Slashdot Mirror
Mozilla Foundation Chief Mitchell Baker Replies
Last week we requested questions for Mozilla Chief Lizard Wrangler Mitchell Baker. A lightly-edited interview transcript is posted below. You also have the option of downloading and listening to an MP3 of the phone conversation itself.
Roblimo: This is Robin "Roblimo" Miller for Slashdot. On the other end is Mitchell Baker, the chief Lizard Wrangler for the Mozilla Foundation. How are you?
Mitchell: I'm great, thanks.
Roblimo: Not just good, great.
Mitchell: Yes, I'm great.
Roblimo: That's good to hear...
Mitchell: You know life is interesting and here we are.
Roblimo: And we heard you just got back from China because we all read your blog.
Mitchell: Ah, yes I did. I really had a great trip to China. I spent five or six months as a foreign student in China many years ago and so going back in the new era was really exciting.
Roblimo: I shouldn't wonder. Well there's a question, let's start off here with one that wasn't the first one and I don't even know if it made the final cut. But people... several wanted to know what browser do you use yourself? (Mitchell laughs) Do you use... which version of Explorer?
Mitchell: I use Firefox.
Roblimo: You don't use Explorer?
Mitchell: I don't use Explorer. In fact I rarely see Explorer... I actually used the Mozilla application suite for quite some time, far longer than many people. And eventually, since our development focus was here on Firefox, I thought I really ought to be a Firefox user as well.
Roblimo: Well, I don't disagree with you. I use it myself. That leads us into the question: What will be the testbed -- that comes from a reader who calls himself Ars-Farsica (laughter) -- his question is, or her question, we don't know, "Now that the Moz suite is apparently non-official, how will new code be tested? Will there be some sort of 'beta' Firefox release for testing? Or a new very minimal piece of code that is a testbed yet not useful to consumers?"
Mitchell: Oh, that's a good question. Let me give a little bit of background first because sometimes those of us who are deep in the project tend to use... just assume contacts that not everyone has. So the name Mozilla, or Mozilla 1.8 or 1.7 or 1.0, 1.4 has always had many different meanings and that's been confusing. So sometimes it just meant the Mozilla code base, sometimes it meant the application suite, sometimes it meant the platform underneath, and we have... we've been working in the last year or so to try and clear up that confusion and we still have... we've made some progress and we have some work to do. So when we have said Mozilla 1.0 or Mozilla 1.4, 1.7, that actually applies to two different things. One is the application layer, you know, the set of user... the things that the user sees and the user interface and the kinds of interactions that the user can have with the program.
Roblimo: What we're looking at right now on our screens.
Mitchell: Yes. We've also used Mozilla 1.0 or Mozilla 1.7 to refer to all the infrastructure technology that sits below that architecturally, but the average human being never sees and doesn't really want to know about; the networking stuff, the layout stuff, how the client talks to the server; all of those things that most consumers just want to work but are never going to interact with directly. And our shorthand for that set of code is Gecko.
So in the past, when we have done a Mozilla Seamonkey 1.4, 1.5, 1.6, 1.7 release or release candidate, we have been testing both the application layer and the Gecko layer underneath it. That will continue. What will happen now is the application layer will change from the old suite known as Seamonkey to Firefox and Thunderbird. So we will continue to do releases that test both pieces. The difference is -- as I've announced, I'm repeating myself -- is that the application layer itself will change. So to answer more specifically, we will not be shipping some minimal piece that's not useful for consumers. For us to get the kind of testing and to produce the quality of product that we need to, we have to ship something usable that those hundreds and thousands of beta testers and then millions of people will actually test and use.
Roblimo: Okay.
Mitchell: Does that sort of make sense?
Roblimo: I think so, and that leads right into a question by kollivier -- that's the Slashdot I.D. that he uses or she uses -- and says he's one of the core developers on wxMozilla and has been wondering how the GRE and Gecko SDK fits into the overall Mozilla framework, and you really just answered it, didn't you?
But his other question, which I'll come to now, says, "I have one more related question because I'm a Mac user." (laughter) "I noticed the hiring of Josh Aas to the Mozilla Foundation and a commitment to improving Mac support, which I was very excited to hear about, and I was wondering if this includes improving the embedding libraries on Mac. Modern Mac apps have significant troubles with the current embedding libraries, which are geared toward OS 8 and 9. I realize this is open source, I'm certainly willing to help in any efforts toward this end and have already made headway towards some patches, but I would need some help and support from the Mozilla project to make this real."
So I guess what he's really asking is, might this help be forthcoming at some point... Mac support?
Mitchell: Well, looking at the embedding libraries is one of the things on Josh's list. It's not currently, I think, right at the top of the list. I think there's some focus on UI bugs, toolkit-related things, maybe some of the Cairo rendering work on Josh's own priority list. But embedding is there and certainly if either this reader or anyone else actually can be helpful, we'd be very interested in that.
One of the reasons that we're excited to have Josh join us is that he's part of the open source community and understands about revealing patches and helping other people make progress with things he's not working on directly. So that turns out to be a long answer for you.
Roblimo: I'm going to add a question in here that isn't on this list of reader-generated questions, but more than one reader said, "Help, I want to help, I'm offering support." What's their best way to get involved?
Mitchell: That depends on where that person's expertise is and how advanced it is. So if this person or anyone else is actually at a point to write useful patches, then that's the single best way.
Now, of course there's a set of people who want to be involved and aren't yet accustomed to our code or need to learn more about it. For those people, there's another series of steps. Some of those -- and again it depends on what people want to do -- some people find a good way in through the QA process: defining bugs and then doing test cases and fixing them. Other people find a good way in through writing extensions.
And there's a number of people -- since the technology necessary to write extensions is a subset of our total technology pool -- who get a good handle on that and write interesting extensions that catch people's eyes and then, through that, look deeper into the product. If someone, for example, already had Toolkit or Cairo or, you know, Mac OS X experience and was able to just jump right in and do things, that's always the single best way.
Roblimo: By the way, the background noise, that was my dog, Terry, who wanted to know if there's any way to translate Mozilla into dog-ish, but I told her no.
Mitchell: Oh well, absolutely. We actually do have Sparky, who joins us here at the office quite a bit. Sparky is, I think, a little terrier and pretty under control sometimes and neurotic the rest. But everytime the phone rings, he does a fair amount of translation.
Roblimo: It sounds like Sparky and Terry would get along. I'm accused of being flippant in these interviews and not treating people with the seriousness they deserve, but (laughter)...
Mitchell: I know, but part of what we do is we try to have fun while we're doing it.
Roblimo: Well, I always thought that was the point. As I recall, Linus Torvalds' autobiography was not entitled "For the Misery of It." I thought he called it "For the Fun of It." But what do I know, right?
Mitchell: Right. But very often people ask us if we're in this for revenge or to go after Microsoft or if that's what we think about, and the answer is no. I mean, who wants to live waking up... at least I don't want to live waking up everyday about revenge. I mean, we're here to try and do something useful and help keep the Internet something fun and useful for us and have a good time while we're at it. So that's really a part of enjoying what we're doing and been able to pursue it.
Roblimo: Codemachine has a tough question titled "Integration and Lost Features." And says, "I'm wondering whether there are any plans to integrate the existing stand-alone applications, and whether this will even be possible now that the Mozilla Foundation is not doing development on all of them. For example, it might be useful to see integration between two or more of Firefox, Thunderbird, Sunbird, Nvu, the address book, and a chat component. But since the Mozilla Foundation does not develop Nvu or Chatzilla any longer (is anyone working on Chatzilla or any XUL chat app any longer?), this won't necessarily be easy. So is there any plan to work with Linspire and other application developers to integrate their work with Firefox and Thunderbird? Will the Mozilla Foundation be doing official extensions that bring some of the suite functionality to the stand-alone products?"
Mitchell: Okay. Well there's a set of questions in there. Let me get started and see how far I can get.
