Mozilla Status, Advocacy, and Mozilla in the Press
with Mike Shaver and Dan Mosedale

11/12/1999

<chrisn> our chat today is with Mike Shaver and Dan Mosedale, talking about Mozilla advocacy...
  could you introduce yourselves, and briefly describe your job with Mozilla?
<shaver> I'm Mike Shaver, and I'm (ir)responsible for developer and community relations
  I help people find the stuff they need to get their jobs done, or help them find jobs to do.
<CGI-BINux> I'm CGI-BINux, I'm a part time CGI programmer and an unofficial mozilla advocate.
<photek> hello
<chrisn> dmose?
<dmose> I'm Dan Mosedale, and I do various random stuff. I started out helping with the IS side of things, more recently have done some tool work, and am currently doing random stuff to help push the browser out the door.
<url> i'm Adam Hicks, a curious end-user in oregon.
<chrisn> ok, anyone else want to introduce themselves?
<BenGoodger> Hi I'm Ben, and I'm a Mozillaholic.
<bergee> i'm bergee, just a mozilla advocate
<TonyG> TonyG - Mozilla Advocate/unofficial UK defender of the faith :)
<Kovu> big time evangelist ;)
<photek> Hervé Renault, mozilla apprentice
<chrisn> I'm chris, and I run mozillaZine
<kerz> I'm jason, and i screwed up mz's rdf file
<jst_home> Hi, I'm Johnny Stenback, DocZilla developer from Finland.
<smfr> Simon Fraser, Mac dude and editor person
* Antisoche <-= also a Chris, essentially lurking
<chrisn> ok, once more with the instructions:
  after introductions, if you have a question, type ? in the channel. You'll be added to the questioner's queue. when called on, you'll get one
  question and a followup. When done, you can type ? to get added back to the queue
  the topic tonight is Mozilla status, advocacy, and Mozilla in the press
  ok, Kovu: first question...
<Kovu> I noticed that the new Netcenter beta matches the new Mozilla UI almost perfectly. I'm confused because I assumed Mozilla's UIs was not NS's
<shaver> I can't really comment on the Netcenter UI
  much of the UI work on Mozilla is done by Netscape engineers
  and if they make a web site that looks like Mozilla, more power to them =)
<chrisn> kovu: followup?
<Kovu> Cool
<chrisn> ok, next question from kerz:
<kerz> Does mozilla support XML?
<chrisn> :)
<shaver> heh
<dmose> yes.
<Ben_Goodger> (lol)
<kerz> Ok, just checking.
<shaver> I think the only version of Mozilla that doesn't support XML very well indeed is whatever build Jesse Berst cons'd up
<dmose> specificly, it includes the James Clark's expat parser.
<shaver> I haven't been able to get my hands on it to verify, though.
<dmose> and the UI itself is XML-based.
<chrisn> kerz: followup?
<kerz> nope
<chrisn> next question from smfr:
<smfr> there's been a lot of marketing spin about the size and performance of the browser ("small and fast")
  is this spin justified?
<shaver> (is spin ever justified?)
<smfr> should I rephrase?
<shaver> OK, so there are a lot of things wrapped up in the phrase ``small and fast''
  no, that's OK
  I know what you mean
  we all certainly want Mozilla to be infinitely small and infinitely fast
  how are we doing so far?
<bergee> very good
  (imho that is
<shaver> our download size is about 5M, which is much better than our predecessor (4.x)
  our memory footprint size is larger
  right now, for example, our layout engine has an overhead of about 14 * character-data-size
  Word has about 12, and the layout team has plans to get it down to 8 or 9
  so that would likely mean about a 30% memory footprint drop if we can pull it off
<dmose> one piece of the puzzle has to do with modularity, too: now you can build a browser which doesn't necessarily have code for extensions that you don't use.
<shaver> yes, that's very true
  I think speed is the more interesting discussion
  when it comes down to how many CPU cycles are needed to display a given page, I don't think we can _ever_ get faster than 4.x
  the reason is simple: we're doing a lot more work, in terms of correctly applying CSS and building a DOM for manipulation
<smfr> right
<shaver> percieved speed, though, is another battle entirely
  and I think that things like incremental flow-and-reflow will let us kick 4.x all over the map there
<smfr> yes
<shaver> the last speed issue -- and I know this one near and dear to smfr's heart -- is startup speed
  right now, we're pretty miserable
  some of that is the cost of the high degree of modularity
<Waldo> Sorry I'm late :)
<smfr> right, but there are still gains to be had at startup
<shaver> but some of that is just waste, I'm sure, so we're working hard on getting rid of those
  hey!
  you knew the answer!
