Weakness of Ada is expensive / security / etc. ? Anything else?

Weakness of Ada is expensive / security / etc. ? Anything else?

[Trish Cayetano]

Hi, here is what I have noted as strengths and weaknesses of Ada. Let me know for any corrections and if you have anything to add... Thank you!

Strength and weaknesses of the language 
Strength
-	Semantics - well defined even in error situations
-	Strong typing - can be used to reduce the scope (and cost) of analysis
-	Quality - successfully used in many high integrity applications 

Weaknesses
-	Ada complete language is too complex. True or False? If true, how?
-	Ada requires expensive compilers for large systems 
-	Ada lacks that web presence greatly. comp.lang.ada is the only location on the internet with an active Ada community.
-	How is security a weakness? (Strength can be a weakness too?)

Re: Weakness of Ada is expensive / security / etc. ? Anything else?

[Simon Clubley]

On 2015-07-17, Trish Cayetano <trishacayetano@gmail.com> wrote:
> Hi, here is what I have noted as strengths and weaknesses of Ada. Let me know for any corrections and if you have anything to add... Thank you!
>
> Strength and weaknesses of the language 
> Strength
> -	Semantics - well defined even in error situations
> -	Strong typing - can be used to reduce the scope (and cost) of analysis
> -	Quality - successfully used in many high integrity applications 
>
> Weaknesses
> -	Ada complete language is too complex. True or False? If true, how?

The Ada language is also nicely modular. This means that you only need to
use the features which are appropriate to the problem at hand. It also
means you can write full Ada programs in a core subset of the language
without having to learn all it's features.

> -	Ada requires expensive compilers for large systems 

That's an incredibly sore point in some parts around here. You are
hereby warned about some of the responses you might get. :-)

For the serious response to this, the expense of the compiler depends
on the support required and the licence conditions of your code. If you
are willing to release the source code for any redistributed binaries
under the GPL more options open up.

However, this one item by itself puts Ada at a major disadvantage when
compared to other options such as C and C++.

> -	Ada lacks that web presence greatly. comp.lang.ada is the only location on the internet with an active Ada community.

There are various resources scattered around the web, but comp.lang.ada
does appear to be the only major place for active discussions (apart
from some discussions on Stack Overflow).

> -	How is security a weakness? (Strength can be a weakness too?)

Is this homework ? (the question seems to be worded in a strange way
and I don't really understand what you are trying to ask here).

Simon.

-- 
Simon Clubley, clubley@remove_me.eisner.decus.org-Earth.UFP
Microsoft: Bringing you 1980s technology to a 21st century world

Re: Weakness of Ada is expensive / security / etc. ? Anything else?

[Trish Cayetano]

On Friday, July 17, 2015 at 6:48:00 PM UTC+8, Simon Clubley wrote:
> On 2015-07-17, Trish Cayetano <trishacayetano@gmail.com> wrote:
> > Hi, here is what I have noted as strengths and weaknesses of Ada. Let me know for any corrections and if you have anything to add... Thank you!
> >
> > Strength and weaknesses of the language 
> > Strength
> > -	Semantics - well defined even in error situations
> > -	Strong typing - can be used to reduce the scope (and cost) of analysis
> > -	Quality - successfully used in many high integrity applications 
> >
> > Weaknesses
> > -	Ada complete language is too complex. True or False? If true, how?
> 
> The Ada language is also nicely modular. This means that you only need to
> use the features which are appropriate to the problem at hand. It also
> means you can write full Ada programs in a core subset of the language
> without having to learn all it's features.
> 
> > -	Ada requires expensive compilers for large systems 
> 
> That's an incredibly sore point in some parts around here. You are
> hereby warned about some of the responses you might get. :-)
> 
> For the serious response to this, the expense of the compiler depends
> on the support required and the licence conditions of your code. If you
> are willing to release the source code for any redistributed binaries
> under the GPL more options open up.
> 
> However, this one item by itself puts Ada at a major disadvantage when
> compared to other options such as C and C++.
> 
> > -	Ada lacks that web presence greatly. comp.lang.ada is the only location on the internet with an active Ada community.
> 
> There are various resources scattered around the web, but comp.lang.ada
> does appear to be the only major place for active discussions (apart
> from some discussions on Stack Overflow).
> 
> > -	How is security a weakness? (Strength can be a weakness too?)
> 
> Is this homework ? (the question seems to be worded in a strange way
> and I don't really understand what you are trying to ask here).
> 
> Simon.
> 
> -- 
> Simon Clubley, clubley@remove_me.eisner.decus.org-Earth.UFP
> Microsoft: Bringing you 1980s technology to a 21st century world

Thank you so much for your response, Simon. 
The security question is my personal question. Since Ada is known to be time-tested, safe, secure and reliable, is there an instance where these could be compromised? 

Re: Weakness of Ada is expensive / security / etc. ? Anything else?

[David Botton]

> The security question is my personal question. Since Ada is known to be time-tested, safe, secure and reliable, is there an instance where these could be compromised?

All security can be compromised. You can only reduce the chances and Ada is one tool amongst others that helps decrease those chances.

David Botton

Re: Weakness of Ada is expensive / security / etc. ? Anything else?

[G.B.]

On 17.07.15 12:19, Trish Cayetano wrote:

> Weaknesses
> -	Ada complete language is too complex. True or False? If true, how?
> -	Ada requires expensive compilers for large systems

Assuming this is for PC style or server based programs,
comparing to Intel or Microsoft without MSDN,
you'd be near $$$ for Ada, too, if Atego still sells a license
the way they once did, for use with smaller projects and moderate
amounts of support only. RR Software's price list is online.

 From what I hear, large systems support for Ada would actually
be cheaper than that first class support you can allegedly
buy from Apple, or Microsoft, or Oracle (at $$,$$$.-).

> -	Ada lacks that web presence greatly. comp.lang.ada is the only location on the internet with an active Ada community.

Well put stackoverflow questions get decent answers,
if not superb answers. Do you need a sticker saying SocialAda
on your phone? What's a large system, then?

> -	How is security a weakness? (Strength can be a weakness too?)

It takes brains and time to write reliable code, perhaps
a little less time if the language lends itself well to it.

Re: Weakness of Ada is expensive / security / etc. ? Anything else?

[Trish Cayetano]

On Friday, July 17, 2015 at 8:06:33 PM UTC+8, G.B. wrote:
> On 17.07.15 12:19, Trish Cayetano wrote:
> 
> > Weaknesses
> > -	Ada complete language is too complex. True or False? If true, how?
> > -	Ada requires expensive compilers for large systems
> 
> Assuming this is for PC style or server based programs,
> comparing to Intel or Microsoft without MSDN,
> you'd be near $$$ for Ada, too, if Atego still sells a license
> the way they once did, for use with smaller projects and moderate
> amounts of support only. RR Software's price list is online.
> 
>  From what I hear, large systems support for Ada would actually
> be cheaper than that first class support you can allegedly
> buy from Apple, or Microsoft, or Oracle (at $$,$$$.-).
> 
> > -	Ada lacks that web presence greatly. comp.lang.ada is the only location on the internet with an active Ada community.
> 
> Well put stackoverflow questions get decent answers,
> if not superb answers. Do you need a sticker saying SocialAda
> on your phone? What's a large system, then?
> 
> > -	How is security a weakness? (Strength can be a weakness too?)
> 
> It takes brains and time to write reliable code, perhaps
> a little less time if the language lends itself well to it.

G.B.,
Thanks for your inputs!

Re: Weakness of Ada is expensive / security / etc. ? Anything else?

[David Botton]

> Weaknesses
> -	Ada complete language is too complex. True or False? If true, how?

Your kidding right? Ada is readable and except for the ridiculous OO syntax fairly intuitive and easier to grasp the big picture than most languages. Have you looked at C++ lately? Or the new king of complex (almost everything can be express in like 10 freakishly different ways Nim?

> -	Ada requires expensive compilers for large systems 

FSF GNAT is a solid free open source professional compiler. Support services available from AdaCore for large or critical systems is _reasonable_ for the market (the fact that they try and hamper the other markets doesn't change their quality of service and reasonable pricing for their niche market.)

> -	Ada lacks that web presence greatly. comp.lang.ada is the only location on the internet with an active Ada community.

Having a central source is a strength not a weakness. However there are actually 4 active discussion groups for Ada, not including OpenStack, etc. Comp.Lang.Ada (The Ada Community Home), #Ada on FreeNode (very active cool people), Google Groups Ada, and actually the Linkedin Ada group.

Check http://GetAdaNow.com, http://LearnAdaNow.com and http://JoinAdaNow.com for more resources.

> -	How is security a weakness? (Strength can be a weakness too?)

You tell me..... You are the one asking the questions.

David Botton

Re: Weakness of Ada is expensive / security / etc. ? Anything else?

[Trish Cayetano]

On Friday, July 17, 2015 at 8:54:14 PM UTC+8, David Botton wrote:
> > Weaknesses
> > -	Ada complete language is too complex. True or False? If true, how?
> 
> Your kidding right? Ada is readable and except for the ridiculous OO syntax fairly intuitive and easier to grasp the big picture than most languages. Have you looked at C++ lately? Or the new king of complex (almost everything can be express in like 10 freakishly different ways Nim?
> 
> > -	Ada requires expensive compilers for large systems 
> 
> FSF GNAT is a solid free open source professional compiler. Support services available from AdaCore for large or critical systems is _reasonable_ for the market (the fact that they try and hamper the other markets doesn't change their quality of service and reasonable pricing for their niche market.)
> 
> > -	Ada lacks that web presence greatly. comp.lang.ada is the only location on the internet with an active Ada community.
> 
> Having a central source is a strength not a weakness. However there are actually 4 active discussion groups for Ada, not including OpenStack, etc. Comp.Lang.Ada (The Ada Community Home), #Ada on FreeNode (very active cool people), Google Groups Ada, and actually the Linkedin Ada group.
> 
> Check http://GetAdaNow.com, http://LearnAdaNow.com and http://JoinAdaNow.com for more resources.
> 
> > -	How is security a weakness? (Strength can be a weakness too?)
> 
> You tell me..... You are the one asking the questions.
> 
> David Botton

Thanks for the feedback and helpful links, David!

