February 2009 Archives

Nate Lanxon, over at C|Net, recently wrote an article titled Opera should give up on desktop browsers where he says "Opera should give up on the desktop browser market and focus its time on developing for mobile phones, media players and similar devices" and "It should take its clearly talented teams of developers and shift its focus to the mobile world where it can really thrive. It should focus on the types of devices it's already winning with -- the Nintendo Wii for one, and even the Archos handhelds -- and claim a dominant position."

I completely disagree. What Nate and others don't get is that Opera needs a desktop presence. Opera's vision of the Web, one I share, is that it's the same Web everywhere. This is really important. But today's mobile Web usage just isn't enough, even if you dominate the mobile market, to have a really thoroughly tested technology stack.

People just don't use the Web on other devices like they do the desktop. Even the most popular mobile Web experience, the iPhone, barely breaks half of one percent of Web usage. With that kind of usage, and the feedback that comes from that usage, a Web browser would not be able to keep up with the ever-changing Web.

Any successful mobile Web browser must embrace the full Web, the same Web that the desktop browsers experience. The only way to do that today is to bring the desktop technologies, proved out by lots of usage, to the mobile space. A mobile-only play just won't cut it.

Opera's desktop browser means millions and millions of desktop users logging hours and hours of testing and providing lots of feedback to the Opera team. That feedback allows them to keep their rendering engine, javascript engine, and other bits of the browser as functional as possible across as many Web sites and Web apps as possible. Take it away, and the Opera mobile efforts will slowly but surely fall behind the "one Web" as Opera calls it.

In addition to proving the technology, the ~30 million or so Opera users on the desktop provide tens of millions of dollars in growing and predictable search revenue from Google (and Ask, Yandex, Amazon, Baidu, and Allegro.) That was almost 25% of Opera's $70M in 2008 revenue.

Suggesting that Opera abandon the desktop when that is the primary proving ground for its core technology and when it's generating ~$16 million dollars a year in revenue sounds kind of silly to me.

This article and several recent other very confused and heavily flawed articles discussing the browser market place raise an interesting point, I think. The Web browser space, desktop and devices, isn't really very well understood by people who aren't directly involved in it. This leads people in the press and on blogs to write all kinds of strange analysis and to draw all manner of wrong conclusions and predictions.

Everyone uses browsers and most tech-savvy folks can follow the basics of browser releases and market share. But the Web browser space is a lot more complicated than that and people should be careful to not over-estimate their understanding of how the market actually operates.

OH, And a little research and asking a few people who do work in the industry the right kinds of questions really does go a long way.

As I'm sure at least some of you know, I'm not a programmer -- not even close. So, when I run into even basic scripting tasks, I have to reach out to a colleague, or in this case my readers, and ask for help.

Here's what I'm doing.

For Air Mozilla streaming broadcasts, I've got a DV camera hooked up to a MacBook Pro via FireWire. The MBP is running Ubuntu 8.10. These are the steps I take to stream video to Air Mozilla.

1. Connect the camera to the laptop and ensure that they're both powered up.
2. Open a terminal and run the command sudo chmod 777 /dev/raw1394 which prompts for the root password. (This seems to be necessary for the user to get access to the FireWire device.)
3. Run the command dvgrab --format raw - | ffmpeg2theora -a 0 -v 5 -f dv -x 320 -y 240 -o /dev/stdout - | tee [EVENT NAME].ogv | oggfwd icecast.mozilla.org 80 [PASSWORD] /[CHANNEL NAME].ogv
4. ctrl+c to stop it all.

I'd like to simplify this so that others in the building can grab the camera and the laptop and stream without having to know any of this.

First, surely it's possible to permanently change permission on the FireWire device or make it so that it automatically gets changed at boot or login. It seems a pretty odd requirement but I don't know how to make the change stick.

Second, I need a simple script that that a user can just double-click on, be promoted for the "event name" (name of the local file that will be saved) the password, and the "channel name" (the name of the stream), hit enter and be streaming live video.