<smfr> that was a good answer, shaver, thanks.
<shaver> you were setting me up!
  I think that the Mozilla 5.0 alpha will be about a 3M download on Linux (including mailnews)
<smfr> ok, thanks!
<shaver> memory footprint will probably be 15-20M, I fear, but maybe less
<Antisoche> /?
<shaver> (sorry for the long answer)
<chrisn> smfr: followip?
<shaver> he better not
<smfr> I agree with everything shaver said, and am glad to hear that layout are still working hard on footprint issues
<chrisn> followup, that is
  ok, next questioner, TonyG
<TonyG> I run a popular UK site were I plug Moz as often as possible. It would be cool if there was an "official" 'Try Mozilla Now' logo or suchlike. Nothing like free advertising :). Any chance of this?
<shaver> ok, this is good
  that's a marketing question
  I like those, believe it or not
  there's a big hole in the Mozilla Story right now, and its name is Marketing
  we need to do a _much_ better job of branding and proselytizing our technology
  we do a pretty good job with developers, I think
<dmose> FWIW, there are already some banners at http://www.mozilla.org/banners.html, and you're free to use any of those.
<shaver> that's very true; I'd forgotten about those
  one thing I _don't_ want is ``best viewed with Mozilla''
  I would much rather see ``best viewed with standards-compliant browsers''
  some day, I hope, that won't mean just Mozilla =)
<dmose> errg.. i mistyped: <http://www.mozilla.org/banners/>
<chrisn> TonyG: followup?
<TonyG> Generally something to go on sites to raise awareness of Mozilla for those philistines out there is needed IMO.
<shaver> yes.
  make some buttons =)
<TonyG> www.nomad-jedi.com has an icon i did based on the Throbber, for example.
  :)
<shaver> ok, thanks
<chrisn> ok, next questioner is me:
  could you go into a little detail about the march towards beta, and when you expect that to happen, based on the current state of moz?
  12/15 was the last estimated date for some sort of beta that I know of...
<shaver> ok, it's time for the schedule dance
  M11 is basically now; the builds go to QA today or tomorrow
  12/15 is a very very special date
<Kovu> :)
<shaver> on that date, The Lizard Willing, we will be releasing the Mozilla Alpha
  ``why not a beta?''
  it's really quite simple:
  ``beta'' means ``feature complete''
  since we honestly don't know exactly which features will be in Mozilla 5.0, we can't be feature complete yet
  you can also think of the Alpha as a ``developer preview''
  we want to get something nice and pretty and stable out there so that people can use it as a base for developing their own plugins or applications with Mozilla technology
  ``ok, fine, when's the damned beta?''
  if all goes according to plan, I think the beta will be in mid-February
  I want a beta for my birthday, which is the 17th
<dmose> specifically, we want to get something that's more or less usable as a day-to-day browser out, in order to help attract more folks to play around with the code.
<shaver> right
<dmose> (referring to the alpha here)
<shaver> I think we're going to get a huge amount of excellent feedback from the Alpha, and we'll need about 2 months to process it all)
  we actually have a new person on the mozilla.org team who's going to help us handle all that feedback
  her name is Christine Begle, and she's actually on vacation until the new year, but I think she's going to be a huge part of our Alpha/Beta success
  (and there _will_ be success. don't make me come over there.)
* dmose chuckles
<shaver> I don't think I can really give a meaningful date for a final release, unfortunately
<kerz> End of next year?
<shaver> I want to, but the alpha->beta->beyond period is too scary (in a good way) to make guesses right now
  I hope that helped.
  oh, before that
  god, man
<chrisn> so, followup: you think that 12/15 will reach dogfood?
<MattyT> end of 22nd century?
