On Thu, Aug 26, 2010 at 8:58 PM, Adam Williamson <awill...@redhat.com> wrote:
> On Fri, 2010-08-27 at 01:14 +0530, Rahul Sundaram wrote:
>> On 08/27/2010 12:20 AM, Adam Williamson wrote:
>> > That wasn't the question. The question was what is the benefit of not
>> > having one. Is it simply that it saves 1.6MB of disk space? If so, uh,
>> > woop?
>>
>> I think, that reverses the responsibility.  If anything is installed by
>> default,  *that* needs a very good justification.  For one thing, it
>> isn't just about space,  I don't want any services running on my system
>> that I don't need and I don't want to take care of updates including
>> security fixes for those software either.
>
> I think that makes sense if we're talking about adding a default, but
> taking one out - especially something that's been default in all Unix-y
> OSes for ever - is a different case.

Does anyone have any stats on what fraction of machines have no need
for an MTA currently? If there were such data it would be a useful
basis on which to make a decision about including or not including an
MTA by default. Certainly I have a need for sendmail (as my choice of
MTA) on every machine I run - partly for sending logs, and also
because I run dovecot and sendmail as a local imap mail system
combination which acts as the basis for mail storage and sending - on
systems where mail is pulled from a pop server this means that the
mail is then stored in a nice format on the local machine and can then
be grabbed from my other machines without further reference to the
external pop server. Similar is true for mail grabbed from an external
imap server.

I wonder how many people would find that their usage concerning mail
would be curtailed somewhat without an MTA?  Or put another way I
wonder what fraction of users would include "yum install sendmail" (or
equivalent) as one of the first actions after an install?

I am with Adam W on this one.

-- 
mike c
-- 
devel mailing list
devel@lists.fedoraproject.org
https://admin.fedoraproject.org/mailman/listinfo/devel

Reply via email to