Re: Current status of Ada?

[Adrian Hoe <abyhoe@gmail.com>]

Hi lou,

I agree with you that most of the links you can find on Ada websites
are dead, these includes some Ada links in my website. Honestly, I
don't have the time to make sure every links in my website are still
up and alive out there. That's the problem. Even some links from other
websites to my Ada projects page became dead after I moved my website
to Wordpress with some remake of uri. It is difficult to keep track of
what one links to and from.

About the compiler, AdaCore has contributed a free gnat compiler to
gnu. It is totally free but users have the option to choose to buy a
license or support seat from AdaCore for support and getting some
"deadly" compiler bugs eliminated.

The reason why compilers maker can't make their compiler totally free
is very reason behind the complex market. SImply, there is not enough
rewards for the compiler makers to open up entirely their compiler to
the free market.

Company like AdaCore and Aonix are major Ada compiler makers. They
have very steep market niche. Their customers are the world's most
advanced software developers and heavy industrial company, e.g.
Lockheed Martin, Boeing, BAE, and etc. Small and mid sized software
development company and one-man shop will not be an interesting profit
maker for AdaCore and Aonix and usually do not play very much
important role and can be neglected somehow.

Take AdaCore for instance, open up the entire compiler market for gnat
by giving free compiler has just become a commitment to the software
development community and not a profit generating activity. A one-man
shop (like me), does not very often need to deal with serious compiler
problems and bugs and most often will not encounter one. One-man shop
can get some of the design problem solved by posting questions to a
usenet like c.l.a. or searching the usenet for already there answers.

Recently, Aonix, another major player has opened up by offering free
compiler to Linux community. Aonix has been providing free compiler
(ObjectAda) to Windows platforms since many years ago. Aonix packages
the best IDE and Ada compiler suite into ObjectAda. But again, Aonix
is still not opening entire ObjectAda to the free market. Simply
because there will be no interest in profit generation. Aonix has a
better advantage over AdaCore is that Aonix offers other suites of
development tools such as StP, TeleUSE, and etc.

People, like me, who has been used to gnat will get lock into gnat
because of compiler specific packages like the gnat packages which are
not very well supported by ObjectAda. Otherwise, the Ada source can be
easily compiled by any Ada compilers. Aonix and AdaCore are just a tip
of the iceberg.

If you look into Apple's business model, you will find Apple is more
successful in open up their development tool to the free market. Apple
has been shipping its integrated development tool, xcode, since Mac OS
X. And recently, as I read from a mailing list, Apple has integrated
gnat 4.0 to its recent xcode release. I still have not get a
confirmation from the mailing list. Can anyone confirm the Apple's
integration? Developers can develop software using xcode for free.
That opens up a very interesting market for Apple. More and more very
nice applications, utilities and widgets have become available to Mac
users for free or for a fee. In return, Apple is rewarded with more
hardware sales. This business model is not applicable and will never
become viable to AdaCore, Aonix and as well as anyother Ada compiler
makers. Sadly to say that.

But on the other hand, you still can use gnat to develop software and
sell. How? You can sell services like, installation, maintenance,
support, training, upgrades and etc which come with your free
application. Or you can sell a hardware which runs a free software
developed using gnat compiler. Company like D-Link has been a
successful example. D-Link sells their routers which have open source
firewall and router software embedded inside their hardware. This way,
you will not violate any GPL or GPL related license which gnat is
released with. This is the whole new approach of software development
industry to generate profit.

In fact, we can see the future as software will be free for everyone.
You just have to pay for the services that you need. If you don't need
that, you just don't have to pay.
--
Adrian Hoe
http://adrianhoe.net