A somewhat more sophisticated script could ask for a quality and size setting, maybe offering a couple of presets.

An even more awesome version would have a simple little GUI so users didn't even have to see the terminal window.

Anyone out there willing to take a stab at this so we can have more Mozillians streaming more live events?

Mitchell Baker has launched a very interesting discussion over at her blog. Rather than talking about specific remedies in the EU Microsoft thing, she's reaching out to our community to come up with a set of principles that should clarify the outcomes we're after.

Principles rather than the detailed prescriptions seems like a great foundation and I encourage you all to participate.

Mitchell Baker -- EC: List of Potential Principles

Last week a big mess broke out at Auctiva, an EBay sellers' tools site.

From what I can tell it went down like this. Auctiva servers get infected. Google and Firefox malware protection reports Auctiva as an attack site. Users freak out. Auctiva tells users it cleaned up the problem. Users report back that Firefox and Google don't agree and still report Auctiva as an attack site. Auctiva tells users to disable Firefox malware protection!!!. Some users bypass Firefox warnings by disabling malware protection or visiting Auctiva with unprotected browsers. Auctiva shuts down site saying the problem isn't yet fully contained.

Maybe I don't have this completely right (please let me know if you know more) but if this accounting is accurate, Auctiva deserves to go out of business.

There is no excuse for a Web site telling its users to bypass or disable core security functions of the browser. None. Ever.

I don't care how confident you are about your own set-up, telling users to disable safety features that apply not just to your site but to all Web sites is completely unacceptable.

I know that if I was putting my money or my private information into a site that advised bypassing browser security measures, I'd cancel that account immediately and never go back. Not only that, I'd tell all my friends, colleagues, and just about anyone that would listen to never do business with them.

There is just no place for that kind of irresponsibility when it comes to user safety and security.

Check it out!

Paul Newman and Sydney Pollack are the ones I will miss most.

We're one week away from the NetApplications February browser usage report, and as I've been doing lately, I'm posting my predictions a week early based on analysis of the daily and weekly data.

February is wrapping up to be one of the more interesting months in a while because of how flat everything looks.

From least to most interesting, here's what I see happening in February.

Opera is going to turn in a pretty flat month, wrapping February at either 0.70% or 0.71%. The long-term trends for Opera lend confidence to this prediction.

Opera's had some modest ups and downs but year over year growth is pretty much negligible. The Web is growing very quickly and Opera's just not able to outpace the Web's overall growth, so while they're surely adding users, they're not able to get a larger piece of the pie.

Chrome appears to be set also for a pretty flat month in February. February will mark the 6th month of Chrome availability. Ending the month with a predicted 1.15% usage share will make clear a nearly horizontal line going back to late December.

From these numbers, it looks like Chrome had a pretty flat first two months, modest growth for the middle two months with the 0.3, 0.4, and 1.0 releases and has fallen back to mostly flat again for the last two months.

Firefox will again post modest gains in February, likely adding about a quarter point to end February with 21.75% usage. It's obviously too early to make 2009 predictions, but I think there's little doubt that Firefox will break 25% this year and may end up closer to 30%.

Internet Explorer will lose more share again in February, but not quite as bad as it's been for the last few months. I'm predicting that IE6 and IE7 combined will net a negative one quarter to one third of a point to close out February at 67.3% usage. I've discussed IE6's rapid decline in recent posts, but another interesting trend is emerging.

IE7 obviously benefited immensely from Windows Update. You can see that from the first big boost after release and then another growth spurt when Microsoft removed the WGA anti-piracy requirement. What I'm more interested in, though, are the flatness of the line over the last 8 months or so and that IE7 never broke 50% usage. With IE 8 just around the corner, it doesn't look like IE7 will ever break 50%. That's pretty amazing.

Finally, the most surprising move of February comes from Safari which I predict will fall back a full quarter point. It's not so much Safari, though, as Mac OS X. I'm predicting that Mac usage overall will take a hit of almost three tenths of a point and that's going to cost Safari for February.

