[LMB] OT: Afterlife
marc.wilson at gmx.co.uk
Mon Nov 2 17:23:41 GMT 2020
On Wed, 28 Oct 2020 15:58:32 -0400, WalterStuartBushell
<proto at panix.com> wrote:
>> On Oct 28, 2020, at 2:36 PM, Zan Lynx <zlynx at acm.org> wrote:
>> This happens in computer programming and debugging all the time, which is my experience. One programmer insists that it cannot happen that way. Which is fine until there's EVIDENCE. At that point continuing to insist is stupid. After enough investigation there's a reason. Even if it seems impossible, it still happened.
>This happened to me. My program went from point A to C without going through B which was necessary by
>program logic. I overflew an array and overwrote an instruction with data that happened to be a jump instruction.
>Actually a subroutine call instruction which stored the return address in the first location of the subroutine.
>That practice in considered harmful today.
>This could not happen in OS that (properly) separated pure code (unwritable) during execution) and data changeable.
>When the pure code is swapped out it doesn’t need to be written out, but the data segments do.
>And several different process can share the same pure code segments.
I'm having exactly this discussion with one of my email suppliers at the
moment. They insist everything is working, and nothing has changed.
Yet my mail client (which I've had for almost 3 decades) cannot retrieve
POP email from their host. It works just fine with several others, and
I can access the mail via webmail or IMAP.
I think it may be time to migrate to Thunderbird.
Now I need to work out how to get all my old emails from Agent into
Thunderbird - that or keep Agent as an "archive", but I'd rather have it
all in one place. To which end, I've managed to duplicate my mail email
address, so that I have both IMAP and POP connections for the same
account (with the POP one set to "leave mail on server") so I can have a
"hard" backup and online access from everywhere.
Humanity without religion is like a serial killer without a chainsaw. -- unknown
More information about the Lois-Bujold