Forgot your password?
typodupeerror

Noise Over Mac OS Market Share "Slip" 481

Posted by kdawson
from the i'll-show-you-flatline dept.
OakDragon writes, "Mac OS market share actually slipped since last September. This reverses a trend in the winter and spring months that showed some slight growth. The actual percentage loss is small: 0.02%. But it may be significant since it follows a solid growth trend. It must be disappointing to Apple and Mac fans to see what is basically a flat line in desktop market share." Mac-oriented sites are pointing out the unreliability of the metrics from Net Applications, which are based on users of the HitsLink service.
This discussion has been archived. No new comments can be posted.

Noise Over Mac OS Market Share "Slip"

Comments Filter:
  • Statistics..... (Score:5, Informative)

    by BWJones (18351) * on Tuesday September 19, 2006 @06:18PM (#16141657) Homepage Journal
    I'd have to say that from my limited sampling, these numbers are very possibly off and a .2% downward change is likely statistically insignificant, especially given their sampling methods.

    Traffic from my blog [utah.edu] primarily from the US shows about 19% of traffic is from the Macintosh (200-900 unique visitors/day). Of all the traffic that hit my blog from the recent Boing Boing posting, it appears that of those that clicked through, over 23% of the clicks were from Macintosh systems and from the traffic I get from Slashdot, about 15% is from Macintosh systems. This limited sampling shows a steady increase in the percentage of Macintosh users that have visited over the past few years.

    Traffic from another site I manage, Webvision [utah.edu] (I know, I know, ....really old design from the early 90's, but it's been low on my priority list for the last four years) was likely the first online textbook receiving much more international traffic (about 1000 unique visitors/day from all over the world) and I have seen the international Macintosh marketshare increase from about 4% to 6.5% of total traffic over the past year.

    Both of these statistics mirror the trends I have seen reported for the platforms marketshare on much wider scales. These are direct measures that I am reporting as opposed to a fee based service like HitsLink whose measures are not as direct. Too bad Google's Zeitgeist no longer reports on platform statistics which were a good measure of overall platform usage from a much wider used resource.

    • "According to Techweb, data gathered by Net Applications shows that the Mac OS had 4.35 per cent of the world's operating system share last December. Now it only has 4.33 per cent.""

      Yet, at the link to the actual data, it says, for August 2006:

      winXP: 84.18%, win2000: 6.54%, Mac: 3.71%, win98: 2.40%, winME: 1.10%, Other: 2.07%

      So, 3.71%, not 4.33%. Looks like The Inquirer is reading the line for April 2006, and not September 2006. Actually, Mac share drops continually during the period December 2005
      • Re: (Score:3, Informative)

        by BWJones (18351) *
        "Other" is actually Linux combined with Mactel (and still others) as they break out the Mactel and MacOS (PPC) separately apparently......

        • Re: (Score:3, Informative)

          by Petrushka (815171)

          I've just put together the complete figures, based on their stats, incorporating both "Mac OS" and "Macintel", since December last year. Mac OS overall is down from a high of 4.49% in April, but consistently up from a low of 4.28% in June.

          Month. . "Mac OS" "MacIntel" Total

          Dec 05 . 4.35 . . 0. . . . . 4.35
          Jan 06 . 4.21 . . 0. . . . . 4.21
          Feb 06 . 4.28 . . 0.03 . . . 4.31
          Mar 06 . 4.29 . . 0.08 . . . 4.37
          Apr 06 . 4.33 . . 0.16 . . . 4.49
          May 06 . 4.19 . . 0.23 . . . 4.42
          Jun 06 . 3.92 . . 0.36 . . . 4.28
          Jul 0

    • Re: (Score:2, Informative)

      by garcia (6573)
      Traffic from my blog [utah.edu] primarily from the US shows about 19% of traffic is from the Macintosh (200-900 unique visitors/day). Of all the traffic that hit my blog from the recent Boing Boing posting, it appears that of those that clicked through, over 23% of the clicks were from Macintosh systems and from the traffic I get from Slashdot, about 15% is from Macintosh systems. This limited sampling shows a steady increase in the percentage of Macintosh users that have visited over the past few years.

