Global Web Stats

Janco has published some web stats that show that other OS has shrunk to 80% of web accesses. That’s down 10 percentage points in a year… Amazing. Those other stats show a shrinkage of the order of 1% per annum. The difference may be that 50% of these stats come from outside the USA. Amen.

M$ may be trying to go “all-in” for the web but it is not off to a good start. Android is. ARM is. 2011 could be a very good year for freedom from monopoly.

One interesting stat is that Konqueror shows 3.8%. KDE must be doing well. Perhaps the converts prefer the comfort of starting in the lower-left corner of things…

About Robert Pogson

I am a retired teacher in Canada. I taught in the subject areas where I have worked for almost forty years: maths, physics, chemistry and computers. I love hunting, fishing, picking berries and mushrooms, too.
This entry was posted in technology. Bookmark the permalink.

11 Responses to Global Web Stats

  1. oldman says:

    “What we can do with thin clients has no comparison to text-only thin clients of the old days. Give it up.”

    There is nothing to give up. I spoke as a personal computer user, not as a member of IT. Your arguments in that regard are quite familiar to me. I have made them myself in the context of our roll out of the pilot of our virtual desktop initiative. When this initiative goes live we will be servicing hundreds of users dispersed all over our campus and possibly overseas as well.

    This having been said, I would respectfully submit that as far as thin clients are concerned, for all their sophistication and power, the modern dedicated thin client is only useful as long as the server it works connects to is reachable. Break that connection and it is just as useless as its far simpler text based ancestor was – period.

    As far as your experiences are concerned I will note the following:

    1) Your classroom workstations are all on the same network segment as the servers. Short of a component failure this is a very easy environment to keep up. Were your thin clients scattered across a large multi segmented network connecting back to centralized servers, the possibilities for down time would increase exponentially as would the troubleshooting.

    2) Keeping 20 students with no expectations amused on your terms is easy in comparison to serving the requirements of a varied population of knowledge workers who demand that particular sets of software be supported, and who have the clout to make it happen. You would find very rapidly that your LTSP/Linux based solution would be rejected out of hand as totally unrelated to the needs of the user.

    To Sum up, As an IT professional charged with assisting on a VDI pilot and roll out, I agree with your observations. As a personal computer user who has experienced and continues working in a centralized computing environment, It is my personal conviction that I personally want to stay far away from what I consider to be a technology that is not relevant to my computing needs. which I personally consider as little more that a modernized dumb terminal.

    And I am not alone.

    Merry Christmas and Happy New Year!

  2. In 30 years we have gone through 20 cycles of Moore’s Law and our processors are no longer a bottleneck. The networks are still a limiting factor but then we had a few kilobits/s and now we have multiple gigabit/s locally. What we can do with thin clients has no comparison to text-only thin clients of the old days. Give it up.

    My old Xeon terminal server can keep 20 students amused. With a modern PC with 4gB of RAM and multiple gigabit/s interfaces I could keep 40 going easily. Because there are a few things we cannot do this way and none of those are prescribed in the curriculum we are good indefinitely using thin clients. The thin client of today is more powerful than the thick clients you claim saved everyone from mainframes.

  3. oldman says:


    I am always amazed by how little of the last 30 years of computing you seem to have gotten! The personal computer arose a a break away from overloaded limited time sharing systems. Their power was in providing dedicated personal resources that did not have to be shared with anyone else!! One did not have to contend with the machinations of the tenders of these shared systems, who became more worried with how to shove the maximum number of users into a system than in providing a good experience . One did not have to listen to how something was too “costly” to be done.

    I have spend a large pile of money over the past 30 years periodically refreshing my computer hardware. I consider it money well spent as it has guaranteed me the resources that I need when I want them, on no ones schedule but my own.

    Now I know you will tell me that most computer users do not “really need all the power that they have” and would be “better off” giving up their personal computers for the modernized version of the dumb terminal that is the think client. I know many people would would think otherwise and would probably tell you to keep your modernized dumb terminal.

  4. but it’s not for lower price:

    A buys a PC for $500 and runs it for 10 years. That’s $50 per year.

