 3 years ago

i.e. set ISO 19115 MD_Metadata.metadataConstraints to https://creativecommons.org/publicdomain/zero/1.0/ or ….

<gmd:MD_LegalConstraints>
    <gmd:useConstraints>
        <gmd:MD_RestrictionCode
          codeList='http://standards.iso.org/ittf/PubliclyAvailableStandards/ISO_19139_Schemas/resources/codelist/gmxCodelists.xml#MD_RestrictionCode'
          codeListValue='otherRestrictions'>otherRestrictions</gmd:MD_RestrictionCode>
    </gmd:useConstraints>
    <gmd:otherConstraints>
        <gmx:Anchor xlink:href='http://inspire.ec.europa.eu/metadata-codelist/ConditionsApplyingToAccessAndUse/noConditionsApply'>no conditions apply</gmx:Anchor>
    </gmd:otherConstraints>
</gmd:MD_LegalConstraints>
nmtoken commented 3 years ago

Should we use full name in title not disco

6footdestiny commented 3 years ago

Which 'parts' of GEMINI - the docs were already releases under some permissive license I think? CC-0 for the 'code' feels less sensible to me.

PeterParslow commented 3 years ago

This isn't about parts of GEMINI - which as you say are released under CC-BY.

This is about encouraging GEMINI users to explicitly state the licence under which their metadata records can be used, with the default/presumption being that metadata is 'free to be reused'.

6footdestiny commented 3 years ago

In which case and given the primary constituency I would have thought OGL was more appropriate?

PeterParslow commented 3 years ago

Perhaps - but not every GEMINI metadata creator is "government". Also, OGL requires attribution, so anyone re-using the metadata would be required to state where they got it from. That may or may not be a good thing. What do others think - about the idea, and about which licence to suggest?

nmtoken commented 3 years ago

I'm not sure CC-0 would be a correct default

You can copy, modify, distribute and perform the work, even for commercial purposes, all without asking permission.

(https://creativecommons.org/publicdomain/zero/1.0/)

Because generally we don't want people to modify the metadata.

PeterParslow commented 3 years ago

That is a good point:

Here's an example that has come across my in-tray today: a certain OS customer ingests OS data into their system, and in parallel to that creates themselves a metadata record that is based on our GEMINI record. It is import that this is a new metadata record - derived from ours - not a modification of the existing one. Specifically, we don't want them keeping the same fileIdentifier if they change anything else (e.g. adding an extra "point of contact" to someone internal to their organisation)!

PeterParslow commented 3 years ago

I'm not sure that CC has a variant which allows "remixing" (as opposed to modification of the original) without requiring "attribution"!

nmtoken commented 3 years ago

Is that an issue?

The metadata has to include the metadata point of contact, so will always be attributed.

PeterParslow commented 3 years ago

The metadata has to include the metadata point of contact, so will always be attributed.

The attribution requirement in any of the Creative Commons license is that the derived work attributes the original author. I would expect the Metadata point of contact of the derived work to be the relevant contact for that derived work, not for the original.

I don't think that CC-0 gives someone the right to modify the original; the problem is that we (the GEMINI record creator) would want to specify that someone creating a derived work HAS TO give it a new File identifier. We may actually be less concerned about being attributed - and although the Creative Commons site says that all their licences require attribution, I can't see that CC-0 does (because they see it as 'not a licence').

I think it's a custom licence (CC-0 or CC-BY plus 'shall use a new File identifier); perhaps we'd have to roll our own, adding that one obligation?

nmtoken commented 3 years ago

What if the default was CC-BY-ND, because generally we don't want to allow derived (meta)data

You are free to: Share — copy and redistribute the material in any medium or format for any purpose, even commercially

But metadata providers could also enter into other arrangements (“dual licensing”) to allow derivations and remixing, such as different file identifiers

(https://creativecommons.org/faq/#can-i-enter-into-separate-or-supplemental-agreements-with-users-of-my-work)

PeterParslow commented 3 years ago

That may be better, but:

archaeogeek commented 3 years ago

Taking this back to first principles since I wasn't at the meeting where this was raised- what's the context for this? What are we (as data and metadata producers) trying to protect (or protect against)? I think we have to be very careful about how this is worded to avoid confusion for end-users (we can't assume any familiarity whatsoever with CC licensing)

PeterParslow commented 3 years ago

As someone who was at the meeting, I think it is not so much to protect/protect against, as to encourage / enable / facilitate.

