On Fri, 2002-04-19 at 10:49, Dan Winship wrote: > On Fri, 2002-04-19 at 10:38, Jason Kohles wrote: > > On Fri, 2002-04-19 at 10:19, Dan Winship wrote: > > > > 4. When "nuke-a-sig" is run, start at the end of the message, and only > > > > read back <n> (user configurable?) lines to get to the delimiter. > > > > > > There's no reason to make it user configurable. And I'd suggest "5" for > > > n. > > > > > That's a good reason to make it configurable, I'd argue that the bigger > > the sig, the more it needs stripping. > > Then that's a lousy reason to make it configurable. Having a > configuration option means "we think some people only want to strip > small signatures, but other people want to strip large signatures too, > so you can decide". > Well, if you want to only strip sigs shorter than 5 lines, and I want to strip sigs larger than that, then apparently "some people only want to strip small signatures, but other people want to strip large signatures too" is true.
> What you're arguing for is "it shouldn't be based on the number of > lines, you just need to have a smarter algorithm" > No, I'm arguing that if it is going to be based on the number of lines, then the number of lines should be configurable. -- Jason Kohles [EMAIL PROTECTED] Senior System Architect Red Hat Professional Consulting http://www.redhat.com/ _______________________________________________ evolution maillist - [EMAIL PROTECTED] http://lists.ximian.com/mailman/listinfo/evolution
