Moving forward on the SWHID specification

[Roberto Di Cosmo]

Dear all,

    now that the SWHID Approved Specification, version 1.0, is out, we must move forward to see if extra features may be needed, and handle the currently open issues. Here is a summary of what is currently in front of us, and a request for your feedback.

Features

A proposed new feature that has been open for a while is the addition of a qualifier to identify fragments of a file content by their byte offset

Proposal to add byte offset as fragment qualifier enhancement

#3 opened on Jul 26, 2022 by rdicosmo

The corresponding pull request (now #28) has been updated to incorporate all previous reviews, and it is ready to be merged unless there are more comments. May you have a look at it and make your comments no later than August 15th 2023?

Breaking changes

An open question is whether we should impose an order on qualifiers. This will break existing SWHIDs present in documents/code by various communities, but one a standard mitigation is to request conformant implementations to accept out-of-order input and only produce in-order outputs.

 2

Impose an order on qualifiers? breaker question

#6 opened on Nov 15, 2022 by zvr

Structure of the standard

Another important point is whether we should keep the standard as it is, or whether we should split it in two, one part for the core SWHID, and another for the SWHID qualifiers.

Splitting into two parts would mean quite some work, starting from the current specification, but may also have advantages.

Q: one standard or two parts? documentation question

#2 opened on Jun 3, 2022 by zvr

Less critical for the moment, is the question whether examples should be moved to an appendix. This is an editing work that can be done at the moment we are happy with the spec, so no feedback is needed now, we keep it open for future memory.

Move examples to an Annex documentation enhancement

#17 opened on Mar 3 by zvr

Documentation

The landing page of swhid.org is not as fancy and well structured as one could wish it was: if anyone on this list is familiar (or knows somebody familiar) with markdown/jekyll/css and wants to propose a better skin or better organization, the source code of the page is here: https://github.com/swhid/swhid.github.io

The website entry page is not very helpful documentation enhancement

#27 opened last month by lunar-debian

 2

Outreach

This last item does not require changes to the specification

 SWHID as a Package-URL type? outreach

#7 opened on Nov 15, 2022 by zvr

Looking forward to your input on all this.

--

Roberto (on behalf of the SWHID specification core team)

[Robbie Morrison]

Hi Roberto, all

Nothing jumps out for me.  If there are people resources, clearly better to address some of these refinements earlier than later.

Robbie

--
You received this message because you are subscribed to the Google Groups "SWHID (Software Heritage Identifiers) discussions" group.
To unsubscribe from this group and stop receiving emails from it, send an email to swhid-discus...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/swhid-discuss/CAJBwKuVbpP9kP-2dV6D8gdZntE11TGYW%3DociL_RYx6hrAv1XTA%40mail.gmail.com.
For more options, visit https://groups.google.com/d/optout.

-- 
Robbie Morrison
Address: Schillerstrasse 85, 10627 Berlin, Germany
Phone: +49.30.612-87617

[Roberto Di Cosmo]

Thanks Robbie!

--

Roberto 

To view this discussion on the web visit https://groups.google.com/d/msgid/swhid-discuss/5bac8e28-f32a-032f-bdc6-399102702ef8%40posteo.de.

[Jeremiah Foster]

Hello,

Thanks for the update. I’d like to understand more about the byte offset and its implications. I don’t feel like I can make an informed opinion at this time so I’ll read up in the issue tracker. 

I’m willing to take a stab at a legible, accessible, brand consistent web site skin when I return to the home office next week.

Regards,

Jeremiah

To view this discussion on the web visit https://groups.google.com/d/msgid/swhid-discuss/CAJBwKuXjinD3QxG8CgzeFuqZ9FqGJfarSa09nsEVP-0nr0H1-w%40mail.gmail.com.