PDA

View Full Version : Linux loses 30% of its users in a month



monsterstack
July 1st, 2009, 05:30 AM
Uh huh. (http://marketshare.hitslink.com/report.aspx?qprid=9&qpdt=1&qpct=4&qpcustom=Linux&qptimeframe=M&qpsp=113&qpnp=14)
http://i717.photobucket.com/albums/ww177/stuffandstuffand/Screenshot-4-1.png
So, yeah, you know all of those "ZOMG LINUZ HAS 1%!!!!!!!!!!!!11111" reports from a while back? Ask yourself how reliable they are before you keep reposting that myth, k?

Edit: someone will tell me that Linux hasn't lost 30% of its users in 5, 4, 3, 2...

Yamamaro
July 1st, 2009, 05:38 AM
You would think it would be more because of the netbook support.

drawkcab
July 1st, 2009, 05:41 AM
if it is in a chart and on some website then it must be true

monsterstack
July 1st, 2009, 05:48 AM
Say now this is pretty breaking news. People are going to blogging about this, flame wars will start, loads of forum topics about this whole debacle will start. The accuracy of Net Applications' reports will receive more criticism and doubt. And then they'll try and mitigate the damage by claiming that they misplaced a decimal point or something. Life goes on!

CharmyBee
July 1st, 2009, 06:02 AM
Guess half the linux consumer world used intel graphics on jaunty.

monsterstack
July 1st, 2009, 06:02 AM
then they'll try and mitigate the damage by claiming that they misplaced a decimal point or something. Life goes on!

Seems they're already on to it. The stats are now listed as being under review. Man, that's a shame.

dtoronto
July 1st, 2009, 06:07 AM
LOL, I would love to know where someone actually gets this kind of data. I mean linux is an open source kernel, so they can't track it by licensing fees. So if a bunch of people download and install it how is someone supposed to track that it was removed off of everyones computer. I love to see statistics like this because you immediately know it's all crap. Truth is that no one knows where the Linux market share really is because it is extremely difficult to track. it is rumored to be somewhere between 1-2 % for desktop an somewhere above 12% for server use.

credobyte
July 1st, 2009, 06:26 AM
Ridiculous ! How can someone ( not to mention, someone without a chance to "track" users ) count us ? I'm not 100% sure but .. even Canonical wouldn't be competent to provide "accurate" data about Ubuntu users.

monsterstack
July 1st, 2009, 06:31 AM
Ridiculous ! How can someone ( not to mention, someone without a chance to "track" users ) count us ? I'm not 100% sure but .. even Canonical wouldn't be competent to provide "accurate" data about Ubuntu users.

Net Applications base their data solely on how browsers identify themselves on select websites. What those websites are is a secret. What their counting methods are is also a secret. It's no surprise then, that many, many people doubt the legitimacy of their claims. On a completely unrelated note, Microsoft gives them lots of money. Make of that what you will. Even so, their data is pretty much taken as authoritative by pretty much every single media outlet there is, and many Linux users themselves actually believe the 1% figure is accurate. Shame, really.

MaxIBoy
July 1st, 2009, 06:44 AM
LOL, I would love to know where someone actually gets this kind of data. I mean linux is an open source kernel, so they can't track it by licensing fees. So if a bunch of people download and install it how is someone supposed to track that it was removed off of everyones computer. I love to see statistics like this because you immediately know it's all crap. Truth is that no one knows where the Linux market share really is because it is extremely difficult to track. it is rumored to be somewhere between 1-2 % for desktop an somewhere above 12% for server use.I believe on servers it's closer to 45% or 50%.

Tipped OuT
July 1st, 2009, 06:46 AM
Guess half the linux consumer world used intel graphics on jaunty.

Very funny, either that, or they've used Windows 7 and can't enough of it!!! :p

lisati
July 1st, 2009, 06:48 AM
How on earth can the number of downloads be used as a reliable guide to how many people are actually using Linux? Hands up those forum users who have downloaded multiple versions and multiple distros!

