May 2009 Archives

I don't know how these internet meme things get started, but maybe this is how.

Deanna and I were just talking about all the great places we've been able to visit and all the sights we'll never forget.

This evening, just as we're about to embark on our second trip to Japan, I asked her which were her favorites. To make it somewhat easier, I asked her about four categories: buildings/architecture, natural wonders, works of art, and animals.

It turns out this didn't make it easier because either one of us could list a half dozen for each category.

The key rule was that we saw the particular item "up close and personal" (and for our purposes, that we saw the item together.) and the instruction that helped us decide was "if you could have ever only seen one item from the category, which would it be?"

Here are her answers, followed by mine.

Deanna:
Architecture/Building: Notre Dame de Paris (Great Wall of China and Himeji Castle were next.)
Natural Wonder: Iguaçu Falls
Work of Art: Nike of Samothrace
Animal: Kangaroo

Asa:
Architecture/Building: St. Basil's Cathedral (Notre Dame a close second)
Natural Wonder: Denali/Mount McKinley
Work of Art: Nike of Samothrace
Animal: Green Sea Turtle (second place would be the Tucans of Brazil )

These were all shared experiences and either one of us might have listed the other's choices without too much prodding.

How would you answer?

We signed about a million papers today and when things are all processed over the next few days, we'll finally be home owners.

It's nine years to the day since we moved from Texas to the San Francisco Peninsula. Kind of cool. Now we're off to Japan for a while. More on that later.

The good folks over at Mozilla labs have just launched a project called Jetpack.

Jetpack is a newly formed initiative and experiment in using open Web technologies to enhance the browser, with the goal of allowing anyone who can build a Web site to participate in making the Web a better place to work, communicate and play.

Firefox extensions have, since their inception, been somewhere between Web development and desktop app development in their complexity and skill requirements. This project is going to really push the barriers down for Web developers and I think open up Firefox extension development to tens of thousands of additional contributors.

If you think the 7,000+ Firefox add-ons available now is pretty awesome, just wait 'till Jetpack takes off!

But don't just sit around waiting. Get involved today!

Oh, and digg it.

We won! Again! :-)

2008 was the year of Firefox 3, which sports an updated look and an updated rendering engine that's several times faster than the one that came before it. Over 8 million copies of the software were downloaded in a 24-hour period in late June as part of Firefox's download day celebration. In 2009 the company continues to push out further revisions that fix bugs and add new features.

It just seems worth saying, and I hope it doesn't come across as too preachy, but the astronauts servicing Hubble are literally risking their lives, from the moment they climb on top of the world's most powerful rocket, through the spacewalks, and back to Earth, and all of humanity owes them thanks for the risks they're taking to advance our understanding of the universe.

It's pretty amazing that there are people who are not only willing, but excited and proud, to put their lives on the line for science and discovery. This isn't like working on a terrestrial telescope. This is life and death for people with families and friends and communities and a thousand meaningful connections. They're putting that all on the line so that we can advance as a civilization and I'm deeply thankful.

Thank you. You're appreciated.

Deanna and I are big fans of Party Down. If you haven't been watching it and you get the chance, it's not just solid entertainment, it's top of the line comedy written and starring some really bright talent. Do check it out.

Not without problems, but the second service to Hubble was a success. With new gyros in place, Hubble is in great shape to give us years more service.

Yay!!

The first spacewalk servicing was a great success. Mission Specialists John Grunsfeld and Andrew Feustel installed the new Wide Field 3 camera, and a critical data handling computer.

This was the most important spacewalk. With these changes, Hubble is nearing good shape to provide years more science.

Tomorrow, they're going to replace the 6 gyroscopes and three batteries. The following day, they'll be installing the Cosmic Origins Spectrograph and starting attempts to repair the totally awesome ACS. After that, they'll try to repair the Space Telescope Imaging Spectrograph, add insulation, and replace the final battery.

So far so good. Keep your fingers crossed. They're gonna need a lot of luck getting ACS up and running again.