Re: Weakness of Ada is expensive / security / etc. ? Anything else?

[Patrick Noffke]

On Friday, July 17, 2015 at 7:54:14 AM UTC-5, David Botton wrote:
> > -	Ada lacks that web presence greatly. comp.lang.ada is the only location on the internet with an active Ada community.
> 
> Having a central source is a strength not a weakness. However there are actually 4 active discussion groups for Ada, not including OpenStack, etc. Comp.Lang.Ada (The Ada Community Home), #Ada on FreeNode (very active cool people), Google Groups Ada, and actually the Linkedin Ada group.
> 
> Check http://GetAdaNow.com, http://LearnAdaNow.com and http://JoinAdaNow.com for more resources.
> 

Here is an excellent online resource:
https://en.wikibooks.org/wiki/Ada_Programming

Re: Weakness of Ada is expensive / security / etc. ? Anything else?

[Shark8]

On Friday, July 17, 2015 at 4:19:30 AM UTC-6, Trish Cayetano wrote:
> Hi, here is what I have noted as strengths and weaknesses of Ada. Let me know for any corrections and if you have anything to add... Thank you!
> 
> Strength and weaknesses of the language 
> Strength
> -	Semantics - well defined even in error situations

And the "Bounded Error" option in the spec. (Which is essentially a step between "well-defined and specified behavior" and "undefined behavior".)

> -	Strong typing - can be used to reduce the scope (and cost) of analysis

This should not be underestimated; you can protect (e.g.) a database against bad input with strong typing; you can enforce validity-checks and formatting with it too.

> Weaknesses
> -	Ada complete language is too complex. True or False? If true, how?

False; while it's a 'big' language it's not difficult and full of gotchas -- if you use it regularly (and 'idiomatically') for a couple of years you're not going to be surprised by much... whereas with C++ you constantly see experts with decade-plus experience being surprised.

> -	Ada requires expensive compilers for large systems 

Semi-false -- This depends on what you're doing. FSF-GNAT is free, but comes with GPL; AdaCore-GNAT is also free but comes with a DIFFERENT GPL. Others like RR-Software or IBM's (they bought Rational) will have a different license.

> -	Ada lacks that web presence greatly. comp.lang.ada is the only location on the internet with an active Ada community.

Semi-true -- others have given places where the Ada programmers hang out; but these places usually have an excellent signal to noise ratio. (There could stand to be more tutorials and such.)

> -	How is security a weakness? (Strength can be a weakness too?)

It's not. Security is something you have to consider while designing the program; and having all the security in the world available means absolutely nothing if you don't use it. -- Ada *does* make a lot of security-related things easier with its strong-typing and, funnily enough, with its timing mechanisms. (You can [theoretically] eliminate the timing side-channel attack by marking the "time+maximal operation duration", do the operation, execute a "delay until X" [where X is the calculated time] and now the timing side-channel doesn't exist because that operation executes with constant time.)

Re: Weakness of Ada is expensive / security / etc. ? Anything else?

[Simon Clubley]

On 2015-07-17, Shark8 <onewingedshark@gmail.com> wrote:
> On Friday, July 17, 2015 at 4:19:30 AM UTC-6, Trish Cayetano wrote:
>> -	Ada requires expensive compilers for large systems 
>
> Semi-false -- This depends on what you're doing. FSF-GNAT is free,
> but comes with GPL; AdaCore-GNAT is also free but comes with a
> DIFFERENT GPL. Others like RR-Software or IBM's (they bought Rational)
> will have a different license.
>

In the case of FSF GNAT, the GPL applies to the compiler toolchain source
only; the RTL has the GMGPL exception (or whatever it's called these days).

Simon.

-- 
Simon Clubley, clubley@remove_me.eisner.decus.org-Earth.UFP
Microsoft: Bringing you 1980s technology to a 21st century world

Re: Weakness of Ada is expensive / security / etc. ? Anything else?

[Jeffrey R. Carter]

I'll stick my nose in here where I think I have something different to say from
the responses so far.

On 07/17/2015 03:19 AM, Trish Cayetano wrote:
> 
> -	Ada complete language is too complex. True or False? If true, how?

That depends on the definition of "too complex". C++ is a popular and widely
used language with an RM longer than the Ada RM, though C++ lacks modules and
concurrency. Clearly if Ada is "too complex" then C++ is also "too complex".
That doesn't stop it from being popular and widely used. Ada will never be
popular, since it supports the way S/W engineers think, while languages like
C/++ are for coders who don't want to think at all before they start coding, and
there are a lot more coders than S/W engineers.

Also, Ada has a number of unnecessary features added in an attempt to be
popular, such as support for programming by extension and anonymous types.
Luckily most Ada features are orthogonal, so you can ignore these features and
use the well designed and less complex remaining features.

> -	Ada requires expensive compilers for large systems 

Requires? Hardly. There are free compilers which can handle systems of any size
as well as any expensive compiler. There may be other project requirements that
drive the project to an expensive compiler, but using Ada does not, in and of
itself, require expensive compilers for any systems.

-- 
Jeff Carter
"Damn it, Jim, I'm an actor, not a doctor."
124

Re: Weakness of Ada is expensive / security / etc. ? Anything else?

[jm.tarrasa]

weakness
- Lack of popularity 

I know popularity shouldn't be a point, but the lack of popularity drives to lack of tools, libraries...
That's a weakness unless it is a niche and very specialized language. And as far as I know Ada is supposed to be general purpose language. 

I don't know how it works with non-PC. But, unless you go for GPL, commercial Ada is really expensive.

If you think  not in the compiler alone, but in a framework, a framework to work properly with Ada compared to frameworks that use other language, Ada is expensive, really expensive and, leaving aside the compiler, much poorer.
GPS is a good IDE but not near Delphi or Visual studio, but prices are.. well.. for deep pockets.

Re: Weakness of Ada is expensive / security / etc. ? Anything else?

[Pascal Obry]

Le vendredi 17 juillet 2015 à 12:28 -0700, jm.tarrasa@gmail.com a écrit
 :
> If you think  not in the compiler alone, but in a framework, a 
> framework to work properly with Ada compared to frameworks that use 
> other language, Ada is expensive, really expensive and, leaving aside 
> the compiler, much poorer.
> GPS is a good IDE but not near Delphi or Visual studio, but prices 
> are.. well.. for deep pockets.

Sure, for deep pockets... It is free! Are you kidding or just very
poorly informed or trying a troll?

-- 
  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: Weakness of Ada is expensive / security / etc. ? Anything else?

[Shark8]

On Friday, July 17, 2015 at 3:00:35 PM UTC-6, Pascal Obry wrote:
> Le vendredi 17 juillet 2015 à 12:28 -0700, jm.tarrasa a écrit
>  :
> > GPS is a good IDE but not near Delphi or Visual studio, but prices 
> > are.. well.. for deep pockets.
> 
> Sure, for deep pockets... It is free! Are you kidding or just very
> poorly informed or trying a troll?

I think he was referring to Delphi and Visual Studio.

Re: Weakness of Ada is expensive / security / etc. ? Anything else?

[Nasser M. Abbasi]

On 7/17/2015 4:53 PM, Shark8 wrote:
> On Friday, July 17, 2015 at 3:00:35 PM UTC-6, Pascal Obry wrote:
>> Le vendredi 17 juillet 2015 à 12:28 -0700, jm.tarrasa a écrit
>>   :
>>> GPS is a good IDE but not near Delphi or Visual studio, but prices
>>> are.. well.. for deep pockets.
>>
>> Sure, for deep pockets... It is free! Are you kidding or just very
>> poorly informed or trying a troll?
>
> I think he was referring to Delphi and Visual Studio.
>

MS now have a free version of Visual Studio

https://www.visualstudio.com/en-us/products/free-developer-offers-vs.aspx

called Visual Studio Community. I downloaded it and used it
(it is needed for using Intel free Fortran compiler, for
students, it is free, since this compiler is integrated with
windows visual studio and has a nice GUI debugger).

Delphi from embarcadero also have a free version called
Delphi Starter Edition, for personal use only.

http://www.embarcadero.com/products/delphi/starter-faq

But the commericals versions are expensive yes.

--Nasser

Re: Weakness of Ada is expensive / security / etc. ? Anything else?

[Trish Cayetano]

On Saturday, July 18, 2015 at 6:41:22 AM UTC+8, Nasser M. Abbasi wrote:
> On 7/17/2015 4:53 PM, Shark8 wrote:
> > On Friday, July 17, 2015 at 3:00:35 PM UTC-6, Pascal Obry wrote:
> >> Le vendredi 17 juillet 2015 à 12:28 -0700, jm.tarrasa a écrit
> >>   :
> >>> GPS is a good IDE but not near Delphi or Visual studio, but prices
> >>> are.. well.. for deep pockets.
> >>
> >> Sure, for deep pockets... It is free! Are you kidding or just very
> >> poorly informed or trying a troll?
> >
> > I think he was referring to Delphi and Visual Studio.
> >
> 
> MS now have a free version of Visual Studio
> 
> https://www.visualstudio.com/en-us/products/free-developer-offers-vs.aspx
> 
> called Visual Studio Community. I downloaded it and used it
> (it is needed for using Intel free Fortran compiler, for
> students, it is free, since this compiler is integrated with
> windows visual studio and has a nice GUI debugger).
> 
> Delphi from embarcadero also have a free version called
> Delphi Starter Edition, for personal use only.
> 
> http://www.embarcadero.com/products/delphi/starter-faq
> 
> But the commericals versions are expensive yes.
> 
> --Nasser

Thanks all for your valuable inputs!!!

Re: Weakness of Ada is expensive / security / etc. ? Anything else?

[David Botton]

> I know popularity shouldn't be a point, but the lack of popularity drives to lack of tools, libraries...

