Google Groups no longer supports new Usenet posts or subscriptions. Historical content remains viewable.
Dismiss

LispWorks Pricing Information from Xanalys

66 views
Skip to first unread message

Rolf Mach

unread,
Nov 30, 2000, 3:00:00 AM11/30/00
to
Xanalys would to announce new LispWorks pricing for Unix, Linux and
Windows, effective immediately.

As many of you know, LispWorks was developed and distributed by
Harlequin Ltd., based in Cambridge, UK. Late last year, a French
company, Global Graphics SA, purchased Harlequin and subsequently formed
Xanalys, Inc. to carry on the LispWorks mantle with many of the same
people who developed the product under Harlequin.

To commemorate our new company, Xanalys has established a new pricing
structure, which we have included below. Xanalys is committed to
developing and supporting the LispWorks products as well as maintaining
a stable and reliable environment for all upcoming versions of all
operating systems.

We invite you to visit our web site at www.xanalys.com for comprehensive
LispWorks product information or send an email to lisp-...@xanalys.com
for detailed pricing and ordering information. For North American
inquiries, please contact Mrs. Brigitte Bovy ++1 781 392 1604 or for
Europe and other locations, please contact Mr. Rolf Mach ++49 +6142
938197

Xanalys’ LispWorks currently has three versions and seven modules;

1). LispWorks Personal Edition for Windows or Linux is free and can be
downloaded at http://www.xanalys.com/software_tools/downloads/index.html

All Xanalys Windows/Linux products support Windows 95, 98, NT, 2000 plus
major Linux configurations.

2). LispWorks Professional Edition for Windows or Linux for $799 USD,
including CLIM and free runtimes.

3). LispWorks Enterprise Edition for Windows or Linux for $1,999 USD,
which includes CLIM, CORBA, CommonSQL and KnowledgeWorks.

4). LispWorks Enterprise Runtimes for Windows or Linux for $300 USD.
This is required when using CORBA or KnowledgeWorks within LispWorks
Enterprise. All other LispWorks Enterprise Edition for Windows/Linux
runtimes are free.

5). LispWorks Developer for Unix for $4,500 USD, which is available for
Solaris, IRIX, HP-UX, AIX, OSF1.

6). LispWorks CLIM option for Unix for $1,150 USD, which delivers full
CLIM 2.0 implementation.

7). LispWorks CORBA option for Unix for $1,500 USD.

8). LispWorks KnowledgeWorks option for Unix - for $1,500 USD. This is a
complete expert system shell: forward and backward chaining rule system
plus Prolog like handling of logic clauses.

9). LispWorks Transducer option for Unix - for $1,500 USD. This is an
optional module for LispWorks Developer for Unix to create runtimes.

10). LispWorks Runtime for Unix - for $750 USD. A license key for each
runtime created by LispWorks Developer & Transducer for Unix.

Please contact us for our academic price list.

--


____________________________________________________________________
XANALYS - www.xanalys.com

Tools for Data Analysis and Intelligence-Led Investigations
Software zur Datenanalyse und kriminologischen Untersuchungen


Rolf Mach
Business Development & Sales Manager, Europe

An der Schaafhansenwiese 6
D-65428 Ruesselsheim, Germany

Phone ++49 +6142 938197
Fax ++49 +6142 938199
rm...@xanalys.com
____________________________________________________________________

Martin Cracauer

unread,
Nov 30, 2000, 3:00:00 AM11/30/00
to
Rolf Mach <rm...@xanalys.com> writes:

>Xanalys would to announce new LispWorks pricing for Unix, Linux and
>Windows, effective immediately.

What about updates? Do they still require a service contract,
uninterrupted since initially purchasing Lispworks?

Martin
--
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
Martin Cracauer <crac...@bik-gmbh.de> http://www.bik-gmbh.de/~cracauer/
FreeBSD - where you want to go. Today. http://www.freebsd.org/

Rolf Mach

unread,
Dec 1, 2000, 3:00:00 AM12/1/00
to
Martin Cracauer wrote:

>
> What about updates? Do they still require a service contract,
> uninterrupted since initially purchasing Lispworks?
>

The most recent patches can be downloaded for free at

http://www.xanalys.com/software_tools/downloads/patch-selection.html

You can expect some news about updates soon.