A few sites picked up on the graph I posted on longterm browser usage trends. Among them was Mashable which had the amusing title "RIP Internet Explorer: 1995-2021".

I'm drawing attention to this post because it repeats a very common and very wrong assertion. Mashable writes, "Whether it’s greater security, more reliability, extensions, or something else, it’s clear that more and more people are choosing Firefox, Safari, and Chrome over its slowly sinking competitor."

The truth is that more and more people are not choosing Safari. As a matter of fact, fewer and fewer people are choosing Safari.

Safari, just like IE, gets virtually all of its usage by shipping as the bundled and default browser with its operating system. (Safari's usage share on Windows, where does represent an actual "choice" is even sadder than Opera's -- completely irrelevant.)

The actual case is that we have a bundled browser on Mac that's losing share to Firefox over time. What does this look like? Well, according to Net Applications, here's what it looks like.

data from Net Applications browser usage reports

As you can see pretty clearly by the trend lines, Safari is losing share on Mac and Firefox is gaining share on Mac.

That's not "more and more people choosing Safari." Exactly the opposite, more and more people are opting out of Safari and choosing Firefox instead.

Now, Safari usage is growing. That's plain from my previous graph. The explanation, though, is not more people choosing Safari; it's more people choosing Mac. That's a very different thing. Having chosen Mac, Safari users, about 27% of them, have opted out of the bundled and default browser and instead chosen Firefox.

That's an even higher conversion to Firefox rate than we're seeing on Windows.

I'm not sure how, without just ignoring all evidence to the contrary, anyone can claim that more and more people are choosing Safari when Safari -- as a real user choice -- is fairing not even as well as Internet Explorer.

As some of you who read this blog regularly are probably aware, Deanna and I are in the process of (trying) to buy a beautiful log cabin in the coastal redwoods of the San Francisco Peninsula.

We were so very close at the end of last month but things have gotten a bit more complicated over the last couple of weeks.

It turns out that we have impeccable credit and we'd found a lender that seemed happy to give us whatever loan we needed. Unfortunately, when the loan went into its final review, the lender appraised the home at less than what we, the seller, and an independent appraiser had agreed it was worth.

Mortgage lenders are obviously a bit timid lately. After a year of sub-prime borrowers defaulting at unprecedented levels, and with the option ARM recasts about to hit hard later this year, lenders have pulled back on all but the safest of loans. That means that if we wanted to buy a million dollar condo in Mountain View, we could probably do it tomorrow. But, trying to buy a log cabin in the woods 45 minutes away from Mountain View is a whole other story.

The place we found and fell in love with is quite special. It's the kind of place that has a somewhat limited market appeal and wouldn't be easy for a lender to get rid of if its occupants were to default. Deanna and I are very qualified borrowers but lenders just don't want to take any chances with properties that might be difficult to sell if they went into default. Because of that, lenders are undervaluing "difficult" properties to keep their risk minimized.

The good news is that we've become quite good friends with the sellers over the last couple of months -- a wonderful couple that has put their heart and soul into this home for the last 15 years, and after meeting with them today it looks like we've got a way to make this all work out.

While it isn't a done deal until the deal is really done, we're optimistic that we'll be the proud owners and occupants of the most beautiful little redwood log cabin on the Peninsula by the end of this month.

Yesterday I blogged that I was concerned about some unexplained spikes in the StatCounter brower share reporting. A few commenter suggested I'd just gotten some intra-day unrepresentative measure and that it would return to normal within a few hours.

Well, it's not getting any better. IE 6, according to StatCounter, has grown well over seven points in just two days. This just doesn't make any sense.

What explains IE 6 making the biggest jump of any browser in the last year or two? Nothing, that's what.

There's something really wrong here and I sure hope that StatCounter sorts it out quickly. It's just not possible that IE 6 is growing -- and growing several times faster than any other browser version out there.

I've read all of the reports, and the reports of the reports, and as far as I can tell Atlantis is just fine.

