To classify “license of a law document” is important for OKFN-projects like OpenDataCensus,
as a general tool for check fundamentals of the openness of country legislation.
So, the first question is “there are a taxonomy of valid licenses for law-docs?”
that is, I am looking for a report or survey where a “licese-analog contract” for law documents was checked.
Thinking about law documents in any country… More contextual details at:
GitHub - okfn/opendefinition: Open Definition source
1 Like
Thank you for an interesting question. I think the problem is more complicated as it asked in question. I’m investigating licensing of Open Government Data for two years. I’ve found some issues:
- Taxonomy of licenses should be improved/created(or rules of citation should be created). Often OGD portals provide link of the license to OKFN page, and they use some “taxonomy” e.g. “Creative Commons Attribution”. It’s not clear which version of the license is applied (e.g. there are a lot of versions of CC BY-SA license which includes also localizations). OKFN page gives you general information but not concrete version: Creative Commons Attribution License (cc-by) - Open Definition - Defining Open in Open Data, Open Content and Open Knowledge. Some links of OKFN page was(still are?) not working.
- Often information of the license is not included to metadata of dataset, or is not clear what license is applied, or information is misleading. In DATA.GOV portal I’ve found that 3 different licenses are provided in metadata of one dataset.
- I’ve investigated 435,682 datasets. Only 56 % of all datasets from the investigated portals are covered by the license. Others are covered by legal notice (27%) or information is not provided/not covered by license or legal notice (17%). Some legal notices are used broadly (e.g. EU legal notice, DATA.GOV legal notice) some are unique (especially in Spain). It’s a question is it possible to create taxonomy for unique legal notices?
sometimes not quite what you expected – the Isle of Man Government OGL was localised from the UK Government OGL but accidentally missed a crown copyright in a footer here and there on “arms length” statutory body websites rather than the national government website
@jonesiom
Hello, sorry to be gone (now stable anb back)…
@martynui,
-
Now I am trying another approach, is a “taxonomy of canonical licenses”, only for an semantic approximation and cluster of similar licenses, see this considerations and procedures.
-
Offering licenses is a new approach for governments, ex. in Brasil today there are a 778398 legislative documents, but no one with the explicit CC0 (attachments, attached-maps, etc. also without CC0)… I am using other approach, see “inferred license” here.
-
Perhaps the cause is the same (as above item 2), and the approch is to interpret (we can do it by a “project goal” and producing simple dataset showing), that is interpret the “inferred license” where no one exist, and interpret “canonical license” where extrange or exotical one exist.
@g_jones, I think that Isle of Man (IM) can add a report inferredLicense-IM.md
here, like Brazil, to show the “real license”, even existing a explicit license in the documents… To show the real “legal interpretation rules” (see item 2 above).
PS: both sorry my English, and not using this tool before… we are waiting for integration.