Gnoga the "Killer App" for Ada

Gnoga the "Killer App" for Ada

[David Botton]

Q) First what is a "killer App"?

A) "a killer application is any computer program that is so necessary or desirable that it proves the core value of some larger technology" - wikipedia

-> Killer App = Have to Have App

Q)  Is Gnoga necessary or desirable enough to get people to notice and use Ada?

A) Before we can answer this, we have to answer desirable to whom?

Q)  So who is the audience we want to be looking at Ada?

Ada already has notice in a very small and specific niche of the software development world. That small niche is not the target of Gnoga (although they certainly would have major advantage using it for remote GUIs). They already have AdaCore actively marketing to them and the financial reward for long term contracts in that world guarantees AdaCore will continue to address that sector for years to come.

AdaCore has always been benevolent and contributed over time their hard work on GNAT to the FSF. The GNAT GPL version gets exposure for Ada for students and for AdaCore's potential customers, but it _hurts_ considerably in the Application space. Unless the Application developer is comfortable already and _familiar_ already with the fact that there are reliable FSF versions.

Understand AdaCore's thinking, (I was at the meeting that made the decision to make public GNAT version GPL only, I disagreed) it is logical for their company. The current market for Ada is their marketing dollars, if their customers can receive a solid unencumbered version of their product, why would they consider support. I have my arguments, but the reality is they were already seeing supported customers using the public versions instead of maintaining support contracts. The problem gets worse as the product gets better, paradox. Result though in the end is that the small app developer is turned off to Ada and the larger developers are more inclined to give their business to AdaCore, understood.

So who is the audience, the people that AdaCore does not market to and _can't_ market to with their current business model. They are Application developers (see next point) with minimal "compiler" support requirements and almost no long term support needs. There _is_ a business model for them and it will be something I will share in the future and I believe many experienced Ada programmers would consider working with me on it in the future, perhaps even AdaCore will consider investment or partnership, but that is another discussion. (I'm one of the few people to have actually made a living in the application space using Ada to do it, so I have what to share when we get there.)

-> Application Developers are the target of the Killer App

Q) What is the difference between Application Developers and Systems Developers?

A) Computer programmers often are grouped into two broad types: application programmers and systems programmers. Application programmers write programs to handle a specific job, such as a program to track inventory within an organization. They also may revise existing packaged software or customize generic applications which are frequently purchased from independent software vendors. Systems programmers, in contrast, write programs to maintain and control computer systems software, such as operating systems and database management systems. These workers make changes in the instructions that determine how the network, workstations, and CPU of the system handle the various jobs they have been given and how they communicate with peripheral equipment such as printers and disk drives. - wikipedia

-> Application developers develop "userland" products, Systems developers develop infrastructure (not a perfect summary, but close enough).

Q) Why target Application Developers for Ada?

A) The reason they should use Ada, everyone on CLA should know and perhaps another thread on future marketing materials for this group can explore more, however the question is why "target" them.

    How many apps on the App store? How many Apps called iOS?
    How many flight control systems? How many games with air planes?
    How many software drivers? How many programs that use them?
    Got the picture?

    Even .1% of the Application space would mean more Ada developers than the language has ever seen
in its lifetime.

There are very smart Application developers (some even as good looking as me too!) There are also some not so smart Application developers (many of them still spell better than I). In the Systems space there are for more people with the engineering backgrounds and education to appreciate Ada from the get go, and so we can understand why that has been where most uptake has been up till now and why they are the easier crowd to introduce Ada too, but they are a minuscule portion of the developers.

So targeting the Application developer is important because

*) Better tools means better quality which means a better world
*) They are the vast majority of people writing software and with numbers comes strength of jobs, products, and security for all.
*) Ada has so much to offer the application space, we need to share the love :)

Q) So how is Gnoga the Killer App for Ada?

1) Since the "bigger" picture is the Application developers
2) They need a reason to look at Ada and all its its own coolness
3) Gnoga with Ada offers the application developer a powerful toolset for secure cloud based computing, mobile apps, desktop apps and web apps the combination not found in any other set of tools in any other language :)

=> It is the "Killer App" for the "Killer Language"

Re: Gnoga the "Killer App" for Ada

[Thomas Løcke]

On 10/06/2014 02:46 AM, David Botton wrote:
> [snipped a bunch of wise words]
>


I wholeheartedly agree with your message David.

