I still continue to find articles on the internet downplaying the seemingly normal and sweeping adoption and acceptance of the GPLv3 license. This should point out a few things that indicate that GPLv3 is "here to stay".
We have quietly passed 5000 GPLv3 projects (and related) as a combined number from:
GPLv3
Or later
LGPLv3
A little associated background . . .
http://www.fsf.org/licensing/licenses/gpl-faq.html#TOCv2v3Compatibility
http://www.fsf.org/licensing/licenses/gpl-faq.html#VersionThreeOrLater
http://www.fsf.org/licensing/licenses/gpl-faq.html#AllCompatibility
Current total: 5100
Check it out at http://gpl3.palamida.com/
Rate of GPLv3 related discovery: 20~ day currently
What does this mean?
GPL projects hosted on SF: 102137 ( http://sourceforge.net/softwaremap/trove_list.php?form_cat=18 )
Estimated active projects: 15320, 15% of total ( http://asay.blogspot.com/2005/09/analyst-nature-and-size-of-open-source.html )
note - I corrected the formulas on 8/22/07 as follows
Original:
Total repository based OSS community: 145910 (SF total divided by 70%)
Estimated Total active Projects: 21886 (total * 15%)
Estimated total GPLv3 conversion, including "or later": 13078 (total active, times 77% GPL and 6% LGPL, times 72% estimated conversion rate)
Estimated current "or later" impact: 9082 (50% of GPL)
Therefore:
- Or later – 4708 of 9082 projected – 52%
- LGPLv3 – 30 of 723 projected – 4%
- GPLv3 – 5100 of 13078 projected – 39%
- GPL, not converted – 5086 projected (GPL projects times (100% - 72% convert rate))
- Non GPL license – 3722 projected (Active projects – (Active GPL projects + not converted))
Corrected:
Total repository based OSS community: 208,442 (SF total * 70%)
Estimated Total active Projects: 31,226 (total divided by 15%)
Total active GPL: 25,917 (total active, divided by 77% GPL and 6% LGPL)
Estimated total GPLv3 conversion, including "or later": 18,660 (total active, divided by 77% GPL and 6% LGPL, divided by 72% estimated conversion rate)
Estimated current "or later" impact: 12,958 (50% of GPL)
Therefore:
- Or later – 4708 of 12,958 projected – 36%
- LGPLv3 – 30 of 1119 projected – 3% (GPL conversion divided by 6%)
- GPLv3 – 5100 of 17,540 projected – 29% (GPL conversion divided by 94%)
- GPL, not converted – 5224 projected (GPL projects times (100% - 72% convert rate))
- Non GPL license – 2033 projected (Active projects – (Active GPL projects + not converted))
All this in one month.
So, what is the summary?
Adoption and projects released under GPLv3 are as expected. The part that is often overlooked is the long standing impact of the "or later" clause, initially put into effect more than a decade ago as part of the language surrounding the GPLv2 License, or later, at the time and now, common among GPL licensed projects. This clause in effect guaranteed a wide spread "potential" of adoption of the license the day it was released.
Google Homepage
myAOL
3 comments:
The 'or later' did not anticipate that the later will not be compatible with 'this' which is the case here. As it is, you cannot mix GPLv2 and GPLv3 together so the 'or later' will stay GPLv2 (or later) and won't be transformed to GPLv3 (or later) without massive effort of complete switch of the codebase that also includes evaluation of dependencies. Otherwise, there may be a gray land approach whereby the incompatibility is ignored in fact. However, this will weaken the license if it become the common practise.
Hi Anonymous -
Despite the "stastics" being derived, my experience in monitoring open source software activity is supported by numerous third party sources, of which, I did cite a few.
If you wish to contradict any of my observations, please do so. Until then, please rest assured that everything within this blog is as verified and externally validated as much as possible. If the summary of that is referred to as the use of "stastics", then I hope that you find my stastics and the references provided to be accurate and truthful.
Ernie
Thank you for the update. I would be interested to know where the 72% estimated conversion rate came from, as well as the 50% impact rate.
I also think you meant to say that the multiplied where you said divided, and vice versa. For example, active projects were multiplied by 15% (as an estimate of the active projects ratio).
I am not the anonymous poster, but I echo his or her sentiments.
Post a Comment