Path: csiph.com!fu-berlin.de!uni-berlin.de!individual.net!not-for-mail
From: rbowman <bowman@montana.com>
Newsgroups: comp.os.linux.setup,comp.os.linux.advocacy,sci.physics
Subject: Re: Linux Crashing
Date: Fri, 1 Oct 2021 20:13:34 -0600
Lines: 23
Message-ID: <irptieF29jqU2@mid.individual.net>
References: <siv07k$us6$3@dont-email.me> <sivg6r$hnl$1@solani.org> <sivn25$mod$3@dont-email.me> <sivvf9$cv7$1@solani.org> <iri694Fi52qU1@mid.individual.net> <YER4J.28697$PE4.10580@fx11.iad> <iri9rbFipbaU1@mid.individual.net> <sj12s2$34c$1@solani.org> <ohj8lg9d377vkgc97rmbllik4rosvrprov@4ax.com> <irja8aFonjiU1@mid.individual.net> <8j29lg5kteuqhn9o9ophsgfp7afi7o2894@4ax.com> <sj2655$ipg$1@solani.org> <68e9lg1q6m6phvi6m7mibhljvib8gdt1hq@4ax.com> <sj2eqn$hbc$1@dont-email.me> <ier9lg9e0eute2d2ol0qt0n7clg59d4m92@4ax.com> <irknihF2empU1@mid.individual.net> <sj45hl$spr$2@dont-email.me> <irlujfF9jncU1@mid.individual.net> <sj52se$clv$1@dont-email.me> <2NH5J.151946$Kv2.4408@fx47.iad>
Mime-Version: 1.0
Content-Type: text/plain; charset=windows-1252; format=flowed
Content-Transfer-Encoding: 7bit
X-Trace: individual.net dICQMEwfFTrc38hX0eA7GwCbxmVEVS+MoXH2NihBpl7mKBB3Lz
Cancel-Lock: sha1:seeUOLSzq1rAZxiZUCHLBiKuofk=
User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:45.0) Gecko/20100101 Thunderbird/45.6.0
In-Reply-To: <2NH5J.151946$Kv2.4408@fx47.iad>
Xref: csiph.com comp.os.linux.setup:4552 comp.os.linux.advocacy:595024 sci.physics:833451

On 10/01/2021 11:43 AM, Charlie Gibbs wrote:
> On 2021-09-30, The Natural Philosopher <tnp@invalid.invalid> wrote:
>
>> On 30/09/2021 15:06, rbowman wrote:
>>
>>> They are about 20 years old, usually work, and are passed on through the
>>> generations because nobody understands what they're doing and are
>>> reluctant to touch them. A lot of legacy Perl is the same.
>> I think the apocryphal story is Gary Kildall on work experience as a
>> student, asked to fix some issues on multiple *books* of assembler, said
>> wouldn't it be easier to rewrite it in FORTRAN'? Too slow they said, but
>> he did, and it ran *faster*.
> A good algorithm written in an inefficient language will run
> rings around a poor algorithm written in assembly language.
>
> The first thing I do when I take over an existing program is
> to figure out what it does.  Then I can throw out the cruft.
> I can usually cut the line count by 30%.

Six weeks later you find the edge case that was handled by a line of 
cruft...