Обсуждение: Re: PostgreSQL and success of OSS
I wanted to first thank you guys for the help so far. I can tell from the number of responses that you guys like/care about PostgreSQL, which is what you would expect on an open source project. I plan on running through the available data like the CVS repositories, and a tool like the one that Guido suggested (http://statcvs.sourceforge.net/) looks helpful. I would also like any other data that you guys can supply me. One data set in particular would be a count of downloads. Really everything that is available for Source Forge projects would be nice. I think that I can probably mine data on bugs from the mailing list, but is there any chance that there is a bug report database which I might be able to get data from? If you are interested here is a link to a similar study that another student in my department did which I am using as a guideline: http://www.cs.colostate.edu/%7ebieman/Pubs/tse05TrungBieman.pdf Again, thanks for the help. -- A pessimist says the glass is half-empty. An optimist says the glass is half-full. An engineer recognizes that the glass is simply too big.
On Fri, Jun 30, 2006 at 12:38:08PM -0600, Brent Wilkins wrote: > I wanted to first thank you guys for the help so far. I can tell > from the number of responses that you guys like/care about > PostgreSQL, which is what you would expect on an open source > project. > > I plan on running through the available data like the CVS > repositories, and a tool like the one that Guido suggested > (http://statcvs.sourceforge.net/) looks helpful. I would also like > any other data that you guys can supply me. Downloads are a very difficult (and, I would argue, somewhat useless, but that's your call) thing to measure because there are so many sources of the software. For example, just about every Linux distribution now comes with it. Almost nobody who isn't a developer of PostgreSQL actually goes to a CVS repository. > One data set in particular would be a count of downloads. Really > everything that is available for Source Forge projects would be > nice. I think that I can probably mine data on bugs from the > mailing list, but is there any chance that there is a bug report > database which I might be able to get data from? The "bug report database" is the archives of the pgsql-bugs, pgsql-hackers and pgsql-patches mailing lists. > If you are interested here is a link to a similar study that another > student in my department did which I am using as a guideline: > > http://www.cs.colostate.edu/%7ebieman/Pubs/tse05TrungBieman.pdf > > Again, thanks for the help. > > -- > A pessimist says the glass is half-empty. An optimist says the glass > is half-full. An engineer recognizes that the glass is simply too > big. A *real* engineer realizes that a safety factor of two may be enough for some applications, but worries occasionally that it may be too small. ;) Cheers, D -- David Fetter <david@fetter.org> http://fetter.org/ phone: +1 415 235 3778 AIM: dfetter666 Skype: davidfetter Remember to vote!
Brent, > I would also like > any other data that you guys can supply me. One data set in > particular would be a count of downloads. Direct downloads for 2005 were 1.4 million. For a list of the problems with this number as an indicator of any real information at all, see: http://blogs.ittoolbox.com/database/soup/archives/the-count-of-downloadia-part-i-5621 Really everything that is > available for Source Forge projects would be nice. I think that I can > probably mine data on bugs from the mailing list, but is there any > chance that there is a bug report database which I might be able to > get data from? You'd need to mine the archives of the pgsql-bugs list and weed out irrelevant discussions, non-bugs, and duplicates. We've discussed adopting a database-driven bug tracker, but so far the advantages don't outweigh the burdens for a good number of our contributors. --Josh Berkus