The small, superficial pits in the ceramic thermal tiles where the right wing joins the body of the craft are nothing to worry about. The pits are probably the result of some insulating foam that broke off from the fuel tank during the launch and just barely grazed the right wing.

The tiles on this part of the craft are some of the thickest and most secure. Atlantis is in great shape and the Hubble repair mission should go forward without any craft safety issues.

Today I put together a chart of browser usage share for the major players, Internet Explorer, Firefox, Safari, and Chrome. It charts the usage breakdown from autumn of 2004 when Firefox 1.0 was released through last month.

data from Net Applications Browser Market Share Report

I've annotated the chart with the major browser version releases (please let me know if you see any errors there) and what's interesting here is that browser releases aren't having any major impact on the macro trends.

Now, this isn't to say that browser releases don't matter. I could imagine a chart that looks radically different were Microsoft or Firefox to not have had the releases they did in this period.

Still, it's interesting that the trends are so very linear and that major releases from Microsoft and Mozilla don't match even the impact of the nominal seasonal changes.

What do you see in this data?

More than three years after Mozilla's community produced hundreds of amazing Firefox television ads and then rallied to fund their placement on TV, Google has produced a few ads of its own and is using its own ad network to air them.

I'm obviously partial to the Firefox ads which were generated by our amazing community and I think they not only beat the Google ad in terms of inspiration but also in production quality. (There's something genuine in them that's missing from Google's professionally produced content.) Some of you may remember Daredevil, Whee!, and This is Hot. Well, those were just a taste of all the great ads that made their debut way back in the Firefox 1.5 era. If you'd like to see more, head over to Firefox Flicks and check out a couple hundred additional spots :-)

Firefox really is a pioneer in terms of grass-roots, community-based marketing for software. It was Spread Firefox, almost five years old now, that really launched the ideas that so many other companies and organizations employ today. Chris Beard, Bart Decrem, Blake Ross, Chris Messina, and countless others don't get nearly enough credit for inventing Spread Firefox and revolutionizing how software reaches users. And Mary, David, Alix, John, Jay, Jamey, Paul, Melissa, Otto, Ken, Tara, (I'm sure I'm missing a few) and the tens of thousands of other Spread Firefox contributors over the years since have continued to break new ground designing, building, and launching some amazing community campaigns.

The New York Times ad, Firefox at 50 Million, Firefox Flicks, the Firefox Crop Circle, Mozilla Campus Reps, Operation Firefox, Mozilla 24, Download Day 2008, Firefox Affiliates, the Community Store, and many more community marketing efforts like those have set a high bar for the rest of the industry.

It'll be interesting to see if the other browser vendors continue to follow our lead here. I certainly think it's a great improvement from the "old ways" and I couldn't be more proud to have played a role in helping to usher in this new era of engagement and participation in marketing.

StatCounter is a relative newcomer to the public Web share stats game and I've been following them pretty closely for about 8 or 9 months now.

I really like their quick and dynamic graphs (though I'd obviously prefer they use Open Web tech rather than Flash) and their 4 billion page views across 3 million websites sounds like a pretty good sample if it's even close to evenly distributed globally.

The problem I have with their numbers is that they have a lot of unexplained spikes that I just can't make heads or tails of. For example, they're reporting that IE 6 gained almost six and a half points in usage from yesterday to today. Looking at a longer-term trend, it just doesn't fit that IE6 would have such an unexplained resurgence.

The clear trend for IE 6, that I can see in the number of accesses to the Mozilla websites, the number of IE 6 users downloading Firefox, the usage percentages from other stats packages, and the obvious movement from older versions of IE to newer versions, is that it's dropping at a steady rate.

Now, in IE 6 does vary more than any other browser between weekend and weekday usage. The remaining ~20 share that IE 6 holds has a much stronger enterprise skew so the "work versus home" difference is more pronounced than with any other browser or browser version. You can see this in the daily stats at both StatCounter and Net Applications so I'm not surprised to see the difference cause a pretty big zig zag up and down as the weeks progress, but this just doesn't look right to me:

I'm really excited to see another stats package publishing its metrics to the public. It's a really helpful service and I hope that more programs (like omniture and sitemeter) will start publishing their aggregate data for public consumption. But. As important as it is to have more data, it's also important that the data is consistent and trendable. I don't know what's up with these random spikes at StatCounter, but I hope that with some time they get things ironed out.

