ENOUGH! Says Nuffer

Just as SCOG kept creeping out of its grave to sue someone else somewhere else, GROKLAW sprang to life long enough to store the PDF of the ruling“It is further ORDERED that, with respect to all remaining claims and counterclaims, SCO is bound by, and may not here re-litigate, the rulings in the Novell Judgment that Novell (not SCO) owns the copyrights to the pre-1996 UNIX source code, and that Novell waived SCO’s contract claims against IBM for alleged breaches of the licensing agreements pursuant to which IBM licensed such source code.” by judge Nuffer. This should lay to rest the last litigation by SCOG. It has no chance, no money, probably not even an address these days and no basis in law to sue IBM over Linux. We all understood that years ago thanks to GROKLAW but SCOG, like some of the trolls here, never gave up trying. Even after its death in bankruptcy, some lawyers kept shuffling papers to use the last pennies. The tables and chairs were sold long ago. Someone bought the rights to sue and this should give them the value they deserved.

It’s nice to know that GROKLAW is not dead, just hibernating. Maybe spring will wake it up. Surely there’s some legal case interesting enough to fire it up, like world v Obamacare, or world v N. Korea, or … Of course, now that M$ loves */Linux and Apple has acknowledged that round corners are not innovative there are fewer lawsuits in the world but until all those who fought GNU/Linux for two decades die of old age, there’s still hope they will seek to humiliate themselves in the courts.

See PDF mysteriously appearing on GROKLAW.

See GROKLAW

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 and tagged , , , , , . Bookmark the permalink.

41 Responses to ENOUGH! Says Nuffer

  1. oiaohm says:

    DrZealot I really hate dealing with complete moron.
    The Genium INET Trading and Clearing workstation
    is not
    OMX’s Genium INET system

    The clients might be c++, .net java and 100 percent custom things. The server in the Genium INET is pure C++.
    http://www.computerworlduk.com/in-depth/open-source/3246835/london-stock-exchange-linux-record-breaking-system-faces-new-challengers/

    DrZealot the .net server of Genium INET crashed and burned for being under performing. Yes London Exchange tried out the Windows solution and is now using Linux and C++.

    http://ref.onixs.biz/cpp-icap-brokertec-handler-guide/
    Sorry “The Genium INET Trading and Clearing workstation” is the reference client. Too slow and under-performing brokerage firms go out and pay for C++ or FPGA.

    That simply isn’t true in any meaningful sense, is it?
    And you’ve never once gone near a Stock Exchange in your life, have you?

    DrZealot. I have been this is why your googling is not going to help you. There are so many reference things done in .net is not funny yet not used in share-trading production.

    HFT stockexchanges sell on how fast a trade can be performed some are in microseconds they can do trades. http://www.wallstreetandtech.com/exchanges/jse-reduces-latency-launches-colocation-service/d/d-id/1268424

    Please note stock exchanges also sell rackspace to traders so they can shorten their wires between the exchange and the automated trading computer.

    HFT stockexchanges development teams are the biggest pack of speed nuts you have ever meet. There universal for HFT is go faster by any means without undermining security. This include downloading and benching different Linux Distributions looking for performance improvements or defects caused by different build options.

    Remember Linux kernel setting and alter performance of software on Linux massively.

    Lets put it this way DrZealot you hear them talk and about what they do the more worried you get about HFT and the more you kinda wish to back the side to have it banned.

    If your “random Linux solution” emits correct FIX/SAXESS, whatever, and it is compliant with various rules to do with timing, capacity, and settlement, then nobody cares what version of Linux (or anything) it might be. Knock yourself out — go random!
    Here you almost hit the Nail on the head. Why share exchanges are so random. Is they are competitive with each other for the shortest possible settlement time. Gentoo is popular in the Linux world with speed nuts. Share Exchanges server development teams happen to be a huge pack of speed nuts always attempting to improve settlement time. Distributions running HFT stock exchanges are mostly distributions designed for speed nuts in other words the full build from source distributions that can have every part custom built for the highest performing form.

    DrZealot like you before visiting an exchange I was expecting to see like Redhat or some other enterprise Linux distribution that has broad testing. Like it or not that is not what stock exchanges use.

    My six month contract at OMX was entirely dedicated to writing acceptance suites for “companies interfacing with the stock exchange,” and I assert that this is complete nonsense.
    This here also shows you are brain dead. DrZealot Exactly why were you coding acceptance test suites. That is right because OMX cannot release list of approved OS’s because everyone is using whatever they like. Something for someone who has not see the traders those acceptance test suites are are only used against the libraries.

    If you were working on “The Genium INET Trading and Clearing workstation” solution in .net then you were partly responsible for the biggest joke in the share trading world.

    Sorry DrZealot working at OMX does not let you on to the trading floor to see exactly what is being used. The ASX walk through is interesting and scary. Because you have trading company representatives and staff from the ASX listing what skills they are in fact looking for in staff. Also showing off what kind of solutions you will be expected to work with.

  2. DrLoser says:

    And to focus on the following alone:

    Why share trading companies interfacing with the stock exchange will be using random Linux solutions.