    B buys a PC for $500 and replaces it with another hotter machine for $400 4 years later. He then buys a third PC for $300 8 years after the initial purchase and he has paid $1200 for perhaps 12 years of use or $100 per year.

    B has paid twice as much as A for IT and they can both read their e-mail and watch YouTube. B has likely tripled the amount of pollution that A causes both in the producing countries and in the using countries. No doubt B has also used more electricity because A’s machine was single-core 32bit and B always upped his cores and RAM and video-card.

    Now, imagine A is part of a family or some organization and he uses a $150 thin client and only upgrades the terminal server. He then gets periodically improved performance while paying less than B. The thin client costs $15 per year. His share of the server costs $30 three times it is replaced. His total cost is $23 per year. While he may give up full-screen video quality, a tiny part of his needs, A gets improved performance over B for everything else and still gets periodic performance boosts.

  5. oldman says:

    “The aspect of the XP machines that interests me is that all of the XP machines can run GNU/Linux and few of them can run “7″. Who is going to take care of them in their old age?”

    Those systems that are under 4 years old with x64 support and the ability to be upgraded to 2Gb RAM should have no problem upgrading to even 64 bit windows 7 . The real issue will be for 32 bit only machines of the vintage up to the age of the ancient history that you are blessed with. Their fate will depend on the owner. If the owner is what what I refer to as computer indifferent, they will continue to use the same system and software until it drops dead, and then go out and buy a new system. The computer literate will long since have replaced their systems with new – often getting more features for the same or lower price.

  6. The aspect of the XP machines that interests me is that all of the XP machines can run GNU/Linux and few of them can run “7”. Who is going to take care of them in their old age?

    In Canada, many of these are recycled to schools and libraries with a new XP licence. Are schools going to want an unsupported OS?

    How many entrepreneurs will relish the possibility of an abundance of free/low-cost PCs to refurbish with GNU/Linux?

  7. amicus_nobrainus says:

    Ray, you’re right! I read the graph wrong.
    So win7 still has less than half the XP share of M$ OS’s. That really must be disappointing for M$ employees and stockholders. M$ has an “End-Of_Life” OS on the street, twice as large as any they’re now trying to sell (vista and 7). M$ can’t even compete well with themselves (and with a 2002 product to boot). Grandpa always did say progress is regress.

    Thanks Ray.

  8. Ray says:

    “I think the “7″ stats in the last graph is just a ridiculous and blatant lie. XP is still the dominant M$ OS.”
    Look closely. In the graph, Windows XP still has 40% of the market, while Windows 7 has 20%.

  9. twitter says:

    The “7” stats may not be a lie, but the conclusion of acceptance is. Because 7 is really just Vista rebranded, the graph only proves that Vista has managed to get another 8% of the market, if the statistics are correct. Most of the 7 people come from Vista, the rest of the Microsoft crowd clings to XP because their software does not migrate to Vista or 7. That makes 7’s growth rate equal to or worse than Vista’s.

    It is possible that botnets have been used to pump up the apparent showing of 7 and Vista. Microsoft regularly engages in channel stuffing and is not above that kind of fraud. Microsoft will be lucky to get the combined market share of Vista and 7 out of single digits.

  10. amicus_nobrainus says:

    I think the “7” stats in the last graph is just a ridiculous and blatant lie. XP is still the dominant M$ OS. I think it was doctored to promote the bloated “7”. And, since “8” will be out in a year, who would be so stupid to dish out money for “7”?

  11. Richard Chapman says:

    I was glad to see Konqueror included in KDE 4.5. I don’t use it very much but it does have some features that Firefox or Chromium don’t have (like user agent on a site-by-site basis). I use all three. Firefox is locked up tight with Cookie Monster and NoScript. Chromium I leave open for everything else.

    Microsoft has a very dysfunctional corporate culture. It’s Bill Gates’ Ego incarnate. It functions best when there’s money to burn. Their culture shows no sign of changing and the money stream isn’t what it used to be. I’m not a big fan of fireworks, but I’ll stick around for this one.

Leave a Reply