Roblimo: Well, there's a bunch of questions that more or less ask that. So I just threw them all into one, as it were.
Mitchell: Okay. So, it may well be the case that one option for delivering Mozilla products is to have a combined Firefox, Thunderbird, maybe a calendar, who knows. Some combination of products that the user gets and installs only once and gets a set. That may well be the case. We're not sure yet; some people want it, many people seem to be happy currently with getting the browser separately. The goal of Firefox and Thunderbird is a new architecture which allows those products to be built separately, to remain separate products. But that is separate from the packaging and delivery mechanism. So we do have in mind that it might make sense at some point to package these and deliver them together. I think that's one of the questions.
Roblimo: And the other one is...
Mitchell: The second question would be, what happens to some of the functionality that used to be in the suite or the stand-alone applications in the suite? So there is a fair amount of work going on with Nvu. Linspire has been active in funding and promoting that work, and Daniel Glazman, an old Netscape and Mozilla contributor, has been doing the work. And we are in close contact with both of those, so when Nvu reaches a 1.0 level we will look and see whether it makes sense to make it a product. It's a very big job, to take a project, you know, and make it into a product and if at some point it makes sense, if we do an integrated bundle, then we would certainly look at Nvu.
Roblimo: Okay.
Mitchell: I don't think that ChatZilla is getting the same kind of attention that Nvu is currently, at least not to my knowledge. So we'll see, we'll see what happens there. I'm still using ChatZilla pretty happily.
Roblimo: I know a number of people including Timothy Lord, Timothy on Slashdot. He uses and loves ChatZilla.
Mitchell: Yes, ChatZilla's really a great example of how you can write an application using XUL that doesn't look at all like a browser, and it can be simple and effective.
Let's see, there was something else in there...
Roblimo: Well, I think we've...
Mitchell: And the third piece, I think -- 'cause this does relate to the Seamonkey suite, is a set of functionality that exists in the Seamonkey suite that does not currently exist in Firefox or Thunderbird. And that the set of people who are very happy using the Seamonkey suite and don't want to switch because they like some of this functionality.
And so there's been a set of questions of, "I want this set of two or three features in Firefox and why isn't it there?" And the reason many of these features aren't there relates to the core design philosophy of Firefox and part of why we think it's successful. One of the long-standing complaints about the application suite is that it's giant, or it's bloated, or it's for power users, or it's got many features that maybe ten percent of the world wants but the other ninety doesn't. And if you have a lot of features that ten percent of the world wants, pretty soon you have a giant program of which any one person is using only a small portion. So our design philosophy for Firefox is to keep our default distribution of Firefox very slim; to have it be this lean product that encapsulates what most people need. And if you want some additional features, rather than having us include them in Firefox where everybody gets them, we use the extension mechanism. So the individual user can go and get the extensions for the features that you want, and you and ten or maybe twelve or fifteen or twenty percent of the world really needs to be effective, but the other eighty percent doesn't. So the answer for those people is there's a set of functionality or preferences in Seamonkey which probably will not show up in a default Firefox distribution. And those features should be made available through extensions.
Roblimo: So there are gonna be... they will or can be made available through extensions? If anybody wants to take up the ball and keep 'em and maintain 'em and love 'em.
Mitchell: Exactly, exactly. And I could certainly see, you know, it's possible that some extensions would eventually become part of the default if they prove to be really important to a lot of people. Tabbed browsing is an example of that, but until that case, we would use the extension mechanism. We worked very hard to make extensions very simple.
Roblimo: Let's move on to a totally different topic. The most popular Slashdot reader, Anonymous Coward, asks, "When will there be a real effort to support SVG -- that is, Scalable Vector Graphics -- and have it turned on in the builds by default?"
Mitchell: Well, there actually is a real effort underway right now with the development effort and a fair amount of planning and thinking. The question that we've looked at is the SVG standard itself is a giant standard and particularly the new version, contradictory to existing standards. And so we have struggled for quite a while to figure out how to integrate the parts of SVG that we need, given that it's part of the standard which contradicts CSS and various other issues. And so that's part of the reason why it's taken some time to figure out what to do here. So I guess the short answer to that is the work is actually going on. We're actively trying to figure out which part of SVG, what do we do with this giant standard? And we're also in the process of trying to figure out when does it get turned on, or how does it get turned on, but I don't have the dates for those now.
Roblimo: Alright. I'm gonna move on and I'm just gonna take a little bit of another question from kollivier. And this is something that came up -- a lot of the questions that were not moderated up to five, but it was a repeated question. Essentially it boils down to, "Yo, we have a whole lot of Mozilla suite installations in our company." Some of them are talking in many thousands of desktops.
Mitchell: Yes.
Roblimo: And what's up with that? "There's gonna be no Mozilla 1.8 and basically we're screwed. Help! What's gonna happen, what should we do?"
Mitchell: Okay. Yeah, this is a good question. Because any time there's a shift from one product line to another product line, it's a difficult and complex set of issues to work through. And there's no perfect way to say, "We're focusing on what we think is the correct future for the project, the new application," and still do everything with the old product line that makes everyone happy. So one thing we've said for quite some time is that the 1.7 releases will be a long-term supported set of releases. This is in with our history. When we released Mozilla 1.0, we said that it would be supported for a long term, that one we supported for about a year.
Roblimo: Let's define long-term.
Mitchell: Yeah. The Mozilla 1.0 was a year.
Roblimo: Okay.
Mitchell: Mozilla 1.4 was our next long-term supported branch, that's been two years. We either just did or will do in the next few days a release on that branch. So that one's been two years. 1.7 is likely to be in that range. And by supported, we mean maintenance and security releases as needed. Plus, we will maintain all of the infrastructure for that product as we always have. This is not the development of needed things on 1.7 and stopping. New feature work will not be occurring on the 1.7 branch. Security and maintenance releases that that customer base needs are part of the plan of record.
Roblimo: Will you consider extending it beyond two years, considering the level of enterprise penetration there is?
Mitchell: We would certainly consider that. It depends on who's still using it two years from now, which distributors are interested in it, and what kind of support they choose to give.
But I think it's always difficult when a product that you're using and accustomed to changes. The things that are different in our case are, well, certainly 1.7 will be supported for some time. Anyone who's actually interested in that branch, 1.7, and what happens to it is welcome to get involved. Now of course, many enterprises do not want to get involved in that, they want someone else to do that. And there certainly are some people who are distributing, particularly in the Linux side, 1.7, who may well choose to maintain and provide support beyond that. Within that several-year time frame, if customers are interested in new features and maintenance, they should look at the new products and think about a migration strategy.
So we recognize that someone using the suite might well want to stay with that suite forever. At the same time, we cannot develop the application suite and the new product lines forever, and do so effectively. So we have new products, which have been well received. We have a commitment to provide some support and we've been working with a key set of early adopters on a migration plan and looking at their issues to see, "Are there any issues with migrating?"
Roblimo: Okay, that should help them out. But right now we're saying two years, maybe if there's a user base that needs it, more, if it's appropriate. That sound about right?
Mitchell: Yeah. I would say that on a timeline, we continue to evaluate who's using things, how many people are out there, what do they really need.
Roblimo: Okay. Let's go on to this one here. I'm not going to read this whole question, the guy has a lot of citations in "Firefox drive-wiping bug took one year to fix?" from user chris59256. "Recently," he says, "I learned of a bug in Windows Firefox versions prior to 1.01 which was fixed in this version. This bug wipes users' hard disks. I've located 15 users who've suffered from this bug. Why did it take over one year to fix this serious bug?"
I'll give you one more paragraph, a short one: "The bug only occurs when a user uninstalls Firefox. A user who uninstalls version 1.0 to prepare for installing version 1.01 is vulnerable. Why has the Firefox homepage not been updated to warn all users about this fact, and offer a safe remedy?"
