Discussion:
Last Voyager engineer retires, NASA needs Fortran/Algol pros
(too old to reply)
RS Wood
2015-10-31 17:40:43 UTC
Permalink
For me the most interesting thing is the comment that the younger tech
staff don't *want* to learn the languages this project now requires.
Pathetic.


http://www.theregister.co.uk/2015/10/30/has_voyager_1_escaped_the_sun_yet_yes_but_also_no_say_boffins/

//--clip
Boffins have debated whether Voyager 1 has left the Solar System for a
few years now, after NASAdeclared it could no longer detect direct
evidence of the solar wind. The absence of the stream of particles
hurtling out from the sun was taken as evidence Voyager 1 had left the
solar system and entered the interstellar medium. But after revisiting
Voyager data, the new paper suggests some odd magnetic field readings
mean the probe is passing through “a more distorted magnetic field just
outside the heliopause, which is the boundary between the solar wind
and the interstellar medium.”
/--clip

http://www.theregister.co.uk/2015/10/31/brush_up_on_your_fortran/

//--clip
In an interview with Popular Mechanics, the manager of NASA's Voyager
program Suzanne Dodd said the retirement of the last original crew
member has left the space agency with a shortage of people capable of
communicating with the 40-year-old craft.
Launched in 1977, the two Voyager crafts rely on mid-70s hardware
powered by purpose-built General Electric interrupt processors. After
38 years in space, the two probes are currently on the outer fringes of
the Sun's influence, heading into interstellar space.
Though most of the instruments onboard the two probes have been
deactivated, both are still able to maintain contact with Earth and
will continue to do so into the 2020's, until their onboard
radioisotope thermoelectric generators no longer function.
In the meantime, NASA needs engineers capable of interacting with the
1970s-era technology, a skillset that includes knowledge of both
Fortran and assembly as well as the ability to command a machine with
just around 68KB of total memory.
"Although, some people can program in an assembly language and
understand the intricacy of the spacecraft, most younger people can't
or really don't want to," Dodd was quoted as saying.
With high-level languages now the standard for developers, knowing how
to fluently code in assembly has become a specialized skill, as has
fluency in dated languages such as Fortran. While obscure, the skillset
is potentially lucrative. Along with NASA's aging fleet of spacecraft,
many businesses still rely on ancient languages such as Fortran or
COBOL for specialized tasks and critical infrastructure. ®
//--clip
Peter Flass
2015-10-31 20:09:35 UTC
Permalink
Post by RS Wood
For me the most interesting thing is the comment that the younger tech
staff don't *want* to learn the languages this project now requires.
Pathetic.
http://www.theregister.co.uk/2015/10/30/has_voyager_1_escaped_the_sun_yet_yes_but_also_no_say_boffins/
//--clip
Boffins have debated whether Voyager 1 has left the Solar System for a
few years now, after NASAdeclared it could no longer detect direct
evidence of the solar wind. The absence of the stream of particles
hurtling out from the sun was taken as evidence Voyager 1 had left the
solar system and entered the interstellar medium. But after revisiting
Voyager data, the new paper suggests some odd magnetic field readings
mean the probe is passing through “a more distorted magnetic field just
outside the heliopause, which is the boundary between the solar wind
and the interstellar medium.”
/--clip
http://www.theregister.co.uk/2015/10/31/brush_up_on_your_fortran/
//--clip
In an interview with Popular Mechanics, the manager of NASA's Voyager
program Suzanne Dodd said the retirement of the last original crew
member has left the space agency with a shortage of people capable of
communicating with the 40-year-old craft.
I can't believe that no one "wants to." What I would believe is that NASA
is not willing to take someone with enthusiasm but no experience with this
specific hardware and train them. Like most employers they want someone
who can step into a position with no learning period.
Post by RS Wood
Launched in 1977, the two Voyager crafts rely on mid-70s hardware
powered by purpose-built General Electric interrupt processors. After
38 years in space, the two probes are currently on the outer fringes of
the Sun's influence, heading into interstellar space.
Though most of the instruments onboard the two probes have been
deactivated, both are still able to maintain contact with Earth and
will continue to do so into the 2020's, until their onboard
radioisotope thermoelectric generators no longer function.
In the meantime, NASA needs engineers capable of interacting with the
1970s-era technology, a skillset that includes knowledge of both
Fortran and assembly as well as the ability to command a machine with
just around 68KB of total memory.
"Although, some people can program in an assembly language and
understand the intricacy of the spacecraft, most younger people can't
or really don't want to," Dodd was quoted as saying.
With high-level languages now the standard for developers, knowing how
to fluently code in assembly has become a specialized skill, as has
fluency in dated languages such as Fortran. While obscure, the skillset
is potentially lucrative. Along with NASA's aging fleet of spacecraft,
many businesses still rely on ancient languages such as Fortran or
COBOL for specialized tasks and critical infrastructure. ®
//--clip
--
Pete
Howard S Shubs
2015-10-31 20:53:27 UTC
Permalink
Post by Peter Flass
I can't believe that no one "wants to." What I would believe is that NASA
is not willing to take someone with enthusiasm but no experience with this
specific hardware and train them. Like most employers they want someone
who can step into a position with no learning period.
What I find amusing is that, as an experienced Fortran programmer on
things from IBM 1130 and newer, I've never heard of these positions.
I've just tried finding them on the ***@Careers site, with no luck.
Perhaps no one is applying because they CAN'T FIND THE DAMN JOBS. But
that'd be too obvious.
artie
2015-10-31 21:12:19 UTC
Permalink
Post by Howard S Shubs
Post by Peter Flass
I can't believe that no one "wants to." What I would believe is that NASA
is not willing to take someone with enthusiasm but no experience with this
specific hardware and train them. Like most employers they want someone
who can step into a position with no learning period.
What I find amusing is that, as an experienced Fortran programmer on
things from IBM 1130 and newer, I've never heard of these positions.
Perhaps no one is applying because they CAN'T FIND THE DAMN JOBS. But
that'd be too obvious.
I'll sing in that chorus -- one of my first programming jobs (some
decades ago) was converting a specialized FORTRAN scientific library
into optimized assembly code (on a PDP-10), resulting in many happy
users of that library.

I also did FORTRAN on the IBM 1130, Univac 1108 (two zeroes!), many 360
systems, SDS Sigma 5, 6, 7, 8, 9, 560, big CDC and Cray boxes.

I'm not dead yet, and I'd love a job doing that kind of thing!

--
Nyssa
2015-11-01 15:42:19 UTC
Permalink
Post by artie
Post by Howard S Shubs
Post by Peter Flass
I can't believe that no one "wants to." What I would
believe is that NASA is not willing to take someone
with enthusiasm but no experience with this
specific hardware and train them. Like most employers
they want someone who can step into a position with no
learning period.
What I find amusing is that, as an experienced Fortran
programmer on things from IBM 1130 and newer, I've never
heard of these positions. I've just tried finding them on
Perhaps no one is applying because they CAN'T FIND THE
DAMN JOBS. But that'd be too obvious.
I'll sing in that chorus -- one of my first programming
jobs (some decades ago) was converting a specialized
FORTRAN scientific library into optimized assembly code
(on a PDP-10), resulting in many happy users of that
library.
I also did FORTRAN on the IBM 1130, Univac 1108 (two
zeroes!), many 360 systems, SDS Sigma 5, 6, 7, 8, 9, 560,
big CDC and Cray boxes.
I'm not dead yet, and I'd love a job doing that kind of
thing!
--
As would I, and no doubt many others with "legacy" experience.

Of course, if NASA went about hiring replacements for their
retired engineers/programmers they would probably fall into
the same pattern that I have encountered all too many times
when applying for other bit fiddler jobs as in....

"Well, you've got the languages and the experience, but it's
not specifically on the machines/os/environment we're using
and your experience isn't in our industry, so it doesn't count."

My FORTRAN days were spent doing stuff for air traffic control
simulators and trainers, weather forecasting systems, and
logistics, so I'd be filtered out of the resume pile by some
HR word-matching algorithm and never have a chance.

That or something like "yeah, you've got the languages, etc.,
but your experience in them is *recent*" and then fall through
the holes again.

I'm sure most of the rest of us old-timers with the right
languages in our back pockets would be told something similar
even if these phantom jobs did surface somewhere on the
NASA site.

BTW anyone know where these phantom jobs are supposed to be
located?

Nyssa, who has seen this play act out many times in her
professional past
Nyssa
2015-11-01 15:43:39 UTC
Permalink
Post by artie
Post by Howard S Shubs
Post by Peter Flass
I can't believe that no one "wants to." What I would
believe is that NASA is not willing to take someone
with enthusiasm but no experience with this
specific hardware and train them. Like most employers
they want someone who can step into a position with no
learning period.
What I find amusing is that, as an experienced Fortran
programmer on things from IBM 1130 and newer, I've never
heard of these positions. I've just tried finding them on
Perhaps no one is applying because they CAN'T FIND THE
DAMN JOBS. But that'd be too obvious.
I'll sing in that chorus -- one of my first programming
jobs (some decades ago) was converting a specialized
FORTRAN scientific library into optimized assembly code
(on a PDP-10), resulting in many happy users of that
library.
I also did FORTRAN on the IBM 1130, Univac 1108 (two
zeroes!), many 360 systems, SDS Sigma 5, 6, 7, 8, 9, 560,
big CDC and Cray boxes.
I'm not dead yet, and I'd love a job doing that kind of
thing!
--
As would I, and no doubt many others with "legacy" experience.

Of course, if NASA went about hiring replacements for their
retired engineers/programmers they would probably fall into
the same pattern that I have encountered all too many times
when applying for other bit fiddler jobs as in....

"Well, you've got the languages and the experience, but it's
not specifically on the machines/os/environment we're using
and your experience isn't in our industry, so it doesn't count."

My FORTRAN days were spent doing stuff for air traffic control
simulators and trainers, weather forecasting systems, and
logistics, so I'd be filtered out of the resume pile by some
HR word-matching algorithm and never have a chance.

That or something like "yeah, you've got the languages, etc.,
but your experience in them is *recent*" and then fall through
the holes again.

I'm sure most of the rest of us old-timers with the right
languages in our back pockets would be told something similar
even if these phantom jobs did surface somewhere on the
NASA site.

BTW anyone know where these phantom jobs are supposed to be
located?

Nyssa, who has seen this play act out many times in her
professional past
Nyssa
2015-11-01 15:44:27 UTC
Permalink
Post by artie
Post by Howard S Shubs
Post by Peter Flass
I can't believe that no one "wants to." What I would
believe is that NASA is not willing to take someone
with enthusiasm but no experience with this
specific hardware and train them. Like most employers
they want someone who can step into a position with no
learning period.
What I find amusing is that, as an experienced Fortran
programmer on things from IBM 1130 and newer, I've never
heard of these positions. I've just tried finding them on
Perhaps no one is applying because they CAN'T FIND THE
DAMN JOBS. But that'd be too obvious.
I'll sing in that chorus -- one of my first programming
jobs (some decades ago) was converting a specialized
FORTRAN scientific library into optimized assembly code
(on a PDP-10), resulting in many happy users of that
library.
I also did FORTRAN on the IBM 1130, Univac 1108 (two
zeroes!), many 360 systems, SDS Sigma 5, 6, 7, 8, 9, 560,
big CDC and Cray boxes.
I'm not dead yet, and I'd love a job doing that kind of
thing!
As would I, and no doubt many others with "legacy" experience.

Of course, if NASA went about hiring replacements for their
retired engineers/programmers they would probably fall into
the same pattern that I have encountered all too many times
when applying for other bit fiddler jobs as in....

"Well, you've got the languages and the experience, but it's
not specifically on the machines/os/environment we're using
and your experience isn't in our industry, so it doesn't count."

My FORTRAN days were spent doing stuff for air traffic control
simulators and trainers, weather forecasting systems, and
logistics, so I'd be filtered out of the resume pile by some
HR word-matching algorithm and never have a chance.

That or something like "yeah, you've got the languages, etc.,
but your experience in them is *recent*" and then fall through
the holes again.

I'm sure most of the rest of us old-timers with the right
languages in our back pockets would be told something similar
even if these phantom jobs did surface somewhere on the
NASA site.

BTW anyone know where these phantom jobs are supposed to be
located?

Nyssa, who has seen this play act out many times in her
professional past
Rich
2015-10-31 22:27:01 UTC
Permalink
Post by Peter Flass
Post by RS Wood
For me the most interesting thing is the comment that the younger tech
staff don't *want* to learn the languages this project now requires.
Pathetic.
http://www.theregister.co.uk/2015/10/30/has_voyager_1_escaped_the_sun_yet_yes_but_also_no_say_boffins/
//--clip
Boffins have debated whether Voyager 1 has left the Solar System
for a few years now, after NASAdeclared it could no longer detect
direct evidence of the solar wind. The absence of the stream of
particles hurtling out from the sun was taken as evidence Voyager 1
had left the solar system and entered the interstellar medium. But
after revisiting Voyager data, the new paper suggests some odd
magnetic field readings mean the probe is passing through ?a more
distorted magnetic field just outside the heliopause, which is the
boundary between the solar wind and the interstellar medium.?
/--clip
http://www.theregister.co.uk/2015/10/31/brush_up_on_your_fortran/
//--clip
In an interview with Popular Mechanics, the manager of NASA's
Voyager program Suzanne Dodd said the retirement of the last
original crew member has left the space agency with a shortage of
people capable of communicating with the 40-year-old craft.
I can't believe that no one "wants to." What I would believe is that
NASA is not willing to take someone with enthusiasm but no experience
with this specific hardware and train them. Like most employers they
want someone who can step into a position with no learning period.
This is most likely the exact reason. The new standard of "the new
hire must come fully trained for _our_ particular stuff". How
companies that expect this manage to hire anyone at all is amazing.
Howard S Shubs
2015-11-01 04:37:07 UTC
Permalink
Post by Rich
Post by Peter Flass
Post by RS Wood
For me the most interesting thing is the comment that the younger tech
staff don't *want* to learn the languages this project now requires.
Pathetic.
http://www.theregister.co.uk/2015/10/30/has_voyager_1_escaped_the_sun_yet_yes_but_also_no_say_boffins/
//--clip
Boffins have debated whether Voyager 1 has left the Solar System
for a few years now, after NASAdeclared it could no longer detect
direct evidence of the solar wind. The absence of the stream of
particles hurtling out from the sun was taken as evidence Voyager 1
had left the solar system and entered the interstellar medium. But
after revisiting Voyager data, the new paper suggests some odd
magnetic field readings mean the probe is passing through ?a more
distorted magnetic field just outside the heliopause, which is the
boundary between the solar wind and the interstellar medium.?
/--clip
http://www.theregister.co.uk/2015/10/31/brush_up_on_your_fortran/
//--clip
In an interview with Popular Mechanics, the manager of NASA's
Voyager program Suzanne Dodd said the retirement of the last
original crew member has left the space agency with a shortage of
people capable of communicating with the 40-year-old craft.
I can't believe that no one "wants to." What I would believe is that
NASA is not willing to take someone with enthusiasm but no experience
with this specific hardware and train them. Like most employers they
want someone who can step into a position with no learning period.
This is most likely the exact reason. The new standard of "the new
hire must come fully trained for _our_ particular stuff". How
companies that expect this manage to hire anyone at all is amazing.
I saw one a month or two ago for a company in LA or San Diego for a
company looking for MACRO-11 programmers. I don't understand how that
company has survived this long. I figure someone retired. They're
supposedly porting to something else, but at the same time, they're
generating NEW MACRO-11 code?!?
Roger Blake
2015-11-01 22:44:55 UTC
Permalink
Post by Howard S Shubs
I saw one a month or two ago for a company in LA or San Diego for a
company looking for MACRO-11 programmers. I don't understand how that
company has survived this long. I figure someone retired. They're
supposedly porting to something else, but at the same time, they're
generating NEW MACRO-11 code?!?
A lifetime ago I used to practically generate MACRO-11 code in my sleep.
I'm surprised there's still a demand for that sort of thing, but really have
no wish to go back to it at this point.
--
-----------------------------------------------------------------------------
Roger Blake (Change "invalid" to "com" for email. Google Groups killfiled.)