Rolf Mach

Dr Nick Levine

unread,
Dec 1, 2000, 3:00:00 AM12/1/00
to Martin Cracauer

> What about updates? Do they still require a service contract,
> uninterrupted since initially purchasing Lispworks?

That looks a little harsh. Do you really mean that, if you once let your
service contract lapse, you cannot obtain future upgrades?

-n

Erik Naggum

unread,
Dec 9, 2000, 9:05:14 AM12/9/00
to
* Dr Nick Levine <n.le...@anglia.ac.uk>
| The interesting difference between LispWorks and Allegro pricing
| remains the issue of charging - or not - for runtimes. Without wishing
| to forestall the debate, it is my opinion what whatever commercial
| difficulties Harlequin as-was had with its lisp products in the past,
| free runtimes was not the cause.

I have tried, but I can't make sense of this last statement. Did the
Lisp product bring in "enough" money that you did not have any need
for royalties? It seems odd that a license to print money would not
be useful to someone, especially someone in commercial difficulties.
Would it not have mattered if they got any money from royalties?
(That means no relevant user base, which contradicts other information
I have.) I also find it strange that one can have opinions about
accounting issues, so I guess you mean that it is your opinion that if
they did charge for runtimes, they would have lost an equal amount of
sales, which I suppose is a valid opinion. However, I think it is
hard to pin anything down on a single cause, so for any contributory
cause, you could easily say "it was not the cause" unless it was so
major a contributory cause that it would at least have equalled the
rest combined. Since that is unlikely, unless the whole business were
focused around royalties, which we know was not Harlequin's model, I
can't make heads or tails of what you're trying to say about Harlequin.

If we apply an interpreter of British understatements and general
vagueness in preference to precision, however, I get the impression
you're trying very hard to have an opinion about another company,
which _does_ have royalties as part of their business model and that
you are trying to invalidate that model by implication.