    My six month contract at OMX was entirely dedicated to writing acceptance suites for “companies interfacing with the stock exchange,” and I assert that this is complete nonsense.

    If your “random Linux solution” emits correct FIX/SAXESS, whatever, and it is compliant with various rules to do with timing, capacity, and settlement, then nobody cares what version of Linux (or anything) it might be. Knock yourself out — go random!

    At NASDAQ (your speciality)/OMX (my speciality), you could even use the Genium INET solution, which appears to run on top of something called …

    Microsoft .NET4

    Prolly just a beta version, though.

  3. DrLoser says:

    This is one of the fun facts since I am in Australia seeing what the NASDAQ system is happens to be possible. USA cit in the USA cannot see it. Sorry DrZealot I have a far better clue than what you every could.

    Ah, ’tis always the way when your mind is unsullied by actually working in the field. No, oioahm, unless you have actually worked in a Stock Exchange (or in a supplier of IT to a Stock Exchange), you remain clueless about all of this. As indeed you do about everything else, for a remarkably similar reason.

    Read this post again your idiot moron. I did include a cite over this very topic.

    Not an entirely convincing cite, since it comes from one Dr John W Lockwood, CEO of a company who sells the things. However, it does appear that FPGAs are used extensively in HFT, and I am therefore wrong. I therefore retract the assertion that people doing this would be “loonies.”

    What, precisely, this has to do with extensive R&D work in Stock Exchanges on alpha- or beta- versions of random Linux distros, I have no idea. To quote your post:

    Sections of the stock exchange staff are encouraged to download what ever Linux they want to use. Yes the alpha and beta testers of new versions. Why share trading companies interfacing with the stock exchange will be using random Linux solutions. Yes different job offerings from stock exchanges ask for this particularly.

    That simply isn’t true in any meaningful sense, is it?

    And you’ve never once gone near a Stock Exchange in your life, have you?

  4. oiaohm says:

    DrZealot I have had the fun

    If in no other area, you are brutally ignorant when it comes to personal experience of stock exchanges.
    Sorry no I have some bad news for you. NASDAQ OMX’s Genium INET system(this is a shared source project) is used by the ASX. The ASX has visitor trips every year for those studying computer science.

    This is one of the fun facts since I am in Australia seeing what the NASDAQ system is happens to be possible. USA cit in the USA cannot see it. Sorry DrZealot I have a far better clue than what you every could.

    That would be insane, but I wouldn’t necessarily put it past these people. A link please, oiaohm.
    http://mrpogson.com/2014/12/19/enough-says-nuffer/#comment-230616
    Read this post again your idiot moron. I did include a cite over this very topic.

  5. DrLoser says:

    So in 2007 you must have never stuck your nose the the high-speed trading section. Of course not you would have never had clearance even to stick your nose in the development room of a stock exchange to work out they don’t use stock distributions for anything other than terminals.

    1-0 me, I think, Fifi.

    I don’t wish to get into the subject of your nose, but let’s be honest. If in no other area, you are brutally ignorant when it comes to personal experience of stock exchanges.

  6. DrLoser says:

    DrLoser some automated share trading are pure custom programmed fpga chips.

    That would be insane, but I wouldn’t necessarily put it past these people. A link please, oiaohm.

    You don’t have one, do you?

    As usual, the rest of the wall of gibberish was … gibberish.

  7. oiaohm says:

    DrLoser some automated share trading are pure custom programmed fpga chips.

    I’m just suggesting that this is of no interest whatsoever to anybody at all. Stock exchanges who use Linux use rock-solid and very probably customised releases from the likes of Red Hat.

    You wild guesses are normal DrLoserl. NASDAQ uses a modified version of the Gentoo Linux distribution for high-speed trading exchange even that is too slow as of 2011. In other-words a build from source distribution so it can be internally audited.

    http://tech.opensystemsmedia.com/fpga/2012/08/fpgas-speed-trading-on-wall-street/ Yes fpga accelerated trading.

    So in 2007 you must have never stuck your nose the the high-speed trading section. Of course not you would have never had clearance even to stick your nose in the development room of a stock exchange to work out they don’t use stock distributions for anything other than terminals.

    Yes you will see some of them saying Novell or Redhat but that is support contract not OS provision. Why every binary source has to be audited to pass a lot of the newer strict regulations from 2012 issue.
    http://www.washingtonpost.com/business/economy/computer-glitch-sent-markets-into-a-tizzy/2012/08/01/gJQAwNsMQX_story.html

    Minor computer glitch in a exchange due to something doing the wrong and it was possibly detectable by a software audit will cost a exchange millions to billions of dollars. There are quite huge IT departments with stockexchanges.

  8. DrLoser says:

    I quit distro hopping after I started using Linux Mint, which was around 2009-11-29 on version 8. Now I am on 17.1 and it is lovely.

    Get thee behind me, Satan!

    “Mint,” indeed. It is the Spawn of the Devil! Does it not refuse to speak of itself with its True Name, in the Sight of the Lord, which shall be that Mankind shall Righteously describe Mint as Nothing Other than:

    Gnu/Linux/Debian/Yummy Ubuntu Bits/Mint?