<shaver> well, dogfood is a really tricky word
  there's been a lot of confusion about it, and some of that is my fault
  there is a project within netscape to get to a ``dogfood'' point
  (where ``Netscape dogfood'' is obviously a Netscape criterion)
  I think that the things that Netscape needs for dogfood are pretty much the same things that Mozilla needs for Alpha/Developer Preview
  so we're on the same track, and all is well
<chrisn> ok
  next questioner is Antisoche:
<Antisoche> hi --
  I wonder if theres anything "you would have done differently" and
  What is in store for 6.0? Any major goals like CSS2, etc?
<shaver> hey!
  that's two questions!
  =)
<Antisoche> or is it just _way_ too early?
  three =)
<shaver> what would I have done differently?
<Antisoche> I don't really know. You can skip that if you'd like
<shaver> it's ok
  hindsight is 20/20, if not better, so I think one thing is to have started on the SeaMonkey plan earlier
* uberfloss is away, bbl
<shaver> we didn't know that we needed to do that, so it's not really something I think of as a mistake, but that might have been nice
<dmose> right. if we could have convinced Netscape to avoid doing a 4.5 release and focus on the free source earlier, that would have been nice.
<shaver> yes, but that's not really a Mozilla decision
<dmose> but they had contractual obligations, unfortunately.
  shaver: true enough.
<shaver> I'm not sure; I think I might be too close to the process right now to give a good answer to that.
  so let's get to 6.0; that's a fun question
<MattyT> hopefully we'll nuke all the most popular rfes ...
<shaver> one of the things that our architecture gives us is the ability to release different pieces at different times
  so I'm not sure what ``Mozilla 6.0'' means
<dmose> well, one thing that's related to that..
<shaver> ``Mozilla 5.0'' is an important baseline, starting-point release for all the components, and the people using them
<Waldo> It's really a 1.0, I think :)
<Antisoche> Well, presuming that "mozilla" is 5, the question is just "what next" Obviously you can't put everything in Moz 1.0, so kinda "what got cut". I just didn't want to type so much
<dmose> is that right now too many of the interfaces in Gecko are currently internal private interfaces. we'd like them to be more modularized
<chrisn> (hold off adding yourself to the queue - we have a lot of people to get through. we'll add more after we get through this list)
<shaver> some things, like XUL2.0 and CSS2/3 and some new modularity hooks in layout, are obvious candidates
<dmose> for the sake of (eg) the MathML folks before the next major browser release
<shaver> there's lots of little things that fell off the 5.0 train, and we need to spend some time getting those cleaned up
  maybe that's for 5.1?
  one thing that I really want to have in 6.0 is a shorter schedule
<chrisn> :)
<shaver> the next time around, we're going to know a lot more about developing with this technology, and this process
  so I'd like to see a 6.0 entering beta around the end of 2000
  (another maybe-not-in-5.0, for-sure-in-6.0 thing is better remote chrome stuff)
<chrisn> ok, next questioner's Kovu:
<Kovu> I'm curious how you'll deal with incompatibility with IE. Is an ActiveX solution a priority or do you guys give much of a rodent's arse? Personally I would hope you would try to convince people not to support that crap by not including compatibility.
<dmose> so there is in fact already an ActiveX plugin for Communicator
  and the old Communicator plugin interface will continue to work in Mozilla
  so it's possible that things are ok with that right now.
  i don't know whether anyone has tried it in recent builds.
<shaver> ActiveX, and other non-standard-but-common web technologies, make for a fun discussion
  on the one hand, Mozilla very much wants to promote standard-friendly solutions to problems
<MattyT> ActiveX is a standard, just a very bad one =)
<zinebot> SeaMonkey changed state from Close to Open
<shaver> on the other, we're very much into flexibility
  I think that all the pieces are in place for someone to provide first-class ActiveX support in Mozill
  they could even use Winelib to let them run in Linux!
<kerz> yay?
<shaver> but I don't think it's a high priority for anyone right now
  (a mixed blessing, indeed)
<chrisn> Kovu: followup question?
<Kovu> Yeah, I guess it's just a question of where you draw the line between supporting bad stuff to enhance compatibility and how that on the other hand might encourage that incompatibility. Thanks, though, good answer.
<shaver> ``we sell rope''
<chrisn> Next questioner is bergee. he asked:
<dmose> but if someone were to contribute ActiveX code, however, we'd certainly be interested in (at the very least) looking at it.
<chrisn> Will you guys ever make a build with just the browser (IE no mail/news/editor stuff) - or are we going to have to make a build like that ourselves?
