Comments on: Reflections on ten releases of Ubuntu http://mdzlog.alcor.net/2009/04/27/ten-releases-of-ubuntu/ a potpourri of mirth and madness Mon, 23 Jan 2012 03:06:44 +0000 hourly 1 http://wordpress.com/ By: mdz http://mdzlog.alcor.net/2009/04/27/ten-releases-of-ubuntu/#comment-1252 Tue, 21 Jul 2009 17:03:33 +0000 http://mdzlog.alcor.net/?p=303#comment-1252 Jef, I won’t be baited in this way. I’ve faced down worse trolls than you. :-)

The only bit of this worth responding to is your assertion that I am the original source of information. I am not. I do seem to have neglected to cite the New York Times article on Ubuntu from January.

If you’d like to question the validity of the reporting, you are welcome to address your concerns to the editor of the Times.

]]> By: Jef Spaleta http://mdzlog.alcor.net/2009/04/27/ten-releases-of-ubuntu/#comment-1251 Tue, 21 Jul 2009 16:16:16 +0000 http://mdzlog.alcor.net/?p=303#comment-1251 You need to realize that all experimental science ultimately relies on estimation. So its a bit ironic for you to call the way your estimation is done as not an exact science. Thanks for the chuckle. You seem to be missing the point. Algorithmic estimation is both exact and is a science as long as your methodology is valid. The methodology used to generate numbers is more important than the numbers themselves.

All this cloak and dagger stuff is getting old.
You know how you are making your estimates.. I don’t.
You have confidence in your estimates, yes? Then you should be equally confident in your methodology, why make such an immense effort to keep it hidden? Its your choice to withhold information with regards to methodology. There is no privacy disclosures by exposing the algorithm that explains how the logged information is parsed.

Writing Shuttleworth personally? That’s a thought. But that suggestion its ultimately a dodge. You stated the numbers in this blog. You are the source of information. You are the one who should be prepared to point to a publicly communicated methodology. The point isn’t for me to get the scripts under NDA. The point is transparency.

-jef

]]> By: mdz http://mdzlog.alcor.net/2009/04/27/ten-releases-of-ubuntu/#comment-1250 Tue, 21 Jul 2009 15:35:29 +0000 http://mdzlog.alcor.net/?p=303#comment-1250 The answers to your first and last questions are both “no, of course not”.

I can’t indulge you further on this point. If you want a copy of the scripts, you’ll have to write to Mark and request their release. I don’t see much value in it, myself. You are quite welcome to disregard the figures if you feel they are not credible enough for you. They are estimates only.

]]> By: mdz http://mdzlog.alcor.net/2009/04/27/ten-releases-of-ubuntu/#comment-1249 Tue, 21 Jul 2009 15:30:43 +0000 http://mdzlog.alcor.net/?p=303#comment-1249 If you’re going to talk in terms of destiny, Jef, you’ll be held accountable for your predictions. Canonical has long said that Launchpad would become open source, and now it has.

I’m sure you will find other things to heckle us over, but I’m glad this one has been put to bed.

]]> By: Jef Spaleta http://mdzlog.alcor.net/2009/04/27/ten-releases-of-ubuntu/#comment-1248 Tue, 21 Jul 2009 15:28:43 +0000 http://mdzlog.alcor.net/?p=303#comment-1248 Be clearer.
Are you counting unique IPs across multiple services?

Which network services specifically?

Are you using a scaling factor of any kind to account for activity you can’t count like mirror activity or the effect of ip address sharing?

What is the time window over which the stated numbers are counted? Is this a running total of unique IPs since 2004?

Write up the methodology…put it in the wiki.
-jef

]]> By: Jef Spaleta http://mdzlog.alcor.net/2009/04/27/ten-releases-of-ubuntu/#comment-1247 Tue, 21 Jul 2009 15:24:10 +0000 http://mdzlog.alcor.net/?p=303#comment-1247 When I wrote that.. I was correct. Its a bit revisionist to suggest otherwise. I’m ecstatic that Canonical came to their senses and openned up Soyuz.

-jef

]]> By: mdz http://mdzlog.alcor.net/2009/04/27/ten-releases-of-ubuntu/#comment-1242 Tue, 21 Jul 2009 10:21:12 +0000 http://mdzlog.alcor.net/?p=303#comment-1242 We log connections from Ubuntu systems to various network services which are provided by Canonical itself.

This is not an exact science. The use of these services is not mandatory for any user, we only identify users by IP address (to protect privacy), and many of these resources (such as security updates) are widely mirrored to local servers.

This is why these figures are always qualified as estimates. In my opinion, they are reasonable estimates based on available data.

]]> By: mdz http://mdzlog.alcor.net/2009/04/27/ten-releases-of-ubuntu/#comment-1241 Tue, 21 Jul 2009 10:12:47 +0000 http://mdzlog.alcor.net/?p=303#comment-1241 It gives me great pleasure, albeit delayed, to point out that you are wholly incorrect on this point.

http://blog.launchpad.net/general/launchpad-is-now-open-source

]]> By: sky http://mdzlog.alcor.net/2009/04/27/ten-releases-of-ubuntu/#comment-820 Tue, 09 Jun 2009 07:59:21 +0000 http://mdzlog.alcor.net/?p=303#comment-820 I would also be very interested to get a direct link to the internet traffic/data you’re referring to.

]]> By: mdz http://mdzlog.alcor.net/2009/04/27/ten-releases-of-ubuntu/#comment-587 Mon, 27 Apr 2009 16:48:32 +0000 http://mdzlog.alcor.net/?p=303#comment-587 Tux Radar published this retrospective which looks at more detail at the changes over these ten releases: http://www.tuxradar.com/content/road-jaunty-look-back-ubuntus-history

]]>