gnu-arch-users
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [Gnu-arch-users] Re: Linus


From: Miles Bader
Subject: Re: [Gnu-arch-users] Re: Linus
Date: Sun, 12 Oct 2003 23:01:02 -0400
User-agent: Mutt/1.3.28i

On Sun, Oct 12, 2003 at 06:10:05PM -0800, Ethan Benson wrote:
> you must make sure the umask gets set such that new dirs get the same
> permissions as the parent, or else the last person to commit will be
> the only one able to commit in the future (since others would be
> unable to move the +revision-lock dir). 
> 
> it can be done with tla help, but it would be less hassle if tla
> helped. but tom has declared it won't, so we have to deal with that.
> 
> and its NOT just as easy as setting the umask globally, that only
> works if ALL your archives/categories/branches/versions use the same
> set of permissions, if they don't you have to come with some method
> that figures out what part of the archive is going to be modified and
> set the umask to match the permissions used there.

`Setting the umask globally' won't satisfy _all_ possible access control
scenarios , but I'll bet it satisfies the majority of those people actually
want.

In particular, I think most free software projects would be fine with
allowing global anonymous read access, as long as its easy to control write
access for some branches.  You could do this by using the group field (which
is inherited from the parent directory if the g+s bit is set) to control
write access, and always making everything world readable, i.e. umask 002.

-Miles
-- 
We are all lying in the gutter, but some of us are looking at the stars.
-Oscar Wilde




reply via email to

[Prev in Thread] Current Thread [Next in Thread]