Markus,

I started from a place where the user agent ids stabilized. A huge swath of 
Maven 2.0.x usage is missing because the user agent isn't easily identified, as 
is the case with Ivy and Gradle in the past. I posit that if I actually went 
back and pulled out all the commons-http-client user agents and used the same 
ratios as the chart shows Maven would probably be closer to 85%. Someone asked 
for past stats so I'm going to go back and do another sampling. Definitely 
going forward this is easy to do as the user agents are accurate if we 
normalize to the initial set used in the first report then we have a decent 
baseline. I'm not worried about the missing block of Maven 2.0.x users because 
as they move forward to a version of Maven that is identifiable they will come 
back into play.

On Sep 29, 2012, at 3:52 AM, Markus KARG <[email protected]> wrote:

> Jason,
>  
> thank you for that concise information. It would be great if you could also 
> publish a quarterly sampled line graph on the same stats, so one could easily 
> identify and trends in this. :-)
>  
> Regards
> Markus
>  
> From: Jason van Zyl [mailto:[email protected]] 
> Sent: Donnerstag, 27. September 2012 20:03
> To: Maven Users List
> Subject: Maven is in no danger of being replaced :-)
>  
> I was curious to see what the breakdown is of Maven, Ivy and Gradle use so I 
> took the block of traffic from last week and filtered down the unique IPs per 
> tool across its versions. If you want to take a look at the data pulled you 
> can see it here: https://gist.github.com/3794963
>  
> I used easily identifiable user agents. As Dan Kulp pointed out in IRC many 
> versions of Maven 2.0.x are not accounted for because we never specified a 
> user agent. So if anything, Maven's portion of the pie is bigger. Older 
> versions of Gradle used Ivy, and older versions of Ivy also didn't identify 
> itself correctly. Let's just say it all comes out in the wash and this is 
> what we're left with. This is only traffic against Maven Central so it's hard 
> to tell what's behind repository managers. But I think this is fairly 
> representative. So as far as I can tell empirically from data gathered from 
> the logs in Maven Central, Maven is in no danger of being replaced anytime 
> soon. I believe that the Aether Ant tasks will ultimately replace Ivy in Ant 
> builds, and Gradle has been around for 5 years (give or take) and it's not 
> really making much of a dent.
>  
> 
> Thanks,
> 
> Jason
> 
> ----------------------------------------------------------
> Jason van Zyl
> Founder & CTO, Sonatype
> Founder,  Apache Maven
> http://twitter.com/jvanzyl
> ---------------------------------------------------------
>  
> happiness is like a butterfly: the more you chase it, the more it will
> elude you, but if you turn your attention to other things, it will come
> and sit softly on your shoulder ...
> 
> -- Thoreau 
>  
>  
> 
> 

Thanks,

Jason

----------------------------------------------------------
Jason van Zyl
Founder & CTO, Sonatype
Founder,  Apache Maven
http://twitter.com/jvanzyl
---------------------------------------------------------

believe nothing, no matter where you read it,
or who has said it,
not even if i have said it,
unless it agrees with your own reason
and your own common sense.

 -- Buddha





Reply via email to