🚀 go-pugleaf

In article <4550@paris.ics.uci.edu> nagel@blanche.ics.uci.edu (Mark Nagel) writes:
>[...] I would like to state that while
>the author *did* mention that you should give access to various
>applications that need to mess with resources and files, he did *not*
>make it clear how important it would be if you did not do so (for
>example, you didn't realize an application needed access until it was
>too late).  [...]

Hi, I'm sorry to hear about your problems with GateKeeper.  The reason
I didn't jump up and down in the documentation about how important failing
to grant privileges to an application could be, is that none of the
applications used by myself, or my testers, died violently when they failed
to get the privileges they needed (Microsoft Excel, under certain conditions,
was the sole exception to this rule).  For myself, I'd always run an application
once, put it through its paces as necessary, and then check the log to see
what had happened.  Usually, it would turn out that the application in
question had worked without interference from GateKeeper.  On occasions when
there were problems, I'd then grant privileges and that would be that.  I
realize that some applications could crash violently on that first test run,
and that they might even cause enough trouble in doing so to put you off 
GateKeeper entirely, but I didn't happen to run into any applications with
such serious problems.  Beyond that, there's not that much I can do about
programs that don't check error conditions - hopefully things will improve
with time.  (In bringing such problems to light, perhaps GateKeeper will 
serve another useful purpose.)

>The problem is that many applications and DA's do not check error
>conditions as thoroughly as they might (especially when such errors
>would normally be extremely rare).

True.

>When extracting an application
>from Stuffit, it is unable to change the file type to APPL, but
>proceeds anyway, etc.  Through several of these mishaps, I managed to
>trash the volume directory for the partition I was in (luckily this
>didn't happen in my root partition!).  I've removed GateKeeper for the
>time being as it is more dangerous than virii.  I'll reinstall it only
>after I manage to build a list of exactly which applications require
>access permissions.  Perhaps a such a list (of popular programs and
>standard programs) could be made part of the documentation?

I don't think GateKeeper can really be said to be as dangerous as virii -
at least you know when you have GateKeeper.  :-)  Seriously, though, it sounds
like you ran a program that was clashing with GateKeeper several times without
checking to see if GateKeeper was reporting any problems - this is asking for
trouble.  It isn't a guaranteed fix, but do use the testing cycle I mentioned
above - if I'm interpreting your account correctly, I think that cycle
could have saved you a lot of trouble.

[I encountered conflicts between StuffIt and GateKeeper as well, but even when
I ran it several times in a row, it politely refrained from crashing.] 

As to building a comprehensive list of conflicting applications, a number of
people have already sent me information concerning programs that need privs. and
consequently need to be added to the release notes.  I am happy to do this, so
please send me your list and help make GateKeeper useful to the world at large.

I'm sorry to hear about your difficulties.  Let me stress that I can help
people with GateKeeper problems more readily if they mail to me directly.
I read comp.sys.mac regularly, but I don't like the thought of missing reports
of important (or even not-so-important) problems just because I hit the 'n' 
key one too many times while reading this newsgroup.

Let me know if I can help,
----Chris Johnson
----Author of GateKeeper