A scapegoat once again :(

A scapegoat once again :(

[Britt Snodgrass]

Seen on DailyTech today:

http://www.dailytech.com/F22+Upgrades+Over+Budget+and+Behind+Schedule+F35+May+Cost+1T+in+Operating+Costs/article21692.htm

"The cause of the delay in delivery stems from the programming
language used called Ada. The Ada language was once a DoD standard,
but the use of the language has waned in the last 15 years. Analyst
Loren Thompson from the Lexington Institute said, "It tends to impede
quick upgrades to the system to which it is the base software."

It’s really sad and disturbing to read such an incorrect and
uninformed statement. How could Ada "impede quick upgrades" unless
they really mean that the coders they've been hiring recently are
intimidated by compiler warnings.

- Britt

Re: A scapegoat once again :(

[Hyman Rosen]

On 5/20/2011 11:13 AM, Britt Snodgrass wrote:
> "It tends to impede quick upgrades to the system to
 >  which it is the base software."

I might naively imagine that impeding quick upgrades to
fighter jet software is a feature, not a bug.

Re: A scapegoat once again :(

[Shark8]

On May 20, 10:22 am, Hyman Rosen <hyro...@mail.com> wrote:
> On 5/20/2011 11:13 AM, Britt Snodgrass wrote:> "It tends to impede quick upgrades to the system to
>
>  >  which it is the base software."
>
> I might naively imagine that impeding quick upgrades to
> fighter jet software is a feature, not a bug.

Yeah.. JetFighter-systems aren't exactly a cost-effective place to
have a
"run it and see if it crashes" approach to development/debugging like
you can do with PHP.

Re: A scapegoat once again :(

[jonathan]

Off topic but mildly entertaining - our friend Mr. Thompson is
is in the news again.  The plot thickens.

http://blogs.forbes.com/bruceupbin/2011/05/24/spacex-responds-to-forbes-contributor-loren-thompson/

Re: A scapegoat once again :(

[jonathan]

On May 20, 4:13 pm, Britt Snodgrass <britt.snodgr...@gmail.com> wrote:
> Seen on DailyTech today:
>
> http://www.dailytech.com/F22+Upgrades+Over+Budget+and+Behind+Schedule...
>

The expert's bio  (from his web page):

Loren B. Thompson is Chief Operating Officer of the non-profit
Lexington Institute and Chief Executive Officer of Source Associates,
a for-profit consultancy. Prior to holding his present positions, he
was Deputy Director of the Security Studies Program at Georgetown
University and taught graduate-level courses in strategy, technology
and media affairs at Georgetown. He has also taught at Harvard
University's Kennedy School of Government.

Mr. Thompson holds doctoral and masters degrees in government from
Georgetown University and a bachelor of science degree in political
science from Northeastern University.

Re: A scapegoat once again :(

[Dmitry A. Kazakov]

On Fri, 20 May 2011 08:31:22 -0700 (PDT), jonathan wrote:

> On May 20, 4:13�pm, Britt Snodgrass <britt.snodgr...@gmail.com> wrote:
>> Seen on DailyTech today:
>>
>> http://www.dailytech.com/F22+Upgrades+Over+Budget+and+Behind+Schedule...
> 
> The expert's bio  (from his web page):
> 
> Loren B. Thompson is Chief Operating Officer of the non-profit
> Lexington Institute and Chief Executive Officer of Source Associates,
> a for-profit consultancy. Prior to holding his present positions, he
> was Deputy Director of the Security Studies Program at Georgetown
> University and taught graduate-level courses in strategy, technology
> and media affairs at Georgetown. He has also taught at Harvard
> University's Kennedy School of Government.
> 
> Mr. Thompson holds doctoral and masters degrees in government from
> Georgetown University and a bachelor of science degree in political
> science from Northeastern University.

Political science, how impressing.

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

Re: A scapegoat once again :(

[Nasser M. Abbasi]

On 5/20/2011 8:13 AM, Britt Snodgrass wrote:
> Seen on DailyTech today:
>
> http://www.dailytech.com/F22+Upgrades+Over+Budget+and+Behind+Schedule+F35+May+Cost+1T+in+Operating+Costs/article21692.htm
>
> "The cause of the delay in delivery stems from the programming
> language used called Ada. The Ada language was once a DoD standard,
> but the use of the language has waned in the last 15 years. Analyst
> Loren Thompson from the Lexington Institute said, "It tends to impede
> quick upgrades to the system to which it is the base software."
>
> It�s really sad and disturbing to read such an incorrect and
> uninformed statement. How could Ada "impede quick upgrades" unless
> they really mean that the coders they've been hiring recently are
> intimidated by compiler warnings.
>
> - Britt


Compared to updating one's Javascript and Matlab scripts and such,
yes, I completely agree with the author, Ada takes more effort to
upgrade, since must recompile things! Very annoying thing to have
to do just to upgrade software.

