bulk is an access matter, form shouldn't be specific to data, machine…

okfn / opendefinition

Open Definition source

https://opendefinition.org/

Creative Commons Attribution 4.0 International

109 stars 123 forks source link

bulk is an access matter, form shouldn't be specific to data, machine… #104

Closed mlinksva closed 9 years ago

mlinksva commented 9 years ago

move bulk should to access
machine-readibility is implied by format must be processable by a floss tool and is otherwise vague
data should not be called out in particular. bulk access and convenient format issues equally apply to other sorts of works, eg streaming-only access to video is bad
preferred form for modification also covers read/parsibility, and is a should not clear to me appropriate for open definition to completely rule out eg pdfs and lossy files as Rob Myers said in https://lists.okfn.org/pipermail/od-discuss/2015-April/001339.html

mlinksva commented 9 years ago

@wolftune I removed bold of 2nd mentions of work in paragraph.

@Stephen-Gates preferred format for modification ought drive toward 3 star (ie 2nd star requirement) publishing. FWIW the terminology is taken from the GPL's " preferred form of the work for making modifications"; I haven't seen a better statement which applies to all kinds of knowledge, but that doesn't mean one can't be made up, suggestions welcome!

wolftune commented 9 years ago

@Stephen-Gates echoing @mlinksva the "preferred" can be interpreted such that if any given party has a form they prefer, it is that form which must be distributed to others. This provides the most flexible and yet strong definition. People can argue about what should be preferred. The real violation is preferring one form for yourself but giving other only some other form.

All that said, I want to see separate references as to what forms of various media are generally recommended for Open works and what defines an Open Format. Those details don't strictly replace the "preferred form" line because even if the format is correct, a compressed / minified / compiled version of the work is not the form preferred for modifying.

Stephen-Gates commented 9 years ago

I agree with "should be provided in bulk" but see this conversation with @rgrp in the OK Forum where he has a stronger view on the provision of bulk data.

wolftune commented 9 years ago

@Stephen-Gates we still need to keep from conflating concerns that are specific to bulk data from things that apply to all Open anything. Having specifications for data in particular seems okay to me in principle, depending on details; as long as it is clearly that.

hlainchb commented 9 years ago

@wolftune this seems to fly in the face of what @mlinksva said to begin with in this thread, which I tend to agree with. I don't see why data is special when it comes to bulk. I may be misunderstanding here. Would specific examples of where bulk would not be appropriate. Assuming we can sufficiently define bulk. :)

mlinksva commented 9 years ago

I agree data isn't special when it comes to bulk. I realized that "as a whole" already covers all media types/subjects/domains, so the world bulk no longer appears.

I'm going to go ahead and merge this now as it seems like progress, though in the end the diff is slight.

If format-specific examples are needed @Stephen-Gates or whoever is interested should open up a new issue or submit a new pull request addressing. Possibly they should go in the OFD document Stephen has primarily been working on rather than the OD?

Stephen-Gates commented 9 years ago

@mlinksva I'm happy with the changes and dealing with format on OFD. Thanks