Linux browser trends examined - Linux

This is a discussion on Linux browser trends examined - Linux ; Linonut wrote: > * thad05@tux.glaci.delete-this.com peremptorily fired off this memo: > >> DFS wrote: >>> >>> Thanks for the ammo, Mercenary: > > Just wanted to point out this term used by DFS, the cheap-shot artist > who nonetheless misses ...

+ Reply to Thread
Page 2 of 4 FirstFirst 1 2 3 4 LastLast
Results 21 to 40 of 70

Thread: Linux browser trends examined

  1. Re: Linux browser trends examined

    Linonut wrote:
    > * thad05@tux.glaci.delete-this.com peremptorily fired off this memo:
    >
    >> DFS wrote:
    >>>
    >>> Thanks for the ammo, Mercenary:

    >
    > Just wanted to point out this term used by DFS, the cheap-shot artist
    > who nonetheless misses the target.


    Actually, I wasn't really bothered by it. Truth is, I am an
    independent consultant, which I've heard referred to as a 'hired
    gun' more than once. Heck, I've jokingly referred to myself as a
    technology whore. Generally, I work on whatever the client needs
    me to.

    The thing is, I've no shortage of Linux work in recent years, so
    I have the luxury of working on the tech I find most interesting.
    It didn't used to be that way.

    >> Actually, looks to me that it could be as much as 2 or 3 per 100 if
    >> you look at a broad enough spectrum of stats.

    >
    > And leave the Microsoft-besotted U.S. out of the equation .
    >
    > Here's another thing. Those stats only get at people using the web for
    > browsing. Also, what if most of the Linux users also used Windows, thus
    > diluting the Linux stats? Are we going to start counting "half-Linux"
    > users?
    >
    > What about people who perforce use Windows during the day, but have
    > Linux systems at home, but are then don't do much browsing, but instead
    > do other stuff with the box.
    >
    > If you just count browser usage, you leave a lot out of the equation.
    > How about NNTP usage? FTP? Robots? Peer-to-peer? On-line gaming?


    True, what web stats can tell you is really the relative presence
    of each platform among the web connected populace, not all PC
    users. Still, that is a valuable thing to know and is useful
    when examining larger trends.

    > And it all takes place in the presence of a still-powerful consumer
    > desktop monopoly that is, as per its former CEO, always "running scared"
    > and using marketing and business tactics that match its emotional state.
    > A company that astroturfs the web with advertising and propaganda.
    > Whose proponents astroturf blogs, article commentary, and other forums.


    > The growth today of Apple systems and Linux systems is something to be
    > welcomed with joy.


    Yes, I welcome additional diversity and competition in the
    marketplace. There is no doubt Linux and Apple have forced MS
    to up there game and indirectly benefited even Windows users.

    I look at this way: If we were watching a tug of war between
    a mouse and an elephant, and the mouse was actually tugging things
    every so slowly his direction, I would be pretty damn impressed
    with that mouse no matter how gradual the progress.

    Thad
    --
    Yeah, I drank the Open Source cool-aid... Unlike the other brand, it had
    all the ingredients on the label.

  2. Re: Linux browser trends examined

    josh fickler wrote:
    >
    > The problem is that w3counter will primarily measure "technical people" and
    > technical people are more likely to be using linux than your average user.
    >
    > It's not all that different from going to a website that's MS-Office centric
    > and using those web stats to draw a conclusion. The typical user attracted
    > to a MS-Office site is more likely to be using Windows.


    How do you figure that? W3counter collects their stats from nearly
    8000 different websites, not just their own. Their is nothing to
    indicate that their stats service is used only by 'technical people'.
    They likely have a cross section of many types of customer sites.

    Thad
    --
    Yeah, I drank the Open Source cool-aid... Unlike the other brand, it had
    all the ingredients on the label.

  3. Re: Linux browser trends examined

    thad05@tux.glaci.delete-this.com wrote:

    > http://www.glaci.com/linuxstats.html
    >
    > Enjoy,
    >
    > Thad


    Thanks for the web page with the nice graphs.