It is only a part. The vendor focus for Ada is in a niche where the tools are either present or they provide all that is needed, So no R&D is focused outside of it for tools except a few open source volunteers, many of whom have left, or reduced contributions, because they feel snubbed by license games that drive down the popularity of the language and greater usefulness of their contributions. (As an example, I only chose Ada in the end for Gnoga because I wanted a last ditch effort to help Ada from become a completely obscure language as the current business models are set to do and the closing up of SigAda efforts, etc. but if I had written Gnoga in Python for example it would make a massive impact and quickly.) 

> That's a weakness unless it is a niche and very specialized language. And as far as I know Ada is supposed to be general purpose language. 

Yes, but small minded thinking is that preventing Ada's wider use will make the niche where it is used more valuable. In the big picture interest in Ada for general purpose use has resulted in its reduction in education environments and over the course of time the niche use will continue to disappear completely since other general purpose languages are attacking that same niche for business as well.

> I don't know how it works with non-PC. But, unless you go for GPL, commercial Ada is really expensive.

No, FSF GNAT works well for professional large projects. Although if large enough you will want as with any language and tool you use support which AdaCore offers at a very reasonable for market rate.
 
> If you think  not in the compiler alone, but in a framework, a framework to work properly with Ada compared to frameworks that use other language, Ada is expensive, really expensive and, leaving aside the compiler, much poorer.

You'd have give specific examples, but Ada is not that far behind in most areas and ahead in others (like Gnoga :)

> GPS is a good IDE but not near Delphi or Visual studio, but prices are.. well.. for deep pockets.

They are different products and just comparing what is the same you will find Microsoft support is not so cheap and Delphi is outright ludicrous in price. (Actually who in their right mind would use Delphi over Lazerous and Free Pascal today anyways and if you are smart you'd use Gnoga over either of those :)

David Botton

Re: Weakness of Ada is expensive / security / etc. ? Anything else?

[Norman Worth]

On 7/17/2015 4:19 AM, Trish Cayetano wrote:
> Hi, here is what I have noted as strengths and weaknesses of Ada. Let me know for any corrections and if you have anything to add... Thank you!
>
> Strength and weaknesses of the language
> Strength
> -	Semantics - well defined even in error situations

		Yes.  This is perhaps Ada's greatest strength.  It is what makes Ada 
safe and often what makes it reliably understandable.  It is why many 
errors that creep in with other languages simply can not happen in Ada. 
  But every Ada programmer has (fairly often) screamed at the compiler 
"I didn't do that!" or "What the hell does that mean!" while staring at 
an obscure error message.  He then clicks in the appropriate place to 
get the ARM reference and, momentarily, is even more confused.  Of 
course, he did make the error, and there is a sound meaning to the error 
message.  Despite the occasional frustration, I don't know of anyone who 
seriously wants to relax the rules.  Since a bit before Ada 95 it seems 
to have become a serious area of study to find ways to make Ada easy to 
use while keeping this strict semantic safety net.

> -	Strong typing - can be used to reduce the scope (and cost) of analysis

		Those of us who are used to it love it.  Others really hate it.  This 
is a matter of training and experience.  It is not always obvious how to 
make strong typing work for you.

> -	Quality - successfully used in many high integrity applications

 > -     Standardization -  Ada is probably the most standardized 
language in existence.  This means that programs and systems can be 
moved from one environment to another with little or no fuss, and they 
work as expected when moved.  It also means that programmers require 
minimum retraining when going from one setting to another.
>
> Weaknesses
> -	Ada complete language is too complex. True or False? If true, how?

		It is not too complex, but it is complex.  Fortunately, it is quite 
modular, and you can use it effectively while learning the more obscure 
or complex features.  I might mention that other languages are also 
becoming very complex, and they do not always share the modularity or 
stright-forward and uniform style of Ada.  Have you looked at the most 
recent versions of Fortran and Cobol?

> -	Ada requires expensive compilers for large systems

		Not at all.  I first used Ada (full Ada 83) on a North Star Z80 
computer with a very cheap commercial compiler (RR Software).  The 
latest version of the language (Ada 2012) is fully supported by the free 
GNAT compiler, which is available for just about all modern computers. 
Even fully supported compilers are competitively priced, and the fully 
supported version of GNAT (from AdaCore) is quite reasonable.  These 
compilers support systems of any size.

> -	Ada lacks that web presence greatly. comp.lang.ada is the only location on the internet with an active Ada community.

		Good point.  Google "Ada language" and you will get a few more 
references.  But links are often out of date, and there is not much 
activity.  Even this forum does not see much action.  The compiler 
makers, like AdaCore, try to keep up some activities of their sites and 
often have interesting articles and tutorials.

> -	How is security a weakness? (Strength can be a weakness too?)

		It isn't.

 > -     It is difficult to find modern and versatile development tools.

		I have had Ada turned down as the development language for projects 
because of this.  GPS is an acceptable, if somewhat awkward, tool for 
GNAT, but it pales in comparison to the tools available for C/C++.

 > -     It is difficult to find interfaces to standard libraries, like 
Motif or the Windows GUI libraries or databases or to the standard 
scientific libraries written for Fortran or C.

		This can be a real problem.  Most big projects rely on standard 
libraries, and almost any user interface requires a GUI these days.  The 
interfaces often exist, but they can be hard to find.  In some cases, 
they are proprietary.  GTKAda provides an interface to the GTK+ 
graphical interface routines for Linux.  Unfortunately, it is hard to 
use compared to Visual C++, and the documentation is difficult.  GNAT 
from AdaCore provides a Windows interface. (I haven't tried it.)  There 
is an old interface to XLib and Motif that used to work, but I haven't 
seen it in years.  GNAT from AdaCore has interfaces to some databases, 
including an ODBC interface, but databases do not seem to be well 
supported in general.

 > -	 Programmers are reluctant to learn and use Ada because they feel 
it limits their employment opportunities.
>

Re: Weakness of Ada is expensive / security / etc. ? Anything else?

[Serge Robyns]

On Monday, 20 July 2015 05:40:55 UTC+3, Norman Worth  wrote:
> On 7/17/2015 4:19 AM, Trish Cayetano wrote:
>   But every Ada programmer has (fairly often) screamed at the compiler 
> "I didn't do that!" or "What the hell does that mean!" while staring at 
> an obscure error message.  He then clicks in the appropriate place to 
> get the ARM reference and, momentarily, is even more confused.

Oh yes! Especially coming from the C class language families.

> > -	Strong typing - can be used to reduce the scope (and cost) of analysis
> 
> 		Those of us who are used to it love it.  Others really hate it.  This 

This is what keeps Ada as general purpose language alive, but for how long?

>  > -     It is difficult to find interfaces to standard libraries, like 
> Motif or the Windows GUI libraries or databases or to the standard 
> scientific libraries written for Fortran or C.
> 
> 		This can be a real problem.  Most big projects rely on standard 
> libraries, and almost any user interface requires a GUI these days.  The 
> interfaces often exist, but they can be hard to find.  In some cases, 
> they are proprietary.  GTKAda provides an interface to the GTK+ 
> graphical interface routines for Linux.  Unfortunately, it is hard to 
> use compared to Visual C++, and the documentation is difficult.  GNAT 
> from AdaCore provides a Windows interface. (I haven't tried it.)  There 
> is an old interface to XLib and Motif that used to work, but I haven't 
> seen it in years.  GNAT from AdaCore has interfaces to some databases, 
> including an ODBC interface, but databases do not seem to be well 
> supported in general.
> 

This is the killer for Ada acceptance.  All other languages provides those libraries and are "usually" well documented.  I had to revert to this forum to help me understanding why my XML SAX validating reader was not working, although I followed the examples posted by AdaCore.  The solution is "read the source", which works when you already know the library well enough but not when you are learning how to use it.

I'm surprised that the AdaCore GNATCOLL SQL library has no "standard" support for Oracle, although it was already there in GNADE from which GNATCOLL SQL was build.  I'm ok to work and contribute back, but this removes me from my own project.  Moreover, the code is quite "shaky" and not resilient at all, which came as a bad surprise to me or a language like ADA.

Re: Weakness of Ada is expensive / security / etc. ? Anything else?

[Shark8]

On Monday, July 20, 2015 at 3:52:03 AM UTC-6, Serge Robyns wrote:
> 
> I'm surprised that the AdaCore GNATCOLL SQL library has no "standard" support for Oracle, although it was already there in GNADE from which GNATCOLL SQL was build.  I'm ok to work and contribute back, but this removes me from my own project.  Moreover, the code is quite "shaky" and not resilient at all, which came as a bad surprise to me or a language like ADA.

GNATCOLL is terrible -- I've *never* gotten it to build w/ DB support and properly link... and that's even after the massive inconvenience of cygwin (needed because the blasted project needs .configure and make).

What's doubly disappointing about that is that GNATCOLL is presented by AdaCore (which presents itself as the public face of Ada) as one of the core useful-libraries and DOESN'T take advantage of Ada's nice portability instead relying on the crap-pile that is make and .configure even when they have/use GPR-files to represent a project themselves. (i.e. not using your own tools is a vote *against* your own product.)

If you adopt the "eat your own dogfood" position, you have an incentive to make that a quality system, fix bugs, and make it easier to use.

Re: Weakness of Ada is expensive / security / etc. ? Anything else?

[David Botton]

Simple Components from Dmitry contains cross platform ODBC and Matreshka (hope spelling correct also has a suite of bindings, Gnoga includes MySQL and SQLLite3. All of which are easy to use portable Ada.

I've not had much success with GNATCOLL, but I wouldn't use either just because of licensing alone. (GPL without runtime exceptions, same goes for most AdaCore "freebies" so make sure you're compliant and your own software is full GPL).

David Botton

Re: Weakness of Ada is expensive / security / etc. ? Anything else?