Safari has benefited a lot from Mac's positive holiday season in 2008 and with less Web browsing going on there, Firefox and Safari are both losing some usage. Safari's got the lions share of Mac so it's taking a bigger hit there and it doesn't have any life at all on Windows where Firefox was more than able to make up that loss.

As I said in the opening, these are all just predictions. It'll be a week more before the February numbers are out. I'll revisit the numbers and any analysis that's impacted by the final numbers in a week.

Mitchell's post got me thinking (and commenting) more about why I think people are mis-reading the Web browser market place.

I don't know how many of you followed my link over there and I think it's almost enough to stand on its own so I'm going to just post my comment here as well.

I’m hearing quite a few claims from the tech press (and others) that we’ve obviously got a healthy environment because Firefox, Safari, and Chrome are all competing well against Microsoft.

The reality is that only Mozilla, with Firefox, has actually managed to pull any significant number of users off of IE.

It’s obviously early for Chrome, so perhaps they will, but they haven’t yet. And Safari has been available on Windows for three times as long as Chrome and it hasn’t even garnered the meager share that Chrome has.

To those who think that “everything is fine,” I say think about the world without Firefox. Without Firefox, Internet Explorer would absolutely dominate with 97% of the Windows desktop.

Let me say that again. Without Firefox, IE would have 97% of the Windows Desktop.

Yes. 97%.

(Safari on Windows has less than half of a percent, Chrome is about one and a quarter percent, and Opera is about three quarters of a percent.)

I do not believe that “everything is fine” when Mozilla is the only thing that stands between Microsoft and it’s dominance over 97% of the Windows desktop.

Go read what Mitchell said: EC Theme: How can there be a problem when Firefox has growing marketshare?

Counting users for a browser or an operating system, or just about anything at scale, is very difficult. LinuxPlanet has an article posted by Steven J. Vaughan-Nichols titled How Many Linux Users Are There (Really)? He spends the bulk of the article basically saying that none of the measures we have are useful and he concludes by pitching a project designed to try to count Linux users with a new tool.

I was a bit disappointed because with a headline like that, I expected some answers :-)

Not one to sit around disappointed, I decided to try to come up with a number myself.

This is all fuzzy math, with a lot of assumptions and lots of rounding, but I think I can do a little better than Steven offered for an answer.

I don't think anyone would disagree that the overwhelming majority of PCs (laptops and desktops being used by human beings) in use today are connected to the Web with Web browsers. I don't know what that number is but let's say it's 90% to 95%. It may even be more, but let's go low for the sake of this investigation.

I'm going to just assert that Firefox represents somewhere between 75% and 90% of Linux browsers in use today. It's probably higher, but that range shouldn't be controversial.

update: for those saying we don't see Linux distro-packaged Firefox update checks, yes, you're right. We do see plug-in blocklist pings and that's what I was measuring. Sorry for the confusion.
OK. So that works out to Firefox Linux browsers representing somewhere between 65% and 85% of all Linux PCs. Let's hold that for a second.

Now, looking at our Firefox numbers, we can positively identify Linux at at least 2% of total Firefox browsers based on update checks. There are some "unidentifiables" in our measuring right now, so I'm going to be generous and allow that most of those unknowns could be Linux. That would put Linux somewhere between 2% and 5% of active Firefox browsers. (Just to let you compare, Mac is ~7% and Windows is the rest.)

Estimating Firefox's total number of active browsers isn't an exact science since we don't track individual machines, but we've worked out over time that it's approximately a 3x multiplier against our "active daily users" count. Right now that count is about 85 million, so we estimate our installed base at ~250 million.

So now we do the math :)

If we look at the "worst case" there are as few as 5.5 million Linux PCs out there.

If we look at the "best case" there are as many as 16.5 million Linux PCs.

That's a pretty big range, but it's certainly a start.

Next, let's look at it from the usage direction. NetApplications, TheCounter.com, and Google Zeitgeist (before they stopped publishing the number) all measure web browser traffic to get OS stats and all put Linux at ~1% of their web browser usage.