Gnoga has the potential to provide Ada with a significant boost,
especially if it works "out of the box" with FSF GNAT.

I sincerely hope Gnoga gains traction.

-- 
Thomas Løcke | thomas@12boo.net | http://12boo.net

Re: Gnoga the "Killer App" for Ada

[tonyg]

On Monday, 6 October 2014 06:42:36 UTC+1, Thomas Løcke  wrote:
> On 10/06/2014 02:46 AM, David Botton wrote:
> 
> > [snipped a bunch of wise words]
> 
> >
> 
> 
> 
> 
> 
> I wholeheartedly agree with your message David.
> 
> 
> 
> Gnoga has the potential to provide Ada with a significant boost,
> 
> especially if it works "out of the box" with FSF GNAT.
> 
> 
> 
> I sincerely hope Gnoga gains traction.
> 
> 
> 
> -- 
> 
> Thomas Lųcke | thomas@12boo.net | http://12boo.net

Its extremely well targeted, and very much the start of a major push, especially if an IDE is completed. I'm using it, and actually its the reason I can use Ada in a job I want done. My only criticisms are totally invalid since you only released it two weeks ago!

I think your objectives are very ambitious but clearly achievable, especially regarding your previous work.

I'm one of the not so smart application developers,  and its nice for you more able developers to help since my subgroup has extra time pressures brought on by sexual demand.

I'm looking forward to seeing how this turns out. I see only good things.

Re: Gnoga the "Killer App" for Ada

[Simon Wright]

tonyg <tonythegair@gmail.com> writes:

> I'm one of the not so smart application developers, and its nice for
> you more able developers to help since my subgroup has extra time
> pressures brought on by sexual demand.

Tell us more!

Re: Gnoga the "Killer App" for Ada

[David Botton]

>My only criticisms are totally invalid since you only released it two weeks ago!

Actually it has not been released yet, I just decided that I wanted people to have access to the code during development, to inspire them and me. We are still a few weeks away from a 1.0 for the API.

Tomorrow will be one month to the first git commit, which was about a week after I started.  A few things from Gnoga.Server are ported (and updated for recent GNATs) from a previous project used in production sites.

My point is, everything very much on schedule. If I stopped now there is enough there for extremely capable apps already.

I'll be back on the tutorials when I am done adding tab views / multi views and layout boxes.

It is important to me that laying out a GUI should be simpler than CSS, which is how it would be done right now.

David Botton

Re: Gnoga the "Killer App" for Ada

[Shark8]

On 10/6/2014 3:29 AM, David Botton wrote:
> It is important to me that laying out a GUI should be simpler than CSS,
> which is how it would be done right now.

Good choice, CSS is crap when it comes to real layout.
It would have been nicer if "they" had chosen something like PostScript 
for handling layout needs rather than hopping on the anemic styling to 
get some of what they wanted.

Re: Gnoga the "Killer App" for Ada

[David Botton]

> It would have been nicer if "they" had chosen something like PostScript 

Interestingly enough some of Gnoga's inspiration is NextStep and Cocoa that at their core render graphics using PS and PDF respectively.

David Botton

Re: Gnoga the "Killer App" for Ada

[dptrash]

Our business decided not to use Gnoga because of the license GPLv3. Any chances for BSD/MIT?

Re: Gnoga the "Killer App" for Ada

[Björn Lundin]

On 2014-10-08 09:22, dptrash@arcor.de wrote:
> Our business decided not to use Gnoga because of the license GPLv3. Any chances for BSD/MIT?
> 

Is not the exception - in practice - the same thing

This is from header of gnoga-gui-screen.ads (randomly picked)

"
As a special exception under Section 7 of GPL version 3, you are
granted additional permissions described in the GCC Runtime Library
Exception, version 3.1, as published by the Free Software Foundation.
"

and

"
As a special exception, if other files instantiate generics from this
unit, or you link this unit with other files to produce an executable,
this unit  does not  by itself cause  the resulting executable to be
covered by the GNU General Public License. This exception does not
however invalidate any other reasons why the executable file  might be
covered by the  GNU Public License.
"

It looks (to me) as just about the same as in gnat pro

--
Björn

Re: Gnoga the "Killer App" for Ada

[Simon Wright]

Björn Lundin <b.f.lundin@gmail.com> writes:

> On 2014-10-08 09:22, dptrash@arcor.de wrote:
>> Our business decided not to use Gnoga because of the license
>> GPLv3. Any chances for BSD/MIT?
>> 
>
> Is not the exception - in practice - the same thing
>
> This is from header of gnoga-gui-screen.ads (randomly picked)
>
> "
> As a special exception under Section 7 of GPL version 3, you are
> granted additional permissions described in the GCC Runtime Library
> Exception, version 3.1, as published by the Free Software Foundation.
> "
>
> and
>
> "
> As a special exception, if other files instantiate generics from this
> unit, or you link this unit with other files to produce an executable,
> this unit  does not  by itself cause  the resulting executable to be
> covered by the GNU General Public License. This exception does not
> however invalidate any other reasons why the executable file  might be
> covered by the  GNU Public License.
> "
>
> It looks (to me) as just about the same as in gnat pro

Not quite the same as GNAT Pro; AdaCore _replaced_ the second form (the
GMGPL) with the first form.

_I_ agree that businesses should be happy with the runtime library
exception; indeed, if not, they shouldn't use GCC at all because the C++
(and C) RTL is licensed on the same terms. But there's a lot of FUD
about the GPL.

Re: Gnoga the "Killer App" for Ada

[David Botton]

 
> Not quite the same as GNAT Pro; AdaCore _replaced_ the second form (the
> 
> GMGPL) with the first form.

Should be identical I took the language out of the AWS git repo that has the exception. I don't have GNAT Pro right now so I can't compare. My AdaCore credentials stopped working at some point and I haven't contacted them about it (I've been on and off contracting with them for projects throughout the years).

So if you can please check to see if language is identical would be appreciated.

Thanks
David Botton

Re: Gnoga the "Killer App" for Ada

[Björn Lundin]

On 2014-10-08 15:17, David Botton wrote:
>  
>> Not quite the same as GNAT Pro; AdaCore _replaced_ the second form (the
>>
>> GMGPL) with the first form.
> 
> Should be identical I took the language out of the AWS git repo that has the exception. I don't have GNAT Pro right now so I can't compare. My AdaCore credentials stopped working at some point and I haven't contacted them about it (I've been on and off contracting with them for projects throughout the years).
> 
> So if you can please check to see if language is identical would be appreciated.
> 
> Thanks
> David Botton
> 

from Gnat Pro 7.1.1 Win32


> ------------------------------------------------------------------------------
> --                                                                          --
> --                         GNAT RUN-TIME COMPONENTS                         --
> --                                                                          --
> --                           A D A . A S S E R T                            --
> --                                                                          --
> --                                 B o d y                                  --
> --                                                                          --
> --         Copyright (C) 2007-2009  Free Software Foundation, Inc.          --
> --                                                                          --
> -- GNAT is free software;  you can  redistribute it  and/or modify it under --
> -- terms of the  GNU General Public License as published  by the Free Soft- --
> -- ware  Foundation;  either version 3,  or (at your option) any later ver- --
> -- sion.  GNAT is distributed in the hope that it will be useful, but WITH- --
> -- OUT ANY WARRANTY;  without even the  implied warranty of MERCHANTABILITY --
> -- or FITNESS FOR A PARTICULAR PURPOSE.                                     --
> --                                                                          --
> -- As a special exception under Section 7 of GPL version 3, you are granted --
> -- additional permissions described in the GCC Runtime Library Exception,   --
> -- version 3.1, as published by the Free Software Foundation.               --
> --                                                                          --
> -- In particular,  you can freely  distribute your programs  built with the --
> -- GNAT Pro compiler, including any required library run-time units,  using --
> -- any licensing terms  of your choosing.  See the AdaCore Software License --
> -- for full details.                                                        --
> --                                                                          --
> -- GNAT was originally developed  by the GNAT team at  New York University. --
> -- Extensive contributions were provided by Ada Core Technologies Inc.      --
> --                                                                          --
> ------------------------------------------------------------------------------





-- 
--
Björn

Re: Gnoga the "Killer App" for Ada

[David Botton]

They just spell out the same exception in blunt terms with their product name. Perhaps I'll go back and add something like that, but is exactly the same thing license wise already.

So IF you use GNAT Pro THEN why not use Gnoga?

