Fog Creek Software
g
Discussion Board




The card reader was two miles away - uphill!

Anyone else have favorite quotes from older programmers who remember how much better everyone coded in the good old days?

"Back when we were doing Fortran 4, we didn't have any of those 'if' statements.  We had to Really know what we were doing."

Boofus McGoofus
Monday, April 19, 2004


"All we had were 1's and 0's.....and sometimes we didn't even have the 1's!"

Mark Hoffman
Monday, April 19, 2004

We had two sticks -- bang 'em together once for "0", twice for "1".

Took about six weeks to balance my checkbook.

Should be working
Monday, April 19, 2004

There's some truth to that.  At least in the 'old days' stupidly verbose solutions were punished more severely.  The recent cheapness of bytes has enabled pasta-oriented programming more than anything else.


Monday, April 19, 2004

What does spaghetti code have to do with storage space?  It's a potential problem in procedural languages of any size...and it's a programmer problem, not a machine problem.

Kyralessa
Monday, April 19, 2004

>What does spaghetti code have to do with
>storage space?

I wasn't just talking about storage space (though memory constraints do select for more efficient programs).  I was talking about the byte-TCO (that is, including input).  The penalty for spaghetti code is higher when it's harder to enter code.  I don't actually want to go back to punch cards, but I've seen too many 100KLOC projects that should have been 20KLOC.


Monday, April 19, 2004

I know it's supposed to be a joke, but my very first programming job was for a Data General OEM using a proprietary language acronymed PLD (go ahead, ask - I dare you).

It involved the use of a great deal of mechanical pencils and reams of paper that eventually got returned when the keyboard input operators typed it in and tried to compile it. 

Of course the keyboard operators usually got stuff wrong, but what I recall was the great care I took given the effort involved to simply get a compile out without errors.  I wouldn't want to go back to then, but I know I'm not as careful now as I was then about double, treble, etc. checking my code.

So, maybe there is at least a grain of truth in it, no?

Then again, maybe I'm just getting lazy in my old age.


 

Mongo (suddenly feeling antediluvian)
Monday, April 19, 2004

"The penalty for spaghetti code is higher when it's harder to enter code."

Not sure I agree. I think a lot of spaghetti code comes from overly clever optimization for size - jumping and branching all over.

Anyway, true story:

I worked for a guy how developed one of the first computer controls for machine tools. With only 4K bytes of code and assembly language it was a challenge. The code that interpolated the seris of points for the machine to move to used self-modifying code. After determining whic direction a move was going to go, opcodes were changed from add to subtract or vice versa as necessary.

We would scream about that now (if we even used assembly!) but it was the only way to make it fit.

sgf
Monday, April 19, 2004

"...my very first programming job was for a Data General OEM using a proprietary language acronymed PLD (go ahead, ask - I dare you)."

http://www.acronymfinder.com/af-query.asp?p=dict&String=exact&Acronym=PLD

I'm gonna have to go with Personnel Lowering Device.  He was programming an elevator.

What is spaghetti code, anyway?  I think people are using differing definitions here.  Mine would be similar to sgf's: code that jumps and branches all over the place and is hard to follow.  It wouldn't necessarily have to be sloppy code except for that.  I think others may be using it merely as a synonym for sloppy code.

Kyralessa
Monday, April 19, 2004

Drumroll, please ...

PLD: Programming Language, Dummy (delivered, of course, with a look of utter disdain and contempt).

The error messages, the _end user_ error messages, also all began "Aw Shit, Error x ..."

I am _not_ making this up, and it still cracks me up, which only goes to show that I may get old, but my sense of humour stays hovering right around puerile.


sgf:

On another note, the guy who wrote this language and the OS around it was one of the earliest machine tool programmers and did many brilliant and scary things ... Was your guy working for Rockwell, by any chance?

Mongo (suddenly feeling antediluvian)
Tuesday, April 20, 2004

I get this sort of thing off my mother.

She's never really approved of the idea of having the computer in the same room as the developers. She certainly doesn't like the sort of rapid "try-it-fix-it" hacking I use to write Perl. Using a computer to syntax check code is just *criminal*...

When she was a developer, the machine was on another site, and they had to put the boxes of cards in the company internal mail.

Katie Lucas
Tuesday, April 20, 2004

AH yes, the old, "If you were really productive they would let you see the lights on the computer" thing.

I believe there was a story in Steven Levy's "Hackers" along those lines.  Folks working for the Navy learning to program, and when they finally got far enough along they got to actually visit the machine.  One woman fainted, overwhelmed by the sheer blinkenlichtness of the machine.

I still get a chuckle when we get visits from doctors evaluating the EMR system we use.  They always stand in awe of the computer room, with all its blinking lights (none of them currently red, knock on wood).  We actually went so far as to install a glass wall, so they can look without having to face the temptation to push buttons.

I admit, with the lights out in the building, it is pretty in a Christmas tree sort of way.

Steve Barbour
Tuesday, April 20, 2004

> Mongo (suddenly feeling antediluvian)

Stop feeling my wife, you b*stard!

Uncle Diluvian
Tuesday, April 20, 2004

True comment from the 80's when I was a novice programmer...appreciated only by assembly language programmers!

"You have 'NOP' instruction now...we had to make do with 'ORA 0' "

Code Monkey
Tuesday, April 20, 2004

Katie Lucas writes:
> I get this sort of thing off my mother.

> When she was a developer, the machine was on another
> site, and they had to put the boxes of cards in the
> company internal mail.

When I was in school, we were able to walk right up and load our own cards into the reader, but as the system got busier near the end of the semester, one frequently didn't get the results of one's "run" back (i.e. a print-out) until the next morning. 

(Yes, we soon learned to be careful with those semi-colons.)

Old enough to be your father (apparently)
Tuesday, April 20, 2004

In the late 60s I took a Fortran 4 class, taught once a week through American University, at a local high school.  At the end of a class we would get a homework assignment (e.g. The Pythagorean Theorem) and hand it in on paper the next week.  The prof would give the classes papers to his secretary who would punch-card them and run them.  The NEXT week we would get back the output ( wrong, of course ), so the week after that we would hand in the correction.  It would be punched and run for the NEXT weeks class.  So, with one small error it would take a month from the time of the assignment to get a correct result!  When I heard that we would all be talking to computers soon, making programming unnecessary, I had no further interest in the field (I changed my mind sometime later!).

Barry Sperling
Tuesday, April 20, 2004

Things are more like they are today than they have ever been before.

Doug Withau
Tuesday, April 20, 2004

Katie Lucas writes:
> I get this sort of thing off my mother.

Me too.  She's actually the source of the quote in the OP.  Normal families argue about money or "why didn't you love me as much as Bobby?"  In our family, we argue about whether gotos are the Spawn of Satan.

She also talks about how everyone triple checked back in the old days because it took 2 days to get a run back and how you never lose that.  She recently took a Java course and called me while debugging a program.  She took great offense when I suggested she count the brackets to make sure she'd closed everything.  I took great glee in learning that that was actually the problem. 

Petty?  Shallow?  Yes, that's me.

Boofus McGoofus
Tuesday, April 20, 2004

Re: watching the buttons light up

We hired a woman a couple years ago -- she had 5-10 years experience programming on the IBM AS/400.  Her first day working for us was the first time she'd ever seen one.  It was, in fact, the first thing she wanted to do after she put her purse down.

Boofus McGoofus
Tuesday, April 20, 2004

*  Recent Topics

*  Fog Creek Home