That's usage, not users/installed base, so people who use the web more, are going to be somewhat over-represented. It's probably a safe bet that Linux users are at the higher end of the usage spectrum so doing the straight math on usage % against total users would give Linux a bit of extra credit. That's OK as long as we keep in mind that this method of counting will paint a somewhat rosier picture for Linux users

The total number of people on the internet is estimated (by the folks that get paid to estimate these things) somewhere between 1 billion and 1.6 billion. We're talking about people here and not browsers. With multiple people sharing a single computer more common than individuals using more than one computer, again the number will come out high.

So we do the math again :)

Working back from usage data and total internet population, we get between 10 million and 16 million Linux users.

There are probably other directions we could approach this, but I don't know what they are and looking at this from a couple of angles, we can already see that there's some agreement about the high end. It's unlikely that there are more than about 16 million Linux PCs in use today.

At the low end, there's less agreement, but I think it's reasonable to just mash those two numbers together and say that there are probably no fewer than about 8 million Linux PCs.

That's still a pretty big range and while I'm usually an optimist, I think if someone asked me how many Linux users there were, I'd just split the different between the high and low and say there were about 12 million of them. That feels about right to me.

I want to restate that this is all fuzzy math, with a lot of assumptions and lots of rounding and mixing up of users and usage and installed base and whatnot. Still, I think it's was a useful exercise and I think it's probably not too far from reality.

What do you think? Does 12M-16M feel right to you? Got a better way to estimate that you'd be willing to share here? Got better data on any of the points I used in my estimate? Think this is all nonsense and not worth thinking about? Please share with us in the comments.

Diamonds are no longer the hardest of the known naturally occurring materials. Apparently the heat and pressure of asteroid impacts and volcanic eruptions can produce significantly harder rocks.

But, are they as pretty?

I've been wondering for a while what we'd use to build the diamond fishing nets out of and maybe now I've got my answer ;)

Not all shit.

Silvia Pfeiffer's recent post on open codecs is a good one.

She really nails why open video is important for an Open Web. I encourage you all to read it through a couple of times.

In just a matter of months, 200+ million people will have a Firefox Web browser that supports an open video codec. It's my hope, and a hope shared with many others, that Firefox can make a significant impact rationalizing video, which has to date been a heavily proprietary and fragmented set of features, with the free, open, and participatory Web.

Deanna, chaque jour je t'aime davantage, aujourd'hui plus qu'hier et bien moins que demain.

I meant to blog this a few days ago but it got lost in my drafts folder.

Firefox 3 surpassed I.E. 6 in global browser usage^* last week.

I think that's a pretty serious milestone. This isn't "all Firefoxes" but a specific version of Firefox (and its Gecko platform).

When you look at it from a developer's perspective the versions actually matter. Coding for I.E. 7 is a lot different than coding for I.E. 6 and coding for Firefox 3 is different than coding for Firefox 2.

With the modern browsers, those differences are fewer between versions and backward compatibility should be better, but it's still a real concern.

Considering how pervasive I.E. 6 was for so many years, it sure is great to see it's finally fallen to third place.

It'll be quite a while before it falls to fourth place; though I do think it will probably break through the 10% floor near the end of this year. By the end of 2011, it could fall to 4th place but it's difficult to say because Safari share continues to be quite fragmented across versions.

*Global usage data from NetApplications Browser Market Share report

Check this out. I think it's pretty badass.

If you want to know what's really going on with Mozilla and the EU Microsoft thing, check out this very helpful FAQ. It will hopefully go some ways to clearing up some confusion that seems to be pervasive in recent reporting on the issue.

I just got through reading a insightful post called Gracefully Upgrading, over at Burningbird's RealTech (Shelley Powers).

It got me thinking. That's pretty much my approach. I've been striving for a clean, effective, and simple design here, with the occasional ill-conceived and annoying flourishes for the more modern browsers.

My latest annoyance is the tilting of post titles using the recently implemented in Firefox CSS transforms. I also threw in there a little bit of text-shadow and I've been considering some SVG flights of fancy as well.