NSA sedition and treason -- http://www.DeathToNSAthugs.com
-----------------------------------------------------------------------------
gareth
2015-11-01 23:54:27 UTC
Permalink
Post by Roger Blake
Post by Howard S Shubs
I saw one a month or two ago for a company in LA or San Diego for a
company looking for MACRO-11 programmers. I don't understand how that
company has survived this long. I figure someone retired. They're
supposedly porting to something else, but at the same time, they're
generating NEW MACRO-11 code?!?
A lifetime ago I used to practically generate MACRO-11 code in my sleep.
I'm surprised there's still a demand for that sort of thing, but really have
no wish to go back to it at this point.
I don't see why not, because the language in which a project is coded
is somewhat irrelevant following a competent design.
jmfbahciv
2015-11-02 13:17:46 UTC
Permalink
Post by gareth
Post by Roger Blake
Post by Howard S Shubs
I saw one a month or two ago for a company in LA or San Diego for a
company looking for MACRO-11 programmers. I don't understand how that
company has survived this long. I figure someone retired. They're
supposedly porting to something else, but at the same time, they're
generating NEW MACRO-11 code?!?
A lifetime ago I used to practically generate MACRO-11 code in my sleep.
I'm surprised there's still a demand for that sort of thing, but really have
no wish to go back to it at this point.
I don't see why not, because the language in which a project is coded
is somewhat irrelevant following a competent design.
That's simply not true. A design is constrained by the machine
lanuguage the software runs on.

/BAH
gareth
2015-11-02 15:35:43 UTC
Permalink
Post by jmfbahciv
Post by gareth
I don't see why not, because the language in which a project is coded
is somewhat irrelevant following a competent design.
That's simply not true. A design is constrained by the machine
lanuguage the software runs on.
Which machine languages are not Universal Turing Machines?
Gene Wirchenko
2015-11-02 16:36:59 UTC
Permalink
On Mon, 2 Nov 2015 15:35:43 -0000, "gareth"
Post by gareth
Post by jmfbahciv
Post by gareth
I don't see why not, because the language in which a project is coded
is somewhat irrelevant following a competent design.
That's simply not true. A design is constrained by the machine
lanuguage the software runs on.
Which machine languages are not Universal Turing Machines?
That is irrelevant. Languages have trade-offs and make some
things easy and some things difficult.

Sincerely,

Gene Wirchenko
gareth
2015-11-02 17:37:15 UTC
Permalink
Post by Gene Wirchenko
On Mon, 2 Nov 2015 15:35:43 -0000, "gareth"
Post by gareth
Post by jmfbahciv
Post by gareth
I don't see why not, because the language in which a project is coded
is somewhat irrelevant following a competent design.
That's simply not true. A design is constrained by the machine
lanuguage the software runs on.
Which machine languages are not Universal Turing Machines?
That is irrelevant. Languages have trade-offs and make some
things easy and some things difficult.
It is very relevant, for that something might be difficult, or perhaps more
involved,
does not denigrate the designed program.
David Hume
2015-11-02 17:28:28 UTC
Permalink
Post by gareth
Post by jmfbahciv
Post by gareth
I don't see why not, because the language in which a project is coded
is somewhat irrelevant following a competent design.
That's simply not true. A design is constrained by the machine
lanuguage the software runs on.
Which machine languages are not Universal Turing Machines?
Washing machine language.
John Levine
2015-11-04 01:28:11 UTC
Permalink
Post by gareth
Which machine languages are not Universal Turing Machines?
All of them. Eventually you always run out of memory.

R's,
John
Walter Bushell
2015-11-11 17:08:37 UTC
Permalink
Post by gareth
Post by jmfbahciv
Post by gareth
I don't see why not, because the language in which a project is coded
is somewhat irrelevant following a competent design.
That's simply not true. A design is constrained by the machine
lanuguage the software runs on.
Which machine languages are not Universal Turing Machines?
All. All machines are finite and there are time considerations in
all applications. An algorerhythm that takes thousands of years
is a non starter.
--
Never attribute to stupidity that which can be explained by greed. Me.
Rod Speed
2015-11-02 20:17:20 UTC
Permalink
Post by jmfbahciv
Post by gareth
Post by Roger Blake
Post by Howard S Shubs
I saw one a month or two ago for a company in LA or San Diego for a
company looking for MACRO-11 programmers. I don't understand how that
company has survived this long. I figure someone retired. They're
supposedly porting to something else, but at the same time, they're
generating NEW MACRO-11 code?!?
A lifetime ago I used to practically generate MACRO-11 code in my sleep.
I'm surprised there's still a demand for that sort of thing, but really have
no wish to go back to it at this point.
I don't see why not, because the language in which a project is coded
is somewhat irrelevant following a competent design.
That's simply not true. A design is constrained by the machine
lanuguage the software runs on.
We aren't talking about machine language there.
jmfbahciv
2015-11-03 12:54:01 UTC
Permalink
Post by Rod Speed
Post by jmfbahciv
Post by gareth
Post by Roger Blake
Post by Howard S Shubs
I saw one a month or two ago for a company in LA or San Diego for a
company looking for MACRO-11 programmers. I don't understand how that
company has survived this long. I figure someone retired. They're
supposedly porting to something else, but at the same time, they're
generating NEW MACRO-11 code?!?
A lifetime ago I used to practically generate MACRO-11 code in my sleep.
I'm surprised there's still a demand for that sort of thing, but really have
no wish to go back to it at this point.
I don't see why not, because the language in which a project is coded
is somewhat irrelevant following a competent design.
That's simply not true. A design is constrained by the machine
lanuguage the software runs on.
We aren't talking about machine language there.
What do you think MACRO-11 generated? IBM 1620 code?

/BAH
Rod Speed
2015-11-03 21:06:37 UTC
Permalink
Post by jmfbahciv
Post by Rod Speed
Post by jmfbahciv
Post by gareth
Post by Roger Blake
Post by Howard S Shubs
I saw one a month or two ago for a company in LA or San Diego for a
company looking for MACRO-11 programmers. I don't understand how that
company has survived this long. I figure someone retired. They're
supposedly porting to something else, but at the same time, they're
generating NEW MACRO-11 code?!?
A lifetime ago I used to practically generate MACRO-11 code in my sleep.
I'm surprised there's still a demand for that sort of thing, but
really
have
no wish to go back to it at this point.
I don't see why not, because the language in which a project is coded
is somewhat irrelevant following a competent design.
That's simply not true. A design is constrained by the machine
lanuguage the software runs on.
We aren't talking about machine language there.
What do you think MACRO-11 generated? IBM 1620 code?
Irrelevant to the DESIGN being discussed.
gareth
2015-11-02 20:48:39 UTC
Permalink
Post by jmfbahciv
Post by gareth
Post by Roger Blake
Post by Howard S Shubs
I saw one a month or two ago for a company in LA or San Diego for a
company looking for MACRO-11 programmers. I don't understand how that
company has survived this long. I figure someone retired. They're
supposedly porting to something else, but at the same time, they're
generating NEW MACRO-11 code?!?
A lifetime ago I used to practically generate MACRO-11 code in my sleep.
I'm surprised there's still a demand for that sort of thing, but really have
no wish to go back to it at this point.
I don't see why not, because the language in which a project is coded
is somewhat irrelevant following a competent design.
That's simply not true. A design is constrained by the machine
lanuguage the software runs on.
Selecting a machine that will not run the program is a significant
failure of system design.
jmfbahciv
2015-11-03 12:54:03 UTC
Permalink
Post by gareth
Post by jmfbahciv
Post by gareth
Post by Roger Blake
Post by Howard S Shubs
I saw one a month or two ago for a company in LA or San Diego for a
company looking for MACRO-11 programmers. I don't understand how that
company has survived this long. I figure someone retired. They're
supposedly porting to something else, but at the same time, they're
generating NEW MACRO-11 code?!?
A lifetime ago I used to practically generate MACRO-11 code in my sleep.
I'm surprised there's still a demand for that sort of thing, but really have
no wish to go back to it at this point.
I don't see why not, because the language in which a project is coded
is somewhat irrelevant following a competent design.
That's simply not true. A design is constrained by the machine
lanuguage the software runs on.
Selecting a machine that will not run the program is a significant
failure of system design.
Not all software projects include buying the hardware. Most projects
already had the computer system.