      I r
      • by BWJones (18351) *
        I really hate when people post their statistics coming from their blogs. It is just as statistically insignificant as saying that there are a large number of Linux based Firefox users visiting Slashdot.

        Well, given that my blog is not really Macintosh centric, those stats should be valid. Just in case though, I posted more international stats from an even more platform neutral site or did you not read that far....

    • I'd have to say that from my limited sampling, these numbers are very possibly off and a .2% downward change is likely statistically insignificant, especially given their sampling methods. Traffic from my blog primarily from the US shows about 19% of traffic is from the Macintosh (200-900 unique visitors/day). ... shows a steady increase in the percentage of Macintosh users that have visited over the past few years.

      They were actually reporting a 0.02% change, which most people would consider noise. Cla

      • Re: (Score:3, Informative)

        by The Bungi (221687)
        Yes twitter... too bad w3schools is not even close to being representative of the real world. Might as well use the statistics from kernel.org.
  • by pilgrim23 (716938) on Tuesday September 19, 2006 @06:18PM (#16141660)
    Forgive! It was the Eye Candy what made me do it!
    • Talking about using CP/M is funny, but, since few Slashdot readers know what CP/M is, they won't understand the joke.

      CP/M is Control Program for Microcomputers, an OS used with 8088 microprocessors back before IBM thought of selling PCs. It was a dog of an OS, mostly because it was unfinished. Back then CP/M was sold by a company that thought printing the original of manuals on a dot-matrix printer with an old ribbon was acceptable practice.

      The Morrow Microdecision came with a Command Line Interface l
      • Re: (Score:3, Funny)

        by Sponge Bath (413667)

        ...few Slashdot readers know what CP/M is

        Crabby Prehistoric Man?

        Woah, no need for buggy whip! It was just a joke.

      • Re: (Score:2, Flamebait)

        by drinkypoo (153816)
        Please don't explain jokes. It doesn't make them funny for the people who only understand them after you explain them, and it makes them a lot less funny to those of us who DO understand them. I've owned two different CP/M machines (A Kaypro 4 and then later an Altos with 8" floppies, two of 'em) and your explanation makes me sad. Also, the majority of CP/M machines used the Z80, not the 8088. This is especially significant because the Z80's instruction set is a superset of that of the 8088 and your 8088 pr
        • You mean the Z80 ran a superset of 8080. The 8088 is an entirely different beast (a 16-bit-internal 8080 sort of.. a real ugly CPU, actually, the Z80 and Z280 were much better successors to the 8080 series than the 8088 was, but that is another story).

          There was an 8088 version of CP/M and even nifty machines like the Compupro10 (1 8088, 4 Z80s) that ran MP/M and eventually Concurrent CP/M (or Concurrent CP/M 8-16 on Compupro hardware, automatically executing programs on the right CPU).

          A Z80 could in no way
      • CP/M was for 8080's originally, not 8088.

        Pilot wasn't a command line interface: it was used to write a basic menuing system for the MD series.

        The command line shipped on the MD series was good old "CCP", the "Console Command Processor".

        ZCPR3 beat the DOS of its day, though, no doubt about it.

        (And, yes, I still have my original MD3, and it still worked last I checked it. But, then it doesnt even have a fan. The only moving parts are the floppy drives.)
    • The article doesn't even get iPod sales correct. It gets the peak month wrong, and it's off by several millions. How can we accurately discuss the results if it doesn't even get public iPod number correct?
  • Who's going to buy a brand new Macintosh when we are just about to go to Rev. B. chips/platforms? Maybe everyone who is not 1-point-oh-averse has already bought a Mac. And everyone else wants x86 2.0.


    • Re: (Score:2, Informative)

      by Gothmolly (148874)
      99% of the computer buying population has no idea what your post even means.
      • Hence names like "Core 2 Duo and "Core Duo."

