[gentoo-user] OT: old farts slum_code enforcement

[gentoo-user] OT: old farts slum_code enforcement

[James]

> Alan McKinnon <alan.mckinnon <at> gmail.com> writes:

> You are a C man.
> Working with bash must be .... excruciatingly painful 

Ah yes, State machine design; not much fittering around with
escaping silly little symbols.....

sh/csh/bash/scripting is not bad.  I just 'lift' the tough stuff
from others mostly. With some codes, like Java, you read
and follow 90%, then there is another code to find and read.
It never ends, on and on and on....  Then instead of one choice
you have 3 or 4 choices....... Maven is a whole nutter beast.... 
Why it is becoming so important is still a wee bit confusing to me.
[1] http://maven.apache.org/


I'm just not use to that sort of world. In embedded, you
over design before you begin coding. You do not have megabytes
of lib stuff to find and read and test  the dozens of variants.

I like to code. It's debugging the stuff that drives me crazy(er?).....
These kids have no respect for us old farts. I remember when
"numerical recipies in C" [2] was the stuff. Now it's C++ or Java. [3]
I have a book on my shelf (where it belongs) on "Concurrency  State
Moels & Java" by Magee and Kramer, 1999 - Wiley. What a hoot!

I'm still looking for "Numerical recipies in Bash" ?
Ju gonna code that up?

Howz your Fortan 90? I did not even know there was such an
ugly beast [4]. I thought Fortran was outlawed decades ago.
I guess nothing ever dies. Physicists have to have a language
to themselves.


Anyway, my_python is comming along...... 
(pist, don't tell anyone, but I almost, (almost) like python).


[1] http://maven.apache.org/
[2] http://www.nrbook.com/a/bookcpdf.php
[3] http://www.nr.com/
[4] https://gcc.gnu.org/wiki/GFortranStandards



hth,
James

Re: [gentoo-user] OT: old farts slum_code enforcement

[Alan McKinnon]

On 15/09/2014 00:21, James wrote:
> 
>> Alan McKinnon <alan.mckinnon <at> gmail.com> writes:
> 
>> You are a C man.
>> Working with bash must be .... excruciatingly painful 
> 
> Ah yes, State machine design; not much fittering around with
> escaping silly little symbols.....

python fixes all of that (see below)

> 
> sh/csh/bash/scripting is not bad.  I just 'lift' the tough stuff
> from others mostly. With some codes, like Java, you read
> and follow 90%, then there is another code to find and read.
> It never ends, on and on and on....  Then instead of one choice
> you have 3 or 4 choices....... Maven is a whole nutter beast.... 
> Why it is becoming so important is still a wee bit confusing to me.
> [1] http://maven.apache.org/


maven is important to enterprise users because Java is important to
enterprise customers. Java is important to them because huge numbers of
apps that enterprise likes use Java. It's not so much the language
itself (any language is almost as good as any other) but the whole Java
ecosystem. I see maven like this:

Maven is to Java what CPAN is to Perl

> 
> 
> I'm just not use to that sort of world. In embedded, you
> over design before you begin coding. You do not have megabytes
> of lib stuff to find and read and test  the dozens of variants.
> 
> I like to code. It's debugging the stuff that drives me crazy(er?).....
> These kids have no respect for us old farts. I remember when
> "numerical recipies in C" [2] was the stuff. Now it's C++ or Java. [3]
> I have a book on my shelf (where it belongs) on "Concurrency  State
> Moels & Java" by Magee and Kramer, 1999 - Wiley. What a hoot!

I'm looking for two books, and modern kids laugh at me when I mention
them. How little they know :-)

Wirth: Algorithms + Data Structures = Programs
Knuth: The Art of Computer Programming (I want the whole complete set of
this one) (in dead tree form)


> 
> I'm still looking for "Numerical recipies in Bash" ?
> Ju gonna code that up?

You'll have to write that one yourself, I doubt anyone has done it yet :-)

> 
> Howz your Fortan 90? I did not even know there was such an
> ugly beast [4]. I thought Fortran was outlawed decades ago.
> I guess nothing ever dies. Physicists have to have a language
> to themselves.

I believe Fortran is still very much alive and well in engineering and
physics - 40 years of number crunching code doesn't just go away by itself

> Anyway, my_python is comming along...... 
> (pist, don't tell anyone, but I almost, (almost) like python).