    I have tabulated some of the data from w3counter starting from the links
    found at
    http://en.wikipedia.org/wiki/Usage_s...rating_systems

    On that page, several Linux measurements during a single month are
    averaged to get a number for that month. I think that's how you are
    making your graphs. However I made my table from the raw data and
    counted the time as the number of days since the beginning of 2007, with
    01/01/07 being day 1. By raw data I mean the pages like

    http://www.w3counter.com/globalstats...ate=2007-05-10
    http://www.w3counter.com/globalstats...ate=2007-05-20
    http://www.w3counter.com/globalstats...ate=2007-05-30

    etc.

    I tabulated data for Linux, Mac, IE6, IE7, and IE6+IE7.

    Date Day Number Linux Mac IE6 IE7 IE6+IE7 Midpoint
    05/10/07 130 1.25 3.91 48.88 17.62 66.50
    05/20/07 140 1.21 3.94 49.54 16.82 66.36 135
    05/30/07 150 1.26 3.77 50.22 16.47 66.69 145
    06/10/07 161 1.27 3.84 49.52 16.96 66.48 155.5
    06/20/07 171 1.32 3.82 49.21 17.09 66.30 166
    06/30/07 181 1.30 3.80 49.12 17.08 66.20 176
    07/10/07 191 1.29 3.74 48.84 17.64 66.48 186
    07/30/07 211 1.32 3.70 48.18 18.23 66.41 201
    08/10/07 222 1.33 3.73 47.59 18.71 66.30 216.5
    08/20/07 232 1.34 3.73 47.26 18.97 66.23 227
    08/30/07 242 1.34 3.77 47.01 19.19 66.20 237
    09/10/07 253 1.37 3.74 46.55 19.59 66.14 247.5
    09/20/07 263 1.37 3.78 46.68 19.54 66.22 258
    10/01/07 274 1.38 3.77 46.10 20.18 66.28 268.5
    10/10/07 283 1.38 3.82 46.01 20.29 66.30 278.5
    10/20/07 293 1.70 4.49 43.76 18.76 62.52 288
    10/30/07 303 1.75 4.54 43.43 18.98 62.41 298
    11/10/07 314 1.74 4.49 43.33 19.22 62.55 308.5
    12/01/07 335 1.77 4.59 42.79 19.57 62.36 324.5
    01/31/08 396 1.84 4.82 40.12 21.67 61.79 365.5
    02/29/08 425 2.01 4.95 39.00 22.67 61.67 410.5

    I'd be happy if you'd like to use these numbers in your graphs. I never
    make mistakes, but you and other COLA readers might want to double-check
    them to be sure they match the numbers on the website.

    The IE6+IE7 and the Midpoint (of the time period) columns are computed
    in my spread sheet, so you could discard those columns and write your
    own expressions to get those numbers if you want.

    I see you have included only Linux graphs ... I think the Mac and IE
    graphs add to the picture ... anyway by using the day-numbered raw data
    for Linux usage, we get a somewhat more precise picture for Linux ...

  4. Re: Linux browser trends examined


    wrote in message
    news:6a5db5-k2o.ln1@tux.glaci.com...
    > josh fickler wrote:
    >>
    >> The problem is that w3counter will primarily measure "technical people"
    >> and
    >> technical people are more likely to be using linux than your average
    >> user.
    >>
    >> It's not all that different from going to a website that's MS-Office
    >> centric
    >> and using those web stats to draw a conclusion. The typical user
    >> attracted
    >> to a MS-Office site is more likely to be using Windows.

    >
    > How do you figure that? W3counter collects their stats from nearly
    > 8000 different websites, not just their own. Their is nothing to
    > indicate that their stats service is used only by 'technical people'.
    > They likely have a cross section of many types of customer sites.


    I was thinking w3schools. Thanks for clearing that up.

    Is this the page you're using to get your 2% linux number from?

    http://www.w3counter.com/stats/platforms/2/20/0

    If so then there's a problem with your theory of 8,000 different websites.
    The most popular OS (Win-XP) shows up as getting just 10,997 visits. That's
    less than 1.5 visits per each of those "8,000" websites they collect data
    from. Linux on that list has a total of 286 visits. You'd think that with
    8,000 websites worth of collective data the numbers would be higher than
    this. A LOT higher.




