Linux Mint Too Successful

Linux Mint distro had DOSed itself by releasing a too-popular edition.“Our main repository isn’t currently able to serve connections to everybody. This can result in errors, timeouts and delays in apt-get, and in your update manager.Please switch to a mirror while we fix this situation” I guess that’s the price of fame and hard work, you get more fame and more hard work to do. Let’s hope they can scale up. The hits on Distrowatch should have been a clue but it’s better late than never to add servers.

See The Linux Mint Blog – High traffic on the package repositories.

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.

13 Responses to Linux Mint Too Successful

  1. ram says:

    Despite liking Linux Mint for a number of reasons, the default Mozilla Firefox browser defintely could be configured more securely. I think most of the blame rests with the Mozilla organisation which clearly has compromised the browser.

  2. oiaohm says:

    DrLoser 4 billion is too high because it exceeds the number of x86 PC in operation. We are currently at 2 billion x86 PC in operation. Problem is Debian desktops don’t just run on what you would call PC units.
    I can understand not finding Debian PC they are in highly strange areas. Boeing powerpc laptop units for system configuration. The last thing you want is aircraft with a virus. So there are a few of these at every major airport on earth.

    Desktop Debian is strange hardware. That is why its not Redhat or CentOS. You want to use strange hardware for security reasons Debian is your choice.

    This is the worry about Linux security its used in some very scary areas.

    DrLoser going by what you can see is a very bad mistake. You could walk around an airport all day and not see the Debian once because you don’t have the security clearance. When you know what Linux systems are interfacing with and controlling is something very serous-ally wrong that Linux desktop security is so under maintained.

  3. DrLoser says:

    I don’t wish to question your grasp of basic mathematics, oiaohm, least of all when the base is 10, but you do realise that the difference between 100 million and 1 billion is a single order of magnitude, don’t you?

    And that the difference between your conservative and optimistic multipliers is either two or three orders of magnitude?

    By what precise statistical formulation did you compress two to three orders of magnitude down into one?

  4. DrLoser says:

    Converstive guess would be add 3 zeros after Distrowatch numbers to get number of systems.

    Is there such a thing as a completely uninformed (I’m not being rude: I’m just pointing out the total absence of information) conservative guess?

    Optimistic would say add 5-6 zeros.

    Is there such a thing as a completely uninformed optimistic guess?

    Debian project size is guess to be somewhere between 100 million and 1 billion desktop systems.

    Funny, that, because I have literally never seen one in the wild. (Although, full disclosure, I now have Wheezy and Jessie on VMs here at home.)

    Let’s be optimistic and round that big fat zero up to one. No, let’s be wildly optimistic and round it up to five.

    Now, I have certainly seen Red Hat and CentOS desktops out in the wild. (I’ve used three or four of each myself.) I’ve probably seen about a couple of hundred. Divide that by five and multiply by your magic number and we have a guess, an informed guess (though one calculated from a thoroughly disreputable baseline) for Red Hat/CentOS desktops out there as:

    Four billion to forty billion.

    Well, could be. Plenty of room there for Debian to expand into!

    Bit of a insanely wide target zone.

    I’m glad you mentioned “insanity,” oiaohm. Because if I’d mentioned it, Robert would have had me up on charges of an ad hominem attack.

  5. dougman says:

    Well, there is a Linux Mint Debian version too

  6. oiaohm wrote, “Wikipedia for example only saw last month 0.5 percent Ubuntu users. We know 3 percent of Internet users are Ubuntu users because their machines reported in. So where did the other 2.5 percent instead.”

    The Wikipedians are not a very good sample of the world’s users of IT. They are mostly English, for instance, a widely but nonuniformly distirbuted sample. If you look at usage by country you see USA greatly over-represented. We know that in schools in USA GNU/Linux is used in a bunch but still a minority of schools. Same in USAian governments. Many consumers in USA have no access to GNU/Linux on retail shelves, probably in 80% of the country. If you look at certain countries in South America, Asia and Europe you can see quite a different result, and those folks, except for India, are not showing up on en.Wikipedia. Consider Uruguay which shows ~15% GNU/Linux for page-views from desktops and Wikipedia shows 22% for Linux but they count Android/Linux as “Linux”… The two stats are not very comparable.

  7. ram wrote, of Mint, “What is there not to like?”

    I don’t like the idea of a tiny community controlling the software on my IT-system. I don’t like the idea of a few mirrors being my source/backup for software.

