Reed Me : fungibility

Self-fulfilling prophecies in computer science education?

reedme — Fri, 12 Dec 2008 00:38:58 GMT

Everybody has heroes. Bjarne is one of mine. If I have to explain why to you, then you wouldn’t understand even if I did try to ‘splain it. Heh. (That comment will be funnier later in this post, if you like irony.)

I’ve done some recruiting events for Microsoft at my alma mater (the University of Houston) and I’ve done my best to influence lots of younger folks to get into computer science, and my experience tracks very closely to what Bjarne describes in this great interview:

[James:] In this interview he speaks frankly about the challenges and problems – and improvements being made – in computer science programs. Among issues like perceptions of offshoring and the need to balance the theoretical with the practical, he addresses complaints by tech companies about the lack of fully qualified CS graduates.

...

[Bjarne:] The US industry could absorb more good developers than there are currently students enrolled in IT-related programs – but not all of those programs and all of those students would qualify as “good” in this context.

Bjarne Stroustrup on Educating Software Developers

Personally, I think “we” (culturally, IT-oriented companies, etc) are going to have to staff jobs with less-than-ideal candidates and train them up rather than count on young kids to bet their futures and careers on computer science education. I was involved with a tax-incented project to provide technical support services back in the ‘90s; it was a project much like the one that I’m proposing, except that more training would be required for my program to turn non-geeks into coders, programmers and someday developers. Interesting. Have I changed my position on fungibility?

The outcome of the project/company (which shall remain nameless) wasn’t any better or worse than the other classic “professional” IT projects/companies that I’ve been involved with, and more successful than most in the long run.

I realize that risk averse companies aren’t likely to try out my rural sourcing ideas (management just thinks I want an excuse to work from a mountain top in Wyoming), so when I’ve accumulated enough capital, I guess I’ll have to try it out myself. The potential upside is good, both tangible and intangible. I especially enjoy seeing people succeed at tasks that they didn’t believe that they could accomplish… all it usually takes is a little encouragement and an opportunity!

In the meantime, I’m glad that smart people like Bjarne are out there working on changing the nature of the self-fulfilling prophecy so that it’s a prophecy that we actually want to come true!

Do you want the bad news first? Or the worse news?

reedme — Tue, 18 Nov 2008 21:16:32 GMT

Oh noes!!

"We're interested in seeing COBOL as a first-class citizen on .NET and on the Azure cloud," [chief technology officer of application modernization at Micro Focus] said. "Wherever cloud computing is going, we want COBOL to be running there." Taking COBOL into the Cloud? (eweek)

Modernization? Srsly?

I'm going to censor myself and not say anything further.

meh.

p.s. If you're @ SCAN or SQL PASS this week, come say howdy. I'm wandering around looking suspicious.

Can you tell a great architect from a bad one?

reedme — Fri, 11 Jul 2008 21:24:38 GMT

Since I'm going to be sitting as a panel member on an MCA review board soon, I'd like to think I can, but apparently some researchers are having trouble...

Highly capable designers have an incentive to choose somewhat more difficult designs to better prove their talent, while less-capable designers have an incentive to choose highly difficult designs to obfuscate their lack of talent, Prof. Siemsen concludes.
Software Designers Strut Their Talent At Cost Of Profit, Says Study

This doesn't even get into the real reasons (Second System Syndrome™, Perfect System Syndrome™, Repeat Business Syndrome™, etc)... It seems to lay the blame at the feet of career concerns, but I'm not entirely sure that's the whole picture, especially when consultants get involved.

Irreducible complexity does exist, but if it seems Too Complicated™, it probably is, and you're probably Doing It Wrong™.

Great architects do The Right Thing™ for their companies or clients: a design that's as simple as possible, and no simpler.

Consultants: You'll never run out of repeat business from satisfied customers if you leave them with something that does what they need AND that they can maintain themselves. I always used to announce during project kick-off meetings, "I am leaving in X days/months/weeks. One of my goals is that you be able to absorb and maintain the deliverables that I'm leaving behind, so if it doesn't compute while we're working together, ASK me to explain it and WRITE down the answers so that you can retain the knowledge for your company."