As always, I'm totally open to design suggestions. Maybe something with font-stretch would be nice ;-)

I want to start off by saying, again, that the comments I post here are my own opinion and shouldn't be taken as the official position of "Mozilla".

I have a question. It's pretty simple but I'm really looking for serious responses from any of you that have been following this issue here or elsewhere.

Let's assume there are two possible outcomes with this EU/EC Microsoft thing. The first outcome is that Microsoft convinces the EU that Microsoft has done nothing wrong and the whole thing goes away. The second outcome is that Microsoft doesn't convince the EU that Microsoft has done nothing wrong and the EU then must come up with a remedy.

In the first case, it's done with and every one moves on. In the second case, something happens.

Now to my question. If the second case happens and the EU is going to develop and enforce a remedy, would you rather that remedy be informed by the other browser vendors (say Mozilla, Apple, Opera, Google, Maxthon, etc.) or would you rather the EU come up with something on its own?

I'm not asking you whether or not you think the EU involvement is a good thing or a bad thing. I'm asking you, now that it's happening, would you rather they go it alone and try to figure this all out themselves or would you rather that other browser vendors had a chance to participate.

My take on it is that it's possible that, if left to figure it out on their own, the EU could come up with a remedy that's far worse than the problem it seeks to solve. I could come up with several of those on my own. Many of you have shared your view of possible horrible remedies here and elsewhere.

I personally would rather that a bad remedy doesn't happen. I think that most people at Mozilla and most interested other browser vendors agree with that. I assume that you all would agree too but that's why I'm asking.

I mean, could you imagine if the EU told Microsoft that they had to ship an OS with no browser or no Internet functionality at all? That would be horrible for everyone. I, as a concerned browser user and as an advocate of hundreds of millions of browser users, would really really want to be in a position to know if something like that was going to happen and I'd want to be able to tell the EU that I think it's a bad idea.

What's your take? Would you rather (personally, or Mozilla on your behalf) know what's happening and be a part of the discussion or would you rather sit on the sidelines and just wait to see what happens?

update: obviously I come at this from the perspective of someone involved with browsers. David Illsley makes a good point in the comments about others having an interest in this issue. So, let me expand slightly and rephrase the question.

Are you for or against third parties with an interest in this issue (say, browser vendors, OEMs, standards bodies, etc.) getting directly involved? A follow-up for those who are against it: What is the better outcome you expect to come from the EU designing a remedy without input from those "interested third parties"?

Saying it, or even printing it, Paul, doesn't make it so.

Open source champion and industry darling Mozilla Corporation has joined the European Commission's demand that Microsoft remove its Internet Explorer (IE) browser from Windows.

That's just not true. Mozilla hasn't called for any specific remedy nor has Mozilla endorsed any proposed remedies.

As some of you all know, I'm been working for a couple years on a project called Air Mozilla. The goal of Air Mozilla was to build on some of what we did with the Firefox 1.0 event and streaming audio and live chat. In 2007, we launched our first streaming video event, Air Mozilla Live, an Interview with Mitchell Baker. That became a regular happening and we produced dozens of live video programs with people from all over the Mozilla project. Then, last year I added our project wide Mozilla Status Meetings to the Air Mozilla schedule.

But with Firefox 3.1 in beta testing, with native Theora video support and the very cool new HTML 5 <video> element, it didn't seem like the best thing to continue streaming our Air Mozilla broadcasts using a Flash-based system. Not only would we be missing the opportunity to test our new features ;-) but we'd be missing the chance to start advocating for a better way, an Open Video way.

It's still very early and the site is just a draft, but I figured I'd announce it sooner rather than later and encourage all of you to take a look at the new Firefox-powered Air Mozilla. The next scheduled live event will be Monday's Mozilla Status Meeting, but I'm also lining up some new interview shows and hopefully some more Mozilla Labs discussions. Until then, you can watch recordings of several of the previous live events.

You'll need Firefox 3.1 beta or newer to play these videos (or a separate client like VLC.) There's still a lot of design work to be done -- this is just the draft to test out the basic functionality, but I'm interested in your feedback.