On Sep 17, 2:35 pm, lou <louisat...@bible-way.org> wrote:
> >The fact that Ada compiler vendors charged outragesous prices
> > for
> > their compilers helped to discourage commercial organizations from using Ada:
> > COBOL, C, C, Pascal, were more affordable.
> > Richard Riehle
>
> This is my very first day on this newsgroup, and I come here after
> about a month of frustration trying to learn more about Ada.
> As a newcomer, I think I can give an important insight as to what
> needs to be done to draw more people into using Ada. First,
> however, I need to say that I found Ada because I was looking for a
> language that supports unicode. It seems that Ada is one of
> very few languages that fully supports unicode. Is that true? If so,
> that is a point that needs to be emphasized. After reading
> some of the articles on the AdaCore website, I became very interested
> in Ada. What has impressed me about Ada: (a) Safety and
> reliability. I'm presently using two applications that have very
> elegant designs (from a user interface point of view), but are
> coded in inferior languages, and as a result have very serious memory
> leaks which their authors seem unable to find. These are not
> flight system control applications, but they are critical for me to
> get my work done. And it is very frustrating when they crash,
> especially if this happens right before a deadline. I wish they had
> been coded in Ada. (b) General purpose. Modern Ada seems to be
> general purpose language like C or C++ (but much safer) which can be
> used to write very serious commercial software. Since it is
> compiled it is fast and harder for someone to steal the source code.
> (c) From an engineering point of view, I like the modular
> structure of the language. The AdaCore website gave me these
> impressions, so I downloaded some free online Ada books and began to
> study, and my interest in Ada has become very serious. But I've
> encountered some serious roadblocks.
>
> So, what are those roadblocks, and what needs to be done to encourage
> more people to program in Ada?
>
> 1. A users group must be very easy to find. Every person or company
> with a web site about Ada should put a link to this users
> group.  It took me a month of doing many web searches to finally find
> an Ada e-mail list. Since subscribing I have not received a
> single post except for the welcome message!  Somehow I chanced to find
> this newsgroup yesterday, and just in time, as I was about
> to give up.  If there is no users group, I am not interested in
> devoting the time to studying the language. In my opinion, a good,
> friendly, helpful users group is very nearly as important as the
> language itself. No matter how good a language is, if I can't get
> help when I have problems, it does me no good.
>
> 2. All links to the Public Ada Library need to be fixed. AdaCore or
> someone should buy the name (url) that was used by the Public
> Ada Library, and make sure that all that source code is still
> available.  I was expecting to be able to find lots of source code
> that I could examine to see how real, working Ada programs are
> written.  Almost every Ada web site I visited had a link to the
> Public Ada Library, and not a link I tried worked. This really makes
> Ada look dead.
>
> 3. AdaCore needs to make their compiler available free of charge for
> commercial as well as non-profit and educational use. It
> seems to me that there is no way this would hurt AdaCore, and would in
> fact help a lot, as it would be the deciding factor to draw
> many new programmers to Ada. I'm still interested in Ada, as I have
> some non-profit uses for it. But, frankly, no small or medium
> size companies, or one man shops in their right minds are going to
> devote themselves to the time and study it takes to learn Ada
> once they learn that to use the compiler for Ada commercially will
> cost $14,000.00 for a one year license!  To be quite blunt,
> that is just being downright unrealistic, especially when there are
> many other languages available for free.  That is forcing people to
> use C++ instead. Mr. Riehle states the truth. OUTRAGEOUSLY PRICED
> COMPILERS IS THE BIGGEST BARRIER TO THE USE OF ADA. I'm still trying
> to decide if it is worth learning, since it appears I would not be
> able to sell any software produced, without paying for a license which
> is totally out of reach. CHARGE FOR SUPPORT NOT FOR THE COMPILER!  :>)
> That would make me happy anyway.
>
> The  big companies needing to produce safe software would be even more
> willing to use Ada and pay that high fee for trustworthy support if
> they knew there were more Ada programmers they could hire when needed.
> And if some of those small and medium sized companies and
> one man shops are given a chance, they may some day grow to be able to
> pay those high support fees too! Meanwhile, let them get
> their support on newsgroups like this one. Once they really start
> making money, they will be more than glad to pay high support
> fees for the professional support only a compiler maker can give. But
> with the compiler priced totally out of reach, there is no
> chance for them to come to that point. "Cast thy bread upon the
> waters: for thou shalt find it after many days."
>
> Other Ada compiler companies can replace "AdaCore" in the above
> paragraph with their own company's name. I predict that the first
> company to make a certified Ada compiler with IDE freely available for
> commercial use will soon dominate, as the result will be
> that eventually most Ada programmers will be experienced in using
> their compiler and IDE, and will go to them when they need safety
> critical
> support.
>
> 4. I'm rather hesitant to make this post, as it appears that it is
> going to expose my e-mail address to the whole world. If this is not
> the case it needs to be made clear. If it is the case that needs to be
> fixed and the fix made clear.