Slashdot Mirror
Linux Distribution Popularity Trends Plotted
DeviceGuru writes "In order to get a sense of the popularity of various Linux distributions over the past several years, LinuxTrends entered their names into Google's search insights tool and grabbed images of the resulting graphs. The graphs display some fascinating trends and bode well for the future of Linux, particularly its ability to adapt to changing requirements and opportunities. What's especially noteworthy is that Android is the first Linux spin to take on a life of its own within consumer devices. It's certainly not the first use of Linux as an OS for devices; what's unique, however, is that it's the first branded Linux-based OS to be widely marketed to consumers."
Not a very accurate measurement IMHO, although its just "popularity" after all:
From TFA:
Bear in mind that the graphs do not represent distribution sales, downloads, or installed base; rather, the data is based entirely on the number of Google searches containing each distribution’s name per unit time as reported by Google’s search insights tool.
I hope google is successful with android in different devices; Android on a ARM netbook for £100, even Microsoft may not be able to crush that.
I am increasingly convinced that Android and WebOS can't really be counted as Linux, any more than Mac OSX can be counted as Mach+BSD
No sig for the moment.
This is utterly meaningless. These aren't graphs of the popularity of the distros, they're graphs of how often people typed their names into a web browser as search terms.
Find free books.
Maybe this is a regional thing, but who the heck uses Suse? I've been rolling out commercial products using Linux for over a decade, and I never see Suse. Ubuntu, sure. That's mostly for the Windows guys who want to look 'lite. But almost never Suse; the last time I saw it here in Silicon Valley was many years ago.
I do consulting, and so I see a lot of what goes on in the Valley. The standard approach is to use Redhat based distros. Fedora for the cutting edge, CentOS when you need need to get something out without paying, and CentOS/Redhat when the paperpushes want the warm fuzzies of support.
This article has absolutely nothing to do with install base, relative to the rest of the market or otherwise. It's solely google trends, and thus completely meaningless.
Why is even Android mentioned ITFA? Android is just than: an embedded, highly tailored, and customized Linux kernel for specific, embedded hardware devices. Comparing distributions with the Linux kernel for end-user consumption in the market made to run on a plethora of hardware and architectures, great. Don't throw Android in there. To me, that's like counting apples in with the oranges.
However, I'm glad there was some sort of data interpretation done and didn't end up being a Ubuntu flamewar. I, myself, started on Redhat 4.x and grew to love (and hate) the RPM packaging system, along with Redhat's idiosyncrasies on the distribution level. I won't say it's been easy trying to find a balance between Fedora, CentOS and RedHat, as far as when to and not-to use bleeding edge; gamble for enterprise support and stick with the community.
Sure, we all know that Android is based on Linux, but is that really how its marketed to normal people? Seems to be that its marketed as the "google phone" or an "iphone killer" or "look at all these apps". If Android is doing well its not so much that Linux is getting a boost so much as that the Linux community should learn the lesson that normal people don't care about mandatory access control, line-rate packet processing, deduplication backup storage, or whatever else we're on about -- they want "apps".
Why is Windows so successful? Not because people give a crap about Windows, but because there is a lot of software that people want to use, or need to use, and its on Windows. Why is Android popular? Because Google made it, it's not locked to AT&T, and There are lots of cool/useful programs for it. And there are lots of cool/useful programmes for it because normal people are willing to pay $1.99 for a program for their cell phone. Desktop linux is "marketed" (if you can call it that) to normal people often times on cost. It's "free". So they'd feel ripped off if they had to pay $1.99 for a program. Thus, no one charges small amounts for desktop linux programmes, and without the market there isn't that much incentive to write them.
So, good for Google and their phone thing that I don't really want, but not sure Android has much at all to do with Linux-as-we-know-it succeeding in any meaningful way.
I'm not so sure about that. People always want some extra feature.
If they install that awesome app in their phone they'll start wondering, "if the phone is like this, imagine the desktop".
Would have been nice for the authors to explain the y-axis scales.
Where's the graph showing Linux's install base compared to the rest of the market?
Yeah, I was wondering about that, and the whole "This bodes well for Linux" bit.
I mean, all the curves are going down. Ubuntu went up at (what appears to be) all the other distros' expense, but they're going down now. Mint may be going up, but not very steadily.
I know, Android is going up. But that's not really Linux---at least, as I understand it, not in the sense that N900 is Linux. Can you run frozen-bubble//wesnoth/sgt-puzzles/quake/openoffice on Android? (I can on my N900)
So, in what sense does it bode well for Linux? Can anyone who reads that out of the data presented in the article explain it to me? If so, thank you very much :-)
Except that the only way that they are even going to be able to make the connection of Linux = Android is by way of the Internet. No one sells Android as Linux. The average Joe would only learn that Android is Linux-based from a technically-geared article or website, and it's also likely that said website would also refer to desktop Linux in comparison. Fuck, most Americans aren't even aware of Android - they see it, they use it, but they don't know it's called Android, or that it even has a name, and when they have a name for it, they always refer to the entire platform by the manufacturer-specific UI. Owners of Motorola DROID phones call the system DROID, in the same way people assume that the web is synonymous to, and also proprietary to, Internet Explorer.