Mitchell: To give an answer to this specific bug, I would need to go back and review this specific bug and all the decisions that were made because I'm not following bug by bug, you know, on this issue. So the reader may be dissatisfied with the answer, but the answers for bugs that are serious, of this type... If they're not fixed right away, it's usually hard to reproduce. In, I don't know, however many millions, thirty million people, however many people have used Firefox, those fifteen people got a very bad experience and we intend to fix it. But it may well be very hard to find and reproduce. Now I'm not saying that's actually the case in this bug because I haven't gone through all the specifics. So it may be that. It may be that it happened only in a very precise set of circumstances, with certain configurations or certain flags or certain other issues that happened. In which case, sorting through those and figuring those out can be very difficult.
Roblimo: Yeah, He's got all his cites and bugzilla over and over again but I think the real question came down to responsiveness. He says, "The bug was reported in bugzilla and discussed without fixing for over one year. At one point a developer didn't remove the dangerous code because he said," and the questioner is quoting here, "This is not an acceptable solution to force on all users because some people make bad assumptions and then don't read dialogs." Now we're just taking his quote, we don't know where he got it, but he continues, "Is Firefox truly ready for 'the masses' when developers maintain this sort of attitude towards users?"
Mitchell: Well, I'm not sure if it is. So one developer has said... sounds to me like there's no easy answer to this. And in looking at it, one-two is figured out. One solution was to do x and it forces on thirty million people a response that might not make sense and might be a bad experience for them to help these fifteen people. That might be the right balance, it might not. That seems to be what this developer, what this quote is saying. And clearly, the person writing in disagrees with that attitude. So I'm not exactly sure; it may be a responsiveness question, it is certainly the case that we are not as responsive as we would like to be in each and every instance. I'm not sure if anybody can be perfect in that regard, but we certainly aren't.
So it's possible it's responsiveness, but I suspect that if it's something that really relates to people's hard disks, it's a question of figuring out what the problem is and figuring out a solution for that subset of people that doesn't have negative implications for a much larger set and then balancing how much negative implication for the larger set makes sense to avoid this problem for a small number of people. I mean, I'm guessing that that's the kind of discussion that's going on here.
Roblimo: It seems to be part of the chronic open source question, whose itch is being scratched. But let's move on anyway.
Mitchell: Well, you know, and that might be it. And if I investigated it, maybe that's what we would come up with. But I don't think that's us because this kind of issue... it's not... that is our itch. So I suspect that there's more to it. I don't think that the developer or our QA folks were uninterested in this stuff. I suspect that there's a level of complexity in here that isn't clear through the quotes and that figuring out the solution and how to implement it and what's the right solution for everyone is not as simple as it appears. I'm not claiming we've done a perfect job here either, because I don't know all the facts, but I don't think that it's just that we're not interested in something like this.
Roblimo: Well, Tuxedo Jack seems to think you're doing a pretty good job. In fact, in his question under the headline, "Corporate-scale Firefox usage," he says this: "Given how easy it is to deploy Firefox across a campus (load it into a Ghost loadset, then deploy in your next periodic reclone), why, in your opinion, are medium-to-large companies loathe to deploy anything but IE, especially given the tendencies of employees to use office machines for distinctly non-work purposes, which often lead to malware infections?"
Mitchell: Well, our experience has been that there are many companies who recognize that their internal content or their intranet or their extranet or whatever, includes or might include IE-specific content. And that in the past, when it wasn't so clear why IE-specific content is so dangerous, many companies ended up either intentionally or because they didn't try hard not to, with IE-specific content. And so now they're worried, well, "Boy, we're interested in deploying Firefox, now we can see the advantages to it. But how do we figure out what content might not work in that case?" But it turns out that public sites are pretty standards-compliant. There's been a chunk of work done at those sites and many of them, maybe some of the banking sites are an exception, but the vast majority of sites work great in Firefox. It turns out that the private internal intranets aren't nearly that far along because they don't have the pressure of the public coming to them. And so companies go round and round and round in trying to figure out, "I'd like to do something new and get out of the track I'm on," but on the other hand, "What do I do? How do I figure out what content doesn't work and know how to fix it?" That's what I think.
Roblimo: And so starting in about 1997, both Jeffrey Zeldman and I both wrote all kinds of articles warning people about getting tied too tightly to Netscape-specific functions rather than being standards-compliant.
Mitchell: Yes.
Roblimo: And you'd think that some of these companies would have learned their lesson then, but I guess not, eh? (laughing)
Mitchell: Well, there's a couple of things going on. One is that there's a question between, you know, moving, developing new things and standards.
Roblimo: Yes.
Mitchell: So there is, I think, a legitimate issue there. And the standards process can slow down development and implementation of new things. But I think the belief that it almost didn't matter and the use of tools that generated IE-specific content without thinking about it, or generating it because ActiveX was so convenient. You know, those are approaches, the dangers of which have now been realized.
Roblimo: Yes. So let's talk about calendars. EvilStein wants to know, "Are there any more plans to put weight behind the calendaring solution? I know that Sunbird exists and there's now Lightning, but the project details are quite vague. The Mozilla Suite could benefit greatly from a fully functional calendar, especially in the small business realm."
So is there any work actively being done on calendaring within Mozilla, the Mozilla organization or Foundation?
Mitchell: There is work being done on Lightning and the Mozilla Foundation provides the infrastructure support for that work -- both Lightning and Thunderbird. The actual coding and development is not being done by Foundation employees. Our mail folks work with those people. So currently the actual -- I think that question might be about the Foundation -- so the Foundation is providing infrastructural development and support for the calendar project. There is work actively going on, architectural and coding and design work all going on, on the calendar solution.
Roblimo: Okay. fm6 asks... let's talk about money for a minute. Money's always good. "What's the significance of Ben Goodger switching his employment from Mozilla Foundation to Google? Is this just a device to offload some of your payroll costs?"
Mitchell: (laughing) Well, the Mozilla project has always had people employed by different employers. It's actually been part of the stated goal for quite some time, so to us that's actually nothing new. You know, there's a chunk of people employed by Netscape for a long time, and Sun, and IBM, and Red Hat all have employed people. Novell employs people and now Google employs people. So it doesn't actually have a significance to us, I think, that it might to other people as well.
Roblimo: Now let's go to our last question, about funding. Could you explain how the Mozilla Foundation currently gets its funding and what your vision is on the long-term funding for open source projects like Mozilla?
Mitchell: Sure. We generate funds in several different ways. There's the classic charitable contribution, which we receive thousands, and we're extremely grateful and they often come with notes from people, which are very heartwarming, about how much difference our products have made in their life on the Internet. Of course, it's hard to support full-time programmers, so we do get funds from a set of companies that are interested in the health of the Mozilla project and so are willing to support the people working for the Foundation as well.
We have generated funds through some -- what's the word I'm looking for -- directed engineering, let's say with Thunderbird, particularly a set of enterprise features that Thunderbird needed. So funding from that funds some of our developers. There's also some engineering and related work that can generate funds and we have arrangements with some of the search providers in the browser itself, which will generate funds as well.
Roblimo: Which search providers?
Mitchell: Well, there's a set. You know, you can see the setup there.
Roblimo: Actually, I've customized mine so hard that I can't see it. (laughter) I'm sorry! I'm one of those...
Mitchell: We come with a default set of five, and that's Google and Yahoo and Amazon and eBay and Creative Commons and some subsets of those.
Roblimo: And as far as companies that are giving financial support, Microsoft and who else?
Mitchell: (laughing) Yeah, right. Microsoft?
Roblimo: (laughing) Well, I like to end things with a little humor.
Mitchell: But actually, I wanted to come back to the money piece, and money and funding and support of open source projects, because I do think that this is a topic that I certainly want to talk about more in the future. And I think other, you know, many people will, because a project like Mozilla really needs people working full time on it.
Roblimo: Yes.
