Re: Policy about packages that have Tk dependent parts.

1997-12-01 Thread Christian Schwarz
On 30 Nov 1997, Rob Browning wrote: > Philippe Troin <[EMAIL PROTECTED]> writes: > > > > Philippe Troin <[EMAIL PROTECTED]> writes: > > > > Two packages instead of one. We have 1400+ packages in main (around 2000 > > > > including non-free and contrib), and I think we should apply care when > > >

Re: Policy about packages that have Tk dependent parts.

1997-12-01 Thread Rob Browning
Philippe Troin <[EMAIL PROTECTED]> writes: > > Philippe Troin <[EMAIL PROTECTED]> writes: > > > Two packages instead of one. We have 1400+ packages in main (around 2000 > > > including non-free and contrib), and I think we should apply care when > > > creating new packages especially when the foo-

Re: Policy about packages that have Tk dependent parts.

1997-11-30 Thread Philippe Troin
On 30 Nov 1997 18:28:23 EST Ben Pfaff ([EMAIL PROTECTED]) wrote: > Philippe Troin <[EMAIL PROTECTED]> writes: > > Two packages instead of one. We have 1400+ packages in main (around 2000 > > including non-free and contrib), and I think we should apply care when > > creating new packages especiall

Re: Policy about packages that have Tk dependent parts.

1997-11-30 Thread Ben Pfaff
Philippe Troin <[EMAIL PROTECTED]> writes: > Two packages instead of one. We have 1400+ packages in main (around 2000 > including non-free and contrib), and I think we should apply care when > creating new packages especially when the foo-tk package is likely to be > a few kilobytes. Why is it a b

Re: Policy about packages that have Tk dependent parts.

1997-11-30 Thread Philippe Troin
On 30 Nov 1997 13:06:00 CST Rob Browning ([EMAIL PROTECTED]) wrote: > Ben Pfaff <[EMAIL PROTECTED]> writes: > > > Rob Browning <[EMAIL PROTECTED]> writes: > > > 1) Create two packages: foo and foo-tk > > > > I can't see any real downsides to this. If `foo' suggests `foo-tk' > > (and `foo-tk'

Re: Policy about packages that have Tk dependent parts.

1997-11-30 Thread Rob Browning
Ben Pfaff <[EMAIL PROTECTED]> writes: > Rob Browning <[EMAIL PROTECTED]> writes: > > 1) Create two packages: foo and foo-tk > > I can't see any real downsides to this. If `foo' suggests `foo-tk' > (and `foo-tk' depends on `foo'), then we've got the best of both > worlds. OK, this seems to be

Re: Policy about packages that have Tk dependent parts.

1997-11-30 Thread Ben Pfaff
Rob Browning <[EMAIL PROTECTED]> writes: > 1) Create two packages: foo and foo-tk I can't see any real downsides to this. If `foo' suggests `foo-tk' (and `foo-tk' depends on `foo'), then we've got the best of both worlds. -- Ben Pfaff <[EMAIL PROTECTED]> <[EMAIL PROTECTED]> <[EMAIL PROTECTED]>

Re: Policy about packages that have Tk dependent parts.

1997-11-30 Thread Bdale Garbee
In article <[EMAIL PROTECTED]> you wrote: : 2) Create a single package suggests Tk rather than depending on or : recommending it I like this option, along with some text in the long description field of the control file explaining that one or more of the utilities use Tk, but that Tk is no

Policy about packages that have Tk dependent parts.

1997-11-30 Thread Rob Browning
I'm now maintaining two packages, hfsutils and sgrep, each of which includes at least one tool that's Tk based. In both cases, bugs have been reported against the packages by people who don't think they should depend on Tk. This is not completely unreasonable, but I wanted to discuss how we shou