Re: Windows CE?

[dewar@gnat.com (Robert Dewar)]

"Robert C. Leif, Ph.D." <rleif@rleif.com> wrote in message news:<mailman.1001142084.4963.comp.lang.ada@ada.eu.org>...
> From: Bob Leif
> To: John McCabe et al.
> This thread is illustrative of the major problem with 
> marketing Ada. Windows CE is obviously a potentially good 
> market for Ada.

See below, our marketing department does not accept that an
unsubstantiated claim by Robert Leif (or anyone else) that something
is obvious is equivalent to fact :-)

One thing you learn quickly in the Ada business (or any
other compiler business) is that there are always people
who are willing to tell you "build this port, it is sure
to be a winner". We have made the mistake of believing
that a couple of times in the past (and that was in cases
when the message was coming from very credible sources).

> Unfortunately, it appears that the compiler companies are 
> reticent about creating a product on internal funding.

Not at all, it is just that we only do this if we have
good evidence that the internal funding is well spent. 
Many significant products have been developed purely with
ACT internal funding, including several ports (e.g. the
Interix port), and tools (ASIS, GLIDE, GVD etc).

> Since the original market for Ada was primarily directed
> toward the Defense Department, the compiler vendors have 
> had little experience with the consumer market.

I am not clear about what you mean by consumer market. Certainly
several of our customers are in what I would
consider to be the consumer market (e.g. cable television).

But if you mean low cost compilers for hobbyists, then it
is not that we have no experience in this market, it is
simply that we think it is not the market for us. If you
think there is a market here, then why not put your money
where your convictions are, and generate such a product.
The nice thing about Free Software is that you don't have
to start from scratch, you can create such a product using
GNAT as the starting point. This has been tried before and
failed, but if you know how to make it work, by all means
give it a try.

> In most companies, the marketing department would 
> determine if a market had a good chance for profitability
> and management would authorize the creation of the 
> product.

That of course is how Ada Core Technologies works, and I
would be surprised if it is not how other companies work.
It is just that we have a different view from you on what
markets have a good chance for profitability.

> In order for Ada to thrive, Ada vendors are going to have 
> to become more entrepreneurial including learning how to 
> attract capital.

Actually Ada Core Technologies is thriving just fine from
revenues without the need to attract capital, and that is
the way we prefer to keep it, since that allows us to make
long term decisions without the pressure of capital providers pushing
for maximizing short term returns. Our
revenue and profitability are on a steady upwards trajectory. We are
not getting rich Microsoft style, but
we have adequate resources to support active future 
development of Ada in those areas that *we* think are
important.

Of course others may disagree, and Robert Leif has always
disagreed, but he seems to be willing only to try to exhort
others to spend their money, and not to spend his own. We
make our own decisions on which markets to address, and
so far these decisions have worked well for us (and we
think for the Ada marketplace). But competition is a great
force for progress, and once again, I think it would be just fine if
Robert would try to see if his ideas could be
put into practice :-)

> I might note that the number of "hobbyists" is 
> potentially large and could be a source of revenue. 

Well there you go, if there is a potentially large source
of revenue that is going untapped because other vendors
are missing it, that sounds like an opportunity you should
persue.

> However, I suspect that a broad Ada market will
> require extremely user friendly tools.

Well for sure everyone is in favor of "extremely user
friendly tools". Of course this broad consensus starts
to breakdown when you try to agree on exactly what that
means. In particular I see a tension between ease of use
and power that is not so easy to address. Our development
is more in the area of supporting large projects than small
scale hobbyists (e.g. the development of the project support facility
in GLIDE). But on the other hand, Martin
Carlisle has done a very nice job of addressing smaller
scale visual development. I am sure Martin would be happy
to get suggestions on how AdaGIDE can be made "extremely
user friendly" if you don't think it is already.

> I also strongly believe that the
> actual core of Ada is inherently easier that that of its 
> competitors: Java, C, C++, and BASIC.

Are you talking ease of use here, or ease of implementation. I am
assuming it must be the former, not
even the most rabid Ada supporter would claim Ada is easier
to implement than C.

As for ease of use, I generally agree, But remember that
ease of use and technical superiority are not guarantees
of success. Look for example at the failure of PL/1 to
displace COBOL despite huge marketing push by IBM. PL/1
is not an ideal language, but it is clearly superior to
COBOL in all respects from a language point of view.

Robert Dewar
Ada Core Technologies

P.S. we are certainly keeping a close watch on the CE
situation, and, as hinted in this thread, it is by no means
impossible to use GNAT on CE today for certain applications. As to
when a full CE port of GNAT Professional appears, that depends on
serious customer
interest, which is gauged by the contacts our sales
department receives, trade show interactions etc (but
not by reading CLA :-)