/BAH
gareth
2015-11-03 13:01:51 UTC
Permalink
Post by jmfbahciv
Post by gareth
Post by jmfbahciv
Post by gareth
Post by Roger Blake
Post by Howard S Shubs
I saw one a month or two ago for a company in LA or San Diego for a
company looking for MACRO-11 programmers. I don't understand how that
company has survived this long. I figure someone retired. They're
supposedly porting to something else, but at the same time, they're
generating NEW MACRO-11 code?!?
A lifetime ago I used to practically generate MACRO-11 code in my sleep.
I'm surprised there's still a demand for that sort of thing, but
really
have
no wish to go back to it at this point.
I don't see why not, because the language in which a project is coded
is somewhat irrelevant following a competent design.
That's simply not true. A design is constrained by the machine
lanuguage the software runs on.
Selecting a machine that will not run the program is a significant
failure of system design.
Not all software projects include buying the hardware. Most projects
already had the computer system.
Then it displays a fundamental lack of competence to design a program that
will
not run on the target machine.

It remains for me, that whatever program I design, it will run whether coded
in
machine code, assembly language, FORTRAN, RTL/2, CORAL, C, Pascal,
FORTH, CRAP, SOAS, or any of a myriad other languages that I have used.
Andrew Swallow
2015-11-03 19:27:11 UTC
Permalink
Post by gareth
Post by jmfbahciv
Post by gareth
Post by jmfbahciv
Post by gareth
Post by Roger Blake
Post by Howard S Shubs
I saw one a month or two ago for a company in LA or San Diego for a
company looking for MACRO-11 programmers. I don't understand how that
company has survived this long. I figure someone retired. They're
supposedly porting to something else, but at the same time, they're
generating NEW MACRO-11 code?!?
A lifetime ago I used to practically generate MACRO-11 code in my sleep.
I'm surprised there's still a demand for that sort of thing, but
really
have
no wish to go back to it at this point.
I don't see why not, because the language in which a project is coded
is somewhat irrelevant following a competent design.
That's simply not true. A design is constrained by the machine
lanuguage the software runs on.
Selecting a machine that will not run the program is a significant
failure of system design.
Not all software projects include buying the hardware. Most projects
already had the computer system.
Then it displays a fundamental lack of competence to design a program that
will
not run on the target machine.
It remains for me, that whatever program I design, it will run whether coded
in
machine code, assembly language, FORTRAN, RTL/2, CORAL, C, Pascal,
FORTH, CRAP, SOAS, or any of a myriad other languages that I have used.
So you were only allowed to work on simple projects. Bigger projects may
involve constructing the buildings to house the computers in several
different countries.
gareth
2015-11-04 11:04:19 UTC
Permalink
Post by Andrew Swallow
Post by gareth
It remains for me, that whatever program I design, it will run whether coded
in
machine code, assembly language, FORTRAN, RTL/2, CORAL, C, Pascal,
FORTH, CRAP, SOAS, or any of a myriad other languages that I have used.
So you were only allowed to work on simple projects. Bigger projects may
involve constructing the buildings to house the computers in several
different countries.
Those exhibiting Builders' Bums do not get to design programs.
Andrew Swallow
2015-11-04 23:48:44 UTC
Permalink
Post by gareth
Post by Andrew Swallow
Post by gareth
It remains for me, that whatever program I design, it will run whether coded
in
machine code, assembly language, FORTRAN, RTL/2, CORAL, C, Pascal,
FORTH, CRAP, SOAS, or any of a myriad other languages that I have used.
So you were only allowed to work on simple projects. Bigger projects may
involve constructing the buildings to house the computers in several
different countries.
Those exhibiting Builders' Bums do not get to design programs.
They do get to project manage the programmers.
Walter Bushell
2015-11-11 17:12:27 UTC
Permalink
Post by Andrew Swallow
Post by gareth
Post by jmfbahciv
Post by gareth
Post by jmfbahciv
Post by gareth
Post by Roger Blake
Post by Howard S Shubs
I saw one a month or two ago for a company in LA or San Diego for a
company looking for MACRO-11 programmers. I don't understand how that
company has survived this long. I figure someone retired. They're
supposedly porting to something else, but at the same time, they're
generating NEW MACRO-11 code?!?
A lifetime ago I used to practically generate MACRO-11 code in my sleep.
I'm surprised there's still a demand for that sort of thing, but
really
have
no wish to go back to it at this point.
I don't see why not, because the language in which a project is coded
is somewhat irrelevant following a competent design.
That's simply not true. A design is constrained by the machine
lanuguage the software runs on.
Selecting a machine that will not run the program is a significant
failure of system design.
Not all software projects include buying the hardware. Most projects
already had the computer system.
Then it displays a fundamental lack of competence to design a program that
will
not run on the target machine.
It remains for me, that whatever program I design, it will run whether coded
in
machine code, assembly language, FORTRAN, RTL/2, CORAL, C, Pascal,
FORTH, CRAP, SOAS, or any of a myriad other languages that I have used.
So you were only allowed to work on simple projects. Bigger projects may
involve constructing the buildings to house the computers in several
different countries.
And we are talking about a confuser on a spacecraft launched long,
long ago (in confuser years) by a much different country. The past
*is* a different country.
--
Never attribute to stupidity that which can be explained by greed. Me.
Bill Evans
2015-11-11 20:35:02 UTC
Permalink
Post by Walter Bushell
And we are talking about a confuser on a spacecraft launched long,
long ago (in confuser years) by a much different country. The past
*is* a different country.
Brings to mind the old guidline: comment your code as though
you were writing it to be understood by a complete stranger,
because six months from now you will *be* that stranger.
--
Bill Evans / Box 1224 / Mariposa, CA 95338 / (209)742-4720
Mail-To: ***@acm.org -- PGP encrypted mail preferred. --
pgpkey.mariposabill.com for public key. Key #: 8D8B521B
PGPprint: 0A9C 3545 8FFF 7501 6265 1519 40FF 76F9 8D8B 521B
Joy Beeson
2015-11-12 03:11:32 UTC
Permalink
Post by Bill Evans
Brings to mind the old guidline: comment your code as though
you were writing it to be understood by a complete stranger,
because six months from now you will *be* that stranger.
I said pretty much the same thing in my book on sewing.

I had a terrible time, once, identifying the "floral linen" one
pattern said it had been made up in. I should have labeled it
"damask".
--
Joy Beeson
joy beeson at comcast dot net
http://wlweather.net/PAGESEW/
Scott Lurndal
2015-11-12 15:31:08 UTC
Permalink
Post by Bill Evans
Post by Walter Bushell
And we are talking about a confuser on a spacecraft launched long,
long ago (in confuser years) by a much different country. The past
*is* a different country.
Brings to mind the old guidline: comment your code as though
you were writing it to be understood by a complete stranger,
because six months from now you will *be* that stranger.
And the longer you program, the shorter that six months becomes :-(
Paul Sture
2015-11-12 17:41:20 UTC
Permalink
Post by Bill Evans
Post by Walter Bushell
And we are talking about a confuser on a spacecraft launched long,
long ago (in confuser years) by a much different country. The past
*is* a different country.
Brings to mind the old guidline: comment your code as though
you were writing it to be understood by a complete stranger,
because six months from now you will *be* that stranger.
I used to spend a lot of time at customer sites, so the ability to be
understood by someone else was firmly in the job description. I
*really* didn't want to get called back at some point in the future due
to having left some incomprehensible code in place.
--
Should not the Society of Indexers be know as Indexers, Society of, The?
-- Keith Waterhouse
jmfbahciv
2015-11-04 13:52:24 UTC
Permalink
Post by gareth
Post by jmfbahciv
Post by gareth
Post by jmfbahciv
Post by gareth
Post by Roger Blake
Post by Howard S Shubs
I saw one a month or two ago for a company in LA or San Diego for a
company looking for MACRO-11 programmers. I don't understand how that
company has survived this long. I figure someone retired. They're
supposedly porting to something else, but at the same time, they're
generating NEW MACRO-11 code?!?
A lifetime ago I used to practically generate MACRO-11 code in my sleep.
I'm surprised there's still a demand for that sort of thing, but
really
have
no wish to go back to it at this point.
I don't see why not, because the language in which a project is coded
is somewhat irrelevant following a competent design.
That's simply not true. A design is constrained by the machine
lanuguage the software runs on.
Selecting a machine that will not run the program is a significant
failure of system design.
Not all software projects include buying the hardware. Most projects
already had the computer system.
Then it displays a fundamental lack of competence to design a program that
will
not run on the target machine.
It remains for me, that whatever program I design, it will run whether coded
in
machine code, assembly language, FORTRAN, RTL/2, CORAL, C, Pascal,
FORTH, CRAP, SOAS, or any of a myriad other languages that I have used.
And the OS developers of this biz made it possible for you to be able to
do your coding job. The language developers made it possible so you
didn't have to think about the hardware. All of these people had to do
designs which had tradeoffs determined by the computing environment so
that you wouldn't have to.