When they do get the bumps (literally) ironed out, I will start to include them more in my regular browser landscape updates and analysis.

Lots of news stories on the Atlantis mission to Hubble are focused on the highly improbably and would be novel failure case where Atlantis needs rescuing by Endeavor. (Because Hubble and ISS are not in similar orbits, in the case of a serious problem with Atlantis, it cannot seek refuge at the ISS.)

That's a shame because the real excitement here is that as the Shuttle program winds down (the final shuttle flight will be next year) Hubble is getting its fifth and last servicing.

Hubble is gonna get a pair of new cameras, including an awesome new Wide Field Camera and the highly anticipated Cosmic Origins Spectrograph. In addition to the upgrades, Hubble's failing guidance system is going to be replaced, along with a new battery and a new downlink computer.

And that's not all. If we're (well, if the astronauts are) really lucky, then we'll also be able to bring the ACS back online. The Advanced Camera for Surveys, which has been down since January of 2007, is responsible for some of the most amazing science and most beautiful pictures over the entire Hubble mission.

That's going to be a tricky repair and I honestly think it's going to take a lot of luck if it's going to happen, but it sure would be sweet.

For those of you who don't know, Theora is the first video codec to be included in Firefox's Open Video implementation. The Theora decoder in Firefox 3.5 betas is what you'll get in Firefox 3.5 final but the Theora encoder, codenamed Thusnelda, is evolving quickly and that's going to mean much higher quality Theora video in the near future. That's going to mean higher quality Open Video in Firefox 3.5!!

The latest Thusnelda update has some great information, but my favorite bit is this headline: "test versions of Thusnelda pulling *ahead* of h264 in terms of objective [PSNR] quality as bitrate increases."

Awesome!

Hit the HQ button and turn up the volume. You won't be disappointed.

In previous blog posts, John Lilly has explained that we cannot identify precisely how many users Firefox has because there is nothing uniquely identifying in the Firefox browser and that's something we're not interested in changing just for the sake of better accounting.

Because of that, we can only estimate. The best method we've come up with is to count the number of security "pings" we get in one day. (Firefox pings a Mozilla server for security changes about once per day) and multiply that active daily users number by three to get active monthly users. (See John's blog post for more on this.)

Over the last couple of weeks, we averaged about 90 million of these pings per day. Multiplied by three, that gives us an active user base of approximately 270 million.

I think that number is probably a bit conservative but it's very likely in the ballpark. (See my earlier post What Does the Web Look Like for some broader estimates.)

We could get an even better number and make this all a bit less opaque and confusing if the other major browser vendors, Microsoft and Apple, would release some of their numbers and estimates. I can't see what value there is in keeping such things a secret so maybe we'll get lucky and they'll open up some.

(digg it)

Here's the graphic that some of you have been asking for:

And yes, the active daily users (ADU) number is an average. Yesterday we had over 98 million ADUs but, as I've discussed in a previous post, total Firefox and total Web users are highest on weekdays but drop sharply on weekends.

It really would be great if Apple and Microsoft would disclose more of their metrics. We'd be able to understand how the Web trends so much better. Until then, at least there's Mozilla metrics :-)

The innovative folks over at the Participatory Culture Foundations have a cool fundraiser going. I think it's a brilliant idea and adopted a line myself.

One of the more interesting dynamics on Web that I've been following is the difference between users and usage on weekdays and weekends. We only understand part of this because we don't know user counts for other browsers, but I think we could have some interesting discussions and maybe even make some guesses at other browsers based on what we do know about Firefox.

Here's what we know:

From our own update ping metrics, we know that the total number of active Firefox browsers drops about 19% on weekends.