        They're the same right?

        Tom
        • Re: (Score:3, Funny)

          by soft_guy (534437)
          I think the name "Core 2 Duo" is actually wonderful, however I wish they could introduce the concept of duality into the name somehow.
      • by Shuh (13578)
        99% of the computer buying population has no idea what your post even means.



        Then I'm in luck! 99% of the computer buying population is not going to read my post.


      • by PopeRatzo (965947) on Tuesday September 19, 2006 @07:33PM (#16142200) Homepage Journal
        I'll tell you what the post means. It means that somebody has paid a press agent to put stories in the media pushing an anti-Apple agenda.

        Let me explain a bit about HitsLink. Their reason for existence is to be a paid "hit"-man for publicity pros. Are you CBS-Viacom or the Radio Industry? Do you need to make it seem to the business community that Howard Stern is tanking on Sirius Satellite Radio? Have Hitslink provide a story saying that the number of Lycos searches for "Howard Stern" are down by X %. Forget the fact that everybody knows that you'd go to Sirius.com if you want to read about Stern. Forget that nobody uses Lycos any more.

        Let's say you are Salem Radio Network and you want it to seem like conservative commentator, former Sec'y of Education and degenerate gambler Bill Bennett's morning show is really happening. Get HitsLink to create a story saying that he's "Number 9 in the nation". Forget that he's just been dumped from the third biggest market in America (Chicago). Forget that the actual listings show that there are 24 talk shows ahead of Bennett's. Let's just round the figures out so that there are 2 or 3 talk shows tied for Number 1, Number 2, etc. So you can say that Bennett is in the Number 9 slot when in reality he is number 24 out of 30.

        It pays to know that nearly every story that you see or hear in the media has been placed there by a press agent or public relations department in the form of a press release, which gets reworked (sometimes) by a "reporter" (really a stenographer) into a "story" which is presented as "news". It pays to know that outfits like HitsLink exist just to spread manure.

        You have to ask yourself if a story like this passes your own "smell test".
  • Between this, the gentoo article, and the global warming article, I'm seeing some local warming right here.
  • Market fluctuates. (Score:3, Insightful)

    by greenguy (162630) <estebandido@nospaM.gmail.com> on Tuesday September 19, 2006 @06:25PM (#16141707) Homepage Journal
    Film at 11.
  • by CaymanIslandCarpedie (868408) on Tuesday September 19, 2006 @06:25PM (#16141714) Journal
    Of course these numbers and not at all scientfic. The change is also completely insignificant. I agree on all of that. However, I have a feeling many who will denounce these statistics would be singing thier praises if they showed a significant gain ;-)
  • by zoftie (195518) on Tuesday September 19, 2006 @06:29PM (#16141744) Homepage
    http://useragentswitcher.mozdev.org/ [mozdev.org]

    set to os x :-D
  • by jmorris42 (1458) * <jmorris@NOSPam.beau.org> on Tuesday September 19, 2006 @06:29PM (#16141746)
    Go dig into the numbers a bit. I'm not a Mac fanboi (see my abuse of one earlier today) but this is a non-story. The site in question is tracking Mac OS and MacIntel seperate, so of course Mac OS is dropping. Add the two together and you get a different picture. They appear not to have fixed the scripts that generate the cute graphs though, because up to now they broke out each OS variation so they could see the migration patterns in Windows versions.
    • by Gothmolly (148874)
      AND, since its a non-story, why is it even accepted on Slashdot? Oh right, the editors don't actually, you know, edit.
    • Re: (Score:2, Informative)

      by Brad Oliver (604118)
      If you click on each month, and combine Mac OS with MacIntel, you get the following:

      Sep 05: 3.74%
      Oct 05: 3.87%
      Nov 05: 4.11%
      Dec 05: 4.35%
      Jan 06: 4.21%
      Feb 06: 4.30% (4.28 + 0.03)
      Mar 06: 4.37% (4.29 + 0.08)
      Apr 06: 4.49% (4.33 + 0.16)
      May 06: 4.42% (4.19 + 0.23)
      Jun 06: 4.32% (3.92 + 0.36)
      Jul 06: 4.29% (3.80 + 0.49)
      Aug 06: 4.33% (3.71 + 0.62)

      The reported 0.02 decline by the Inquirer is the difference of Dec 05 (4.35) vs Aug 05 (4.33). I'm not sure why Dec 05 was chosen as the comparison m

  • by Anonymous Coward
    Mac-oriented sites are pointing out the unreliability of the metrics from Net Applications, which are based on users of the HitsLink service.

    Yet if it proved the opposite they wouldn't question its reliability at all, and would bring it up every chance they get.
    • Re: (Score:2, Insightful)

      by devjj (956776) *
      And the Mac-haters would be complaining about how the statistical methods are flawed. In the end, it's all the same.
  • "Mac OS market share actually slipped since last September."

    The statistic is affected by the $200 PC computers and $500 laptop PCs that are being sold.
  • Does this mean that Apple is beleaguered again?
  • by Danathar (267989) on Tuesday September 19, 2006 @06:38PM (#16141807) Journal
    Quite frankly I don't want to see OS X have some huge marketshare. I'd prefer the platform to have enough marketshare that developers can make money and Apple to make a profit, but not big enough for Virus writers and spyware authors to care (the way it is now).

    Why does OS X have to have an increasing marketshare to remain successful?
    • Disclaimer: I am a very happy owner of a Mac Pro.

      I'd prefer the platform to have enough marketshare that developers can make money and Apple to make a profit, but not big enough for Virus writers and spyware authors to care (the way it is now).

      I honestly have never understood this idea that Macs would suddenly get more interest from Virus writers if they had market share.

      If you were a cracker and you saw these pompous Apple commercials, saw the Apple trolls that say that Apple can do no wrong, and saw all
      • Re: (Score:3, Insightful)

        by Psychotext (262644)
        I can see your argument, but the simple fact is that most of the exploits out there on the web were written to generate money, not kudos. Ignoring if mac really is more secure or not (btw, you might want to look at how many vulnerabilities they recently patched), the simple fact is that at some point it's going to end up on some mafia-esque hacker's radar and that's when the stream of viruses and vulnerabilities will appear.

        Same thing with firefox. Initially it wasn't worth bothering with, and everyone sa
      • Re: (Score:3, Insightful)

        by drsmithy (35869)
        I honestly have never understood this idea that Macs would suddenly get more interest from Virus writers if they had market share.

        There are a few important reasons for this (which may or may not all apply)

        1. Return on Investment: Where is the benefit in gaining access to a machine only one out of every 100 people (roughly) uses and which is even more uncommon in business environments ?

        2. Infection rates: Any "virus" infection is going to spread far, far more slowly on Macs than PCs. Heck, there's a pr

  • beleaguered (Score:5, Funny)

    by ElephanTS (624421) on Tuesday September 19, 2006 @06:45PM (#16141857)
    Bloody hell! It's back to beleaguered then.
  • This is pretty absurd. The market for 10.4 Tiger is pretty much done, because anyone who wanted to upgrade to to Tiger did so last year.

    Anyone buying a Mac in the last year and a half got Tiger for 'free.' So who is left to buy a Tiger upgrade? If they waited this long, why buy it now rather than waiting for Leopard in a few months?

    Statistics are worthless if they are presented by idiots who don't even know what the numbers mean.

    ----
    www.roughlydrafted.com
  • by askegg (599634) on Tuesday September 19, 2006 @06:54PM (#16141923)
    What is the fascination with market share?

    What's the thinking here? More market share must mean more sales and therefore more profit? Apple seems to be making plenty of money, so what does more market share gives you, or is it just a measure of how many customers you did not get?