My most favourite language of all time!
Runs about as fast as reasonably complex shell code, but because it's an
interpreter and not executed directly by the shell, there's no auto
globbing and weird bash expansion going on. Effect = all that
mind-bending psychosis-inducing escape nonsense just falls away.
printf reduces to something a mortal human might even grok.
control structures look sane - no fi, esac, or elif. Do's and while's
look like real do's and while's

I could go on, but you get the idea :-)



-- 
Alan McKinnon
alan.mckinnon@gmail.com

Re: [gentoo-user] OT: old farts slum_code enforcement

[J. Roeleveld]

On Monday, September 15, 2014 07:21:18 AM Alan McKinnon wrote:
> On 15/09/2014 00:21, James wrote:
> >> Alan McKinnon <alan.mckinnon <at> gmail.com> writes:
> >> 
> >> You are a C man.
> >> Working with bash must be .... excruciatingly painful
> > 
> > Ah yes, State machine design; not much fittering around with
> > escaping silly little symbols.....
> 
> python fixes all of that (see below)

I like state machines.

> > sh/csh/bash/scripting is not bad.  I just 'lift' the tough stuff
> > from others mostly. With some codes, like Java, you read
> > and follow 90%, then there is another code to find and read.
> > It never ends, on and on and on....  Then instead of one choice
> > you have 3 or 4 choices....... Maven is a whole nutter beast....
> > Why it is becoming so important is still a wee bit confusing to me.
> > [1] http://maven.apache.org/
> 
> maven is important to enterprise users because Java is important to
> enterprise customers. Java is important to them because huge numbers of
> apps that enterprise likes use Java. It's not so much the language
> itself (any language is almost as good as any other) but the whole Java
> ecosystem. I see maven like this:
> 
> Maven is to Java what CPAN is to Perl

Actually, enterprise customers just want big packages supplied by the vendor 
without having to do any further heavy lifting. That allows them to complain 
to the vendor when the product doesn't work as expected. (Never mind it works 
as designed)

> > I'm just not use to that sort of world. In embedded, you
> > over design before you begin coding. You do not have megabytes
> > of lib stuff to find and read and test  the dozens of variants.
> > 
> > I like to code. It's debugging the stuff that drives me crazy(er?).....
> > These kids have no respect for us old farts. I remember when
> > "numerical recipies in C" [2] was the stuff. Now it's C++ or Java. [3]
> > I have a book on my shelf (where it belongs) on "Concurrency  State
> > Moels & Java" by Magee and Kramer, 1999 - Wiley. What a hoot!
> 
> I'm looking for two books, and modern kids laugh at me when I mention
> them. How little they know :-)
> 
> Wirth: Algorithms + Data Structures = Programs
> Knuth: The Art of Computer Programming (I want the whole complete set of
> this one) (in dead tree form)

Did you try google? :)

> > I'm still looking for "Numerical recipies in Bash" ?
> > Ju gonna code that up?
> 
> You'll have to write that one yourself, I doubt anyone has done it yet :-)

Are there enough "old farts" left to generate sufficient interest?

> > Howz your Fortan 90? I did not even know there was such an
> > ugly beast [4]. I thought Fortran was outlawed decades ago.
> > I guess nothing ever dies. Physicists have to have a language
> > to themselves.
> 
> I believe Fortran is still very much alive and well in engineering and
> physics - 40 years of number crunching code doesn't just go away by itself

Fortran and Cobol refuse to die...
I still encounter them regularly.

> > Anyway, my_python is comming along......
> > (pist, don't tell anyone, but I almost, (almost) like python).
> 
> My most favourite language of all time!
> Runs about as fast as reasonably complex shell code, but because it's an
> interpreter and not executed directly by the shell, there's no auto
> globbing and weird bash expansion going on. Effect = all that
> mind-bending psychosis-inducing escape nonsense just falls away.
> printf reduces to something a mortal human might even grok.
> control structures look sane - no fi, esac, or elif. Do's and while's
> look like real do's and while's
> 
> I could go on, but you get the idea :-)

Python is nice, I should really look into that sometime.