[NiGHTS]

On Monday, July 20, 2015 at 1:35:37 PM UTC-4, Shark8 wrote:
> On Monday, July 20, 2015 at 3:52:03 AM UTC-6, Serge Robyns wrote:
> > 
> > I'm surprised that the AdaCore GNATCOLL SQL library has no "standard" support for Oracle, although it was already there in GNADE from which GNATCOLL SQL was build.  I'm ok to work and contribute back, but this removes me from my own project.  Moreover, the code is quite "shaky" and not resilient at all, which came as a bad surprise to me or a language like ADA.
> 
> GNATCOLL is terrible -- I've *never* gotten it to build w/ DB support and properly link... and that's even after the massive inconvenience of cygwin (needed because the blasted project needs .configure and make).
> 
> What's doubly disappointing about that is that GNATCOLL is presented by AdaCore (which presents itself as the public face of Ada) as one of the core useful-libraries and DOESN'T take advantage of Ada's nice portability instead relying on the crap-pile that is make and .configure even when they have/use GPR-files to represent a project themselves. (i.e. not using your own tools is a vote *against* your own product.)
> 
> If you adopt the "eat your own dogfood" position, you have an incentive to make that a quality system, fix bugs, and make it easier to use.

Here are a few comments on my personal experience with GNATCOLL and the greater AdaCore code base.

So I learn the language, fall in love (of course), then try to dive in and make something useful as a way to get experience in this great human accomplishment of a language. So I go to AdaCore (the main google result to most queries involving installation of Ada). Then you are presented with a page with a "Build your own package" button. A laymen would likely not understand what this means. Compilers, Linkers, Tools, IDEs all have a clear single point installation or at the very least some guide on what to do to get started. In AdaCore's world the most you will get is a huge list of checkboxes and "On Windows, you can use the Extract facility of the Winzip utility." Thats literally all you get, that and "If you have problems downloading or installing GNAT GPL or SPARK GPL Edition contact us at gnat-gpl@adacore.com". Oh really? I can ask installation questions to this email address for GNAT GPL?? Been there, done that. Here is the response I got:

"We do not provide support for the use of GNAT GPL, and do not provide help
on rebuilding from sources which requires some specific expertise." -- AdaCore Support.

I love the irony. I pointed out the irony. They did not respond to it as expected.

Now any average person would run away as soon as they see the long meaningless checklist of out-of-context packages, or perhaps after downloading all of them become even more intimidated by the massive downloaded archive and its convoluted rats-nest of archives and the occasional executable (If this is a Windows installation).

So a Windows user would be excited to see an executable and run it to get this installation started. Nope. You need a compile environment. If you are like me trying to get the GTK installed as well as GNATCOLL, you need to install MinGW with MSYS first (select all ada packages!), then install all targets (3 installers) in a very specific order changing all the default directory to the MSYS (NOT MinGW!) root folder, then use a command prompt to finalize the installation of the GNATCOLL library only after ensuring that the environment paths are correctly pointing to the right root (it defaults to MinGW). 

To get these packages installed it took me 2 weeks of trial and error. The Windows installation took me nearly as long as the Linux installation. Each had their own unique challenges, and each required that the order of installation be perfect. But what drives me crazy is how many dependencies they REQUIRE which are not initially stated by any documentation on the matter. Its up to you to figure this complexity out.

Now in my limited experience working with the AdaCore GDK (GUI) library I was disappointed with the behavioural differences when compiling between Linux and Windows. I am forced to switch between both platforms and compile & test on both when adding new GUI elements because some components behave wildly differently between the two platforms. Additionally other components in the standard library are completely missing yet fully documented as if it was expected to be there (For instance Listbox). Combine that with the strangeness I sometimes encounter working in the library (the messagebox package has a mysterious undocumented "Arg5" System.Address parameter mentioned no where in the entire world) and the lack of resources to help with usage (my only friend was the example code, C-based documentation, and the GDK source) and what I ended up with is a hugely unproductive and soured opinion on the Ada path as a solution to serious application development.

GNATCOLL got me excited when I read about how they leverage the features of Ada's strong typing mechanism to protect SQL queries. But when you want to actually use it the documentation is literally non-existent. The only places you can look is in the GNATCOLL source code and a few places here and there on the internet. I had posted a question recently on this newsgroup regarding not being able to DELETE anything in GNATCOLL. It went unanswered, and I'm not surprised. This is very depressing to me. The library has so much raw potential. It can be a market killer -- completely dominate the world of RDBMS, and yet it's difficult to install, nearly undocumented, and seemingly waiting to die alone.

I can't speak for the other libraries, but I have been creating my own library interfaces to compensate for a lack of out-of-the-box library support. I'm so thankful that Ada is so easy to interface with C libraries and that GprBuild is another shining human accomplishment.

Well that's it for my rant on the AdaCore's non-insightful way to get newcomers to use their products.

Re: Weakness of Ada is expensive / security / etc. ? Anything else?

[Norman Worth]

On 7/20/2015 4:12 PM, NiGHTS wrote:
> On Monday, July 20, 2015 at 1:35:37 PM UTC-4, Shark8 wrote:
>> On Monday, July 20, 2015 at 3:52:03 AM UTC-6, Serge Robyns wrote:
>>>
>>> I'm surprised that the AdaCore GNATCOLL SQL library has no "standard" support for Oracle, although it was already there in GNADE from which GNATCOLL SQL was build.  I'm ok to work and contribute back, but this removes me from my own project.  Moreover, the code is quite "shaky" and not resilient at all, which came as a bad surprise to me or a language like ADA.
>>
>> GNATCOLL is terrible -- I've *never* gotten it to build w/ DB support and properly link... and that's even after the massive inconvenience of cygwin (needed because the blasted project needs .configure and make).
>>
>> What's doubly disappointing about that is that GNATCOLL is presented by AdaCore (which presents itself as the public face of Ada) as one of the core useful-libraries and DOESN'T take advantage of Ada's nice portability instead relying on the crap-pile that is make and .configure even when they have/use GPR-files to represent a project themselves. (i.e. not using your own tools is a vote *against* your own product.)
>>
>> If you adopt the "eat your own dogfood" position, you have an incentive to make that a quality system, fix bugs, and make it easier to use.
>
> Here are a few comments on my personal experience with GNATCOLL and the greater AdaCore code base.
>
> So I learn the language, fall in love (of course), then try to dive in and make something useful as a way to get experience in this great human accomplishment of a language. So I go to AdaCore (the main google result to most queries involving installation of Ada). Then you are presented with a page with a "Build your own package" button. A laymen would likely not understand what this means. Compilers, Linkers, Tools, IDEs all have a clear single point installation or at the very least some guide on what to do to get started. In AdaCore's world the most you will get is a huge list of checkboxes and "On Windows, you can use the Extract facility of the Winzip utility." Thats literally all you get, that and "If you have problems downloading or installing GNAT GPL or SPARK GPL Edition contact us at gnat-gpl@adacore.com". Oh really? I can ask installation questions to this email address for GNAT GPL?? Been there, done that. Here is the response I got:
>
> "We do not provide support for the use of GNAT GPL, and do not provide help
> on rebuilding from sources which requires some specific expertise." -- AdaCore Support.
>
> I love the irony. I pointed out the irony. They did not respond to it as expected.
>
> Now any average person would run away as soon as they see the long meaningless checklist of out-of-context packages, or perhaps after downloading all of them become even more intimidated by the massive downloaded archive and its convoluted rats-nest of archives and the occasional executable (If this is a Windows installation).
>
> So a Windows user would be excited to see an executable and run it to get this installation started. Nope. You need a compile environment. If you are like me trying to get the GTK installed as well as GNATCOLL, you need to install MinGW with MSYS first (select all ada packages!), then install all targets (3 installers) in a very specific order changing all the default directory to the MSYS (NOT MinGW!) root folder, then use a command prompt to finalize the installation of the GNATCOLL library only after ensuring that the environment paths are correctly pointing to the right root (it defaults to MinGW).
>
> To get these packages installed it took me 2 weeks of trial and error. The Windows installation took me nearly as long as the Linux installation. Each had their own unique challenges, and each required that the order of installation be perfect. But what drives me crazy is how many dependencies they REQUIRE which are not initially stated by any documentation on the matter. Its up to you to figure this complexity out.
>
> Now in my limited experience working with the AdaCore GDK (GUI) library I was disappointed with the behavioural differences when compiling between Linux and Windows. I am forced to switch between both platforms and compile & test on both when adding new GUI elements because some components behave wildly differently between the two platforms. Additionally other components in the standard library are completely missing yet fully documented as if it was expected to be there (For instance Listbox). Combine that with the strangeness I sometimes encounter working in the library (the messagebox package has a mysterious undocumented "Arg5" System.Address parameter mentioned no where in the entire world) and the lack of resources to help with usage (my only friend was the example code, C-based documentation, and the GDK source) and what I ended up with is a hugely unproductive and soured opinion on the Ada path as a solution to serious application development.
>
> GNATCOLL got me excited when I read about how they leverage the features of Ada's strong typing mechanism to protect SQL queries. But when you want to actually use it the documentation is literally non-existent. The only places you can look is in the GNATCOLL source code and a few places here and there on the internet. I had posted a question recently on this newsgroup regarding not being able to DELETE anything in GNATCOLL. It went unanswered, and I'm not surprised. This is very depressing to me. The library has so much raw potential. It can be a market killer -- completely dominate the world of RDBMS, and yet it's difficult to install, nearly undocumented, and seemingly waiting to die alone.
>
> I can't speak for the other libraries, but I have been creating my own library interfaces to compensate for a lack of out-of-the-box library support. I'm so thankful that Ada is so easy to interface with C libraries and that GprBuild is another shining human accomplishment.
>
> Well that's it for my rant on the AdaCore's non-insightful way to get newcomers to use their products.
>
I agree that lack of support for Oracle, by far the most common 
commercial database system, is a serious problem.  Perhaps there are 
copyright and licensing issues.  If someone wants to take up the 
challenge, the people at Oracle might be cooperative - might be.