TL;DR: No one except for a handful of retards will make that connection. No worries.
I know it's not as popular as it once was, but after all it's the grandfather of the group. Back when I first started playing around with linux in around 95/96 the only book(s) you could really find included a copy of Slackware. I purchased two books, one with RedHat and one with Slackware. I know one of the two had a kernel in around 1.2.13 or so, and the other 1.1.something. But that was back in the days where getting X to work was part skill and part magic, among many other things that weren't nearly as easy as what you can do today. Again, just based on age and the fact it was one of the biggest Distros in years passed and helped (in my mind) pave the way for a lot of the newer distro's, I don't believe it should be in a 2nd tier but in the 1st tier myself.
Ubuntu took off from heavy advertising. Advertising and marketing works, and the initial "get as many free disks as you want " shipit deal helped there as well.
Now, what they failed to do is capitalize on this advertising. They should immediately (as soon as they saw they had a hit on their hands) have gone to selling their own machines. Relying on dell. some *mart, some online mom and pop store, etc is not the same.
They don't need every hardware config under the sun, just maybe six machines total, don't go crazy there. This needs to be like an Apple effort, just using open source, integrated hardware and software in nice packages.
They need a netbook (go for ARM, make the breakthrough), a full laptop with optical drive, an entry level cheapish desktop, a higher end workstation/gaming type desktop, expensive but all top shelf components, don't be scared, just do it, a SOHO tower server perhaps that comes with NAS, then a rackmount enterprise server.
They use the LTS releases as the target OS for those machines. The main hired on devs use those machines all the time, so they really grok how to make them fly and get rock solid.
Now, the support is two tier, you own a bonafide branded Ubuntu machine, you get first tier support. Everyone else gets second tier (which is what Ubuntu has now, pure anarchy with hardware, good luck if it works or not, go wade through the forums pages deep with "help me plz").
first tier, separate forum, and the devs, or other offical hired on dudes, DO read the forums there and respond. And quickly. You take the people's money, you answer their questions and fix problems as fast as possible.
First tier branded machines get the REAL "just works" treatment. The rest is like now, good luck with your hardware, might work, might not, go haunt the second tier forums, see what needs to be done there. If you want the every six month bleeding edge releases, oh well, good luck. sure you can run it..but don't expect the same support as they give people who have paid for the hardware and software. Hardware they sell should stay supported for two LTS cycles. That's more than enough now a days for people to milk their hardware out. Chances are, if the hardware and software was really a good match, it would work longer than that, but officially, make it two cycles.
And they should be able to keep cost competitive in this, as they have the resources for economies of scale and some good Q&A before making hardware selection for the branded machines. If local mom and pops can assemble and sell generic machines, so can a big company like Canonical.
Every new LTS release, new hardware comes out, and it "just works", everything, wireless, all of it.
*Most* people don't give a rat's ass about upgrading their OS and machines every six months, look how many people and businesses are still running XP and some older hardware. They want "just works" and "finally, I got this freekin SOB computer figured out and can use it now" over bleeding edge every other week something new is added/updated, and something old that worked, stops working. That gets way annoying to the other 99% of the humans out there who aren't serious devs/hobbiests. Real annoying.
You want linux mainstream on the desktop, or you want to keep it for hardcore nerds only, choose one.
Not to mention some distros have much better organized wikis and documentation than others, so their users don't use google as often when searching for help as opposed to simply searching directly on the website. For instance, Arch has a very well organized wiki requiring maybe 4 clicks to find information on anything from installation to sound troubleshooting, while Ubuntu has separate pages and directories for each release and they like to rename stuff making google your best bet since most information is in forum threads instead of wikis.
How much money are you willing to bet? ... but most of all because the Google Trends statistics are only trends, and normalised at that. They're only valid to judge how a search term varies over time compared to itself, not to anything else.
There are many reasons why there is unlikely to be a significant correlation, including:
- Distros that try not to do things their own way, and where users are far more likely to search on a package than their own distro name when they run into problems. (Personally, I tend to search for "packagename -Ubuntu" whenever I search, both because the noise-to-information ratio is higher for Ubuntu users, and because Ubuntu does everything their own way, so the risk of valid answers not applying is high.)
- Distros that have their own search facilities and support portals likely generate fewer Google hits.
- News releases generating search hits.
They also have a successful desktop OS.
Or perhaps I am a realist that has grown tired of treating Linux like a religion.