February 27, 2004

Partial Dev Day review

Dev Day '04 Review

I can't speak to the entire day because I arrived just before noon, but I did take some notes (on my superhandy Tablet PC) on some of the afternoon presentations.


QA Presentation
Asa Dotzler gave the QA presentation as usual, and brought up some pretty important points:

When the Foundation was spun off from Netscape, the Foundation basically lost their regular QA staff and now relies a great deal more on the Mozilla community for QA. Unfortunately, volunteers are typically unable (or unwilling) to devote a great deal of time to QA and in particular be regularly, consistently available.

As a result, the Foundation needs more regular testers to run daily smoketests, run smoketests for new releases the week or so prior to a release, and people to help run bug day in #mozillazine on irc.mozilla.org.

One audience member noted that credit for testers in the actual release might help motivate QA folks. If nothing else, I think this is a good idea because it would help people provide concrete proof on their resumes of the work they do for Mozilla.org.

One question during Asa's discussion of Talkback having landed in Mozilla (not yet Firefox) for the first time since the Netscape days was that it would be nice to have an open source version of Talkback available. Asa seemed open to an open source alternative, but expressed doubt that such a task would be easy or forthcoming in the near future.

Asa is working to simplify bug reporting with automatic smoketest testcases and needs volunteers to help test it.

Mozilla and Rich Internet Applications
George Coa gave a presentation on using Mozilla to create Rich Internet Applications, which are Internet-based applications that basically don't suck. He used as an example Hotmail vs. Outlook; the former has a lousy interface and the latter has a nice interface. I guess the point in Rich Internet Applications is to allow Internet-based applications such as Hotmail for example to have interfaces and usability of the same quality of applications such as Outlook that are client-based instead of server-based.

Coa says that SVG support (which isn't finished in Mozilla) is blocking the potential for rich Internet applications. Completing SVG support would allow Mozilla to fend off Microsoft's C#-based XAML interface language that is expected to be included in the next version of Windows (codenamed Longhorn). Doing so would allow Mozilla to offer a cross-platform Rich Internet Application solution that's open source three years or so before Microsoft offers it for a price and only for Windows.

Posted by kovu at February 27, 2004 8:30 PM
Comments

Simplifying bug reporting would be an important step in the right direction. There are, after all, more people using Moz/Firefox/Camino who would be inclined to click on a bug report buttona and submit a short report on what happened than there are people who are willing to dig through the present bug reporting system with the silly voting system. Prioritization of bug fixes is a management function. If an important item is "broken" does it matter how many/few people bother to vote for it?

Same could be said of the feature set in the development. To people not involved in the development of the browsers that the same "mistakes" are being repeated with each of the projects rather than taking a step ahead and adding features which are essential to the browser.

With the changes that have occurred it is more important now than ever to have a management/business plan that is rational.

Posted by: RB on March 2, 2004 1:11 PM

fp

Posted by: fp on March 3, 2004 3:00 AM

From the web developers point of view. I defintely agree on point to get proper support for SVG and of course good IDE. ;)

Posted by: JP on March 3, 2004 6:44 AM

It has taken us only 20 years to get back to terminal computing, except now we have a nifty GUI interface and they call it "progress." I think our values are misplaced :)

Posted by: S.R. Prozak on March 3, 2004 11:03 AM

Actually terminal computing is long overdue.. The only reason companies strayed from it was because M$ wanted them to.. Terminal computing is an excellent way for public school cpu labs and smaller buisnesses to save money by buying one "good" computer and many inexpensive ones.. If mozilla can help that out, then what's the harm? :-)

Posted by: AnImAl on March 3, 2004 4:50 PM
Post a comment