July 2005 Archives

Three weeks ago I posted a short article titled Linux Not Ready for the Desktop where I called Linux "a very capable product with a very limited audience" and I suggested that there were four areas Linux needed to improve in order to attract "Regular People" in significant numbers: migration, stability, simplicity, and comfort.

The article generated quite a bit of feedback, both here at this blog and over at slashdot -- where it got noticed by Nat Torkington, Program Chair for this year's OSCON who asked me if I'd be interested in updating the article based on all of the comments and giving it as a keynote. I accepted and will be presenting Linux - In Search of the Desktop on Friday, August 5th, at 8:45 AM.

Rather than updating the original article with replies to all of the criticism I'd received, I decided to do a quick reply to the top few most common complaints and then start a mini-series of posts to dig further into each of the four points, migration, stability, simplicity, and comfort.

If you're at OSCON, and you'd like to see me wrap this all up into a 20 minute keynote, do stop in on Friday morning. If you can't make that but you're interested in chatting, I'll be at the Convention Center all week long.

Having finished that series of articles, and again read through the 1500+ replies, I think there are just two pieces I'd like to spend a bit more time on. I'll be following up with a couple of blog posts covering those two areas. Expect the first one tomorrow: a response to the "Linux doesn't need more (Windows) users" feedback that I received so much of. If I can find time on Tuesday or Wednesday, I hope to post on the second area, what I love about Linux.

Thanks again to everyone that read this series and especially those of you who gave feedback. It's been fun.

In my original article, I said that "Linux must feel comfortable to Windows users. Most people using computers today have been at it for a while now and they've been at it on Windows. Don't mess with their basic understanding of how things work."

First, it's important to clarify that I am definitely not saying that Linux should be (or even could be) a Windows clone. Just as Firefox is not and should not be an IE clone, Linux needs to be its own while also being as comfortable as possible to Regular People coming over from Windows.

A fine corollary can be found in our move from the legacy Mozilla Application Suite to Firefox. When Mozilla was being developed, a major focus was to appeal to the existing user base which was mostly Netscape Communicator 4.x users. Those users were comfortable with the keyboard shortcut Ctrl+L to focus the addressbar. IE has always used the Alt+D shortcut to focus the addressbar and a non-trivial number of IE users had complained that that simple little shortcut habit was causing them pain when they moved to Mozilla. Mozilla refused to change so as not to "break" the Communicator users. By the time Firefox was born, the pool of Netscape Communicator, Mozilla 1.0, and Netscape 6/7 users that were potential Firefox converts probably numbered in the low millions (if that.) The pool if IE users that were potential Firefox converts, on the other hand, numbered in the hundreds of millions. We added the Alt+D shortcut for focusing the addressbar. Was this a critical change? Probably not. Was it important enough that by doing it we made millions of IE users a bit comfortable on Firefox and so more likely to stick with Firefox? Absolutely.

I bring this up because I took quite a bit of heat for suggesting that something like the reversed positions of OK and Cancel buttons in the common dialogs on Linux should be changed to accommodate the much larger potential audience of Windows users. Is it absolutely critical? No. But I argue that, like that Mozilla shortcut key, there isn't sufficient benefit to being different to warrant making Regular People feel even a little bit uncomfortable when making the decision about whether or not to save a document. And I'm also not saying that Linux should be as inconsistent as Windows is. This is a great opportunity to take what Regular People are comfortable with (the Windows button ordering) and improve it by doing it consistently across all dialogs. Linux can be comfortable for Regular and better than Windows.

In addition to the OK and Cancel buttons being reversed, I also mentioned the issue of the Linux clipboards. While it may be that having two distinct clipboards is a useful feature for some people, it will be disconcerting to most Regular People and I'm arguing that it's not worth it if it makes Linux's largest potential audience feel even the least bit uncomfortable. This is also not some kind of horrible deal breaker for getting Regular People to use linux, but when you add up a lot of these minor discomforts, it makes for something that feels overall too foreign and uncomfortable. That's to be avoided if you want those people to hang around long enough to appreciate all of the benefits of the Linux desktop.

Other more serious examples of unnecessary divergence from the Windows desktop with little (or no) benefit to Regular People include the Gnome change to what they called "spacial Nautilus" - stripping Nautilus of the file browser and moving to a paradigm with multiple open folders on the desktop, and the main panel move from the bottom of the screen to the top. Both of these chances are very jarring to Windows converts and are of questionable value to most Regular People. If the feature, once adjusted to, offers a far superior experience then maybe it would be worth it. If you can't say for certain that it does offer significant value, then there is no good reason to break the habits of Regular Users and give them something uncomfortable.

I'm sure there are more examples, both minor and major, of differences between Linux and Windows - changes that don't offer significant benefit but do make Linux uncomfortable to Windows users. I'm not cataloging every issue here because I simply don't have the time.

How does Linux improve here? The challenge is to find those areas where it's valuable to change and make the transition as easy as possible (through good documentation, intuitive or easily learned UI, etc.) and to find those areas where change doesn't offer enough benefit and make those areas as comfortable as possible. It is not necessary to be a clone, but it's foolish to deviate from what Regular People expect when the value of that deviation isn't extremely high.

I'll have at least two more posts on this topic (hopefully before OSCON) where I'll summarize my four main points, and hopefully be able to further respond to some of the comments and criticism. Stay tuned.

I've been playing with IE 7 beta 1 since last night and I've got a few quick thoughts. First, I suspect this is really more like an "alpha" than a beta. Second, their tabbed browsing UI is pretty nice. It's discoverable, it's easy to learn, and they've solved the overflow problem very nicely (we should do that!) Third, this new toolbar arrangement is going to take a _lot_ of getting used to.

Too bad for all the windows 98, windows 2000, and windows XP pre-SP2 users out there that they won't get this upgrade. Good thing they can get a better browser (today!) with Firefox ;-)

Reading Steven J. Vaughan-Nichols eWeek article titled IE 7 Beta: Improved Security, Resembles Firefox I noticed this nice little typo down near the end.

I just got word from a friend at Yahoo! that the Yahoo! Toolbar for Firefox has achieved it's 1.0 status. You can read what he has to say about it over at the Yahoo! Search blog.

Great work, guys!