Mitchell: And that means someone has to provide funds. You have to raise funds somehow. And once there's money in the picture, money is great because we all need it to live on. Money tends to make people suspicious, if there's any money floating around. So I think one of the issues that open source projects, including Mozilla, will have to look at in the future is: given the need to generate enough funds to sustain ourselves and pay people, i.e. the need to generate funds, how does that change the dynamics of the project? I don't know the answer to that, but I do know that if you're generating any funds, people tend to get suspicious of you and maybe rightly so. And so I think there's gonna be a whole round of discussion on exactly that.
Roblimo: Thank you very much for being our guest on Slashdot, Mitchell Baker. And I'm Robin Miller, known to far too many people as Roblimo, and thank you all for listening.
Mitchell: Well, thank you! I hope we can do it again before too long.
Roblimo: This is Robin "Roblimo" Miller for Slashdot. On the other end is Mitchell Baker, the chief Lizard Wrangler for the Mozilla Foundation. How are you?
Mitchell: I'm great, thanks.
Roblimo: Not just good, great.
Mitchell: Yes, I'm great.
Roblimo: That's good to hear...
Mitchell: You know life is interesting and here we are.
Roblimo: And we heard you just got back from China because we all read your blog.
Mitchell: Ah, yes I did. I really had a great trip to China. I spent five or six months as a foreign student in China many years ago and so going back in the new era was really exciting.
Roblimo: I shouldn't wonder. Well there's a question, let's start off here with one that wasn't the first one and I don't even know if it made the final cut. But people... several wanted to know what browser do you use yourself? (Mitchell laughs) Do you use... which version of Explorer?
Mitchell: I use Firefox.
Roblimo: You don't use Explorer?
Mitchell: I don't use Explorer. In fact I rarely see Explorer... I actually used the Mozilla application suite for quite some time, far longer than many people. And eventually, since our development focus was here on Firefox, I thought I really ought to be a Firefox user as well.
Roblimo: Well, I don't disagree with you. I use it myself. That leads us into the question: What will be the testbed -- that comes from a reader who calls himself Ars-Farsica (laughter) -- his question is, or her question, we don't know, "Now that the Moz suite is apparently non-official, how will new code be tested? Will there be some sort of 'beta' Firefox release for testing? Or a new very minimal piece of code that is a testbed yet not useful to consumers?"
Mitchell: Oh, that's a good question. Let me give a little bit of background first because sometimes those of us who are deep in the project tend to use... just assume contacts that not everyone has. So the name Mozilla, or Mozilla 1.8 or 1.7 or 1.0, 1.4 has always had many different meanings and that's been confusing. So sometimes it just meant the Mozilla code base, sometimes it meant the application suite, sometimes it meant the platform underneath, and we have... we've been working in the last year or so to try and clear up that confusion and we still have... we've made some progress and we have some work to do. So when we have said Mozilla 1.0 or Mozilla 1.4, 1.7, that actually applies to two different things. One is the application layer, you know, the set of user... the things that the user sees and the user interface and the kinds of interactions that the user can have with the program.
Roblimo: What we're looking at right now on our screens.
Mitchell: Yes. We've also used Mozilla 1.0 or Mozilla 1.7 to refer to all the infrastructure technology that sits below that architecturally, but the average human being never sees and doesn't really want to know about; the networking stuff, the layout stuff, how the client talks to the server; all of those things that most consumers just want to work but are never going to interact with directly. And our shorthand for that set of code is Gecko.
So in the past, when we have done a Mozilla Seamonkey 1.4, 1.5, 1.6, 1.7 release or release candidate, we have been testing both the application layer and the Gecko layer underneath it. That will continue. What will happen now is the application layer will change from the old suite known as Seamonkey to Firefox and Thunderbird. So we will continue to do releases that test both pieces. The difference is -- as I've announced, I'm repeating myself -- is that the application layer itself will change. So to answer more specifically, we will not be shipping some minimal piece that's not useful for consumers. For us to get the kind of testing and to produce the quality of product that we need to, we have to ship something usable that those hundreds and thousands of beta testers and then millions of people will actually test and use.
Roblimo: Okay.
Mitchell: Does that sort of make sense?
Roblimo: I think so, and that leads right into a question by kollivier -- that's the Slashdot I.D. that he uses or she uses -- and says he's one of the core developers on wxMozilla and has been wondering how the GRE and Gecko SDK fits into the overall Mozilla framework, and you really just answered it, didn't you?
But his other question, which I'll come to now, says, "I have one more related question because I'm a Mac user." (laughter) "I noticed the hiring of Josh Aas to the Mozilla Foundation and a commitment to improving Mac support, which I was very excited to hear about, and I was wondering if this includes improving the embedding libraries on Mac. Modern Mac apps have significant troubles with the current embedding libraries, which are geared toward OS 8 and 9. I realize this is open source, I'm certainly willing to help in any efforts toward this end and have already made headway towards some patches, but I would need some help and support from the Mozilla project to make this real."
So I guess what he's really asking is, might this help be forthcoming at some point... Mac support?
Mitchell: Well, looking at the embedding libraries is one of the things on Josh's list. It's not currently, I think, right at the top of the list. I think there's some focus on UI bugs, toolkit-related things, maybe some of the Cairo rendering work on Josh's own priority list. But embedding is there and certainly if either this reader or anyone else actually can be helpful, we'd be very interested in that.
One of the reasons that we're excited to have Josh join us is that he's part of the open source community and understands about revealing patches and helping other people make progress with things he's not working on directly. So that turns out to be a long answer for you.
Roblimo: I'm going to add a question in here that isn't on this list of reader-generated questions, but more than one reader said, "Help, I want to help, I'm offering support." What's their best way to get involved?
Mitchell: That depends on where that person's expertise is and how advanced it is. So if this person or anyone else is actually at a point to write useful patches, then that's the single best way.
Now, of course there's a set of people who want to be involved and aren't yet accustomed to our code or need to learn more about it. For those people, there's another series of steps. Some of those -- and again it depends on what people want to do -- some people find a good way in through the QA process: defining bugs and then doing test cases and fixing them. Other people find a good way in through writing extensions.
And there's a number of people -- since the technology necessary to write extensions is a subset of our total technology pool -- who get a good handle on that and write interesting extensions that catch people's eyes and then, through that, look deeper into the product. If someone, for example, already had Toolkit or Cairo or, you know, Mac OS X experience and was able to just jump right in and do things, that's always the single best way.
Roblimo: By the way, the background noise, that was my dog, Terry, who wanted to know if there's any way to translate Mozilla into dog-ish, but I told her no.
Mitchell: Oh well, absolutely. We actually do have Sparky, who joins us here at the office quite a bit. Sparky is, I think, a little terrier and pretty under control sometimes and neurotic the rest. But everytime the phone rings, he does a fair amount of translation.
Roblimo: It sounds like Sparky and Terry would get along. I'm accused of being flippant in these interviews and not treating people with the seriousness they deserve, but (laughter)...
Mitchell: I know, but part of what we do is we try to have fun while we're doing it.
Roblimo: Well, I always thought that was the point. As I recall, Linus Torvalds' autobiography was not entitled "For the Misery of It." I thought he called it "For the Fun of It." But what do I know, right?
Mitchell: Right. But very often people ask us if we're in this for revenge or to go after Microsoft or if that's what we think about, and the answer is no. I mean, who wants to live waking up... at least I don't want to live waking up everyday about revenge. I mean, we're here to try and do something useful and help keep the Internet something fun and useful for us and have a good time while we're at it. So that's really a part of enjoying what we're doing and been able to pursue it.
Roblimo: Codemachine has a tough question titled "Integration and Lost Features." And says, "I'm wondering whether there are any plans to integrate the existing stand-alone applications, and whether this will even be possible now that the Mozilla Foundation is not doing development on all of them. For example, it might be useful to see integration between two or more of Firefox, Thunderbird, Sunbird, Nvu, the address book, and a chat component. But since the Mozilla Foundation does not develop Nvu or Chatzilla any longer (is anyone working on Chatzilla or any XUL chat app any longer?), this won't necessarily be easy. So is there any plan to work with Linspire and other application developers to integrate their work with Firefox and Thunderbird? Will the Mozilla Foundation be doing official extensions that bring some of the suite functionality to the stand-alone products?"
