Re: A bit of electioneering on the <canvas> charter issue

On Nov 20, 2007, at 4:05 PM, Karl Dubost wrote:

>
> Le 21 nov. 2007 à 08:26, Maciej Stachowiak a écrit :
>> While some weight should be given to having a serious shipping  
>> implementation with nontrivialn market share, I hope I do not need  
>> to point out that weighing opinion strictly by market share is  
>> clearly anti-competitive.
>
>
> /me wonders what will be your stake if a big majority of web  
> designers and web developers were requesting to pull out some  
> features of the specification. In terms of business market shares,  
> they represent a lot more than browser developers. A full business  
> and market has been developed around their *own* practices.

Obviously we should consider it if such a request was made en masse. I  
think such an occurrence is unlikely, for the following reasons:

1) A subset of elite web designers sometimes oppose new features, for  
various reasons. Usually these are ideological (a proposed HTML  
feature is "too presentational") or due to the cost of updating their  
education and advocacy practice. But web practitioners as a whole  
largely seem to want new features to be available, and certainly do  
not mind them being available as long as old features are not removed.

2) Web designers clearly want more functionality than open web  
technology can provide today; they vote with their feet by using  
proprietary technologies like Flash or ActiveX, semi-proprietary  
technologies like Java, or nonstandard but widely implemented features  
of the open web technology stack.

3) In my six-year career as a browser engine developer, I cannot  
recall a single customer or developer request to remove an engine  
feature. But we get requests for new engine features (standard or not)  
all the time. I do recall complaints when we removed features that  
were so broken that they were creating compatibility problems. For  
example we had partial emulation of IE's quirky handling of DOM  
attributes where they are reflected directly as JavaScript attributes,  
this actually caused sites to break because they expected either the  
IE or the Firefox behavior, not our halfway compromise. But we still  
had some complaints when we removed it.

So, I'm wondering, are there actually specific features that a  
majority of web designers and web developers doesn't want? What are  
these features? Where is the evidence that web developers want them  
removed? Apple and the WebKit Project aim to give web developers the  
tools they need to make great web documents and applications, so we'd  
really like to know if HTML5 is on the wrong track.

> I guess it is why Sam asked for defining the "we".

In this context, by "we" I meant "the HTML Working Group".

> It is why often a charter is being defined to avoid that the will of  
> people or a group is being taken over by a few individuals. It is  
> usually to help to guarantee the life of the group. It is part of  
> the social process.

But decisions on the charter are made by fewer people than ordinary  
Working Group decisions. So I don't see how deciding things at the  
charter level prevents the will of the people from being denied. If  
anything, it seems like a very narrowly construed charter could only  
block the will of the group.

Regards,
Maciej

Received on Wednesday, 21 November 2007 00:31:43 UTC