I think all flight control systems, nuclear plants software and
such should be rewritten from scratch in HTML, CCS, python,
Javascript, flash, and other new cool, modern tools.

This way one can upgrade the software by editing the page on the fly,
and see the effect show up instantly !  Easy upgrade path, edit and try,
edit and try, edit and try, this is how things are supposed to be.

The above is the modern way of developing software. No more painful
compiling and such. Scripting is the way to go.

The author is right on. Compiling and fixing compiler error and then
linking is so like 1980's and not cool anymore.

--Nasser

Re: A scapegoat once again :(

[Hyman Rosen]

But look at this: <http://www.swep-eds.com/XD%20Ada/XD%20Ada.html>

     XD Ada
     XD Ada Cross Compilers

     In partnership with HP (formerly Digital Equipment Corporation)
     the Software Engineering and Products Group has developed the
     world�s premier family of Ada cross-development toolsets, XD Ada.

     XD Ada is targeted to the Motorola M68000 family (including the
     MC683xx micro-controllers) and MIL-STD-1750A microprocessor
     architecture.

     Product options exist for multiprocessor communications and
     debugging, and for Emulator and Simulator support.

     Hosted on OpenVMS Systems, XD Ada is fully integrated with the
     OpenVMS Debugger and HP DECset CASE tools, providing a unique and
     comprehensive software development environment. XD Ada is available
     on OpenVMS VAX, Alpha and Integrity Servers.

     XD Ada produces excellent code quality through the use of powerful
     global optimisation techniques. XD Ada delivers outstanding
     performance on real time projects against proven industry-accepted
     benchmarks.

     XD Ada is the standard Ada development cross-compiler for the
     four-nation Fighter Aircraft Replacement programme through the
     Eurofighter consortia. In the USA, it was selected by Boeing for
     their 777 and F-22 applications.

     Users of XD Ada can be assured of comprehensive, long-term support
     on current or specific versions of the tools.

How much do you want to bet that the F-22 software still needs to be
developed on this ancient programming environment? If so, it's hardly
any wonder that using it would cause significant delays.

Re: A scapegoat once again :(

[Georg Bauhaus]

On 20.05.11 17:59, Hyman Rosen wrote:

> How much do you want to bet that the F-22 software still needs to be
> developed on this ancient programming environment? If so, it's hardly
> any wonder that using it would cause significant delays.

TTBOMK, one or two companies, one in Switzerland, are actively developing
and selling efficient VAX or Alpha virtualization (if OpenVMS on
Intel can't be used) that runs on standard PC hardware.

So you'd get Common Desktop Environment (on X); good tools for
Ada development are then not difficult to obtain.

Re: A scapegoat once again :(

[Phil Clayton]

On May 20, 4:59 pm, Hyman Rosen <hyro...@mail.com> wrote:
> But look at this: <http://www.swep-eds.com/XD%20Ada/XD%20Ada.html>
>
>      ...
>
>      XD Ada is the standard Ada development cross-compiler for the
>      four-nation Fighter Aircraft Replacement programme through the
>      Eurofighter consortia.

Surely this stopped being the case for Eurofighter/Typoon Tranche 2
due to introduction of PowerPCs?  I have a feeling that the text is
now out of date.


> How much do you want to bet that the F-22 software still needs to be
> developed on this ancient programming environment? If so, it's hardly
> any wonder that using it would cause significant delays.

XD Ada M68000 is available for the CHARON-VAX emulator environment -
what's the problem? :)

Ah yes, no Ada 95...

Re: A scapegoat once again :(

[Vinzent Hoefler]

Phil Clayton wrote:

> On May 20, 4:59 pm, Hyman Rosen <hyro...@mail.com> wrote:
>> But look at this: <http://www.swep-eds.com/XD%20Ada/XD%20Ada.html>
>>
>>      ...
>>
>>      XD Ada is the standard Ada development cross-compiler for the
>>      four-nation Fighter Aircraft Replacement programme through the
>>      Eurofighter consortia.
>
> Surely this stopped being the case for Eurofighter/Typoon Tranche 2
> due to introduction of PowerPCs?

Yes, it did. There's still some level of support for Tranche 1, though.


Vinzent.

