Re: employment with ada

["Marin David Condic" <mcondic.auntie.spam@acm.org>]

Simon Wright <simon@pushface.org> wrote in message
news:x7vof2hjoq4.fsf@smaug.pushface.org...
>
> Does realtime, embedded work involve GUIs? databases?
>
Not often, but if there were a few million Ada developers out there making
PC apps with Ada, nobody would object that they can't find Ada developers.
If there were a few million Ada/PC developers with good quality compilers
and support tools on the PC backing them up, people might be tempted to
start doing realtime/embedded work in Ada where they had a PC SBC as their
target. If Ada compiler vendors were making money hand over fist selling
PC-based compilers, they might be tempted to retarget them to other boards
for embedded use.

Success breeds success. At this point, I don't think it would be easy or
likely to get Ada adopted in most embedded efforts because of market
resistance and lack of adequate environments to do the job. Its a lot easier
to get it adopted in areas with less resistance and less demand for specific
tools. You don't need nearly as much to get a PC based app up and running so
Ada doesn't have to provide as much to be competitive in that market. If you
got acceptance in one market, its easier to then bleed over into other
markets. Start with what is achievable and work from there.


> How much safety-critical embedded avionics software gets written in C?
>
>
Lots.

We don't help ourselves if we delude ourselves into thinking that Ada is
some giant success out there in some invisible market that if only we could
lift the veil we'd see a multi-billion dollar business. Anything can be a
"success" if we define "success" narrowly enough. (Q: How many engine
controls for STOVL aircraft, designed in Palm Beach Gardens, Florida are
programmed in Ada? A: All of them. Does this make a difference to most Ada
developers looking for work?) Even if Ada had *all* the safety-critical
embedded avionics software development in the US, this wouldn't be a really
big market compared to the overall software market. Since that market is
small, the people making decisions about where to go with tools, etc., get
driven by costs and other factors to go use what the rest of the world is
using. Arguing that Ada is "better" for this kind of work is interesting,
but not compelling. Quality of the language is only one factor that has to
be considered. Making Ada more of a market force that didn't leave us poor
slobs building engine controls in P.B.G., FL feeling like we were off in a
desert somewhere might make the decision to use Ada an easier one.

I'm a big fan of Ada, but I think it needs to get off bottom-dead-center in
*some* market and start dominating it if it is going to have good long-term
prospects. "Realtime/Embedded" as a market looks like a tough nut to crack
and the Ada vendors don't have the budget to attempt to address it on
speculation. Some subset of the PC app market would be an easier target and
I think that developing libraries, GUIs, databases, etc., to suit some niche
in there would be a lot more doable.

MDC
--
======================================================================
Marin David Condic
I work for: http://www.belcan.com/
My project is: http://www.jsf.mil/

Send Replies To: m c o n d i c @ a c m . o r g

    "Going cold turkey isn't as delicious as it sounds."
        -- H. Simpson
======================================================================