I found installing GNAT on Windows to be very easy and effective.  You 
just download the package, switch to its toplevel directory, and execute 
the installer.  The default settings work properly for most Windows 
installations, and all of the paths are correctly established.  You do 
not have to fuss with MinGW or MSYS; they are installed and linked to 
the package correctly and automatically.  What is difficult about it is 
realizing that you want the GNAT GPL package, that you must specify 
Windows (and the correct version of Windows) when asking for the 
download, and that this must be installed first.  Installing GTKAda and 
ASIS are a great deal more difficult - possible, but difficult. 
GNATCOLL was never designed for Windows, and I have not tried to port 
it.  GNATCOLL actually has some errors in the source code.  They are 
easy to correct, but they shouldn't be there at all.  Actually 
installation on Linux can be more difficult than installation on Windows

Re: Weakness of Ada is expensive / security / etc. ? Anything else?

[Dmitry A. Kazakov]

On Mon, 20 Jul 2015 15:12:30 -0700 (PDT), NiGHTS wrote:

> Now in my limited experience working with the AdaCore GDK (GUI) library I
> was disappointed with the behavioural differences when compiling between
> Linux and Windows.

Never heard of AdaCore GDK. What is this?

> GNATCOLL got me excited when I read about how they leverage the features
> of Ada's strong typing mechanism to protect SQL queries.

The concept was interesting, at least as an alternative to embedded SQL
with patching the compiler.

-- 
Regards,
Dmitry A. Kazakov
http://www.dmitry-kazakov.de

Re: Weakness of Ada is expensive / security / etc. ? Anything else?

[NiGHTS]

On Tuesday, July 21, 2015 at 2:54:11 AM UTC-4, Dmitry A. Kazakov wrote:
> On Mon, 20 Jul 2015 15:12:30 -0700 (PDT), NiGHTS wrote:
> 
> > Now in my limited experience working with the AdaCore GDK (GUI) library I
> > was disappointed with the behavioural differences when compiling between
> > Linux and Windows.
> 
> Never heard of AdaCore GDK. What is this?
> 

I meant the Gnome Toolkit (GTK). I often confuse the abbreviation because their library has both abbreviations in use.

Re: Weakness of Ada is expensive / security / etc. ? Anything else?

[Dmitry A. Kazakov]

On Tue, 21 Jul 2015 06:31:21 -0700 (PDT), NiGHTS wrote:

> On Tuesday, July 21, 2015 at 2:54:11 AM UTC-4, Dmitry A. Kazakov wrote:
>> On Mon, 20 Jul 2015 15:12:30 -0700 (PDT), NiGHTS wrote:
>> 
>>> Now in my limited experience working with the AdaCore GDK (GUI) library I
>>> was disappointed with the behavioural differences when compiling between
>>> Linux and Windows.
>> 
>> Never heard of AdaCore GDK. What is this?
> 
> I meant the Gnome Toolkit (GTK). I often confuse the abbreviation because
> their library has both abbreviations in use.

OK, but then GtkAda is not to blame for any differences GTK might have.
Which mostly non-existent, and if any, then rather due to GTK-themes
selected, or not, by the user (you).

All GUI toolkits are themed in these days, which is bad in my view as well
as from the ergonomic/automation POV.

But they are hugely beloved by people dying to see green texts on magenta
backgrounds...

-- 
Regards,
Dmitry A. Kazakov
http://www.dmitry-kazakov.de

Re: Weakness of Ada is expensive / security / etc. ? Anything else?

[NiGHTS]

On Tuesday, July 21, 2015 at 12:59:24 PM UTC-4, Dmitry A. Kazakov wrote:
> OK, but then GtkAda is not to blame for any differences GTK might have.
> Which mostly non-existent, and if any, then rather due to GTK-themes
> selected, or not, by the user (you).
> 
> All GUI toolkits are themed in these days, which is bad in my view as well
> as from the ergonomic/automation POV.
> 
> But they are hugely beloved by people dying to see green texts on magenta
> backgrounds...
> 

If it was theme related differences I wouldn't be complaining about it. Here is an example for you: Word wrap simply doesn't work on labels in Linux but it does work in Windows. Same code (literally), wildly different results. Here's another: When using Tree_View.Set_Cursor with Start_Editing => True, you can't see what you are typing in Windows (looks like uneditable text) yet it works just fine in Linux.

There are other examples I could give you but I can't remember them all on account of that I now try to avoid the things that cause structural inconsistencies. Mind you none of them are style / color related. 

I would love to hear some workarounds to some of these problems since I am actively working on such a project for multiple platforms on Ada with GNATCOLL and GTKAda involved. (Testing platforms: Kubuntu 15.04 x64 / Windows 7 x32)

Re: Weakness of Ada is expensive / security / etc. ? Anything else?

[Dmitry A. Kazakov]

On Tue, 21 Jul 2015 10:50:27 -0700 (PDT), NiGHTS wrote:

> On Tuesday, July 21, 2015 at 12:59:24 PM UTC-4, Dmitry A. Kazakov wrote:
>> OK, but then GtkAda is not to blame for any differences GTK might have.
>> Which mostly non-existent, and if any, then rather due to GTK-themes
>> selected, or not, by the user (you).
>> 
>> All GUI toolkits are themed in these days, which is bad in my view as well
>> as from the ergonomic/automation POV.
>> 
>> But they are hugely beloved by people dying to see green texts on magenta
>> backgrounds...
> 
> If it was theme related differences I wouldn't be complaining about it.
> Here is an example for you: Word wrap simply doesn't work on labels in
> Linux but it does work in Windows.

Label text wrapping may depend on the widget size requisition. So you
should set the label size explicitly. It could be a GTK bug as well,
because it is highly unlikely that anybody would use that for text
rendering, there is Gtk_Text_View for that. And because GTK 3 reworked most
of widget sizing.

> Same code (literally), wildly different results.

GTK is full of bugs, especially GTK 3. But that has nothing to do with Ada.

> Here's another: When using Tree_View.Set_Cursor with
> Start_Editing => True, you can't see what you are typing in Windows (looks
> like uneditable text) yet it works just fine in Linux.

It is again GTK stuff, not Ada.

>There are other examples I could give you but I can't remember them all on
> account of that I now try to avoid the things that cause structural
> inconsistencies. Mind you none of them are style / color related. 

Style is much more than colors. All widget style properties can be changed.
 
> I would love to hear some workarounds

There is GtkAda mail list.

> to some of these problems since I am
> actively working on such a project for multiple platforms on Ada with
> GNATCOLL

I cannot tell anything for that, I never used it. I dropped using
DB-specific bindings a decade ago. ODBC does the job. There is no serious
reasons to use DB-specific bindings.

-- 
Regards,
Dmitry A. Kazakov
http://www.dmitry-kazakov.de

Re: Weakness of Ada is expensive / security / etc. ? Anything else?

[Björn Lundin]

On 2015-07-21 21:44, Dmitry A. Kazakov wrote:

> I cannot tell anything for that, I never used it. I dropped using
> DB-specific bindings a decade ago. ODBC does the job. There is no serious
> reasons to use DB-specific bindings.

Yes there is.
When the platform you are forced to work on
is not supporting odbc in a trustworthy way,
like Aix.

Unixodbc 'might' work on aix, but not so I
would put any serious business into it.

Oracle with binding to pro*c or OCI works well.
(clumpsy, but works)


--
Björn

Re: Weakness of Ada is expensive / security / etc. ? Anything else?

[Georg Bauhaus]

"Dmitry A. Kazakov" <mailbox@dmitry-kazakov.de> wrote:

> ODBC does the job. There is no serious
> reasons to use DB-specific bindings.

Unfortunately, ODBC works only half way,
because there is only half of a meaning
in SQL standard: try quoting a name portably.

Can you place a single quote in an SQL
comment to be sent with a request?

Re: Weakness of Ada is expensive / security / etc. ? Anything else?

[Shark8]

On Tuesday, July 21, 2015 at 4:02:57 PM UTC-6, Georg Bauhaus wrote:
> 
> Unfortunately, ODBC works only half way,
> because there is only half of a meaning
> in SQL standard: try quoting a name portably.

Heck, try writing non-trivial CREATE statements [w/ relations and constraints].

Re: Weakness of Ada is expensive / security / etc. ? Anything else?

[Dmitry A. Kazakov]

On Tue, 21 Jul 2015 16:44:43 -0700 (PDT), Shark8 wrote:

> On Tuesday, July 21, 2015 at 4:02:57 PM UTC-6, Georg Bauhaus wrote:
>> 
>> Unfortunately, ODBC works only half way,
>> because there is only half of a meaning
>> in SQL standard: try quoting a name portably.
> 
> Heck, try writing non-trivial CREATE statements [w/ relations and constraints].

ODBC was not designed to supersede SQL, unfortunately. You must fight SQL
and DB-specific variations of, regardless ODBC. And yes, ODBC has lots of
problems starting with basic types being dependant on 32 or 64-bit targets.
But, again, there is no serious reason to use DB-specific bindings.

-- 
Regards,
Dmitry A. Kazakov
http://www.dmitry-kazakov.de

Re: Weakness of Ada is expensive / security / etc. ? Anything else?

[Dmitry A. Kazakov]

On Tue, 21 Jul 2015 22:01:26 +0000 (UTC), Georg Bauhaus wrote:

> "Dmitry A. Kazakov" <mailbox@dmitry-kazakov.de> wrote:
> 
>> ODBC does the job. There is no serious
>> reasons to use DB-specific bindings.
> 
> Unfortunately, ODBC works only half way,
> because there is only half of a meaning
> in SQL standard: try quoting a name portably.

> Can you place a single quote in an SQL
> comment to be sent with a request?

Never use literals, always bind variables.

-- 
Regards,
Dmitry A. Kazakov
http://www.dmitry-kazakov.de

Re: Weakness of Ada is expensive / security / etc. ? Anything else?

