A practice at places like Eclipse (at least before Git/Gerrit, not sure, what hurdles exist for DM to move e.g. "data" to Git, maybe officially leaving the Incubator?) used in JIRA/Bugzilla is attaching a patch. We may encourage that, even for very large files like DeviceData.xml, a patch should only contain diffs and be relatively small.
@Bertrand/Radu, do you know, if Apache Git also includes Gerrit at least for projects that want to use it? Werner On Mon, Aug 4, 2014 at 11:16 PM, Werner Keil <[email protected]> wrote: > There are some JIRA tickets similar to "Pull Requests" on GitHub for ODDR, > but even with a larger number of contributors, this may be a bit tedious. > > I've pointed that out a few times, so fully agree here, a more > "interactive" way would be better. > > Wikipedia once offered to "crawl" and collect UA data, not sure, if that > ever came and where it is now? > IMHO (not being a lawyer, but quite exposed to licensing stuff not just > via ODDR but even more around JCP, Eclipse, etc.) places that simply gather > User Agents without doing any processing (I believe there was at least one > site even offering some Web API for that) seem fine to use as sources. We > must not use any actual repository neither WURFL (beyond the point Bertrand > captured, but its structure is worthless now anyway and not compatible with > anything other than its propritary API) nor DeviceAtlas or others. > > The initial "Open_DB" / OMA sources were OK, but I don't believe they are > still actively maintained either. > > So we can > > - Gather raw User Agent Data and process it into the right format via > some crawler/service > - Offer a Web UI and/or Service for people to contribute that way. As > long as they somehow confirm to own the data (either paper or online) that > seems OK, it may be just the class of contributor that uses JIRA here. If > they use a service that contributes THEIR User Agent and interprets it, > then that seems fine, we may only need some security, Capcha, etc. to avoid > abuse. > > WDYT? > > On Mon, Aug 4, 2014 at 11:07 PM, eberhard speer jr. <[email protected]> > wrote: > >> -----BEGIN PGP SIGNED MESSAGE----- >> Hash: SHA1 >> >> hi, >> >> I really don't think JIRA is the place for that. >> May have an alternative 'soon' if you don't mind that it's a .net site >> running here. >> But, JIRA is not good. >> >> With regard to the sources, those may prove of some value, I was >> thinking as more primary sources. >> >> I think we require a more permanent 'database' and build relations >> with various sources. >> >> esjr >> -----BEGIN PGP SIGNATURE----- >> Version: GnuPG v2.0.22 (MingW32) >> Comment: Using GnuPG with Thunderbird - http://www.enigmail.net/ >> >> iQEcBAEBAgAGBQJT3/YKAAoJEOxywXcFLKYc0nkH/jOdFTFs8maUR59v9wh1uVnM >> npWnivmUSkGByq0NaLT0TCda4wjwJXmO7veSjQCIgiR0gVbYe0+xcnLIb7G2Iaoi >> Vz9PftnWTWPNvIIL4JbQIw0gsk0+7046yNVuDS0l8dSJj+Tk1I9TxLt/06zF6NfD >> vkz8cP05p+IP+w5ecNWdwge66ZaCebG2KypUDzPIB8YoQpqKM18+QDlmBrPIvrJC >> McHCzQ5HHX+cJQCh7jA+Sd1ok/Zf/ckNxouV7Vr+l+JmvHNBXWbDfq1/f24Lbitt >> Qem7LHiirSqLQ4WJua+fYq18esNqJ364nBZMh1jszmBKZxEbnBBwNws5Nbb/HfU= >> =KILw >> -----END PGP SIGNATURE----- >> > >