Mitchell: Okay. Well there's a set of questions in there. Let me get started and see how far I can get.
Roblimo: Well, there's a bunch of questions that more or less ask that. So I just threw them all into one, as it were.
Mitchell: Okay. So, it may well be the case that one option for delivering Mozilla products is to have a combined Firefox, Thunderbird, maybe a calendar, who knows. Some combination of products that the user gets and installs only once and gets a set. That may well be the case. We're not sure yet; some people want it, many people seem to be happy currently with getting the browser separately. The goal of Firefox and Thunderbird is a new architecture which allows those products to be built separately, to remain separate products. But that is separate from the packaging and delivery mechanism. So we do have in mind that it might make sense at some point to package these and deliver them together. I think that's one of the questions.
Roblimo: And the other one is...
Mitchell: The second question would be, what happens to some of the functionality that used to be in the suite or the stand-alone applications in the suite? So there is a fair amount of work going on with Nvu. Linspire has been active in funding and promoting that work, and Daniel Glazman, an old Netscape and Mozilla contributor, has been doing the work. And we are in close contact with both of those, so when Nvu reaches a 1.0 level we will look and see whether it makes sense to make it a product. It's a very big job, to take a project, you know, and make it into a product and if at some point it makes sense, if we do an integrated bundle, then we would certainly look at Nvu.
Roblimo: Okay.
Mitchell: I don't think that ChatZilla is getting the same kind of attention that Nvu is currently, at least not to my knowledge. So we'll see, we'll see what happens there. I'm still using ChatZilla pretty happily.
Roblimo: I know a number of people including Timothy Lord, Timothy on Slashdot. He uses and loves ChatZilla.
Mitchell: Yes, ChatZilla's really a great example of how you can write an application using XUL that doesn't look at all like a browser, and it can be simple and effective.
Let's see, there was something else in there...
Roblimo: Well, I think we've...
Mitchell: And the third piece, I think -- 'cause this does relate to the Seamonkey suite, is a set of functionality that exists in the Seamonkey suite that does not currently exist in Firefox or Thunderbird. And that the set of people who are very happy using the Seamonkey suite and don't want to switch because they like some of this functionality.
And so there's been a set of questions of, "I want this set of two or three features in Firefox and why isn't it there?" And the reason many of these features aren't there relates to the core design philosophy of Firefox and part of why we think it's successful. One of the long-standing complaints about the application suite is that it's giant, or it's bloated, or it's for power users, or it's got many features that maybe ten percent of the world wants but the other ninety doesn't. And if you have a lot of features that ten percent of the world wants, pretty soon you have a giant program of which any one person is using only a small portion. So our design philosophy for Firefox is to keep our default distribution of Firefox very slim; to have it be this lean product that encapsulates what most people need. And if you want some additional features, rather than having us include them in Firefox where everybody gets them, we use the extension mechanism. So the individual user can go and get the extensions for the features that you want, and you and ten or maybe twelve or fifteen or twenty percent of the world really needs to be effective, but the other eighty percent doesn't. So the answer for those people is there's a set of functionality or preferences in Seamonkey which probably will not show up in a default Firefox distribution. And those features should be made available through extensions.
Roblimo: So there are gonna be... they will or can be made available through extensions? If anybody wants to take up the ball and keep 'em and maintain 'em and love 'em.
Mitchell: Exactly, exactly. And I could certainly see, you know, it's possible that some extensions would eventually become part of the default if they prove to be really important to a lot of people. Tabbed browsing is an example of that, but until that case, we would use the extension mechanism. We worked very hard to make extensions very simple.
Roblimo: Let's move on to a totally different topic. The most popular Slashdot reader, Anonymous Coward, asks, "When will there be a real effort to support SVG -- that is, Scalable Vector Graphics -- and have it turned on in the builds by default?"
Mitchell: Well, there actually is a real effort underway right now with the development effort and a fair amount of planning and thinking. The question that we've looked at is the SVG standard itself is a giant standard and particularly the new version, contradictory to existing standards. And so we have struggled for quite a while to figure out how to integrate the parts of SVG that we need, given that it's part of the standard which contradicts CSS and various other issues. And so that's part of the reason why it's taken some time to figure out what to do here. So I guess the short answer to that is the work is actually going on. We're actively trying to figure out which part of SVG, what do we do with this giant standard? And we're also in the process of trying to figure out when does it get turned on, or how does it get turned on, but I don't have the dates for those now.
Roblimo: Alright. I'm gonna move on and I'm just gonna take a little bit of another question from kollivier. And this is something that came up -- a lot of the questions that were not moderated up to five, but it was a repeated question. Essentially it boils down to, "Yo, we have a whole lot of Mozilla suite installations in our company." Some of them are talking in many thousands of desktops.
Mitchell: Yes.
Roblimo: And what's up with that? "There's gonna be no Mozilla 1.8 and basically we're screwed. Help! What's gonna happen, what should we do?"
Mitchell: Okay. Yeah, this is a good question. Because any time there's a shift from one product line to another product line, it's a difficult and complex set of issues to work through. And there's no perfect way to say, "We're focusing on what we think is the correct future for the project, the new application," and still do everything with the old product line that makes everyone happy. So one thing we've said for quite some time is that the 1.7 releases will be a long-term supported set of releases. This is in with our history. When we released Mozilla 1.0, we said that it would be supported for a long term, that one we supported for about a year.
Roblimo: Let's define long-term.
Mitchell: Yeah. The Mozilla 1.0 was a year.
Roblimo: Okay.
Mitchell: Mozilla 1.4 was our next long-term supported branch, that's been two years. We either just did or will do in the next few days a release on that branch. So that one's been two years. 1.7 is likely to be in that range. And by supported, we mean maintenance and security releases as needed. Plus, we will maintain all of the infrastructure for that product as we always have. This is not the development of needed things on 1.7 and stopping. New feature work will not be occurring on the 1.7 branch. Security and maintenance releases that that customer base needs are part of the plan of record.
Roblimo: Will you consider extending it beyond two years, considering the level of enterprise penetration there is?
Mitchell: We would certainly consider that. It depends on who's still using it two years from now, which distributors are interested in it, and what kind of support they choose to give.
But I think it's always difficult when a product that you're using and accustomed to changes. The things that are different in our case are, well, certainly 1.7 will be supported for some time. Anyone who's actually interested in that branch, 1.7, and what happens to it is welcome to get involved. Now of course, many enterprises do not want to get involved in that, they want someone else to do that. And there certainly are some people who are distributing, particularly in the Linux side, 1.7, who may well choose to maintain and provide support beyond that. Within that several-year time frame, if customers are interested in new features and maintenance, they should look at the new products and think about a migration strategy.
So we recognize that someone using the suite might well want to stay with that suite forever. At the same time, we cannot develop the application suite and the new product lines forever, and do so effectively. So we have new products, which have been well received. We have a commitment to provide some support and we've been working with a key set of early adopters on a migration plan and looking at their issues to see, "Are there any issues with migrating?"
Roblimo: Okay, that should help them out. But right now we're saying two years, maybe if there's a user base that needs it, more, if it's appropriate. That sound about right?
Mitchell: Yeah. I would say that on a timeline, we continue to evaluate who's using things, how many people are out there, what do they really need.
Roblimo: Okay. Let's go on to this one here. I'm not going to read this whole question, the guy has a lot of citations in "Firefox drive-wiping bug took one year to fix?" from user chris59256. "Recently," he says, "I learned of a bug in Windows Firefox versions prior to 1.01 which was fixed in this version. This bug wipes users' hard disks. I've located 15 users who've suffered from this bug. Why did it take over one year to fix this serious bug?"