I posted, as part of my original article, that the third major problem hindering Linux adoption on the desktop was its lack of simplicity. I said "Just because you can include a feature doesn't mean that you should. Just because you can provide a user preference doesn't mean you should." This was difficult lesson to learn for me, and it took watching a lot of "Regular People" interacting with computers to understand it. You and I are not Regular People and we don't interact with computers like Regular People. We all really like computers and often don't mind the challenge of fixing computers when they break or troubleshooting problems for friends and family. Regular People don't enjoy troubleshooting problems. They don't want to be asked to guess what the right setting is and they don't want to customize their computer experience beyond the very basics. They just want it to work and if they've got it working, they sure don't want to risk breaking it.

So, where does Linux fall down when it comes to simplicity. First and most obvious is that Linux is not one but hundreds of different desktop operating systems that are all quite distinct. Until Linux has enough market share that OEMs start shipping PCs to Regular People with Linux pre-installed, prospective users are going to have to hunt down and find Linux on their own. Having so many choices is actually quite off-putting for most Regular People. Supposing they finally dig through the choices and pick a distribution, before they can start using it, they're often asked whether they want the KDE desktop or the Gnome desktop -- yet another choice that makes a new user wonder whether or not he's doing the right thing. Another opportunity to say "I just don't know" and walk away.

When the potential new Linux user settles on even a relatively easy to use distribution like Ubuntu (my current favorite of the Regular People focused distributions,) they're still faced with a the complexity of a text-mode installer that can't be launched from a running Windows system and that asks too many questions that could be answered by simply polling their current OS. After installation they get a nice and clean desktop, but as soon as they open the main applications menu, they're confronted by a Games menu with almost 20 video games they'll probably never play, a Sound and Video menu that contains multiple programs sharing the same icon and about 8 apps when three or four (or one or two,) would probably do -- and Ubuntu is leaner here than pretty much every other distro I've used. The next menu down, System Tools contains items like a floppy formatter, a tool that lets you report your hardware setup to Ubuntu, a bug reporting tool, and something called a "Root Terminal". Step over to the System menu and things get even crazier with options for configuring the user's online CD information database, desktop themes and window manager, and the look and feel of the login screen! Oh, and which menu was it, System Tools, Preferences, or Administration, that had the mouse settings and why wasn't that on the same menu as the printer settings.

Linux is asking users to learn an entirely new system and starting with way too much at the top level. It's simply not necessary to provide access to all of these items at the top level of the main menus (or in menus at all.)

Most of the popular distributions seem to run with the "more is better" approach -- including every application that they can manage to package up. A few are starting to come around and pick best of breed apps to feature, but even this just addresses a small piece of the complexity that's so daunting to Regular People.

Digging a bit deeper into the system, you'll find this same abundance of geeky configuration choices overwhelming the few key settings users might actually want to adjust. Opening the preferences for what should be the simplest of applications, Text Edit (gEdit,) reveals more choices than the any reasonable user would ever want or need. At what point was it decided that the Linux version of Notepad should offer users the ability to configure the exact number of Undo actions he wants the application to provide (and how is anything less than "infinite" a reasonable default for this setting?) It's not just Text Edit that suffers, either -- do Regular People really need over 100 options to configure something as simple as an IM client?

I know that some of this may sound like picking nits, but this is Ubuntu, the most Regular People focused distro I've used, well ahead of the competition, and they still put "Open Terminal" as the first item in the desktop context menu! Open Terminal. If opening a terminal is still important enough that it was necessary to put it at the top of the desktop context menu, then it's going to be a long time before many Windows users move over to Linux.

So how does Linux improve here. First, the user shouldn't have to answer more than a few basic questions to get through an install. Second, just because Linux can offer a feature or a service or a config option doesn't mean it deserves a position on the main menus. The things that most people don't need most of the time shouldn't be mingling on the menus with the things that most people need most of the time. Third, application developers need to make some of the hard choices and stop falling back on the "make it a pref" solution that seems to be all too popular in most software these days.

Taking just a few more steps to remove unnecessary complexity will go a long way to improve the appeal of the Linux desktop for Regular Users. My fourth and final segment in this Linux on the Desktop series, which should be posted in the next few days, will cover the issue of user comfort.

Today, Jeremy from LinuxQuestions.org interviewed me for LQ Radio. You can check it out here.

Seventy-Five Million Firefox Downloads! Wow. (see the counter over at SpreadFirefox.com.)

In just eight and a half months, we've gone from our first download to our 75,000,000th download. Thanks to everyone out there making this happen.

We've got a great new tool coming online to help us distribute some of our Mozilla/Firefox/Thunderbird functional testing load. It's called Litmus and we'd like to get more people using it but right now it's just plain ugly (completely my fault for not having time to pretty it up.)

You can take a look at the still experimental tool by visiting Litmus. You'll want to enter your system information, pick one of the two groups of tests (right now it's smoketests and basic functional tests) and then pick a sub-group and hit submit.

All of these pages could use the loving hand of a good designer and I'm just far too busy right now to tackle this work. If any of you are interested in helping out here, please let me know with an email - asa@mozilla.org

The Discovery launch went off without a hitch. What a beautiful shot of the external fuel tank separating that was. Wow. Spectacular!

We're just about 8 hours from the planned launch of the space shuttle Discovery. If you've got NASA TV, tune in. If not, you can grab a feed over at the NASA Television site.

It looks like we're going to be doing a Mozilla 1.7.11 pretty quickly in order to fix bugs 300749 and 301917. If you are a suite user and have any other concerns related to the 1.7.x releases, feel free to let me know here.

We have one fix in hand and are looking into the second issue today. I'll have updates, here or over at the Quality blog as soon as we get candidate builds.

In my original article about Linux on the desktop, the second major point I brought was API stability and an easy and uniform software installation story. I said, "The second problem that blocks massive Linux Desktop growth is stability. I don't mean the not crashing kind of stability, I'm talking about a stable API that doesn't require the user jump through hoops when they want to download a new application from download.com."

There are two basic methods of installing software on Linux, using the package system from your distribution, or downloading source and compiling. Compiling is right out for Regular People coming from Windows. It's too complicated for most and too slow for all. Using the distro's package system seems quite reasonable when you're just talking about updating the core OS but starts to fall down when Regular People want or need to install applications that aren't available or easily accessible from their distro's repository. Windows users are comfortable going to Download.com and downloading and installing an application. They feel safe and secure going to a software vendor's site, Mozilla or Adobe for example, and downloading and installing an application. The current Linux systems take the application vendor, a powerful tool for building user trust and confidence on the Linux desktop, and pull it out of the picture almost entirely.

