NetNews Usenet Archive 1993 #3/NN_1993_3.iso/spool/comp/dcom/modems/19959

Text File | 1993-01-25 | 5.4 KB | 118 lines

Path: sparky!uunet!valinor.mythical.com!n5ial!jim
From: [e-mail] (Jim Graham)
Newsgroups: comp.dcom.modems
Subject: Bad Experience Seeking 2400 bps, v.42bis
Message-ID: <[e-mail]>
Date: Sun, 24 Jan 93 18:35:42 GMT
References: <[e-mail]>
Distribution: world
Organization: Me? Organized? Hah! :-)
Lines: 106

In article <[e-mail]>
[e-mail] writes:

> When a user buys a modem labeled "MNP/V.42/V.42bis" today he assumes a
> traditional design. I think the RPI approach would be considered
> inferior and modem vendors seem to share this opinion.

it is inferior.  there's no real way to say it isn't....  after all,
if I want to use my USR DS on another computer, I just unplug the modem
from one, plug it into the other, and I'm set.  with RPI-based modems,
I have to transfer all of the software for the modem, too.  this could,
of course, mean porting if I don't happen to be transporting to a
computer/OS combination already provided with the modem.

> You won't
> anywhere see a RPI modem advertised as needing additional driver
> software, at most this fact is mentioned in the small print.

if the following assumption doesn't work out to be true, I'd avoid the
modem in question like the plague:

      if the modem supports V.42, it supports V.42.  whether
      the software is an EPROM within the modem or code on the
      computer (or in a dumb terminal...how would you do that,
      anyways?  oh well, Rockwell must have a way) is
      irrelevant.  they obviously have the software ported to
      every possible hardware/software platform (including dumb
      terminals, which do still exist and are used....), in
      device-driver form so you can use it with whatever comm
      software you want (N/A on dumb terminals).

btw, I include network management ports on various telecom equipment in
the dumb terminal category, since you can't exactly write your own code
for them).

> Also, not all the world is a DOS machine. Not all the world even is a
> PC. There are different hardware platforms and especially in the PC
> market lots of different operating systems. Rockwell cannot provide RPI
> support for all of these.

sure they can.  it'd be a b*tch to do so, but that was their choice,
wasn't it?  :-)

> Does Rockwell provide a portable (say C code) MNP/LAPM/V.42bis RPI
> driver template that could rather easily be ported to different
> environments? If so, this would ameliorate the situation.

personally, I wouldn't want to port it myself --- that's their job.
now, if they're not doing so, they deserve to lose customers.  oh, and
I wouldn't want to wait more than a day or so for them to port it.
after all, if I bought it, I'd expect it to already work as advertised.

> However, my
> current state of information and impression is that there is only RPI
> driver support for MS-DOS and both driver code and RPI specifications
> are proprietary.

sounds to me like my original assumption is false.  therefore, I know I
won't be recommending any of these modems....people generally don't want
to be locked down by proprietary bullshit anymore.  Rockwell obviously
hasn't noticed the trend away from proprietary solutions and toward
standards in recent years (even ibm has noticed this...they haven't done
too much about it in some cases, but they've noticed it!).

if you buy this modem, but then still have to buy another modem for
applications (or platforms) not supported by it, what have you really
bought?  an oversized paperweight?

> I already received requests by concerned users how to use the RPI driver
> with their favorite non-RPI-sensitive DOS (!) comms application. These
> people were tricked into thinking they'd buy a full MNP/V.42bis modem
> only to find out that they can't use this important feature in their
> environment. Pardon the language but on behalf of those users I'm
> _pissed_off_.

this is no different than the less-reputable modem vendors selling a
V.22bis (2400 bps) modem with V.42/V.42bis and selling it as a
``9600 bps'' modem.  it's unethical as hell, and it's aimed at the
uninformed users who don't know better.  once they get stuck with the
modem, it doesn't matter, because they can't return it.  and most of
the types who get stuck like this don't really have a wide enough
audience for their complaints to do too much damage.

> However, I
> doubt that licensing RPI technology explicitely forces a modem vendor to
> plainly state that the resulting product is only able to deliver its
> claimed performance with the supplied software driver.

doing so would then interfere with their ability to sell to the
uninformed user, as they would then be informing said user.  in other
words, they'd lose the sale.  as I said, unethical as hell, but (as far
as I know), sadly enough, it's legal.

all of this is why I insist on going along when friends/family buy
modems (not to mention being careful about who I do/don't recommend
professionally).
   --jim

--
#include <std_disclaimer.h>                                 73 DE N5IAL (/4)
------------------------------------------------------------------------------
INTERNET: [e-mail]  |  [e-mail]
          [e-mail]       (OLD): [e-mail]
AMATEUR RADIO: n5ial@w4zbb        AMTOR SELCAL: NIAL       ICBM: 30.23N 86.32W
------------------------------------------------------------------------------