MaxIBoy
July 1st, 2009, 06:52 AM
I believe I have about 20 different Linux/BSD ISOs (most never even used) spread out over multiple computers. One of my computers is dual-booting two distros. It used to be that two of them were.

credobyte
July 1st, 2009, 06:54 AM
Net Applications base their data solely on how browsers identify themselves on select websites. What those websites are is a secret. What their counting methods are is also a secret. It's no surprise then, that many, many people doubt the legitimacy of their claims. On a completely unrelated note, Microsoft gives them lots of money. Make of that what you will. Even so, their data is pretty much taken as authoritative by pretty much every single media outlet there is, and many Linux users themselves actually believe the 1% figure is accurate. Shame, really.

I really don't care, I'm on the list or not .. I use it for myself, not for charts ! Regarding .NET - yes, some part of that may be true, tough, not everything.

monsterstack
July 1st, 2009, 06:54 AM
How on earth can the number of downloads be used as a reliable guide to how many people are actually using Linux? Hands up those forum users who have downloaded multiple versions and multiple distros!

That's a great point and all, but it's kinda irrelevant seeing as nobody does that any way.

jrusso2
July 1st, 2009, 07:06 AM
I believe on servers it's closer to 45% or 50%.

Source please?

gymophett
July 1st, 2009, 07:06 AM
That's a great point and all, but it's kinda irrelevant seeing as nobody does that any way.

Lots of people do, including me, unless I am interpreting this wrong.

Eisenwinter
July 1st, 2009, 07:10 AM
The question is, why do you even care? I certainly don't. Wow, so 30% less people use Linux, why does that matter?

LookTJ
July 1st, 2009, 07:12 AM
Source please?
http://news.netcraft.com/archives/web_server_survey.html

monsterstack
July 1st, 2009, 07:13 AM
The question is, why do you even care? I certainly don't. Wow, so 30% less people use Linux, why does that matter?

http://jamie-online.com/random-jamz/wp-content/uploads/2009/06/facepalm.jpg

jrusso2
July 1st, 2009, 07:29 AM
http://news.netcraft.com/archives/web_server_survey.html

This is just apache which runs on more then just Linux. Including Windows and BSD.

I am asking for the total percentage of Linux servers used.

jomiolto
July 1st, 2009, 07:31 AM
OMG! Linux gains 9% market share in Europe! It must be true, it's in a graph! ;)

(source: http://gs.statcounter.com/#os-eu-daily-20090601-20090701)

monsterstack
July 1st, 2009, 07:44 AM
This is just apache which runs on more then just Linux. Including Windows and BSD.

I am asking for the total percentage of Linux servers used.

Google. It's just a few clicks away. Or if you like, have a look at this IDC report (http://www.idc.com/getdoc.jsp?containerId=prUS21399008) from last year:


Overall Server Market Standings, by Vendor

IBM held onto its number 1 spot in the worldwide server systems market with 33.2% market share in factory revenue for 2Q08 growing factory revenue by 13.8% year over year. This growth was driven by solid performance from its System z and System p servers. HP maintained the number 2 spot with 27.4% share for the quarter, growing revenue 3.1% compared to 2Q07. HP's growth stemmed from strong Integrity server and BladeSystem performance. Dell captured the third position with factory revenue growth of 14.1%, increased their market share by 0.9 points year over year. Sun held the number 4 position with a factory revenue decline of 7.2% year over year.

Top-Level Server Market Findings

* Linux servers posted year-over-year revenue growth of 10.0%, for a total of $1.9 billion in the quarter. Linux servers now represent 13.4% of all server revenue, up from 9.4% a year ago.

* Unix servers experienced year-over-year revenue growth of 7.7%. The high-end enterprise segment of the Unix market was strongest of all three segments (volume, midrange enterprise and high-end enterprise), as worldwide Unix revenues totaled $4.6 billion in 2Q08, representing 32.7% of quarterly server spending. Unix servers account for the second-largest segment of spending, by operating system in the worldwide server market.

* Microsoft Windows server revenue was $5.1 billion in 2Q08, showing 1.7% year-over-year growth and comprising 36.5% of all server revenue in the quarter. Windows servers account for the single largest segment of spending, by operating system, in the worldwide server market.

* IBM's System z servers running z/OS experienced the second consecutive quarter of positive revenue growth, with 31.7% year-over-year growth in 2Q08 to $1.6 billion. IBM mainframes running the z/OS operating system accounted for 11.8% of all server revenue in 2Q08.