/BAH
gareth
2015-11-04 15:00:21 UTC
Permalink
Post by jmfbahciv
And the OS developers of this biz made it possible for you to be able to
do your coding job.
Not so, for I work with the native machine by preference.
Post by jmfbahciv
The language developers made it possible so you
didn't have to think about the hardware.
I always think about the hardware, having started out as a radio ham and
then
qualifying in electronics with a specialisation in computer design.

With one exception 25 years ago which was a data base for a book publisher,
all my projects
have been hardware related. Any real computer scientist worthy of the name
is fully au fait with how a computer works, anyway.
Post by jmfbahciv
All of these people had to do
designs which had tradeoffs determined by the computing environment so
that you wouldn't have to.
A non-sequitur.
Rod Speed
2015-11-04 21:56:23 UTC
Permalink
Post by jmfbahciv
Post by gareth
Post by jmfbahciv
Post by gareth
Post by jmfbahciv
Post by gareth
Post by Roger Blake
Post by Howard S Shubs
I saw one a month or two ago for a company in LA or San Diego for a
company looking for MACRO-11 programmers. I don't understand how that
company has survived this long. I figure someone retired. They're
supposedly porting to something else, but at the same time, they're
generating NEW MACRO-11 code?!?
A lifetime ago I used to practically generate MACRO-11 code in my sleep.
I'm surprised there's still a demand for that sort of thing, but
really
have
no wish to go back to it at this point.
I don't see why not, because the language in which a project is coded
is somewhat irrelevant following a competent design.
That's simply not true. A design is constrained by the machine
lanuguage the software runs on.
Selecting a machine that will not run the program is a significant
failure of system design.
Not all software projects include buying the hardware. Most projects
already had the computer system.
Then it displays a fundamental lack of competence to design a program that
will
not run on the target machine.
It remains for me, that whatever program I design, it will run whether coded
in
machine code, assembly language, FORTRAN, RTL/2, CORAL, C, Pascal,
FORTH, CRAP, SOAS, or any of a myriad other languages that I have used.
And the OS developers of this biz made it possible for you to be able to
do your coding job.
Wrong again. I used no OS on the PDP8S, the serial one, that
I used to measure fluorescent decay to below 1ns levels.
Post by jmfbahciv
The language developers made it possible so
you didn't have to think about the hardware.
I didn’t use any language, I used assembler.
Post by jmfbahciv
All of these people had to do designs which
had tradeoffs determined by the computing
environment so that you wouldn't have to.
They in fact designed the hardware so that it would
do a decent job with any of the commonly used
languages used in that segment of the industry.

And fucked that up pretty comprehensively too
at times like when DEC concentrated on Focal
instead of Basic for various reasons etc.
Greymaus
2015-11-01 11:58:03 UTC
Permalink
Post by Rich
Post by Peter Flass
Post by RS Wood
For me the most interesting thing is the comment that the younger tech
staff don't *want* to learn the languages this project now requires.
Pathetic.
http://www.theregister.co.uk/2015/10/30/has_voyager_1_escaped_the_sun_yet_yes_but_also_no_say_boffins/
//--clip
Boffins have debated whether Voyager 1 has left the Solar System
for a few years now, after NASAdeclared it could no longer detect
direct evidence of the solar wind. The absence of the stream of
particles hurtling out from the sun was taken as evidence Voyager 1
had left the solar system and entered the interstellar medium. But
after revisiting Voyager data, the new paper suggests some odd
magnetic field readings mean the probe is passing through ?a more
distorted magnetic field just outside the heliopause, which is the
boundary between the solar wind and the interstellar medium.?
/--clip
http://www.theregister.co.uk/2015/10/31/brush_up_on_your_fortran/
//--clip
In an interview with Popular Mechanics, the manager of NASA's
Voyager program Suzanne Dodd said the retirement of the last
original crew member has left the space agency with a shortage of
people capable of communicating with the 40-year-old craft.
I can't believe that no one "wants to." What I would believe is that
NASA is not willing to take someone with enthusiasm but no experience
with this specific hardware and train them. Like most employers they
want someone who can step into a position with no learning period.
This is most likely the exact reason. The new standard of "the new
hire must come fully trained for _our_ particular stuff". How
companies that expect this manage to hire anyone at all is amazing.
In this country, in non-computer areas, it usually means "I want this job for a
particular person, but we have to advertise"
--
greymaus
.
.
...
Walter Bushell
2015-11-11 17:15:27 UTC
Permalink
Post by Greymaus
In this country, in non-computer areas, it usually means "I want this job for a
particular person, but we have to advertise"
In America for computer jobs also. Ads may require 5 years experience
for an environment that was just released less than a year ago. Ergo,
the fix is in.
--
Never attribute to stupidity that which can be explained by greed. Me.
Andrew Swallow
2015-11-12 00:40:14 UTC
Permalink
Post by Walter Bushell
Post by Greymaus
In this country, in non-computer areas, it usually means "I want this job for a
particular person, but we have to advertise"
In America for computer jobs also. Ads may require 5 years experience
for an environment that was just released less than a year ago. Ergo,
the fix is in.
Or the manager is having to recruit despite the anti-personal department.
Scott Lurndal
2015-11-02 14:29:39 UTC
Permalink
Post by Peter Flass
Post by RS Wood
In an interview with Popular Mechanics, the manager of NASA's Voyager
program Suzanne Dodd said the retirement of the last original crew
member has left the space agency with a shortage of people capable of
communicating with the 40-year-old craft.
I can't believe that no one "wants to." What I would believe is that NASA
is not willing to take someone with enthusiasm but no experience with this
specific hardware and train them. Like most employers they want someone
who can step into a position with no learning period.
Let's see:

Hand the first: A job at Google/Twitter/FB/LI using modern technology and getting
paid beaucoup bux.

Hand the second: A job at NASA working with dead technology and getting
paid close to zilch.

Seems an easy choice.
Osmium
2015-11-02 15:06:27 UTC
Permalink
Post by Scott Lurndal
Hand the first: A job at Google/Twitter/FB/LI using modern technology and getting
paid beaucoup bux.
Hand the second: A job at NASA working with dead technology and getting
paid close to zilch.
Seems an easy choice.
I have the feeling that anyone over 30 years old automatically has about 10
strikes against him on your first list. So any one person isn't given the
choice you imply.

BTW, what is LI?
D. Aaron Sawyer
2015-11-02 15:10:23 UTC
Permalink
Post by Scott Lurndal
Hand the first: A job at Google/Twitter/FB/LI using modern technology and getting
paid beaucoup bux.
Hand the second: A job at NASA working with dead technology and getting
paid close to zilch.
Seems an easy choice.
I have the feeling that anyone over 30 years old automatically has about 10 strikes against him on your first list. So
any one person isn't given the choice you imply.
BTW, what is LI?
LinkedIn - the "professional" social network (a higher^H^H^H^H^H^Hmore nearly literate class of spam, IME)
https://www.linkedin.com/
Osmium
2015-11-02 15:27:04 UTC
Permalink
Post by D. Aaron Sawyer
Post by Osmium
BTW, what is LI?
LinkedIn - the "professional" social network (a higher^H^H^H^H^H^Hmore
nearly literate class of spam, IME)
Thanks. Those damned people did an amazing job of data mining on me. They
presented me with the name of a woman I know in my personal life and said
"Wouldn't you like to talk to her?". I absolutely can not imagine a linkage
they could find. But they did. I also get similar messages for professional
associates I know, but I can understand those.

I only signed up at my niece's request. She asked me to "friend" her, or
whatever they call it, so she could glow brighter. She recently got a new
degree.
BartC
2015-11-02 15:35:44 UTC
Permalink
Post by D. Aaron Sawyer
Post by Osmium
BTW, what is LI?
LinkedIn - the "professional" social network (a higher^H^H^H^H^H^Hmore
nearly literate class of spam, IME)
https://www.linkedin.com/
And all this time I thought it was Linkedln.
--
Bartc
Scott Lurndal
2015-11-02 19:21:18 UTC
Permalink
Post by Osmium
Post by Scott Lurndal
Hand the first: A job at Google/Twitter/FB/LI using modern technology and getting
paid beaucoup bux.
Hand the second: A job at NASA working with dead technology and getting
paid close to zilch.
Seems an easy choice.
I have the feeling that anyone over 30 years old automatically has about 10
strikes against him on your first list. So any one person isn't given the
choice you imply.
I'm regularly soliciated by all of the aforementioned companies and
I'm considerably over 30. I won't argue that they tend to prefer
younger (and thus less expensive) employees, but the don't limit themselves
by any means.
Charles Richmond
2015-10-31 20:26:50 UTC
Permalink
Post by RS Wood
For me the most interesting thing is the comment that the younger tech
staff don't *want* to learn the languages this project now requires.
Pathetic.
Everything is programmed in assembly language on the two Voyager spacreafts.
How does knowing FORTRAN or ALGOL help with that???