    > Thad
    > --
    > Yeah, I drank the Open Source cool-aid... Unlike the other brand, it had
    > all the ingredients on the label.




    --
    Posted via a free Usenet account from http://www.teranews.com


  5. Re: Linux browser trends examined

    thad05@tux.glaci.delete-this.com wrote:
    > amicus_curious wrote:


    > A larger install base also
    > adds to the viral growth through peer advocacy and increased
    > visibility.


    That is an exponential-growth phenomenon: the growth rate is
    proportional to the existing number.

  6. Re: Linux browser trends examined

    josh fickler wrote:
    > "Rick" wrote in message
    > news:13u7ef3t0k24nab@news.supernews.com...
    >> On Fri, 21 Mar 2008 08:48:57 -0500, DFS wrote:
    >>
    >>> Gregory Shearman wrote:
    >>>> DFS wrote:
    >>>>> Thanks for the ammo, Mercenary: after 10+ years of trying to give
    >>>>> itself away, less than 1 in 100 users chooses Linux.
    >>>> An exclusive club.
    >>> ...of woman haters http://lonien.de/wjl/images/jpg/p200602260034.jpg
    >>>
    >>>
    >>>
    >>>> When people have a choice to buy preloaded linux then things will
    >>>> improve. But choice is pretty hard to come by.
    >>> Wal-mart just chose to remove one such choice because when offered a
    >>> choice people didn't choose that choice.

    >> No, they took the macines off the shelves because of low margins. They
    >> are still on the web.

    >
    >
    > Which is precisely the point. The in-store linux computers was inventory
    > that Walmart had to buy, stock and sell. Except that nobody was interested
    > in buying it. Everything Walmart sells has low margins. They make it up in
    > quantity so this computer is no different. They dropped it because it sat
    > there and collected dust.


    except for the sold out part.

    >
    > As far as the computers Walmart sells on their website. For starters, how
    > many people do you know that bought their computer from Walmart.com? I've
    > never heard of anyone doing this. So it's not going to be any huge boom to
    > linux adoption.


    It may, or may not, but WalMart still has them for sale.

    >
    >
    > Selling the computer online is hardly a testiment for linux. Walmart simply
    > "advertises" the computer on the website. If someone were to actually buy
    > one of these Walmart would take the order, process it, do the billing but
    > the computer would ship directly from Everex. Walmart neither stocks the
    > computer or does anything more than take the order for the computer. It's
    > hardly different from someone selling a "linux computer" on eBay. This in no
    > way means that Walmart believes in linux systems or that they are actually
    > selling in any significant numbers. Walmart simply lists them on the website
    > pretty much the same way that eBay lists items. The actual shipping and
    > delivery of the item is done by someone else.
    >
    > (Walmart doesn't have a huge pile of these linux machines sitting somewhere
    > in a warehouse.)


    No ****, Sherlock. WalMart does sell them, however, and there are a LOT
    of things WalMart doesn't sell.

    --
    Rick

  7. Re: Linux browser trends examined

    In article <47e3b30d$0$25198$ec3e2dad@news.usenetmonster.com>,
    "amicus_curious" wrote:
    > >
    > > No one really knows how many people use linux.
    > >

    > Nor does it really matter.


    I find it amusing, though. After all, scientist and others routinely
    estimate things like the number of pollock in the Bering Sea (so they
    can set fishing quotas), or the number of members of various endangered
    species, so they can plan development and conservation efforts.

    Or closer to human activity, they have a good idea of things like the
    value of the marijuana crop in California, the size of the cocaine
    trade, and things like that.

    These are things that should be very hard to count, due to remoteness,
    or due to the participants going to great lengths to be undetected. Yet
    we count them.

    But counting Linux users--well, that's beyond us!?

    > 0%, 1%, 2%, even 3% are effectively zero in terms of sparking any
    > mass migration to Linux on the desktop. For marketing purposes, you
    > have two categories, i.e. Windows and "others".


    You've got Windows, Mac, and other. Mac is small compared to Windows in
    percentage, but its user base skews toward some more desirable
    demographics. Mac does well among the trendsetters on the creative side
    (artists, musicians, photographers, film makers) *and* it is does well
    among the alpha geeks. That gets you two groups on the leading edge,
    that are highly influential in determining what the masses do.

