October 2004 Archives

Watching the US presidential election has never been so interesting. Actually way more interesting than those of many banana republics out there. Enough for the OSCE's Office for Democratic Institutions and Human Rights (ODIHR) to deploy an Election Observation Mission there (and they have already published a puzzling report [PDF] following their first evaluation visit).

After Bush who bans everybody but North Americans from his site (but I can still see ya, Junior!), now 58,000 Florida ballot papers go missing just days before the deadline, leading to a flurry of lawsuits. Florida, governed by Jeb Bush, once against will shine with hints of voter fraud, dirty tricks, facing election fracas by enforcing its bogus felon list again, enjoying e-vote snafu and bugaboo with electronic voting machines just as they are turned on (oh, and don't miss the chimpanzee demonstrating how to erase votes on the higly secured Diebold machines, but no need to worry now that a Diebold rep runs the elections in a California County). Oh, and I forgot about "cutting edge" user interface design.

Meanwhile in New Jersey, a coalition of private citizens and local elected officials plan to file a lawsuit to block the state's use of electronic voting machines, because they "cannot be relied upon to protect the fundamental right to vote".

And with more electoral codes than states in the US, there might be other surprises, such as if Colorado voters scrap the winner-takes-all electoral college system, in favour of a more proportional solution. Making every vote count in a democratic election, who would have thought of that before?

Indeed, let's stop complaining and recognize how far ahead the US are compared to our old democracies: technology-driven elections that can potentially be determined before a human being casts a vote, a system where you don't need to get the majority of votes to be elected, where the highest Justices are politically affiliated and can nominate the winner in case the lawsuits don't work... No, really, watching this from my middle-aged city in old Europe, I'm impressed.

The Register reports that Bush's website is blocking visitors from outside the US, citing an article from Netcraft which monitors the site from various points worldwide. Netcraft has noted that since Monday, www.georgewbush.com is reachable for visitors coming from within North America, while others (like me today) are greeted with a rude "Access Denied: You don't have permission to access http://georgewbush.com on this server." error message (a "403 forbidden" HTTP response for the geeks out there).

Bush's reelection campaign site is distributed via the Akamai CDN which provides a significant boost to bandwidth and resilience to flash crowds and attacks. This makes this ban even more surprising, IMO.

Well, may be that's the proof we were all waiting that, if re-elected, Bush would respect his promise to favor the moderate Republicans over the neocons, and consequently show some consideration for the foreigners out there!

Remember: promises bind only those who listen to them!

Update: ah! Being a customer of Akamai myself, I did the trivial test of finding the origin server via origin.georgewbush.com and, bingo! There you are, the origin server does not apply the ban. This further confirms my opinion that this ban is deliberate, and utterly stupid, politially speaking. Junior's site is also accessible at http://65.172.163.222/. Amateurs (private French joke).

Less than two weeks ago, Eric Meyer published a smart slideshow system he calls S⁵ for Simple Standards-Based Slide Show System. It didn't take long for Anil Dash to produce a S5 Movable Type template that builds on it to use MT for slideshows. The published template turns an entire weblog into a single slideshow with each post being a slide. It would be interesting to tweak it to host several slideshows into one weblog, for example by using categories or some other way (one could think of entering an entire presentation within a single post instead of one post per slide).

Thanks to Boom Chicago of Amsterdam, here is how Junior will win Florida thanks to Jeb Bush as governor, and voting machine maker Diebold contributing to the Republican party. Pick either Windows Media or QuickTime.

Some people at Amazon might have received an angry phone call from Cupertino. Amazon put Tiger on pre-order with a ship date of March 31, 2005. Oops!

Mac Central unveils a great hack to allow screen spanning on low-end Macs that are supposed to do only screen mirroring -- they have a picture of a new iMac G5 driving a second flat screen independantly from the iMac's own screen. It occurs that Apple ships video cards that have spanning capability, but locks them to mirroring through the firmware. The hack, which simply changes that firmware setting, is reversible and a utility is provided to handle it without requiring Terminal skills.