    With unfortunate bits of systemd embedded, of course.

  9. dougman says:

    I quit distro hopping after I started using Linux Mint, which was around 2009-11-29 on version 8. Now I am on 17.1 and it is lovely.

    http://jimlynch.com/linux-articles/is-distrohopping-on-the-decline-in-linux/

  10. DrLoser says:

    Majority of Linux users don’t distribution hop either. So most Linux users are not interested in what Distrowatch is doing.

    A cite would be too much to ask, Fifi?

    I hate to repeat myself, but you’ve got lazy beyond belief. You don’t cite your extraordinary claims in the first place, and now you’ve just given up entirely.

    Oh, and incidentally, your conclusion does not in any way follow from your premise, does it?

  11. DrLoser says:

    DrLoser when were you in stock exchanges. I guess before High-frequency Trading and some of the nasty glitches. Alpha/Beta network for testing software is not connected to the main production networks.

    I didn’t say it was. You were the one who claimed that this is a common occurrence. I have no more evidence than you (and you have the habitual zero), and I can’t say that there aren’t people squirreled away somewhere in the back-offices of the typical stock exchange who are fully permitted to download each and every alpha release, as they choose.

    I’m just suggesting that this is of no interest whatsoever to anybody at all. Stock exchanges who use Linux use rock-solid and very probably customised releases from the likes of Red Hat. They don’t piss around with the sort of crap that you are suggesting.

    DrLoser when were you in stock exchanges. I guess before High-frequency Trading and some of the nasty glitches. Alpha/Beta network for testing software is not connected to the main production networks.

    Guess again, Fifi. 2007.

    When were you in real life? I’m guessing, never.

  12. DrLoser says:

    At last we are free of automobile and fire-arms analogies.

    It’s more than that. Debian has a huge repository. It’s like a bank with $free money. It’s very attractive.

    It isn’t really even remotely like that, is it, Robert? It’s more like a bank filled with Czarist bonds from 1913. (Or pick your vintage and bonds.)

    You seem to have given up your strange assertion that Debian is worth it “at any price” and settled on the obvious price-point of zero.

    It’s not that “attractive” even then, though, is it? I don’t see a mad rush of people beating down the doors here.

  13. DrLoser says:

    Stock brokers doing high High-frequency trading are basically one of the performance nuts of the computer world as well as normally being programmers.

    There isn’t actually any such thing as a “stock broker doing high-frequency trading,” oiaohm. Apparently the concept of milliseconds is equally as foreign to you as the ability to count desktops into the hundreds of millions.

    Even a casual observer would recognise that these are automated systems. And even a casual observer would realise that the people who construct the systems (the programmers) are an entirely distinct set from the people who run the models that make the decisions. (Big clue: these people couldn’t program a computer to save their lives.)

    Is speaking the truth entirely foreign to you, Fifi?

  14. oiaohm says:

    dougman
    Stock brokers do not have the time to download Linux distros, nor do they even care. They are interested in closing the next sale of stock!
    High-frequency trading is done by computer. Every mill-second faster can result in more profit.

    Stock brokers doing high High-frequency trading are basically one of the performance nuts of the computer world as well as normally being programmers.

    Since High-frequency Stock Broker trades are done by automated software with them just monitoring they are the ones that do have the time to download or customize a Linux Distribution. Some also use custom program FPGA. This is the problem stock exchange has to interface with a random collection of crap these days.

    There is not a single stock exchange in the entire world that would let their employees download a random unauthorized Linux distro, however fine it might be, inside a Production network.
    DrLoser when were you in stock exchanges. I guess before High-frequency Trading and some of the nasty glitches. Alpha/Beta network for testing software is not connected to the main production networks. 1999 is when High-frequency Trading appears. This is also when trading OS consistency and authorized OS goes out the Window. Each High-frequency trader in a company may be using a completely different OS.

  15. oiaohm wrote, of reasons for not distro-hopping, “too much of a pain learning different package management.”

    It’s more than that. Debian has a huge repository. It’s like a bank with $free money. It’s very attractive. Many of the small distros don’t even have the storage space to hold that repository let alone the bandwidth needed to distribute it. Then there’s the Social Contract and Debian policies. Those are very valuable to users. Other distros have nothing like that. RedHat has a great reputation but then there’s the per-whatever charges. All kinds of things differentiate distros. There’s nothing quite like Debian. Forking it would be an awesome task, probably taking many years to gain any share. Like all FLOSS, there may be enough heavy users of Debian to make the fork feasible but it will be a long time before it has much influence in the wider world. It’s probably easier to do custom installations of Debian than to adopt a new fork/distro.

  16. dougman says:

    Stock brokers do not have the time to download Linux distros, nor do they even care. They are interested in closing the next sale of stock!

  17. DrLoser says:

    Sections of the stock exchange staff are encouraged to download what ever Linux they want to use. Yes the alpha and beta testers of new versions. Why share trading companies interfacing with the stock exchange will be using random Linux solutions. Yes different job offerings from stock exchanges ask for this particularly.