David Botton

Re: Gnoga the "Killer App" for Ada

[Simon Wright]

David Botton <david@botton.com> writes:

> They just spell out the same exception in blunt terms with their
> product name. Perhaps I'll go back and add something like that, but is
> exactly the same thing license wise already.
>
> So IF you use GNAT Pro THEN why not use Gnoga?

Or FSF GCC, for that matter. This example is from GCC 4.9.0 gnat.ads:

-- GNAT is free software;  you can  redistribute it  and/or modify it under --
-- terms of the  GNU General Public License as published  by the Free Soft- --
-- ware  Foundation;  either version 3,  or (at your option) any later ver- --
-- sion.  GNAT is distributed in the hope that it will be useful, but WITH- --
-- OUT ANY WARRANTY;  without even the  implied warranty of MERCHANTABILITY --
-- or FITNESS FOR A PARTICULAR PURPOSE.                                     --
--                                                                          --
-- As a special exception under Section 7 of GPL version 3, you are granted --
-- additional permissions described in the GCC Runtime Library Exception,   --
-- version 3.1, as published by the Free Software Foundation.               --
--                                                                          --
-- You should have received a copy of the GNU General Public License and    --
-- a copy of the GCC Runtime Library Exception along with this program;     --
-- see the files COPYING3 and COPYING.RUNTIME respectively.  If not, see    --
-- <http://www.gnu.org/licenses/>.                                          --
--                                                                          --
-- GNAT was originally developed  by the GNAT team at  New York University. --
-- Extensive contributions were provided by Ada Core Technologies Inc.      --

Re: Gnoga the "Killer App" for Ada

[David Botton]

 
> Or FSF GCC, for that matter. This example is from GCC 4.9.0 gnat.ads:

That looks identical but instead of "this library" spells out name.

Re: Gnoga the "Killer App" for Ada

[Simon Wright]

David Botton <david@botton.com> writes:

>  
>> Or FSF GCC, for that matter. This example is from GCC 4.9.0 gnat.ads:
>
> That looks identical but instead of "this library" spells out name.

This is true of the first para.

The second and fourth paras are identical.

The third para of the GNAT Pro version is

-- In particular,  you can freely  distribute your programs  built with the --
-- GNAT Pro compiler, including any required library run-time units,  using --
-- any licensing terms  of your choosing.  See the AdaCore Software License --
-- for full details.                                                        --

whereas the third para of the FSF version is

-- You should have received a copy of the GNU General Public License and    --
-- a copy of the GCC Runtime Library Exception along with this program;     --
-- see the files COPYING3 and COPYING.RUNTIME respectively.  If not, see    --
-- <http://www.gnu.org/licenses/>.                                          --

The GNAT Pro version is more explicit; the FSF language is the same as
that used in other FSF GCC libraries (and I suspect that AdaCore were
strongly encouraged to use it when the change to GPLv3 occurred).

Even the GNAT Pro language isn't that good. It seems to promise that
using the GNAT Pro compiler entitles you to ignore the licensing terms
of any library you choose to include in your proprietary code! I quite
like the idea of merging the GNAT Pro language of the third para above
with the GMGPL language in the fourth para of your Gnoga; that would,
for example, cover the situation for people who build Gnoga with GNAT
GPL.

Of course IANAL.


Robert Dewar used to claim (at the time when AdaCore were making their
public releses pure GPL but hadn't completely updated the source of
libraries like ASIS) that the copyright statements in source files were
irrelevant and overridden by the overall copyright statement in the
distribution. I think this claim may not have been accurate (otherwise
why would FSF require copyright statements in each file?)

Why GPL for GNAT hurt(s) the Application Space

[David Botton]

>>  I think this claim may not have been accurate (otherwise 
>> why would FSF require copyright statements in each file?) 

I think the entire GNAT GPL idea is suspect, however, I and I think most others are not about to challenge it. I had my vote at an AdaCore meeting against the idea and what was, is.

As I wrote, GNAT GPL _hurt_ and _hurts_ the Application programming space for Ada, even though for AdaCore it "may" have helped to stem the tide of customers leaving to use the public releases. Sure it is better that GNAT GPL exists than nothing. Since that would hurt all of Ada. Certainly I appreciate that it is AdaCore that fuels the development for these years of a free compiler. So don't get me wrong. I understand, appreciated and even like AdaCore.

