The end of an era: Internet Explorer drops below 50% of Web usage | Hacker News Reader

The end of an era: Internet Explorer drops below 50% of Web usage

arstechnica.com

238 pointsllambda89 comments

It had been a while since I stepped back and looked at IE6 usage, so I pulled some numbers off of a site I manage just now to see where things were.

In January 2007 IE6 represented 64% of total site visits.

In October 2011, IE6 represented 1.4% of visits.

Wow.

I don't think that stat is particularly meaningful. Your FF1 usage figures have I'm sure dropped off far more severely.

It's incredibly meaningful in deciding whether it makes sense for us to continue supporting IE6 or not. And IE7 appears to be not far behind.

Here's me thinking you were commenting on the story at hand ... silly me.

Meanwhile in Russia:

http://www.liveinternet.ru/stat/ru/browsers.html

IE is around 20%

I wonder why Russia leads the way moving off IE. My guess: in corporate sector pointy-haired bosses don't actually have control over which programs system administrators get to install, so they install (and users get to use) whatever they know works best.

Check out the Philippines: IE at 8.6%

http://gs.statcounter.com/#browser-PH-monthly-201011-201110

Most third world countries have low percent of IE users (almost always lower than the mean, generally much lower), it's not a strange phenomenon. IE usage works as a (very coarse) estimator to percent of the population with regular Internet access. Try it yourself.

I personally think it happens because in those countries, in order to have access to technology you need to be from the upper socioeconomic class. Usually those users have a higher level of understanding about technology and hence choose to customize their browser.

I know a few people from the "upper socioeconomic class" and they are even more likely to not upgrade/use "the best browser", and instead just use what came with their laptop/desktop/whatever. I'm talking GPs, Lawyers, Accountants, Investment Bankers.

I have no data to base my sweeping generalisation on, other than personal experience

In a lot of countries there are government web sites and services that forces citizens to use Windows and specific versions of IE. Some of them used Activex, vbscript etc, The worst example I remember is an important web service for drugstores in Turkey that required Microsoft Java VM!

Apparently this is not the case for Russia.

A weird case where the more developed nations are being held back by their early adoption!

It's not particularly unusual, other examples are developing nations jumping straight to online banking (skipping chequebooks) or mobile phones (skipping copper phone cables):

http://en.wikipedia.org/wiki/Leapfrogging

Have you ever tried to book a rail ticket in Turkey via the Turkish State Railways website?

"Using TCDD Online ticket selling project requires internet explorer 5.0 or higher upgrades!"

Similar deal with banks in South Korea typically requiring ActiveX plugins and the like: http://blog.mozilla.com/gen/2010/04/28/the-security-of-inter...

South Korean banks seem to be oblivious to security research and best practices. ActiveX plugins that install several buggy DLL libraries are only the tip of the iceberg. 64-bit IE fails to open most banks' websites. IE 9 does not work at all. I was even sent a codebook (it has a list of colour-coded numbers which hash to codes, on paper!) by my South Korean bank for internet banking. I thought it was a somewhat archaic practice but physical books of codes do constitute "something you have" for authentication. Then I discovered that my partner had the same book. I disabled internet banking after that.

In Russia there is (or at least, was) mass pirating of Windows -- maybe MS turned off the updates.

What does Windows update have to do with IE usage?

Anyway, "no updates for pirated Windows" is a myth I hear repeated, but last time I checked Windows Update doesn't require passing Windows Genuine Advantage check.

> What does Windows update have to do with IE usage?

If Windows Update didn't work, that would make it less likely that users upgrade IE, which increases the usage of older versions of IE. And users using older versions of IE might well choose to upgrade to some other browser, rather than to just a newer version of IE, particularly when they find that IE won't handle the sites they want to visit.

in russia, opera is decreasing too. it is always a wonder of mine, why from russia to kazakhstan; ex-soviets, there has been so much love for opera.

Opera worked well on crappy dial-up connections. Then, you begin to appreciate its other features (e.g. first browser with tabbed browsing etc.) Then you use it by force of habit until it pisses you off in a major way, like not rendering the sites that other browsers have no trouble with. That's my experience.

As recently as five years ago, I thought this would never happen. The argument was something like: "All Microsoft has to do is put out a good-enough browser. Every time people buy their next computer, IE will be the default and they won't bother to switch. The only way would be if Apple gets a big part of the desktop market, and that's very unlikely."

Wrong in so many big ways: - Completely missed mobile browsers - People aren't upgrading their desktops as often anymore - The browser is the most important app, so people are motivated to get the best one - Apple's growth

I love this industry.

