> It was announced that these changes will happen on the way to 3.0 > early enough. All 2.9X releases prepare 3.0, that is what the > announcement mail says.
It is not about what Eric Raymond says, it is about what best practice is in the open source world. And this is not. And I maintain that it was neither early enough nor properly announced. > Also - as a developer - I think you should > read the -dev list instead... No, I don't read the GTK+, Cairo, sqlite, libcurl dev lists either. And I don't have to because they adhere to well established best practices. > The gpsd manpage says since a year: Hehe, that's a good one. I will right tomorrow start to re-read all man-pages :) > It was a bad idea not to use libgps even before that was mentioned > there. The protocol is around since 15 years, the begining of gpsd and long before Eric Raymond got his fingers into it. Anyway, I would be interested in the rationale of why the protocol is bad apart from the fact that Eric is an uber-coder who wants me to use his tangle-fangled lib. At least the success of TCP/IP and HTTP tells me that protocols are a fantastic thing. > I know you're whining constantly that the library was buggy, Hehe, no. I didn't whine, I just kicked that crappy pile of junk out of the code of tangoGPS. And it is certainly not going to return. > > See bug #573848 for a bug filed against tangoGPS. > > > > Suggested solution: revert to 2.90 and make a new version 3.0 with > > the new API. > > No chance. Too many changes, too many bugfixes. This is again amusing, as Eric boasts that in 20.000 lines of code were only 2 bugs. Anyway, feel free to close the bug if you haven't yet done. I'll remove gpsd as position supplier in one of the next releases. -- To UNSUBSCRIBE, email to [email protected] with a subject of "unsubscribe". Trouble? Contact [email protected]
