comp.lang.ada
 help / color / mirror / Atom feed
From: kevin  cline <kevin.cline@gmail.com>
Subject: Re: Choosing a new language
Date: Mon, 7 Jan 2008 09:25:16 -0800 (PST)
Date: 2008-01-07T09:25:16-08:00	[thread overview]
Message-ID: <b5b7d63b-1af2-440f-80ac-d520b1301083@d4g2000prg.googlegroups.com> (raw)
In-Reply-To: 1m3pn3pntfj99vd85gedvplrlonv96p1ea@4ax.com

On Jan 3, 1:32 am, Tim Roberts <t...@probo.com> wrote:
> kevin cline<kevin.cl...@gmail.com> wrote:
>
>
>
> >As if there were such a thing as an 'Ada programmer'.  Any decent
> >programmer should be productive in Ada long before their security
> >clearance is approved.
>
> That's only true because the security clearance process has become so
> complicated.  Ada is not a trivial language by any means.  Even an
> experienced C programmer is going to find enough sharp edges to send him
> back to the reference manuals on a regular basis.
>
> >The real problem the DoD has is that defense work is not attractive to
> >the best and brightest.
>
> Bull crap.  You don't HEAR about them because of that same security
> clearance issue, but some of the most complicated and certainly some of the
> LARGEST computing systems in the world come out of the DoD.  You don't
> create reliable large systems using a corral full of bright-eyed college
> new hires.

I didn't say anything about what the DoD built, or attempted to
build.  I meant that
the most talented young programmers find companies like Google and
Amazon or other
startups more attractive than defense work.  I worked at a Defense
software startup
in Dallas for ten years.  I know how it worked.  Organizations like
Texas Instruments D-Seg hired a lot of new graduates, mostly from
second-tier midwestern
public schools, and put them to work writing defense systems.

With cost-plus contracting, companies bill the DoD by the hour, making
a fixed fee for
each hour charged.  As long as a programmer has the necessary
credentials, their productivity
makes no difference to the company's income.  Once the contract has
been won, labor saving
suggestions have no value.  I know of at least one case where a very
talented programmer
realized that several man-years of manual effort could easily be
automated, but his suggestion was
rejected because it would have left a dozen cut-and-paste programmers
with no work.

With that sort of grind-it-out project management, talented people who
came
to Dallas to work for TI or E-Systems didn't tend to stay in defense
very long.
Many were cherry-picked by the growing telecomm industry, where a
talented developer
could make a huge difference to the bottom line.



  reply	other threads:[~2008-01-07 17:25 UTC|newest]

Thread overview: 27+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2007-12-28 15:23 Choosing a new language Rico Secada
2007-12-28 17:15 ` Joachim Durchholz
2007-12-28 17:18   ` smallpond
2007-12-28 20:54 ` John Nagle
2007-12-28 22:57   ` George Neuner
2007-12-29  9:30     ` Joachim Durchholz
2007-12-29  9:37       ` Paul Rubin
2007-12-29 18:16         ` Joachim Durchholz
2007-12-29 18:22           ` John Thingstad
2007-12-30 11:41             ` Joachim Durchholz
2007-12-29 13:41     ` Stephen Leake
2007-12-28 22:49 ` Gary Scott
2007-12-29  5:29 ` george.priv
2007-12-29  6:07 ` byte8bits
2007-12-29 22:56   ` Samuel Tardieu
2007-12-30  2:07     ` Rico Secada
2007-12-30  9:02       ` Pascal Obry
2007-12-30 20:52       ` Jacob Sparre Andersen
2007-12-31  4:38   ` George Neuner
2008-01-02 19:36     ` kevin cline
2008-01-03  7:32       ` Tim Roberts
2008-01-07 17:25         ` kevin cline [this message]
2007-12-29 12:35 ` bearophileHUGS
2007-12-29 15:11 ` Achim Schneider
2007-12-29 17:58   ` Arnaud Delobelle
2007-12-29 18:39     ` John Thingstad
2007-12-29 15:40 ` j.khaldi
replies disabled

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox