RE: Review of the Reformatted Recommendations (was Re: New W3C Web Site Launched)

As I've said, I strongly agree, and would like to add an additional point:

Keep in mind that various groups have, for good or bad reasons, augmented 
the standard CSS with customizations or additions, and the new base CSS is 
mangling some of these.  Compare the renderings of constraints, principles 
and good practice notes from the old version of the Architecture of the 
World Wide Web (attachment ArchdocOriginal.jpg) [I can't even get a 
hyperlink for this in the old format!!!] and the new [1] 
(ArchdocBroken.jpg).   The layout is clearly broken, and the color coding 
of the boxes is also lost.

Furthermore, those boxes were intentionally parallel in layout to the ones 
used for "Stories" (other two attachments).  First of all, the layout of 
the stories themselves is now pretty bad (no padding on the bottom), but 
the similarity with good practice notes, constraints, and principles is 
also now less apparent.

I request that all of this be reverted immediately.  It's very 
unprofessional looking as it stands, and arguably misleading. Furthermore, 
fixing the architecture document and various findings to "work" with the 
new CSS would be time consuming and unnecessary IMO, and who's to say 
nobody would try another change later?

The larger observation here is:  don't assume that the decoupling of style 
and content is sufficiently robust in practice that you can go swapping 
stylesheets on deployed documents.  It is possible to achieve this if the 
authors of every document are aware from the start that this might happen 
and if they plan for it, but then it can become extremely difficult to 
meet the demand for features specific to particular sorts of documents 
(principles, constraints, and good practice notes in the case of AWWW). 
The safe assumption is, IMO:  act as if each publication is permanently 
bound to its stylesheets unless you have specifically verified that 
changes aren't going to compromise the quality of the result. Furthermore, 
don't assume that you can legislate this away be telling groups to live 
with only a bounded set of stylesheet classes used in limited ways. That's 
a fine thing to try for when content is quite repetitive in nature or 
tightly controlled, but W3C publications cover a very wide range of 
subject matter and audiences;  such a restriction would not be practical 
for them.

Noah

[1] http://www.w3.org/TR/webarch/

--------------------------------------
Noah Mendelsohn 
IBM Corporation
One Rogers Street
Cambridge, MA 02142
1-617-693-4036
--------------------------------------








Michael Champion <Michael.Champion@microsoft.com>
Sent by: chairs-request@w3.org
10/15/2009 12:10 PM
 
        To:     Robin Berjon <robin@robineko.com>, Christopher B Ferris 
<chrisfer@us.ibm.com>, "spec-prod@w3.org" <spec-prod@w3.org>
        cc:     "chairs@w3.org" <chairs@w3.org>, Ian Jacobs <ij@w3.org>, 
W3C Members <w3c-ac-members@w3.org>, Michael Rys <mrys@microsoft.com>, 
(bcc: Noah Mendelsohn/Cambridge/IBM)
        Subject:        RE: Review of the Reformatted Recommendations (was 
Re: New W3C Web  Site Launched)


I agree with Robin and Jim that there wasn't enough consultation with WGs 
before their publications were reformatted.  I'd suggest immediately 
rolling back the changes to the *specs'* formatting (not the overall 
website -- there are issues, but I will follow Ian's advice to be 
patient).   Then, follow a process such as Robin suggests to work through 
the issues and let WGs opt-in -- or at least opt out-- of the new CSS. If 
you need guinea pigs, use submissions, Recommendations without active WGs, 
etc., but leave active WGs in control of both the form and content of 
their specs.

I personally like the new look of the documents and believe that most WGs 
will eventually opt-in, but the team really needs to respect the consensus 
process and the principle that the WGs own the specs they produce.

-----Original Message-----
From: w3c-ac-forum-request@w3.org [mailto:w3c-ac-forum-request@w3.org] On 
Behalf Of Robin Berjon
Sent: Thursday, October 15, 2009 7:33 AM
To: Christopher B Ferris; spec-prod@w3.org
Cc: chairs@w3.org; Ian Jacobs; W3C Members
Subject: Review of the Reformatted Recommendations (was Re: New W3C Web 
Site Launched)

Hi all,

in the absence of a list specifically tailored for editors, I'd like to 
suggest that we can move this discussion to spec-prod@w3.org which seems 
to be the closest logical location.

All but a few of the W3C Recommendations listed at:

  http://www.w3.org/TR/tr-status-stds.html

have been reformatted to match the look of the new site. In many cases 
this has broken them with various degrees of severity (in some cases 
rendering them largely unusable). Surely, users can go to the previously 
published version if they happen to need a functional document, but it's 
not something that they're likely to guess (unless they read the small 
note at the bottom of all those documents).

I don't think that I'm being particularly grouchy or demanding if I state 
that running live breaking experiments on documents that are expected to 
be stable and authoritative at their canonical URLs is a rather bad 
situation, that we should work together to address as quickly as possible.

I have already heard several people who had reviewed beta.w3.org being 
surprised at the changes made to the Recommendations. It seems rather 
clear to me that this part of the new site has not received anywhere near 
the amount of validation that it ought to have.

So in the spirit of reaching consensus that we are all familiar with, and 
in order to help the Team out as it pushes through this huge redesign 
effort that is in pretty much every other one of its aspects absolutely 
fantastic, to get all the editors past and present who are willing to help 
to discuss ways of addressing the current breakage swiftly. I would think 
that anyone would naturally be welcome to help, but I single out editors 
as they are after all those whose blood and tears and paper cuts from a 
thousand man-hours of last comments build these documents and donate them 
to W3C. They know the kinks and the warts, and they've generally had no 
other option but to listen to their users at great length.

Amongst the topics that I would like to see resolved as part of this 
discussion are:

  - Should this experimentation be performed on live Recommendations at 
their canonical URLs?
  - Should old documents be updated at all? If yes, should the WGs in 
charge handle them?
  - Do TRs need to have the site navigation included or are they 
standalone?
  - Is it okay to have the logos of commercial companies on TRs?
  - Should the SotD and paraphernalia be pushed to the end?

And of course any other concern that editors may bring up. Personally, I 
agree that the idea behind most of the changes has merit, but I believe 
that this is being rushed out unbaked, and that the quality of our 
production is taking a hit because of it.

WDYT?

--
Robin Berjon
  robineko - hired gun, higher standards
  http://robineko.com/