<shaver> that's a hard question
<dmose> the editor is a slightly different question than the mail/news
<shaver> mailnews is easy: separating it out is a packaging issue, and you'll probably see a different package for mozilla-mailnews in the Beta, if not the Alpha
<dmose> because the editor is used for text fields in the browser. so large chunks of that code will already be necessary.
<shaver> editor is hard: we use lots of the editor code for our text widgets, so you can't just rip out mozilla/editor and get going
  I might like to see a cleaner factoring of the editor code to produce a minimal editor widget (text only, etc.), but I don't know how hard that is, or what the gain would really be
<MattyT> I imagine the html editor is used in mailnews and not browser ...
<shaver> <textarea type="text/html"> is an emerging defacto standard, though, so...
<chrisn> next questioner is cls, actually...
<cls> you mentioned modularity, would it be possible to reduce the runtime footprint by not loading (at all) any "advanced" HTML features that the user does not want. (Advanced in this case being things beyond HTML 3.2 or HTML 2.0 even)
  basically, are there any plans to make bits of the layout engine more modular?
<shaver> there are certainly plans to make the layout engine modular is some ways
  splitting XUL and HTML is one such way
  I'm not sure if anyone has looked at HTML3.2/HTML2.0, etc.
  it would be interesting to see how much code you could gain that way, and at what cost in speed for the full-HTML4 case
  to some extent, though, this is an operating system problem:
  if you're not using the <FRAME> tag, then the code related to it should stay out of memory, dammit =)
<chrisn> cls: followup question?
<cls> what does it take to become an "official" member of mozilla.org? :)
<shaver> that's not a followup, you cheater, but I'll ask anyway
  s/ask/answer/
  basically, you apply
<MattyT> cls: you need to hork the tree at least once
<cls> ok
<shaver> if someone is interested in being part of the core mozilla.org administrative team, they would need to make explain to the existing core what they want to do
  and then it's like any interview process: lots of embarrassing questions and stuff
<chrisn> ok, next questioner is Antisoche:
<Antisoche> me?
  I think I'm done, thanks.
<chrisn> ok
  then on to Waldo....
<Waldo> With all the *insane* anti-mozilla spinning going on lately, how does an open source project best combat FUD and disinformation? Can we expect AOL to be involved in or leading a media/education campaign any time soon, and if so will it be coordinated/planned with advocates in the OSS community?
<kerz> you type damn fast
<dmose> I think we can expect
<shaver> heh, yes, it is *insane*
<Waldo> mavis beacon rocks! :)
<shaver> I think that you'll notice that our press balance has been more positive of late
<dmose> to get help from many of the different companies involved in Mozilla, including AOL.
  some will be in the form of join press releases, some of which we've seen recently
<shaver> the _best_ thing we can do to get good press is release a browser that kicks ass and takes names
<Kovu> Hallelujah!
<shaver> there will still be bad press; everything gets bad press
<skJain> Me says yay to that also
<shaver> now, we also need to work the system a little better, I suspect
<dmose> but if any mozilla contributors would like to donate marketing muscle, we'd be interested
<shaver> I do what I can as the press point-man for mozilla.org, but there's only so much I can do
<chrisn> :) uh - hello?
<shaver> there's definitely room for more people doing stuff like chrisn and ben_goodger do
<Kovu> I'm trained in journalism and willing ;)
<shaver> that's the kind of advocacy and marketing that I want to see lots of
<chrisn> shaver: I'd love to have more to do on that front
<shaver> it's (usually =) ) even-handed and fair, and helps us a lot
  it's great for morale, too
<chrisn> shaver: just like to have a plan behind it somehow
<shaver> I think chrisn does a _spectacular_ job right now
  even slashdot is starting to follow his pro-Mozilla lead =)
<zinebot> Just appeared in MozillaZine (http://www.mozillazine.org/): &lt;font color=&quot;red&quot;&gt;Developer Chat going on NOW!&lt;/font&gt;
<dmose> Kovu: you could consider writing an article about why mozilla is in fact succeeding, and submitting it for publication in the more mainstream press, for example
<zinebot> Just appeared in freshmeat.net (http://freshmeat.net): thttpd 2.05
<chrisn> they finally are!
<shaver> we'll show them. we'll show them all.
<dmose> .
