Compiler business prospects

[Georg Bauhaus]

AP reports:
"White House officials say President Barack Obama will call on Congress on
Tuesday to pass legislation that, among other measures, would eliminate tax
rates on capital gains for investments in small businesses and extend for a
year the ability of all businesses to immediately deduct all of the costs of
equipment and software purchases."

Does this mean that more compilers will be sold during the coming months?

[Alan Jump]

Only if Congress passes it. A President can "call for" anything under the sun, but if Congress doesn't pass a bill to bring that into reality, it's just a call.

[an...@att.net]

You every head of ready-made projects

The changes would only help the general population software aka ready to
use, such as: accounting packages or other business related packages or
educational learning software.

Microsoft will get a boost, but for Ada or Linux nothing. Because Linux is
still free for the most part. And Ada software, is too specific or
design for a single user application.

[Gautier write-only]

On 31 jan, 16:56, a...@att.net wrote:
> You every head of ready-made projects
>
> The changes would only help the general population software aka ready to
> use, such as: accounting packages or other business related packages or
> educational learning software.
>
> Microsoft will get a boost, but for Ada or Linux nothing. Because Linux is
> still free for the most part. And Ada software, is too specific or
> design for a single user application.

Are you assuming one cannot make a single-user application in Ada ?
_________________________
Gautier's Ada programming
http://gautiersblog.blogspot.com/search/label/Ada
NB: follow the above link for a valid e-mail address

[Robert A Duff]

I can call for compilers from the vasty deep.
But will they come when I call?

- Bob

[Erich]

As a newbie who has started to appreciate Ada (after an initial
painful learning phase) I'd say that Ada doesn't need another compiler
but rather an *affordable* cross-platform GUI toolkit that can be used
for the development of commercial software.

By affordable I mean 100$ to 300$. I would love to use Ada for larger
standalone GUI applications but the main choices QtAda and GtkAda are
simply out of question for a hobby shareware developer like me. :(

[Yannick Duchêne (Hibou57)]

Someones don't agree, and I understand the reasons why… but still two
things: there is no GUI toolkit which could please everyone (just think
about portability issues: GDI/Windows, GTK/Ubuntu, Cocoa/Mac, Qt/BSD, and
some others), and there already is a handy to use and portable GUI
interface, which is a web page in a web browser (all major platform comes
with at least a modern browser). Unless you need real‑time graphics or
real‑time audio, a web page is a good material for a GUI (you will just
need a socket binding and some serialization to talk with the UI).

--
“Syntactic sugar causes cancer of the semi-colons.” [1]
“Structured Programming supports the law of the excluded muddle.” [1]
[1]: Epigrams on Programming — Alan J. — P. Yale University

[Gautier write-only]

If you want to program on Windows only, you have a nice toolkit here
http://sourceforge.net/projects/gnavi/
for 0$. You can make standalone executables (0 DLL!).
GWindows is used in freeware and business applications, so why not
shareware as well ?

[Stephen Leake]

I don't follow; both of those are available in free versions, and can be
used for commercial software.

If you are asking for _support_, rather than just for the source code,
that is a different matter; support is expensive, and rightly so.

If you are asking for non-open-source commercial software, that is also
a different matter.

Can you be more precise about what your problem is?

What sort of software are you hoping to make money from, and why are
unable to use QtAda or GtkAda for that?

--
-- Stephe

[Yannick Duchêne (Hibou57)]

Le Thu, 02 Feb 2012 10:45:40 +0100, Stephen Leake
<stephe...@stephe-leake.org> a écrit:

> Can you be more precise about what your problem is?
>
> What sort of software are you hoping to make money from, and why are
> unable to use QtAda or GtkAda for that?

At least about QtAda, as far as I know, it's OK for non‑open source.


> QtAda Professional Edition — distributed under terms of the
> GNAT Modified General Public License (GMGPL), suitable for
> commercial/proprietary software development. For more
> information please contact sa...@qtada.com.
Source: http://www.qtada.com/en/licensing.html

Providing the OP enjoy Qt (I personally don't, except for platforms where
it's the native toolkit).

[Erich]

> If you are asking for non-open-source commercial software, that is also
> a different matter.
> Can you be more precise about what your problem is?