    IMHO, the problem is you can not make a product that will please everyone. Apple has decided to make a certain kind of product - looks cool, well designed, easy to use and at a premium price.

    I guess it depends on how you classify your market. If you are talking portable mp3 players in the USA, then Apple has around 80% of the market (their figures).

    If you mean "laptop computers" then the field is wide open to every man and his dog that can bolt a machine together - including the el cheapo models who compete on price alone. This is akin to putting Mercedes, Audi and Lexus in the "car market" and wondering why their share is so low (hint: you are including Hyundai and others). This is not the same market. Who are the premium computer manufacturers? IBM might be there, Dell isn't.

    As long as Apple continues to focus on making their products this way they will have a following and will generate profits - to hell with market share.
  • So, "the Mac OS had 4.35 per cent of the world's operating system share last December. Now it only has 4.33 per cent." Is that, by any chance, a share of a "market" that consists mostly of corporations and IT departments?

    The Mac has always had that problem. "Market share" depends entirely on how you choose to define the market. Among people who don't want Apple computers, Apple's market share is small.

    Cessna has a market share of about 4% of the airplane market (Cessna has revenues of $3.5 billion, Boeing $
    • There's a shit market? What?!?!?! Why did nobody tell me about this before... I've lost a fortune. My output is full of shit. Just look at most of my posts.
  • "It must be disappointing to Apple and Mac fans to see what is basically a flat line in desktop market share."

    Not as bone chilling as the news that "Other" (that's French for Linux, boysngirls) has about as many users as WinME.

    "Sloppy metrics" is the understatement of the decade.

  • by camperslo (704715) on Tuesday September 19, 2006 @07:07PM (#16142018)
    IIRC even Steve Jobs said something early this year about being surprised at sales being stronger than expected. The PPC to Intel transition is a major one and is an excellent long term move. But it did cause some people to hold off on buying waiting for more native software, and allowing time for the shaking out of any minor glitches in the first products. A few probably also held off on buying when they heard that the Core 2 chips were coming.

    It is pretty obvious that the move was a wise choice and that both Macintosh users and Apple will be better off long term. The appeal of the new generation of machines can be expected to increase over time. In addition to new features in the OS, it is reasonable to expect that 10.5 will bring even better performance. It'll likely make better use of multiple CPU cores, use the GPU horsepower for other tasks, use the Core 2 supplemental SSE3 instructions (I've heard them called both SSSE3 and SSE4), and use of the 64-bit capabilities. The software for Windows support will also be more mature (Apple's utility is currently beta).

    The release of Vista will likely bring an increase in the number of people pondering new machines instead of just an OS upgrade. With Apple being more visible than in the past some of those people will opt for getting Macs instead (either solely for the Apple experience, or to run Windows too). Some may also be playing wait and see with Vista. If it isn't really, really, wonderful, it'll help Apple.
    • Re: (Score:3, Interesting)

      by maynard (3337)
      That's definitely the case. I just bought a MacBook to replace my ti powerbook g4 800, which I wouldn't have done had one of the hinges not finally given out. I had budgeted this purchase for next year, to wait out a bugfix in the next generation MacBook. But -- oh, well! Shit happens. Fortunately, the MacBook is good. Performace is excellent for the price, and all the software (except classic) still works like a champ! What more could I ask?
  • What apple really cares about (and what matters for OS adoption) is how many people are making serious or primary use of OS X. What these numbers show is what portion of web browsing is done in "OS X"

    Now given the recent release of boot camp, parrells and similar programs it seems likely that a significant percentage of OS X users will spend say 5% or more of their time in windows. If the growth in market share for OS X is usually less than the average percent of time OS X users have started spending in w
  • by OnanTheBarbarian (245959) on Tuesday September 19, 2006 @07:12PM (#16142043)
    Anyone who thinks that a 0.02% change is likely to be statistically significant has to be smoking crack. Of course, with enough users and a rigorous enough methodology, it's possible, but I doubt it.
    • Re: (Score:3, Funny)

      by Weedlekin (836313)
      Why use 0.02% when one could invent a more authoritative-looking 0.020773091522850061%. After all, If we're going to use survey results that are below the survey's own quoted statistical noise, then why not pad them with some extraneous yet no less meaningful digits to impress mathematically challenged press types?
  • Personal experience (Score:3, Interesting)

    by codemachine (245871) on Tuesday September 19, 2006 @07:31PM (#16142181)
    In my personal experience, I know more and more people buying Apple who never have in the past. Especially in the notebook computer market.