Regular People do not want to get all of their software from a single source -- their distribution. They appreciate finding and downloading new gems, from around the web. A simple example would be something like Mozilla's Thunderbird application. Many Linux distributions ship Evolution as the default installed email application and a user who happened to read a positive article about Thunderbird would find herself at the Mozilla website looking over the Thunderbird feature set and reviews. She should be able to download Thunderbird right there and have it just work. With today's Linux desktop, this will fail more often than it will work because software developers cannot support even the most popular few distributions with a single download.

Another problem is that not all distros can keep up with every software package in a timely manner. When Regular People read about the latest update to a piece of their software, they should be able to get it, right then, and from the vendor that created that software. As of the writing of this article, several major distributions aren't yet providing the latest release for Firefox (1.0.6). With today's Linux, the Firefox release team cannot possibly create packages for every distribution and every package system out there and they should not have to. They create one package on Windows and it works out of the box on every Windows release since Windows 98SE (released more than 6 years ago.) Supporting even the last three years of Linux, even one distro, with a single package just isn't feasible today. Supporting the last one or two years with a single package isn't without some pain and Mozilla's a fairly large and capable organization.

How are these issues remedied. First, backward compatibility must be a real priority. This means doing whatever it takes to avoid breaking compatibility when revving libraries and carrying around specific compatibility libraries when compatibility has to be broken. It's certainly more of a maintenance headache than what most Linux distro vendors want, but until that change happens, software makers are deterred from producing high-quality products and maintaining those on Linux and Regular People are a lot more likely to be frustrated by the software install process. Second, there needs to be a lot more cross-distro compatibility or a lot fewer distros. This will make it much easier for software vendors to target the Linux platform and will make it much easier for Regular People to "shop around" for software.

Making improvements in these two areas will go a long way to improving the experience Regular People have with the Linux desktop. My next article will discuss the third issue keeping Regular People off of the Linux desktop -- unnecessary complexity.

Wow. I'm mentioned on /. again.

Take a look at this very cool Google Maps mod. Read about it at Raj Kaimal's blog.

Mozilla Foundation just celebrated its two year anniversary this week. What an exciting two years it's been. Thanks to everyone who has helped make our efforts so successful.

The first point I covered in the original article was migration. I said "For 'Regular People' to adopt Linux (which usually means leaving Windows) Linux is going to need a serious migration plan. It will need to install on machines next to Window, leaving that completely intact and easy to return to, and carry over all or nearly all of the user's data and settings."

The overwhelming majority of potential converts are currently using Windows and they have set up their desktop and each of their primary applications to work the way they want the things to work. They don't want to put that at risk in any way. They hate troubleshooting computers and will avoid anything that sounds like it might disrupt their semi-stable environment. A "Regular People" Linux (which I'll call "RPL" in these articles) must not threaten their existing environment at all.

So what specifically is needed here and how can it be accomplished. The RPL distribution will need to install beside the Windows installation, leaving Windows fully intact and functional (and give the new user a lot of confidence in that -- don't even bother offering to over-write their Windows install. Seriously.) The install CD should include a graphical partitioning wizard that will explain to the user that he needs to free up space on his Windows machine and that this will not harm his Windows installation in any way. This utility should autorun when the user inserts the RPL CD and should clearly explain to the user how much space is available on the user's disk and recommend to him the necessary amount of space required for the RPL installation. After the user agrees, the tool should then perform the necessary work to set up the partition(s) without further user interaction and, upon conclusion, kick off the actual RPL installer.

Most Regular People are not going to copy their files and settings (if they can even find them) onto a CD or USB device and manually carry them over to the Linux installation and then reorganize them. Neither are they going to figure out how to mount their Windows file system, locate and copy over those files and settings, so it is up to the Linux installation to handle that. Even if they haven't done serious customization, they still have information like fonts, printers, timezone, desktop shortcuts, desktop wallpaper, language, files, bookmarks, cookies, history, passwords, email, contacts, recently accessed file lists in apps, windows size and position, etc. that exists in their current environment and that needs to exist in the new environment. If the desktop and applications they arrive at in their new Linux installation have these bits, the user will be a lot more likely to actually use it and to stick with it.

How does Regular People Linux deliver this. After installation, on the initial launch of RPL, the user should get a migration wizard that offers to copy the his Windows data -- making it very clear that this is a copy and not a transfer so the user doesn't feel he's risking anything. The wizard should list each of "Web browser settings," "Email settings," "Instant Messenger settings," "Office suite settings," "General desktop settings," and "My Documents, Pictures, Music, and Videos." Next to each of the generic labels, the wizard should display the actual names of the specific applications that are being migrated so the user can understand the relationship between his Windows applications and the receiving applications. The migration tool will need to be able to mount the Windows partition, migrate the various settings and copy the user's files over to his new RPL installation. While that's happening, the migration wizard could offer a brief tour that explains the RPL applications that will replace the user's Windows applications and point out where his documents will be located.

The individual migrators may be difficult to write, but at least two of them, for Firefox and Thunderbird, are already mostly done and could be easily repurposed for this task. Migrators should carry over as much as is available from Windows itself (and the Windows applications) and as much as the receiving clients can handle -- including toolbar organization, recently accessed documents, applicable options/preferences/settings, and desktop/start menu/quicklaunch toolbar shortcuts to those applications.

The copying of user files should be a fairly easy task. They should be copied into a similar folder structure as they are in the user's Windows filesystem (probably a "My Documents" folder with "My Pictures," "My Music," and "My Videos" sub-folders.)

These migration features are the first major steps to getting Regular People to give Linux a try and stick with Linux in any reasonable numbers. My next article will focus on the second set of issues that are necessary for widespread adoption by Regular People -- API stability and a familiar application install experience.

It's been ten days since my article, Linux Not Ready for the Desktop and I've finally made it through the 300+ comments here at my blog, the handful of blog posts that trackbacked me, and the nearly 1000 comments at the slashdot posting.

I'm happy to have received so much feedback. Thank you all (well, almost all of you.)

Before I start my mini-series covering each one of the four areas I believe need serious improvement for "Regular People" to move to Linux in any serious numbers, I wanted to take a minute and address a few of the responses I got.

One of the common replies I got was basically "Linux doesn't need more users, so don't change anything to accommodate Windows users." While that view seemed to be held by decent number of people, I couldn't disagree more. We need a competitive desktop marketplace and Linux could be a player in that market. The Windows mono-culture is a threat to security and an impediment to progress. There is currently a window of opportunity for Linux to compete in the mainstream and bring some diversity back into the system. We need it.

Another set of responses said "We don't need your Linux bashing," "Are you working for Microsoft," or something similar. Well, I didn't think I was bashing and I certainly don't work for Microsoft. I was offering what I thought to be constructive criticism for attracting and holding onto new users from the Windows world. I've got a bit of experience with that task working on the popular open source Mozilla and Firefox projects for the last 6 years.

