Hi all (and matthew we need your guidance/help) We really want to get pharo 1.0 license clean. It would be bad that 1.0 is not. But 1.0 should get out before summer. Here is a collection of all the information I collected on the topic: Here what matthew wrote I've done the first step and done a full audit, using Yoshiki's tools [1], of all the code in Squeak 3.10.2, with the exception of four packages: - Monticello - SUnit, TestRunner, and SUnitImproved - Universes - Traits I also applied all of yoshiki's etoys changes to 3.10.2. The algorithm that Yoshiki seemed to follow when cleaning etoys is: - If the method can be easily deleted, delete it - Otherwise, use the historical image [1] to revert to the latest version before the problematic author - Fix any bugs this causes I think we can use the same procedure. Perhaps Yoshiki will comment. So, to help with the relicense, grab the historical image[1], check out the license audit [2], leave a note on mantis about what you think you'll work on, and start rolling out change sets. For your convenience, all of the license audits are included both in the relicensing tools download and on mantis. I'll try to keep both up to date as we progress with the relicense [1] The tools I've been using to do the relicense: http://ftp.squeak.org/4.0alpha/squeak4.0-relicensingTools.zip see also: http://lists.squeakfoundation.org/pipermail/squeak-dev/2008-September/131586... http://www.squeaksource.com/MethodAuthorship.html [2] The 4.0 relicense mantis issue: http://bugs.squeak.org/view.php?id=6989 The most up-to-date list of ok contributors is at http://www.squeaksource.com/311/MethodAuthorship-kph.17.mcz See the class-side methods in Authorship, category data here is another email from matthew On Wed, Apr 22, 2009 at 11:10:12AM +0200, stephane ducasse wrote:
I need your guidance for the MIT conversion effort of Pharo Do you have some cycles to help me and telling me how to proceed? I'm on holidays (so working remotely so I may allocate some time on that).
I can help you after school gets out in a week. I need to do a quick count of how many lines of code need to be fixed in squeak for submission to SFLC. I can do the same for Pharo. I don't know how strenuous you want to be in your license conversion. The next 3 steps in the license conversion process are, in no particular order: - get a document from yoshiki documenting how etoys was relicensed (and, by extension, cuis). The board will overview it with the help of SFLC and determine if it is good enough. - The board will publish a statement about what is and is not a trivial-enough change to ignore. - I will count up the lines of code to be converted, and determine how many of them are trivial and non-trivial You should chat with Randal Schwartz about what you need to decide before you can start relicensing, and what you need to document in case your relicense needs to be reviewed in the future. Yoshiki and Juan did not do this and thus we have no evidence yet that Yoshiki's relicense is strong enough. I documented the priorities for squeak 4.0 in this email: http://lists.squeakfoundation.org/pipermail/squeak-dev/2009-February/134083.... You should probably figure out your priorities for the relicense as well, so you know whose standards you need to meet. Since that email was written, Randal has indeed heard back from SFLC, but I haven't heard what they had to say yet. -- Matthew Fulmer -- http://mtfulmer.wordpress.com/