    I wouldn't be surprised if that is what most everyone else here is seeing as well.

    Sometimes these studies aren't an exercise in what the truth is in the real world, especially if they are funded by those who don't like what is happening in the real world.
  • by ursabear (818651) on Tuesday September 19, 2006 @07:31PM (#16142184) Homepage Journal
    As to TFA, I have a question... There are lots of Slashdotters that can probably answer this for me pretty well: Isn't .02% statistically negligible, WRT a market trend report?

    Mod me OT on this one, It's fine with me.

    I'm always amazed at the vitriol that spews forth on this subject. Although, frankly, post threads like those in response to this article are always interesting to read (and sometimes funny).
    IMVHO, use what machine and OS you like, like what machine and OS you use (if you have a choice). It isn't the chip, the windowing system, the kernel, or the manufacturer... it's what it does for you personally. I like Solaris, Fedora, Mac OS (any, really), XP, 2000, Irix, HP/UX... well, just about any of them. The hardware is always a relative benchmark to me. If I like it, and it works great without kicking me in the pants every time I try to use it, then I use it. I enjoy my little Blade 100 as much as my VAIO as much as my iMac G5. Like what you use, and use what you like.
    • Re: (Score:3, Insightful)

      As to TFA, I have a question... There are lots of Slashdotters that can probably answer this for me pretty well: Isn't .02% statistically negligible, WRT a market trend report?

      That information is not available. You see, the source data was not presented, only the results without and details of the methodology. This is PR, not science and is designed to influence people who pay attention to PR, instead of look at scientific data. The fact that you know what statistically significant means, is indication

  • by grappler (14976) on Tuesday September 19, 2006 @08:25PM (#16142540) Homepage
    From one of the linked articles
    http://www.coolest-gadgets.com/20060919/is-apple-l osing-os-share/ [coolest-gadgets.com]

    Web usage by only HitsLink subscribers is just a small random sampling and has nothing to do with overall market share.

    If that is really a random sampling, it has everything to do with overall market share. But it isn't. It is a sample of the market which subscribes to HitsLink. That's not a random sample.
  • by frdmfghtr (603968) on Tuesday September 19, 2006 @09:35PM (#16142976)
    Warning: Rant to follow

    Unless you're Ballmer or Jobs or a Linux distro company, does it really matter? I mean, really, really matter?

    Do I, as a OS X user, see any sort of effect if OS X usage goes up or down?

    In case you're wondering...no.

    I guess I just get tired of Linux fanboys declaring that "we must get this to the desktops of the unwashed masses" or the Mac fanboys stomping around saying how much Microsoft is copying from OS X into Vista, and the Microsoft fanboys sitting around all smug with their favorite OS enjoying a practical monopoly status.

    You use what works best for what you want to do, market share be damned. I use OS X for some things and WinXP for others because they each have their strengths in different areas. If John and Jane Public can easily get their digital photos of Junior's 8th birthday party by simply plugging their camera into their Windows box and pressing a button, more power to them. If you develop the Next Great Thing in an Unbuntu environment, congratualtions.

    If a WinXP platform did what I want it to do as well as, or better than, OS X for a better value then I would have stuck with WinXP. If the engineering tools I need to use every day worked on a Linux platform as easily as on an WinXP or OS X platform, I would have stuck with Linux.

    I coouldn't care less if OS X market share changed 0.02%, up, down, or sideways.

    I'm done ranting.

Real Users hate Real Programmers.

Working...