A third common reply was "Linux should not become a Windows clone." I agree. Linux should not (and probably can not) become a Windows clone. The very successful Firefox web browser is not an IE clone. It does, however, make moving from IE to Firefox a much easier task than moving from IE to any other web browser and this has been a major factor in its success. Just to take one example, if Ben Goodger hadn't implemented a kick-ass migrator for IE, I doubt we'd have any more than a small portion of the Firefox users we have today. Regular People do not like to fiddle with or troubleshoot computers. Asking them to do that is taking what could otherwise be a gentle learning curve and turning it into an impenetrable brick wall.

The last of the easily categorized responses to the article was the charge that I'm not informed enough to make these claims, that I've either not spent enough time on Linux or that I've only every used Fedora Core 4. Well, I've been using Linux for six years. I've tried Red Hat versions going back to 5.2. I've used several versions of Debian over the years. I've used SUSE starting with 7.0 and all the way up through the current 9 and NLD. I've spent a good bit of time on all of the Fedora releases. In the last year I've tried two versions of Ubuntu, including the current release. I tried Lindows when it was first released and recently got a copy of Linspire 5 and a subscription to their software catalog. I've tried out just about every major distro for the last six years and while I did complain specifically about Fedora, my arguments stand for all of these distributions. Some are clearly going to be easier for Regular People than others, for example, I think Ubuntu (which shows the most promise) would be considerably easier for my mom to adjust to than unixy distros like Slackware or performance oriented Gentoo. But even Ubuntu does not make it easy enough for Regular People and is missing out on millions of potential customers because it does not give Windows users the transition that most of them need.

I'll be following this post with the first in my mini-series of four posts dealing with each of the major issues I highlighted in the original article: migration, stability, simplicity, and comfort.

For the last several months, we've been discussing the versioning of the upcoming Firefox release.

One major consideration in this decision was that the sheer volume of changes in the Firefox core (Gecko) made a minor .1 increment seem misleading. While it may not be obvious by looking simply at release dates, today's Gecko core of Firefox has seen nearly 16 months worth of changes compared to what shipped in Firefox 1.0. This is because we created our Gecko 1.7 branch (the branch from which Firefox 1.0 shipped) back in April of 2004. At that time, Gecko development on the trunk continued and very little of that work was carried over to the 1.7 branch to be included in Firefox 1.0.

Another consideration was that we've made some major improvements to the Firefox application, especially in the update and extension systems that warrant more than a minor version bump. Calling it 1.1 would suggest to most users that this was a minor update when in fact it is quite major and all 1.0 users really should move forward for a much improved product.

update: it seems there is some confusion about the beta naming. The beta is going to be called "Firefox 1.5 Beta" not "1.4 [anything]". The beta's "1.4" versioning is necessary for our extension and update testing but will not be a part of the "common name" for release. It will be called "Firefox 1.5 Beta".

If you're a Pluck user (or you think you might become one) you will definitely want to head over to addons.mozilla.org and grab the Pluck RSS Reader 2.0 (Beta 2). It's a very well done Firefox extension that gives you convenient access to a powerful tool for organizing your online world. Do check it out.

I'm sure most of you already know, but just in case you don't, developer.mozilla.org has gone live. Devmo is "is dedicated to providing solid information, documentation, and community for Mozilla and Firefox developers of all types." A huge thanks to Deb Richardson for making this happen.

If you were a big fan of the advanced searching options in the old Mozilla search sidebar, there's a great new project over at mozdev called Searchsidebar that will probably make your day. It's still in the early development stages but I'm sure they could use some help testing.

If you're a del.icio.us user, and you're not already using Firefox's very cool quicksearch feature for fast access to del.icio.us bookmarks, head over to dmiessler.com for a couple of cool tips.

update: Reports have come in that this isn't working. it would be great if one of you all could look into this and see if you can get it working!

After my recent blog post on the Linux desktop, I was invited to keynote the article at the O'Reilly Open Source Conference. Since I was already headed up to Portland to present on Mozilla (with Chase) I agreed. So, if you're interested in seeing me get pummled with rotten vegetables (or you want to participate) for about half an hour, get yourself over to OSCON in the early morning of August 5th.

Oh, and I've been "published" at ZDNet.

And just in case anyone was wondering, I still believe in the points I made in the original post. More on that in upcoming days.

Lawrence Lessig uses Firefox :D

PC World is reporting that Firefox took nearly another full percentage point away from IE in the month of June.

Firefox increased its market share to 8.71 percent, up from 8 percent in May, while IE's share shrank to 86.56 percent from 87.23 percent, NetApplications.com, an Aliso Viejo, California, maker of applications for monitoring and measuring Web site usage, said Thursday in a statement.

Since the beginning of the year, Firefox has increased its market share every month between 0.5 percent and 1 percent, mostly at the expense of IE, according to NetApplications.com, which compiles its browser usage data from more than 40,000 Web sites monitored by its HitsLink.com service.

Rounding out the top five browsers in June were Apple Computer Inc.'s Safari, with 1.93% market share; America Online Inc.'s Netscape, with 1.55%; and Opera Software ASA's Opera, with 0.59%.

And we're not done yet.

Chris Aillon, over at his blog tries to suggest that I'm a hypocrite for saying that Linux (and Fedora in particular) are failing to provide stability for third party developers. That's bullshit. When we realized we broke this API, we sat down, developed a plan to address it ASAP and are spinning up new builds as we speak.

When I complained to Chris the other day about his Fedora project dropping support for Mozilla's Firefox releases (by not including compat-libstdc++ version whatever,) his answer was that I wasn't supposed to use Mozilla's Firefox, I was supposed to get my Firefox from Fedora and Fedora only.

Chris, the difference is that we try hard not to break people and when we do - like in this case - we drop everything else and go about fixing it. You simply washed your hands and said "not our problem" when you broke Mozilla's Firefox on Fedora. Sure we can do better, but at least we're trying and not telling our users to get over it.

Another win for Mozilla applications, and millions of users, Yahoo!'s desktop search now supports Thunderbird. Thanks very much to Duke Fan and all of the great folks over at Yahoo's desktop search team. Also thanks to Jeremy for bitching about this shortcoming in the earlier release ;-)

Please post your comments from linux not ready for the desktop here. Thanks.

