SOMEONE!$%#%#!%#%@ fund Jovial to Ada anything

SOMEONE!$%#%#!%#%@ fund Jovial to Ada anything

[Gregory Aharonian]

    I run a business partly known for knowing something about converting codes
across languages in a semi-quasi-pseudo-automatic way. About once a month for
the last few years (with bursts around RFPs), I get phone calls asking about
Jovial to Ada translators. For people who don't do, this is an obscene idea,
but for the managers and software engineers that call, this is a real problem
affecting their Ada transition efforts.
    To my knowledge, there are no cost-effective Jovial to Ada translators
available, at least in the 100% complete translation. There are rumors of
80%+ translators, but even so, what is usually left from such translators
to be processed manually costs as much as redoing everything.
    According to my estimates, and two tech reports, a useful tool for these
people would cost a few hundred thousand dollars. None of the contractors who
have contacted me, individually or jointly, seem willing to put up the money
to have such a tool developed. I've contacted the Language Control Facility
and they don't seem to know of anyone inside the Air Force willing to put up
the money.
    I figure that there is at least 5 million lines of Jovial that people will
be converting to Ada now and over the rest of the decade, at an enormous cost
without an standard tool. And as disgusting as people claim translators or
convertors are, a few good language specialists could put together a tool that
helps as much as possible.
    I've received enough phone calls to know there is enough of a demand to
make this worthwhile for the AJPO to fund someone to do something, but
probably not enough demand for a company to develop a product. If anyone on
the net has a good suggestion on what to do, I've got plenty of names of
people who will be your best friend.

Gregory Aharonian
Source Translation & Optimization

Re: SOMEONE!$%#%#!%#%@ fund Jovial to Ada anything

[Karl A. Nyberg]

> ... know there is enough of a demand to make this worthwhile for the AJPO
> to fund someone to do something, but probably not enough demand for a
> company to develop a product.

If you think it's such a !$%#%#!%#%@ good idea, why don't YOU (certainly a
subset of SOMEONE!$%#%#!%#%@) write one?  If there's the kind of market that
you imply, you stand to clean up.  Put YOUR money where your mouth is.  Get
your unnamed contacts/contractors to put the money up.  Or as my friend Fred
says, qwitcherbitching.

At least quantify your "knowledge" to justify having the government spend
money that even private industry won't.  Hard facts sell.  Innuendo walks.

Sounds like another Liberal Massachusetts Democrat to me... :-)

-- Karl --

-- 
Karl A. Nyberg				karl@grebyn.com
Post Office Box 497			Grebyn Corporation
Vienna, VA 22183-0497 USA		+1-703-281-2194

Re: SOMEONE!$%#%#!%#%@ fund Jovial to Ada anything

[cis.ohio-state.edu!magnus.acs.ohio-state.edu!zaphod.mps.ohio-state.edu!ua]

In article <9107092046.AA21359@grebyn.com> karl@grebyn.com (Karl A. Nyberg) wri
tes:
>> ... know there is enough of a demand to make this worthwhile for the AJPO
>> to fund someone to do something, but probably not enough demand for a
>> company to develop a product.

>If you think it's such a !$%#%#!%#%@ good idea, why don't YOU (certainly a
>subset of SOMEONE!$%#%#!%#%@) write one?  If there's the kind of market that
>you imply, you stand to clean up.  Put YOUR money where your mouth is.  Get
>your unnamed contacts/contractors to put the money up.  Or as my friend Fred
>says, qwitcherbitching.

>At least quantify your "knowledge" to justify having the government spend
>money that even private industry won't.  Hard facts sell.  Innuendo walks.

But private industry may not be willing to fund it because they don't know
the market size for the major military languages, i.e. JOVIAL and CMS-2.
NOT because they are unwilling to risk money!  And that market size is hard
to figure out because the existence of many military programs is limited to
a need to know basis.

        While I'm not sure that ADJPO ought to fund any tool building, I think
that it would be useful for them to publish overall statistics on the amount
of JOVIAL code which is in current use, perhaps broken down into number of
projects in each of various size ranges.  In other words, some way for a privat
e
company to make a good business case that the market is large enough to justify
spending the $.  Same thing for CMS-2 code and perhaps FORTRAN code

Rich Hammond