The decline of programmers in the U.S.

[Katja]

Guy shows up with a shoe box of punch cards! I may be old, but I am not that old.

Yeah, only we old fogies know the terror of dropping our decks.

[crypticgimp]

When I worked at NCAR, we scripted in python, perl, etc...but the climate forecasting codes are all written in Fortran (running on supercomputers).

interesting. we have some few climate predictor folks at RIT that were able to predict earthquakes/hurricanes etc(actually a lot of the images on the news after what happened in haiti were generated from the guys at rit) but they use c/c++ i believe.

but it seems to be a function of where you go/who you're with as each person/group thinks their method is "the best" where my thinking is that you use the program/tool that best fits the job. you wouldnt nail in a nail with an axe, for example

[Katja]

but it seems to be a function of where you go/who you're with as each person/group thinks their method is "the best" where my thinking is that you use the program/tool that best fits the job. you wouldnt nail in a nail with an axe, for example

Legacy also plays a big role. Many of the climate model codes have been around for decades.

[crypticgimp]

Legacy also plays a big role. Many of the climate model codes have been around for decades.

true... but so have many other algorithms which can be implemented in different languages. just because it has existed and mainly in use in one language doesn't necessarily mean that is THE language to use.

[garvey]

What makes a good programmer?

I'm curious to hear opinions on this.
I've worked as programmer but I never considered myself to be any better than competent.
Yet I worked with people who seemed to be naturally gifted at it and could crank out high quality code at incredible pace.
I also noted, through the years, that programming skill seems to co-exist with musical talent in many individuals. i remember spending some time "on the beach" at one of the consulting firms I worked for and 80% of the folks waiting to go to their next gig seemed to be in bands of one kind or another.

[Foolish Old]

true... but so have many other algorithms which can be implemented in different languages. just because it has existed and mainly in use in one language doesn't necessarily mean that is THE language to use.

The legacy question is usually one of a reluctance to abandon previous investments.

[Katja]

true... but so have many other algorithms which can be implemented in different languages. just because it has existed and mainly in use in one language doesn't necessarily mean that is THE language to use.

While almost every programmer believes that porting or refactoring legacy code is a good idea, in aerospace we've found that given a working legacy code base, porting almost never pays for itself. It's fun to do, though.

[hlh]

I'm curious to hear opinions on this.
I've worked as programmer but I never considered myself to be any better than competent.
Yet I worked with people who seemed to be naturally gifted at it and could crank out high quality code at incredible pace.
I also noted, through the years, that programming skill seems to co-exist with musical talent in many individuals. i remember spending some time "on the beach" at one of the consulting firms I worked for and 80% of the folks waiting to go to their next gig seemed to be in bands of one kind or another.

I think programming is like speaking a language. Music is a language. Many people who are good at "languages" and develop these areas of the brain that contribute to this, likely can apply those abilities to other "languages".

Of course, programming can be a great high paying job with lots of consulting opportunities (ie. part time/off hours/telecommuting as well), allowing people to pursue their hobbies with more vigor.

[willingtocope]

Yet I worked with people who seemed to be naturally gifted at it and could crank out high quality code at incredible pace.
I also noted, through the years, that programming skill seems to co-exist with musical talent in many individuals.

You're right that like music, programming is really an art...not a science.

As to "high quality code at an incredible pace", I've found that can be deceiving. In my opinion, high quality code is easily maintained and/or modified as the needs of the system change. (Some studies have shown that as much as 99% of the true cost of a system is in the maintainence). In my experience, quickly written code is harder to maintain.

[t8burst]

You're right that like music, programming is really an art...not a science.

As to "high quality code at an incredible pace", I've found that can be deceiving. In my opinion, high quality code is easily maintained and/or modified as the needs of the system change. (Some studies have shown that as much as 99% of the true cost of a system is in the maintainence). In my experience, quickly written code is harder to maintain.

I don't think he was talking about quickly written code. If you make light bulbs, your best light bulb maker may be 10% better than your average one. Your best programmer can be orders of magnitude better than your average one. It is not about number of lines of code, it is about the art. Some people can look at a problem and do it in 10 lines and an hour where an average person may take days and 100s of lines.

[willingtocope]

Your best programmer can be orders of magnitude better than your average one. It is not about number of lines of code, it is about the art. Some people can look at a problem and do it in 10 lines and an hour where an average person may take days and 100s of lines.

I certainly agree. The "art" is in the "chunking"...the ability to break a complex problem into small, coherent, and maintainable pieces...and the abiity to name those pieces so that their function is apparent.

If you can define coherent "chunks" in your mind, you can write maintainable and useful code...quickly.

[markg]

This is so true.

My Dad used to say that a desk top PC these days has more computer power then existed in the ENTIRE WORLD, when he started programming. Something like that..... Crazy.....

In college one of my instructors shared that when he was pursuing a graduate degree in computer science at UCLA the lab was absolutely freezing inside. It was so cold, and a true testament to how far technology has come, as a lab worker he used to sleep on the CPU during breaks because it was warm.

[willingtocope]

he used to sleep on the CPU during breaks because it was warm.

Would you beleive that in the 70's and 80's some of the cpus were actually water cooled because they generated so much heat.

[t8burst]

Would you beleive that in the 70's and 80's some of the cpus were actually water cooled because they generated so much heat.

The CPU on my gaming machine is water cooled

[BSgimp]

Hello,

I am a Computer Science student and I am about to graduate this winter. Do you guys have any advice for the likes of me. I read the thread and see some dont really hire people straight out of college. It is a scary time ahead of me as I transition into industry. I am not sure what to expect. I study in City Univeristy of New York and program here didnt touch anything else other then assembly, C++, C# and Mysql. I learned a lot about Comp architecture and deisgn but I am not sure how that applies to Software Eng jobs. I think I have sharp critical thinking skills and I have many ideas for projects I want start after I graduate but our team conists only of my friends from college. We dont have a professional setup ..just our minds and perseverance.

You guys seems to understand the industry well. What would you say to fellow gimp aspiring to have a career in Cmputer Science. I love it and I want to make a living this way.