--
Joost

Re: [gentoo-user] OT: old farts slum_code enforcement

[Alan McKinnon]

On 15/09/2014 08:17, J. Roeleveld wrote:
>>> Ah yes, State machine design; not much fittering around with
>>> > > escaping silly little symbols.....
>> > 
>> > python fixes all of that (see below)
> I like state machines.
> 


Oops, I wasn't 100% clear;

It's silly escaping rules that python fixes, not state machines



-- 
Alan McKinnon
alan.mckinnon@gmail.com

Re: [gentoo-user] OT: old farts slum_code enforcement

[Peter Humphrey]

On Monday 15 September 2014 07:21:18 Alan McKinnon wrote:

> I believe Fortran is still very much alive and well in engineering and
> physics - 40 years of number crunching code doesn't just go away by itself

Confirmed. (Remember Blake's Seven?) It's the main application language in a 
multimillion-pound grid control system I was involved with 20 years ago. 
Fortran 66, that is, not 90. Nine other languages were used too - even 
including Pascal or something like it. Just the mainframes cost millions.

> I could go on, but you get the idea :-)

What, you?  :-)

-- 
Regards
Peter

Re: [gentoo-user] OT: old farts slum_code enforcement

[Alan McKinnon]

On 15/09/2014 11:00, Peter Humphrey wrote:
> On Monday 15 September 2014 07:21:18 Alan McKinnon wrote:
> 
>> I believe Fortran is still very much alive and well in engineering and
>> physics - 40 years of number crunching code doesn't just go away by itself
> 
> Confirmed. (Remember Blake's Seven?) It's the main application language in a 
> multimillion-pound grid control system I was involved with 20 years ago. 
> Fortran 66, that is, not 90. Nine other languages were used too - even 
> including Pascal or something like it. Just the mainframes cost millions.
> 

My favourite Fortran story:

One of the Mariner spacecraft yonks ago was sent a routine regular
course correction. It flew off at some wild angle and was never seen
again, and because the antenna was pointing in the wrong direction comms
could not be re-established. Oops. Expensive spacecraft down the drain....

The bug was eventually trace to a comma instead of a period in some
Fortran code (made a huge difference). Note that many code reviews
missed the bug, and the compiler accepted it as valid syntax



>> I could go on, but you get the idea :-)
> 
> What, you?  :-)
> 


-- 
Alan McKinnon
alan.mckinnon@gmail.com

Re: [gentoo-user] OT: old farts slum_code enforcement

[gottlieb]

On Mon, Sep 15 2014, Alan McKinnon wrote:

> My favourite Fortran story:
>
> One of the Mariner spacecraft yonks ago was sent a routine regular
> course correction. It flew off at some wild angle and was never seen
> again, and because the antenna was pointing in the wrong direction comms
> could not be re-established. Oops. Expensive spacecraft down the drain....
>
> The bug was eventually trace to a comma instead of a period in some
> Fortran code (made a huge difference). Note that many code reviews
> missed the bug, and the compiler accepted it as valid syntax

Perhaps instead of

    Do  15 i=1,10

a loop from 1 to 10

we had

   Do  15 i=1.10

which is the same in fortran as

   do15i = 1.0

an assignment statement.

allan

Re: [gentoo-user] OT: old farts slum_code enforcement

[Alan McKinnon]

On 15/09/2014 20:59, gottlieb@nyu.edu wrote:
> On Mon, Sep 15 2014, Alan McKinnon wrote:
> 
>> My favourite Fortran story:
>>
>> One of the Mariner spacecraft yonks ago was sent a routine regular
>> course correction. It flew off at some wild angle and was never seen
>> again, and because the antenna was pointing in the wrong direction comms
>> could not be re-established. Oops. Expensive spacecraft down the drain....
>>
>> The bug was eventually trace to a comma instead of a period in some
>> Fortran code (made a huge difference). Note that many code reviews
>> missed the bug, and the compiler accepted it as valid syntax
> 
> Perhaps instead of
> 
>     Do  15 i=1,10
> 
> a loop from 1 to 10
> 
> we had
> 
>    Do  15 i=1.10
> 
> which is the same in fortran as
> 
>    do15i = 1.0
> 
> an assignment statement.

If memory serves if was indeed something like that.


-- 
Alan McKinnon
alan.mckinnon@gmail.com

[gentoo-user] Re: OT: old farts slum_code enforcement

[Grant Edwards]

On 2014-09-15, Alan McKinnon <alan.mckinnon@gmail.com> wrote:

> My favourite Fortran story:
>
> One of the Mariner spacecraft yonks ago was sent a routine regular
> course correction. It flew off at some wild angle and was never seen
> again, and because the antenna was pointing in the wrong direction
> comms could not be re-established. Oops. Expensive spacecraft down
> the drain....
>
> The bug was eventually trace to a comma instead of a period in some
> Fortran code (made a huge difference). Note that many code reviews
> missed the bug, and the compiler accepted it as valid syntax