I'll give you one more paragraph, a short one: "The bug only occurs when a user uninstalls Firefox. A user who uninstalls version 1.0 to prepare for installing version 1.01 is vulnerable. Why has the Firefox homepage not been updated to warn all users about this fact, and offer a safe remedy?"
Mitchell: To give an answer to this specific bug, I would need to go back and review this specific bug and all the decisions that were made because I'm not following bug by bug, you know, on this issue. So the reader may be dissatisfied with the answer, but the answers for bugs that are serious, of this type... If they're not fixed right away, it's usually hard to reproduce. In, I don't know, however many millions, thirty million people, however many people have used Firefox, those fifteen people got a very bad experience and we intend to fix it. But it may well be very hard to find and reproduce. Now I'm not saying that's actually the case in this bug because I haven't gone through all the specifics. So it may be that. It may be that it happened only in a very precise set of circumstances, with certain configurations or certain flags or certain other issues that happened. In which case, sorting through those and figuring those out can be very difficult.
Roblimo: Yeah, He's got all his cites and bugzilla over and over again but I think the real question came down to responsiveness. He says, "The bug was reported in bugzilla and discussed without fixing for over one year. At one point a developer didn't remove the dangerous code because he said," and the questioner is quoting here, "This is not an acceptable solution to force on all users because some people make bad assumptions and then don't read dialogs." Now we're just taking his quote, we don't know where he got it, but he continues, "Is Firefox truly ready for 'the masses' when developers maintain this sort of attitude towards users?"
Mitchell: Well, I'm not sure if it is. So one developer has said... sounds to me like there's no easy answer to this. And in looking at it, one-two is figured out. One solution was to do x and it forces on thirty million people a response that might not make sense and might be a bad experience for them to help these fifteen people. That might be the right balance, it might not. That seems to be what this developer, what this quote is saying. And clearly, the person writing in disagrees with that attitude. So I'm not exactly sure; it may be a responsiveness question, it is certainly the case that we are not as responsive as we would like to be in each and every instance. I'm not sure if anybody can be perfect in that regard, but we certainly aren't.
So it's possible it's responsiveness, but I suspect that if it's something that really relates to people's hard disks, it's a question of figuring out what the problem is and figuring out a solution for that subset of people that doesn't have negative implications for a much larger set and then balancing how much negative implication for the larger set makes sense to avoid this problem for a small number of people. I mean, I'm guessing that that's the kind of discussion that's going on here.
Roblimo: It seems to be part of the chronic open source question, whose itch is being scratched. But let's move on anyway.
Mitchell: Well, you know, and that might be it. And if I investigated it, maybe that's what we would come up with. But I don't think that's us because this kind of issue... it's not... that is our itch. So I suspect that there's more to it. I don't think that the developer or our QA folks were uninterested in this stuff. I suspect that there's a level of complexity in here that isn't clear through the quotes and that figuring out the solution and how to implement it and what's the right solution for everyone is not as simple as it appears. I'm not claiming we've done a perfect job here either, because I don't know all the facts, but I don't think that it's just that we're not interested in something like this.
Roblimo: Well, Tuxedo Jack seems to think you're doing a pretty good job. In fact, in his question under the headline, "Corporate-scale Firefox usage," he says this: "Given how easy it is to deploy Firefox across a campus (load it into a Ghost loadset, then deploy in your next periodic reclone), why, in your opinion, are medium-to-large companies loathe to deploy anything but IE, especially given the tendencies of employees to use office machines for distinctly non-work purposes, which often lead to malware infections?"
Mitchell: Well, our experience has been that there are many companies who recognize that their internal content or their intranet or their extranet or whatever, includes or might include IE-specific content. And that in the past, when it wasn't so clear why IE-specific content is so dangerous, many companies ended up either intentionally or because they didn't try hard not to, with IE-specific content. And so now they're worried, well, "Boy, we're interested in deploying Firefox, now we can see the advantages to it. But how do we figure out what content might not work in that case?" But it turns out that public sites are pretty standards-compliant. There's been a chunk of work done at those sites and many of them, maybe some of the banking sites are an exception, but the vast majority of sites work great in Firefox. It turns out that the private internal intranets aren't nearly that far along because they don't have the pressure of the public coming to them. And so companies go round and round and round in trying to figure out, "I'd like to do something new and get out of the track I'm on," but on the other hand, "What do I do? How do I figure out what content doesn't work and know how to fix it?" That's what I think.
Roblimo: And so starting in about 1997, both Jeffrey Zeldman and I both wrote all kinds of articles warning people about getting tied too tightly to Netscape-specific functions rather than being standards-compliant.
Mitchell: Yes.
Roblimo: And you'd think that some of these companies would have learned their lesson then, but I guess not, eh? (laughing)
Mitchell: Well, there's a couple of things going on. One is that there's a question between, you know, moving, developing new things and standards.
Roblimo: Yes.
Mitchell: So there is, I think, a legitimate issue there. And the standards process can slow down development and implementation of new things. But I think the belief that it almost didn't matter and the use of tools that generated IE-specific content without thinking about it, or generating it because ActiveX was so convenient. You know, those are approaches, the dangers of which have now been realized.
Roblimo: Yes. So let's talk about calendars. EvilStein wants to know, "Are there any more plans to put weight behind the calendaring solution? I know that Sunbird exists and there's now Lightning, but the project details are quite vague. The Mozilla Suite could benefit greatly from a fully functional calendar, especially in the small business realm."
So is there any work actively being done on calendaring within Mozilla, the Mozilla organization or Foundation?
Mitchell: There is work being done on Lightning and the Mozilla Foundation provides the infrastructure support for that work -- both Lightning and Thunderbird. The actual coding and development is not being done by Foundation employees. Our mail folks work with those people. So currently the actual -- I think that question might be about the Foundation -- so the Foundation is providing infrastructural development and support for the calendar project. There is work actively going on, architectural and coding and design work all going on, on the calendar solution.
Roblimo: Okay. fm6 asks... let's talk about money for a minute. Money's always good. "What's the significance of Ben Goodger switching his employment from Mozilla Foundation to Google? Is this just a device to offload some of your payroll costs?"
Mitchell: (laughing) Well, the Mozilla project has always had people employed by different employers. It's actually been part of the stated goal for quite some time, so to us that's actually nothing new. You know, there's a chunk of people employed by Netscape for a long time, and Sun, and IBM, and Red Hat all have employed people. Novell employs people and now Google employs people. So it doesn't actually have a significance to us, I think, that it might to other people as well.
Roblimo: Now let's go to our last question, about funding. Could you explain how the Mozilla Foundation currently gets its funding and what your vision is on the long-term funding for open source projects like Mozilla?
Mitchell: Sure. We generate funds in several different ways. There's the classic charitable contribution, which we receive thousands, and we're extremely grateful and they often come with notes from people, which are very heartwarming, about how much difference our products have made in their life on the Internet. Of course, it's hard to support full-time programmers, so we do get funds from a set of companies that are interested in the health of the Mozilla project and so are willing to support the people working for the Foundation as well.
We have generated funds through some -- what's the word I'm looking for -- directed engineering, let's say with Thunderbird, particularly a set of enterprise features that Thunderbird needed. So funding from that funds some of our developers. There's also some engineering and related work that can generate funds and we have arrangements with some of the search providers in the browser itself, which will generate funds as well.
Roblimo: Which search providers?
Mitchell: Well, there's a set. You know, you can see the setup there.
Roblimo: Actually, I've customized mine so hard that I can't see it. (laughter) I'm sorry! I'm one of those...
Mitchell: We come with a default set of five, and that's Google and Yahoo and Amazon and eBay and Creative Commons and some subsets of those.
Roblimo: And as far as companies that are giving financial support, Microsoft and who else?
Mitchell: (laughing) Yeah, right. Microsoft?