    With Debian I have a much larger and more diverse community. It’s like wanting to live in a city. Do you choose a city of 50K, 100K, or 300K people. 50K is marginal in so many ways. ~100K is approaching ideal in that you can have almost anything you want from the community. When you get to millions of people you do get other problems like crime being a problem. There is an optimal size. For a distro, Debian has ~1K developers and that seems to be quite manageable (despite systemd…). Debian has hundreds of mirrors too, so they can serve well just about anyone anywhere. I was working in the Arctic and never felt isolated.

    Do we even know who is contributing to Mint? They say they have two maintainers… They are almost totally dependent on the hard work of Debian. Why not use the original?

    Linux Mint’s website begins with a lie: “Linux Mint is the most popular desktop Linux distribution and the 3rd most widely used home operating system behind Microsoft Windows and Apple Mac OS.” There’s no basis for that. Look at Wikimedia’s stats. They are at less than 1% of Ubuntu GNU/Linux. It’s not only a lie but a very big one. Debian is slightly ahead of Mint on Wikipedia.

    Really, if these guys want to improve on Debian by tweaking the UI, why don’t they join Debian and create a branch there?

  8. ram says:

    My company has moved Internet client machines over to Mint 17.1 (“Rebecca” MATE). It has alot going for it. The graphical interface is clean, intuitive, and fast. The defaults are preset to sensible values, a few things we needed to adjust, but most defaults are fine. The browsers are up to date and include all the necessary multimedia CODECs. The LibreOffice suite is state-of-the-art, JACK audio all works as do the most common media creation tools.

    The distribution is available as a torrent download, and the installation was fairly painless, quite a bit simpler than installing straight Debian or CentOS, or even Knoppix.
    Mint uses the apt-get, gdebi, and Synaptic package managers so those familiar with Debian based distributions will be familiar with the package handling and updates.
    What is there not to like?

  9. oiaohm says:
    DrLoser of course off on wild assumptions. Popcorn uses popularity-contest package that is Debian option in tracking system.

    That 100000+ confirmed unique systems to be using Debian stable last week who opt-in to be tracked. 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. Yet distrowatch show under 2000 hits. Converstive guess would be add 3 zeros after Distrowatch numbers to get number of systems. Optimistic would say add 5-6 zeros. Debian project size is guess to be somewhere between 100 million and 1 billion desktop systems. Bit of a insanely wide target zone.

    Ubuntu tracks by ubuntu time server access. Company behind Ubuntu are reporting 100 million plus active Desktop systems on the Internet. 100 million is about 3 percent of the Internet userbase using Ubuntu alone.

    If Mint is as larger than Ubuntu and Debian then Desktop Linux should have crossed 10 percent market share on the Internet.

    There are some confirmed Linux number out there. None of the numbers are absolutely complete. One thing absolutely for sure Linux is not 1% market share on the Internet. Any website stat reporting less than 3 percent must not have sites that interest Linux users. 5 to 10 percent most likely.

    Wikipedia for example only saw last month 0.5 percent Ubuntu users. We know 3 percent of Internet users are Ubuntu users because their machines reported in. So where did the other 2.5 percent instead.

    This is the problem. The evidence to prove exactly what I suspected all the long is starting to turn up. Linux users on-line don’t behave like Windows users. As more solid numbers turn up I just suspect more and more it will show Linux as being under-counted.

  10. dougman says:

    The DistroWatch Page Hit Ranking statistics are a light-hearted way of measuring the popularity of Linux distributions and other free operating systems among the visitors of this website. They correlate neither to usage nor to quality and should not be used to measure the market share of distributions. They simply show the number of times a distribution page on was accessed each day, nothing more.

    The figures in the third column of each table represent the average number of hits per day (HPD) for the specified period.

    All it says is that Linux Mint is more popular then the other Linux OS’s.

  11. DrLoser says:

    (Page hits, not downloads. Whatever. Let’s assume one download of the distro per page hit.)

  12. DrLoser says:

    I’m not entirely sure that a sudden rush to the fons et origens, as opposed to a mirror site, implies popularity, Robert. Here, from your own cite, are the daily figures for downloads from DistroWatch:

    Last year: 2960
    Last 6 months: 2359
    Last 3 months: 2497
    Last month: 2658

    Call us back when you meet somebody who actually uses the thing, why don’t you?

Leave a Reply