However, I think worth making a _very clear_ statement if not for those working on http://gnuada.sourceforge.net/ (you all have my praise) and producing a Mac FSF Gnat version, I would never have bothered to write Gnoga, I would never bother with Ada public projects again, I would have just used another (unnamed language and dev environment) that I have used for numerous projects since being away from working on public Ada projects (although I have been doing private Ada work, even did some for AdaCore).

I had the idea for Gnoga even before websockets came out years back. I even had some tests running using alternate methods. The point is I didn't feel the desire anymore to try and promote or push Ada and use if it meant you were locked in to the GPL (despite the fact that if Ada was popular enough I'd feel the opposite about it!)

No one wants to waste their time and until I felt I had a viable tool suite that didn't have the GPL virus in it, Ada was out for public projects for me.

I even wrote and used for production an entire Rails for Ada, never bothered releasing it. (I've ported some of the code into Gnoga and more will come) in this in between time.

I am an Application developer. I've written systems stuff, some compiler stuff, etc., I just don't enjoy it.

I am a software _artist_ with engineering skills. Coding is meditation, magic and music.

I like to make noise and say look at this cool stuff, try this. I want to share my music, I want to share my enthusiasm for the mystical creation of new code.

So when I think that I can't share my code with anyone since no one will be able to use it (remember the Application space doesn't have the 10s of 1000s  for support contracts) why bother...

Sometimes I wonder if I had bother completing the few lines of code left to GNAVI for Windows if it would have been the VB killer, or if I ported GWindows to Linux and Mac bringing easy VB like dev everywhere in Ada, maybe not since Windows _died_ and the Web was born, but who knows.

So as long as I think there is an audience for my music, Gnoga rolls on :)

David Botton

yes I have a big head, yes I think too much of my work and what I am doing and what it could be, yes sometimes people think it means I think less of their work (never the case), but dreams make reality, and ambition produces results. So be it, I don't get paid to do this somethings got to motivate change :)

Re: Why GPL for GNAT hurt(s) the Application Space

[Simon Wright]

David Botton <david@botton.com> writes:

> However, I think worth making a _very clear_ statement if not for
> those working on http://gnuada.sourceforge.net/ (you all have my
> praise) and producing a Mac FSF Gnat version

Thanks!

Where can one get the Windows equivalent? (gnuada should provide it, but
I'm not in a position to work on it myself).

Re: Why GPL for GNAT hurt(s) the Application Space

[Björn Lundin]

On 2014-10-09 11:21, Simon Wright wrote:
> David Botton <david@botton.com> writes:
> 
>> However, I think worth making a _very clear_ statement if not for
>> those working on http://gnuada.sourceforge.net/ (you all have my
>> praise) and producing a Mac FSF Gnat version
> 
> Thanks!
> 
> Where can one get the Windows equivalent? (gnuada should provide it, but
> I'm not in a position to work on it myself).
> 

There is one in minGW

--
Björn

Re: Why GPL for GNAT hurt(s) the Application Space

[G.B.]

On 09.10.14 00:09, David Botton wrote:
> No one wants to waste their time and until I felt I had a viable tool suite that didn't have the GPL virus in it, Ada was out for public projects for me.

If there is a desire to not have that source based reciprocity
(however asymmetrical it may actually be) shouldn't this create
an opportunity for a compiler company to offer a judiciously priced
Ada 2005 compiler?
  That is, whenever the "sales department" can reliably determine that

- a $$$$ support contract won't make sense for this customer,

- support will, by agreement, be based on goodwill, or none,
   at the discretion of the company

- use will, by agreement and in fact not become freeloading
   (as has been the case, apparently, at AdaCore)

- no misdemeanor needs to be expected, like overzealously
   comparing compilers' features in public

they would offer a command line tool for less than VS Pro, say.

Re: Why GPL for GNAT hurt(s) the Application Space

[Simon Wright]

"G.B." <bauhaus@futureapps.invalid> writes:

> On 09.10.14 00:09, David Botton wrote:
>> No one wants to waste their time and until I felt I had a viable
>> tool suite that didn't have the GPL virus in it, Ada was out for
>> public projects for me.
>
> If there is a desire to not have that source based reciprocity
> (however asymmetrical it may actually be) shouldn't this create
> an opportunity for a compiler company to offer a judiciously priced
> Ada 2005 compiler?
>  That is, whenever the "sales department" can reliably determine that
>
> - a $$$$ support contract won't make sense for this customer,
>
> - support will, by agreement, be based on goodwill, or none,
>   at the discretion of the company
>
> - use will, by agreement and in fact not become freeloading
>   (as has been the case, apparently, at AdaCore)
>
> - no misdemeanor needs to be expected, like overzealously
>   comparing compilers' features in public
>
> they would offer a command line tool for less than VS Pro, say.