Apparently, there are some people who are unsure that they can take a metadata record, ingest it into their system, use it as a base for any sort of processing - including creating derived metadata records to describe derived datasets. The guy who raised this concern suggested that explicitly stating the licence under which the metadata record is made available could ease this situation.

I felt (& I think other "GEMINI folk" present at the time) felt that 'of course that's what you should do', hence the idea that GEMINI records default to allowing that kind of thing.

nmtoken commented 6 months ago

Just revisiting this and moving forward BGS metadata will be explicit in stating metadata is provided under CC BY-ND 4.0.

archaeogeek commented 2 months ago

This might be better as guidance rather than an explicit instruction. Assigning to @nmtoken with assistance from @PeterParslow to create initial page

nmtoken commented 2 months ago

Intending to implement in XML like:

<gmd:resourceConstraints xlink:title="MetadataConditions">
    <gmd:MD_LegalConstraints>
        <gmd:useConstraints>
            <gmd:MD_RestrictionCode codeList="gmxCodelists.xml#MD_RestrictionCode"
                codeListValue="otherRestrictions"/>
        </gmd:useConstraints>
        <gmd:otherConstraints>
            <gmx:Anchor xlink:href="https://creativecommons.org/licenses/by-nd/4.0/">Metadata is
                distributed with a CC BY-ND 4.0 (Attribution-NoDerivs 4.0 International)
                licence</gmx:Anchor>
        </gmd:otherConstraints>
    </gmd:MD_LegalConstraints>
</gmd:resourceConstraints>
PeterParslow commented 2 months ago

Surely it should use gmd:metadataConstraints (a child of MD_Metadata) not gmd:resourceConstraints as a child of either MD_DataIdentification or SV_ServiceIdentification?

nmtoken commented 2 months ago

You are right! that would be better so...

<gmd:metadataConstraints>
  <gmd:MD_LegalConstraints>
     <gmd:useConstraints>
        <gmd:MD_RestrictionCode codeList="gmxCodelists.xml#MD_RestrictionCode"
           codeListValue="otherRestrictions"/>
     </gmd:useConstraints>
     <gmd:otherConstraints>
        <gmx:Anchor xlink:href="https://creativecommons.org/licenses/by-nd/4.0/">Metadata is
           distributed with a CC BY-ND 4.0 (Attribution-NoDerivs 4.0 International)
           licence</gmx:Anchor>
     </gmd:otherConstraints>
  </gmd:MD_LegalConstraints>
</gmd:metadataConstraints>
PeterParslow commented 1 week ago

Now we need to agree where to put that XML fragment and how to introduce what it's about!

  1. Where to put it? I think it belongs in 2.2.14 https://agiorguk.github.io/gemini/1048-uk-gemini-encoding-guidance.html#_limitations_conditions_and_licences, alongside the other "licence" bits. It's almost another sub-section in / near there.

It would need some text, such as:

"ISO 19115 allows a specific license declaration for the metadata itself. Generally, part of the reason to publish structured metadata is so that others can ingest it and manage it within their systems alongside the data, you may want to use something like CC BY-ND, which allows people to transform / remix the metadata but not to re-distribute the result. You should encourage anyone who does transform the metadata to create and use a new File identifier. The ISO 19115 metadata constraint element should appear towards the end of the file, after all the GEMINI elements. This position is set by the ISO 19139 schemas, based on the order implied in the ISO 19115 standard."

This approach deliberately avoids adding "Metadata constraints" as an additional GEMINI element. But it does break new ground by giving advice concerning a bit of ISO 19115 which isn't actually in GEMINI - so it may be "cleaner" to add a new element, id = 54.

  1. we could introduce the whole matter of 'enabling & controlling what people do with your metadata record" in https://agiorguk.github.io/gemini/1052-metadata-guidelines-for-geospatial-data-resources-part-1.html. Probably as part of a section on "the purpose(s) of metadata"!