Towards Mozilla 1.0
Tuesday June 26th, 2001
Gervase Markham recently posted his feelings on what a 1.0 release of Mozilla would be. Gerv has sent us the follow-up to that posting, including much of the feedback he received. To read it, click the full article link. Once you have read through it, we welcome you to post your feelings on what you think a 1.0 release would have. [As Gerv says, please don't post your favorite list of bugs, only the criteria for choosing what bugs to fix.]
#101 Re: Re: Cut the bull or put the bull in?
Wednesday June 27th, 2001 6:13 PM
You are replying to this message
"How do you measure the size of IE 5.5?" Why don't we start with the download size?
"Also, why should a Linux Mozilla user (or someone on any other OS) care about performance in relation to IE 5.5?" Just because they're not running an OS with Windows doesn't mean that they should accept a browser that is slower than IE5.5 on a comparable win/mac system.
"Why pick version 5.5?" Why Not? You have to pick some target. I could easily retort every target you make with "Why pick that?", but doing so doesn't produce any valuable concerns.
"Do you use Mozilla's -turbo to mitigate the IE preloading factor or not?" Personally, I run with the -turbo option. But again, you set a goal, and you hit it. If you say "startup time should be as fast as IE with the -turbo option" then you hit it. If you decide to set your goal differently, then you do so.
"Do you take into account that Mozilla may be doing far more with a page (supporting <LINK>, doing hover correctly) than IE?" Do you take into account that IE supports other features that Mozilla does not? Like favicon.ico. Of course not. Zdnet's ibench doesn't take that into account, and neither is any other benchmark program. I can't reiterate this enough, "You pick a goal, and you hit it"; goals can be so simple. Like, "beat IE 5.5 on machine X in Zdnet's ibench".
"How do you do proper metrics on things which take fractions of a second, when you can't compile IE with timing?" In that case, how does any timing benchmark ever get created? Did you know that they do happen, regularly? C|net posted benchmarks for page loading. This is no big technical feat.
"What do you suggest?" I suggest that you pick a benchmark and beat IE in it. Any benchmark. I know my team regularly puts out benchmark results versus AIM 4.3 (I work on the NIM sidebar in the commercial arm of Mozilla).
"If we only _try_ to hit it, what's the point of having it at all?" What's the point of having any guideline? Are you really going to freeze the API's? (probably not) Are you really going to achieve 100% Good Net-Keeping Award? (probably not). Everyday mozilla engineers are asked to triage their bugs to the milestones they think they can fix those bugs by. Why bother (you might ask)? Since they're not going to hold the release for all the bugs! Why not just ignore the milestone field in bugzilla until the bug is actually fixed, then go back and set when the bug really was fixed! The reason, because we *want* to try.
"Would you hold the 1.0 release if the performance figures were a bit off those you'd pulled out of thin air but everything else was in place?" I take offense that you say I pulled performance figures out of thin air. For starters, I never mentioned any performance figures. But if I were going to use performance figures, they wouldn't come out of thin air; and I would hold the 1.0 release to hit those figures, just as much as any other guideline.
"Mozilla.org has made promises in this area; would you be happy with breaking them?" Mozilla.org never, in the beginning, promised a 1.0. Mozilla was, originally, meant as a developers browser, not meant for the consumer. Since that has changed, Mozilla has decided to also promise a 1.0. Anyone remember the Milestone system before this? M1 - M30, with really no end. So if you want to know if I think it's okay that Mozilla break it's promise. Ask the WSP. One of the promises was about developing a browser in a reasonable amount of time. Why don't you read <http://www.mozilla.org/apology.html> for a good idea of what mozilla.org promised (for a 5.0, note that Mozilla 1.0 had been acknowledge as having only taken 7 months to ship!). Or read <http://groups.google.com/…c=1&selm=an_353682358>
In short, yes. I think most people would be happy if Mozilla delivered what they promised long ago.