Mozilla at W3C: review of Web Applications WG Charter
April 30th, 2010
This post is by David Baron, who is a developer at Mozilla and is Mozilla’s representative to the W3C Advisory Committee.
I wanted to start blogging about one of the ways Mozilla interacts with the W3C: reviews of charters and proposed recommendations in the Advisory Committee (which has one representative per W3C member company). Sometimes I find these somewhat awkward to write, since the W3C requires a single response on behalf of Mozilla. So I want to blog about these reviews to let the Mozilla community know how we’re interacting with the W3C and have the chance to provide feedback about that interaction. Additionally, I think blogging about these reviews provides some more visibility into the W3C process.
So I’ll start with the review of the Web Applications WG Charter, which closes today.
A working group charter defines what the working group is supposed to work on and what the group is allowed to work on. This charter is a rechartering of an existing working group that we participate in. The Web Applications Working Group is, along with the HTML and CSS working groups, one of core working groups for browser standards at W3C.
It’s responsible for designing and maintaining APIs that are not part of HTML and CSS, which includes things like the DOM and XMLHttpRequest, but now extends to many new technologies (see the charter for a full list).
Since the charter was already discussed and agreed on within the working group, we were already largely happy with the proposed charter. However, that doesn’t mean that we’re interested in everything in this charter. In particular, we’re not particularly interested in the widgets work going on in this working group, nor, because of the dangers of either relying on a single SQL implementation or trying to standardize an SQL dialect, in the SQL Database work. However, we don’t want to hold up the chartering process by objecting to these items being in the charter.
Therefore, our response said that:
- We:
suggest changes to this Charter, but support the proposal whether or not the changes are adopted (your details below). - Regarding the proposed revisions to section 3.1.2: we have a slight preference for the original over the revised version, but we are happy with either.
- We would also support dropping Web SQL Database from the charter; we don’t see how this specification will achieve two independent interoperable implementations.
- We:
intend to participate in the Web Applications Working Group - We:
intend to review drafts as they are published and send comments,
intend to develop experimental implementations and send experience reports,
intend to develop products based on this work, and
intend to apply this technology in our operations. - We expect to implement (or have already implemented) many of the deliverables of this working group. We are not planning to implement the Web SQL Database, nor any of the Widgets specifications. (That’s not to say that we are planning to implement all of the rest, though.)
April 30th, 2010 at 3:00 pm
[...] This post was mentioned on Twitter by Deb Richardson, Arun Ranganathan. Arun Ranganathan said: @davidbaron explains our commentary about the WebApps WG to the W3C: http://bit.ly/charterrev Mainly: no widgets. No SQL Database. [...]
May 1st, 2010 at 3:18 am
[...] This post was mentioned on Twitter by Deb Richardson and Arun Ranganathan, Planet Repeater. Planet Repeater said: Mozilla Standards: Mozilla at W3C: review of Web Applications WG Charter http://dlvr.it/j6Gq [...]
May 4th, 2010 at 12:09 am
David,
How do you handle these challenges, I can imagine, in the case of an organization like Mozilla (or any big companies).
1. Do you get reviews from the people who are supposed to work on the specific parts mentioned by the charter?
2. Do you plan in advance who will be potentially available for committing work to the Working Group (in terms of comments and spec reviews)?
3. How Mozilla validates its development choices. For example, here you’re saying “No Web SQL database, no widgets”, is it defined by the lack of resources? or by a more general technical choice in a roadmap?
That is part of the questions I always had when I was in W3C Team. Life of a working group can be really constraining for work schedules and the right balance is difficult to achieve. Sometimes 20% of time seems to be nothing. I’m painfully aware of it, now that I’m not 100% at W3C and having a swallowing-time-job.
May 4th, 2010 at 9:06 am
@karl (this is Arun responding):
1. Yes, David typically solicits reviews from folks in the WG in question.
2. Yes, we generally plan in advance who might be on call to review specifications.
3. In the case of Web SQL Database, we were pretty much mostly in consensus that hitching our wagons to SQLite wasn’t the right choice for a web API (nor was stringifying SQL statements in JS desirable to developers), and that the web needed a better database model. We think IndexedDB sounds promising. And as for widgets, we’re pretty much in consensus that there are more important things to work on.