Date   : Mon, 16 Jan 1984 11:17:00-PST (Monday)
From   : Tony Li <Tli@usc-eclb>
Subject: BDOS Function 37 possible bug.

    Date: 16 January 1984 04:58 EST
    From: Jerry E. Pournelle <POURNE @ MIT-MC>
    To:   Tli @ USC-ECLB
    cc:   Info-Cpm @ BRL-VGR
    Re:   BDOS Function 37 possible bug.
    Return-path: <POURNE@MIT-MC>
    Received: from MIT-MC by USC-ECLB; Mon 16 Jan 84 01:59:12-PST

    Dear Mr. Li:
       Perhaps you know best.

As I stated directly in my first comment.  I DO NOT KNOW BEST.  I DO
NOT KNOW CP/M-80 EVENT TO THE EXTENT THAT THE READERS OF THIS LIST DO.

       Me, I will not use Fn 37; I have seen too many disk
    directories trashed.

You may be right.  If indeed there is a bug, I would like to know
about it.  Bugs should be common knowledge.  When DRI finds out about
a bug, it publishes a small in-house sheet, and then considers it a
feature.

       Re: how, it is NOT triggered ONLY by writing to
    improperly opened/closed files.  There is NO (known to me)
    simple algorithm for preventing the fn 37 bug from biting you,
    even taking meticulous care.  Or so say several sources I trust.

May we talk to the sources?  I'm sure that the net would appreciate
more info on the problem.

       You are welcome to continue using fn 37 with its
    undocumented features.  My apologies for bringing up the
    subject. 
       I really ought to know better by now.

Yup, you're right.  We've argued before Jerry, and we've never gotten
anywhere but flaming.

Cheers,
Tony ;-)