NetNews Usenet Archive 1992 #27/NN_1992_27.iso/spool/comp/dcom/fax/2142

Internet Message Format | 1992-11-19 | 5.2 KB

Path: sparky!uunet!mtndew!friedl
From: [e-mail] (Stephen Friedl)
Newsgroups: comp.dcom.fax
Subject: Re: MultiTech Fax Modems could run Gnu NetFax if you call
Message-ID: <[e-mail]>
Date: 20 Nov 92 05:57:39 GMT
References: <[e-mail]> <[e-mail]> <[e-mail]>
Organization: Steve's Personal machine / Tustin, CA
Lines: 97

Keith Smith and I have been "debating" the class 2 standard in
this public forum, and it is obvious that he and I are talking
about only partially related topics.

His topic is why standards are good things, and in that respect
I believe we are in complete agreement: they promote uniformity
where it matters, fosters competition (keeps prices down and
quality up), and makes moving things around much easier.  When
I am wearing my VSI*FAX "porting" hat, I promise you that I am
thinking heavily about standards such as POSIX and ANSI C.

The other topic is whether the benefits of standards accrue to
everybody, and I contend that they do not necessarily do so.
For instance, just because ANSI C promises to help make code
more portable, it doesn't mean that everybody will want to do
this.  Somebody doing embedded software development for an 8051
microcontroller just won't care about ANSI at all because the
code is inherently tied to the underlying hardware.  This is not
to say that the standard is bad, just that this is not one of
the applications that drove the need for standardization in the
first place.

Now for fax.  The main benefit of the class 2.0 standardization
is that we should be able to plug in brand X modem and have it
work with our software.  Some people are very interested in this,
but my experience is that UNIX fax vendors just don't get
very excited about it.

Part of the problem is the definition of "support".  If you are
thinking about the technical requirements to make software talk
to a modem, then undoubtedly class 2.0 will make this much
easier.  I believe that we will always have some local variations
that require per-modem workarounds, but this will hopefully be
a very small matter.

The broader definition of "support" that I mean includes things
unrelated to the command set.  Providing real commercial support
for a UNIX fax modem requires a great deal of work beyond the
driver, and most of the UNIX fax software companies just aren't
that interested in going down this road for almost no return.

After the software knows about the modem (which might be the
easy part), we have to develop cabling specifications for all
the machines that we support, we have to make notes of all the
appropriate DIP switch settings, make matrices of ROM upgrade
fixes, and get all of this to the tech support department.  When
a customer says "it doesn't work", we have to spend a lot of
time finding out whether it is our problem or that of the modem,
and these can sometimes take a very long time.

We have to develop some kind of relationship with the vendor so
that if we find a bug we have some decent chance of getting it
fixed.  Recall that the software company usually takes the heat
for ALL bugs.

My point is that why should we go through all this work when it
won't mean any additional sales?  Lots of people ask us if they
can use their Supra or Practical Peripherals or whatever modem
with VSI*FAX, and we never get much heat when we say "no".  We
just tell them that they have to get the Multi-Tech, and it just
never seems to be a problem.

One very key difference between DOS fax and UNIX fax.  The DOS fax
user tends to invest dramatically less in his or her setup than a
UNIX buyer does.  Many people with DOS fax just want it to play
around or send letters occasionally.  Few buy a fax modem primarily
for fax.

Under UNIX, the cost of the modem is typically only a small part
of the entire effort.  The UNIX fax software almost always costs
more than the modem (especially on larger machines), and even this
is dwarfed by the labor required to integrate the software into the
appropriate application.

I know of lots of companies that have invested literally hundreds
of hours in their fax add-on software that runs on top of VSI*FAX, so
the difference in price between a cheap $100 modem and a real one is
essentially noise at the bottom of a financial statement.  Saving a
couple of hundred dollars per modem is not worth more than a couple
of hours of hassle to anybody involved.  You are buying a solution
to a business problem, not a piece of hardware.

Yes, there are price sensitive users out there, and with 2.0 I expect
that we will pick up some low-end modem as part of the product line,
but we aren't really too excited about it.  We have to go through all
the same painful learning that we went through the first time with
almost no return.  It remains to be seen whether we will be able to
find Multi-Tech quality in a $100 or $200 modem.

Disclaimer: I do contract UNIX facsimile software development for
            V-Systems, but I do not speak for them or anybody else.

--
Stephen J Friedl | Software Consultant | Tustin, CA |     [phone]
3b2-kind-of-guy  | I speak for me ONLY |   KA8CMY   | uunet!mtndew!friedl

Attorneys are paranoid because they assume everybody else is dishonest too