W3C home > Mailing lists > Public > site-comments@w3.org > October 2009

Previous specs still exist [Was: New W3C Web Site Launched]

From: Ian Jacobs <ij@w3.org>
Date: Wed, 14 Oct 2009 22:20:49 -0500
To: Jeff Schiller <codedread@gmail.com>
Message-Id: <6641E5A9-92D5-452D-A9A6-DCAD4773E75B@w3.org>
Cc: site-comments@w3.org, chairs@w3.org, Robin Berjon <robin@robineko.com>

On 14 Oct 2009, at 7:51 PM, Jeff Schiller wrote:

> In case it isn't immediately obvious, the src of that <img> should be:
>
> src="images/coords/InitialCoords.png"
>
> from what I can tell (at least that's what my offline copy has).
> Thank god I saved this specification locally long ago.

Hi everyone,

Please note that all the specs you've published still exist at their  
original URIs. We published brand new ones at brand new URIs.

Nothing was moved or lost for the existing publications.

  _ Ian

>
> Regards,
> Jeff
>
> On Wed, Oct 14, 2009 at 7:48 PM, Jeff Schiller <codedread@gmail.com>  
> wrote:
>> <img alt="Example InitialCoords - SVG's initial coordinate system"
>> src="//afs/w3.org/pub/WWW/TR/2003/REC-SVG11-20030114images/coords/ 
>> InitialCoords.png"
>> width="300" height="100"/>
>>
>> Sorry to be blunt, but frankly I'm just amazed that this "live
>> re-styling" was allowed and didn't require some sort of approval by
>> the chairs.
>>
>> Jeff
>>
>> On Wed, Oct 14, 2009 at 7:42 PM, Jeff Schiller  
>> <codedread@gmail.com> wrote:
>>> I'm going to echo Robin's "surprise" here.  I did not expect to have
>>> problems with specs because the W3C website design changed.   
>>> Websites
>>> are different than technical specifications.  One is fluid, the  
>>> other
>>> is not expected to be, especially after the specification has been
>>> released.  I will also echo Robin that specification re-styling  
>>> should
>>> have been done in a sandbox somewhere, then reviewed with each WG,
>>> then released.
>>>
>>> On the one hand, thank you for restoring http://www.w3.org/TR/SVG11/
>>>
>>> On the other hand, http://www.w3.org/TR/SVG11/coords.html (for
>>> instance) is missing all of its images.  This is just the first  
>>> thing
>>> I noticed.
>>>
>>> I hope the web team can "power through" all these changes as quickly
>>> as possible so that W3C doesn't lose credibility.
>>>
>>> Regards,
>>> Jeff
>>>
>>> On Wed, Oct 14, 2009 at 11:52 AM, Robin Berjon  
>>> <robin@robineko.com> wrote:
>>>> Hi Ian,
>>>>
>>>> On Oct 14, 2009, at 18:28 , Ian Jacobs wrote:
>>>>>
>>>>> We had a beta test period for some time. Going live was intended  
>>>>> to get
>>>>> more feedback (which is happening).
>>>>> We will fix things as we go. If the new templates prove  
>>>>> unfixable, we'll
>>>>> remove them.
>>>>
>>>> I do not question that that approach is right for the general site;
>>>> requirements for standards are different though. Cool standards  
>>>> don't change
>>>> under your feet. I strongly urge the Team to consider things that  
>>>> live under
>>>> /TR/ as being a completely different use case and a largely  
>>>> different crowd
>>>> than the rest of the W3C website.
>>>>
>>>> And if you do insist on running live tests inside TR, why run  
>>>> them on the
>>>> stable, important documents and not on unstable and less  
>>>> important ones?
>>>> Presumably, their formatting requirements are the same, while the  
>>>> impact of
>>>> issues is lesser.
>>>>
>>>>> We've kept the previous documents available at their original  
>>>>> URIs. We
>>>>> have new URIs for the reformatted specs. So people who wish to  
>>>>> refer to the
>>>>> dated spec can continue to do so. The "latest version" URI takes  
>>>>> you to the
>>>>> reformatted versions.
>>>>
>>>> At the very least would you consider switching that around so  
>>>> that the
>>>> latest version would point to the latest version that actually  
>>>> reached
>>>> consensus in the WG in charge of publishing them and was endorsed  
>>>> by the