-- 
A C program is like a fast dance on a newly waxed dance floor by people carrying
razors.
   --  Waldi Ravens

Re: A scapegoat once again :(

[Vinzent Hoefler]

Hyman Rosen wrote:

> But look at this: <http://www.swep-eds.com/XD%20Ada/XD%20Ada.html>
>
[...]
>      XD Ada is the standard Ada development cross-compiler for the
>      four-nation Fighter Aircraft Replacement programme through the
>      Eurofighter consortia. In the USA, it was selected by Boeing for
>      their 777 and F-22 applications.

Well, that info is a little bit outdated, I can assure you - at least
for the European part.

>      Users of XD Ada can be assured of comprehensive, long-term support
>      on current or specific versions of the tools.
>
> How much do you want to bet that the F-22 software still needs to be
> developed on this ancient programming environment? If so, it's hardly
> any wonder that using it would cause significant delays.

I can tell you about "delays" (actually it's more like a pipe-line) in
software upgrades in Avionics. The paperwork to be done is enormous. Even
if you "hack" in a little code to just change a minor aspect of the software
it needs to get tested at various stages, formalized with a lot of design,
test, and $whatever documentation and maybe it gets flight-tested at some
point in time. But before it actually starts running inside a real
customer-delivered aircraft, I'd estimate a time-span of 24 months since
the first change in requirements were settled.

And that's not likely to change with _any_ programming language.


Vinzent.

-- 
A C program is like a fast dance on a newly waxed dance floor by people carrying
razors.
   --  Waldi Ravens

Re: A scapegoat once again :(

[Dmitry A. Kazakov]

On Fri, 20 May 2011 08:44:48 -0700, Nasser M. Abbasi wrote:

> On 5/20/2011 8:13 AM, Britt Snodgrass wrote:
>> Seen on DailyTech today:
>>
>> http://www.dailytech.com/F22+Upgrades+Over+Budget+and+Behind+Schedule+F35+May+Cost+1T+in+Operating+Costs/article21692.htm
>>
>> "The cause of the delay in delivery stems from the programming
>> language used called Ada. The Ada language was once a DoD standard,
>> but the use of the language has waned in the last 15 years. Analyst
>> Loren Thompson from the Lexington Institute said, "It tends to impede
>> quick upgrades to the system to which it is the base software."
>>
>> ItοΏ½s really sad and disturbing to read such an incorrect and
>> uninformed statement. How could Ada "impede quick upgrades" unless
>> they really mean that the coders they've been hiring recently are
>> intimidated by compiler warnings.
> 
> Compared to updating one's Javascript and Matlab scripts and such,
> yes, I completely agree with the author, Ada takes more effort to
> upgrade, since must recompile things! Very annoying thing to have
> to do just to upgrade software.

That reminds me an old saying about training astronauts to climb trees,
that gets them closer to the Moon.

He probably meant that upgrades written in C or Java do not work and
customers are refusing to upgrade from old working Ada software, a boring
obstacle indeed.

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

Re: A scapegoat once again :(

[Georg Bauhaus]

On 20.05.11 17:44, Nasser M. Abbasi wrote:
> On 5/20/2011 8:13 AM, Britt Snodgrass wrote:
>> Seen on DailyTech today:
>>
>> http://www.dailytech.com/F22+Upgrades+Over+Budget+and+Behind+Schedule+F35+May+Cost+1T+in+Operating+Costs/article21692.htm

> Compared to updating one's Javascript and Matlab scripts and such,
> yes, I completely agree with the author, Ada takes more effort to
> upgrade, since must recompile things! Very annoying thing to have
> to do just to upgrade software.

Trying to find reason in this, other than social, political,
or ignorance of change engineering.  Switching Java class
files as an alternative, not whole systems?  Even technically?

The phrase in question is a claim.

Made by an analyst, good heavens, stated like its said by an expert
(see jonathan's research on this), or someone representing
experts who would not have a need to give reasons for their
claims.

But more importantly, why are tax payers not holding
*journalists* accountable?!  US citizens, if not NATO
citizens, are paying for a most advanced staffed weapon system.
And they silently accept it when masses of their money are dealt
with this kind of journalistic accuracy?

Adventuring into Java class file replacement might
appear to proceed faster than a whole program build,
test, etc.  If he is thinking of Java.   I'm not sure
the target hardware is on the list of JIT supporting
JVM platforms, though ...


The people's voice, in a comment regarding Ada:
"They really need to contemporize their software of they will
always be playing catchup."

Can someone please tell voters the age of the most
commonly used programming language in embedded systems?
Or of C++?

