Re: Feedback sought on reorganizing the list of W3C Technical Reports

> On 12 Jan 2021, at 18:08, Dominique Hazael-Massieux <dom@w3.org> wrote:
> 
> Le 11/01/2021 à 13:25, Ivan Herman a écrit :
>> The concept of 'families' sounds great to me. Two immediate comments:
>> 
>> - It would be great if a 'family' had its own stable W3C URI (with some
>> reasonable resource returned when dereferencing it) and would not only
>> appear as a structure within the web site. I always had problems when,
>> in a references, a presentation, etc, people were asking me to give a
>> reference to, say, RDF. Some of the groups published an 'overview'
>> document on TR, others maintained a wiki page or a page on date space,
>> etc; I think it would greatly help if a family was a real and systematic
>> resource.
> 
> That's unlikely to be part of the first phase of redesign; we've been
> investigating that idea on the "spec series" level rather than "spec
> family", but maybe this would need revisiting. One of my concern (again
> based on past experience) is how stable can we expect the families to be
> - if there is going to be non-infrequent merge / split / shuffling of
> families, then maintaining these stable URIs will become a hassle.

Yes, there is this danger. But I am not sure it is that frequent (I may be wrong).

It is a pity if we cannot do this; I think it would be a lost opportunity for a longstanding issue...

> 
>> - I am not sure if we will be able to retain a strict hierarchy, ie,
>> that a specific TR would belong to a single family. (I am not sure that
>> was your intention, but that is how the table seems to be structured).
> 
> Yes, that's the intention - and part of what I want to confirm in this
> discussion is that this design constraint is realistic moving forward.
> 
>> There are overlaps: just from my own practice we have the DPUB-ARIA work
>> that is as much a publishing document as an ARIA one,
> 
> I categorized that document very specifically in the ARIA family since
> the document itself states "This document is part of the WAI-ARIA suite
> described in the WAI-ARIA Overview."
> 
> I absolutely agree that this document also relates to publishing; but
> for sake of simplicity and maintainability, my current proposal doesn't
> include managing multiple hierarchies.
> 
> (to put it otherwise, I don't think I can push this proposal in time for
> the current redesign phase if it needs to include multiple hierarchies)

Will we also have some sort of a tagging alongside these 'families'? If there are tags, and related search features, then that is fine.


> 
>> we have overlaps
>> between web annotations and some publishing work again, etc. There may
>> be a "primary" family, mostly due to the WG that develops it, but there
>> should be links to other families, too.
> 
> For sake of clarity, my current proposal is about these "primary"
> families; I doubt there would be an easy way to integrate multiple
> families in TR and keep the page relatively usable - so if we ever go
> the multiple families path, my assumption is that they would appear in a
> different context than the TR page.
> 

As I said above: maybe we can leave the 'family' as a main organizing principle, but also allow adding tags to documents as a further approach to search.

The community is pretty used by now to use filters with many tags… and possibly more tags than what we have now. Tags can evolve over the years, without touching the 'families'


> So I guess the question becomes whether you think going forward only
> with these primary families (either for now or for good) is an improved
> way of presenting our work or not.

My feeling is yes (as a primary means of organization), but with an additional search facility around tags


> 
> Thanks!
> 
> Dom
> 


----
Ivan Herman, W3C 
Home: http://www.w3.org/People/Ivan/
mobile: +33 6 52 46 00 43
ORCID ID: https://orcid.org/0000-0003-0782-2704