eberard@ajpo.sei.cmu.edu (Edward Berard) (02/29/88)
For purposes of this immediate discussion, I will define a commercial
application of Ada technology as any "non-weapons system" application
of Ada technology. This definition includes the use of Ada technology
for financial systems for the military and embedded applications in
the private (non-government) sector. While this definition is not as
complete as I would like it to be it will suffice for this message.
Most applications of Ada technology, worldwide, are commercial. For
example, in Europe, depending on whose figures you accept, four to
fifteen times as much money is spent on commercial Ada applications
compared to military applications. In Japan, virtually all Ada work is
commercial. However, in the United States (U.S.), Ada is still thought
of largely as "a Department of Defense (DoD) language."
What I am proposing is the blatant commercialization of Ada
technology. By commercialization, I mean establishing Ada technology
in the commercial marketplace, at least in the U.S., to the extent
that much more money is spent on commercial (non-weapons systems) Ada
applications than on military Ada applications.
I contend that if this is accomplished, all parties (the U.S.
government, the providers of Ada-related products and services, the
consumers of Ada technology (i.e., those who must develop, test, and
maintain Ada applications), and the public at large) will benefit. Let
me explain further.
The U.S. Department of Defense is the largest single consumer of
computer technology on the face of the earth (or so it is claimed).
However, it has also been observed that DoD expenditures on software
represent only about 5% (five percent) of the total annual
expenditures on software in the U.S. It should be obvious that the
commercialization of Ada technology would benefit the providers of
Ada-related products and services -- if nothing else, at least by
opening new markets.
The U.S. Government has been attempting to "ignite the private
sector's interest in Ada technology" for some time now. Examples of
these attempts include STARS (Software Technology for Adaptable
Reliable Systems), the Ada Software Repository (ASR), and the Software
Engineering Institute (SEI). The general idea is that the DoD would
expend resources on establishing Ada technology until such time as the
private sector began to make substantial (and self-sustaining)
investments in the technology.
The benefits to the U.S. Government of the commercialization of Ada
technology are many. Obviously, the government could reduce the amount
of resources expended in promoting and "seeding" the technology. (The
private sector seems to be much more proficient at this anyway.)
Further, due to the increased availability of Ada-related products and
services that would result from the commercialization of the
technology, the DoD would find fewer obstacles to its Ada mandate.
The benefits to the consumers of Ada technology are primarily related
to the increase in the availability and quality of Ada-related
products and services. By increasing the demand (i.e., in the
commercial sector) for Ada-related products and services, we increase
the financial incentives for investing in the technology. This results
in more, and hopefully better, products and services from which the
consumers can choose.
Since the public at large are the ones who pay the bills (e.g., via
taxes), the reduction in the expense of mounting an Ada project due to
increased availability of quality products and services should have a
positive impact on the cost of the project. Further, given that the
main thrust of the Ada effort is more reliable, higher quality, and
less expensive software, the public gets a better product for its
dollar.
Why have I bored you with this discussion? Recently, I have been
attempting to help some commercial Ada clients, and in the process of
trying to get some questions answered, I found some glaring
deficiencies in the Ada marketplace, i.e.:
1. While many vendors of technology have active and
well-established third party product programs (e.g., Apple's
Certified Developer program and Sun's Catalyst program), few, if
any, in the Ada community have such programs. For example, how
many Ada compiler vendors have established programs which help
those who wish to develop commercial products with their
compilers?
2. There are any number of newsgroups, mailing lists, etc. which
allow detailed discussion of products and services. For example,
consider the communication channels open to Macintosh users
(e.g., comp.sys.mac, comp.sys.mac.programmers, info-mac, and
Delphi). Users of Ada technology do not enjoy the same freedom
of discussion as do their counterparts on other mailing lists. I
am not necessarily advocating a change to comp.lang.Ada.
3. In the U.S., there are no publications, which are totally
dedicated to Ada technology, which accept advertising. Vendors
of Ada products and services must either have enough money to
cover the potential marketplace by many ads in many different
publications, or hope that direct mailing to the SIGAda mailing
list will bring in new customers.
I also noted some curious attitudes among current users of Ada
technology. The classic example is the selectively deaf Ada advocate.
This is an advocate who recites a litany of the advantages of the
technology (e.g., higher productivity, portability, and reusability),
yet does not use the technology for any project for which it is not
mandated. Another example is the "Ada is only good for ..." people.
These are people with limited imaginations who say things like "Ada is
only good for embedded applications.", "Ada is only good for
non-real-time applications." and "Ada cannot be used for artificial
intelligence applications."
One of the barriers to the commercial acceptance of Ada technology is
the assumption that the only appropriate use of the technology is for
projects on which it is mandated.
I am not advocating an abandonment of "traditional" Ada applications.
What I am advocating is an increase in the discussion of the
commercial applications of Ada technology. In addition, the commercial
sector has many ideas and concepts which I feel could be beneficial to
the classic Ada community.
-- Ed Berard
(301) 695-6960