<Waldo> My followup-- there was some discussion a few weeks back on mozillazine about using more contests to generate slogans, banners, icons, etc.. Ok, not so much a question as a comment... Oh yeah, and Chrisn-- you do an awesome job!
<chrisn> thanks!
<shaver> (that wasn't a very good answer, I fear. sorry)
<chrisn> ok, next question is from kerz:
<kerz> What is mozilla.org's official mascot?
<MattyT> uh oh
<chrisn> hehehe
<shaver> our official mascot is pavlov
<Kovu> oh the green or red thing again!
<chrisn> heh
<dmose> that would be the red dragon that you see on the mozilla web pages
<MattyT> green!
<Kovu> T-rex
<shaver> that would be a good poll, if it hasn't been one already
<dmose> the green dragon is quite nice, but he is and always has been very much associated with netscape
<Kovu> dragon, pshaw
<skJain> Inversing the colors in the image you get a pretty shade of blue
<chrisn> shaver: ok, will do
  just don't stuff the ballot box!
<MattyT> shaver: I've suggested it already =)
<shaver> but at this point, I think that changing the Mozilla colours or style would be harmful to what little brand-value we already have
  I should point out that I don't think we can really change
<chrisn> maybe we could have Dave Titus do up a cartoon version of the red T-Rex
<shaver> but I'm still interested in seeing the poll results
<chrisn> so that it could be used in various settings easily, like the green moz was
<shaver> that would be a decent compromise
<dmose> if Dave's interested in that, that would be a neat thing
<chrisn> I have to talk to him anyway. I'll ask him about it
<Kovu> dmoz.org still uses the green BTW
<chrisn> kerz: followup?
<kerz> followup: What are the licence(s) on mozilla.org images?
<dmose> fwiw, for various reasons, dmoz.org is fact not associated with mozilla.org
<shaver> the license on the images is a really good question
  I'd like to say ``as free as the code'', but I should really look into getting that clarified
* shaver adds to his todo list
<Kovu> dmose: maybe another reason to leave the green
<dmose> Kovu: yes
  .
<chrisn> ok - next in the queue was me, but my question was asked by someone else a while ago, so we're on to cls, if he has another question:
<cls> yep
<shaver> cls is full of it
<cls> there have been murmurings that mozilla is over-engineered and will never be finished/released. (I suppose no open source project truly is). Do you think this perception would have been different if development had continued on the classic
<shaver> I mean, of questions
<cls> instead of dropping nearly everything and basically starting over 6 months after the project started?
<shaver> if we had continued on the classic, we would have shipped 5.0
  I'm pretty sure of that
  but 5.0 would have been _vastly_ different
<Kovu> (piece of crap, methinks)
<shaver> it would have been seriously constrained by the 4.x heritage
  I spent months hacking on Classic layout and DOM and CSS stuff, so I speak from painful experience here
  as to the over-engineering accusations, there are relatively few people outside of the developers who are familiar enough with the code to give an opinion that I would respect
  I think some of the developers would agree that parts are over-engineered
  I could probably point to a few things that I'd like to trim a bit =)
  but I don't think it's a serious issue
  most of the requests we get in terms of architecture are for _more_ engineering, like additional hooks and modularity indirections
  4.x was arguably not as ``over-engineered'', but it was also not nearly as pleasant to work with
  we'd be seeing an entirely different set of complaints, I think =)
<chrisn> ok, we have just a few more questions in the queue for this evening...
  skJain is up next:
<kerz> he's gone
<chrisn> ack
  ok
  photek is up next
<photek> do you expect MS to release a competing IE 6 or whatever the number in reaction against Moz 5 soon after you release it ?
<shaver> do I get to give the IE-competition rant now?
<chrisn> yes, please!
<kerz> look out
<dmose> that isn't something we've been giving much thought to
<Kovu> (IE has accomplished its goal, and is on back, back burner, methinks)
<shaver> first, I gotta say that I have no idea what MS will do with their versioning, and I try to spend as little time as possible thinking about it
  one thing I _hope_ they'll do is release an IE that has higher-fidelity standards support; maybe Mozilla can help push them in that direction
  but now
  I am not at all motivated by competing with IE
<TonyG> p
<shaver> if IE were to be taken off the market tomorrow, our mission would be no less important
  the job here is to provide a high-quality, open-source browser
<dmose> that is cross-platform
<shaver> yes
  what _might_ change our mission would be if a browser like IE was magically ported to lots of platforms and released under an open source license
  in the meantime, I think we'd best keep working. =)
<chrisn> photek: followup?