Roblimo: (laughing) Well, I like to end things with a little humor.
Mitchell: But actually, I wanted to come back to the money piece, and money and funding and support of open source projects, because I do think that this is a topic that I certainly want to talk about more in the future. And I think other, you know, many people will, because a project like Mozilla really needs people working full time on it.
Roblimo: Yes.
Mitchell: And that means someone has to provide funds. You have to raise funds somehow. And once there's money in the picture, money is great because we all need it to live on. Money tends to make people suspicious, if there's any money floating around. So I think one of the issues that open source projects, including Mozilla, will have to look at in the future is: given the need to generate enough funds to sustain ourselves and pay people, i.e. the need to generate funds, how does that change the dynamics of the project? I don't know the answer to that, but I do know that if you're generating any funds, people tend to get suspicious of you and maybe rightly so. And so I think there's gonna be a whole round of discussion on exactly that.
Roblimo: Thank you very much for being our guest on Slashdot, Mitchell Baker. And I'm Robin Miller, known to far too many people as Roblimo, and thank you all for listening.
Mitchell: Well, thank you! I hope we can do it again before too long.
A lightly-edited interview transcript is posted below.
:(
I would have thought Slashdot would have left the all important question: "Who the fuck are you and why do I care?" in but apparently they didn't
Why is it half of the time when I click on a "read more" link for a story with no comments I repeatedly am told "Nothing for you to see here. Please move along."?
Help Brendan pay off his student loans
an MP3.
here comes the care police
The answer was in the blurb, at the risk of being redundant, it's:-
A lightly-edited interview transcript is posted below.
Because you come here far, far too often.
"Nothing can shake my belief that this world is the fruit of a dark god whose shadow I extend." - Emil Michel Cioran
Basically this is another interview of a Mozilla worker, who google will pinch to work on gbrowser. You can't beat big business...
Your threshold settings by chance?
* Winners compare their achievements to their goals, losers compare theirs to that of others.
You must be adicted to slashdot, reloading so you see this message everytime
No talk of IE7 or how they plan to take on a new pre-installed browser. That makes me sad. I guess maybe a lack of CSS2 support in IE7 might help Mozilla... but that's a big might.
-dave
http://millionnumbers.com/ - own the number of your dreams
Is it just me, or is 'Mitchell' supposed to be 'Michelle'?
Funny how often users etc bitch about open source and here we're using mp3.
Way to go.
-- Note: If you don't agree with me, don't bother replying. I won't read it.
You managed to avoid insulting and harrassing the interviewee (is that a word?) like you've done in the past. Very good.
Problems like that are the true test of integrity in open source software. Having the organizational character to step up and warn people that uninstalling/installing your app can totally delete everything on their drive is noble. Being blase' about it is just bullshit. If your organization wants credibility then they have to take responsibility for their screwups, even if they do give away their product for free.
I'm flattered that the interview ended with my question about funding.
This is extremely important to my company as we make free software tools and we try to do this with funding from commercial projects.
I don't promote my company on Slashdot and it does not matter - this is a general issue - but we've been making free software since 1995 or so.
It's not an easy combination but we're getting some really interesting work the last year or so. There are companies that spend significantly more on licenses for software than it would cost to build open source alternatives. This creates a market opportunity for companies like ourselves.
It does take quite a lot of courage for a large company to deliberately pay for an open source development. But there are smart CIOs out there.
And yes, I agree with Mitchell Baker that funding is essential. Love of the work is a large part of the reason for doing it, but love of success and fame and fortune are also a big part of it.
Sig for sale or rent. One previous user. Inquire within.
Roblimo: I want to have your babies.
He hit the telephone so hard until it breaks!
You obviously missed this one
Thats Ars-Fartsica, the art of farting. I assure you if you are ever sitting next to me on a plane I will make this painfully obvious to you.
Nvu developper is Daniel Glazman, not Glassman. See his weblog
I know one of the points of Firefox was that it's seperated from the Mozilla email client etc, and that's a good thing, but one way of tacking the integration issue is to make everyone else a Firefox extension. I'm not sure how efficient that mechanism is for integrating apps as big as or larger than Firefox but it would be good to have a Thunderbird extension, alongside the Calender one (or have Firefox include Thunderbird and have that Thunderbird have a Calender extension). The more stuff you bundle into Firefox then presumably the better they can co-operate, and the less I need to minimize my browser and deal with windows. There could be a Firefox `windows explorer` extension, for instance, or a Windows application launcher extension. Perhaps some of these have already been done, but not, as far as I'm aware, in the context of the browser utterly controlling the desktop.
... never ever met a man (american, brit) who had this type of intonation and speech rythm (not even speaking about the pitch of his/her voice....)
this sounds absolutely and ultimately like a woman, maybe from some huge company in the marketing/manager level.
this doesnt sound at all like a white male student. and i assume that this mitchell guy is some "wasp"-type guy, or am i completely mistaken?
maybe he had some sex/genger change/reversal.
it really sounds like if that would be the case.
Ahh, open source chronic -- an unattainable panacea. I've search low and high, but what I've found is not exactly what I'm seeking. The obvious place to look returns a 404 error -- not the 420 I had hoped for. I've performed numerous searches, but haven't gotten many hits. Please help, I'm buggin'!
High Times,
Letter
Now that will liven up your iPod shuffle.
sulli
RTFJ.
To lighten the load on the /. image server here is a torrent of the MP3 file.
mitchell_baker_interview_cut.torrent
Roblimo: I shouldn't wonder. Well there's a question, let's start off here with one that wasn't the first one and I don't even know if it made the final cut. But people... several wanted to know what browser do you use yourself? (Mitchell laughs) Do you use... which version of Explorer?
We don't need to read "how are you", "fine", etc. ...
Also next time remove "and another question is...".
Just do: user questionN, answerN,
I agree, Robleimo is a terrible interviwer and this interview was useless. I read the first quarter and quickly scanned the rest. Nothing caught me as very interesting.
WURD!!
How come "Who are you and why the fuck should I care?" wasn't one of the questions asked?
http://www.thecompassionclub.org/
WTF is right! What's with you moderators calling this informative. If I say your mom is a .... is that informative?
Yooouuu shooould liiisten toooo thhhee Emm Peeee Threeeee. Robbblimmmoooo haaass thhhee mooosttt booooring tooone evver.
You don't even need to listen to the whole thing. Just the start:
Roblimo: How are you?
Mitchell: I'm great, thanks.
Roblimo: Not just good, great.
Mitchell: Yes, I'm great.
Roblimo: That's good to hear...
I read that and imagined it was some lighthearted quick banter. Listen the the interview. It sounds like Roblimo has recently taken a large dose of NyQuil.
Oh, and:
Mitchell: I don't use Explorer. In fact I really see Explorer...
That's, as most people probably guessed, RARELY. Not "really".
All the installer needs to do (and many installers already do this) is create a log of files installed, and uninstall only those files. Look at Unix; if they nuked /usr/local everytime someone uninstalled an application... But that doesn't happen, because the installers on those platforms keep track of what was installed where.
An uninstaller *should* do what the user expects it to, which is just uninstall the program, no matter where the installer put the files. If it's too much work to undertake in the short term, as a stopgap, they could do something like forbid installation into C:\ (or any root prefix). But just calling the user stupid doesn't help Mozilla any. Most commercial apps do these sorts of things just fine, and if open source apps want greater acceptance in the marketplace, they should too. (Note that with my suggestions above, no "dumb user" warning messages needed!)
I dunno what that is. So i'll say yes.
When you die, on your deathbed, you will receive total consciousness. So I got that goin' for me, which is nice.
It's a simple thing really, and one lots and lots of other install builders have gotten right.
Indeed. Most other install builders go so far as to know exactly which files they are installing, and the uninstallers only remove those exact files - and only remove the directories if they are left empty.