    You have never worked for a stock exchange of any variety whatsoever … well, maybe Prize Bull semen, but other than that … and I have, oiaohm.

    There is not a single stock exchange in the entire world that would let their employees download a random unauthorized Linux distro, however fine it might be, inside a Production network.

    Far be it from me to call you out as a blatant liar. I prefer to call you out as a deluded fantasist whose extraordinarily limited cognitive abilities are completely at odds with the financial world as presently constituted.

  18. DrLoser says:

    Majority of Linux users don’t distribution hop either. So most Linux users are not interested in what Distrowatch is doing.

    A cite would be too much to ask, Fifi?

    OK, I’m going to restrict myself in the most _ridiculous_ way possible. *I’m going to limit myself to recent cites by Robert.*

    David E. C. Gettis, Esq. explicitly mentions the following distros:

    * Bodhi Linux
    * Mint
    * Debian
    * Crunch Bang
    * Lubuntu

    Given his approbatory quote, “Lubuntu has become my new home,” I believe we can safely describe this as “distro-hopping.”

    Not that he seems entirely satisfied with the results. I won’t spoil the fun for you. Read the post yourself!

  19. oiaohm says:

    Why? Because the corporate market is not very closely aligned to the Distrowatch market, basically.
    Majority of Linux users don’t distribution hop either. So most Linux users are not interested in what Distrowatch is doing.

    A quick note on “distro-hopping,” btw. I happen to think it’s a good thing, particularly if you are going to choose a Linux desktop.
    Really 70+ percent of Linux Distributions are basically Debian or Ubuntu customized. Most deb system users will not hop to a deb based and most rpm based users will not hop to a deb base why its too much of a pain learning different package management.

    The reality is distribution hoping is starting to die out.

    But, let’s face it, it makes this “100,000+ unique seats” claim sound a bit silly, doesn’t it? How long does each seat last on average? Eighteen months?
    Sorry DrLoser is stupid guessing again popcon data says 6-10 years is a systems lifespan running debian showing up on popcon. popcon data collection has been running for quite a long time. Ubuntu and Centos equal to popcon also shows the same lifespan ranges. Number of Linux users distro hopping is less than 1 percent if collected stats are anything to go by. Most find a distribution and settle in.

    The known lifespans of Linux solutions is why claiming Windows is only a 50 dollar cost is so far out its not funny. Windows + extended support matches to Linux Distrobution.

    It’s quite possible that Debian has a longer purchase and higher loyalty, but we’re still talking shifting sands here.
    Data basically suggests this is the case. Longer purchase and higher loyalty.

    “100000+ confirmed unique systems” but then DrLoser changes it to ” 100000+ unique seats”
    Debain installs with X11 installed could be running thin terminals. So attempting to say 100000+ unique seats is data short. We don’t have clear data on seats. To get seats I would have to run the popcom data again looking for any signs of thin terminal usage then allow for it. Of course DrLoser changing words with incompetence again.

    You yourself have pointed out that there are biases inherent in Distrowatch. They’re innocent biases, but one has to take them into account.
    Of course you are incompetent here. The issue with Distrowatch is called method bias. All I was doing was the methods you need to check for method bias.

    Its not innocent to blindly take figures and attempt to make a point of view without understanding the error factors. DrLoser its fraudulent usage of numbers or moron with no brains using numbers.

    As you can imagine, OMX, like all stock exchanges the world over, actively encourages employees to download their Linux desktops from absolutely any source that they feel like.

    I mean, hey, what do they have to lose?

    https://www.linkedin.com/jobs2/view/11112789
    You really need to watch the jobs DrLoser. Share trading companies use Debian, Redhat and Suse for high speed share trading desktops by what they ask for in job applications. Window is too slow as well.

    Sections of the stock exchange staff are encouraged to download what ever Linux they want to use. Yes the alpha and beta testers of new versions. Why share trading companies interfacing with the stock exchange will be using random Linux solutions. Yes different job offerings from stock exchanges ask for this particularly.

    “actively encourages employees” kinda needed a all in there. General users will be Linux Google with Goobuntu.

  20. DrLoser says:

    A quick note on “distro-hopping,” btw. I happen to think it’s a good thing, particularly if you are going to choose a Linux desktop.

    But, let’s face it, it makes this “100,000+ unique seats” claim sound a bit silly, doesn’t it? How long does each seat last on average? Eighteen months?

    It’s quite possible that Debian has a longer purchase and higher loyalty, but we’re still talking shifting sands here.

  21. DrLoser says:

    That may well be true for larger organizations with a lot of IT-inventory but the small guys for whom most of us work do need all the help they can get…

    What an interesting choice of poster child, Robert. For starters, he’s a lawyer. We all love lawyers. Lawyers are almost as cuddly as marketroids, aren’t they?

    Secondly, David E. C. Gettis, Esq — I love this bogus Esq that a certain type of American lawyer insists upon — didn’t start off as a lawyer. He started off as an “entrepreneur.” Clearly not a very successful one.