That really sounds like FSF GCC. Unless people will really take fright
at "GPLv3" without reading on to "with GCC Runtime Library Exception".

Re: Why GPL for GNAT hurt(s) the Application Space

[G.B.]

On 09.10.14 15:18, Simon Wright wrote:
> That really sounds like FSF GCC. Unless people will really take fright
> at "GPLv3" without reading on to "with GCC Runtime Library Exception".

FSF GCC still does not "feel good":

World views aside, and FUD aside, consider the lack of one
conveniently installable package like AWS on Windows™ producing
anger. Anger is a strong motive, as is convenience.

Also, on Windows™, GCC is not perceived to be native if it sits
on top of Cygwin or MingW. These and Windows™ itself require
licensing issues. It is less important, I thinkm how these are resolved
than how much convincing the resolution takes. (Robert Dewar,
on stage, explains how much time it had taken AdaCore to get
a company lawyer understand "You have the source" when the
lawyer asked for source code escrow. I'm mentioning this because
it isn't technical rationality that is at work in cases like these.)

I understand that packaging of an affordable "feel good" compiler
has failed to be a commercial success in the past. But if a
compiler maker sells a copy to a "cleared" individual, it should
help him free up some personal money, and help Ada be viable
outside the niche. Would this be so bad?


And another thing:
how long Apple is Apple going to tolerate non-LLVM compilers on Mac?

Re: Why GPL for GNAT hurt(s) the Application Space

[Simon Wright]

"G.B." <bauhaus@futureapps.invalid> writes:

> World views aside, and FUD aside, consider the lack of one
> conveniently installable package like AWS on Windows™ producing
> anger.

I will consider including AWS in the next Mac OS X release; but there
seem to be quite a lot of user-specific choices (LDAP? (which) SSL?), so
perhaps it's not a good idea. It takes about 5 minutes to build.

That was the GPL 2014 distro, which is of course pure GPL. Is the source
from the git repo pure GPL?

> Also, on Windows™, GCC is not perceived to be native if it sits
> on top of Cygwin or MingW. These and Windows™ itself require
> licensing issues.

I think it uses mingw (whose licensing terms are very open -
MIT-like). A program that used the cygwin libraries would have to be
GPL.

Not sure the users need install mingw themselves? (I mean, it could be
included in a .msi if someone were to produce one, so long as the mingw
licensing notice was included).

Windows licensing issues?

Re: Why GPL for GNAT hurt(s) the Application Space

[G.B.]

On 09.10.14 19:11, Simon Wright wrote:
> Windows licensing issues?

In general, you'd have to make sure that everything
to be distributed with an .exe is redistributable, i.e.,
does not include any MS code unless that's permitted
by a Microsoft EULA (wherever the latter is legal,
and applies).  (MS's *free* tools do not allow that kind
of free distribution, if that's what many here would
want. AFAIK.)

http://www.mingw.org/wiki/mingw

has good news WRT GCC, but also draws some attention to
MS's rights.

Re: Why GPL for GNAT hurt(s) the Application Space

[Pascal Obry]

Le jeudi 09 octobre 2014 à 16:43 +0200, G.B. a écrit : 
> World views aside, and FUD aside, consider the lack of one
> conveniently installable package like AWS on Windows™ producing
> anger. Anger is a strong motive, as is convenience.

First there is a Win32 script to build now. Second I refuse to bear the
non acceptance of Ada on me because of that :) And we are a whole
community, doing a script to build AWS has been proposed by no one, I
would have been happy to include it or help building it.

-- 
  Pascal Obry /  Magny Les Hameaux (78)

  The best way to travel is by means of imagination

  http://v2p.fr.eu.org
  http://www.obry.net

  gpg --keyserver keys.gnupg.net --recv-key F949BD3B

Re: Why GPL for GNAT hurt(s) the Application Space

[David Botton]

I don't think he meant to say that AWS was the reason :) I'd argue it will be reason people WILL take up Ada as more tools and packages are built on AWS and Gnoga.

