Am Montag, 17.03.03 um 23:04 Uhr schrieb David R. Morrison:
Hi Max.Yeah I understood that was your intention.
Well, my idea is that we do something fairly sensible, but then offer the
user the chance to see the file we changed and edit it some more if they
want.
This non-CLI setup is mainly intended for people who haven't used the command line much and so will likely only have the default setup.
And I agree with that goal and support it :-) But it's especially those users for which the tool has to be "fool proof" (and no I do not want to imply that somebody w/o unix experience is a fool, that'd be far from me, unix *is* arcane to a high degree). That's why I mentioned my concerns
More advanced users will have to modify the default behavior to fit
their circumstances (as they do now, of course, with "manual" instructions).
Yup. Furthermore, more advanced users know how to modify their default behavior, or how to cope with problems. It's not them I fear for, nor them why I wrote what I wrote....
If you are trying to get to the angle that the people who will want to use our tool will not have a complex .tcshrc / .cshrc - then I will disagree. Yes, 99% of them won't have one, but I am willing to bet high amounts of money that there are those who have a complex .cshrc. E.g. because a friend installed it for them, or maybe some 3rd party tool we don't know created it, etc. And it's this 1% which will run into problems. Dealing with border cases like this is what takes most of the time in software development.
I.e. I could write down a simple Cocoa app that does what you suggested in a couple hours, but doing one which is fail proof in 99% (or ideally 100%) of all cases, and which has a highly intuitive and appealing user interface, is a much harder task.
One thing which Ben Reed and I were talking about in this regard is that
it would be good to put in a little check in /sw/bin/init.csh to stop
it from getting loaded twice (set a FINKINIT environment variable or
something like that).
Yes that would be a good idea, and might be a good solution for this whole tool, and some other issues. Since I see no way to reliably find out if /sw/bin/init.csh is being loaded, this would nicely solve the problem, I guess. We still would want to do any checks we can in order to avoid having 20 "source" lines in the .cshrc files of our users :-)
Max
-------------------------------------------------------
This SF.net email is sponsored by:Crypto Challenge is now open! Get cracking and register here for some mind boggling fun and the chance of winning an Apple iPod:
http://ads.sourceforge.net/cgi-bin/redirect.pl?thaw0031en
_______________________________________________
Fink-devel mailing list
[EMAIL PROTECTED]
https://lists.sourceforge.net/lists/listinfo/fink-devel
