On Tue, 3 May 2011 11:38:21 -0400, Shmuel Metz (Seymour J.) wrote: >In <listserv%[email protected]>, on 05/03/2011 > at 09:35 AM, Paul Gilmartin <[email protected]> said: > >>I've long wondered, if sequence numbers are so valuable, why haven't >>they spread outside the progeny of unit record systems? > >Because all source code is stored on the progeny of unit-record >equipment. They've spread to tape and disk; what more do you want? > Tunnel vision.
I don't want them. But I was referring to OSes that haven't the leg^H^H^H heritage of a unit record world: Linux. OS X. Solaris. And there's even another inexplicably popular OS out there. >Now, if you want to complain that simple integers are too simplistic >and that sequence numbers should be tree structured, you'll get no >argument from me. > CDC did this, sort of. They had 90-column source records with PTF-ID.LINE-NUMBER in 73-90. Of course, they were no longer sortable. And they used an update protocol rather than the downdate mentioned by Lynn. Downdate has the advantage that the source is instantly accessible without the need to run a filter. And the disadvantage that if your current instance is ever modified outside the downdate delta protocol you're really SOL. -- gil ---------------------------------------------------------------------- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [email protected] with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