I'd be interested in seeing figures comparing Europe to other continents. If you remember, Europe forced Microsoft to give the user a choice between browsers when Windows boots up for the first time.

Europe has more Firefox usage, but the rate of change in IE (or Firefox or Chrome) usage doesn't seem to have been affected materially by the government intervention.

It's also quite variable within the EU, compare UK to Germany for example.

http://gs.statcounter.com/#browser-eu-monthly-200807-201110

http://gs.statcounter.com/#browser-DE-monthly-200807-201110

http://gs.statcounter.com/#browser-GB-monthly-200807-201110

StatCounter (http://gs.statcounter.com/) paints a different picture where IE crossed that line an year ago. It's numbers are reasonably consistent with data from sites I manage.

Chrome is growing wildly in Brazil, envy us: http://gs.statcounter.com/#browser-BR-monthly-201010-201110

edit: Russia wins :(

This happens right as Microsoft's anti-trust restrictions are lifted and they are forcing OEMs to make it really hard to boot Linux. Not a good omen. The next Windows will have IE tightly integrated, but hopefully it has lost enough ground that things won't be like the old days.

Isn't this already the case with Metro in Win8? You can only use IE.

I suppose it's possible that they will try and force their market share back similar to "must not duplicate existing functionality" with apple stuff.

The problem with this whole thing is that generally capitalism and innovation work best when there are few restrictions on competition so I wonder what will happen if everything becomes so locked down by a few companies who will prioritise their products over competing ones even if the competing ones may have something else to offer.

Could this lead to mass stagnation like it did with IE6 but on a larger scale?

If this is the case then an open alternative such as Linux would actually become much more attractive to independant software vendors because there they can compete on merit.

metro is what I was referring to

Someone pinch me.

Is that a light I see at the end of the tunnel?

No, it's Google and their monopoly of your privacy.

Yep - you know it's the "end of days" when you look back on Microsoft as the hey-day of freedom in computers!

See the shiny aluminium laptop - don't worry about not being allowed to own the stuff on it - look shiny.

We have a funny logo some days - just enter your kid's details here for marketing purposes - look funny logo.

Give me a break. You're just making things up. By all means wear your tinfoil hat and keep your pitchfork sharp, but try to keep at least one toe on the ground.

Please. All that Google paranoia belongs in the same thread as UFOs and bigfoot. Let Google and Facebook's linguistic algorithms scan my every email. So long as there's no human doing it, I could care less.

I would like things better if I were the only anything scanning my emails, but we all know that will never happen.

I could never deal with that much spam.

Google feels no pity, no fear, no pain, and it will not stop, ever ....

Oh, wrong movie.

So has anyone found any privacy-compromising code in the open-source Chromium? Nope?

What about Firefox or Opera(which are mostly Google-funded now)? Nope?

Anyone? Bueller?

What's your expectation? That IE < 9 will sink to a number that you won't have to support? That clearly isn't going to happen for many, many, many years. Microsoft could file chapter 7, today, and you'd still see IE as a large-enough-to-matter browser share for years to come.

The really interesting piece of information is the percentage of clients that support advanced CSS3 attributes without needing any kind of polyfills, and have faster Javascript engines. That percentage appears to be going up very, very quickly. We've stopped designing for "IE first", and instead are targeting browser capabilities. Modernizr for development, then extremely targeted Javascript and CSS for production.

FYI, I estimated 5 years when I was talking on reddit about how IE damaged XHTML, and by then XHTML was already 10 years old.

Why not? Google doesn't support IE7 for Google Apps. They claim that they will only support the two most recent versions of any browser. Should we expect that they will drop IE8 support once IE10 ships? For personal use, people need to be nudged to use a modern browser. The average person probably just uses the browser that ships with their machine and doesn't even know to upgrade.

Depends how many XP users there will be left, the '2 versions' nonsense is just that, nonsense.

XP users can only use IE8 however. So XP market share and the number of their clients still using XP will drive what Google does and does not support.

Most of all, I'm happiest the world is moving forward to recognizing and rewarding more modern and updated browsers with their usage, whatever those might be.

IE wise, I'm a fan of not having to code around the weak links. Currently, I use enough standard libraries not to worry about IE too much. Still, between different IE versions, I even have to deal with bugs only specific to 6,7,8 which for some reason don't show up in other browsers.

The ability to use cutting edge stuff in modern browsers does get limited programming around the weakest links (IE) instead of the strongest.