Sure, but I'm not sure if it's wise, because presumably this has been
discussed already in the past. Look, I have nothing against the GPL
and am even planning to release some smaller projects under the GPL,
but the fact of the matter is that you *cannot* successfully sell
shareware or run a small business under the GPL because someone will
grab the source code and give his version away for free. That's just a
fact.

BTW, there is no real problem for me because I'm already using Racket
successfully for my GUIs. I was just a bit disappointed that there
seems to be no way to use Ada for that. Languages like the following
ones offer free (as in beer) LGPL bindings to GUI toolkits: C++, tcsl,
Perl, Ruby, Racket, FreePascal, ... just to name a few. For successful
shareware business the key is to provide executables for each major
platform that run immediately out of the box with a native and well-
crafted GUI. Toolkits like Qt make that possible but unfortunately Ada
is the only language I've seen so far where you have to pay for
bindings that allow non-GPL distributions.

> Source: http://www.qtada.com/en/licensing.html
> Providing the OP enjoy Qt (I personally don't, except for platforms where
> it's the native toolkit).

Yes, but how much is the GMPL license? I suppose the problem is that
QtAda must be based on the commercial version of Qt, because otherwise
you wouldn't be allowed to statically link to the Qt libraries, and of
course I understand when people want a return for their investment.

[Ludovic Brenta]

Erich <jo...@peppermind.com> writes:
>> If you are asking for non-open-source commercial software, that is
>> also a different matter. Can you be more precise about what your
>> problem is?
>
> Sure, but I'm not sure if it's wise, because presumably this has been
> discussed already in the past. Look, I have nothing against the GPL
> and am even planning to release some smaller projects under the GPL,
> but the fact of the matter is that you *cannot* successfully sell
> shareware or run a small business under the GPL because someone will
> grab the source code and give his version away for free. That's just a
> fact.

See http://fosdem.org/2012/schedule/event/keynotes_model_success

You *can* sell GPL software and make a profit. That's just a fact.

>> Source: http://www.qtada.com/en/licensing.html
>> Providing the OP enjoy Qt (I personally don't, except for platforms where
>> it's the native toolkit).
>
> Yes, but how much is the GMPL license? I suppose the problem is that
> QtAda must be based on the commercial version of Qt, because otherwise
> you wouldn't be allowed to statically link to the Qt libraries, and of
> course I understand when people want a return for their investment.

AFAIU, Qt has been licensed under LGPL since 2009.

--
Ludovic Brenta.

[Erich]

>
> Seehttp://fosdem.org/2012/schedule/event/keynotes_model_success

>
> You *can* sell GPL software and make a profit. That's just a fact.

I really didn't want to get into this discussion but I have to reply
to this, because your claim is completely misleading. The talk you
link to is about AdaCore. Sure you can make money with business-to-
business software, especially when it involves very special know-how,
but you do *not* make the money by selling the software. You are paid
for the consulting, product service, and maintenance and *not* for the
software.

I'm talking about making moderate amounts of money with non-business-
to-business, generic end-consumer software - where an executable is
deployed to Joe, the user, from a box on a shelf or via download and
payment by a payment service provider (Paypal, Kagi, etc.). This kind
of software does not require any consulting or services, because it
works out-of-the-box and does not require any skills to operate. For
example, I'm currently working on a word processing application with
unique features for creative writers such as style checking, "dark
room" minimal user interface, automatic backup and versioning,
automatic project management, extremely fast notetaking and search
functions, etc. The only way you could make money with that under GPL
was with a ridiculously large user base (>1 million, which means a few
people might pay for charity) or, say, by convincing the US
government to buy 10000 copies+5 year maintenance contract. <-- Joke

Anyway, I welcome it when other people make money with GPL software
but in many of not most cases, and certainly in mine, this is an
utterly unrealistic perspective.

> AFAIU, Qt has been licensed under LGPL since 2009.

LGPL does not allow static linking with the executable. I was assuming
that QtAda statically links with the Qt libraries, and anyway GMPL/
=LGPL, both of which seems to imply that the GMPL version of QtAda
must be based on the commercial version of Qt.