    Thirdly, he’s a “solo practitioner.” I believe you have had four separate careers, Robert, which is three more than me, so you might be able to shed light on this.

    When was the last time you worked for a solo practitioner?

  22. DrLoser wrote, “the corporate market is not very closely aligned to the Distrowatch market, basically”.

    That may well be true for larger organizations with a lot of IT-inventory but the small guys for whom most of us work do need all the help they can get and I would bet quite a few visit Distrowatch. e.g. A lawyer:
    “After over two years of running Linux, I’m never going back to Windows. The flexibility, reliability, speed, and instant availability of free software has become addicting.
    I first tried Linux out of frustration with the copy-protection practices of Microsoft and other Windows-based programs. Every time I upgraded my computer, I had to re-validate my copy of Windows, which sometimes required a phone call to Microsoft and other vendors. There are few things more frustrating than begging a vendor for permission to use a product that I already paid for! Now when I upgrade my computer, Linux magically recognizes my new hardware, and I’m up and running again.”

    He recently went distro-hopping after Ubuntu annoyed him. One of the huge advantages for any organization large or small is the absence of lock-in with GNU/Linux.

  23. DrLoser says:

    I missed out OMX (the Swedish Stock Exchange), btw.

    As you can imagine, OMX, like all stock exchanges the world over, actively encourages employees to download their Linux desktops from absolutely any source that they feel like.

    I mean, hey, what do they have to lose?

  24. DrLoser says:

    DrLoser you said you had Sean and used a Redhat desktop. That floats around the 500 mark on Distrowatch at the moment. Reactos also floats around 500. So for you claim to hold Redhat desktops has to be as common as Reactos Desktops we both know that is not true.

    Specious reasoning, oiaohm. You yourself have pointed out that there are biases inherent in Distrowatch. They’re innocent biases, but one has to take them into account.

    Of the four commercial operations I have worked with that used Red Hat, I can’t think of one that would have troubled Distrowatch’s numbers. Well, maybe the embedded software consultancy down the road: they had five Red Hat desktops. (Out of seven, total.)

    The biggest two, both being Telco suppliers, pretty much took their RHEL desktops images straight from the obvious place — Red Hat. (I wouldn’t swear that the earlier of the two didn’t use a Red Hat version earlier than RHEL 2.1, so don’t please let’s start quibbling about nomenclature.) That’s a total of ~50 seats + ~100 more offshore in the first case, and ~30 dev + ~30 back-office/support in the second.

    So, I probably underestimated the number of Red Hat desktops I’ve seen. But, and trust me on this one, only a small percentage would have shown up on Distrowatch.

    Why? Because the corporate market is not very closely aligned to the Distrowatch market, basically.

  25. oiaohm wrote, “You really cannot make a any percentage of usage guess from Distrowatch”

    Distrowatch is not about usage but about inquiries. How many are reading about one or another particular distro beyond the headlines? At best those stats are an indication of rate of growth or competitivity or mindshare. Typically, a newbie will be introduced to some distro somehow and if there’s anything difficult/challenging/awkward about it, the newbie will naturally wonder if there are alternatives. I remember starting with Caldera and when they folded went to Mandrake. They tried to roast CD-drives at one point and tried to go “commercial” or something so I tried Slackware, Fedora, LTSP (then based on RedHat), Ubuntu GNU/Linux and finally Debian GNU/Linux. I wasted a lot of years, believing the FUD about Debian. It’s not hard to install and it’s certainly very flexible. I think it’s superior to Ubuntu GNU/Linux in several ways and Ubuntu is quite successful. Along the way I spent many happy hours reading on Distrowatch but now that I’ve discovered Debian there’s little need to visit except to read those stats… I don’t even need to click on Debian there because I already know a lot about it, yet I’m a user and that’s not what the stats there indicate, usage.

  26. oiaohm says:

    Please note DrLoser Ubuntu could be counting Goobuntu and other internal based Ubuntu if they have not changed clock server.

  27. oiaohm says:

    But … if we take the Distrowatch downloads as a proxy for “unique end-user desktop” percentages amongst Linux distros
    You really cannot make a any percentage of usage guess from Distrowatch .

    DrLoser you said you had Sean and used a Redhat desktop. That floats around the 500 mark on Distrowatch at the moment. Reactos also floats around 500. So for you claim to hold Redhat desktops has to be as common as Reactos Desktops we both know that is not true. Redhat desktops are way more common than Reactos ones.

    Distrowatch is more how many news articles mention them.

    Your “pessimistic guess” was 1 in 1000, leading you to a bottom end of ~100 million. Consistent. Absurd, yet consistent.

    Your “optimistic guess was between 1 in 100,000 and 1 in 1,000,000.
    Those are distrowatch multiplexers when you try to work it out. Distro watch caps out at 2358. So 2358000 to 2358000000 for Mint. Popcon is only 1000 or 10000 or 100 million to 1 billion.