Most here on <a.f.c.> can do FORTRAN IV or FORTRAN 77 coding. And ALGOL is
*not* so different from Pascal. So where's the problem???

Perhaps NASA and JPL should consider a paradigm shift. Turn the Voyager
missions into "open source" projects. Let anyone join in and submit
changes. Have two or three people at NASA that manage the changes actually
made to the spacecraft. Release executable code for a Voyager spacecraft
emulator so the contributors will have a way to test changes. Something
along the lines of the way that the Linux kernal is managed...

Heck, the way that the U.S. government disrespects NASA, you may find that
NASA has to go to "crowdsourcing" to support its projects!!! :-)
--
numerist at aquaporin4 dot com
Hils
2015-10-31 22:32:49 UTC
Permalink
Post by Charles Richmond
Post by RS Wood
For me the most interesting thing is the comment that the younger tech
staff don't *want* to learn the languages this project now requires.
Pathetic.
Everything is programmed in assembly language on the two Voyager
spacreafts. How does knowing FORTRAN or ALGOL help with that???
Most here on <a.f.c.> can do FORTRAN IV or FORTRAN 77 coding. And ALGOL
is *not* so different from Pascal. So where's the problem???
Perhaps NASA and JPL should consider a paradigm shift. Turn the Voyager
missions into "open source" projects. Let anyone join in and submit
changes. Have two or three people at NASA that manage the changes
actually made to the spacecraft. Release executable code for a Voyager
spacecraft emulator so the contributors will have a way to test
changes. Something along the lines of the way that the Linux kernal is
managed...
That's a great idea, and also consistent with "citizen science".
Post by Charles Richmond
Heck, the way that the U.S. government disrespects NASA, you may find
that NASA has to go to "crowdsourcing" to support its projects!!! :-)
Crowd-coding?
Michael Black
2015-11-01 00:47:53 UTC
Permalink
Post by Hils
Post by Charles Richmond
Post by RS Wood
For me the most interesting thing is the comment that the younger tech
staff don't *want* to learn the languages this project now requires.
Pathetic.
Everything is programmed in assembly language on the two Voyager
spacreafts. How does knowing FORTRAN or ALGOL help with that???
Most here on <a.f.c.> can do FORTRAN IV or FORTRAN 77 coding. And ALGOL
is *not* so different from Pascal. So where's the problem???
Perhaps NASA and JPL should consider a paradigm shift. Turn the Voyager
missions into "open source" projects. Let anyone join in and submit
changes. Have two or three people at NASA that manage the changes
actually made to the spacecraft. Release executable code for a Voyager
spacecraft emulator so the contributors will have a way to test
changes. Something along the lines of the way that the Linux kernal is
managed...
That's a great idea, and also consistent with "citizen science".
Post by Charles Richmond
Heck, the way that the U.S. government disrespects NASA, you may find
that NASA has to go to "crowdsourcing" to support its projects!!! :-)
Crowd-coding?
Aren't those called "Hackathons"?

Michael
Theo Markettos
2015-11-02 11:05:08 UTC
Permalink
Post by Charles Richmond
Perhaps NASA and JPL should consider a paradigm shift. Turn the Voyager
missions into "open source" projects. Let anyone join in and submit
changes. Have two or three people at NASA that manage the changes actually
made to the spacecraft. Release executable code for a Voyager spacecraft
emulator so the contributors will have a way to test changes. Something
along the lines of the way that the Linux kernal is managed...
From what I read, the computers are custom, and what they are isn't properly
written down. Or was, but documentation has been lost. And it's not just
about the computers, but how they interact with the hardware systems and the
existing loaded software. That means that there is no emulator, and the
only example you could take apart to see how it works is a rather long way
away.

So it seems to be more of a reverse engineering job, with the caveat that
finding out whether you were right is a very high risk affair. I suspect
that if there was an accurate 'Voyager spacecraft emulator' then they would
be home and dry, but the task is more 'read several million pages of
documentation to work out what it does, and hope you aren't missing any'.

But still, crowdsourcing the research might be a good idea. Though it may
be the case that a holistic view is required, especially if the subsystems
don't decompose neatly.

Theo
RS Wood
2015-11-02 11:33:40 UTC
Permalink
Post by Theo Markettos
So it seems to be more of a reverse engineering job, with the caveat that
finding out whether you were right is a very high risk affair. I suspect
that if there was an accurate 'Voyager spacecraft emulator' then they would
be home and dry, but the task is more 'read several million pages of
documentation to work out what it does, and hope you aren't missing any'.
But still, crowdsourcing the research might be a good idea. Though it may
be the case that a holistic view is required, especially if the subsystems
don't decompose neatly.
Theo
In this case, rather than crowdsourcing the wider developer community
in general it seems they should bite the bullet, pitch the job to the
limited number of people with the appropriate background, and then pay
whatever he/she/they ask. If this is really truly a reverse
engineering effort, it's going to require someone with a very specific
set of capabilities/knowledge. And you'll (you *should*) pay dearly
for it, kicking yourself the whole time for having been so stupid as to
have lost the docs etc. It's called "paying stupid tax" for a reason :)
Charles Richmond
2015-11-02 21:12:36 UTC
Permalink
Post by Theo Markettos
Post by Charles Richmond
Perhaps NASA and JPL should consider a paradigm shift. Turn the Voyager
missions into "open source" projects. Let anyone join in and submit
changes. Have two or three people at NASA that manage the changes actually
made to the spacecraft. Release executable code for a Voyager spacecraft
emulator so the contributors will have a way to test changes. Something
along the lines of the way that the Linux kernal is managed...
From what I read, the computers are custom, and what they are isn't properly
written down. Or was, but documentation has been lost. And it's not just
about the computers, but how they interact with the hardware systems and the
existing loaded software. That means that there is no emulator, and the
only example you could take apart to see how it works is a rather long way
away.
No, I think that there *are* some manuals about the instruction set of the
computers. And I'll bet there is a copy of the hardware still here on earth
and used for testing purposes. All this only makes sense.
Post by Theo Markettos
So it seems to be more of a reverse engineering job, with the caveat that
finding out whether you were right is a very high risk affair. I suspect
that if there was an accurate 'Voyager spacecraft emulator' then they would
be home and dry, but the task is more 'read several million pages of
documentation to work out what it does, and hope you aren't missing any'.
But still, crowdsourcing the research might be a good idea. Though it may
be the case that a holistic view is required, especially if the subsystems
don't decompose neatly.
So my basic premise... that NASA should try a new paradigm for supporting
Voyager... is something you can sign up for... right???
--
numerist at aquaporin4 dot com
Freddy1
2015-11-03 00:53:02 UTC
Permalink
Post by Charles Richmond
Post by Theo Markettos
Post by Charles Richmond
Perhaps NASA and JPL should consider a paradigm shift. Turn the Voyager
missions into "open source" projects. Let anyone join in and submit
changes. Have two or three people at NASA that manage the changes actually
made to the spacecraft. Release executable code for a Voyager spacecraft
emulator so the contributors will have a way to test changes. Something
along the lines of the way that the Linux kernal is managed...
From what I read, the computers are custom, and what they are isn't properly
written down. Or was, but documentation has been lost. And it's not
just about the computers, but how they interact with the hardware systems
and the
existing loaded software. That means that there is no emulator, and the
only example you could take apart to see how it works is a rather long
way away.
No, I think that there *are* some manuals about the instruction set of the
computers. And I'll bet there is a copy of the hardware still here on earth
and used for testing purposes. All this only makes sense.
I recall that some of that hardware was modified after the fact( IE, memory
failures that had to be programmed around. Probably some other stuff. ) So
even the on-planet hardware duplicate differs from the actual probe. I hope
that they remembered what spacecraft parts those were.

Freddy,
also experiencing minor hardware failures.
--
Assembled in Mexico.

/|>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>\|
/| I may be demented \|
/| but I'm not crazy! \|
/|<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<\|
* SPAyM trap: there is no X in my address *
Nyssa
2015-11-01 15:50:27 UTC
Permalink
Post by artie
Post by Howard S Shubs
Post by Peter Flass
I can't believe that no one "wants to." What I would
believe is that NASA is not willing to take someone
with enthusiasm but no experience with this
specific hardware and train them. Like most employers
they want someone who can step into a position with no
learning period.
What I find amusing is that, as an experienced Fortran
programmer on things from IBM 1130 and newer, I've never
heard of these positions. I've just tried finding them on
Perhaps no one is applying because they CAN'T FIND THE
DAMN JOBS. But that'd be too obvious.
I'll sing in that chorus -- one of my first programming
jobs (some decades ago) was converting a specialized
FORTRAN scientific library into optimized assembly code
(on a PDP-10), resulting in many happy users of that
library.
I also did FORTRAN on the IBM 1130, Univac 1108 (two
zeroes!), many 360 systems, SDS Sigma 5, 6, 7, 8, 9, 560,
big CDC and Cray boxes.
I'm not dead yet, and I'd love a job doing that kind of
thing!
As would I, and no doubt many others with "legacy" experience.

Of course, if NASA went about hiring replacements for their
retired engineers/programmers they would probably fall into
the same pattern that I have encountered all too many times
when applying for other bit fiddler jobs as in....

"Well, you've got the languages and the experience, but it's
not specifically on the machines/os/environment we're using
and your experience isn't in our industry, so it doesn't count."

My FORTRAN days were spent doing stuff for air traffic control
simulators and trainers, weather forecasting systems, and
logistics, so I'd be filtered out of the resume pile by some
HR word-matching algorithm and never have a chance.

That or something like "yeah, you've got the languages, etc.,
but your experience in them is *recent*" and then fall through
the holes again.

I'm sure most of the rest of us old-timers with the right
languages in our back pockets would be told something similar
even if these phantom jobs did surface somewhere on the
NASA site.

BTW anyone know where these phantom jobs are supposed to be
located?

Nyssa, who has seen this play act out many times in her
professional past
Paul Sture
2015-11-04 11:45:33 UTC
Permalink
Post by Nyssa
Post by artie
I also did FORTRAN on the IBM 1130, Univac 1108 (two
zeroes!), many 360 systems, SDS Sigma 5, 6, 7, 8, 9, 560,
big CDC and Cray boxes.
I'm not dead yet, and I'd love a job doing that kind of
thing!
As would I, and no doubt many others with "legacy" experience.
Of course, if NASA went about hiring replacements for their
retired engineers/programmers they would probably fall into
the same pattern that I have encountered all too many times
when applying for other bit fiddler jobs as in....
"Well, you've got the languages and the experience, but it's
not specifically on the machines/os/environment we're using
and your experience isn't in our industry, so it doesn't count."
My FORTRAN days were spent doing stuff for air traffic control
simulators and trainers, weather forecasting systems, and
logistics, so I'd be filtered out of the resume pile by some
HR word-matching algorithm and never have a chance.
Yes, seen that in action. Some industry sectors are a 'bit special'
here - if you haven't come across it before you might not be
comfortable with the amount of politics involved somewhere like the
UK's NHS. Unfortunately the HR types seem to think that this
applies everywhere.
Post by Nyssa
That or something like "yeah, you've got the languages, etc.,
but your experience in them is *recent*" and then fall through
the holes again.
Where *recent* seems to mean "3 months". Just because you take
a break working in another environment doesn't mean that you
automatically forget your previous knowledge. My experience
here tells me that if anything I *know* which bits of my
knowledge are likely to be rusty after a prolonged break so
know what areas need refreshing, and will address that off my own
bat thankyouverymuch.

Specific software versions or hardware platforms are another
sticking point. From the fuss that various recruitment agencies
fed me during the transition from VAX to Alpha one would have
thought that none of my VAX experience was relevant to Alpha, yet
once I got my hands on one I was truly impressed with how little
had changed, even in the bowels of system management utilities.
Post by Nyssa
Nyssa, who has seen this play act out many times in her
professional past
Me too :-(
--
The major difference between a thing that might go wrong and a thing
that cannot possibly go wrong is that when a thing that cannot possibly
go wrong goes wrong it usually turns out to be impossible to get at or
repair. -- HHGTTG
J. Clarke
2015-12-19 01:55:49 UTC
Permalink
Post by Nyssa
Post by artie
Post by Howard S Shubs
Post by Peter Flass
I can't believe that no one "wants to." What I would
believe is that NASA is not willing to take someone
with enthusiasm but no experience with this
specific hardware and train them. Like most employers
they want someone who can step into a position with no
learning period.
What I find amusing is that, as an experienced Fortran
programmer on things from IBM 1130 and newer, I've never
heard of these positions. I've just tried finding them on
Perhaps no one is applying because they CAN'T FIND THE
DAMN JOBS. But that'd be too obvious.
I'll sing in that chorus -- one of my first programming
jobs (some decades ago) was converting a specialized
FORTRAN scientific library into optimized assembly code
(on a PDP-10), resulting in many happy users of that
library.
I also did FORTRAN on the IBM 1130, Univac 1108 (two
zeroes!), many 360 systems, SDS Sigma 5, 6, 7, 8, 9, 560,
big CDC and Cray boxes.
I'm not dead yet, and I'd love a job doing that kind of
thing!
As would I, and no doubt many others with "legacy" experience.
Of course, if NASA went about hiring replacements for their
retired engineers/programmers they would probably fall into
the same pattern that I have encountered all too many times
when applying for other bit fiddler jobs as in....
"Well, you've got the languages and the experience, but it's
not specifically on the machines/os/environment we're using
and your experience isn't in our industry, so it doesn't count."
My FORTRAN days were spent doing stuff for air traffic control
simulators and trainers, weather forecasting systems, and
logistics, so I'd be filtered out of the resume pile by some
HR word-matching algorithm and never have a chance.
That or something like "yeah, you've got the languages, etc.,
but your experience in them is *recent*" and then fall through
the holes again.
I'm sure most of the rest of us old-timers with the right
languages in our back pockets would be told something similar
even if these phantom jobs did surface somewhere on the
NASA site.
BTW anyone know where these phantom jobs are supposed to be
located?
Nyssa, who has seen this play act out many times in her
professional past
This seems like a good place to hang this.

I'm a quant at a Fortune 100 financial services company in Springfield,
MA. We need another one.

The main thing we're looking for is someone who knows how to write code
in some language, we don't care much which, who has the attitude "I'm
going to figure this out and get it working or die trying". We don't
expect an applicant in the modern world to know Fortran or APL or be
able to drive a mainframe, but if you do it will count hugely in your
favor.

The only hard-and-fast requirements are that you have bachelors or
higher in some technical field or other, that you pass an assessment (a
few flowchart questions and a few word problems), and that there's
nothing in your background that would legally bar you from working in
the financial services industry.

As someone who came to this position from engineering, I can say that
the hard part is likely to be the business logic, which is different
from anything one encounters in the physical sciences. The programming
itself is mostly straightforward, once you figure out what the thing is
supposed to be doing. But figuring out what it's supposed to be doing
can be a bear.

This isn't a "lost in the sea of IT" position--there are four quants in
the group and you'll make a fifth--we're high enough up that we're on a
first name basis with the Chief Actuary, who reports directly to the
board, and part of the job is doing calculations that are shown directly
to the board to assist them in making decisions (another part is
adjusting the production system in accordance with those decisions).

The environment is IBM Z/Architecture with VS Fortran and XL C. Also
PC-based APL, and the usual modern office tools. Code base goes back to
the '60s, but we add new products with some regularity. There's an
attempt in process to move off of the current system to a new one and
supporting that activity will also be part of the job (we aren't writing
that system, but part of our job is to keep that system honest).

The job carries the usual benefits.

