On Saturday 15 January 2005 12:41, foser wrote: > You define it quite correct : 'optional functionality from the > _package_' . This is no option defined in the source.
Why does "package" have to equal "source"? Why can't package equal "Gentoo package"? And why do you feel the need to make your definition so ... tight? Restrictive? Unimaginiative? Beaurocratic? What benefit do our users get from you choosing to split this particular hair? > Would you rather have 3k of USE flags you can't deal with to begin with. You know, I actually don't find it difficult to deal with the ~300 global USE flags and ~830 local USE flags that we currently have. All I do is emerge -uav <package> and look at the USE flags listed in the output. *This* is where I actually deal with USE flags, and this way means that I only have to deal with a few at a time. This approach makes the number of USE flags in Portage largly irrelevant. Instead, it's the number of flags that packages support which becomes important. Apart from the odd anomoly (like PHP), the use flags / package ratio is very small, and should be within the capacity of most people to cope with. I think there's some value into making USE flags of the form <category>/<flag>, to make it easier to find (for example) all flags which would make a system more secure, or all the flags which activate support for different types of database. I believe that we could move over to a scheme like that without requiring any code changes to Portage itself. And, it has the advantage of making it possible to add more and more USE flags to Portage ;-) > Increasing USE flags to no end is at least as much choice inhibiting as > restricting some choice. USE flags are a strong point of Gentoo, but > it's actually getting weaker and weaker all the time. You obviously have a problem, and I have to respect that. But it's not a problem that I understand or encounter for myself. > You misunderstand the goals of this 'camp' I think. I think the 'camp' consistently fails to present any positive suggestions about what can be done to address the issues that you have. You're fighting a battle with no clearly definied objective, no clear strategy for achieving the objective, and - most importantly - not a single line of code to change Portage's behaviour. It's no wonder you're not making any progress on this. Instead of saying that John shouldn't have added this automated capability via USE flags, you'd get a lot further (and, frankly, be more respected) if you suggested how else he *could* have done it. And even further if you coughed up some code to make it possible. At the moment, USE flags are the only per-package mechanism available to users to indicate their choices. Maybe we need per-package FEATURES? That would seem a more appropriate place for John's symlink support? > The line of thought > is more that USE flags are getting abused for things they weren't meant > for. I personally believe this particular one is one of them. The world hasn't ended just because John added a USE flag. Best regards, Stu -- Stuart Herbert [EMAIL PROTECTED] Gentoo Developer http://www.gentoo.org/ http://stu.gnqs.org/diary/ GnuPG key id# F9AFC57C available from http://pgp.mit.edu Key fingerprint = 31FB 50D4 1F88 E227 F319 C549 0C2F 80BA F9AF C57C -- -- [email protected] mailing list