    --
    --Tim Smith

  8. Re: Linux browser trends examined

    Moshe Goldfarb wrote:
    >
    > I think the BBC study which pegged Linux at about 0.8 percent is probably
    > in the ballpark because the BBC is not a techie site where Linux would show
    > inflated numbers nor is it a really a site targeted at specific types of
    > people.
    >
    > It's worldwide although I suspect the highest market share is from GB and
    > the European community.
    > This should actually help Linux as Linux seems to be more popular over
    > there than in USA.


    I agree the BBC site probably provides better stats than most, but to
    some extent it will suffer some of the same demographic issues as any
    other single web site. Aggregating stats from many different web
    sites undoubtedly provides a more accurate picture. Stats from a
    really popular search engine would run a close second though. Too
    bad google no longer posts that sort of info.

    > At any rate, no matter how you slice it, the market share for desktop Linux
    > is horrendous.
    >
    > It's free and virtually nobody is using it.
    >
    > The Linux community needs to sit down and suck it up and figure out what is
    > wrong with Linux and fix it.


    Considering Linux is continually improving and expanding, I would say
    that is exactly what is happening.

    >
    > Releasing one faulty distribution after another is not the way to fix it.


    Tell you what, you publish your manifesto or design specs or whatever
    for the Ultimate Path to Victory(tm), and the entire Linux community
    can rally around it and march in lock-step.

    Sorry, but as messy as it is, open experimentation with the freedom
    to fail is just how progress happens in this community. It is like
    capitalism or democracy or the scientific process... try and
    centralize the control or streamline them too much, and you just
    cut the heart out of them. It is in the diversity of opinion
    that the rare insights are born.

    Really, all these calls for unity and a single grand plan for
    the traditionally decentralized Linux community is really rather
    pointless. Might as well try and teach cats to line dance.

    Thad
    --
    Yeah, I drank the Open Source cool-aid... Unlike the other brand, it had
    all the ingredients on the label.

  9. Re: Linux browser trends examined

    In article ,
    thad05@tux.glaci.delete-this.com wrote:
    > You can keep saying that to yourself if it makes you feel better,
    > but as Linux marches into solid single digit territory with a clear
    > long term growth trend, it makes software vendors that much more
    > likely to include Linux in their cross platform plans. Increased


    Someone needs to come up with a working revenue model first. How many
    successful end-user for pay applications are there on Linux currently?

    Until that happens, cross platform plans for people developing for
    desktop users will continue to mean "Windows and Mac", not "Windows,
    Mac, and Linux".

    --
    --Tim Smith

  10. Re: Linux browser trends examined

    Hadron wrote:
    >
    > The thing that scares me is that you take this 0.2% growth during Linux
    > prime time as a positive thing. It is disastrous.


    You know there was a time when server Linux was claiming only
    0.2% of the install base over a similar time period, and yet
    that doesn't seem to have stopped it from becoming a serious
    server solution. The fact that Linux is gaining ground one
    the desktop even when faced with an entrenched competitor like
    Microsoft and then Apple as well... that doesn't exactly
    scare me.

    If you look at it another way, Linux has added better than 40%
    to its desktop install base in the past year. Even conservative
    estimates would put that at a gain of 5 to 10 million users.
    If there is any network effect at play at all, we can expect
    the expanded install base to yield faster growth.

    I think the next few years are going to be fun times for us
    Linux geeks.

    Thad
    --
    Yeah, I drank the Open Source cool-aid... Unlike the other brand, it had
    all the ingredients on the label.

  11. Re: Linux browser trends examined

    thad05@tux.glaci.delete-this.com wrote:

    > You can keep saying that to yourself if it makes you feel better,
    > but as Linux marches into solid single digit territory


    Normally you're a straightforward kind of guy, but I see you have a naughty
    sense of humor!!!

    0.6%...0.8%...1.1%.... solid territory!



    > with a clear
    > long term growth trend, it makes software vendors that much more
    > likely to include Linux in their cross platform plans. Increased
    > application choice makes the platform more viable to users which
    > in turn reinforces the growth trend. A larger install base also
    > adds to the viral growth through peer advocacy and increased
    > visibility.