If after corresponding with you I think you're a good candidate, your
resume goes to the Quantitative Director, not to HR, bypassing the usual
filters.
Andy Burns
2015-12-19 11:19:39 UTC
Permalink
Post by J. Clarke
If after corresponding with you I think you're a good candidate, your
resume goes to the Quantitative Director, not to HR, bypassing the usual
filters.
Is knowledge of VIXAL-4 a pre-req?
RS Wood
2017-05-04 10:40:06 UTC
Permalink
Post by RS Wood
http://www.theregister.co.uk/2015/10/31/brush_up_on_your_fortran/
//--clip
In an interview with Popular Mechanics, the manager of NASA's Voyager
program Suzanne Dodd said the retirement of the last original crew
member has left the space agency with a shortage of people capable of
communicating with the 40-year-old craft.
Launched in 1977, the two Voyager crafts rely on mid-70s hardware
powered by purpose-built General Electric interrupt processors. After
38 years in space, the two probes are currently on the outer fringes of
the Sun's influence, heading into interstellar space.
Though most of the instruments onboard the two probes have been
deactivated, both are still able to maintain contact with Earth and
will continue to do so into the 2020's, until their onboard
radioisotope thermoelectric generators no longer function.
In the meantime, NASA needs engineers capable of interacting with the
1970s-era technology, a skillset that includes knowledge of both
Fortran and assembly as well as the ability to command a machine with
just around 68KB of total memory.
"Although, some people can program in an assembly language and
understand the intricacy of the spacecraft, most younger people can't
or really don't want to," Dodd was quoted as saying.
With high-level languages now the standard for developers, knowing how
to fluently code in assembly has become a specialized skill, as has
fluency in dated languages such as Fortran. While obscure, the skillset
is potentially lucrative. Along with NASA's aging fleet of spacecraft,
many businesses still rely on ancient languages such as Fortran or
COBOL for specialized tasks and critical infrastructure. ®
//--clip
I posted the above back in October 2015. Interesting to see something
similar in early 2017. Again, from the Register:

//--clip
Feed: The Register
Title: Fortran greybeards: Get your walking frames and shuffle over to NASA
Author: Richard Chirgwin
Link: http://go.theregister.com/feed/www.theregister.co.uk/2017/05/04/fortran_greyb
eards_get_your_walking_frame_and_shuffle_over_to_nasa/
Date: Thu, 04 May 2017 03:28:13 -0400

Space agency has US$55k in prizes to those who can accelerate old code

NASA wants scientific computer experts to take a look at one of its oldest
software suites in the hope they can speed it up.…

//--clip

NASA is serious about bringing in FORTRAN developers. And now we have some
specific tasks - looking for performance improvements in old code. Wonder
if FORTRAN can do SMP these days?
Rich
2017-05-04 10:53:30 UTC
Permalink
Post by RS Wood
NASA is serious about bringing in FORTRAN developers. And now we
have some specific tasks - looking for performance improvements in
old code. Wonder if FORTRAN can do SMP these days?
Yes:
http://scv.bu.edu/computation/bluegene/IBMdocs/compiler/xlf-10.1/html/xlfopg/smpsrc2.htm#smpsrc2

There's a set of four comparitive examples (the one above being one
of the set) here:
http://scv.bu.edu/computation/bluegene/IBMdocs/compiler/xlf-10.1/html/xlfopg/sampprg.htm

FORTRAN, it seems, is still quite extensively used in scientific number
crunching.
Paul Sture
2017-05-05 11:44:03 UTC
Permalink
Post by Rich
Post by RS Wood
NASA is serious about bringing in FORTRAN developers. And now we
have some specific tasks - looking for performance improvements in
old code. Wonder if FORTRAN can do SMP these days?
http://scv.bu.edu/computation/bluegene/IBMdocs/compiler/xlf-10.1/html/xlfopg/smpsrc2.htm#smpsrc2
There's a set of four comparitive examples (the one above being one
http://scv.bu.edu/computation/bluegene/IBMdocs/compiler/xlf-10.1/html/xlfopg/sampprg.htm
FORTRAN, it seems, is still quite extensively used in scientific number
crunching.
According to a Scientific Computing course I did a couple of years ago,
very much so.

The links you give are IBM specific. I

If you want to try it at home, there are two parallel processing methods
for GNU Fortran (aka gfortran, binaries available for Linux, Windows,
macOS and more):

<https://gcc.gnu.org/wiki/GFortran>

OpenMP:

<https://sites.google.com/site/gfortransite/>

OpenMP official site:

<http://www.openmp.org/specifications/>

MPI:

<http://www.mcs.anl.gov/research/projects/mpi/tutorial/mpibasics/index.htm>


The flavour used by the Pleiades Supercomputer can be found at the bottom of
this page:

<https://www.nas.nasa.gov/hecc/resources/pleiades.html>


Operating Environment

Operating system: SUSE® Linux®
Job scheduler: Altair PBS Professional®
Compilers: Intel and GNU C, C++ and Fortran
MPI: SGI MPT

Note: Intel has its own C++ and Fortran compilers.

<https://software.intel.com/en-us/intel-compilers/>
--
Everybody has a testing environment. Some people are lucky enough to
have a totally separate environment to run production in.
Andrew Swallow
2017-05-05 00:15:27 UTC
Permalink
On 04/05/2017 11:40, RS Wood wrote:
{snip}
Post by RS Wood
Space agency has US$55k in prizes to those who can accelerate old code
NASA wants scientific computer experts to take a look at one of its oldest
software suites in the hope they can speed it up.…
//--clip
NASA is serious about bringing in FORTRAN developers. And now we have some
specific tasks - looking for performance improvements in old code. Wonder
if FORTRAN can do SMP these days?
Check you data structures when speeding up Fortran. Since it does not
have them they were not planned. I/O is the other slow item.

NASA may have used some bit handling, possibly by calling subroutines.

Just recompiling the program to run on an ARM or X'86 microprocessor is
likely to produce a speed up. Although check word length.
J. Clarke
2017-05-05 08:29:17 UTC
Permalink
Post by Andrew Swallow
{snip}
Post by RS Wood
Space agency has US$55k in prizes to those who can accelerate old code
NASA wants scientific computer experts to take a look at one of its oldest
software suites in the hope they can speed it up.?
//--clip
NASA is serious about bringing in FORTRAN developers. And now we have some
specific tasks - looking for performance improvements in old code. Wonder
if FORTRAN can do SMP these days?
Check you data structures when speeding up Fortran. Since it does not
have them they were not planned. I/O is the other slow item.
NASA may have used some bit handling, possibly by calling subroutines.
Just recompiling the program to run on an ARM or X'86 microprocessor is
likely to produce a speed up. Although check word length.
Speed up compared to what? And what makes you think they aren't running
whatever code they're running now on x86? And if they aren't, what makes
you think that an x86 can be adapted to the task? They may be dealing with
inaccessible hardware.
Richard Kettlewell
2017-05-05 08:49:59 UTC
Permalink
Post by J. Clarke
Speed up compared to what? And what makes you think they aren't running
whatever code they're running now on x86? And if they aren't, what makes
you think that an x86 can be adapted to the task? They may be dealing with
inaccessible hardware.
There is no need for anyone to speculate, they specify the supercomputer
it runs on.
--
http://www.greenend.org.uk/rjk/
Paul Sture
2017-05-05 11:10:01 UTC
Permalink
<snip>
Post by J. Clarke
Post by Andrew Swallow
Just recompiling the program to run on an ARM or X'86 microprocessor is
likely to produce a speed up. Although check word length.
Speed up compared to what? And what makes you think they aren't running
whatever code they're running now on x86? And if they aren't, what makes
you think that an x86 can be adapted to the task? They may be dealing with
inaccessible hardware.
<https://www.nasa.gov/aero/nasa-issues-a-challenge-to-speed-up-its-supercomputer-code>

"For more information about this challenge, the FUN3D software, or the
Pleiades supercomputer, send an email..."

Pleiades Supercomputer details here:

<https://www.nas.nasa.gov/hecc/resources/pleiades.html#url>

System Architecture

Manufacturer: SGI
161 racks (11,472 nodes)
7.25 Pflop/s peak cluster
5.95 Pflop/s LINPACK rating (#13 on November 2016 TOP500 list)
175 Tflop/s HPCG rating (#9 on November 2016 HPCG list)
Total CPU cores: 246,048
Total memory: 938 TB
2 racks (64 nodes total) enhanced with NVIDIA graphics processing units
(GPUs)
184,320 CUDA cores
0.275 Pflop/s total
1 rack (32 nodes total) enhanced with Intel Xeon Phi co-processors (MICs)
3,840 MIC cores
0.064 Pflop/s total
--
Everybody has a testing environment. Some people are lucky enough to
have a totally separate environment to run production in.
Loading...