Minutes Sep 7 2011
Attendees: Jeff Licquia (LF), Robert Schweikert (SUSE)
- Retry from last week: LSB 5.0 and SDK.
Robert: FHS 3.0 feedback?
Jeff: Mostly editorial, no recent stuff. Just posted a blog on linux.com, which hopefully will generate any more needed interest. Only content objection was along the lines of "you're not doing /run, are you?"
Jeff: Running the numbers with the fake uplifts. Some numbers make sense; some really don't. For example: 257 headers in GTK+, 1125 interfaces, 1064 types. Could be man-years with those numbers to do the whole GTK+ uplift (including GDK, GObject, Cairo, etc.). Also, a fake import of D-Bus returned zero interfaces.
Robert: Yeah, the numbers don't sound right. It might make more sense to just do the GTK+ set of uplifts, and then use that as the basis for estimating the time.
Jeff: Could be. Will put another week into trying to understand what I'm doing, but without any better understanding next week, that might be the best way forward.
Jeff: SDK is also something to update. Some obvious big bugs.
Robert: Understand the need for the SDK, but it seems too complex and fragile.
Jeff: A lot of that comes from the interaction of libbat and the SDK. May be a good time to make libbat a first-class member of the SDK.