Path: utzoo!news-server.csri.toronto.edu!cs.utexas.edu!swrinde!elroy.jpl.nasa.gov!ames!uhccux!uhunix1.uhcc.Hawaii.Edu!todd
From: todd@uhunix1.uhcc.Hawaii.Edu (Todd Ogasawara)
Newsgroups: comp.lang.prolog
Subject: Re: Arity's so-called upgrade
Message-ID: <11954@uhccux.uhcc.Hawaii.Edu>
Date: 14 Mar 91 21:24:19 GMT
References: <11768@uhccux.uhcc.Hawaii.Edu> <WEISS.91Mar13134309@toucan.lcs.mit.edu>
Sender: news@uhccux.uhcc.Hawaii.Edu
Organization: University of Hawaii
Lines: 168

In article <WEISS.91Mar13134309@toucan.lcs.mit.edu> weiss@theory.lcs.mit.edu (Paul G. Weiss) writes:
>This forum is not and should not become a substitute for direct
>communication with Prolog vendors regarding product features and
>technical support.  Arity has always welcomed suggestions and

First, I want to say that I am the person who posted the original message.
I also want to say that I've communicated with Paul a few times years ago
and found him to be very cordial and helpful. So, I hope everyone
understands that what I say here or elsewhere about Paul or Arity is never
intended to be a flame. Rather, it is intended to be information (good, bad
or otherwise) from a long time paying customer (I've been using Arity since
version 3.2 which was released in 1984) to other current and potential
Arity customers.

Next, I want to say that I disagree with Paul about this... I hope he
noticed the following lines from my original message.  (I wrote in my
original message):

]I wrote and mailed a
]reasonably toned letter of dissatisfaction to Meredith Bartlett (VP of
]Marketting at Arity) two days before I learned that I was not going to get
]any new manuals (which I hoped would clear up a few changes they made to
]some proprietary predicates). Now, that I know I am not getting any new
]documentation, I am outraged at what they have sold to me under what I
]consider "false pretenses."

]If any one else is similarly outraged at Arity, please write a letter to
]them and say so. If anyone is interested in sending a group letter, please
]let me know and I will coordinate such an effort. I am going to wait for a
]response to all the questions I posed in my first three page letter to
]Meredith before drafting a stronger letter of complaint.

Please note that I did write a letter directly to Meredith at Arity at a
time that (1) I was very dissatisfied with what I saw and (2) I thought I
would get documentation that would at least clear up the documentation
related problems (there are other problems that documentation alone will
not clear up but those issues were also addressed in my letter to
Meredith).

Please also note that I asked people to write to me directly if they wanted
to write a group letter to Arity. Again, I did not ask for a flame fest
here. I was asking for other Arity users to contact me in order to approach
Arity with a coherent set of complaints and also to show Arity that it was
not a matter of a single dissatisfied customer shouting in the wind. I also
asked people to write DIRECTLY to Arity.

I am disappointed that Paul chose to ignore the fact that I took great
pains to ensure that a flame war DID NOT start due to my posting. I was and
still am simply interested in making sure that Arity customers discuss any
concerns they have directly with Arity just as I am doing to make sure that
they hear our concerns.

I am also disappointed that Paul feels that this forum is not appropriate
for discussing Prolog vendor specific issues. I have seen many discussions
about Quintus Prolog, PDC Prolog (formerly Borland Turbo Prolog) and other
vendor specific versions of Prolog discussed here. And, in fact, I see many
other vendors' implentation of programming languages discussed actively in
other news groups (take a look at comp.windows.ms.programmer if you want to
see a really active discussion about the recently released Borland C++ 2.0
[I just received my upgrade of Borland C++ 2.0 and have been following the
threads over there daily]). I think that this is the appropriate forum to
discuss aspects of vendor implementations of Prolog.

>complaints from its customers in order to be better able to
>serve them.  Indeed, our current version, version 6.00,
>incorporates several features suggested by our users.

I'm sure that's true. I noticed a few things in 6.00 that I found useful
also. The point is, though, that the upgrade notice and the price indicated
that the move from version 5.1 to 6.0 was a major one. Version 6.0 is not a
major evolution from 5.1. At best it should have been called version 5.2
and cost much less than the ~$127 I paid. How can a product be considered a
major release change if the company says that no new manuals are needed
because the product hasn't changed enough? And yet, I now have something
like 30 pages of documentation changes to wade through between release 5.0
(when the manuals were released) and 6.0? That just doesn't make sense.