Re: A scapegoat once again :(

[Simon Wright]

"Nasser M. Abbasi" <nma@12000.org> writes:

> I think all flight control systems, nuclear plants software and
> such should be rewritten from scratch in HTML, CCS, python,
> Javascript, flash, and other new cool, modern tools.

And then send Loren Thompson on the test flight, please.

Re: A scapegoat once again :(

[Vinzent Hoefler]

Simon Wright wrote:

> "Nasser M. Abbasi" <nma@12000.org> writes:
>
>> I think all flight control systems, nuclear plants software and
>> such should be rewritten from scratch in HTML, CCS, python,
>> Javascript, flash, and other new cool, modern tools.
>
> And then send Loren Thompson on the test flight, please.

YMMD. :)


Vinzent.

-- 
A C program is like a fast dance on a newly waxed dance floor by people carrying
razors.
   --  Waldi Ravens

Re: A scapegoat once again :(

[Robert Love]

On 2011-05-20 10:13:38 -0500, Britt Snodgrass said:

> Seen on DailyTech today:
> 
> http://www.dailytech.com/F22+Upgrades+Over+Budget+and+Behind+Schedule+F35+May+Cost+1T+in+Operating+Costs/article21692.htm

"The 
> 
> cause of the delay in delivery stems from the programming
> language used called Ada. The Ada language was once a DoD standard,
> but the use of the language has waned in the last 15 years. Analyst
> Loren Thompson from the Lexington Institute said, "It tends to impede
> quick upgrades to the system to which it is the base software."
> 
> It’s really sad and disturbing to read such an incorrect and
> uninformed statement. How could Ada "impede quick upgrades" unless
> they really mean that the coders they've been hiring recently are
> intimidated by compiler warnings.
> 
> - Britt


And by "once again" are you refering to the Secretary of the Air Force 
being told that Ada delayed SBIRS High by LM management?

Or is there some other horror story.

Re: A scapegoat once again :(

[Vinzent Hoefler]

Britt Snodgrass wrote:

> Seen on DailyTech today:
>
> http://www.dailytech.com/F22+Upgrades+Over+Budget+and+Behind+Schedule+F35+May+Cost+1T+in+Operating+Costs/article21692.htm
>
> "The cause of the delay in delivery stems from the programming
> language used called Ada. The Ada language was once a DoD standard,
> but the use of the language has waned in the last 15 years. Analyst
> Loren Thompson from the Lexington Institute said, "It tends to impede
> quick upgrades to the system to which it is the base software."

Yeah right. What an idiot.

Maybe they should do it all in pluggable Javascript, some C++ for the
time-critical parts and a bit of C# as glue code.

I can imaging the face of the pilot upon

"The flight-control software has stopped working.
  Windows is checking for a solution..."

For the sake of the pilot let's just hope that the ejection seat control is
hardwired and not controlled by the same "easily upgradable" crap.


Vinzent.

-- 
A C program is like a fast dance on a newly waxed dance floor by people carrying
razors.
   --  Waldi Ravens

Re: A scapegoat once again :(

[Gautier write-only]

On 20 mai, 17:13, Britt Snodgrass <britt.snodgr...@gmail.com> wrote:
> Seen on DailyTech today:
>
> http://www.dailytech.com/F22+Upgrades+Over+Budget+and+Behind+Schedule...
>
> "The cause of the delay in delivery stems from the programming
> language used called Ada. The Ada language was once a DoD standard,
> but the use of the language has waned in the last 15 years. Analyst
> Loren Thompson from the Lexington Institute said, "It tends to impede
> quick upgrades to the system to which it is the base software."
>
> It’s really sad and disturbing to read such an incorrect and
> uninformed statement. How could Ada "impede quick upgrades" unless
> they really mean that the coders they've been hiring recently are
> intimidated by compiler warnings.

If it can reassure you, I don't think the person behind that statement
knows what a compiler warning is :-). It sounds more like standard "IT
fashionista" speech from a not very technical person. Typically they
want "long-lasting technologies", the "newest technology", and "quick
upgrades" at the same time. They just cannot imagine that there could
be a contradiction in all that. It would mean they have a knowledge of
what makes a technology suitable for long-term projects.

Fortunately, now that time is passing and IT is stabilizing, you can
show with examples and nice Powerpoint slides why flashy technologies
are not suitable for projects due to last for more than 5 years. In
many cases, upgrades with MS technologies just don't work. In a recent
case a few days ago (porting a legacy .NET+SQL software to a new
server), the upgrade wizard stopped with a very verbose statement
"Type mismatch". No more info, like for instance the location. Bye-
bye. Fortunately I have made a replacement in Ada. So I was able to
say: "Don't worry, we'll install the old MS framework since the
software itself won't be updated; we have a replacement that avoids
using technologies with programmed obsolescence." :-)

G.