    With "advocates" like these twits around here, Linux needs some real
    friends.


    > Of course there will not be a mass migration... as I've said many
    > times, it takes years to churn an entrenched install base. Just
    > look at how long Netware stuck around long after the common wisdom
    > said it was 'dead'. But the desktop Linux trend is clear, and
    > just like server Linux, is defying the predictions of the critics.
    > Linux is not 'dying out' or 'going nowhere'. It is clearly
    > improving and growing.


    So's my belly. Doesn't mean I won't die out.



    > I find it interesting that success for MS has now essentially
    > been redefined from 'crushing Linux' to 'holding off the mass
    > migration'.


    You weren't kidding - you *have* been drinking something.




  12. Re: Linux browser trends examined

    Hadron wrote:

    > And its numbers tie in more with the BBC which says 0.8%.
    >
    > A roughly 0.2% boost during Ubuntu hysteria. One user in every 500.
    >
    > This is not good whichever way you want to cut and dice it.


    Lucky for Linux developers they don't demand a return on their time or
    effort.



  13. Re: Linux browser trends examined

    josh fickler wrote:
    > wrote in message


    >> You've got some sort of problem with the w3counter.com numbers other
    >> than they are larger than you like? It seems they use a rather large
    >> and diverse sample and are thus rather credible.

    >
    > The problem is that w3counter will primarily measure "technical
    > people" and technical people are more likely to be using linux than
    > your average user.


    thad know this already, but he thinks he can play Mini-Advocate and throw a
    spitball past us.



  14. Re: Linux browser trends examined

    * Tim Smith peremptorily fired off this memo:

    > In article ,
    > thad05@tux.glaci.delete-this.com wrote:
    >> You can keep saying that to yourself if it makes you feel better,
    >> but as Linux marches into solid single digit territory with a clear
    >> long term growth trend, it makes software vendors that much more
    >> likely to include Linux in their cross platform plans. Increased

    >
    > Someone needs to come up with a working revenue model first. How many
    > successful end-user for pay applications are there on Linux currently?
    >
    > Until that happens, cross platform plans for people developing for
    > desktop users will continue to mean "Windows and Mac", not "Windows,
    > Mac, and Linux".


    If true, that's a shame, since cross-platform coding can be pretty easy
    once you get used to it.

    I wonder what kind of "checks" Bill is referring to in the sig.

    --
    Programs today get very fat; the enhancements tend to slow the programs down
    because people put in special checks. When they want to add some feature,
    they'll just stick in these checks without thinking how they might slow the
    thing down.
    -- Bill Gates

  15. Re: Linux browser trends examined

    * Tim Smith peremptorily fired off this memo:

    > In article <47e3b30d$0$25198$ec3e2dad@news.usenetmonster.com>,
    > "amicus_curious" wrote:
    >> >
    >> > No one really knows how many people use linux.
    >> >

    >> Nor does it really matter.

    >
    > I find it amusing, though. After all, scientist and others routinely
    > estimate things like the number of pollock in the Bering Sea (so they
    > can set fishing quotas), or the number of members of various endangered
    > species, so they can plan development and conservation efforts.
    >
    > Or closer to human activity, they have a good idea of things like the
    > value of the marijuana crop in California, the size of the cocaine
    > trade, and things like that.
    >
    > These are things that should be very hard to count, due to remoteness,
    > or due to the participants going to great lengths to be undetected. Yet
    > we count them.


    No, we do not.

    > But counting Linux users--well, that's beyond us!?


    When you use the word "counting", you're making a mistake.

    It is estimation, not counting.

    --
    Your most unhappy customers are your greatest source of learning.
    -- Bill Gates

  16. Re: Linux browser trends examined

    * thad05@tux.glaci.delete-this.com peremptorily fired off this memo:

    > Sorry, but as messy as it is, open experimentation with the freedom
    > to fail is just how progress happens in this community. It is like
    > capitalism or democracy or the scientific process... try and
    > centralize the control or streamline them too much, and you just
    > cut the heart out of them. It is in the diversity of opinion
    > that the rare insights are born.
    >
    > Really, all these calls for unity and a single grand plan for
    > the traditionally decentralized Linux community is really rather
    > pointless. Might as well try and teach cats to line dance.