[Georg Bauhaus]

On 22.07.15 08:13, Dmitry A. Kazakov wrote:
> On Tue, 21 Jul 2015 22:01:26 +0000 (UTC), Georg Bauhaus wrote:
>
>> "Dmitry A. Kazakov" <mailbox@dmitry-kazakov.de> wrote:
>>
>>> ODBC does the job. There is no serious
>>> reasons to use DB-specific bindings.
>> try quoting a name portably.
> Never use literals, always bind variables.
>
Can you portably use ODBC like this:

   SELECT ... FROM ? WHERE ...

instead of:

   SELECT ... FROM "T" WHERE ...
   SELECT ... FROM `T` WHERE ...
   SELECT ... FROM [T] WHERE ...

Re: Weakness of Ada is expensive / security / etc. ? Anything else?

[Dmitry A. Kazakov]

On Wed, 22 Jul 2015 09:04:57 +0200, Georg Bauhaus wrote:

> On 22.07.15 08:13, Dmitry A. Kazakov wrote:
>> On Tue, 21 Jul 2015 22:01:26 +0000 (UTC), Georg Bauhaus wrote:
>>
>>> "Dmitry A. Kazakov" <mailbox@dmitry-kazakov.de> wrote:
>>>
>>>> ODBC does the job. There is no serious
>>>> reasons to use DB-specific bindings.
>>> try quoting a name portably.
>> Never use literals, always bind variables.
>>
> Can you portably use ODBC like this:
> 
>    SELECT ... FROM ? WHERE ...
> 
> instead of:
> 
>    SELECT ... FROM "T" WHERE ...
>    SELECT ... FROM `T` WHERE ...
>    SELECT ... FROM [T] WHERE ...

Yes.

-- 
Regards,
Dmitry A. Kazakov
http://www.dmitry-kazakov.de

Re: Weakness of Ada is expensive / security / etc. ? Anything else?

[Graham Stark]

What's needed here, I think, is a very simple abstraction layer above native bindings, just dealing with binding variables, getting a connection, and the like, for each different database (including ODBC). A JDBC clone, essentially. Actually GNATColl has the germs of this, in the GNATCOLL.SQL.Exec package, though it only works with Postgres at the moment. I use just these low-level bits of GNATColl and it works well, I think. Having that is key to getting more ambitious schemes to work, as with Torque, Hibernate, etc. in the Java world.

Graham

Re: Weakness of Ada is expensive / security / etc. ? Anything else?

[Dmitry A. Kazakov]

On Wed, 22 Jul 2015 02:29:24 -0700 (PDT), Graham Stark wrote:

> What's needed here, I think, is a very simple abstraction layer above
> native bindings, just dealing with binding variables, getting a
> connection, and the like, for each different database (including ODBC).

ODBC is exactly such a layer. The native DB client is encapsulated by the
ODBC driver.

-- 
Regards,
Dmitry A. Kazakov
http://www.dmitry-kazakov.de

Re: Weakness of Ada is expensive / security / etc. ? Anything else?

[G.B.]

On 22.07.15 09:37, Dmitry A. Kazakov wrote:
> On Wed, 22 Jul 2015 09:04:57 +0200, Georg Bauhaus wrote:
>
>> On 22.07.15 08:13, Dmitry A. Kazakov wrote:
>>> On Tue, 21 Jul 2015 22:01:26 +0000 (UTC), Georg Bauhaus wrote:
>>>
>>>> "Dmitry A. Kazakov" <mailbox@dmitry-kazakov.de> wrote:
>>>>
>>>>> ODBC does the job. There is no serious
>>>>> reasons to use DB-specific bindings.
>>>> try quoting a name portably.
>>> Never use literals, always bind variables.
>>>
>> Can you portably use ODBC like this:
>>
>>     SELECT ... FROM ? WHERE ...
>>
>> instead of:
>>
>>     SELECT ... FROM "T" WHERE ...
>>     SELECT ... FROM `T` WHERE ...
>>     SELECT ... FROM [T] WHERE ...
>
> Yes.

Interesting.

As I, and many others, tend to see syntax errors or other
errors when a statement having ? in a place of a table's
name is sent via ODBC, I'm sure many would like to see how.

Re: Weakness of Ada is expensive / security / etc. ? Anything else?

[Dmitry A. Kazakov]

On Wed, 22 Jul 2015 12:12:33 +0200, G.B. wrote:

> On 22.07.15 09:37, Dmitry A. Kazakov wrote:
>> On Wed, 22 Jul 2015 09:04:57 +0200, Georg Bauhaus wrote:
>>
>>> On 22.07.15 08:13, Dmitry A. Kazakov wrote:
>>>> On Tue, 21 Jul 2015 22:01:26 +0000 (UTC), Georg Bauhaus wrote:
>>>>
>>>>> "Dmitry A. Kazakov" <mailbox@dmitry-kazakov.de> wrote:
>>>>>
>>>>>> ODBC does the job. There is no serious
>>>>>> reasons to use DB-specific bindings.
>>>>> try quoting a name portably.
>>>> Never use literals, always bind variables.
>>>>
>>> Can you portably use ODBC like this:
>>>
>>>     SELECT ... FROM ? WHERE ...
>>>
>>> instead of:
>>>
>>>     SELECT ... FROM "T" WHERE ...
>>>     SELECT ... FROM `T` WHERE ...
>>>     SELECT ... FROM [T] WHERE ...
>>
>> Yes.
> 
> Interesting.
> 
> As I, and many others, tend to see syntax errors or other
> errors when a statement having ? in a place of a table's
> name is sent via ODBC, I'm sure many would like to see how.

Ah, you meant table names? I see.

Well, that is not a literal, it is an identifier. Binding does not work for
them. And you want too much from SQL. SQL is a very low-level language
where tables are not objects and have no type (to have common operations
where a table could be a parameter)

-- 
Regards,
Dmitry A. Kazakov
http://www.dmitry-kazakov.de

Re: Weakness of Ada is expensive / security / etc. ? Anything else?

[G.B.]

On 22.07.15 12:31, Dmitry A. Kazakov wrote:
>
> Ah, you meant table names? I see.
>
> Well, that is not a literal, it is an identifier. Binding does not work for
> them. And you want too much from SQL. SQL is a very low-level language
> where tables are not objects and have no type (to have common operations
> where a table could be a parameter)

... T1 INNER JOIN T2 ...

has infix operator INNER JOIN which requires T1 and T2 to
be of proper type, i.e. tables, or views, or compatible
relations produced via SELECT etc...

They aren't of some named type, but their type is grammatically
implied and needs structural equivalence.

Granted, there are no meta-types for objects from the
universe having objects of table kinds in it; also, one
cannot define JOIN. Not too unusual to not be able to
define language in language, I think.

Re: Weakness of Ada is expensive / security / etc. ? Anything else?

[NiGHTS]

> Label text wrapping may depend on the widget size requisition. So you
> should set the label size explicitly. It could be a GTK bug as well,
> because it is highly unlikely that anybody would use that for text
> rendering, there is Gtk_Text_View for that. And because GTK 3 reworked most
> of widget sizing.

It affects things like tooltips (built in) and various other things that inherently use labels. I've explicitly set widths to all inherited widgets including the label itself and this had no effect. I made sure Word wrapping was enabled and followed all correct guidelines on the subject. Again it works in Windows, just not in Linux.


> GTK is full of bugs, especially GTK 3. But that has nothing to do with Ada.

I am not blaming Ada. Its a beautiful language. I merely suspect that GtkAda's thick bindings have something to do with it. I am not dismissing the possibility that GTK 3 is also at fault, but in my mind I feel that the GTK core library is probably more tested than GTKAda (A misconception maybe? I would love to be wrong about this.)


> Style is much more than colors. All widget style properties can be changed.
> 

Well you were talking about "themes", and no doubt styles involve both color and structure. On that note I am quite disappointed by how limited the GTK CSS support is with regards to things like setting margins and other structural parameters. They simply don't work. I can only use the CSS to define colors, fonts, and backgrounds. I've been forced to hard code structural things like margins which I feel is bad form as it should belong in the style sheet.


> There is GtkAda mail list.
> 

I've looked and couldn't find it. I would love to know what the address is if you don't mind me asking.


> I cannot tell anything for that, I never used it. I dropped using
> DB-specific bindings a decade ago. ODBC does the job. There is no serious
> reasons to use DB-specific bindings.

GNATCOLL's database library was not designed to be DB-Specific. From what I understand of their design philosophy is that they are offering an strongly-typed Ada-esque approach to work with SQL queries and other common database tasks using a common seamless API. They only support two databases currently but it was designed to support whatever database is needed for the job. If done correctly you should be able to switch databases without changing code in GNATCOLL. It would have been nice to have added support for ODBC, but the GNATCOLL is designed to facilitate interfacing to anything including ODBC.

I fell in love with the GNATCOLL SQL philosophy. Unfortunately the designers have not done a good job of generating interest in the project. In my case I can't get any support whatsoever with using that library and it seems like no one here takes it seriously. Its a shame.

Re: Weakness of Ada is expensive / security / etc. ? Anything else?

[Dmitry A. Kazakov]

On Tue, 21 Jul 2015 17:17:38 -0700 (PDT), NiGHTS wrote:

>> Label text wrapping may depend on the widget size requisition. So you
>> should set the label size explicitly. It could be a GTK bug as well,
>> because it is highly unlikely that anybody would use that for text
>> rendering, there is Gtk_Text_View for that. And because GTK 3 reworked most
>> of widget sizing.
> 
> It affects things like tooltips (built in) and various other things that
> inherently use labels.

They reworked tooltips in GTK 3, I don't know how the new implementation
works, e.g. if you can insert widgets into a tooltip. But I suppose that
Gtk_Label could have issues with size computation interaction with its
container, the tooltip window in this case. Non-wrapping would be a
consequence of that.

> I've explicitly set widths to all inherited widgets
> including the label itself and this had no effect. I made sure Word
> wrapping was enabled and followed all correct guidelines on the subject.
> Again it works in Windows, just not in Linux.

You can always take a look in the GTK sources. Sometimes it helps.

>> GTK is full of bugs, especially GTK 3. But that has nothing to do with Ada.
> 
> I am not blaming Ada. Its a beautiful language. I merely suspect that
> GtkAda's thick bindings have something to do with it.

