There is a story (probably exaggerated) that when the board of Hitachi saw
the bill from I.P.Sharp associates for the port of SHARP APL to Hitachi
mainframes, they (after receiving the necessary medical attention) wrote to
Ian Sharp and said "PLEASE send us an invoice for a million dollars
immediately!". And IPSA was a relatively successful commercial operation for
a decade or two. Of course, as someone pointed out earlier, it was an odd
age, where programming was given away for free to people who were
subsequently able to burn a hundred dollars per CPU second on your mainframe
(and some of them were "multicore", even in that age). IPSA grew at an
incredible pace without an actual marketing operation because in the late
1970's APL Timesharing on the worlds' first packet-switched network was a
tool of such awesome power that it didn't need a salesman. And then, when
the PC and internet arrived, *poof*.

Buying books and training is fine, but it doesn't address the need for an
ongoing, reliable source of income that would allow the company to invest in
product development. What I would do is look at my business and say
something like: My j-based business is worth (say) $100,000. I pledge not a
tithe, but one fifth of a tithe, to support my "spiritual foundation". Write
to jsoftware and ask them to draw up a contract for $2,000 per annum and
start invoicing you. Or better still, say you have five customers - a
contract which obliges you to pay $400 per annum per customer using your
product or service based on j. Or some other number that makes you feel
comfortable that you are getting a good deal and supporting the technology
that makes it all possible. If they won't do it, then I DO suggest that you
seriously consider switching to Dyalog APL ;-)

If you don't think you are reducing your costs by 2% by using j, then I
would suggest that you are taking a completely unacceptable technology risk
by using it and I suggest that you stop immediately! Fortunately, what our
customers generally tell us is not that "It's wonderful, APL saved me 10%
last year", they are saying "I would never have gotten this project off the
ground at ALL without your product". So 2% sounds like a good deal to me :-)

This only applies if you DO have a commercial business based on J. But I'm
told that J is growing faster than APL, so there must be SOMEONE making
money out there...(?)

As a customer of array language vendors in the past, I have held several
meetings to find ways for them to send us bigger invoices. I bought a j
source code license in a previous life - and was a proud commercial sponsor
of the sparse array project. When working for a company that was planning to
float on NASDAQ(*), we managed to negotiate the sale of some stock options
to our vendor at the time. Not "out of the goodness of our hearts", but
because it is always in your interest to ensure the health of anyone you
intend to rely on. 

I'm sorry to be going on about this, and I promise to disappear for a while
now. If I have moved anyone in the direction of writing checques to
jsoftware, I would be very pleased. Sure, we're "competitors" but we have
much more in common and I actually believe that the health of j is both in
the interests of me (it's cool, mind bending stuff!) and the company that I
work for (array language vendors of the world, unite!).

You may find it hard to believe, given all this talk about "business cases",
but I enjoy writing code in APL, or doing mental gymnastics trying to
decipher J, much more than being a "businessman". But I grew up with APL, I
can't imagine life without it and I'm damned if I'm going to let it wither
and die because people aren't willing to ask for money, or contribute
appropriately. Companies like MatLab and Mathematica have made lots of money
by essentially selling applications. APL/J vendors take a more purist stance
because the core language itself *is* more valuable (and at Dyalog, we also
intend to remain a technology vendor and leave application-building to our
customers). 

But we (MUST!) demand a fair share of the revenue. The intrinsic value of J
and APL technology is measured in billions.

/ Morten

(*) The bubble burst and we had to settle for a $150M sale to Cognos (so the
underlying product is now back at IBM, from whence it started its journey 20
years ago, which is very funny - and also that as a result of this and other
acquisitions, IBM is now running APL+Win, Dyalog, and J (and APL2 - and who
knows what else). But the options DID increase in value.

-----Original Message-----
From: Raul Miller [mailto:[email protected]] 
Sent: 27. maj 2009 19:39
To: Chat forum
Subject: Re: [Jchat] More APL (was: No More APL)

On Wed, May 27, 2009 at 8:15 AM, Morten Kromberg <[email protected]> wrote:
> However: And this is something that I think the J community should rethink
> (and then DO something about if it really cares): Unless its users find a
way
> to start paying substantial amounts of money to jsoftware so that this
company
> can hire some apprentices and start growing, I really don't see how it can
survive
> in the long term.

That's an interesting thought, however...

I have recently (within the last year) started incorporating J in some small
projects at work.  However, at the moment, it's less of a factor here than
some
other software which we paid $55 for.  I have some hope that this will
change,
over time, and at some point we may reach out to jsoftware for training.

I do not know what else I could buy from them.  (They do not have a cafe
press
shop, for example, and the only obvious "buy" link I can find from the
home page is their consulting page which I am not prepared to pay for
at this time.  I know I could buy some books, but I have already bought
most of the ones I wanted.)

-- 
Raul
----------------------------------------------------------------------
For information about J forums see http://www.jsoftware.com/forums.htm

----------------------------------------------------------------------
For information about J forums see http://www.jsoftware.com/forums.htm

Reply via email to