[Yannick Duchêne (Hibou57)]

Le Thu, 02 Feb 2012 19:19:21 +0100, Erich <jo...@peppermind.com> a écrit:

>
>>
>> Seehttp://fosdem.org/2012/schedule/event/keynotes_model_success
>>
>> You *can* sell GPL software and make a profit. That's just a fact.
>
> I really didn't want to get into this discussion but I have to reply
> to this,

Sorry for beeing a bit out of topic for this Usenet, just wanted to give
you some lecture. Seems a Wikipedia page already discussed the topic:
http://en.wikipedia.org/wiki/Commercial_open_source_applications
May be worth reading for people joining either yours or Ludodiv's opinion.

[Erich]

> Sorry for beeing a bit out of topic for this Usenet, just wanted to give
> you some lecture. Seems a Wikipedia page already discussed the topic:http://en.wikipedia.org/wiki/Commercial_open_source_applications
> May be worth reading for people joining either yours or Ludodiv's opinion.

The page confirms what I said. All programs on this web page are
business-to-business solutions -- without a single exception!

Anyway, I really didn't want to get into this topic and want to
apologize to other readers of this group. I can and will still use Ada
for backends and time critical code.

[Tero Koskinen]

On Wed, 1 Feb 2012 04:24:27 -0800 (PST)
Erich <jo...@peppermind.com> wrote:
> By affordable I mean 100$ to 300$. I would love to use Ada for larger
> standalone GUI applications but the main choices QtAda and GtkAda are
> simply out of question for a hobby shareware developer like me. :(

You could buy Janus/Ada Personal edition with Claw from
RR Software for $295. (http://www.rrsoftware.com/)

It is only for Windows, but shouldn't matter that much since Linux
people don't pay for software anyway and Apple wants everything on OS X
to be coded in Objective-C. ;)

As a bonus, you can make your code open, since not that many has
Janus/Ada and CLAW and those who have most likely are not interested
in forking your project and making money for themselves.

--
Tero Koskinen - http://iki.fi/tero.koskinen/

[Shark8]

On Feb 3, 1:11 pm, Tero Koskinen <tero.koski...@iki.fi> wrote:
> On Wed, 1 Feb 2012 04:24:27 -0800 (PST)
>

+1 for Applied Security Through Obscurity. ;)

[an...@att.net]

What I was getting to with the single user concept is. Where are the
businness or general public application written in Ada that the end
user can buy at their local online software store.

Also:

Programmers do not want to spend their life re-writing code written in
Ada. That is, every new language release requires 80% to 95% of the Ada
code to be rewritten. Even the yearly release of Adacore's GNAT can have
so many updates that it can cause a complete rewrite of an application
even between language releases.

And Ada is still concurrent but the programmer of today require
multiple-thread and multiple cpu languages. Ada 2012 has some multiple-cpu
controls but those type of features are still very limited. In truth, Ada 95
should of had been the first true multiple-thread and multiple cpu language.
And until these two systems are fully functional and I would say even
going beyond all other languages Ada programming will stay crippled to a
limited number of concurrent applications.

Then you have features that are included in other languages, but either
not available or not maintain in Ada. Such as links for OS graphics and
direct hardware devices such as USB ports. Yes, there are binding for
SDL and openGL but are those current. And still most programmers would
prefer Windows DirectX graphics, or Apple's graphics.

As for Ada, a plus would be a common graphics package that the programmer
could use to create an application on one system and then just compile
for another system without modify the code. Yet, today no package exist!

So, without these and other features added to Ada, no governmental tax or
altered law will increase Ada compiler being sold.

[Yannick Duchêne (Hibou57)]

Le Sat, 04 Feb 2012 04:43:03 +0100, <an...@att.net> a écrit:
> Programmers do not want to spend their life re-writing code written in
> Ada. That is, every new language release requires 80% to 95% of the Ada
> code to be rewritten.

Are you talking about the Ada which is the topic of this Usenet ?

On the contrary, all languages release are done with backward
compatibility in mind (Randy often referred to this requirement here at
comp.lang.ada). The Reference Manual always points potential
incompatibilities, and these are not legions.

[AdaMagica]

On 4 Feb., 04:43, a...@att.net wrote:
> Programmers do not want to spend their life re-writing code written in
> Ada. That is, every new language release requires 80% to 95% of the Ada
> code to be rewritten. Even the yearly release of Adacore's GNAT can have
> so many updates that it can cause a complete rewrite of an application
> even between language releases.

Again some incredible junk from anon. I rehosted Ada 83 embedded
software (several hundred thousand LoC) to different hardware with Ada
95, then moved on to Ada 2005, without much ado.

