July 11, 2004

11:30 PM

Are there any reports, stats, or studies which might give me a more accurate picture of the percentage of people who use Windows or the Mac OS at home? What I'm curious about is how the numbers break down after you remove the vast number of PCs / Windows machines purchased by offices, and the rather inaccurate way that market share is determined.
posted by theNonsuch to (19 comments total)

The University of Texas has home vs. office o.s. usage breakdowns [sample/methodology info] from a 2002 survey they did of students and employees. Obviously not at all suitable for generalization, but since you asked for "any" info I guess it's worth a mention.
posted by nakedcodemonkey at 11:47 PM on July 11, 2004

(2003 update of same)
posted by nakedcodemonkey at 11:51 PM on July 11, 2004

(similar surveys conducted 2002-2004 by the University of New Brunswick)
posted by nakedcodemonkey at 11:55 PM on July 11, 2004

the rather inaccurate way that market share is determined.

Do tell.
posted by NortonDC at 12:18 AM on July 12, 2004

That UNB survey is pretty worthless, by the raw numbers they had about 300 people (on a campus of 10,000) answer the thing. Also, there were 8 people who said they used a macintosh at home and 10 who said the primary OS of their computer was Mac OS. I want to meet those other 2.

NortonDC: Market share is a stupid number because it includes every piece of junk sold with a celeron processor to outfit a call centre or other such use and includes it in the same category as a Power Mac, yet people use the Market Share stat to bash apple. Installed base,especially within a particular domain, would be a much better measurement if you're looking for actual computer use.
posted by Space Coyote at 12:54 AM on July 12, 2004

Also, there were 8 people who said they used a macintosh at home and 10 who said the primary OS of their computer was Mac OS. I want to meet those other 2.

Not improbable, especially if they're from the CS department. PowerPC Macs do run Linux. Also < ahref="http://www.computing.net/beos/wwwboard/forum/460.html">some versions of BeOS.
posted by nakedcodemonkey at 1:15 AM on July 12, 2004

*passes Mac users box of tissues*

Anyway, here's some numbers from Distributed.Net, which of course tends toward the power-user end of the market.
posted by falconred at 1:17 AM on July 12, 2004

Argh. Let's try that last one again.

Also < a href="http://www.computing.net/beos/wwwboard/forum/460.html">some versions of BeOS.
posted by nakedcodemonkey at 1:17 AM on July 12, 2004

(Crud.)
posted by nakedcodemonkey at 1:19 AM on July 12, 2004

Also some versions of BeOS

/crud
posted by sebas at 3:52 AM on July 12, 2004

I think accurate home market share figures would be impossible to calculate. For example, my old man uses a cube, and my mum uses both an original iMac and an ancient powermac 7600. How do these machines make the overall market share figures?

I think this is also the reason for Apple's low market share; As a general rule Apple hardware is used for much longer than comparative PC hardware.
posted by derbs at 6:28 AM on July 12, 2004

What's up with those distributed.net numbers? PowerPC *Rhapsody* as having the third spot, yet MacOSX hasn't been called Rhapsody for at least five years!

I also hate the way market share numbers are skewed. Every cash register with Windows installed is counted in the "Market Share" numbers, but these machines will likely never have a large-scale consumer application installed on them. Software companies point to this number and say they won't develop for the Mac.

That said, it would be incredibly difficult to be accurate when counting Windows PCs anyway, since Whitebox shops sell parts, and likely don't tell anyone about their full machine sales numbers...
posted by tomierna at 6:40 AM on July 12, 2004

This is a great question, shame there isn't a better answer. It's possible I just hang out with cooler ;) people, but the Mac/PC breakdown @ home among people I know skews more toward 50-50.

Assuming that internet access is fairly consistent across platforms, wouldn't an analysis of a large consumer site (Amazon, Yahoo) over a weekend (when people are at home) come close to a reasonable estimate?
posted by mkultra at 7:09 AM on July 12, 2004

... or metafilter ;)
posted by Space Coyote at 7:44 AM on July 12, 2004

nakedcodemonkey, it's true that other operating systems run on Mac hardware, but that's not what Space Coyote was talking about when he said he wanted to meet the other 2. In the UNB survey, 8 people said they used Macs and 10 said they used Mac OS--meaning 2 people are under the impression that they're running Mac OS on non-Mac hardware.
posted by Acetylene at 8:19 AM on July 12, 2004

perhaps the people space coyote wants to meet are running mac OS on home-built generic boxes rather than official macs? there's a lab across the street, PI favors macs, but prefers to build her own systems. probably one in three systems are actually mac-branded, the rest look like the average home-built beige box windows PC but run mac OS.
posted by caution live frogs at 10:42 AM on July 12, 2004

If someone was building custom mac hardware on this campus I think I'd have met him or her by now :) I think it's just someone answering the survey and getting messed up somehow
posted by Space Coyote at 2:21 PM on July 12, 2004

Internet hit statistics broken out by OS will reflect installed base instead of sales, and they will eliminate the point of sale systems and their ilk that you seek to remove from consideration.

Operating Systems Used to Access Google - May 2004:
Mac     3%

TheCounter.com's hit tracker for web sites - Browser OS Stats Tue Jun 1 00:05:01 2004 - Wed Jun 30 23:55:05 2004
Mac     1%

W3 Schools - OS Browser Platform Statistics June 2004
Mac     2.5%

Does that cover it?
posted by NortonDC at 4:27 PM on July 12, 2004

... Except that it doesn't cover "at home" vs. "at work".
posted by Space Coyote at 4:44 PM on July 12, 2004