    Bill Gates didn't teach cats to line dance -- he just made sure they
    could find only his brand of cat food.

    --
    It turns out Luddites don't know how to use software properly, so you should
    look into that. -- The reason we come up with new versions is not to fix
    bugs. It's absolutely not. It's the stupidest reason to buy a new version I
    ever heard. When we do a new version we put in lots of new things that
    people are asking for. And so, in no sense, is stability a reason to move to
    a new version. It's never a reason.
    -- Bill Gates, http://www.cantrip.org/nobugs.html

  17. Re: Linux browser trends examined

    josh fickler wrote:
    >
    > I was thinking w3schools. Thanks for clearing that up.


    No problem.

    > Is this the page you're using to get your 2% linux number from?
    >
    > http://www.w3counter.com/stats/platforms/2/20/0


    I got it from here:

    http://www.w3counter.com/globalstats.php

    And then selected other samples using the date pulldown on the
    page.

    > If so then there's a problem with your theory of 8,000 different websites.
    > The most popular OS (Win-XP) shows up as getting just 10,997 visits. That's
    > less than 1.5 visits per each of those "8,000" websites they collect data
    > from. Linux on that list has a total of 286 visits. You'd think that with
    > 8,000 websites worth of collective data the numbers would be higher than
    > this. A LOT higher.


    The URL I listed above has a bit more detail on it, including the
    following text regarding their methodology:

    This report was generated 02/29/2008 based on the last 20,873,774
    unique visits to 7,930 websites. The last 25,000 page views to each
    website are analyzed to identify unique visits. Some visits may occur
    before the month of the report. Search engine market share is
    computed as percentage of search-referred traffic to analyzed sites.
    Live Search includes MSN searches. Partner sites are not included in
    these shares.

    Cheers,

    Thad
    --
    Yeah, I drank the Open Source cool-aid... Unlike the other brand, it had
    all the ingredients on the label.

  18. Re: Linux browser trends examined

    Linonut writes:

    > * Tim Smith peremptorily fired off this memo:
    >
    >> In article <47e3b30d$0$25198$ec3e2dad@news.usenetmonster.com>,
    >> "amicus_curious" wrote:
    >>> >
    >>> > No one really knows how many people use linux.
    >>> >
    >>> Nor does it really matter.

    >>
    >> I find it amusing, though. After all, scientist and others routinely
    >> estimate things like the number of pollock in the Bering Sea (so they
    >> can set fishing quotas), or the number of members of various endangered
    >> species, so they can plan development and conservation efforts.
    >>
    >> Or closer to human activity, they have a good idea of things like the
    >> value of the marijuana crop in California, the size of the cocaine
    >> trade, and things like that.
    >>
    >> These are things that should be very hard to count, due to remoteness,
    >> or due to the participants going to great lengths to be undetected. Yet
    >> we count them.

    >
    > No, we do not.


    Yes we do.

    These figures are indeed counted. The count might only be an estimate of
    total but it is indeed a count.

    >
    >> But counting Linux users--well, that's beyond us!?

    >
    > When you use the word "counting", you're making a mistake.


    No. He isn't.

    >
    > It is estimation, not counting.


    Wrong. It is counting.

    --
    "There is no such thing as Intellectual Property"
    Mark Kent
    Head of Technology Strategy, BT Global
    COLA Hypocrite

  19. Re: Linux browser trends examined

    DFS wrote:
    >>
    >> The problem is that w3counter will primarily measure "technical
    >> people" and technical people are more likely to be using linux than
    >> your average user.

    >
    > thad know this already, but he thinks he can play Mini-Advocate and throw a
    > spitball past us.


    No, josh was simply confusing w3counter with w3schools. Thats been
    all cleared up now. W3counter aggregates stats from almost 8000
    web sites, and I highly doubt all of them are just geeky technical
    sites.

    Thad
    --
    Yeah, I drank the Open Source cool-aid... Unlike the other brand, it had
    all the ingredients on the label.

  20. Re: Linux browser trends examined


+ Reply to Thread
Page 2 of 4 FirstFirst 1 2 3 4 LastLast