Hi, This might worth a look. One interesting point is this stuff: "This calculator estimates how much it would cost to hire a team to write this project from scratch." Result? $5,364,474 with default values :) http://ohloh.net/projects/3189 Programming language usage and codebase history are also interesting. And they have some (false?) findings about licences used. Maybe they care to explain what files they identified with these licenses in more detail. Gabor
Yes, that's quite a bit more than Plone, which is worth $257. http://ohloh.net/projects/70
Hi,
This might worth a look. One interesting point is this stuff:
"This calculator estimates how much it would cost to hire a team to write this project from scratch."
Result? $5,364,474 with default values :)
http://ohloh.net/projects/3189
Programming language usage and codebase history are also interesting. And they have some (false?) findings about licences used. Maybe they care to explain what files they identified with these licenses in more detail.
Gabor
It's interesting... The main problem that comes to mind is that, for example, if I integrate junit into Drupal, commit the code, and I've just put the entire development effort of junit's worth of rank into the system. Not accurate. What's more that site (ohloh.net) has some difficult licensing, and you shouldn't be using it. For example: - The site you are using doesn't allow anyone below the age of 18 years old to register with the site ("You must be 18 years of age or older to register as a member and use the Services.") so I'm not sure these results CAN be accurate. What's more, Open Source projects are not limited to 18+. I mean, 13 I can understand due to legal concerns in the US about COPPA, but 18? What the...? - Any use of the site forfeits every right you have - "When you post Content, you hereby grant us a royalty-free, perpetual, irrevocable, fully sublicensable, worldwide, non-exclusive right to use, reproduce, modify, translate, adapt, publish, create derivative works of, transmit, distribute, perform, display, delete (in whole or in part), and incorporate the Content for any purpose and without acknowledgement to you." So don't post anything interesting, anything you ever intend to publish, etc, because these guys can use it without ever giving source credit (that's unconscionable in my opinion - essentially it's a license for plagiarism for anything anyone posts). All-in-all, this site about Open Source doesn't seem to very "Open" itself - in fact it seems like they spent more time and money on the legal contract than the site contents. I'm sure the inaccuracies that can be shaken out (I'm not convinced they all technically can - crawling project is VERY complicated) will be in time, but the "corporate" license is a little tough pill to swallow. I'm sure Scott and everyone are good people and all, but be careful what you post up on the site - they've got a HUGE legal contract behind it with a couple big downers. Sorry, I know its off topic, but I was pretty shocked when I read the terms. Jonathan On Oct 22, 2006, at 11:10 AM, John VanDyk wrote:
Yes, that's quite a bit more than Plone, which is worth $257.
Hi,
This might worth a look. One interesting point is this stuff:
"This calculator estimates how much it would cost to hire a team to write this project from scratch."
Result? $5,364,474 with default values :)
http://ohloh.net/projects/3189
Programming language usage and codebase history are also interesting. And they have some (false?) findings about licences used. Maybe they care to explain what files they identified with these licenses in more detail.
Gabor
-- Jonathan Lambert Principal | FireBright, Inc. Email: j@firebright.com -------------------------------------------------- "Victory goes to the player who makes the next-to-last mistake." - Chessmaster Savielly Grigorievitch Tartakower (1887-1956)
It is interesting look at for sure, but there are some discrepancies. For example, looking here at me (kbahey) http://www.ohloh.net/projects/3189/contributors?sort=man_months_reverse&page... I have 2.0 man years, and code total is 1290. However, a few users below me "flow" has 0 code total, with no language listed, but still 1.8 man years. Uwe has 1.8 man years, with 4 code total. John Vandyk has 2.1 man years, but code total is 503. Robert Douglass has 22862 code total, but only 1.9 man years. It becomes weirder when you sort by code total: http://www.ohloh.net/projects/3189/contributors?sort=code_total_reverse There is no obvious place where they explain what these metrics mean or how they are derived. Confusing to say the least.
On Sunday 22 October 2006 12:21, Khalid B wrote:
It is interesting look at for sure, but there are some discrepancies.
For example, looking here at me (kbahey) http://www.ohloh.net/projects/3189/contributors?sort=man_months_reverse&pag e=2
I have 2.0 man years, and code total is 1290. However, a few users below me "flow" has 0 code total, with no language listed, but still 1.8 man years. Uwe has 1.8 man years, with 4 code total. John Vandyk has 2.1 man years, but code total is 503. Robert Douglass has 22862 code total, but only 1.9 man years.
It becomes weirder when you sort by code total: http://www.ohloh.net/projects/3189/contributors?sort=code_total_reverse
There is no obvious place where they explain what these metrics mean or how they are derived.
One way they possibly could be derived is using something like the equivalent to `cvs annotate` to get lists of user commits for each file. This would account for people that have spent 1.8 man years writing 4 lines of code. Maybe his code got replaced or committed by someone else. Only the last modifications to a line would show up. Really, without going through each and every line of EVERY commit, it is impossible to gauge how much code each person has spent. And people write code at different speeds too. --Vernon
Hi, Gabor Hojtsy wrote:
Hi,
This might worth a look. One interesting point is this stuff:
"This calculator estimates how much it would cost to hire a team to write this project from scratch."
Result? $5,364,474 with default values :)
http://ohloh.net/projects/3189
Programming language usage and codebase history are also interesting. And they have some (false?) findings about licences used. Maybe they care to explain what files they identified with these licenses in more detail.
This is quite interesting, and I really do not believe that I am at the top of this list ahead of Dries. But I am guessing that I this is only on the contributions repo as I really don't think that I have contributed nearly 8000 more lines of code that Dries. Well I hope not. This is too scary. Gordon.
Of course it is the entire repository. Otherwise, only Dries, Steven, Kjartan, Drumm and killes would be listed. Instead everyone, including translations, modules, themes and documentation are included. The $5.3 million is also for the entrie project. Hope to see this split by core and contrib, but only after they fix/clarify how their metrics work, otherwise it is still questionable.
This is quite interesting, and I really do not believe that I am at the top of this list ahead of Dries.
But I am guessing that I this is only on the contributions repo as I really don't think that I have contributed nearly 8000 more lines of code that Dries. Well I hope not.
This is too scary.
Gordon.
Hi, Khalid B wrote:
Of course it is the entire repository.
hmm. But I find this quite shocking that I have contributed the most lines of code, by a significant number of lines. Also remember that Dries, Steven, Kjartan, Drumm and kills are committing code that they did not write in the form of patches supplied by other people. I know I do this too with EC, but not as much as them. The only explanation that I could come up with is that core was not included. but I see it is.
Otherwise, only Dries, Steven, Kjartan, Drumm and killes would be listed. Instead everyone, including translations, modules, themes and documentation are included.
The $5.3 million is also for the entrie project. Hope to see this split by core and contrib, but only after they fix/clarify how their metrics work, otherwise it is still questionable.
I would love to see this. Gordon.
This is quite interesting, and I really do not believe that I am at the top of this list ahead of Dries.
But I am guessing that I this is only on the contributions repo as I really don't think that I have contributed nearly 8000 more lines of code that Dries. Well I hope not.
This is too scary.
Gordon.
!DSPAM:1000,453c2ac2144563366512726!
participants (6)
-
Gabor Hojtsy -
Gordon Heydon -
John VanDyk -
Jonathan Lambert -
Khalid B -
Vernon Mauery