While that is a good story, Wikipedia refers to it as "a
legend", and NASA explicitly identifies a different cause.
According to NASA failure was caused by faulty beacon equipment
combined with an omitted hyphen in a data editing program.

http://nssdc.gsfc.nasa.gov/nmc/spacecraftDisplay.do?id=MARIN1:

   The failure was apparently caused by a combination of two factors.
   Improper operation of the Atlas airborne beacon equipment resulted
   in a loss of the rate signal from the vehicle for a prolonged
   period. The airborne beacon used for obtaining rate data was
   inoperative for four periods ranging from 1.5 to 61 seconds in
   duration. Additionally, the Mariner 1 Post Flight Review Board
   determined that the omission of a hyphen in coded computer
   instructions in the data-editing program allowed transmission of
   incorrect guidance signals to the spacecraft. During the periods
   the airborne beacon was inoperative the omission of the hyphen in
   the data-editing program caused the computer to incorrectly accept
   the sweep frequency of the ground receiver as it sought the vehicle
   beacon signal and combined this data with the tracking data sent to
   the remaining guidance computation. This caused the computer to
   swing automatically into a series of unnecessary course corrections
   with erroneous steering commands which finally threw the spacecraft
   off course.  
   
-- 
Grant Edwards               grant.b.edwards        Yow! I want the presidency
                                  at               so bad I can already taste
                              gmail.com            the hors d'oeuvres.

Re: [gentoo-user] OT: old farts slum_code enforcement

[the]

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA256

On 09/15/14 02:21, James wrote:
> Howz your Fortan 90? I did not even know there was such an ugly
> beast [4]. I thought Fortran was outlawed decades ago.

Outlawed. I love how this word is used in the same sentence
with "fortran".

> I guess nothing ever dies. Physicists have to have a language to
> themselves.

Unfortunately. First of all technically they are programers.
It looks like this kind of programmers have a hard time
learning another programming language.
Second, major generation/simulation packages are
moving on to c++.

http://home.thep.lu.se/~torbjorn/Pythia.html
http://www.geant4.org/geant4/

> Anyway, my_python is comming along...... (pist, don't tell anyone,
> but I almost, (almost) like python).

What about Perl?

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v2.0.22 (GNU/Linux)

iQEcBAEBCAAGBQJUFqPiAAoJEK64IL1uI2haddkH/24SHcJ+4q9ypRaq7A9W0MAn
d1jpDlfwJnevLDgnYhbcAuaWLj2b3Vyok4J+yZ3CsdVx9VH89KaCvOJDkNMyE40G
j2ai2ysV51ECP3ZfK5S+vACNCML/77nYzvhXsnQfzCrZ5DLxaMdUucCbXHsMNzlW
f/UkeTuvt+xm6nCD/zDZ2ILq1PQd9CoukEipfnPMBVwSK2pCuY2HR3glScL2G7DP
mbk1txHqvixqpbrSFcSytIDTf2fGCCEA5C8MXJ3BMSvFADjAGe+tmSxya1cxbksW
RGAL5bMGCK1xG7p75QAi1vnpcJ81vl6CAINbBGI2+uJsZeTUVc98i5COBNHObig=
=ITg6
-----END PGP SIGNATURE-----

Re: [gentoo-user] OT: old farts slum_code enforcement

[Volker Armin Hemmann]

Am 15.09.2014 um 10:31 schrieb the:
> On 09/15/14 02:21, James wrote:
> > Howz your Fortan 90? I did not even know there was such an ugly
> > beast [4]. I thought Fortran was outlawed decades ago.
>
> Outlawed. I love how this word is used in the same sentence
> with "fortran".
>

what is wrong with fortran?

> > I guess nothing ever dies. Physicists have to have a language to
> > themselves.
>
> Unfortunately. First of all technically they are programers.
> It looks like this kind of programmers have a hard time
> learning another programming language.
> Second, major generation/simulation packages are
> moving on to c++.

and that is better why?

>
> http://home.thep.lu.se/~torbjorn/Pythia.html
> http://www.geant4.org/geant4/
>
> > Anyway, my_python is comming along...... (pist, don't tell anyone,
> > but I almost, (almost) like python).
>
> What about Perl?
>

it is not a language. It is evil incarnated.