Just about 14 hours from now, the space shuttle Discovery will be lifted into the air by a pair of massive solid fuel rocket boosters. I will be at work looking for the best online feed available. I'm armed with a handfull of NASA TV feeds and looking for alternatives if those are clogged.

If you've got suggestions for live streams other than NASA TV, please let me know. Thanks.

update: launch postponed :(

Tonight, you can finally get your hands on Deer Park Alpha 2.

With this release comes the now enabled by default "fast back (and foward)" feature which makes moving backward and forward in history blazing fast. We've also enabled the "Reporter" tool for reporting broken websites (you can find this in the Help menu.) In addition, tab drag and drop reordering made it into this alpha release, so did a major improvement to pop-up blocking, and also, error pages have replaced error dialogs.

You can read more about the release at the release notes page and you can download it from here

You all asked and I've (finally) answered.

"Now that we are in July, do you still think we can make a 2.0 or even 1.5 release by the end of the year?"

Martin, we'll make our next major release around the end of summer. Major releases beyond that will probably be scheduled for early next year though we're not firm on that.

"When a bug is marked WONTFIX, what, if any, is the decision process leading up to that? Do developers make that decision independently, or would do they consult other devs first? If a decision has been made in the past to not implement xyz, what, if anything, would prompt a review of that decision?"

Rishi, it really depends. A bug will be marked wontfix for a number of reasons. Sometimes it's because fixing the bug would result in issues that are worse than the original problem. Other times it's a feature we do not intend to implement in Firefox. We have a system of Module Owners and Module Peers. They are ultimately responsible for decisions about bugs and features in their modules. You can read more at the module owners page. Module Owner decisions are pretty much final.

"microsoft always gets the blame for having the style scrollbar-color css tags because there IE only, but mozilla has the moz-opaticy and moz-borderroundess (i may not be accurate with those) dont you think there is a contridiction there?"

No, I don't think there's a contradiction. The specification contains, for example, opacity and it most certainly allows user agents to experiment with and prototype new css features as long as they preface it with a -vendor flag and we do this with -moz-border-radius. With -moz-opacity, that's a slightly different story. It used to be -moz-opacity, but when it came into sufficient conformity with the specification, the -moz was dropped and now it's just opacity like the spec calls for.

"i know this is your personal blog as well as mozilla based but do you have to get your blogs checked in any way before you post them?"

Absolutely not.

"What marketing efforts are you making to ensure that people upgrade to [the latest version]? Why isn't spreadfirefox pushing this harder?

Updating is important and we can always do more to get people to update, but the charter of Spread Firefox is to spread Firefox, not to work with existing users to keep them secure. We can certainly use Spread Firefox, along with all of the other outlets we have, to get the word out about updates, but the real solution is to get the update mechanism working so that this word of mouth effort isn't required. That's coming with Firefox 1.1 and all users of 1.1 going forward will have a much better update experience.

"Do you have plans to create a unique session for each browser and all tabs within each browser? If so, when can we expect to see this change? If you do not plan on making a change for your session managing design, would you tell us how developers like myself can write web applications to work well with your browser without having to architect our own session managing framework?

Steve, I do not believe that there are any plans to have each browser run in it's own process. I'm not sure what your application does and I'm not a developer so I really can't say how you would work around this (or even why you would need session management frameworks.)

"When will we see 1.0.5?"

Now :-) here.

"When will we see 1.1? :-P"

We should be shipping the next major release of Firefox by the end of summer if things go well for us.

"Will FF 1.1 pass the Acid2 test? If not, should we expect it in 1.5?

Limulus, Acid2 is not on the list of requirements for the next major Firefox release and we're not far enough into the planning for the subsequent release for me to say whether or not it will be on that list. update: one of our Gecko experts has posted in the comments saying "Asa, I think it's safe to promise that the next major Firefox release after 1.1 will pass Acid2 :-) "

"when will the skins support the full gui instead of just icons and colors ?? currently, I find it a bit ridiculous."

I'm guessing you don't understand the full capabilities of Firefox skins or I don't understand your question.

"One of the things that I've been thinking about recently is "sustainability" of large and/or important FLOSS projects in the middle-to-long term (i.e., attracting new developers to take over when the current crop "retire"). I'm curious if you can provide an "estimate" of developers within the Mozilla project as a whole (because Gecko/Core is just as important as the end-user apps themselves) who do "significant" work but who are not former Netscape employees."

Smokey, I haven't done any math on this. I'm also not quite sure what it is you're trying to discover. Would you count people who, like me, were volunteers on the project _before_ they were paid by Netscape (or IBM, or Oracle, or Novell, or Red Hat, or Sun, or other major corporate contributors)? If I'm guessing correctly and you're wondering about how many people are working on Firefox/Mozilla/Gecko right now who were not original Netscape employees, I'd say it's an overwhelming majority. If you're wondering about people who were never paid a dime by anyone to work on Firefox/Mozilla/Gecko, that's a minority (especially since the Mozilla Foundation has been hiring up many of the contributing volunteer developers over the last couple of years.)

"Do you think that MoFo in particular is now being hypocritical in implementing tags like <canvas>?

No, I do not. The standards bodies serve an important purpose. They hash out new features and do as much as possible to future proof them and get them well documented. They do this in public so that all vendors can participate and can chose to implement (or not) that feature with a clear understanding of what it means. They do not design and specify new features with the intent of creating vendor lock-in to proprietary (and often difficult or impossible to understand/implement)features. Canvas is a multi-vendor specification that meets the above criteria for openness. I hope that Web Forms will follow the same path and be adopted by all the major browser vendors.

"Why is it that there are many bugs that have patches, reviews, and even approval for check-in, but never actually get checked in? For the Deer Park alpha 1 cycle, it looks like there are still 15 bugs or so that have approval but no checkin, and for Deer Park alpha 2 the number looks like 40 or so (I am sure some will be checked in once things are opened up a little more after tha a2 release, but seems like a large number)"

Well, they fall into a couple of categories. Some belong to developers who do not have check-in rights and they must wait on someone who is willing to take that responsibility. (Remember that when you check in code, you have to stick around and watch the Tinderbox until all of the major platforms cycle green and you're responsible for any bustage that happens during that cycle.) Another reason might be that the patch belongs to someone who is hard at work on a more pressing issue and he just hasn't gotten around to landing it. Before we ship a milestone, we take a quick look at all the "ready to land" patches and make sure we've got any that are critical to our release.

