| Re: [eclipse.org-architecture-council] SO survey |
On 21 Mar 2016, at 11:15, Christian Campo wrote:
I am NOT really good at interpreting statistics but it looks to me that you compare 26% from a developer developer survey to 9.1 % from a overall desktop browsing statistics. (that is web servers analyzing the client OS)
Yes, I know that - but SO are using their responses to the survey, not their actual detected statistics. If they did I would not be arguing against this stat ;)
Many factors where this can go wrong.
- the browser client might send a wrong user agent (firefox even on mac can pretend to be windows)
Not something most people would do.
- Windows users do more websurfing than OS X users
eh - why? and even if they did you don't get counted multiple times coming from the same ip/session.
If I look in technical talks into the audience you see a lot of Macs (undeniable) I believe a lot more that 9.1 %.
Yes, but these definitely do not reflect the real world. Majority of developers on the planet does not get to go to tech talks.
And BTW Visual Studio could also mean Visual Studio Code which also runs on Mac.
Which have only existed a few months. I doubt the broader world moved that fast.
/max
christian
Von: <eclipse.org-architecture-council-bounces@xxxxxxxxxxx<mailto:eclipse.org-architecture-council-bounces@xxxxxxxxxxx>> on behalf of Mickael Istria <mistria@xxxxxxxxxx<mailto:mistria@xxxxxxxxxx>>
Antworten an: "eclipse. org-architecture-council" <eclipse.org-architecture-council@xxxxxxxxxxx<mailto:eclipse.org-architecture-council@xxxxxxxxxxx>>
Datum: Montag, 21. März 2016 um 10:06
An: "eclipse. org-architecture-council" <eclipse.org-architecture-council@xxxxxxxxxxx<mailto:eclipse.org-architecture-council@xxxxxxxxxxx>>
Betreff: [eclipse.org-architecture-council] SO surveyOn 03/21/2016 08:24 AM, Max Rydahl Andersen wrote:
Just one comment here - that survey looks so completely foobar it hurts.The OS stats goes against every other stat available on this area.
If OSX really is 26% of their users then its weird 35% prefer a Windows only dev environment (Visual Studio).Unless stack overflow would also publish the actual OS usage stats they track I seriously
distrust that survey.Some other indicators allowing to question the reliability and usefulness of generalizing this survey results:
* OSX is not 26% of the IDE market, but 9 to 10% according to various metrics ( https://en.wikipedia.org/wiki/Usage_share_of_operating_systems#Desktop_and_laptop_computers )
* 55% _javascript_ on SO; when TIOBE says 2.33% (as main programming language)
* Eclipse is downloaded about 7 to 8M times for each maintenance version. We can estimate a user base of about 7 to 8M people for Eclipse IDE (some download it multiple times, some other simply try updates so I count 1 dl == 1 user as a rough estimation). Unfortunately, I didn't manage to get download stats for IntelliJ nor Android Studio; I vaguely remember some 2 millions users of IntelliJ last year but didn't manage to find out the sourceThis survey result actually show that SO survey are not representative of the whole development world, it's only representative of the set of people who answered this survey, and this set is not a good sample of the whole developers population.
If someone knows a good way to get Android Studio and/or JetBrains IDEs download stats, please share the link.
Mickael Istria
Eclipse developer at JBoss, by Red Hathttp://www.jboss.org/toolsMy bloghttp://mickaelistria.wordpress.com - My Tweetshttp://twitter.com/mickaelistria
compeople AG
Untermainanlage 8
60329 Frankfurt/Main
fon: +49 (0) 69 / 27 22 18 0
fax: +49 (0) 69 / 27 22 18 22
web: www.compeople.deVorstand: Jürgen Wiesmaier
Aufsichtsratsvorsitzender: Christian GlanzSitz der Gesellschaft: Frankfurt/Main
Handelsregister Frankfurt HRB 56759USt-IdNr. DE207665352
eclipse.org-architecture-council mailing list
eclipse.org-architecture-council@xxxxxxxxxxx
https://dev.eclipse.org/mailman/listinfo/eclipse.org-architecture-councilIMPORTANT: Membership in this list is generated by processes internal to the Eclipse Foundation. To be permanently removed from this list, you must contact emo@xxxxxxxxxxx to request removal.