[prev in list] [next in list] [prev in thread] [next in thread] 

List:       mutt-users
Subject:    Re: request for SMTP integration (was Re: Mail using non-local SMTP server.)
From:       Jeremy Blosser <jblosser () firinn ! org>
Date:       2001-05-17 11:47:43
[Download RAW message or body]

Louis-David Mitterrand [vindex@apartia.ch] wrote:
> * On Wed, May 16, 2001 at 11:38:25AM -0600, Charles Cazabon wrote:
> > Louis-David Mitterrand <vindex@apartia.ch> wrote:
> > 
> > > > Trust me, if you open the door to basic MTA functionality,
> > > > the rest will slowly creep in ("feature creep"), bloating mutt.
> > > 
> > > Features is what we want and need. Features is why we use software in
> > > the first place. Evolve or die: software is not unlike a living being
> > > for that matter.
> > 
> > False assumptions lead to false conclusions.  Software is nothing like a
> > living creature; a piece of software I install today will continue to work
> > exactly the same, with no loss of functionality, for the rest of time.
> 
> Basic history proves you wrong.
> 
> Sure you can still use mailx, sed, cat and ed to perform your daily
> tasks but I suspect you'd get fired the minute your boss discovered
> you're losing so much time by not using the proper tools.

Erm, no, my boss pays me money to use those tools to save our department
ridiculous amounts of time.  Having small tools that do one thing and do it
well, and connecting them via pipes other forms of redirection, and being
able to mix and match these modular utilities quickly, and reuse them with
minor modifications for similar tasks over and over... this is the epitome
of efficiency.

Why the heck do you use any unix tools if you're so opposed to the core
concepts that made them what they are?

> Deal with it. Software has to do more every day. Your computer is not a
> friggin' museum where you can marvel at the quaintness of old pieces of
> codes that have stayed pure. It's a tool and it better be efficient.

Deal with it.  Tools have to do more every day.  Your toolbox is not a
friggin' museum where you can marvel at the quaintness of old tools
[hammers, screwdrivers, pliers] that have stayed pure.  It's a [single]
tool and it [singular] better be efficient.

> One only need to look at mutt's origins. Michael Elkins was frustrated
> with ELM developers' inability to evolve, so he moved on to create mutt.
> Guess who is the leading unix mailer today?

As with many of your arguments in this thread, this is apples v. oranges.
The Un*x philosophy is 'do one thing and do it well'.  Elm had ceased to do
that one thing well.  Yes, utilities should add features as they are
needed, but only those features that fit their core function.  Sed should
strive to coninually be a better stream editor.  Vi should strive to
continually be a better visual editor.  Mutt should continually strive to
be a better mail user agent.  But since ME started it Mutt has ALWAYS stuck
with doing one thing and doing it well, it has never screwed around with
stuff that belonged in other programs.  And that is one of the main reasons
WHY it is one of the leading unix mailers.  Many people who understand unix
use it explicitly for this reason, and many others use it because it just
works and sucks less, both of which are primarily true because it focuses
on doing what it does well and leaves other programs' functions to other
programs.  I'd even go so far as to say that all the selling points of Mutt
are tied either explicitly or implicitly to this philosophy.

> Software dies when its developers are taken by self-righteousness,
> sclerosis and elitism. I detect the faint stench of these symptoms on
> mutt-dev.

Elitism has been rampant on mutt-dev for about as long as it has existed.
If anything it's less present now than it ever was.  Mutt has never been
after luser mindshare, it's been after being the best damn MUA for those
that make it.  It has succeeded, and if other people find it useful, good
for them, but people shouldn't be so quick to criticize a process as
failing when it produced the very thing they seem to find useful.

> > Continual addition of features may be what _you_ desire, but it isn't
> > what anyone _needs_.  That way madness lies, as evidenced by the
> > commercial proprietary model of software which convinces users they
> > "need" to upgrade continually.  The real reason for this, of course, is
> > the producer's need for a continuous revenue stream.
> 
> A comparison with commercial software is highly irrelevant. Although I
> recently wrote a business plan and used almost _every_ MS-Word feature I
> could find and was glad they were present (I know, I need to read that
> damn LyX tutorial). 
> 
> Look at how total computing newbies can make passable PowerPoint
> presentations in very little time. Software should work that way.

Gar.  Powerpoint is the leading cause of utterly contentless meetings that
either put people to sleep or cause them to be unable to digest any
information that is more than 3 bullets with a star wipe and 2 tone
background.  Making newbies able to inflict these on their co-workers in
place of actual presentations with content that take a bit longer to
produce is not a good thing.  And yes, I sit through a zillion of these
things a month at work, I'm well aware of what they get used for.

> Let's take a car. Would you argue that it need less features? It needs
> more, if only to make the ride more comfortable, safer, more reliable.
> Yes cars are becoming more complex, heavier, but I'm glad they have all
> these features.

Complexity and features are not inherently bad, and that's not the
argument.  The argument is that features should go with the programs they
make sense for and not all just end up in one program, because that
produces lockin and inflexibility and in the end a loss of features.  Your
argument for an MTA in Mutt makes as much sense as saying that people
shouldn't have to worry about finding a road to drive their car on, their
car should just produce pavement for them wherever they want to go.  Too
bad if the road it produces is too soft or too bumpy for them, they can buy
a whole new car to fix that.  And forget about being efficient and reusing
the roads by multiple people or only producing roads where they are
actually needed.

-- 
Jeremy Blosser   |   jblosser@firinn.org   |   http://jblosser.firinn.org/
-----------------+-------------------------+------------------------------
the crises posed a question / just beneath the skin
the virtue in my veins replied / that quitters never win

[prev in list] [next in list] [prev in thread] [next in thread]