Seems like this guy is the worst programmer ever on earth.

[Stephen Leake]

Erich <jo...@peppermind.com> writes:

>> If you are asking for non-open-source commercial software, that is also
>> a different matter.
>> Can you be more precise about what your problem is?
>
> Sure, but I'm not sure if it's wise, because presumably this has been
> discussed already in the past. Look, I have nothing against the GPL
> and am even planning to release some smaller projects under the GPL,
> but the fact of the matter is that you *cannot* successfully sell
> shareware or run a small business under the GPL because someone will
> grab the source code and give his version away for free. That's just a
> fact.

That's what the music industry said about CDs, until Apple proved them
wrong with iTunes.

If you are selling a useful product (for instance an Android app), the
majority of people will be willing to pay something for it, rather than
use an illegal version. Hmm, I guess "illegal" is the wrong word here;
if you are using GPL, the free version will not be illegal.

You have to sell it at a price they are willing to pay. You may have to
offer support, or upgrades, in addition. You have to advertise to get
people's attention.

The people offering a free product will not be able to offer either
support or upgrades, nor will they be able to afford advertising.

I have no idea what the license is for most of the apps I've installed
on my Android; even if they are GPL, I can't be bothered to get the
source and compile it. I'd far rather pay someone else to do that.

> BTW, there is no real problem for me because I'm already using Racket
> successfully for my GUIs. I was just a bit disappointed that there
> seems to be no way to use Ada for that. Languages like the following
> ones offer free (as in beer) LGPL

Your issue is free as in freedom, not free as in beer; you do not pay
for the GUI bindings. The issue is what freedoms you have regarding
distribution when you use them.

>> Source: http://www.qtada.com/en/licensing.html
>> Providing the OP enjoy Qt (I personally don't, except for platforms where
>> it's the native toolkit).
>
> Yes, but how much is the GMPL license?

I suggest you email sa...@qtada.com and find out! It might well be worth
it for your business.

> I suppose the problem is that QtAda must be based on the commercial
> version of Qt, because otherwise you wouldn't be allowed to statically
> link to the Qt libraries, and of course I understand when people want
> a return for their investment.

The "problem" is that people need to eat, so they need to get paid, same
as you. If you are making money by using other people's work, it is only
reasonable that they get a cut. Maintaining QtAda is work!

--
-- Stephe

[Stephen Leake]

Erich <jo...@peppermind.com> writes:

>>
>> Seehttp://fosdem.org/2012/schedule/event/keynotes_model_success
>>
>> You *can* sell GPL software and make a profit. That's just a fact.
>
> I really didn't want to get into this discussion but I have to reply
> to this, because your claim is completely misleading. The talk you
> link to is about AdaCore. Sure you can make money with business-to-
> business software, especially when it involves very special know-how,
> but you do *not* make the money by selling the software. You are paid
> for the consulting, product service, and maintenance and *not* for the
> software.

+2

Ludovic, you were spreading FUD.

--
-- Stephe

[Stephen Leake]

Qt is _not_ QtAda

--
-- Stephe

[Stephen Leake]

Tero Koskinen <tero.k...@iki.fi> writes:

> On Wed, 1 Feb 2012 04:24:27 -0800 (PST)
> Erich <jo...@peppermind.com> wrote:
>> By affordable I mean 100$ to 300$. I would love to use Ada for larger
>> standalone GUI applications but the main choices QtAda and GtkAda are
>> simply out of question for a hobby shareware developer like me. :(
>
> You could buy Janus/Ada Personal edition with Claw from
> RR Software for $295. (http://www.rrsoftware.com/)
>
> It is only for Windows, but shouldn't matter that much since Linux
> people don't pay for software anyway and Apple wants everything on OS X
> to be coded in Objective-C. ;)

You left out the Android market. Which is Linux way down underneath, but
most people don't care about that.

I tried to import the Android GUI into Ada, but GnatJVM isn't quite up
to that, and AdaCore doesn't see enough market to support fixing it.

So you are stuck with the free Java GUI.

> As a bonus, you can make your code open, since not that many has
> Janus/Ada and CLAW and those who have most likely are not interested
> in forking your project and making money for themselves.

That's a nice angle :). I can just see an add for CLAW:

Keep your software safe! Use our totally obscure system to compile
it!