"Regarding testing and QA, a topic I know you spend a lot of time with. What is your thoughts regarding specific functionality testing like smoke tests and other automated testing versus the idea of providing stable builds that people download and use on a regular basis as a form of testing? I really like the later approach, it proves Firefox in the "real world", fits the development philosophy of incremental improvements with a constantly usable platform, and seems to be the best method of testing a very complex system in an even more complex environment. I would really like to hear your thoughts on this approach. I have more than just a passive interest in this topic, we are currently working on some resources for aiding nightly/hourly build users with regression testing/bug triaging"

Darin, I think both automated testing and interactive testing are extremely valuable. Because we've traditionally had hundreds to thousands of nightly build users who report bugs when they find problems in ad hoc testing, we've put more energy into finding ways to organize and utilize their feedback than we have into test automation. At the same time, it would be wonderful to be able to automate the morning smoketests that Tracy Walker has been performing on each product on each platform for about 4 years now (he deserves a break from that :-) I think they both have their place. We really do need to get some good automation together to test our layout, JavaScript, security infrastructure, networking, etc. Until that happens, we're overly dependent on nightly and milestone build user/tester feedback and that's led to problems in the past.

"Extensions are not only a great way to allow 3rd party developers to easily add to the product, but also a good source of innovation for core features should they be incorporated into the product in the longer term. For example, I think Ben Goodger was hoping to get MiniT in, and you blogged ResizeableTextarea. Such useful yet non-bloating features would be great to have in the default install - is this happening?"

Yes. We're always evaluating extensions. I have probably played around with about 600 Firefox extensions over the last two years and have recommended we take some of those into the core. A tab reordering feature, based on MiniT has already landed in Firefox and we're currently investigating a couple of other small feature additions from extensions.

"I was "awarded" a SFX '25 million downloads coin' but I've never received one in the post. I know of another member who received nothing but an empty envelope. I am of course a little disappointed that it never arrived, all I want to know is can I even expect anything? I'm certainly not demanding an answer, after all, I was privileged to be awarded the rare coin in the first place. My blog posts (SFX, blakeross.com) e-mails, they've all gone unanswered... =o( Curses on whoever stole it...

Craig, I'm terribly sorry to hear this. If you'll send me your mailing information, I'll put another coin in the mail ASAP.

"Do you think that [privacy options] are implemented in such a way that normal user can get with them (I mean are cookies, cache, entries, history in vocabulary of normal surfer)?"

Ivan, I don't think that "Normal People" use preferences at all. Nor do I think most Normal People would understand what they were doing if they did start mucking about in cookies, cache, history, etc.

"How much of the back-end of your site did you write? And the front-end for that matter."

I didn't write any of the backend of this blog. It lives on Mozillazine's installation of Movable Type. I wrote the bulk of the structure and style of the front-end though.

"It's long been maintained by those working with Mozilla that Gecko is much more secure than IE. But this year there have been 44 security vulns in Mozilla products, compared with 25 for Microsoft's products. What are your thoughts on those statistics? Also - have the numbers of Firefox downloads started slowing down significantly yet, are they continuing at their impressive pace?"

I think Firefox and the Gecko technologies that underly Firefox make it the safest mainstream browser available. Security is a process, not a product and when we find problems, we fix them. Our downloads aren't slowing down at all, as a matter of fact, in the last two weeks it seems like they've sped up some :-)

"It should be so much easier than it currently is to sync profiles (and extensions) between computers (if you're running the profile from differently named directories on the two computers). Why can't I simply move the profile from one computer to another? It would be great if Firefox incorporated an "import profile" feature. Will this ever happen?

I don't know. It's certainly not on our list of requirements for any of the next major releases.

"Will Spreadfirefox.com ever become more user friendly?"

Yes, it will.

"Will you increase, maintain, or reduce your efforts at annoying the Opera community to new heights for July?"

Apparently some of them have very thin skin and see malice where there is none. I don't intend to change how I blog to protect the more fragile among us.

"What feeds do you subscribe to?"

I subscribe to the standard ego feeds: google news, feedster, and pubsub searches for "asa", "firefox", and "thunderbird". I also subscribe to feeds from most of my Mozilla colleagues as well as JWZ's APOD scraped feed, and Gizmodo.

"Will Camino's new automatic "Block web advertising" (Camino 0.9a1) feature make it into Firefox soon? It's pretty effective already."

Not that I'm aware of, Dex. I'll look into it though.

"Should Ben still be lead developer of Firefox? I don't mean only because he now works at Google, but I mean that because you simply don't seem him to be active in bugzilla."

Ben is the Firefox lead developer. If his module peers have issues with that, I'm sure they'll bring it up with him.

"Any thoughts about integrating some of the features provided by extensions into the FF "core"

Yes, we are always evaluating these Firefox based projects and where it makes sense, we will incorporate their work in Firefox. That being said, in most cases, it makes the most sense that they continue on as extensions.

"Does the way you are constantly picking on Opera and lying about it show that you are worried that it's a better browser?"

That must be it. And the 7 million people that downloaded Opera 8 are clearly overtaking the 70 million people that downloaded Firefox 1.0.

"Hmmm, how does this thing work Asa? Aren't you suppose to answer or do you wait another few days to "collect" more questions? I'm asking since I'm very curious about some of these questions :)"

Matt, I wait until I've got a good collection of questions and then until I have a free couple of hours to respond to them all and then I write up and post this blog post :-)

"It seems as though Mozilla product releases have slowed down considerably since 1.0 releases happened. I seem to recall an article a while back with Mitchell Baker stating the goal would be to have regular product releases every quarter or so. Now, unless you count security releases in that mix, things have shifted from that mentality a lot. Could you expand on some of the contributing factors that may have lead to the slow down in product release?"

Quarterly "final" releases for Firefox were probably never in the cards. Most users (and when you're reaching many tens of millions of users, you're dealing with more "regular" people that when you're reaching the first couple million early adopters) don't want to change their software every three months. We do intend to continue to make releases happen every two to three months, in the form of alphas and betas, because we rely on these for more widespread feedback and testing, but I think we're much more likely to see Firefox major releases more spaced out.

"Are there plans to increase the updating of the Mozilla Wiki pages or is something else going to fill the gap left since the 1.0 releases? I very much understand having mscott or others post on MozillaZine directly is a huge time commitment, but since Mozilla products are open source, having better community communication helps with feedback, support, and planning. Could you provide any other suggestions on where to go for more information regarding releases and what pages could I point Mozilla Newbies to (other than MozillaZine or your site)?"

