Re: [scheme-reports-wg2] Threads and futures Linas Vepstas (19 Mar 2022 05:54 UTC)
Re: [scheme-reports-wg2] Threads and futures Marc Nieper-Wißkirchen (19 Mar 2022 08:24 UTC)
Re: [scheme-reports-wg2] Threads and futures John Cowan (19 Mar 2022 18:57 UTC)
Re: [scheme-reports-wg2] Threads and futures Linas Vepstas (19 Mar 2022 20:04 UTC)
Re: [scheme-reports-wg2] Threads and futures Marc Nieper-Wißkirchen (19 Mar 2022 22:11 UTC)
Re: [scheme-reports-wg2] Threads and futures Linas Vepstas (20 Mar 2022 07:50 UTC)
Re: [scheme-reports-wg2] Threads and futures John Cowan (20 Mar 2022 22:34 UTC)
Re: [scheme-reports-wg2] Threads and futures Marc Feeley (21 Mar 2022 04:49 UTC)
Re: [scheme-reports-wg2] Threads and futures Marc Nieper-Wißkirchen (21 Mar 2022 06:28 UTC)
Re: [scheme-reports-wg2] Threads and futures Marc Nieper-Wißkirchen (21 Mar 2022 06:54 UTC)
Re: [scheme-reports-wg2] Threads and futures Ray Dillinger (21 Mar 2022 19:00 UTC)
Re: [scheme-reports-wg2] Threads and futures Dr. Arne Babenhauserheide (21 Mar 2022 16:54 UTC)
Re: [scheme-reports-wg2] Threads and futures Marc Nieper-Wißkirchen (21 Mar 2022 16:04 UTC)
Re: [scheme-reports-wg2] Threads and futures Taylan Kammer (23 Mar 2022 14:20 UTC)
Re: [scheme-reports-wg2] Threads and futures Marc Nieper-Wißkirchen (23 Mar 2022 14:28 UTC)
Re: [scheme-reports-wg2] Threads and futures Dr. Arne Babenhauserheide (23 Mar 2022 15:40 UTC)
Re: [scheme-reports-wg2] Threads and futures Marc Nieper-Wißkirchen (23 Mar 2022 15:34 UTC)
Re: [scheme-reports-wg2] Threads and futures Dr. Arne Babenhauserheide (23 Mar 2022 22:08 UTC)
Re: [scheme-reports-wg2] Threads and futures Ray Dillinger (01 Apr 2022 23:04 UTC)
Re: [scheme-reports-wg2] Threads and futures Per Bothner (01 Apr 2022 23:21 UTC)
Re: [scheme-reports-wg2] Threads and futures Ray Dillinger (01 Apr 2022 23:28 UTC)
Re: [scheme-reports-wg2] Threads and futures Per Bothner (01 Apr 2022 23:50 UTC)
Re: [scheme-reports-wg2] Threads and futures Ray Dillinger (02 Apr 2022 00:01 UTC)
Re: [scheme-reports-wg2] Threads and futures Per Bothner (02 Apr 2022 00:35 UTC)
Re: [scheme-reports-wg2] Threads and futures Linas Vepstas (10 Apr 2022 23:46 UTC)
Re: [scheme-reports-wg2] Threads and futures Linas Vepstas (27 Mar 2022 16:54 UTC)

Re: [scheme-reports-wg2] Threads and futures Ray Dillinger 21 Mar 2022 18:59 UTC


On 3/21/22 06:54, Marc Nieper-Wißkirchen wrote:
> Am So., 20. März 2022 um 23:34 Uhr schrieb John Cowan <xxxxxx@ccil.org>:
>
>> Because MIT decided that the comp-sci course every student takes should now be about writing glue rather than data structures and algorithms, a plausible choice in the current environment.
> Thinking that leads to such decisions is fatal in my opinion.
>

While not inclined to entirely disagree, it must be acknowledged that
the majority of programmers are now in fact in the business of writing
glue. 

I believe that "computer science" indicates that someone ought to be
able to reason about code and prove important facts about it.  They
should be able to design programs and make choices of data structures or
methodologies with an eye to efficiency, maintainability, likely future
requirements, likely future maintainers, and general ease of use.

But what are the graduates actually doing?  Most of the entry-level jobs
start with demands like "glue this database onto the back end of that
web server using the same failover and scaling software we're using in
this other project, and write some scripts to produce hit counts and
customer engagement statistics for all of these pages (in a format
suitable for auto-feeding as data into our customer engagement optimizer
software).  Also we need to trigger auto-login to all our affiliates'
sites so they can cookie-track the users across all our other
affiliates' sites, and make sure your autologin trigger works across all
seven of these browsers, defeating this blocking plugin that's become
popular on those three and taking advantage of this additional
capability of any of these three other add-ons if any of them are
installed on the four of these browsers that can install them...."

Designing programs is architecture.  It's clean.  You get to choose the
shape of all the parts.  You get to choose what problem it solves and
what solution is acceptable.  You get to decide what loads each part
will bear and how. And once code is written it always continues to do
the same thing.  But actually working in the field these days is more
like trying to build a tower where you have to use pentagonal bricks and
you have grease instead of mortar, and then you have to keep the tower
upright while the construction site has regular earthquakes and people
hang large weights off of it at weird angles all the time.  The
pentagonal bricks aren't supposed to be a problem, because, see, we've
got all these adapters and frames for the septagonal bricks two sizes
smaller that we were using last year, and you can just use those.... Oh,
and if the grease isn't sticky enough I guess we could theoretically get
some silly putty instead, but management probably won't spring for the
expense.... so just try using heavier grease, okay?

Bear