[XaraXtreme-dev] Xara's Expectations

[Jed Frechette <jedfrechette@xxxxxxxxx>]

Motivated by a recent report that only ~27% of patches to Mozilla projects 
were submitted by volunteers leading up to the Firefox 2 release [1] I 
started wondering exactly how many outside contributions Xara has received 
during the last 11 months. How far short of Xara's expectations have we, "the 
community", come up?

First a couple of caveats. I do not intend this post to be antagonistic or 
imply that anyone is not acting in good faith and I'm sad that the discussion 
has reached the level that I feel I need to include that disclaimer. 
Furthermore, the analysis provided below is not scientific in anyway. It was 
performed very quickly by someone with only a loose grasp on who the main 
players are so I'm sure someone with more inside knowledge could do a much 
better job. In fact, I would encourage anyone to make corrections as some of 
the numbers I've quoted below may very well be wrong or misleading.

Translations [2]
------------
15,732 messages translated
7 languages
3 languages with > 95% UI translated (not including English)

This would appear to be the area that has been able to attract the most 
outside support. As far as I could tell most if not all of this work has been 
done by volunteers. Even though I am only fluent in English I would like to 
thank everyone who has contributed to this. You all rarely get the 
recognition that you deserve.

Downstream Packages
-------------------
Debian & allies (non-free)
Fedora (recently removed)
...

I know other distros have packages available, unofficial and official, but I 
was to lazy to do much searching.

Bugzilla Activity 2006-03-19 to 2007-02-19
---------------------------------------
368 bugs modified (not created) [3]
112 closed, assigned to Xara employees [4]
103 closed, assigned to Bugzilla Mailing List [5]
67 closed, assigned to non-Xara employees [6]
415 bugs reported [7]
202 reported by non-Xara employees [8]

The distinction between employees and non-employees was made based on the 
simplistic assumption that all Xara employees will have "xara" in their email 
address. Not surprisingly, the vast majority of the closed bugs assigned to 
non-Xara employees are assigned to Alex Bligh. A much larger number of people 
account for the bugs reported by non-Xara employees, however.

In addition to the things that I've tried to quantify above I'm sure Xara has 
received other benefits from their open source move during the last year. For 
example, myself and at least a few other people have purchased their Windows 
products explicitly to support further open source development. They have 
also received a certain amount of publicity and goodwill, although the latter 
seems to be in danger of disappearing.

Although it is clear that a huge number of outside developers have not signed 
on to the project, all in all, when I look at the the above numbers they 
don't look to bad. This is especially true in light of the fact that we are 
talking about the first year of a partially free project that is not included 
in most Linux distributions and is relatively unknown even on the platform 
where it has been sold for years. Despite these handicaps Xara has clearly 
been able to generate interest in the project, otherwise the current 
discussion wouldn't be happening. As for the assertion that people are only 
interested in "stealing" CDraw for their own projects, I find that a little 
bit ridiculous.

Ultimately, however, it doesn't really matter what I think as Xara clearly 
does not feel that their expectations have been met. Which brings me back to 
my original question: What were Xara's expectations going into this 
experiment? Assuming that the release of CDraw was contingent upon some level 
of outside involvement, I can't find any statements on the website that 
explicitly say this, how many more contributions would have been required to 
make this happen? Would 2 more developers actively contributing have been 
enough, 5, 10, 100? How many more accepted patches would have been required: 
20, 50, 100, 1000? Were any fixed benchmarks ever set or was this more like 
the definition of pornography, "I know it when I see it"?

Once again, these observations aren't meant to inflame anyone any further. I'm 
simply hoping that we can learn something from this experiment so that the 
next time a similar situation arises it can be handled more smoothly.


[1] This number is almost certainly low because only developers who had 
submitted 50 or more patches were included. 
http://blog.mozilla.com/seth/2007/02/13/community-participation-stats/
[2] https://launchpad.net/xaralx/trunk/+pots/xaralx
The tinyurl links below point to the results of Bugzilla searches that require 
an account to view.
[3] http://tinyurl.com/29wdjt
[4] http://tinyurl.com/33g9va
[5] http://tinyurl.com/2quqqq
[6] http://tinyurl.com/2zd2x5
[7] http://tinyurl.com/36shuk
[8] http://tinyurl.com/347x9h

Sincerely,

-- 
Jed Frechette
http://jdfrechette.alturl.com

3140 Dead, 23,417 Wounded