We're always working on that. I am concerned with the lack of communication and am looking forward to seeing more communication -- even within the development teams. It's time consuming, though. We're trying to do more of it with developer and QA blogs, the Wiki, DevMo, and other outlets. We're open to suggestions for squeezing more minutes out of the day ;-) We do take this seriously and there's no doubt that it needs improving. I, personally, fully understand the benefits and know that even though it often feels like a burden and a big cost of time, it's worth because of the return that we get in being better coordinated and getting more feedback sooner.

Thanks for all your questions. If you had a serious question that I didn't get to, or you have follow-up questions, please hold off until I open the next installment of Ask Asa. Feel free to discuss these questions and answers here, but I probably won't be getting heavily involved. Thanks again and I hope that you all are still finding this a useful feature.

Last night, Firefox crossed the 70,000,000 downloads mark. Wow.

update: Please post your comments here (this page is getting too long and slow.)

After reading Massimo Sandal's recent article, The Firefox Target And The TuxMini, I knew I was going to be in trouble because I just had to blog my contrary position. This post isn't a point by point reply to Sandal's article, but rather a more general disagreement based on my experiences with Linux, Mozilla and Firefox.

It's probably worth pointing out that I'm not a "Linux person." I've only been using Linux for about 6 years. I've been using Windows for about twice that long, and I've been using Macintosh for about 20 years. I'm not really loyal to any one OS; I've used what my employer or school offered or required and when that wasn't a consideration I've used what I thought was most convenient.

OK. So what's wrong with Linux that makes it not ready for the desktop. I've tried KDE and Gnome desktops but my latest is FC4 so my criticism is focused on that (and Gnome) but I think KDE distros suffer just as bad if not worse. The issues fall into four basic categories, migration, stability, simplicity, and comfort. These issues each cover both technical capability shortcomings as well as usability failings.

The first issue, migration, is pretty serious. For "Regular People" to adopt Linux (which usually means leaving Windows) Linux is going to need a serious migration plan. It will need to install on machines next to Window, leaving that completely intact and easy to return to, and carry over all or nearly all of the user's data and settings. Regular People may be willing to take a look at Linux, but as long as all of their data and settings still lives in Windows, they're not going to stay very long -- no matter how appealing it might be. We learned this lesson in the Mozilla world. It wasn't until we implemented a very capable migration system in Firefox, which carried over the user's IE favorites, cookies, history, passwords, etc. that Regular People started moving over in serious numbers -- and staying (and bringing others over.) Linux needs to do the same. It's clearly a much bigger task for an entire OS and all of its major applications to accomplish, but it simply has to be done. When Regular People fire up the Linux desktop for the first time, the browser, office suite, email client, IM client, file manager, etc, each need to carry over as much as possible of the Windows application settings and all or very nearly all of the user data. Without this, the hill is just too steep to climb and Regular People will not make the climb.

The second problem that blocks massive Linux Desktop growth is stability. I don't mean the not crashing kind of stability, I'm talking about a stable API that doesn't require the user jump through hoops when they want to download a new application from download.com. A user should be able to install Fedora Core 4 and go grab the latest Firefox release from Download.com and have it work without the need for finding and installing compat-libstdc++ or whatever. Developers may think it's cool to reuse as much code as possible but the user doesn't care whether it was Linux that failed to include the necessary compatibility components or Mozilla that failed to make the build work for that particular dot release of libstdc++. Regular People expect to be able to download software, install it, and have it just work. Asking them to figure out complex system library and kernel compatibility issues is a one way ticket off of their desktop.

The third issue is a lack of simplicity. Just because you can include a feature doesn't mean that you should. Just because you can provide a user preference doesn't mean you should. I don't want to start a desktop war but I really gotta say to the distros, pick a desktop and be happy. Regular People shouldn't have to (guess or learn enough to) choose between Gnome and KDE when they're installing your product. Regular People don't need 15-20 mediocre games in a highly visible Games menu at the top of the Applications list. And what is a Regular Person to think when confronted with a choice between Helix Player, CD Player, and Music Player? Does the Music Player not understand CDs? What's "Helix" mean? Gedit has about 30 user preferences spread across 5 tabs in a preferences window -- Notepad has about three. You and I know that the difference between Settings and Preferences is that one is system wide and one is per-user but Regular People don't know that and shouldn't need to know that. If the Regular Person doesn't have access to it because it's a system wide setting, then why put that entire menu of options in front of him. If the Regular Person has equal access to both, then why are they split? It's just a confusing mess.

The final major issue is comfort. Linux must feel comfortable to Windows users. Most people using computers today have been at it for a while now and they've been at it on Windows. Don't mess with their basic understanding of how things work. Regular people do not know what it means to "mount a drive" and they shouldn't have to. Regular People don't want their OK and Cancel buttons reversed -- tossing out years of finely tuned muscle memory. Regular People shouldn't have to learn what /home means or how it differs from My Documents. Regular People don't want two clipboards that seem to constantly overwrite each other. Linux UI fundamentals need a reworking to match the habits that Windows users have been building over the last decade. Get the users first, then try to teach them a better way (if you've got one.) Putting things in the "right" place for Windows users will go a long way. You can never do too much to ease the transition.

I think of Linux today the way I think of Mozilla 1.0 from just a few years ago: a very capable product with a very limited audience. If Linux makes major inroads on the desktop, it will probably be as a result of the same kind of focus that put Firefox on tens of millions of desktops, a focus on migration, stability, simplicity, and comfort.

update: I just ran acros Kim Bauters' article which hits on some other great points.

update: Please put your comments here (this page is getting too long and slow.)

I think that Robert Scoble makes a very good point (and not just because he mentions me in making that point) in his article Bloggers Are Gonna Have To Be Journalists.

Be your own editor and aggregate your truth out of the cloud of variously biased viewpoints.

No, I haven't forgotten that I owe you all answers to Ask Asa. They're coming, eventually.

We're getting very close to Deer Park Alpha 2. We need your help testing these out to ensure they're good enough for the alpha release. Head over the the Quality blog for more details.

At the end of a long week, Josh Aas and I sat down on the couch we call the MoFo lounge and hacked on his skateboard for a few minutes. You can see the results over at Josh's blog

Steve Rubel, who I got to finally meet at Gnomedex, has the goods on the upcoming Yahoo blog search capability.

Eric, over at ROACHFIEND.COM, is soliciting ideas for a new Firefox extension. Head over and tell him what you'd like to see in a new extension and maybe you'll get lucky :-)

Get it while it's hot!.

After initial evaluation, it seems like the spell checking is broken on trunk Firefox builds. It works with our 1.0.x releases (and the latest Firefox 1.0.5 test builds.)