And for the curious, my set-up is pretty simple. I'm Theora encoding DV from a Panasonic AG-DVX100B on my MBP, pushing that to an IceCast server, and then using the <video> tag to display it on the page. The recordings are from the live stream and could have been somewhat better quality if I wasn't trying to live stream them :-)

Exciting news from our good friends at the Participatory Culture Foundation: Miro 2 is here!

For those of you who don't know, Miro is a very cool desktop program that makes it super-easy to find and watch video content from the Internet.

(You can even subscribe to bittorrent rss feeds! How did I ever get along without that?!)

There's just no cooler video tool available for locating, acquiring, and watching TV shows, movies, YouTube shorts, just about anything.

Oh, and the people behind Miro are awesome. They're a not for profit organization with a global community of volunteers (just like Mozilla!) working their butts off to create a more open and democratic video space.

Get Miro, and if you like it and the mission of the PCF appeals to you, get involved.

A few days ago I linked to a post by Mitchell Baker about the European Commission and Microsoft. Since then, there have been a lot of comments there as well as a few other blog posts and articles and their associated comments.

I've read a good number of silly suggestions, some personal insults, some obvious straw men, a bit of hand-wringing, and even a couple of good ideas, but I haven't seen a lot of serious discussion of the problem itself. I think that's a necessary pre-requisite to any discussion about possible remedies and I'm hoping this post will help start that discussion by asking you all to look at one specific piece of the problem.

How precisely has tying advantaged Microsoft.

(As I said above, I think it's only after answering this that we can reasonably discuss whether or not there is some way to level the playing field.)

To get at any kind of answer for this, I think one has to first understand something about the PC OEM channel.

The PC OEM channel (companies like HP, Dell, Lenovo, Toshiba, Sony, etc.) isn't very well understood by most people and that includes me. But it does seem like it's possible to make some progress discussing it even with a somewhat simplified frame.

OEMs build personal computers by assembling a lot of different pieces. PC have hard drives, motherboards, processors, memory, optical drives, etc. Today, there is relatively healthy competition in the market for each of those components. There's competition among manufacturers of memory, processors, drives, etc, and not just across the different OEMs, but often within a single OEM. HP, for example, will regularly ship computers containing processors from both major manufacturers, Intel and AMD. And when you look across the different PC OEMs, the combinations and permutations of those components increase even more dramatically.

The fierce competition between Intel and AMD processors, between Seagate and Western Digital hard drives, between Crucial, Corsair and PNY memory makers (just a few of many examples) is one of the reasons that we're all enjoying much more powerful computers at considerably less cost today than we were a decade ago.

There is one component of the PC computer, however, that doesn't face competition in the PC OEM channel and that's the operating system. Right now, Microsoft Windows is the overwhelming dominant OS for each of HP, Dell, Lenovo, Toshiba, Sony, etc. It is this overwhelming dominance of the PC operating system business that led both the United States and the European Commission to determine that Microsoft does indeed have a monopoly and so must abide by a special set of laws and regulations that apply to companies with monopolies.

One of those regulations, in really simple terms, is that they cannot use that monopoly to advantage themselves in other markets.

So how precisely does Microsoft's monopoly in PC operating systems advantage Internet Explorer?

To me, it looks rather simple. Because the PC OEMS are already distributing Windows (and paying Microsoft for that privilege) anything that Microsoft adds to Windows gets "free" distribution and placement through the OEMs. It's this free distribution and placement by the PC OEMs that other browser vendors just cannot compete with. There is no equivalent alternative channel to reach customers and there is no serious incentive, except perhaps cash, for the PC OEMs to offer an alternative to Microsoft's bundled programs.

(As an aside, the only other Web browser besides Firefox to gain significant share is Apple's Safari browser which enjoys a similar distribution and placement advantage on the Mac OS X operating system as I.E. does on Windows.)