From Net Applications we see that Firefox usage goes up by about 10% on weekends. (Other usage tracking sites confirm the trend, though they differ in degree.)

So what's going on here? The number of active Firefox installs drops pretty seriously on weekends, but Firefox usage goes way up.

There's only one simple explanation, and that is that the total number of active users on the Web drops precipitously on weekends, but that decline isn't as steep for Firefox as it is for IE.

There are two reasons for this, I believe. First, I think that there are a lot of people who only (or mostly) access the Web from work and those work machines are locked down in terms of browser choice, heavily favoring IE. When those people go home on the weekend, the total number of people active on the Web drops dramatically and negatively impacts IE's active user number and IE's usage share.

The second reason, I think, is that people who access the Web from both work and home, while they may be stuck with IE at work, have made the choice to move to Firefox at home. So when those people go home, IE again loses users and share, but Firefox picks up some users and share.

I think it's probably more the former than the latter, but I wouldn't discount those home choice-makers completely. Are there other explanations for this dynamic?

Now, I'm just not good at math, but I'll bet some of you all are. Maybe one of you could help me take this a bit further.

As I said above, on weekends Firefox users drop by 19% while usage grows by 10%. We also know that IE usage drops 7% on weekends. If we just look at Firefox and IE for a moment, is it possible to work backwards and figure out how much IE's active users drops off on weekends?

We're one third of the way through 2009 and that's plenty of time for me to make a prediction wild ass guess at where the major browsers will end the year.

Firefox started the year with just over 21% of all browser usage, according to Net Applications. The overwhelming majority of that share was Firefox 3.0 users. My prediction is that Firefox ends 2009 with 26.5% share, with most users on Firefox 3.5.

Internet Explorer started 2009 with more than two thirds of all browser usage, about 68%. IE users were split better than two to one between IE 7 and IE 6. IE 8 was just released and is quickly taking share from IE 7. I'm predicting that Internet Explorer ends 2009 with a 62% share. I see IE 7 at about 26%, IE 8 at 22%, and IE 6 at 14%.

Safari was going pretty strong in late 2008, but started the new year on a slower note at about 8% usage. With the release of Safari 4.0 this year, I think we'll see Safari finally nudge up against 10%. I'm predicting an ending share at 9.5%. It'll be split pretty badly with Safari 3.0 holding a quarter percent, Safari 3.1 taking 1.75%, 3.5% for Safari 3.2, and about 4% for Safari 4.0.

Chrome started 2009 with a bit more than 1% of browser usage. I think we'll see Chrome hit 2.5-3% by the end of the year. Chrome's got a much shorter track record so it's harder to predict, but I wouldn't be surprised if they had 2.5% at their 1 year anniversary and 3% by the end of the year.

So those are my guesses. Where do you see browsers at the end of 2009?

About two weeks ago I posted my predictions for the Net Applications browser usage report for the month of April.

In that post I predicted that Firefox would end April at 22.55% usage. The final report says 22.48%. Not too far off. Firefox had a great weekend at Easter and an otherwise nominal month of solid growth.

I predicted that IE would drop 7/10ths of a point to 66.12%. I pretty much nailed that one. IE ended April with 66.10% of browser usage according to Net Applications. IE8 is starting to grow to significant numbers but not making up the share lost at IE7 and IE6.

Safari did end the month essentially flat, losing 2/100ths of a point where I had predicted it would gain 2/100ths of a point.

Chrome I predicted would gain 1/10th but with what I attribute to the heavy banner advertising on Google's homepage, they managed to gain nearly 2/10th in share. It looks like Google's huge advertising effort netted them about 1/10th of a point on top of their nominal growth. Not bad, but also not terribly good considering that they put banners for Chrome on the most visited site on the Web for the better part of a month.

I predicted that the perennial last-place finisher Opera would have a flat month. That was apparently too generous as Opera came in with a slight loss.

All in all, a pretty normal month. Firefox continues it grow at the expense of IE. The alternative browsers all sit mostly flat, with the exception of Google Chrome which showed some modest gains thanks to a month of serious advertising.