"IBM regained the top spot in Unix market share on the strength of its Power-based System p and merged Power Systems families, growing revenue nearly 25.7% in the quarter and gaining 5.1% points in year-over-year comparisons," said Steve Josselyn, research director for Enterprise Platforms at IDC. "Sun took second position with 31.1% share, posting a drop of 5.6% points from a year ago, and HP rounds out the top three with 25.8% share and a gain of 1% point. Overall, the Unix market remains a significant source of revenue and competition among the top three suppliers."

13.4% of servers worldwide apparently run on Linux. So there you go. What I think gets people is that they make the mistake of assuming that servers == internet. Servers aren't just used for the internet, y'know. It is probably true to say that Linux runs most of the internet than any other OS, but as with all things in life, things aren't that simple. IDC have been called out for their methodology as well. These things are very hard to do accurately.

Leed
July 1st, 2009, 07:50 AM
I don't think anyone who fully changes to Linux will ever go back, maybe change distros, but not more.

It's just the effort to change that scares many away

-GreyWolf-
July 1st, 2009, 07:54 AM
That's a lot of Bull... how the hell did they get that kind of data if there's no basis. I just laugh at it.

FuturePilot
July 1st, 2009, 07:57 AM
There's no accurate way to measure Linux market share which is why I completely stopped paying attention to such statistics.

areteichi
July 1st, 2009, 08:17 AM
OMG! Linux gains 9% market share in Europe! It must be true, it's in a graph! ;)

(source: http://gs.statcounter.com/#os-eu-daily-20090601-20090701)

It might be related to this news:
http://www.webinfrance.com/police-france-microsoft-linux-201.html
http://www.newsfactor.com/story.xhtml?story_id=13000CYN8S0K&page=1

check the statistics for France (http://gs.statcounter.com/#os-FR-daily-20080701-20090701).

Eisenwinter
July 1st, 2009, 08:23 AM
http://jamie-online.com/random-jamz/wp-content/uploads/2009/06/facepalm.jpg
Then you're an idiot.

monsterstack
July 1st, 2009, 08:26 AM
Then you're an idiot.

You fail for two reasons:

For thinking that Linux losing 30% of its marketshare in a single month is somehow not a big deal.
For thinking that Linux HAS lost 30% of its marketshare in a single month.

monsterstack
July 1st, 2009, 08:34 AM
It might be related to this news:
http://www.webinfrance.com/police-france-microsoft-linux-201.html
http://www.newsfactor.com/story.xhtml?story_id=13000CYN8S0K&page=1

check the statistics for France (http://gs.statcounter.com/#os-FR-daily-20080701-20090701).

Holy Cow (http://gs.statcounter.com/#os-FR-daily-20090601-20090701) [statcounter.com]. These stat sites and services are so terribly bad.

Giant Speck
July 1st, 2009, 09:15 AM
Damn. It looks like Linux isn't as barely popular as it used to be. I might as well erase it from my hard drive and go back to using Windows 7.

madjr
July 1st, 2009, 09:17 AM
sorry to break the news but MJ is dead dude

oh and linux hasnt lost 30%, just that we're not interested in visiting hitslink.com anymore (you know marketshare.hitslink.com)

in fact i see we are doing good over here (much better counter service than hitslink IMHO):

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

benj1
July 1st, 2009, 09:38 AM
why are people so bothered about statistics?

first this is obviously wrong, linux hasn't lost 30% market share, hell the entire linux market share probably falls within the margin of error.

second why does it matter, so what if we have 0.6% market share or 0.9% market share, everybody will still argue over it. Even if miraculously everybody agreed we had reached 1% market share, we aren't suddenly going to magically get good flash support, itunes for linux etc etc etc, the only important thing i suppose is the long term trend, which is up.

so just keep calm and carry on.

monsterstack
July 1st, 2009, 09:58 AM
I really should have made it more obvious that the very idea that Linux could just lose 30% of its users in a single month was absurd in my original post. Maybe I could have even made the fact that these sorts of tests are hopelessly inadequate and inaccurate a little bit more obvious as well. [Sigh.]

areteichi
July 1st, 2009, 10:14 AM
I really should have made it more obvious that the very idea that Linux could just lose 30% of its users in a single month was absurd in my original post. Maybe I could have even made the fact that these sorts of tests are hopelessly inadequate and inaccurate a little bit more obvious as well. [Sigh.]