Some PC OEMs are willing to ship additional software beyond what they get from the Microsoft Windows bundle. Unfortunately that's cost prohibitive for even the wealthiest browser vendors because the OEMs charge ridiculous amounts for that distribution and simply don't offer the kind of desktop placement that Microsoft gets with Internet Explorer.

So we've got a tilted, twisted, screwed up playing field where Microsoft, thanks to its operating system monopoly, gets free distribution and optimal placement while all of the other browser vendors have to pay exorbitantly for distribution and even then cannot get optimal placement.

This is a pretty serious problem. And it's a problem that doesn't seem to offer any obvious fixes.

There are of course many other factors that play a role in the Web browser market, but all other factors aside, the PC OEM channel is no doubt the biggest and most entrenched advantage for Microsoft and a very serious disadvantage for Mozilla, Opera, Google, or anyone else trying to compete with Microsoft's Internet Explorer.

So, if we just look at this specific piece of the system, Microsoft's PC OEM cost of distribution advantage, is there anything that can be done?

Are there any changes that could either remove Microsoft's advantage, or remediate other vendors' disadvantage, or some combination of the two?

What say you all?

For this post (there may be others later) I'm only interested in comments that address this question or offer further insight into the PC OEM channel. Please do not go off topic. Thanks.

Every once in a while I run across a prediction so amazing that I can't help but marvel.

Today's bit of brilliance comes to you from someone named Dave at a Web site called Scrpting News.

In the struggle to achieve simplicity for OpenID, I came to the conclusion that it has to be built into each web browser.... Imho, the browsers that need to do this are: Firefox, MSIE, Safari and Chrome. If they do it the rest will follow.

Yes, Dave, if the browsers that combined represent ~99% of the Internet all agree to implement something, it's a safe bet that the others will follow.

Way to go out on a limb there, Dave. Such boldness.

I've said here some of what Mitchell makes much clearer at her latest blog post.

This is the part that really stands out for me and echoes a lot of what I've been trying to say to people claiming that the Web browser market is healthy and competitive.

Equally important, the success of Mozilla and Firefox does not indicate a healthy marketplace for competitive products. Mozilla is a non-profit organization; a worldwide movement of people who strive to build the Internet we want to live in. I am convinced that we could not have been, and will not be, successful except as a public benefit organization living outside the commercial motivations. And I certainly hope that neither the EU nor any other government expects to maintain a healthy Internet ecosystem based on non-profits stepping in to correct market deficiencies.

Go read the rest if you haven't.

No matter the successes we have with Firefox, without clear indications of a strong _commercial_ competitor, it's pretty obvious to me that Microsoft's control of the PC OEM channel continues to give them unfair advantage.

(And no, Safari is not competing in any meaningful way with I.E. Against I.E., Safari has gained about one third of one percent over the course of almost two years.)

Just watched E.T. the Extra Terrestrial for the first time since I saw it in the theater almost 27 years ago.

Magical.

I'm really surprised how much of it stayed in my head for so long. Spielberg really does hold up well.

Henry Thomas' role was as convincing to me 27 years later as it was when I was 8. Drew Barrymore was, of course, perfectly cute.

The story and the effects hit me in all the same places they did so long ago. Funny, exciting, scary, and sad. Pretty amazing.

Next up, The Goonies, and Stand by Me.

If Opera manages to get this new JS engine into a shipping product any time soon, that will definitely leave IE in a sad class all its own. Go Opera. Go fast!

Lots of other blogs and tech pubs have articles up today on the latest market share numbers that don't paint a pretty picture for I.E.