Besides as I just asked nice and you had a script in hours prooof enough that is not the case. 

The idea is that Ada for small companies and projects is not so attractive with the current situation. Something that I think as a community we can change.

I have not approached AdaCore yet about the work I am doing and my roadmap. However while it has been a mixed bag of reception for many of my ideas perhaps they will see a viable business and community value in returning to a different license and direct support for some community projects that are not their core business. We will see.

Regardless I believe as I mentioned I have a good business model that can produce a sustainable business for serious Ada developers in the application space and enough to help fund non gpl encumbered Ada distros. 

David Botton

Re: Why GPL for GNAT hurt(s) the Application Space

[Luke A. Guest]

David Botton <david@botton.com> wrote:

> I think the entire GNAT GPL idea is suspect, however, I and I think most
> others are not about to challenge it. I had my vote at an AdaCore meeting
> against the idea and what was, is.
> 

I've challenged it many times. So you're not the only one. It really
doesn't help that GNAT is the only compiler available freely. Unlike other
languages.

I also don't consider different licensed GNAT's as different compilers,
they are the same just under different licenses and at different versions.

> As I wrote, GNAT GPL _hurt_ and _hurts_ the Application programming space for Ada, even 

There is no application space really, no mobile, no small embedded (hobby
or otherwise). David Howells has started a runtime project based on my work
on a bare metal runtime. 

I'm also working in stuff to try to promote more use of Ada but if this
fails it'll have been a massive waste of time and I'll have to start
looking for something better or try to design my own language.

Luke

Re: Why GPL for GNAT hurt(s) the Application Space

[tonyg]

On Wednesday, 8 October 2014 23:09:58 UTC+1, David Botton  wrote:
> >>  I think this claim may not have been accurate (otherwise 
> 
> >> why would FSF require copyright statements in each file?) 
> 
> 
> 
> I think the entire GNAT GPL idea is suspect, however, I and I think most others are not about to challenge it. I had my vote at an AdaCore meeting against the idea and what was, is.
> 
> 
> 
> As I wrote, GNAT GPL _hurt_ and _hurts_ the Application programming space for Ada, even though for AdaCore it "may" have helped to stem the tide of customers leaving to use the public releases. Sure it is better that GNAT GPL exists than nothing. Since that would hurt all of Ada. Certainly I appreciate that it is AdaCore that fuels the development for these years of a free compiler. So don't get me wrong. I understand, appreciated and even like AdaCore.
> 
> 
> 
> However, I think worth making a _very clear_ statement if not for those working on http://gnuada.sourceforge.net/ (you all have my praise) and producing a Mac FSF Gnat version, I would never have bothered to write Gnoga, I would never bother with Ada public projects again, I would have just used another (unnamed language and dev environment) that I have used for numerous projects since being away from working on public Ada projects (although I have been doing private Ada work, even did some for AdaCore).
> 
> 
> 
> I had the idea for Gnoga even before websockets came out years back. I even had some tests running using alternate methods. The point is I didn't feel the desire anymore to try and promote or push Ada and use if it meant you were locked in to the GPL (despite the fact that if Ada was popular enough I'd feel the opposite about it!)
> 
> 
> 
> No one wants to waste their time and until I felt I had a viable tool suite that didn't have the GPL virus in it, Ada was out for public projects for me.
> 
> 
> 
> I even wrote and used for production an entire Rails for Ada, never bothered releasing it. (I've ported some of the code into Gnoga and more will come) in this in between time.
> 
> 
> 
> I am an Application developer. I've written systems stuff, some compiler stuff, etc., I just don't enjoy it.
> 
> 
> 
> I am a software _artist_ with engineering skills. Coding is meditation, magic and music.
> 
> 
> 
> I like to make noise and say look at this cool stuff, try this. I want to share my music, I want to share my enthusiasm for the mystical creation of new code.
> 
> 
> 
> So when I think that I can't share my code with anyone since no one will be able to use it (remember the Application space doesn't have the 10s of 1000s  for support contracts) why bother...
> 
> 
> 
> Sometimes I wonder if I had bother completing the few lines of code left to GNAVI for Windows if it would have been the VB killer, or if I ported GWindows to Linux and Mac bringing easy VB like dev everywhere in Ada, maybe not since Windows _died_ and the Web was born, but who knows.
> 
> 
> 
> So as long as I think there is an audience for my music, Gnoga rolls on :)
> 
> 
> 
> David Botton
> 
> 
> 
> yes I have a big head, yes I think too much of my work and what I am doing and what it could be, yes sometimes people think it means I think less of their work (never the case), but dreams make reality, and ambition produces results. So be it, I don't get paid to do this somethings got to motivate change :)