>>>> Membership? A lot of resources out there point to the latest  
>>>> version instead
>>>> of the dated one (as does Google in most cases).
>>>>
>>>>> Instead, I ask your patience while we fix bugs (which one should  
>>>>> expect
>>>>> during a significant upgrade such as this one). If you need the  
>>>>> stable
>>>>> previous specs in the meantime, those URIs still work.
>>>>
>>>> I am more than happy to be patient and to help out with the  
>>>> creation of new
>>>> templates. I merely ask that we don't play Russian roulette with  
>>>> documents
>>>> that worked and that are widely referenced. I am somewhat  
>>>> surprised (to put
>>>> it nicely) that the same organisation that deliberately inflicts  
>>>> dated URIs
>>>> upon the world would toy with the product of consensus so  
>>>> carelessly.
>>>>
>>>>> On the question of "google on every page" we discussed this  
>>>>> issue quite a
>>>>> bit. We certainly don't have the resources to write our own  
>>>>> search engine.
>>>>> And offering N search options to users (in a gesture to be more  
>>>>> neutral) is
>>>>> not really a service to users. We talked to google about  
>>>>> dropping their logo
>>>>> requirement and they let us know that that would not be possible.
>>>>>
>>>>> Regarding twitter and identi.ca, we are already using 2 rather  
>>>>> than one.
>>>>> If we end up setting up our own microblog service at W3C, then  
>>>>> we might
>>>>> promote it instead. But all of that would require more resources  
>>>>> than we
>>>>> have currently allocated.
>>>>
>>>> Again, the general website and the specifications are different  
>>>> things. I'm
>>>> perfectly happy with those things in the general site. I would be  
>>>> happy with
>>>> ads on the general site  that'd make the W3C some useful money.
>>>>
>>>> The specifications, on the other hand, are authority documents. I  
>>>> have
>>>> absolutely nothing against Google, but W3C specifications aren't  
>>>> Google
>>>> specifications. There is enough confusion in the community  
>>>> already about who
>>>> drives what.
>>>>
>>>>
>>>>> I prefer to keep going and work out the bugs. The advantages of  
>>>>> the new
>>>>> templates for TRs include:
>>>>>
>>>>> * integrated into the rest of site
>>>>
>>>> I think that's a bug. Specifications aren't pages just like other  
>>>> pages in
>>>> the site. We shouldn't be trying to give the impression that they  
>>>> sit on the
>>>> same level, which is what the current layout does.
>>>>
>>>>> * status section has been moved down so people can begin reading  
>>>>> more
>>>>> quickly
>>>>
>>>> I'm not convinced that that's a good change either  see other  
>>>> thread in
>>>> chairs@.
>>>>
>>>>> There are some challenges in ensuring we don't break formatting;  
>>>>> we will
>>>>> continue to investigate and fix those.
>>>>> If this experiment does not bear fruit, we will roll back.
>>>>
>>>> Is there at least a date at which we plan to make a call as to  
>>>> whether the
>>>> experiment was a failure or not? Is there a process of any sort  
>>>> telling us
>>>> who's making the call and who we can appeal to? Is there any plan  
>>>> to engage
>>>> and involve the people who actually write the specifications? The  
>>>> people
>>>> writing specification production tools?
>>>>
>>>>> But given the largely positive feedback we've received, I'd like  
>>>>> to keep
>>>>> plugging ahead for a short while.
>>>>
>>>> Positive feedback on the site in general should be taken  
>>>> separately from
>>>> feedback on the specs, I hope.
>>>>
>>>> --
>>>> Robin Berjon
>>>>  robineko  setting new standards
>>>>  http://robineko.com/
>>>>
>>>>
>>>>
>>>>
>>>>
>>>
>>
>

--
Ian Jacobs (ij@w3.org)    http://www.w3.org/People/Jacobs/
Tel:                                      +1 718 260 9447
Received on Thursday, 15 October 2009 03:20:57 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Wednesday, 24 October 2012 16:21:32 GMT