I think we all know that.
We are just trying to talk more fruitfully despite its inaccuracy. For there may have been some cause for its drop (Windows 7 approaching its release, schools entering summer break, etc.). Just because something is unreliable doesn't mean it's not worth talking about it altogether.

BbUiDgZ
July 1st, 2009, 10:24 AM
if it is in a chart and on some website then it must be true

considering most linux users block things like coremetrics, doublelcick, neilson netratings and google-analitics... it must be true!!

Delever
July 1st, 2009, 10:43 AM
Of course - they decided that this is perfect time to buy Vista.

Johnsie
July 1st, 2009, 01:51 PM
I don't go by those stats... I go by the number of people I actually know personally who use Linux as their main O/S. The answer is 0 and I know quite alot of people. It's all Mac and Windows. Linux still has a reputation for being hard to use and configure, as well as not supporting several popular software and hardware products.

There are billions of people out there in the world. Compare that to the size of the Ubuntu community. Yes there are other distros out there but Ubuntu is one of the top 3.

Whether the user base has grown or shrunk, it is clearly evident from the world around us that Linux is still a minnow in the operating systems world. Ubuntu hasn't caught on and this not a serious competitor to Windows or OSX. The reasons are simple: lack of advertising, no coolness factor, no killer app, no pre-installed machines on the hight street and it's tricky for non-techs to install

Sublime Porte
July 1st, 2009, 02:13 PM
w3counter (http://www.w3counter.com/globalstats.php) (an independant counter, as opposed to those funded by Microsoft) does show Linux use fell by about 0.03% last month (may), but nothing significant. 2.16 to 2.13 isn't a big deal, normal fluctuation.

Biochem
July 1st, 2009, 02:27 PM
Google. It's just a few clicks away. Or if you like, have a look at this IDC report (http://www.idc.com/getdoc.jsp?containerId=prUS21399008) from last year:


* Linux servers posted year-over-year revenue growth of 10.0%, for a total of $1.9 billion in the quarter. Linux servers now represent 13.4% of all server revenue, up from 9.4% a year ago.

13.4% of servers worldwide apparently run on Linux. So there you go. What I think gets people is that they make the mistake of assuming that servers == internet. Servers aren't just used for the internet, y'know. It is probably true to say that Linux runs most of the internet than any other OS, but as with all things in life, things aren't that simple. IDC have been called out for their methodology as well. These things are very hard to do accurately.

And you my friend is confusing revenue and use. Assuming they are taking into consideration licence cost (easiest) and not the whole TCO, Windows and Mac License cost a lot of money while you can get Linux for free. Therefore you can have a a million more Linux server than the other contender combined and still have only 13.4% of the revenue pie.

All to say that numbers don't lie but but stats can, especially when they are manipulated by concerned party.

Sealbhach
July 1st, 2009, 02:44 PM
Nobody uses Linux in Antarctica:

http://i43.tinypic.com/fmnqiw.jpg

You'll notice somebody had a Mac but they brought it home with them when they left.

.

swoll1980
July 1st, 2009, 03:01 PM
I don't go by those stats... I go by the number of people I actually know personally who use Linux as their main O/S. The answer is 0

I know at least 15 people that use Linux as their main os. I guess it just depends on the people you know. 4 of them are members of my family. (they don't live with me, or each other)

moster
July 1st, 2009, 03:29 PM
Calm down. If linux was meant to fall, it would happened long time ago.

Look at another stats and believe in what you want :)

http://www.w3schools.com/browsers/browsers_os.asp

benj1
July 1st, 2009, 03:54 PM
Nobody uses Linux in Antarctica:

http://i43.tinypic.com/fmnqiw.jpg

You'll notice somebody had a Mac but they brought it home with them when they left.

.

well i think its great that the 30+ antartic outposts, with 1000 to 5000 inhabitants are helping to save the planet by sharing just one computer ;)