0.02 :)

Stats for deviantart.com, which skews towards a younger audience: http://cl.ly/1v0w2Z1H0C2E28162n3W

I'm getting very similar numbers at ticalc.org, which skews hard toward high-school students.

I wonder if "acceptability at work" also would skew those vs ars or the internet at large. My gut would be that much of the IE browsing includes browsing at work.

The university site I manage still has IE in the lead. Luckily the majority are running IE8 with IE9 close behind.

Do you have stats for the different IE versions?

Yes: http://cl.ly/1U0j0V2T0Q410j3J1t0O

When we started developing our web-app 18 months ago we decided from the outset not to support IE. It's just too much of a headache for a heavy JavaScript site.

We develop to Chrome and Firefox. This basically gives us Safari and Opera, too.

Demographic still makes all the difference. I have one site that gets ~1m visits/year, targeting older folks. IE numbers: 2007, 86% 2010, 67% YTD, 62% (ie6 still 4%)

Not that I'm dismissing this statistic, but android browser is included in the pie chart so I'm guessing mobile safari (i.e. iPhone and iPad) are included in this sample as well.

In my opinion, this is a little misleading. Sure, IE use has decreased overall, but its deterioration is also due to a growing mobile user base. It's worth pointing out that there are likely much disparate data for the matter, a few of which are:

http://en.wikipedia.org/wiki/Usage_share_of_Web_browsers

http://gs.statcounter.com/

http://www.netmarketshare.com/

With the last one, browsers are broken down by desktop versus mobile/tablet and desktop usage still puts IE at over 50%. I guess it just depends on where one looks for the information.

Eh, I'm going the other direction. I've picked up IE9. It has a few small negatives, but it is pretty fast and light.

The irony is I occassionally have to install Firefox because some website isn't compatible with IE9.

>some website isn't compatible with IE9 //

Have you examples of where the website is using non-standards complaint HTML/CSS and this is causing IE9 to fail?

... you mean that IE9 isn't compatible with a website. It's been documented and my experience that while IE9 does fix somethings IE8 got wrong, it also breaks on lots of things even IE8 got right. It's a huge headache and not really any kind of step forward

IE9 is good! I say this with my web developer hat on - my current site works best in IE9. It’s got the smoothest animation by quite a long way.

Just a pity it didn’t get text-shadow.

Just yesterday I had a chat with a guy who's an entrepreneur in tech sector, and kept asking me: "What's wrong with IE? I'm using it for years and never had a problem? I don't want to install Chrome unless you can tell me what is the benefit of it." (and then he dismissed all the arguments I had for better browsers).

I kinda gave up on such people. Life is too short to convince someone who doesn't want to, and better spent with making awesome stuff. :)

With Microsoft only supporting ActiveX technology to enable most features of their Remote Web Workplace Sharepoint websites, IE is going to be around awhile still. Almost one quarter of my small business clients for my service business use Small Business Server.

And until I move away from Exchange Server (email), I still keep a Windows virtual environment going just to run Outlook on.

The web will be a better web when IE6 and IE7 don't need to be supported. Think of all the time that was wasted working to support them. Following Steve Jobs' analogy of saving lives through speeding up boot time a few seconds: how many people has IE6/7 killed?

The worldwide mobile browser market share chart is iffy. They're suggesting that Safari Mobile (i.e. iPhone + iPod + iPad) accounts for 62% of the market. The numbers are based on web-page visits to sites that are part of the 'net applications' network[1]. I suspect that Android, Blackberry and Symbian devices are being seriously undercounted, either because they're more likely to be sold with a limited data-plan or because they're more popular amoung users in countries that are less likely to visit these sites. In short: I'm sceptical that iOS is almost 2/3 of the worldwide mobile browser market.

[1] http://www.netmarketshare.com/mobile-methodology.aspx

Anecdotally I browsed the web as little as possible on my Nexus One. Prior to that on my 3GS and now on my 4S I browse the web a lot more, because Safari is far better than Android's browser. The screen being nicer certainly doesn't hurt, why does text look so horrible on the Nexus One at 480x800?

It sounds at least plausible to me. Not to mention that many people who buy iPhones want to do lots of stuff with it, whereas many Android customers just got it because it was free with their contract, not necessarily because they want to do everything it's capable of.

It measures page visits, it can't count all those symbian/bada browsers sitting unused in people's pockets. They're not in the market.

The "browser usage" market share is a very inaccurate and pointless stat by default. Do developers really care if a user uses 5 hours a day his browser or 1 hour? Wouldn't they care more about how many use a certain browser?