    The distro watch numbers are too wide to be useful. Even 1000 multiper has to be questioned with Reactos having about 500. So I would now have to be 10 or a 1 000 000 multiplier. So 23580 to 2358000000 for Mint. The values are that large covering so much space they are worthless to even use to guess what Linux is even more popular. Reactos known users is less than 5000 yet it can show numbers about the same as Red Hat.

    The error rate is that large something with a count of 1 in 200+ something place could have more users than something in number 1 place on Distrowatch.

    I’m not even sure that the ration of Popcorn installers isn’t nearer to 1 in 5, although I have equally no basis for this guess.
    You can tell how many popcon users are on VM DrLoser. Particular packages are installed for VM operations inside a VM. I have already subtracted that number to the 100 000. So that is 100 000+ on bare metal. Yes popcon is complete package lists so running on a VM shows up. Some how I don’t think you would put a popularity package on a server and the mix of packages do suggest desktop usage.

    Its about 100 000 bare metal with a X11 environment counted by popcon. So that is a number of Debian users we know exist.

    The best place to count would the the time servers. Unfortunately most distributions use pool.ntp.org this includes debian. Ubuntu is an exception ntp.ubuntu.com so we have some solid counts from Ubuntu.

    100+ million is just ubuntu all-own counted by their own timeserver. This will count if or if not installed in a virtual machine.

    Ubuntu equal to popcon only reports about 50 000. So I am being down right conserve saying Debian is most likely 100 million+ it would not be more than 1 billion.

    Please note Ubuntu server edition and desktop edition do message ntp server slightly differently.

    DrLoser you are the one who believes in wild guessing with numbers.

    I am not in fact wild guessing. I am taking what are called educated guesses from known data. But sometimes the data is that bad when you start applying correction values it starts looking like a wild guess. This is because the data is not good enough quality. As soon as you put all the corrections required on Distrowatch it is worthless.

    DrLoser I would like to know where there is somewhere with more quality data than I have.

  28. DrLoser says:

    Ubuntu surveys tell us a lot of interesting things. Like less than 0.1 percent bought their computer without an OS or with Linux. Between 75 and 80 percent of Ubuntu user run Windows, OS X or wine.

    Ah, yes. I was so entranced by oiaohm’s complete inability to handle large numbers that I forgot this bit.

    Neither one of those “things” is of any interest whatsoever, is it?

  29. DrLoser says:

    And the absurdity continues.

    100 million users equals about 5 percent of the market.

    For desktops world-wide, yes. And you’re already at Robert’s giddy heights of Linux desktop usage in markets such as Ethiopia, Venezuela, maybe even China or India … but not world-wide.

    And not a single one of these markets is the exclusive province of Debian, vis-à-vis the desktop.

    And here is actually where we can make use of the despised Distrowatch numbers Let’s take those distros that are fairly consistent on a daily basis over the last year, shall we? I have no idea what they are. Hang on, I’ll look them up.

    The top nine seem to be hardy perennials. Along with Debian, we have Mint, Ubuntu, Fedora, OpenSuSe, Mageia, Arch, elementary and CentOS. I’m doing the Cause a massive favour here by omitting both RedHat (proper) and Goobuntu (and other “in-house” desktops, eg LiMux).

    Do The Math, as Robert likes to say. Debian comes out at a very creditable 1 in 8 of these downloads.

    But … if we take the Distrowatch downloads as a proxy for “unique end-user desktop” percentages amongst Linux distros, this suggests that oiaohm’s already ravingly mad pseudo-mathematical guess for Linux desktops worldwide has to be inflated by a factor of eight.

    Which means that, on the numbers oioahm insists upon, and even though they don’t match whatever he scribbled on the white-board, he is apparently claiming:

    Linux desktops worldwide number somewhere between ~800 million and ~8 billion.

    Sometimes, oiaohm, when the numbers your theory comes up with just scream “D’Oh!” at you …

    … it’s time to discard the theory.

    Eat a donut. It will do you good.

    Ubuntu surveys tell us a lot of interesting things. Like less than 0.1 percent bought their computer without an OS or with Linux. Between 75 and 80 percent of Ubuntu user run Windows, OS X or wine.

  30. DrLoser says:

    My numbers were not off of wild guess from hell Distrowatch hit counter.

    So it was off Popcorn after all. You could have saved us that interminable seventy one word (yes, I used wc) first sentence, and the incredibly tortuous syntax, by just admitting it.

    A word before going into your calculations once more. The idea that only one out of a thousand downloaders will “opt in” is farcical/b>.

    1) It isn’t an opt-in, web-page sort of thing. It’s a “do you want to install this software?” thing. In the middle of an installation. People tend to say “yes” to this sort of question.
    2) And that would be randomly selected people saying “yes.” Here we have “self-selected” people being asked the question. Pretty much all of them made a personal choice to download Debian.
    3) Heck,
    I even nearly said “yes,” both times. The other thing that motivates people to say “yes” is that they’re being given something for free, and just being registered on a usage count doesn’t seem like an onerous thing to pay back.

    I actually felt bad about not installing Popcorn, and I thought about it, and then I decided that I would be unhelpful if I did, because a couple of barely-used VMs would skew the figures in the wrong way.