One thing leads to another... leads to fired?

reedme — Sat, 19 Apr 2008 02:04:08 GMT

I'm not sure what I'm supposed to get out when people start sending me links like this:

10 Signs Your Company Wants You Gone (eweek.com)

Why Your Boss Doesn’t Want You to Telework (webworkerdaily.com)

If the second one is actually #11 from the first...?

C'est la vie.

Heh. Should be fun picking tags for this post.

Are you fungible? Or are you a funge?

reedme — Fri, 08 Feb 2008 02:53:25 GMT

A flamewar that I've been having with Steve (over @ SQLServerCentral.com) got me thinking about developer fungibility again. It was something that I struggled with when managing teams of developers in previous lives. (Thankfully, I don't manage anybody anymore. Not even me. Heh.)

It almost seems like humans (at least those in IT) have a natural tendency toward insectlike specialization in skills and behavior. Whether it's an aversion to continuous learning or the comfort of being an expert within a niche, I'm not sure. Maybe multiple reasons. I'll leave it to the colleges of education and organizational learning to NOT figure it out. [Please ignore the author's personal biases. -ed]

What worked for us (in one previous life) was a hybrid approach to agile development that combined XP and Scrum seemed to help keep my developers working as generalists across the entire codebase (~150M lines of code). They still had to be prodded to "volunteer" for tasks in areas of the codebase that they weren't familiar with, but... It was good to be the boss! Now, remember that this was back in 2000 and following, before "agile" became a cool buzzword... and there was both senior management and developer resistance to it, which I didn't entirely understand.

Management seems schizophrenic about developer fungibility, or perhaps just doesn't understand the learning cost of deep expertise in an individual technology. Managers want to be able to plug any developer into any task and get results; at the same time, managers want to be able to pigeonhole skill sets to make them easier to check off on an HR worksheet as needed. Seems to be orthogonal to me, but... both desires seem to be a fundamental desire for simplicity in managing developers. Since you can't have both, it would seem to behoove management to require developers to rotate around and not become insects. It's really difficult to interview for fungibility, though. Especially since it's not a word commonly taught outside of MBA school.

Developers, on the other hand, don't seem to have any ambiguity about it. They/we (in large part, not all of them/us) are averse to constantly learning new things. It takes time. It doesn't seem to have measurable value. And it's got a shelf-life. If you don't believe me, you've never dusted off code that you wrote a year ago and wonder "who was smokin' what when they wrote THAT"? Of course, check-in comments don't lie (about who did the check-in). Despite the fact that a fungible developer is more valuable to an organization long-term, the incentive to overspecialize is still strong. While the talented geek pool is still too small for the demand, it will take an individual choice not market forces to change the behavior of the developer population. Don't believe me? There are still companies hiring COBOL programmers and developers who refuse to write anything but command-line interfaces...

Not sure where I was going with this or why I started, but I guess that's normal... Hrm. Maybe it resonated because I've seen a lot of developer behavior lately that seems to be a bias toward insectlike specialization with absolutely no interest in things that aren't in a specific, narrow niche. To the degree where it even influences product direction. Things outside the "known" are apparently scary to many people. Don't look out the windows! (Take the pun any way you like.)

In Dave's perfect world, management needs to tell HR to get stuffed and figure out how to interview for general talent and the ability to learn quickly AND developers need to actively seek positions where they'll constantly be challenged with new and different work. But nobody asked for my opinion. You get what you paid for it!

I personally don't want to go the way of the typewriter repairman, which is why I'm always trying to be at the leading edge; and I actually happen to enjoy learning new stuff and plumbing the depths of other peoples' code, but I wish that more prospective employees did -- because someday I'll probably start another technology company and be looking to hire some fungible developers and I hope by then that the colleges and universities are turning out lots and lots more flexible-thinking coders.

Yes, they're both words. It's called a vocabulary. Look it up! :-P fungible | funge