<photek> i agree with you but i tell you this because the press seem to think that version number is synonym with "better quality" (it's silly i know)
<chrisn> next questioner is MattyT:
<shaver> the numbering game is one that I think ``downstream'' companies like AOL/Netscape will be able to plpay
  play better
  let's leave that to them.
<MattyT> well open source is synonymous with "release often"
<photek> ok
<MattyT> after the original full "stable" release, what process is likely to be undertaken, will it be like linux and there be separate branches for stable and unstable releases?
<shaver> I think so
<MattyT> and how would that go with linear milestone numbers
<shaver> I would like to see the 5.0 release be followed very closely by a branching
  we may very well have to rethink our milestone numbering =)
<chrisn> MattyT: followup?
<MattyT> yep
  when we shut down features hard for alpha/beta, will there be a branch for contributors to work on still?
<shaver> I hope we don't need a branch
  when we get down into hard-core polish mode, I'd like to think that lots of developers will be able to work on extensions
<MattyT> ok so switches will be used ...
<shaver> if you're doing the IRC client or a new network protocol or MNG support, you don't have to get locked down at the same time
  I think the mozilla/extensions directory will be locked down very late indeed, if at all
  for contributors who are working on core layout and JS and widget, like lots of engineers at Netscape, they'll have to follow the lockdown
<chrisn> Waldo's up next:
<Waldo> It seems likely that there could be an official "gecko/nglayout inside" certification to brand commercial products that use the layout engine. Who would own/control such a thing?
<shaver> branding is hard. let's use the shopping button.
<dmose> that's an interesting question
<shaver> it would be nice if there was a programme for that
  but deciding what it actually means would be hard
<dmose> for one, some companies are more likely to be interested in branding a specific downstream distribution
<shaver> I think you'll see some organizations branding gecko/nglayout and providing support for those branded things
<Waldo> Maybe you'd say "Netscape(TM) Gecko compliant" but that could turn to chaos pretty quick..
<shaver> well, I wouldn't be saying ``Netscape(TM)'' anything =)
<Kovu> {maybe back to your "Standards-compliant" idea}
<dmose> Waldo: right. what do you mean by turning to chaos?
<Waldo> I mean if you had more than one version of certification.. everyone certifying their own gecko..
<MattyT> waldo: mozilla nglayout you mean =)
<shaver> I'd love to have people say ``my software kicks ass because of Mozilla technology''
  I'm just not sure how to keep that meaningful
<Kovu> {maybe a W3C-certification}
<shaver> the Mozilla license and process is designed to let people take very small parts, or the whole pile
<MattyT> shaver: it doesn't matter if it's meaningful, if it's good for the mozilla name =)
<dmose> this also gets back to our current need for marketing muscle. if any of you would care to devote time and thought to how such a mozilla branding might work, that could bge worthwhile
<shaver> it might be bad for the mozilla name!
  if someone takes the layout engine and breaks CSS with an ``optimization'', I don't want them calling that ``Mozilla Gecko''
<MattyT> but bad news is better than no news
<Waldo> I think it's important to have mozilla advocacy consistant and "on message" because the anti-mozilla forces are highly organized.
<shaver> not yet sure where to draw the line
  the anti-mozilla forces, though, are generally ill-informed
  that's our advantage: facts =)
<chrisn> true
<Waldo> except for one big one in redmond
<MattyT> there are no anti-mozilla forces, just ignorant people
<Kovu> good product = good press
<shaver> MS hasn't done much in the way of anti-Mozilla press, if that's what you mean
<MattyT> and ignorant people can be educated
<shaver> I think they're probably waiting, like everyone else, to see what we get done
<Waldo> Maybe i'm just really paranoid :)
<MattyT> and MS think Mozilla is a failure I think
<shaver> and when Mozilla 5.0 is done, I'm ready to take all comers
<Waldo> I have to assume MS is behind everything bad
<shaver> =)
<Waldo> heeh
<Kovu> they don't want to call attention to it
<chrisn> ok, folks we need to wrap it up for this evening. I'd like to thank Mike and Dan for stopping by and answering questions (and for staying later)