No. GtkAda is semi-thick, script-generated. Most bugs are GTK's own.
Problems you described IMO cannot be GtkAda's, per design of. If you had a
GtkAda bug, the effect would be program crash. E.g. I guess that Set_String
of Gtk_Tree_Store is such a bug.

> I am not dismissing
> the possibility that GTK 3 is also at fault, but in my mind I feel that
> the GTK core library is probably more tested than GTKAda (A misconception
> maybe? I would love to be wrong about this.)

GTK is poorly maintained and a quite mess. My encounter with bug reports
was very disappointing.

GtkAda designers do read GtkAda mailing list, they even respond there!
(:-)) The only problem with GtkAda maintenance is that the new release
comes out once a year. It would not be a problem, if building it under
Windows from the sources (they are freely available from the code base at
any time) weren't such a huge trouble.

>> Style is much more than colors. All widget style properties can be changed.
> 
> Well you were talking about "themes", and no doubt styles involve both
> color and structure. On that note I am quite disappointed by how limited
> the GTK CSS support is with regards to things like setting margins and
> other structural parameters. They simply don't work. I can only use the
> CSS to define colors, fonts, and backgrounds. I've been forced to hard
> code structural things like margins which I feel is bad form as it should
> belong in the style sheet.

GTK 2 had resource files. They dropped them and replaced by CSS. RC files
were bad CSS is worse.

>> There is GtkAda mail list.
> 
> I've looked and couldn't find it. I would love to know what the address is
> if you don't mind me asking.

gtkada@lists.adacore.com

> I fell in love with the GNATCOLL SQL philosophy. Unfortunately the
> designers have not done a good job of generating interest in the project.

Well, it depends on the potential customers. Personally, I am pretty
allergic to any code generators and don't want to contaminate my tool chain
with that. Then, I don't use anything but ODBC or *statically* linked
SQLite3.

> In my case I can't get any support whatsoever with using that library and
> it seems like no one here takes it seriously. Its a shame.

You should get GNAT Pro, if you want support. Otherwise, you must look into
the sources. Doing that you will probably join me in my resentment toward
code generators...

-- 
Regards,
Dmitry A. Kazakov
http://www.dmitry-kazakov.de

Re: Weakness of Ada is expensive / security / etc. ? Anything else?

[Shark8]

On Wednesday, July 22, 2015 at 12:54:50 AM UTC-6, Dmitry A. Kazakov wrote:
> 
> 
> GTK 2 had resource files. They dropped them and replaced by CSS. RC files
> were bad CSS is worse.

I agree; CSS is terrible, *especially* regarding layout.
Insofar as providing persistent values for layout/style (well, properties, really) I rather like Delphi's DFM format.

Re: Weakness of Ada is expensive / security / etc. ? Anything else?

[Brian Drummond]

On Tue, 21 Jul 2015 17:17:38 -0700, NiGHTS wrote:

> 
>> There is GtkAda mail list.
>> 
>> 
> I've looked and couldn't find it. I would love to know what the address
> is if you don't mind me asking.
> 
Dmitry gave you the address for posting to it.
http://lists.adacore.com/mailman/listinfo/gtkada
will allow you to subscribe, and links to the archives

-- Brian

Re: Weakness of Ada is expensive / security / etc. ? Anything else?

[Egil H H]

On Tuesday, July 21, 2015 at 12:12:32 AM UTC+2, NiGHTS wrote:
> GNATCOLL got me excited when I read about how they leverage the features of Ada's strong typing mechanism to protect SQL queries. But when you want to actually use it the documentation is literally non-existent. 

The GNATCOLL documentation is included in the source bundle. The SQL documentation (./docs/sql.rst) alone is a couple thousand lines, including simple examples.

Re: Weakness of Ada is expensive / security / etc. ? Anything else?

[Serge Robyns]

On Monday, 20 July 2015 20:35:37 UTC+3, Shark8  wrote:
> On Monday, July 20, 2015 at 3:52:03 AM UTC-6, Serge Robyns wrote:
> > 
> > I'm surprised that the AdaCore GNATCOLL SQL library has no "standard" support for Oracle, although it was already there in GNADE from which GNATCOLL SQL was build.  I'm ok to work and contribute back, but this removes me from my own project.  Moreover, the code is quite "shaky" and not resilient at all, which came as a bad surprise to me or a language like ADA.
> 
> GNATCOLL is terrible -- I've *never* gotten it to build w/ DB support and properly link... and that's even after the massive inconvenience of cygwin (needed because the blasted project needs .configure and make).
> 
> What's doubly disappointing about that is that GNATCOLL is presented by AdaCore (which presents itself as the public face of Ada) as one of the core useful-libraries and DOESN'T take advantage of Ada's nice portability instead relying on the crap-pile that is make and .configure even when they have/use GPR-files to represent a project themselves. (i.e. not using your own tools is a vote *against* your own product.)

I join you in this double disappointment.  Trying to make things work with ORM requires trial and error to understand what part of the Python script doesn't like your database configuration.  It took my days to make a useful session using sqlite, dropping some features supposed to be.  Yes, I was shocked by crashes without any explanation of tools to be used in high reliable environments.  And this as you said from the "face" of Ada.
 
> If you adopt the "eat your own dogfood" position, you have an incentive to make that a quality system, fix bugs, and make it easier to use.

But who will then do the coding on the project itself :-(

Re: Weakness of Ada is expensive / security / etc. ? Anything else?

[Florian Weimer]

* Trish Cayetano:

> -	Ada requires expensive compilers for large systems 

What is expensive?  $300 per year for a single seat, including
developer support?

Re: Weakness of Ada is expensive / security / etc. ? Anything else?

[Paul Rubin]

Florian Weimer <fw@deneb.enyo.de> writes:
> * Trish Cayetano:
>> -	Ada requires expensive compilers for large systems 
> What is expensive?  $300 per year for a single seat, including
> developer support?

What compiler is that?  I thought GNAT Pro was orders of magnitude more
expensive.  There are apparently no other Ada 2012 compilers, but maybe
older versions suffice.

Re: Weakness of Ada is expensive / security / etc. ? Anything else?

[Florian Weimer]

* Paul Rubin:

> Florian Weimer <fw@deneb.enyo.de> writes:
>> * Trish Cayetano:
>>> -	Ada requires expensive compilers for large systems 
>> What is expensive?  $300 per year for a single seat, including
>> developer support?
>
> What compiler is that?

GNAT with a developer subscription from Red Hat.

Re: Weakness of Ada is expensive / security / etc. ? Anything else?

[Paul Rubin]

Florian Weimer <fw@deneb.enyo.de> writes:
>> What compiler is that?
> GNAT with a developer subscription from Red Hat.

Is that basically the FOSS segment of GNAT, with some support added?
Does the target system have to be Red Hat (or at least Linux), as
opposed to embedded targets?  Otherwise it sounds too good to be true.

Re: Weakness of Ada is expensive / security / etc. ? Anything else?

[Florian Weimer]

* Paul Rubin:

> Florian Weimer <fw@deneb.enyo.de> writes:
>>> What compiler is that?
>> GNAT with a developer subscription from Red Hat.
>
> Is that basically the FOSS segment of GNAT, with some support added?

Yes, as far as I can see, it's the same bits you get as part of Red
Hat Enterpise Linux (and CentOS is built from pretty much the same
sources).  It's based on the FSF version, so you get the GNAT run time
exception.

> Does the target system have to be Red Hat (or at least Linux), as
> opposed to embedded targets?

Production support (after deployment) is separate, and Red Hat will
support deployment on Red Hat Enterprise Linux only, I assume.

So this is certainly not an option for small-device embedded
development, or high-assurance software.  But for those markets,
AdaCore's pricing is not unreasonable, I think.

Re: Weakness of Ada is expensive / security / etc. ? Anything else?

[Paul Rubin]

Florian Weimer <fw@deneb.enyo.de> writes:
>> Is that basically the FOSS segment of GNAT, with some support added?
> Yes, as far as I can see, it's the same bits you get as part of Red
> Hat Enterpise Linux (and CentOS is built from pretty much the same
> sources).  It's based on the FSF version, so you get the GNAT run time
> exception.

Fair enough, though that level of support doesn't seem that useful
(downloading FSF GNAT to my laptop and getting it working wasn't
difficult).  I guess GPL SPARK is not included.

> Production support (after deployment) is separate, and Red Hat will
> support deployment on Red Hat Enterprise Linux only, I assume.

I wonder if anyone is using it for real.

> So this is certainly not an option for small-device embedded
> development, or high-assurance software.  But for those markets,
> AdaCore's pricing is not unreasonable, I think.

The number I heard was at least 10x higher than single-seat licenses of
competing C/C++ and Forth products that I know of.  It does come with 5
seats minimum, but that's not of much use to a solo developer.  I also
don't know how much more you have to pay for the add-on tools that make
the product more interesting (SPARK Pro, CodePeer, etc.)

I'm not exactly criticizing AdaCore since they're probably doing the
right thing for their target market (aerospace companies, high-dollar
critical systems, etc).  But it seems to leave an unfilled need for the
low-budget embedded developer who wants a better language than C.

Re: Weakness of Ada is expensive / security / etc. ? Anything else?

[Florian Weimer]

* Paul Rubin:

> Florian Weimer <fw@deneb.enyo.de> writes:
>>> Is that basically the FOSS segment of GNAT, with some support added?
>> Yes, as far as I can see, it's the same bits you get as part of Red
>> Hat Enterpise Linux (and CentOS is built from pretty much the same
>> sources).  It's based on the FSF version, so you get the GNAT run time
>> exception.
>
> Fair enough, though that level of support doesn't seem that useful
> (downloading FSF GNAT to my laptop and getting it working wasn't
> difficult).

Well, the subscription comes with developer support.  (The
self-support option is $99 per year, with the same packages as far as
I know.)

> I guess GPL SPARK is not included.

It's not.

>> Production support (after deployment) is separate, and Red Hat will
>> support deployment on Red Hat Enterprise Linux only, I assume.
>
> I wonder if anyone is using it for real.