Especially on the mobile side, the stats get very skewed, because browser usage will be much higher on a tablet (read: iPad) than on a smartphone (iPhone/Android/Symbian/whatever). So say 10 million iPads could get as much "browser usage share" as 50 million iPhones. The iPads, although much fewer than iPhones, could easily double up the "iOS Safari" market share in this type of stat. That's why these stats don't make any sense.

I had to go back and re-read but couldn't find anything relating to it: why did you assume they used hours of usage as a metric versus the number of visits to the site?

... and good riddance!

Any tips on getting friends and relatives to switch? Mentioning security concerns seems to give the best results for me, but using fear makes me feel dirty.

Fresh install of whichever browser when they buy a new computer works best but that would make you feel dirty too ;)

well finally there will be some legit competition, not based on who has the underlying operating system but on real power, speed and features :)

They lead with:

"A couple of interesting things happened in the world of Web browser usage during October."

But at the end:

"Ars Technica's unusual usage figures are not surprising when considering its audience: visitors to the site tend to be technologists and early adopters"

Many people are likely to just assume they have a source other than their own usage statistics and ignore the long winded explanation at the bottom. The red flag for me was the mobile browser statistics.

That's because the bulk of the article (to which the headline applies), is based on Net Applications' globally-captured data ( http://www.netmarketshare.com/ ), while the last part of the article (to which that disclaimer applies), is based on the Ars site's browser usage stats.

Is there another reference to net applications aside from the graph footers? I had to go back to find what you mentioned but thanks for pointing that out.

The main website I run is at 56% IE users and that has gone up in the last 4 months instead of down.

Which version has gone up?

IE9

I wonder if at least some of them are enterprises who switched after the Asa/Kaply fiasco.

Glad I live to see this day. :)

W3Schools says that IE usage is at 22.9% in Sept 2011.

http://www.w3schools.com/browsers/browsers_stats.asp

The bias in the user population for a site that teaches web development is going to be huge. The link you provide states:

"W3Schools is a website for people with an interest for web technologies. These people are more interested in using alternative browsers than the average user. The average user tends to the browser that comes preinstalled with their computer, and do not seek out other browser alternatives."

So many people still on IE 6, I pity you!

I don't buy it.

My company manages a number of websites; the site with the most homogenous use across all demographics shows IE at 80% usage.

Here's a shot from the last couple of days: http://a.yfrog.com/img614/5741/ls8.png

Edit: IE usage breakdown: http://a.yfrog.com/img612/6052/f2yf.png

It's natural for many people here to think whatever they are doing must be what everyone else is doing, but the average Internet user does not have a MacBook Pro running Chrome. And the ones that spend the most money on the Internet certainly can't be pigeonholed that way.

Recently I worked with a high-fashion site in New York that you'd think must get most of their web hits from iPads and modern browsers, but no. More than half their traffic and sales came from IE6/7/8 users browsing the site from a corporate PC during their lunch hour.

Agreed. Our (mostly non-tech industry) audience shows a similar browser breakdown.

Wikipedia shows IE at only 35% of it's current traffic, and they are a top 10 site.

http://stats.wikimedia.org/wikimedia/squids/SquidReportClien...

It would be interesting to know the story behind some of those stats - I wonder why those people are using FF1.0? (or indeed if, perhaps that's just the UA string they set).

Is your site homogeneous across all demographics across the world?

Because IE share in the US is higher than its share worldwide. The article is about worldwide share.

US, certainly. I should have established the context.

I work for an e-commerce company and our breakdown with 200,000 unique visitors a month is 54% IE, 19.22% Safari, and 14.36% Firefox, and 8% Chrome.

IE 6 is < 1% and IE 7 is < 5%.

That sounds high. Most sites I've seen have been in the 30-50% range. I have a tech oriented site that has less than 5%.

That's interesting; we run a range of sites from law firms to a VOIP switch producer to a big hospital system. IE usage ranges from 60-85%. Oddly enough the VOIP switch producer has the highest IE utilization, but we did a bunch of analysis and basically they have a lot of vendors and clients using PCs that are locked into using IE.

In case anyone is interested. In japan (young people): http://i.imgur.com/HQVwv.jpg

Usage varies wildly according to demographics. The largest site I manage has IE at 44%, and it's a quite "non-technologist" market. My blog has IE at 1%, tied with Android, and a webapp is at 81% Chrome.

This has everything to do with the audience of a particular site.

Our small business portfolio websites see an IE share of above 75%.

Our "Web 2.0" web-app sites see shares as low as 9%.

Reply on news.ycombinator.com