<div dir="ltr">Related to the sort discussion, there's an oral history of Duane Whitlow, founder of SyncSort, which was a big deal in IBM shops in the 70s. (and perhaps later; I lost track)<div><a href="https://archive.computerhistory.org/resources/access/text/2013/05/102702251-05-01-acc.pdf">https://archive.computerhistory.org/resources/access/text/2013/05/102702251-05-01-acc.pdf</a></div></div><br><div class="gmail_quote gmail_quote_container"><div dir="ltr" class="gmail_attr">On Sat, Jan 18, 2025 at 8:00 AM Bakul Shah via TUHS <<a href="mailto:tuhs@tuhs.org">tuhs@tuhs.org</a>> wrote:<br></div><blockquote class="gmail_quote" style="margin:0px 0px 0px 0.8ex;border-left:1px solid rgb(204,204,204);padding-left:1ex">On Jan 18, 2025, at 7:16 AM, Larry McVoy <<a href="mailto:lm@mcvoy.com" target="_blank">lm@mcvoy.com</a>> wrote:<br>
> <br>
> On Sat, Jan 18, 2025 at 04:51:15PM +0200, Diomidis Spinellis wrote:<br>
>> I'm sure the mainframe sort programs did some pretty amazing things and<br>
>> could run circles around the puny 830 line Unix Seventh Edition sort<br>
>> program. The 215 page IBM DOS VS sort documentation that John Levine posted<br>
>> here is particularly impressive. But I can't stop thinking that, in common<br>
>> with the mainframes these programs were running on, they represent a mindset<br>
>> that has been surpassed by superior ideas.<br>
> <br>
> I disagree. Go back and read the reply where someone was talking about<br>
> sorting datasets that spanned multiple tapes, each of which was much<br>
> larger than local disk. sort(1) can't begin to think about handling<br>
> something like that.<br>
> <br>
> I have a lot of respect for how Unix does things, if the problem fits<br>
> then the Unix answer is more simple, more flexible, it's better. If<br>
> the problem doesn't fit, the Unix answer is awful.<br>
> <br>
> cmd < data | cmd2 | cmd3<br>
> <br>
> is a LOT of data copying. A custom answer that did all of that in<br>
> one address space is a lot more efficient but also a lot more special<br>
> purpose. Unix wins on flexibility and simplicity, special purpose<br>
> wins on performance.<br>
<br>
Mainframes had usage based pricing, not unlike what you pay for renting<br>
resources in the cloud, so performance really mattered. Also note that<br>
users use whatever computing resources they have available to get their<br>
job done, ideally at the lowest cost. Elegance of any OS architecture<br>
is secondary, if that.<br>
<br>
</blockquote></div>