Re: [gentoo-user] OT: old farts slum_code enforcement

[Alan McKinnon]

On 15/09/2014 13:02, Volker Armin Hemmann wrote:
> Am 15.09.2014 um 10:31 schrieb the:
>> On 09/15/14 02:21, James wrote:
>>> Howz your Fortan 90? I did not even know there was such an ugly
>>> beast [4]. I thought Fortran was outlawed decades ago.
>>
>> Outlawed. I love how this word is used in the same sentence
>> with "fortran".
>>
> 
> what is wrong with fortran?
> 
>>> I guess nothing ever dies. Physicists have to have a language to
>>> themselves.
>>
>> Unfortunately. First of all technically they are programers.
>> It looks like this kind of programmers have a hard time
>> learning another programming language.
>> Second, major generation/simulation packages are
>> moving on to c++.
> 
> and that is better why?
> 
>>
>> http://home.thep.lu.se/~torbjorn/Pythia.html
>> http://www.geant4.org/geant4/
>>
>>> Anyway, my_python is comming along...... (pist, don't tell anyone,
>>> but I almost, (almost) like python).
>>
>> What about Perl?
>>
> 
> it is not a language. It is evil incarnated.


Perl is an experiment in brain<->cpu mapping.

Depending on your POV it has either wildly succeeded beyond all
expectations, or is an abysmal failure


"More than one way to do it". Ugh.

But, there is one very beautiful thing in perl, it's the "unless" keyword



-- 
Alan McKinnon
alan.mckinnon@gmail.com

Re: [gentoo-user] OT: old farts slum_code enforcement

[Peter Humphrey]

On Monday 15 September 2014 13:02:01 Volker Armin Hemmann wrote:
> Am 15.09.2014 um 10:31 schrieb the:
> > Outlawed. I love how this word is used in the same sentence
> > with "fortran".
> 
> what is wrong with fortran?

Nothing. Especially if you still have an 80-hole card punch and reader lying 
around  :-)

> > Second, major generation/simulation packages are
> > moving on to c++.
> 
> and that is better why?

Has your career not shown you that newer is /always/ better? I used forever to 
have young Turks proposing new ways of doing things, and I'd ask why change? 
What will be better? Not once did I get a good answer - but the "improvement" 
was always made anyway and called a "modernisation".

> > What about Perl?
> 
> it is not a language. It is evil incarnated.

Or if it were to be called a language, it would be a write-only language.

-- 
Regards
Peter

Re: [gentoo-user] OT: old farts slum_code enforcement

[the]

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA256

On 09/15/14 15:02, Volker Armin Hemmann wrote:
> what is wrong with fortran?

I don't know. Perhaps you can tell us.

OT(childhood psychological trauma){
I find it unreadable and had unpleasant experience
with fortran code. I actually had to **pick a compiler**
to the existing code in order to compile and get expected
results. After a few painful loops of "get a new compiler" -
"compile" - "fail" I was told that ifort should handle the
situation. I downloaded ifort (will RMS ever forgive me?)
and found out that my processor didn't support all the
instructions required for running ifort (SSE2). Of course
it (code failing to compile) is programmer's fault.
It was a part of my job so I couldn't just lay it off and
forget about it.
};

>> Unfortunately. First of all technically they are programers. It
>> looks like this kind of programmers have a hard time learning
>> another programming language. Second, major generation/simulation
>> packages are moving on to c++.
> 
> and that is better why?

What is better?
Suddenly, modelling in physics isn't computational mathematics
and it's easier to think in object rather than procedures.


-----BEGIN PGP SIGNATURE-----
Version: GnuPG v2.0.22 (GNU/Linux)

iQEcBAEBCAAGBQJUFublAAoJEK64IL1uI2ha5KEH/iBEdbQqrc0UDQFcJnZXYTVB
evtVw0skLbCrH8ir9IgDjMwofGHQ1ReQhDQdmEf/KyIg0kdEGE4dm19SN9ihwEgn
8JUr1Q1eHZAjhTBF0UE3GtKTgr9jrQ1PaqH9evjGQ5i5TCLZdR4BCoPbvv2e1rr7
90EgMPU5dPYd3aJqpOa+Ns0f7pc07RaDl2iWg9BAfJquxnTJFppcKqBi/dUfynB5
MKgjGuHmV4F62QOhUNS1dAmhVNm+PPUE31wSWMyegbdq3Yi5o5+jVzbDBKPwn/+4
jvXBcIPr+IMtl3vMAvQmA5DMM6Th1GhSblJrLSLHxFDvDNqctfrwbzAWa+j4OHo=
=3Qk0
-----END PGP SIGNATURE-----