    I’m not even sure that the ration of Popcorn installers isn’t nearer to 1 in 5, although I have equally no basis for this guess.

    But that wasn’t your guess, was it, oiaohm?

    Your “pessimistic guess” was 1 in 1000, leading you to a bottom end of ~100 million. Consistent. Absurd, yet consistent.

    Your “optimistic guess was between 1 in 100,000 and 1 in 1,000,000.

    You seem to have missed out the magic multiplier for “one billion unique Debian desktops” there.

    But you do seem to have claimed, via an optimistic prediction, that there might be as many as one hundred billion out there.

    If you still can’t see how spectacularly silly your claims make you sound, then I really can’t help you here.

  31. oiaohm wrote, “That is about the end of how useful Distrowatch is from a statistical point of view.”

    Exactly. Collecting statistical data was never the intention but it the best place to learn about the features of diverse distros for the newbie of the distro-hopper.

  32. oiaohm says:

    100 million – 1 billion is based off Debian project own statistical collection popcon cross references with Ubuntu know 100million and their popcon equal that shows more than 1 in 10000 and less than 1 in 1000 users take part in popcon style(yes ubuntu runs its own equal to popcon) collection so 100 000 unique is systems counted on popcon is at least 100 million users and less than 1 billion. The average rates of taking part in voluntary optin software surveys is between 0.1 and 0.01 percent this has also being found true with Windows software stat collections so the correction here is not even strange .

    My numbers were not off of wild guess from hell Distrowatch hit counter. Debian numbers have terrible large correction values but they are way better than Distrowatch for working out how many users there really is. Also a 1000 to 10000 correction value is most likely the best you are going to get.

    Conservative Numbers are the lowest possible case with known errors.
    Optimistic is the highest number that could be generated by know errors.

    Problem you have here is when RHEL had 10 million documented desktop users Distrowatch showed a magic value of 10 also there have been a few showing 100 on distrowatch when they have had known counts in the above 10 million. So this creates the evil 10^5 or 10^6 error correction factor for Optimistic for Distrowatch. That starts having absolutely impossible numbers bit come out.

    The fact it error results in generating an optimistic value greater than the number of x86 PC in usage tell you that you cannot use that source.

    Debian desktops worldwide number, on a conservative estimate, ~100 million, and on an optimistic estimate, up to ~100 billion.
    This is not what I said. DrLoser to go around quoting that Linux is only 1 percent you legally should have the documents to back your claim.

    Linux desktop users are over 10 percent of the market. But due to 80% of them duel booting or vm running you could expect only 2 percent to show up in some forms surveys. Problem here is the value should not be less than 2 percent unless another error is coming into play.

    Ubuntu surveys tell us a lot of interesting things. Like less than 0.1 percent bought their computer without an OS or with Linux. Between 75 and 80 percent of Ubuntu user run Windows, OS X or wine.

    100 million users equals about 5 percent of the market. So you could claim about 1 percent of the market uses Ubuntu exclusively. The survey numbers back this claim. You can expect Debian due to it being about the same size that it also has 1 percent of exclusive users. Realistically about 8 percent of users will be like you DrLoser with Linux in a virtual machine or duel booting just using the users of Debian and Ubuntu. So majority you may not spot.

    Remember less than 1 percent of a large networks computers need to go rogue to cripple the network.

    Distrowatch is useful for a few things.
    1) Relationship between distrobutions. Like what percentage are Debian relations this is over 70 percent of them.
    2) Number of active distributions. This has been reducing since the year 2011 with a peak of 323 and its currently 285 distributions left at a lost of about 10 per year its going to be a while before that trims down to sane levels. It took over 20 years to end up with too many distributions and it will take 20 years for it to drop to sane numbers.

    That is about the end of how useful Distrowatch is from a statistical point of view.

    Please note Linux Desktop does not include Android in these numbers. Android instantly take you up to over 50 percent of market. If you were aiming as a hardware maker to only make 1 driver to sell to the largest market share you could the choice would be Linux Kernel due to Android + Linux Desktop any other choice would restrict your sales options more.

  33. lpbbear says:

    Once again Pog posts an article regarding something of importance to Linux/Open Source folks……thanks Pog, good news.

    Once again DrSlapsHisGumsALot posts his usual off topic rubbish, not just once but 4 posts and counting.

    It wouldn’t be so bad if a few visitors here would simply respond with the following to shills like DrBlabALot…………..

    “crickets”

    (try it, just resist the urge to respond to the moron, not so hard)

  34. DrLoser says:

    Come on what are you going for having to owe everyone here apologies and hoping no one sues you??

    Quite the reverse, oiaohm. I genuinely hope that you do sue me. In fact, I strongly suggest that you do so. You have, after all, stated outright that I am a criminal: if you can’t put me behind bars, where apparently I belong, the least you can do is to cripple me financially.

  35. DrLoser says:

    DrLoser mathematical guesses from poor data always contain at least 1 insane value. Note I said from 100 million to 1 billion.

    Whilst waving your hands wildly in the air, yes. I’m happy to accept that you did so.