CIO.com » Microsoft's Internet Explorer Loses More Share, Slides to New Low  |  Inquisitr.com » Internet Explorer Loses More Ground While Other Browsers Grow  |  The Unofficial Apple Weblog (TUAW) » Apple market share continues to climb, Windows drops  |  Ars Technica » Report: Apple nabs (almost) 10% market in January  |  PC Magazine » IE's Loss is Firefox, Safari, Chrome's Gain  |  TechFlash: Seattle's Technology News Source » Chart: Microsoft's IE market share slips for sixth straight month  |  CNET News » IE slips further as Firefox, Safari, Chrome gain  |  vnunet.com » Windows users show signs of defection  |  The Inquirer » Linux, Mac OS gain from Windows losses  |  Infopackets.com » Internet Explorer Loses Ground in Browser Battle  |  MacUser News » OS X nabs 10% of the net  |  FavBrowser » Web Browsers Nostradamus, Year 2010  |  IT PRO » Mac OS X and Microsoft's betas gain market share  |  FavBrowser » Internet Explorer, Opera Loses, Firefox, Safari, Chrome Gains, Jan 2009  |  IGM » Mac OS nears 10% Web Usage; Safari Gains at the Expense of IE, faster than Firefox; iPhone Gains on Linux  |  National Business Review (NBR) » Microsoft browser share hits record low  |  seattlepi.com » Chart: Internet Explorer's market share keeps falling  |  VentureBeat » Internet Explorer’s browser market share shrinks — because IE6 is finally dying  |  Software Journal » IE market share drops again  |  PC World » Projection: IE’s Days of Domination Are Numbered  |  Digital Daily | AllThingsD » IE: Ya Slippin’  |  Martin English Whatsup » Internet explorer losing ground to firefox and safari  |  Software As She’s Developed » IE Market Share Slips  |  ReadWriteWeb » Soon, Majority of Web Users Will No Longer Use IE  |  Download Squad » Internet Explorer market share falling like a ton of bricks  |  Tech Web » Mac OS X Share Up, Windows Down

I'm sure there are some that I missed, so if you know of others, please post them in the comments here.

The NetApplications January numbers are out and tell pretty much the story I predicted last week. You could go read that or just keep reading here to see my assessment with the final numbers.

November and December were really strong months for Firefox. Many suggested those gains were only attributable to increased home usage over the holidays -- and that Firefox would fall back to pre-holiday levels in January.

That didn't happen. Firefox not only held onto the gains it made at the end of 2008, it added a fraction of a point to finish January at 21.53%.

Chrome's now been available for 5 months, 2 of them as a finished product. It ends January at at a somewhat uninspiring 1.12% of the global market. Chrome still hasn't been able to put in full a month that matches the excitement and usage of those first few days way back in September when it hit 1.16%.

Now all of that's probably from website downloads so we'll have to wait and see if they can do significantly better with Google Pack and that OEM distribution everyone's been talking about. Maybe February will be their big month?

Safari continues to ride Mac OS X adoption to increased usage share and closes January with 8.29% of the global browsing market. Nothing really unexpected there given Mac's recent strong showings.

Opera actually lost market share but so little that it doesn't require any generosity to say it was basically flat at 0.70%. The good news for Opera, as I said a while back, is that with the total browser market growing so quickly, even with their share remaining flat or falling slightly, their absolute number of users is still increasing and so probably is their Google and Ask search revenue.

Finally, and no surprises here, I.E. is the big looser again in January. It drops another half of a point this month to end January at a new low f 67.55%. Every month in recent memory has been a negative one for Internet Explorer's total share. I like that trend. I expect we'll see another pretty bad month in February before I.E. 8 ships, something might slow their fall a bit.

update: I almost forgot one of the most exciting bits of information in the latest report. Internet Explorer 6, the scourge of the Web, has finally dropped under 20% global share.

At the rate it's falling, we'll surely see I.E.6 dip below 10% by the end of 2009. It can't happen soon enough. In addition to the sharp decline of I.E. 6, Microsoft isn't having a lot of luck with I.E. 7 either. It's been basically flat at about 47% for the last six months. We'll have to wait and see if I.E. 8 is enough to get them back some momentum.

NetApplication's Global Market Share Statistics is based on aggregate data from ecommerce, corporate, and content websites from dozens of countries in regions including North America, South America, Western Europe, Australia, the Pacific Rim, and Parts of Asia and represents approximately 160 million unique monthly visitors.

update 2: Looks like I jumped the gun a little bit. I've adjusted the numbers to match what I believe is the final tally.