That's a question for Red Hat sales.  I couldn't possibly comment.

>> So this is certainly not an option for small-device embedded
>> development, or high-assurance software.  But for those markets,
>> AdaCore's pricing is not unreasonable, I think.
>
> The number I heard was at least 10x higher than single-seat licenses of
> competing C/C++ and Forth products that I know of.

But these products have unbundled support, and you pay only for
software licenses, right?  So the comparison isn't really fair.  For
GNAT, the equivalent self-support option is priced at a very
competitive $0.

> I'm not exactly criticizing AdaCore since they're probably doing the
> right thing for their target market (aerospace companies, high-dollar
> critical systems, etc).  But it seems to leave an unfilled need for the
> low-budget embedded developer who wants a better language than C.

I doubt it's commercially reasonable to support such developers, at
least not without constraining them to a specific (small set) of
target platforms.

Re: Weakness of Ada is expensive / security / etc. ? Anything else?

[Paul Rubin]

Florian Weimer <fw@deneb.enyo.de> writes:
>> The number I heard was at least 10x higher than single-seat licenses of
>> competing C/C++ and Forth products that I know of.
> But these products have unbundled support, and you pay only for
> software licenses, right?  So the comparison isn't really fair.

The other products do have good support, though I don't know how to
compare it with Adacore support, which sounds very high end.

> For GNAT, the equivalent self-support option is priced at a very
> competitive $0.

No I don't think so, there's a lot of software not included with the
free system, there's lots of unsupported processors that the commercial
package supports, and there is also the GPL runtime issue.  I'm fine
with the idea that someone wanting to ship a non-free product shouldn't
whine about having to use non-free tools, but once you've made the step
into proprietary toolchains, it's ok to be cost conscious when comparing
them.

>> it seems to leave an unfilled need for the low-budget embedded
>> developer who wants a better language than C.
>
> I doubt it's commercially reasonable to support such developers, at
> least not without constraining them to a specific (small set) of
> target platforms.

Sure, I wouldn't expect support for every weird processor out there, but
there are many non-Ada compiler vendors who support a reasonable target
range affordably to small developers.  There doesn't seem to be anything
comparable for Ada.

Re: Weakness of Ada is expensive / security / etc. ? Anything else?

[Florian Weimer]

* Paul Rubin:

> Florian Weimer <fw@deneb.enyo.de> writes:
>>> The number I heard was at least 10x higher than single-seat licenses of
>>> competing C/C++ and Forth products that I know of.
>> But these products have unbundled support, and you pay only for
>> software licenses, right?  So the comparison isn't really fair.
>
> The other products do have good support, though I don't know how to
> compare it with Adacore support, which sounds very high end.

Can you name a few examples?  I have no idea what kind of tools people
use in the embedded world, and what the support and licensing
agreements are like.

Re: Weakness of Ada is expensive / security / etc. ? Anything else?

[Paul Rubin]

Florian Weimer <fw@deneb.enyo.de> writes:
>> The other products do have good support, though I don't know how to
>> compare it with Adacore support, which sounds very high end.
>
> Can you name a few examples?  I have no idea what kind of tools people
> use in the embedded world, and what the support and licensing
> agreements are like.

http://www.forth.com/embedded/index.html
http://mpeforth.com/xc7.htm
http://www.keil.com/
http://bytecraft.com/

Re: Weakness of Ada is expensive / security / etc. ? Anything else?

[Florian Weimer]

* Paul Rubin:

> Florian Weimer <fw@deneb.enyo.de> writes:
>>> The other products do have good support, though I don't know how to
>>> compare it with Adacore support, which sounds very high end.
>>
>> Can you name a few examples?  I have no idea what kind of tools people
>> use in the embedded world, and what the support and licensing
>> agreements are like.
>
> http://www.forth.com/embedded/index.html

Haven't looked at these.

> http://mpeforth.com/xc7.htm

Support for that is surprisingly affordable, around $600 per year for
a cross-compiler:

<http://www.mpeforth.com/pricelist.htm>

That's surprisingly affordable for an embedded solution, but it is
Forth.  Depending on what the application does, royalties may apply,
though:

<http://www.mpeforth.com/vfxlicence.htm>

> http://www.keil.com/

Here's a leaked price list:

<http://www.tagor.rs/files/cenovnici/Keil_pricelist_USD_2012.pdf>

Based on that, it appears that licenses cost several thousand U.S.
dollars.  The product seems to come with support, though, but it's not
clear for which time frame.  If you need the yearly fee option for
continuous support, the pricing is roughly in AdaCore's ballpark.

> http://bytecraft.com/

These compilers are not *that* expensive if you need only specific
targets:

<http://www.mouser.com/catalog/645/usd/260.pdf>
<http://www.tec-i.com/ByteCraftLtdCompilers.htm>

There is no information about the support included.

Re: Weakness of Ada is expensive / security / etc. ? Anything else?

[Jeremiah]

On Sunday, July 26, 2015 at 3:41:19 PM UTC-4, Florian Weimer wrote: 
> > http://www.keil.com/
> 
> Here's a leaked price list:
> 
> <http://www.tagor.rs/files/cenovnici/Keil_pricelist_USD_2012.pdf>
> 
> Based on that, it appears that licenses cost several thousand U.S.
> dollars.  The product seems to come with support, though, but it's not
> clear for which time frame.  If you need the yearly fee option for
> continuous support, the pricing is roughly in AdaCore's ballpark.

The prices listed there are slightly higher than they are now (difference in prices is in the low hundreds).  The 1 year license is a "1 year of use" while the really expensive one is a "perpetual use".  If you want updates or support after the first year, you have to buy that separately even if you get the "perpetual" license.  The free version of this compiler only allows up to 2k of compiled code, which is pretty limiting.

I use a lot of Microchip parts...typically the PIC24 series.  C compilers range from the early thousands down to nearly free.  I use Custom Computer Services series compilers which run pretty cheap.  I by all of the different lines and the IDE at once for a one time fee of $600-700 and if I want the yearly updates, I have to pay $275 for the first license and $130ish for any additional licenses per year.  If you don't need all the different targets or the IDE, you can buy command line versions of the specific target lines you want (there are 4 different lines) for much less.  And you don't need the yearly updates if you have a good build (also, if there is a really bad problem, they will sometimes let you update for free).  The free compilers for these type of parts typically only have optimization limits but not code size limits like the Keil tools.

I long for the day I can get an Ada compiler for the 16-bit processors in the Microchip lines, but I doubt it will ever come.  They have a very different architecture from most GCC targets, so it wouldn't be as simple as porting an Arm or MIPS target.  The 32 bit parts are MIPS like, but they are not as well designed as the 16bit parts in terms of size and power consumption, which is a big deal in the part of the embedded industry that I work in.  I hate the feeling that it is just so out of reach.   

Re: Weakness of Ada is expensive / security / etc. ? Anything else?

[Stefan.Lucks]

[-- Attachment #1: Type: text/plain, Size: 915 bytes --]

On Mon, 20 Jul 2015, Florian Weimer wrote:

> Date: Mon, 20 Jul 2015 21:33:01 +0200
> From: Florian Weimer <fw@deneb.enyo.de>
>> I guess GPL SPARK is not included.
>
> It's not.

But it is not difficult to *develope* your software using gnat/GPL (with 
SPARK support) and then *compile* the final production executable using 
gnat/FSF. The compilers are essentially the same. Well, the fronend of 
gnat/FSF appears to be a bit behind the gnat/GPL frontend. When Ada 2012 
was fresh, gnat/FSF needed some time to include new Ada features already 
supported by gnat/GPL. But this issue is moot, now ... at least until the 
next revision of the Standard.

Stefan


--------  I  love  the  taste  of  Cryptanalysis  in  the morning!  --------
www.uni-weimar.de/de/medien/professuren/mediensicherheit/people/stefan-lucks
----Stefan.Lucks (at) uni-weimar.de, Bauhaus-Universität Weimar, Germany----

Re: Weakness of Ada is expensive / security / etc. ? Anything else?

[EGarrulo]

I am new to Ada, but so far I have noticed a few things...

*Strenghts*:

- compiles to native executables, hence no need for runtimes;

- no need for runtimes, hence your application will start quickly;

- executables are small (as a comparison, compilers for other languages may produce executables of many megabytes for simple programs);

- fast executables (comparable to C);

- generic code does not carry a runtime penalty (like in Java);

- consistent design of the language;

- everything is explicit, hence the compiler does not perform any clever tricks beyond your back (like in C++).

*Weaknesses*:

> -	Ada complete language is too complex. True or False? If true, how?

Rather than "complex", I would say "rich", because there are lots of features that are meant to tune your application in different ways, but you can safely ignore them while you learn (this is not the case with C++).

> -	Ada requires expensive compilers for large systems 

Even if this were the case -- and it is not, because of the FSF GNAT -- who would develop large systems without enough financial resources?  And when you are developing large systems, you should know that other (missing) qualities of your tools may cost you much more than what you saved by relying on the wrong tool because it was cheap.  On the other hand, expensive compilers are a problem for small shops, but this is not the case anyway for Ada, as said.

> -	Ada lacks that web presence greatly. comp.lang.ada is the only location on the internet with an active Ada community.

Web presence doesn't impress the knowledgeable professional.  What matters is whether you can get answers or not, and the quality of those answers (the so called "signal to noise ratio").

Other weaknesses:

- books are outdated: AFAIK, there is only one (recommended) book that is based on Ada 2012; all other available books are based on Ada 95 (while the basics  may be still relevant, you may wish to learn modern techniques from the start);

- the GNAT GPS is good, but lacks some polishing (and I am using a version that  freezes randomly); I don't know what are the alternatives... maybe I will try Emacs;

- manual memory management (garbage collection is optional, but in practice not available; apparently there are implementations of the "Shared Reference/Weak Reference" pattern, but I have yet to investigate them).

- lack of introspection for generics: you have to "spoon-feed" the compiler to instantiate a generic unit, when the compiler should infer types on its own: this makes the code more verbose than it should.