    I didn’t wish to question your calculations too closely at the time, but since you seem so fond of them, I shall do so now. They revolved around a baseline and a (spurious) exponent. You defined the (spurious) exponent with considerable precision: 10^3 at a conservative guess, and 10^5 or 10^6 at an optimistic guess.

    You were a little less precise about the baseline. It isn’t clear whether you meant the “100000+ confirmed unique systems … opted in to be tracked” via Popcorn, or the 2000+ from Distrowatch.

    It ought to be clear, because you actually specified that you were “adding zeros” to a Distrowatch number. Possibly a completely different Distrowatch number, because I can’t get 100 million – 1 billion out of 2000, unless I actually take the optimistic exponent.

    It might have been the Popcorn figure, which would make more sense, because they represent a base at a given point of time and not an averaged hits-per-day. And your “argument,” such as it was, for the exponents was as follows:

    Yes the popularity-contest package you have to manually choose to install so this is only really a small percentage of debian users who go to the effort of installing the package.

    Taking the Debian “confirmed unique etc” base number and applying your exponents, oiaohm, I come to the conclusion that you are suggesting (though not correctly quoting, given the three orders of magnitude I mentioned) the following range:

    Debian desktops worldwide number, on a conservative estimate, ~100 million, and on an optimistic estimate, up to ~100 billion.

    “Mathematical guesses from poor data always contain at least 1 insane value?”

    What does that even mean?

    Is there some fundamental axiom of statistics that I am missing here?

  36. oiaohm says:

    DrLoser now you owe Robert and Me both apologies for fraudulent representation. Come on what are you going for having to owe everyone here apologies and hoping no one sues you??

  37. oiaohm says:

    DrLoser mathematical guesses from poor data always contain at least 1 insane value. Note I said from 100 million to 1 billion.

    First oiaohm asserts that there are billions of Debian desktops out there
    Again this is another one of your lies to suit your point of view. You love fraudulent representation using other peoples name DrLoser. It was you DrLoser who said 4 billion redhat users it you with the insane point of view.

  38. DrLoser says:

    The ruling elite is not afraid of the terrorists, it’s actually secretly grateful to them for providing the excuse for their whole surveillance programme.

    No, no, no, Dougie. You have to try harder than that. It’s totally incoherent. You haven’t made an attempt to define “the ruling elite,” have you? And what about the “terrorists,” eh? And the very least you could do when ascribing things to the “whole surveillance program” is to mention the odd detail here and there.

    You barely sound like a paranoid nutter at all. How are you going to join the Big Boys without at least mentioning Black Helicopters?

    You might at least have mentioned the Department of Homeland Security (thank you, President Bush!). But given where you live, you probably make a lot of money out of these bozos.

    Now, this is the bloodshot-eyed, frothy sort of mania you want to aim at:

    The hereditary elite are behind the “terrorists”. Anybody really fighting their colonial exploits is working in Russia, China, India, and Brazil.

    See? Starts with “hereditary,” which is always a good idea. (I would have started with twelve-feet tall extra-terrestrial lizards, myself, but it’s a broad church.) Drops “colonial” in for no readily apparent reason. Can’t quite get the BRICs in order, and is a little behind the times, but has presumably just recently caught up with Jim O’Neill back in 2001, and Jim is in no way part of the hereditary elite.

    Now, this is the perfectly circular model of paranoia that you should be aiming for, Dougie.

    When they strike, with billions of troops and thousands of nuclear missiles, it will not be called “terrorism”. We’ll be lucky if anybody is left to call it anything at all!

    First oiaohm asserts that there are billions of Debian desktops out there (a harmless delusion), and now ram claims that the “hereditary elite” has billions of troops just waiting on standby.

    Well, I beg to differ.

    The clear and present danger to Western Civilisation is:

    Amazonian Tree Frogs.

    Think about it. These poor little buggers have been threatened with extinction, not merely by the “hereditary elites,” but also by one or more members of the BRICs — naturally under remote control from somewhere or other.

    Amazonian Tree Frogs will fight back, I tell you! It is a matter of life or death for them! And don’t forget:

    1) They are almost all very, very poisonous.
    2) In every single photo you see online, they are unspeakably cute.

    Ladies and gentlemen, I conclude my case. And also, have you ever seen an Amazonian Tree Frog with a foreskin?

    They’re all Jewish, I tell you!

  39. dougman says:

    $500 trillion in combined assets, the world is a chessboard and we are their pawns.

  40. ram says:

    The hereditary elite are behind the “terrorists”. Anybody really fighting their colonial exploits is working in Russia, China, India, and Brazil. When they strike, with billions of troops and thousands of nuclear missiles, it will not be called “terrorism”. We’ll be lucky if anybody is left to call it anything at all!

  41. dougman says:

    Groklaw will stay down to avoid the very same reason that happened with Lavabit.

    The ruling elite is not afraid of the terrorists, it’s actually secretly grateful to them for providing the excuse for their whole surveillance programme.

    https://www.youtube.com/watch?v=XKjW2dUTMRI

Leave a Reply