Borland, on the other hand, charged me $99+shipping to go from Turbo C++
1.0 to Borland C++ 2.0. For this price, I received what I consider to be a
evolutionary change from the previous version (it looks and behaves
differently [and better for that matter]). I was provided with completely
new documentation that reflects the changes in the product (to be fair, it
also came with a README file that had documentation changes -- but the
changes were, perhaps, 2 pages of material, not 19 to 30 like Arity's).
Another interesting comparison between these two upgrades that I received
about two weeks apart is that Borland's sample code modules compile as
indicated in their documentation. Arity's, on the other hand, do not
compile without modification to the source code that are not indicated
anywhere in the README-type files on the disk. This was a minor, but
annoying discovery as I tried to understand a few items in V6.0 by looking
carefully at the sample files (which tend not to display features new to
V6.0 anyway, I discovered).

I'm not trying to claim that Borland has provided me with a "perfect tool"
by any means. However, so far, I do feel like I have received an upgrade
that reflects the major change in version number (from C++ 1.0 to 2.0) and
that I was charged relatively fairly (~$100) for what I was provided.

>Arity is very proud of the latest release and the response from
>users has been favorable.  For criticism, inquiries, complaints,
>praise, or anything else we may be reached directly by any of
>the following means:
>
>Telephone:	(508) 371-1243		(9am - 5pm Eastern Time)
>Fax:		(508) 371-1487
>BBS:		(508) 369-5622		
>US Mail:	Arity Corporation
>		29 Domino Drive
>		Concord, MA 01742

Again, I urge people who are dissatisifed with what Arity has provided in
V6.0 to do what Paul says. Contact Arity directly and let them know that
you are dissatisfied and why ("why" is the important part). They also have
an 800 number that is listed on their letters and in their advertisements
in AI Expert.

Finally, I want to say that in the many years I have dealt with Arity I
have found their staff to be professional and responsive. In fact, Meredith
left a message on my home answering machine yesterday and I am sure that we
will have a productive discussion when we are able to touch bases with each
other in the near future. I also want to say that their product was the
only MS-DOS Prolog available at the time (1984) that allowed me to have a
"real" Edinburgh style Prolog for PCs to develop with. Although that
statement is no longer true, I would prefer to stay with Arity because (1)
I have come to respect the staff I have interacted with, (2) I prefer not
to have deal with a whole new set of proprietary Prolog extensions
(menuing, etc.), and (3) quite frankly, I've invested something like $2,000
out of pocket over the years and want to see a return on my investment on
this product.

Again, I am disappointed that Paul's opinions and mine about using this
forum for a discussion are so far apart. I still believe my statements are
true and my actions appropriate. I still hope that despite Paul's feelings
on this matter, that the result of these discussions will be productive and
constructive and ultimately benefit those of us who are Arity customers.

Perhaps a constructive P.S.: Before getting Arity 6.0 I had hoped that I
would get an integrated development environment (what Borland referes to as
IDE) like Borland's Turbo C++ environment (and in fact Arity documents say
that their environment is Turbo-like; I disagree with this statement).
However, I have been able to get around some of the environmental
shortcomings in Arity by starting to modify the Sage Professional Editor [SPE]
(I am using V1.0 and awaiting the V1.1 upgrade) to tune it to Arity. It
is quite simple to launch a compile job from Sage's editor (something
you can also do from Borland C++'s editor but not from Arity's). If any SPE
users are out there, I can give them the modifications I made to their
errorfix.pel that allows you to run APC from SPE.  SPE also provides me
with multiple edit windows on screen simultaneously, mouse control in the
editor, and many other features which I feel are necessities in a modern
programming environment. Now, that I know that Arity's environment isn't
suitable for my programming environment desires, I am adding more features
to SPE to allow me to use it as my environment for working with Arity. I'm
already able to bring up the various Arity HLP files in various windows of
SPE while editing source code. The next step is to add context sensitive
features that will lead me to the appropriate place in the appropriage HLP
file when I see an ERR message or need to check on syntax. I've also tried
running Arity V6.0 and Sage simultaneously under Windows 3.0 in 386Enh mode
and found that that works relatively well (Arity does see the Expanded
memory emulation available when running DOS programs under 386Enh mode,
btw). My only problem is that the DOS windows that appear on my SuperVGA
screen are pretty small and my aging eyes can't see the characters as well
as I would like :-).

--
Todd Ogasawara, U. of Hawaii
UUCP:		{uunet,ucbvax,dcdwest}!ucsd!nosc!uhunix!todd
BITNET:		todd@uhunix
INTERNET:	todd@uhunix.UHCC.HAWAII.EDU