Re: A scapegoat once again :(

[Gautier write-only]

On May 20, 5:13 pm, Britt Snodgrass <britt.snodgr...@gmail.com> wrote:

> Seen on DailyTech today:
>
> http://www.dailytech.com/F22+Upgrades+Over+Budget+and+Behind+Schedule...
>
> "The cause of the delay in delivery stems from the programming
> language used called Ada. The Ada language was once a DoD standard,
> but the use of the language has waned in the last 15 years. Analyst
> Loren Thompson from the Lexington Institute said, "It tends to impede
> quick upgrades to the system to which it is the base software."
>
> It’s really sad and disturbing to read such an incorrect and
> uninformed statement. How could Ada "impede quick upgrades" unless
> they really mean that the coders they've been hiring recently are
> intimidated by compiler warnings.

It is especially wrong considering upgrades: it is difficult to have
quicker upgrades than with Ada (at least, GNAT):
- Install the new GNAT (a few minutes)
- *That's it* ! No need to spend a week upgrading sources or running
upgrade wizards like with MS tools for instance.
______________________________________________________________________________
Gautier's Ada programming -- http://gautiersblog.blogspot.com/search/label/Ada
NB: follow the above link for a valid e-mail address

Re: A scapegoat once again :(

[Ludovic Brenta]

Gautier write-only wrote:
> Britt Snodgrass wrote:
>
>> Seen on DailyTech today:
>>
>>http://www.dailytech.co/F22+Upgrades+Over+Budget+and+Behind+Schedule...
>>
>> "The cause of the delay in delivery stems from the programming
>> language used called Ada. The Ada language was once a DoD standard,
>> but the use of the language has waned in the last 15 years. Analyst
>> Loren Thompson from the Lexington Institute said, "It tends to impede
>> quick upgrades to the system to which it is the base software."
>>
>> It’s really sad and disturbing to read such an incorrect and
>> uninformed statement. How could Ada "impede quick upgrades" unless
>> they really mean that the coders they've been hiring recently are
>> intimidated by compiler warnings.
>
> It is especially wrong considering upgrades: it is difficult to have
> quicker upgrades than with Ada (at least, GNAT):
> - Install the new GNAT (a few minutes)
> - *That's it* ! No need to spend a week upgrading sources or running
> upgrade wizards like with MS tools for instance.

They were not talking about upgrading the development environment.
They were talking about upgrading the embedded life-critical software
on board the aircraft.  That does not diminish the utter stupidity of
the statement.  I think there are at least two idiots at work here:
one inside the project who has heard of Ada and wants to get rid of
it, and one "expert" who doesn't know what software is and repeats
whatever the first one said.  Both should be held accountable for
their incompetence.

--
Ludovic Brenta.

Re: A scapegoat once again :(

[Adam Beneschan]

On May 23, 5:25 am, Ludovic Brenta <ludo...@ludovic-brenta.org> wrote:

> I think there are at least two idiots at work here:
> one inside the project who has heard of Ada and wants to get rid of
> it, and one "expert" who doesn't know what software is and repeats
> whatever the first one said.  Both should be held accountable for
> their incompetence.

Then there's the journalist who repeated the words of the "expert"
without doing any checking---apparently not noticing that the expert's
background is in political science.  So I make it at least three.

                         -- Adam

Re: A scapegoat once again :(

[Rick]

While I share the sense of righteous indignation coming through the
postings in this thread, I cannot help but feel that they are wasted.
After all, postings on this group preach to the converted.

I would be much happier if these comments were getting to the people
responsible for publishing the idiocy which instigated this
conversation.  Getting to the Editors who allow their staff to write
unchallenged drivel would have a much more profound effect.

Now we have the issue, who do we write to?

Re: A scapegoat once again :(

[AdaMagica]

On 24 Mai, 09:41, Rick <rickdu...@gmail.com> wrote:
> While I share the sense of righteous indignation coming through the
> postings in this thread, I cannot help but feel that they are wasted.
> After all, postings on this group preach to the converted.
>
> I would be much happier if these comments were getting to the people
> responsible for publishing the idiocy which instigated this
> conversation.  Getting to the Editors who allow their staff to write
> unchallenged drivel would have a much more profound effect.
>
> Now we have the issue, who do we write to?

There are a lot of comments on that page, and in fact a few from
people who know what they are talking about. Especially EmbeddedSwEng.
But will this Shane McGlaun read the comments, let alone forward them
to the idiots involved?