home *** CD-ROM | disk | FTP | other *** search
- Newsgroups: comp.dcom.modems
- Path: sparky!uunet!cs.utexas.edu!sdd.hp.com!ux1.cso.uiuc.edu!uchinews!machine!chinet!les
- From: les@chinet.chi.il.us (Leslie Mikesell)
- Subject: Re: Boom! You're Dead.
- Message-ID: <1992Jul24.031631.26450@chinet.chi.il.us>
- Date: Fri, 24 Jul 1992 03:16:31 GMT
- References: <2043@tymix.tymnet.com> <1992Jul18.142320.4269@qiclab.scn>
- Organization: Chinet - Public Access UNIX
- Lines: 44
-
- In article <1992Jul18.142320.4269@qiclab.scn> leonard@qiclab.scn.rain.com writes:
- >tnixon@hayes.com writes:
- >>Of course, this means that there is
- >>a fixed probability that your modem will inadvertently escape when
- >>you're transmitting random data (such as a compressed file) or any
- >>file that happens to mention the given character sequence
-
- Adding the guard time lowers the probability but doesn't eliminate
- it. And, as many of us have experienced, when using such modems
- on the host side of a connection, they will drop in response to
- the echoed characters regardless of the time delay so it doesn't
- help a bit there.
-
- >Well, aside from the bit about "Hayes Improved Escape Sequence" with
- >"Guard Time" as a description for a feature that is over a decade old,
- >it *does* sound like somebody was asleep at the switch.
-
- Yes, maybe back in 1980 someone actually had a reason to put the
- modem in command mode in the middle of a connection. Maybe they
- were using software that didn't offer the ability to drop DTR.
-
- >They actually *don't check* for anything but the tripled character?
- >*Unbelievable*!
- >
- >BTW, you seem to imply that the makers of these brain-dead modems are
- >actually pushing this bug as a *feature*? Is that true?
-
- It would be a feature for a modem to have *no* default in-band escape
- to command mode these days (actually it would have always been better
- that way). On the rare occasions when you felt the urge to be able
- to spuriously interrupt your communications (perhaps you are debugging
- and want to query the registers) you could always send the enabling
- command explicitly before dialing. It's hard to believe that no one
- has gone this route as the simple solution.
-
- Now that that's settled, does anyone want to talk about modems that
- refuse to stop dialing once they've started even though you drop DTR?
- (This is with &D3 set - they drop a completed connection on DTR loss).
- And then they insist on sending back the result code, even though
- the program that was trying to dial has given up and gone away. Does
- anyone have a list of modems that don't act this way?
-
- Les Mikesell
- les@chinet.chi.il.us
-