Our community of 1.0.5 test build testers found a regression yesterday and we've fixed it and have a new round of builds to test. If you're a 1.0.x user and can spare some time to upgrade to this latest (and hopefully final) test build, we could definitely use your eyeballs.

And just to make it fun, if you find a regression that we have to fix for 1.0.5, I'll send you a crappy Firefox t-shirt ;-) This shirt is a limited edition (because the screen print was good but not up to our quality standards of "great") and if you want, I'll even autograph it ;-) The photo was taken with the crappy camera on my phone so even though the shirt is crappy, it's not quite as crappy as the crappy photo suggests.

You can get the scoop on where to get and how to test these 1.0.5 test builds over at the Mozilla Quality blog

This has to be one of the coolest applications I've ever seen. If you've got Firefox, you really have to check this out. Seriously.

(link via John Udell.)

We're getting close to our 1.0.5 releases and we could use your help testing those bits before they ship. If you've got some time, please head over to the Mozilla Quality blog and lend us a hand.

Yesterday, the Pew organization released a report (PDF) which says "18% of internet users say they have started using a different Web browser to avoid software intrusions." This was in response to the question "Have you, personally, done any of the following to avoid getting unwanted software programs on your computer?" where "Started using a different internet web browser." was one of the six choices. The survey was conducted mid May through mid June.

I'd be extremely surprised if people were moving to IE so the conclusion I draw from the report is that lots of people, nearly one in five, are fleeing IE to get away from software intrusions. I'm also going to assume that the niche browsers Safari and Opera (the number three and four most used browsers,) make up at least a small chunk of that.

You can read more about this study in the AP report. This data and analysis seems to back up the general trend we're seeing in all of the surveys of browser usage. Firefox is taking a big bite out of IE and at a pretty quick pace.

It appears that Dafont.com is blocking Firefox+adblock users. I sure hope that this is a mistake and not another case of a website trying to tell users how they should view the web.

Webmasters, listen up. Respect your users or you will lose them. With Firefox, the user is no longer just a spectator, he's a participant. Play nice or face extinction. Seriously.

Gervase (pronounced like "service") Markham is now an official part time employee of the Mozilla Foundation. It seems kind of silly to say "welcome aboard" since he's been working in the Mozilla world for as long as I have, but getting paid to work on something you love is certainly a bonus. This also means he'll be able to spend more time, 2 days a week, on Mozilla work and that can only be good. Read more over at Gerv's.

If you're a Movable Type user (like me) then you'll appreciate this article, Make out with FireFox. The article introduces a spellchecker and handy greasemonkey scripts.

(link via anil dash)

I've posted the instructions for testing the new Deer Park update system over at the Mozilla Quality blog. If you're interested in helping us test out this new 1.1 feature, please head over there and give it a spin. Thanks!

Ben and Darin gave a quick demo of their Firefox Update work last week and today I was able to update from yesterday's nightly Deer Park build to this morning's Deer Park build using the new Update system.

I just created a short video clip of the update process and as soon as I edit it down so it's not such a huge file, I'll get it posted.

This isn't fully enabled in the nightly builds just yet. One of us will be posting over at the Mozilla QA blog when we're ready for more widespread testing.

Go read this.

Susan Kitchens has been blogging the Comet Bash all evening. Great stuff. Thanks, Susan!

Head over to her site for play by play on the Deep Impact mission's most exciting moments.

update SMACK!!!!!!

I just read over at the Googlebar mailing list archive that Google will shortly be releasing an official Google Toolbar for Firefox. Nice.

How many of you all use toolbars? I get by with Firefox's built in search field and our advanced find in page feature but I know a lot of people, especially IE converts, got used to the Google Toolbar features and will appreciate this.

Anything that helps to ease the transition from IE to Firefox is a good thing in my book.

In my attempt to add an RSS 2.0 feed, I temporarily broke things for some of my feed consumers. I think everything is working now. If it's not, I'm hoping that someone will let me know. Thanks.

We're well aware that the update system we had in place for Firefox 1.0 was sub-par and a major focus for the upcoming 1.1 release has been building a world-class update system that "just works" for our users.

Accomplishing this work is no small feat and we learned a lot from our experiences with the 1.0.x releases. The new system will be much more robust and capable on both the client side and the server side. From the client end, we've got a much improved user experience that should take away pretty much all of the pain of updating Firefox 1.1 for both minor versions and major versions. On the server side, we've increased our capacity and will have the capability to serve "patch" updates which should make things easier for everyone.

This new system will also provide benefits for our testing community. If all (or even most) goes according to plan, nightly build users will be able to get the latest nightly bits through the same update mechanism which will make testing nightly builds a bit more convenient and, more importantly, will help us get a lot more testing on the update system itself. This nightly updating system should be in place within the next few days and will be available to our 1.1 Alpha 2 users (preffed off by default). 1.1 Alpha 2 is just the first testing stage, though, and the 1.1 Beta is where we'll introduce the full set of capabilities (including patching) to the testing community.

I'm super-excited about these improvements and will do my best to keep you all up to date on our progress and the opportunities for early testing. Stay tuned.

Over the last few weeks, questions about the origins of tabbed browsing have come up several times. I thought it would be worth a short post to put my thoughts on the subject in front of a larger audience for scrutiny (and flames ;-)

Tabbed browsing is neither a Firefox nor Opera invention. Firefox and Opera fans, both, should step back from any claims to this invention.

The first real tabbed browser with any significant presence on the web was Netcaptor, created by the very talented Adam Stiles way back in 1997.

The next major implementation of tabbed browsing was the work of HJ van Rantwijk with MultiZilla, a tabbed browsing extension for Mozilla that copied pretty much everything that Adam had done in Netcaptor. HJ launched this extension for Mozilla back in 2000.

In September of 2001, Dave Hyatt added a tabbed browsing mode to Mozilla. This feature was release in Mozilla 0.9.5 in October of 2001

In December of 2001, Opera Software released version 6 of its Opera browser which was the first version to contain a genuined tabbed browsing mode (along with its SDI and MDI modes).

In September of 2002, Phoenix 0.1 (which would eventually be renamed to Firefox) shipped its first release which contained the most usable tabbed browsing implemenatation to date ;-)

In January of 2003, Apple introduced the Safari web browser which contained a very nice tabs implemenation.

In May of 2005, Microsoft announced that IE 7 (due later in the year) would have a tabbed browsing interface. The MSN team at Microsoft shipped a tabs-capable toolbar for IE 6 in June of 2005.

Am I missing anything in this timeline?

Yes, it's time again for another installment of Ask Asa. Ask away!