:)

--
-- Stephe

[Ludovic Brenta]

On Feb 5, 12:23 am, Stephen Leake <stephen_le...@stephe-leake.org>
wrote:
> Erich <j...@peppermind.com> writes:
>>> See http://fosdem.org/2012/schedule/event/keynotes_model_success

No. The FUD is in thinking *a priori* that this business model is
"impossible" and presenting that as a "fact". What I said was not
FUD, it was perfectly correct and it is a proven fact that one can
sell libre software and make a living, like AdaCore has proven for
decades. However this is not true in all cases as Robert Dewar
himself admitted during his keynote speech at FOSDEM this Saturday.
One case where this business model is not sustainable is when the
product is expensive to make but must be sold at a low price; this is
true of movies and video games among other things.

It was not obvious from the OP that he was envisioning selling cheap
software for consumers. The fact that he wants to run a small company
does not imply that the products inexpensive; I myself have worked and
still work in a small company selling expensive products and services
to big corporate customers. I believe this is in fact the case of
many regulars of comp.lang.ada.

More generally, I personally believe that small companies cannot
compete by selling cheap products in the mass market and that this
applies not only to software but to pretty much everything; this is
because the key to succeeding in the mass market is to control
distribution channels (i.e. retail stores). This is much more work
than a small company can handle and a one-person company wants to
spend their time developing, not marketing. A small company that goes
to the consumer market usually ends up being controlled (and
pressured) by their distributors. It is easy to see this in bicycle
shops and their suppliers: the small ones thrive and survive only
because they sell expensive high-end products, expert advice and
premium customer support, thereby avoiding the mass market. Mind you,
these are not "facts" but only my perception.

--
Ludovic Brenta.

[Ludovic Brenta]

Stephen Leake wrote on comp.lang.ada:

> Ludovic Brenta <ludo...@ludovic-brenta.org> writes:
>>>> Source:http://www.qtada.com/en/licensing.html
>>>> Providing the OP enjoy Qt (I personally don't, except for platforms where
>>>> it's the native toolkit).
>>>
>>> Yes, but how much is the GMPL license? I suppose the problem is that
>>> QtAda must be based on the commercial version of Qt, because otherwise
>>> you wouldn't be allowed to statically link to the Qt libraries, and of
>>> course I understand when people want a return for their investment.
>>
>> AFAIU, Qt has been licensed under LGPL since 2009.
>
> Qt is _not_ QtAda

Correct but irrelevant. If you want to program using QtAda, you're
going to have to link your program against both QtAda and the
underlying Qt (and possibly other libraries too), therefore you must
consider how the licenses of both Qt and QtAda (and the other
libraries, if any) affect your program. Qt is under LGPL at no cost.
QtAda uses the classical dual-licensing scheme: GPL at no cost, GMGPL
at extra cost.

--
Ludovic Brenta.

[Yannick Duchêne (Hibou57)]

Le Mon, 06 Feb 2012 11:40:05 +0100, Ludovic Brenta
<lud...@ludovic-brenta.org> a écrit:

> No. The FUD is in thinking *a priori* that this business model is
> "impossible" and presenting that as a "fact".

One talks about an absolute appreciation, the other about a practical
appreciation.

It just happens that although possible, this is hardly done and unlikely
to work for most. People arguing that's impossible are wrong about the
whole, true, while people arguing that's possible, are wrong about
effectiveness too, when they forget to provide any kind of evaluation
(like providing probability of success, which is known to be very‑very low
with that model).

It happens too, most people complaining about that fact, will fall in the
common case where it's not possible. If that was more likely possible,
believe people who mean they would be the first glad to go for the
adventure (I'm of these ones too :-P ).

May be you are both right and both wrong and precision was lacking (Ada
designers failing to provide relevant specifications, what a shame :-(
... just teasing :-P ).

[Yannick Duchêne (Hibou57)]

Le Mon, 06 Feb 2012 11:40:05 +0100, Ludovic Brenta
<lud...@ludovic-brenta.org> a écrit:

> More generally, I personally believe that small companies cannot
> compete by selling cheap products in the mass market and that this
> applies not only to software but to pretty much everything; this is
> because the key to succeeding in the mass market is to control
> distribution channels (i.e. retail stores).

That's a clever analysis.

Will keep it as a note somewhere.