Tim Bray, what a Canadian computer programmer thinks about that American election:

The rest of the world is largely convinced that George W. Bush is an ignorant malevolent jackass. Here’s a story by way of highlight; we recently had an old friend over for dinner, and we were talking over the US election and he burst out “Americans are decent people and I like them, but if they go and re-elect that #*!^!% #$*!%* &*@!$%!, there’s just no excuse and the #!%$* with ’em!” I would say that there are a lot of people, not just in Canada, who feel about the same.

Indeed, I keep hearing the same all over the place in Europe.

Urchin is a software editor that provides web analytics software and is pretty well implanted in the hosting world. Until now, they provided software packages that one would run on their premises or through their ISP. But their latest version is now provided "on demand" in ASP mode, which I can't help but think that this confirms the trend to use web beacons over log analysis software for web analytics -- as I advised previously -- which also is a perfect candidate for an ASP business.

A bit of warning about Urchin 6 "On Demand", its base price is not the lowest in this market, it's good for only 100,000 page views per month(*) and the extra cost of $1 per mille is absolutely, utterly, ridiculously expensive compared to what you can get with more powerful services from more experienced competitors in this business. I reckon there is a gap below the 2M page views per month, a lower limit for the big players, and Urchin is attacking this entry-level market. But they should revise their scale. For a comparison, WebSideStory with its HitBoxProfessional service starts at $35/month for 50,000 page views per month and $0.65 per mille.

And a bigger warning: always test several solutions thoroughly. All suppliers have different offerings and ways to crunch the metrics, and it's not easy to find the one that suits your needs.

(*) which is the level of traffic this very site receives right now.

Adam Bosworth contrasts his definition of the platform to the one of Microsoft (emphasis is mine):

The platform of this decade isn't going to be around controlling hardware resources and rich UI. Nor do I think you're going to be able to charge for the platform per se. Instead, it is going to be around access to community, collaboration, and content. And it is going to be mass market in the way that the web is mass market, in the way that the iPod is mass market, in the way that a TV is mass market. Which means I think that it is going to be around services, not around boxes. I postulate, still, that 95% of the UI required for this world will be delivered over the browser for the same reason that we all still use a steering wheel in a car or have stayed with << < | > >> for so long. Everybody gets it. But this will, by definition, be an open platform because the main value it has is in delivering information and communication. Notice that the big players, Amazon, eBay, and Google have already opened up their information through Web API's. It is Open Data coupled with Open Communication built on top of Open Source that will drive the future, not Longhorn.

I believe in most of it -- especially in services tuned for community, collaboration and content -- except for the "95% of the UI required for this world will be delivered over the browser" because, unless the web standards grow up to offer a significant improvement in terms of UI, the user experience coming out of a browser is still a disgrace compared to what any modern OS can offer to desktop applications. See iTunes for an example of what can be done with a desktop application connected to web services.

P.S. the title is purposely connected to my earlier posts about Browsers War II, since it's pretty clear that Microsoft has seen the "web as a platform" as a significant threat against its "Windows as The Platform" model, and therefore had to move from ignoring it to "embrace and extend" it.

Jeffrey Veen has written an excellent piece on making a better open source CMS. If you're into content management systems, go read it. If you're a CMS developer, stop what you're doing now, go read it and, for once, try to understand what Jeffrey says. Even if yours is not open source, those advices still apply.

I'm confident all those ugly systems we see today will eventually disappear. They are to content management what the markup-ridden text editors we had to use 20 years ago on terminals hooked to mini systems are to MS Word. In 20 years from now we'll have something a zillion times easier to use that's part of the standard office tools.

Meanwhile, it is still exceedingly desperating to see that getting a CMS developed in a reasonable budget and time frame is more complicated than launching a manned mission to Mars.