Personally, I think the whole business model of the software industry
is rotten to the core. Microsoft is not even a contributory cause --
Bill Gates isn't smart enough to invent or create something like this.
It used to be the norm that software was essentially free (gratis) and
just vehicles to move hardware, which was tangible enough to make it
easy for the anal-retentive beancounters to count and weigh and such,
especially those in the government offices in charge of squeezing the
juice of out the produce of society. I think the software crisis has
been created by the tax laws and government officials who were unable
to understand the value of the computer industry products. Since the
Western business world is operating very closely in a trigger-response
pattern relative to changes to the tax laws, the regulators are fully
to blame for the inability of societies to value software properly,
and hence the "protection" the software industry got in the Y2K scare,
the ability to skirt warranty and every other consumer protection law,
the non-constitutionality of the War on Software Piracy (which is much
worse than the absolutely insane War on Drugs measured in results and
costs and accepted loss of personal freedom), and numerous additional
deep-rooted and pandemic consequences of bad (working to destroy) and
sometimes evil (intended to destroy) policy and ignorant idiots with
too much power. One of the consequences is that software development
is paid for over the marketing budgets, like cheap plastic pens with
logo imprints that work no longer than you remember where you got it
and T-shirts of so low quality that they couldn't have been sold as
clothes -- when you give some trinket away as a marketing gimmick, of
course you don't want to warrant, support, or maintain it, and of
course nobody would even dream of paying royalties for it, they can
just get another cheap pen or T-shirt or shirnk-wrapped software
package of the shelf somewhere else. Microsoft, the ruling king of
mass marketed trinketware, are just surfing on the same tsunami that
lifted Taiwan and Hong Kong from poverty into an inflated economy that
just _had_ to crash down and wash out _enormous_ values some day. We
see the same situation in the so-called Internet business and lots of
people have written articles and books about why it cannot succeed in
the long term the way it operates today. In the meantime, producers
of fine pens and quality clothes and excellent software suffer because
they cannot brand their products with their own logos in large print
and give them away while waiting for the wash-out. While the users
are waiting for quality software, lobbyists have even succeeded in
exempting Internet companies from sales tax, rewarding incompetence
and waste, computers and software have remarkably short lifetimes in
the generally accepted accounting principles and tax laws, often with
depreciation to near zero before Moore's law could possibly apply,
rewarding "investment" in new hardware as a very good way to reduce
your taxable profits for the entire business market. Like an economy
where the inflation is so high you'd better spend your money while it
is still hot off the printing presses or it'll be worthless, you have
to be inordinately smart and probably rich in real values to recognize
what the _real_ values will be a few years down the line. But like
all inflation, the inflation in the software industry is caused by the
government policies of _devaluation_ of the reference valuables, which
when push really comes to shove is the ability of the average citizen
to secure a reasonable living after his ability to work hard enough to
produce more than he consumes has been depleted by the passage of time
-- or, in other words: Our trust in a stable future. That is the real
victim of the ongoing novelty craze and the give-away culture that has
resulted from massively retarded policy decisions shortly after WWII,
when the government goal was to rebuild and instill hope in the future
near term. The government role has since been reversed because the
near term future of its past policies are now _our_ past. Instead of
being a guarantor of stability and long-term safety that each of us
cannot build or even maintain on our own, policies in the information
technology industries have turned into guarantors of instability and
short-term profiteering, effectively betting the future on the fun we
can have today, a massive lottery where everybody loses, especially
the guy who wins $25 million and discovers that everybody else has to
_continue_ to play (read: lose money to) the lottery for him to get
monthly installments. The software industry has turned into a
pyramid game because the government valuation strategies for software
have penalized longevity. It has absolutely _nothing_ to do with the
so-called "rapid pace" of the technological development. It isn't
rapid and I'll dispute a claim of general development, too. It's all
about marketing old ideas in new and ever more shiny wrappings, and
nobody does that better than Microsoft today. Their "innovation" is
_purely_ restricted to more shiny wrappings, because that is where the
money is in today's market. Here's how to destroy their power: Force
software and hardware acquisition costs to be depreciated over no less
than five years, and treat support, royalties, and other lifetime
costs as financial costs, effectively loans to the vendor. There will
be lots of excruciating pain lining up for the stupid people who bet
their future on the short-term fun and the quick buck, but that's just
life: Quick bucks sometimes vanish with no notice whatsoever and if
you reap enormous profits in the short term, you're supposed to take
care of the long-term saving yourself, anyway. Government was set up
to protect society as a whole and all individuals, not only some of
them, and it currently protects only some individuals at the expense
of our future and the long-term stability of society as a whole.
Worse, it's the government that is playing the lottery with its
citizens and random fate will determine which us have to pay the big
prize the government decides to collect if it isn't stopped from
betting our future on low-quality software by rewarding incompetence
and short-term profiteering in the information technology industry.

Programming never was just about making some trinket please a user or
a programmer, but building infrastructure and long-term solutions and
improvements to the human condition. The global interdependency on
software makes low-quality software even in the home and small office
a serious threat to our ability to maintain safe and secure living
conditions. E.g., viruses, which threaten both financial stability
and vital information, are caused by Microsoft, not malicious hackers
like that self-serving jerk Bill Gates constantly whines, as we all
_know_ that bad people exist and will always exist and form a threat
to our safety unless we take precautions against them. Responsibility
for the failure to take such precautions cannot be placed anywhere but
on the creator of the vehicle of the malice -- ironically recognized
by the U.S. court system for everything _but_ software. The unique
perspective of longevity and stability of a programming language like
Common Lisp is what _may_ save us all through the impending crash, and
the ability to save up enough cash to survive the wash-out is crucial
to carrying us through.

To bring this back to the concrete level: The commercial difficulties
that Harlequin had, no matter what their causes, is a strong indicator
that they some if not all factors of their operations were not smart
enough. I'm loathe to exonerate free runtimes and no royalties just
because someone waves his hand and opines indirectly that something
else was "the cause". Instead, I think software _should_ be subject
to continued payments as if they were services actually produced by
the people who produced the software. The legalistic reasoning behind
this is that software is not static like hardware, and that it takes
effort and incredible levels of foresight and intelligence to ensure
that software continues to work under changing conditions, which means
that royalty implies warranty and vice versa, and that this warranty
cannot be paid for through purchase of a product alone because of the
varying cost of the warranty according to the risks involved in the
different uses it sees. The only time you should accept _not_ to pay
royalties is when you have accepted to assume all costs and all risks
of ownership yourself, and if you do that, you're a goddamn fool, in
_any_ business relationship, meaning that whoever got away with
selling you shrink-wrapped software without normal product warranty is
a con artist and a fraud and should hae been prosecuted and shut down
by the government. Instead, we have the government perpetrating the
worst possible fraudulence towards its own citizens by allowing the
continued operation of software companies that threaten the very
fabric of society through an open-ended license to engage in what
would have been criminal neglect in any other industry.

