I disagree with combining the IE5 and 5.5 results into one number. Although it may seem like a straightforward matter of addition, due to rounding, the actual total could be off by .01% either way. For example, the current total could actually be closer to 0.13% (or 0.11%) than 0.12%. While this may seem minor, in the absence of any compelling reason to combine the numbers, I feel they should be split back out. -- 01:21, 26 September 2008 (UTC)
Why even bother listing anything older than IE6? They have negligable market share (all less than 0.1%) Thelem (talk) 23:42, 1 March 2009 (UTC)
184.108.40.206 (talk) 17:56, 7 September 2011 (UTC) The total % added up is 104% which is an entirely different problem.
The link used as a reference here is currently showing a "we are preparing a report" page; before the beginning of July, this showed a nice graph backing up the numbers we had here. One IP tried vandalizing this article by putting in deliberately bogus numbers; I reverted their changes but another IP (from a completely different geographic location) put in numbers that look reasonably plausible but are not supported by any external reliable sources. I will revert this in a few days back to the old May numbers if we can't get a copy of a survey backing up the numbers here; the survey has been in preparation since the beginning of July. Samboy (talk) 14:46, 3 July 2009 (UTC)
Net Applications changes
Net Applications started country weighting this month, which has lead to big shifts in the reported share. In addition, Internet Explorer 5.5 has dropped off the chart — I wasn't sure how to record this, so I just put in <0.03% (0.03% being the lowest shared listed). —Stephen Morley (talk) 07:19, 1 August 2009 (UTC)
This is the only resource of global statistics that I can find that splits out the IE market share by version. W3CShools also has a statistics page that breaks it out by IE version, however this data is collected by the users of that site, who tend to be web developers. This tends to make that source bias.
- There are at least three others . Net Applications has VERY high IE6 numbers; too high to trust IMHO. -- Limulus (talk) 10:53, 4 January 2010 (UTC)
- According to each one's own description (currently):
- w3counter: "based on the last 15,000 page views to each website tracked by W3Counter. W3Counter's sample currently includes 40,557 websites". While that equates to a healthy 600 million views, that's a very small network of sites and a tiny time window. What if they happen to take the snapshot a few hours after a new version of Chrome is released and everyone's trying it out?
- statcounter: "StatCounter Global Stats are based on aggregate data collected by StatCounter on a sample exceeding 15 billion page views per month collected from across the StatCounter network of more than 3 million websites". Seems like a much better choice, most importantly the drastically larger array of websites. You're very unlikely to have it skewed to, say, tech-savvy users: There just aren't 3 million Engadgets and Gizmodos in the world.
- statowl: "Our data is made up of an average of 28 million unique visitors per month" and "92% of web sites serve a predominantly United States market (this is an area we would looking to add more diversity in the upcoming months)" are some of the interesting facts. No statement on how many sites they cover. It could be lots of tiny sites, or only a few big ones. Either way it's probably a poor representation. It's most certainly a poor indicator of world usage, given that it's 92% skewed to the US.
- And yes, it's well known that w3schools is heavily skewed towards the tech savvy (by their own admission!). I would say our official choice of these should be statcounter, since they base it on the biggest data set, and have the largest diversity of sites. 220.127.116.11 (talk) 19:01, 5 October 2010 (UTC)
I don't know which statistics to believe: http://www.w3schools.com/browsers/browsers_stats.asp (IE combined hovers around 30%); statcounter claims IE has a 52% (see http://gs.statcounter.com/) and mashable seems to support net aplplication's numbers (see http://mashable.com/2010/02/01/browser-stats-chrome/). My favorite stat is this one: http://www.michaelvandaniker.com/labs/browserVisualization/ —Preceding unsigned comment added by 18.104.22.168 (talk) 18:16, 28 July 2010 (UTC)
Net Applications pay wall
The reports from Net Applications now only show the top five items, with the more detailed data hidden behind a pay wall. This means that the data for Internet Explorer versions before version 6 are no longer available without payment.—Stephen Morley (talk) 06:51, 1 August 2010 (UTC)