(For a little insight on this post, I invite you to read Towards Browsers War II.)

Technographics, or the statistics about the technologies used to access a web site, are part of the metrics that every webmaster should follow on a regular basis. Typical technographics include browsers and operating systems, as well as screen sizes and client-side technologies (such as plug-ins, cookies, javascript, java, etc.).

There has been a recent surge of reports that the market share of Internet Explorer is falling while those of alternative browsers such as Mozilla, Firefox or Safari are rising. According to WebSideStory (as reported by Linux World AU and Silicon.com) which aggregates statistics from 25 million unique browsers through its outfit StatMarket, IE's market share has slowly dropped from 95.73 percent on June 4 to 94.73 percent on July 6, while Mozilla and Netscape's combined market share has increased by 26 percent, rising from 3.21 percent of the market in June to 4.05 percent in July. According to WebSideStory, this is the first time such downward trend appears since they started measuring in 1999.

Personally, I started to observe something in January 2004 on Capgemini's main site that I think is worth sharing with my peers.

The following figures show the evolution of the market shares of the top five browsers -- respectively, Internet Explorer, the Mozilla family, Safari, Opera and Konqueror (aggregated by versions and platforms) -- used by the visitors of www.capgemini.com between April 2003 and September 2004:

Taking out a discrepancy in Sept 2003 (the spike for Safari seems to coincide with the release of Safari 1.1, not sure about the spike for Opera though), one can observe, starting around January 2004, a clear and increasing fall in IE's share with a simultaneous and accelerating increase for the Mozilla family plus a rather constant increase for Safari. I think I can reasonably infer that there is a real trend installed here and that the share of IE is being eroded mostly by Mozilla and, to a lesser extent, by Safari. The fact that this trend comes six months before WebSideStory starts seeing it on its broad base may be related to the technical nature of this site's audience, people who are bigger users of IT technologies than the average business user, more exposed therefore more sensitive to security risks, and probably more informed and open to alternatives. Among those people are also many influencers who can (and surely do) play a significant role to spread the word about those alternatives. However it should be noted that those visitors using their company computer usually have little to no influence on their "standard" set of software, which normally increases inertia and traditionally plays in favor of Microsoft (so this fact goes, intuitively, against the observable trend).

As I wrote before, statistics taken from a single web site should be taken with a grain of salt, especially the percentages above, as they relate only to the specific audience of this site. It is therefore important to put things in context, which is for this site:

Type of site and market

Corporate site, business-to-business, IT consulting/technology/outsourcing services

Audience

Clients, prospects, job seekers, analysts, journalists, shareholders, members of the company -- mostly business people with a majority of technical/computer-savvy profiles. International audience coming from all over the world

Top 4 Operating Systems (all versions)

Windows: 97.6%, Linux: 1.73%, Mac OS: 0.46%, Sun OS: 0.11% -- the market shares of OSs has remained quite stable during the same period, and therefore should not have had any significant influence on the change of browser

As a web manager I would like to see more stats from peers published on the web. It's one thing to hear the buzz on how great your life will be if you use web standards and forget about browsers, it's another to actually back up this idea with figures. That's why I'm doing this and I welcome links to similar stats (please add them in the comments or trackback this entry).

This image published by StatMarket (a subsidiary of web analytics outfit WebSideStory) is quite telling. Are we seeing a new browser war? Is Microsoft Internet Explorer starting to lose its hegemony on the web? Is Mozilla the new phoenix rising from the ashes of the Netscape browser?