Then again, people have their daily routines to go through, not enough
cash to think long-term, and generally lead lives they do not control,
so they don't hae time to concern themselves with what they fear might
be huge policy issues well outside their graspability, but which
really comes down to this: If the U.S. Government continues to protect
the essentially fraudulent operation of the software industry and
continue to support corporations which deny and ridicule the concept
of product warranty, it bets the future of its citizens on a whimsical
idea that safety and stability are _unnecessary_ in the software world
and therefore that there _are_ no risks worth defending and preparing
against, meaning: The government doesn't care what happens to any of
its citizens 10-20-40 years down the line, but it is impossible for
any one of us to prepare ourselves for more than a few years ahead,
and the more complex the infrastructure we rely on, the fewer years,
and that affects the entire globe. The U.S. Government is gambling
with the lives of billions of people when it protects the software
industry from having to conform to quality measures and allows the
industry to run pyramid games and using future funds to pay for past
mistakes. This is criminal recklessness on a global scale, and no
significant number of voters will notice until their pension funds and
social security evaporate due to the impending crash in the protected
industries.

Paying royalties to one company when you have the option of not paying
royalties to another company is not going to change the big picture,
but a company with commercial difficulties _not_ demanding royalties
is tantamount to _encouraging_ further recklessness. Being a parasite
off of the marking budgets of large corporations who try to get paid
for something entirely different is not my idea of supporting longevity
of anything, so using marketing-funded tools is very far from what I
consider rational use of my time and money, not the least because the
marketing-funded tools will have to continue to amaze and thrill the
statistical fraction of the mass market in marketing terms, not in any
quality terms that a product that has to pay for itself from every one
of its customers would have to do.

Sorry for the gravity of the message. My cat's back from the hospital
with about 7 lives to go, and longevity issues tend to crop up with me
when the threat of its absence is most clear and present. Have a nice
weekend, though -- I know I will.

#:Erik
--
"When you are having a bad day and it seems like everybody is trying
to piss you off, remember that it takes 42 muscles to produce a
frown, but only 4 muscles to work the trigger of a good sniper rifle."
-- Unknown

Dr Nick Levine

unread,
Dec 12, 2000, 6:12:44 AM12/12/00
to
Erik Naggum wrote:
>
> * Dr Nick Levine <n.le...@anglia.ac.uk>
> | The interesting difference between LispWorks and Allegro pricing
> | remains the issue of charging - or not - for runtimes. Without wishing
> | to forestall the debate, it is my opinion what whatever commercial
> | difficulties Harlequin as-was had with its lisp products in the past,
> | free runtimes was not the cause.
>
> I have tried, but I can't make sense of this last statement.

My apologies, let me try again.

Harlequin failed to turn its lisp products into a commercial success.
I'm not sure I want to go into the reasons for that in a public forum -
I'd need to be convinced that it was going to help more people than it
hurt. But I think I can state with confidence and without upsetting
anyone that these reasons were much more fundamental than the question
of whether or not to charge for lisp runtimes. In other words, that
question was never put to the test because the lisp business never
developed to the point where it would have made any significant
difference to the outcome.

> If we apply an interpreter of British understatements and general
> vagueness in preference to precision, however, I get the impression
> you're trying very hard to have an opinion about another company,
> which _does_ have royalties as part of their business model and that
> you are trying to invalidate that model by implication.

Not at all, I assure you. I think these are two different approaches to
what is essentially a business decision. Both are potentially valid.
[Thinking of "other companies", I had an entertainingly energetic
discussion with Fritz Kunze on this very subject last year, which I
think I may have lost. On the other hand Fritz was paying for the
lunch... ;-)]

> Personally, I think the whole business model ... customers would have to do.

Sorry, I haven't had time to read all this yet. I was up disgustingly
late last night devising a multiple-choice test for my Modula2 students,
and it seems to have sapped my energies.

Hope you cat's on the mend.

- nick

0 new messages