Well I'm listening and enjoying it so far, and I think others are too... Rock on David

Re: Gnoga the "Killer App" for Ada

[David Botton]

> > Our business decided not to use Gnoga because of the license GPLv3. Any chances for BSD/MIT?


*** Did you look at the FAQ? I discuss this there.

If you want to consider a large donation to the project (I have a lot of plans for it) I'd be happy to give you any license you want for your company alone, although I think that silly (not the donation part) since you can use Gnoga in closed source projects.

However as pointed out this is __GMGPL v3__ not GPL v3 which means you can use in closed source applications, just if you extend / change the library itself _and_ you distribute your app beyond your walls, you need to make just _that_ source available to others. That is the reason I don't use BSD/MIT, I don't want a 3rd party company building on Gnoga with out sharing :) (Even though they could get away with it to a degree even with GMGPL)

All the code I've written for the community to date has been with GMGPL and they have been used in 100s of closed source project (no exaggeration on that number since I know of one company with over 100 alone using GWindows).

I'd be happy to discuss with anyone you like any issues regarding it. 

> It looks (to me) as just about the same as in gnat pro

It is identical to it.

David Botton

Re: Gnoga the "Killer App" for Ada

[Tero Koskinen]

8.10.2014 16:11, David Botton wrote:
>>> Our business decided not to use Gnoga because of the license 
>>> GPLv3. Any chances for BSD/MIT?
> 
> 
> *** Did you look at the FAQ? I discuss this there.
...
> However as pointed out this is __GMGPL v3__ not GPL v3 which means 
> you can use in closed source applications, just if you extend / 
> change the library itself _and_ you distribute your app beyond your 
> walls, you need to make just _that_ source available to others. That 
> is the reason I don't use BSD/MIT, I don't want a 3rd party company 
> building on Gnoga with out sharing :) (Even though they could get 
> away with it to a degree even with GMGPL)
> 
> All the code I've written for the community to date has been with 
> GMGPL and they have been used in 100s of closed source project (no 
> exaggeration on that number since I know of one company with over
> 100 alone using GWindows).

As a guy who releases most of his Ada code under BSD/MIT/ISC,
my experience has been that companies using open source Ada projects
with permissive licenses in their closed source projects are
willing to collaborate with the project and provide their patches
back to the project.

Maintaining a separate fork is a huge task and these days companies
tend to realise it.

Also, if you distribute your project under GPL or GMGPL, you should
actively do checks that companies using your code actually honor
the license, otherwise GPL/GMGPL is pretty pointless.

And if you trust that the companies comply with the license terms,
you most likely could trust that the same companies contribute
back to your BSD/MIT/ISC licensed projects.

Another point is that, as we have seen here, GPL'd version
of GNAT (GNAT GPL 20xx) has harmed the image of the Ada quite much.
The language gets a lot of bonus points if you can say
"You have this big amount of code under permissive BSD/MIT/ISC license
ready to be used in any project you wish."

Yours,
 Tero

Re: Gnoga the "Killer App" for Ada

[David Botton]

> Gnoga has the potential to provide Ada with a significant boost,
> 
> especially if it works "out of the box" with FSF GNAT.

It works on debian sid, no change
It works with GNAT FSF on Mac, no issue

I know that it is working on most people's window's installs, but that may be a project I or someone else will need to do, i.e. a working FSF windows package.

I am just as happy for there to by AdaCore GPL versions and it working on their installs as well. In principle I GPL, just Ada and even Ada and Gnoga are not going to be reason enough for most to restrict their licensing situations (and sometimes not even an option).

In particular I hope Gnoga gets AdaCore more business just as I hope it produces work for everyone else. Many larger companies will find they want AdaCore's model for their projects with Gnoga as well.

Most of us will find work with Gnoga using other models :) When ready will bring up again.

David Botton