A little irony aside, having crossed the path of Netscape at the time its browser was the hegemonic one, I've been looking at the browsers market shares on a regular basis with a professional interest for quite some time now. Back in 1997, the folks at Netscape had already identified one small but important fact of the IT life: more than 70% of the users would never change the default settings of their PC (OS and applications). There is no mystery why Netscape was keen to sign partnerships with PC manufacturers and get their browser installed by default with its default home page pointing to www.netscape.com. Alas, this very fact of life played mechanically in favor of Microsoft when it came with its own browser, Internet Explorer, non accidentally entrenched within Windows as the default choice (with MSN for home page). If 70% of your users are happy with basically not having to make a choice, better be the one who chose for them, isn't it? This is how Microsoft displaced Netscape in the browsers market shares (and triggered the monopoly trial, but this is another subject). The first browsers war ended with the clear victory of Microsoft and, ultimately, the death of Netscape. In the past two years, people remembered Netscape mostly to enjoy (with reason) the disappearance of its legacy browsers, and AOL put the last nail in the coffin in July 2003 by killing the Netscape browser division it inherited when Netscape was split between them and Sun.

What happened in the past years?

From the external observer looking towards Redmond, it seems that Microsoft is not in a hurry to develop its browsers, when it does not simply kill them like IE for Mac OS. IE for Windows is now quite old and there are no plans to release a new version until the next version of Windows, Longhorn, ships sometimes in 2006. So long for innovation. Updates to IE seem only triggered by more and more aggressive security issues that plague Windows PCs in droves.

Meanwhile, the Mozilla Foundation, born from the defunct Netscape browser division at AOL, continued its efforts to develop a comprehensive suite of internet products around its Gecko rendering engine, notably the Firefox web browser, the Thunderbird email client and the 200-pound Mozilla browser/email/IRC/newsgroup/HTML editing suite. With the ambition to appeal to the masses, they (at last) made a marketing effort and relooked their site and products to make them attractive to the non-geeks out there. Those efforts helped place the Mozilla products as a serious alternative to Microsoft's ones, serious enough that more and more influencers are suggesting to look for alternatives to IE (e.g. Walt Mossberg/WSJ, CNet switchers list, Gartner, or the Spread Firefox campaign).

Although this will be deemed as anecdotal by some, it should be noted that Apple's entrance in the browser market with Safari, and its aggressive push to spread it amongst the Mac OS users, led to the quick fall of the market share of Mac IE (but also, I think, somehow limited the spread of the Mozilla products on Mac OS).

What gives?

It can be easily argued that the balance has shifted in favor of the alternative solutions to IE. I also think that there is a limit to what users can accept when, at the same time, they are confronted with more and more evidence that there are better alternatives than what they are presented with by default, and a loss of confidence in how secure they can rest while surfing the net and reading their emails with Microsoft's products. The last stroke, i.e. the risk of being attacked by just viewing a picture, is quite scary for the layman. I would love to have refreshed statistics about what percentage of users never change the default software preferences, but I would be surprised if it stands at the same level as in 1997. Logically, IE (and possibly Outlook) is at high risk of falling.

Indeed, if we believe the recent reports, the market share of IE is falling in favor of alternative browsers, mainly the Mozilla family⁽¹⁾. Take your pick at: Linux World Australia, Silicon.com, the W3School stats or those of Ars Technica. I'm myself seeing a significant trend building in this direction on my biggest site (more on that later but you can easily infer which one I'm referring to). Note that stats excerpted from specific web sites are to be taken with a grain of salt, since they are never representative of more than those sites audiences. However, aggregated stats such as the ones provided by web analytics ASPs or prominent places are significant in terms of trends. (Unfortunately, the Google Zeitgeist does not show the browsers trends anymore.)

To conclude, it's always been dangerous to build a site to a specific browser, but it's never been more dangerous to believe that IE is the only browser that matters. You probably already know where I'm heading to: when building or redesign a web site, your only safe bet is to adopt web standards and a design that is open to the broadest possible set of clients. And not just browsers, mind you, but also spiders, newsreaders and who knows what people are using to pull your content out.

If Microsoft keeps its stance to not upgrade IE before Longhorn ships, the next two years of the Browsers War II will be very interesting to watch.

(1) i.e. all Gecko-based browsers: Mozilla, Firefox and Netscape.