Email Program

Paul McGuire ptmcg at austin.rr.com
Mon Mar 2 09:36:49 CET 2009


On Mar 2, 1:11 am, Paul Rubin <http://phr...@NOSPAM.invalid> wrote:
> Paul McGuire <pt... at austin.rr.com> writes:
> > As to your question of whether Python can be used to write an e-mail
> > client, or to create a programming language, I assure you both are
> > possible.  But also, given your unfamiliarity with Python, both are
> > well beyond your skills for some time yet, and you are nowhere near
> > reaching any of Python's limits any time soon.  
>
> Mitch Kapor (of Lotus 1-2-3 fame) spent a lot of money hiring
> very sharp Python programmers to write an email client called
> Chandler, but from what I understand, progress so far has been
> disappointing, at least in part for performance reasons.

Yipes, have you read the book ("Dreaming in Code")?  Chandler's
problems were much more organizational than technical.  The staff was
loaded with Netscape/Microsoft/Apple alumni, well-respected, but also
well-heeled, perhaps a little too comfortable and complacent with
their IPO and stock option money, not hungry enough.  And Kapor's
leadership from the top was too easy-going, not enough urgency or
drive.  The project would tread water for months at a time, waiting
for some collective eureka to resolve philosophical design questions,
while the operational burn rate burned on.  Kapor was both majority
investor and CEO/COO, which would make it difficult for Kapor the
Manager to see core problems when Kapor the Investor could just
plaster over them with more money.  But when even the added cash ($5MM
initial + $2.5MM add-on) was burned through, the basic problems still
remained and ultimately, no product.  (A partial package has been
released - for $7.5MM one would hope so!)

When starting a project, you have three priority factors: time, money,
and scope.  You have to pick the top 2, and the third has to give.
Chandler did not have any time driver, fuzzy scope, and *lots* of
money, maybe too much.  If the money had been tighter, the deadlines
and scope would have had to get a lot crisper.

Python performance may be the reason why the delivered package is
sluggish, but it is far from being the root reason for the project's
lackluster results.

-- Paul



More information about the Python-list mailing list