Which means a lot of uninstallers leave the program directory behind, with eg preferences files (which weren't installed by the installer, or have been changed since). But it's the safe approach to err on the side of caution, as a default, you then change the installer to delete extra files/directories that you KNOW will be left behind. (The cache directory would be another example).
fud, notfud, yes, no, maybe
...The one question I wanted an answer to (asked by the wxMozilla core developer about the separation of GRE and XULRunner, etc. from Mozilla) didn't really get asked because Roblimo claimed Mitchell had answered it in the previous question. But of course, he didn't.
This is a really important question for a lot of developers: When can we integrate Gecko in a project without (a) it depending fully on Mozilla or (b) including a whole entire of the GRE bundled?
it is here...
Sweet Jesus on a motorbike...could we do without the friggin' introductory retard-to-retard small talk on the next interview??
"Fine, I can hear you now, Dimitri. Clear and plain and coming through fine. I'm coming through fine too, eh? Good, then. Well then as you say we're both coming through fine. Good. Well it's good that you're fine and I'm fine. I agree with you. It's great to be fine."
GET FREE APPLE STUFF!
Awww, Rob is hitting on Mitchell. Isn't that cute?
Made me laugh - +5 FUNNY
...issues.
I was very interested in hearing about the influx of Mozilla vulnerabilities and Mozilla's direction on security.
Loading...
What kind of Question is that? This is an important interview, why waste Baker's time with nonsense? I would hope that someone's important question did not get left out because of that.
Trix are for kids!
No application should delete any files that it didn't create, Unless it is designed for that purpose. End Of Subject.
..."
"If a user is advanced enough to change the install directory
You are making the incredibly incorrect assumption that only a knowledgable user can make the change.
The Kruger Dunning explains most post on
I've never taken a look at Slashcode, but could it be a caching problem? Trying to access a nonexistence or not-generated-yet cache.
No uninstaller should ever delete any file that the corresponding installer didn't place there. Period. There's this neat high-tech solution that's been gaining favor since - well, actually, since there have been installers, period - called "keeping a log of what files you installed".
The Windows Installer does this automatically for the developer. Given that all Windows programs MUST use the Windows Installer for their installation programs (which means Firefox can't call itself a Windows program), this shouldn't be a very hard task to manage.
The best part is that the uninstaller doesn't even remove the profile directory. Were I to uninstall Firefox right now, I'd have 30MB of crap left hanging around my hard drive. That's larger than the browser itself. Why? Because it doesn't - at the very least - erase the cache on uninstall. Hell, it doesn't even ask you if you want to delete your profile!
Saying that "it's OK to delete the entire hard drive" because the user has to take extreme measures to do it is bullshit. The installer should never even have to ask "should I remove the installation directory". It should know enough to remove all the files it installed and then remove the directories if they're empty. If they aren't, warn the user, but NEVER DELETE THE DIRECTORY.
The dialog box reads something like "Really really uninstall?" and so most users just hit "yes" because they have no idea what it's asking, and most certainly don't connect it with "erase the entire hard drive".
This is a HUGE FLAW on behalf of the Mozilla team and one that MUST be remedied.
"Who are in control, they are not in control of anything - they don't even control themselves!" - Glen Beck
Good timing or what? Go and grab it from getfirefox.com.
The transcript has numerous error in transcription that actually change what Mitchell says, so please instead hear her actual words.
She survived AOL until they laid her off because she had not become appropriately assimilated. What makes you think Google would have better luck. Those of the Mozilla developers who shift employers, but not loyalty earn a deep respect from me.
What The Transcript Says:
What He Probably Really Said:
WTTS:
WHPRS:
A while ago I had the chance to talk with a Chatzilla developper on irc, and he said the lack of progress on Chatzilla could be attributed to the fact XULRunner wasn't yet released (at the time) and that they were waiting for it.
:
XULRunner description
XULRunner is a single installable package that can be used to bootstrap multiple XUL+XPCOM applications that are as rich as Firefox and Thunderbird. [http://wiki.mozilla.org/XUL:Home_Page]
I don't know in what state the app currently is, but I was told it would greatly simiplify the packaging of Chatzilla, similarly to ffx.
Slashdot isn't exactly one of the most open websites around. Just sayin'.
He reminds me of someone, but I'm not sure I can put my finger on exactly who...
None of that was ever addressed, except maybe a little about Moz 1.8.
Not to mention the format was terrible -- it prevented long questions and had too much nonsense chat -- and Roblimo never challenged her. He was her buddy, not a journalist. Next time, just reprint a press release; who needs an interviewer?
This should help put a face to the name.
You sly dog: you got me monologuing! - Syndrome
Why are missing pages 404 errors? Since they happen so often shouldn't they be higher on the list? are there 403 more important or encountered errors? Am i totally off base in my assumptions?
Can you be Even More Awesome?!
a browser with everything added in and that utterly controls the desktop.
Sound like Windows + IE + MS Office.
Since the mail/news in 1.8b1 was a bit unstable I decided to migrate to Thunderbird which feels unpleasant and clunky, just like IE/Outlook, and then I find that changing things is also wrongly/poorly documented and hard eg ~/.thunderbird not ~/.mozilla/thunderbird at the top level, where are the hidden prefs documented, how do you change the
CANCEL OK
button order back to OK CANCEL in the chrome.
I am disappointed.
Downloading the installer.tar.gz files, moz is 13M and firefox is 8M. Not a huge difference. In either case, if you are on a slow connection you go out for coffee while it downloads and if you are on a fast connection it loads almost instantly. When running, ff uses a little less RAM, but again not enough to matter much. So what's the big deal?
If you want an unbloated browser, the dillo rpm is 280k. Loading this article nested, threshold 2, in ff uses 32M, in dillo 5M. That is a difference worth talking about.
The point is not to rag on firefox. The question is why people who know better advance empty technical arguments to support the change. The developers who started phoenix were quite up-front that politics and bureaucracy were a major motivation. They said that working inside the vast mozilla structure was stifling and not any fun. Now somehow the project that started as a way to work outside the main browser development process has become the main browser development process. ("Everybody's trying to be so alternative, but alternative is so main stream" - JBE)
Can anyone explain what is really going on?
No, actually it's called Just for Fun.
Before you walk a mile in someone's shoes, you should insult them so you know how they are and what they're doing.
who's Robin and who's Mitchell?
What an odd pair.
WTPOUAWYHTTOTWPA
What's the point of using acronyms when you have to type out the whole phrase anyways?
In the 24th minute, the bits in italics are left out of the transcript:
Roblimo: Now we're just taking his quote, we don't know where he got it. Is Firefox truly ready fo the masses when developers maintain this sort of attitude toward users? So basically, this is a responsiveness question --
Mitchell: Well, I'm not sure if it is. Or, I'm not sure if it's just a difference in opinions. So one developer has said...
Smoooooth.
Never let a bloke interview a bird. Seriously.
Feel free to adjust the above for sexual preferences/prejudices.
Mate, this is slashdot, please don't upset the OGG zealots! :)
errors in transcription... I didn't know Microsoft made Speech Recognition software..
It's a 45 rpm mp3. You're using 33 rpm.
-Dave
The competitive advantage of Netscape/Mozilla over IE & Outlook has always been the way the suite worked so nicely and efficiently together. It is a corporate standard at many companies; including mine. If you have to cobble together Firefox/TBird/SomeHTMLeditor and get all the roving profiles working you might as well go MS. I just don't understand. :(
Listen To The F#$%n' mp3 file should be the new RTFA now.
Opera now includes SVG in it's latest beta (SVG 1.1 Tiny).
:)
http://www.opera.com/pressreleases/en/2005/03/16/
Maybe that might kick Mozilla along a bit with SVG support
I'm a perfectionist but I'm trying to cut back.
Why has the above post been modded funny, surely it's informative?
This on-the-phone format is extremely irritating and way less informative than the written list of questions that used to be the standard for interviews.
Back to the old way PLEASE!!