Reader Feedback to "Deliver Us From Genius"

Editor's Note: Here's what readers had to say about Linda Hayes' Oct. 17 column titled "Deliver Us From Genius":

I doubt seriously that you've written much code. I am a developer with QA experience. SQA, that is. The thought that SEI and creativity or SEI and bleeding edge are incompatible is simply not true. SEI is about control and repeatability. You can implement SEI at the car wash as well as in exploring the hidden truths about dot net. SEI is more about software management than about genius. If you want to place blame, you'll have to look someplace else. -- Adrian

Linda's Response: Thanks Adrian. l appreciate your feedback! I take it as a compliment that you took the time to write.

Believe it or not, I actually developed a software application called Petroware that was marketed over a period of 15 years to over 600 companies (and is still in use today), but I confess it was in the dark ages of 1975-1980. What was interesting about it was that when the PC came out we spent a huge amount of money to completely rewrite it, using all the latest (at the time) technology, then ended up throwing it all away and buying a compiler instead so the old code would run on the new platform instead. The reason? The new tools were not yet mature enough.

It was not my intent to place any blame anywhere, just to say that offshore development is a real threat and a huge part of it is pure economics. I agree (and don't think I said) that SEI is not incompatible with either bleeding edge or genius, only that it is a discipline that places the process above the individual.

I do admit, however, that I intentionally took a controversial tone...precisely to elicit some type of response. Looks like it worked. :-)

So...what do you think about offshore development and how we can combat it?

Adrian's Response: Yes, I guess you got me! I have a passion for quality. In the history of mass email, I have probably responded to only one other solicitation, and that was political.

In the ten or so years since I've left the quality profession, I have yet to work for a company that has a clue about software development processes. I have also learned along the way, that very few people want to learn. Most of my time in SQA was for a large software development house. Without quality processes, the product - including several million lines of code - would have not made it out the door. At the time, they were also examining the possibility of offshore development with Russia. We actually hired a consultant from a newly formed Russian Tech Consortium as part of a national deal to help post-Communist Russia get on its feet. The exchange never progressed beyond hiring one person for one year, mostly because of language barriers. They couldn't write a specification properly. Also, the Russian company was too new to be process oriented. But the desire to make it work was impressed upon everyone from up higher in the corporate structure. At the time, there wasn't much fear because every estimate indicated that developer demand would always exceed supply.

Now that the economic bubble has burst, I can see why this would become a concern again. What I lack to offer a proper response to your question however, is knowledge of how much US development has been going on offshore in the last ten years. I can only assume it has been on the rise, but by how much and in what relation to total development I don't know. In general however, I would say there are two fronts to fight. First, dispense with the notion of a free market economy. In any other sector, there isn't one, so why would we try to lay that load on the shoulders of developers. The point being, let's regulate. Corporations are not known for their national spirit. They only respond to the law ( somewhat ). Secondly, educate management. The cost benefit analyses that process improvement brings, must be understood and enforced at the highest level. Most low level managers, even if they had understanding and conviction, are not motivated to take action that would benefit the company a day after a product they are responsible for delivering has been released. Ahh, the old QA battleground. So glad I left! The truth is that most smaller companies that I've worked for are inefficient enough in the use of their human resources that other factors besides software quality would be better targets for overall improvement to the bottom line. -- Adrian

Linda's Response: Thanks, Adrian, for a thoughtful response. I'm with you - the SQA battle is a long, hard, and often hopeless struggle - not because we don't know what to do and how to do it, but because corporate priorities and management incentives don't factor quality into the equation. Maybe the "new new economics" (not to be confused with the old new economics in which money was free and the attitude was don't worry, be crappy) will make a difference. Or maybe not...

My compliments on a very thoughtful and insightful article on the strengths of off-shore development teams. I am a lawyer practicing in the IT area and have numerous experiences with US cowboy geniuses.

Recently read Pankaj Jalote's Software Project Management in Practice, a book which is a full case study of an Infosys development. As you indicate in your article, these non-US outsourcers are advanced on the CMM scale, and it shows up in their work and the value they add for customers.

My compliments on your article; it is great to read people who don't mind saying things that aren't popular. -- Kenneth

Linda's Response: Thanks, Kenneth for the kind comments. As you pointed out, my point was not too popular and not all of the responses I received were complimentary. Sometimes it's what we don't want to hear that is what we need to hear the most!

My compliments on this excellent article and the topic (Deliver us from genius - Comdex Marketplace). Very succinctly put - Picasso to paint highway stripes!! That says it all. -- Madhu

Linda's Response: Thanks, Madhu. I'm glad you enjoyed it...not everyone did, as you might imagine.

You didn't mention the ownership of the (Indian) programmers' works. Still westerners. The web is rapidly reducing the game to fewer parameters. .NET will do more of that for services, which is what software largely delivers..

When they predict that 50% of software developers will be gone in five years, they don't mean replaced by Bangalore. They mean software will be on the periphery of important code, with ASP's reducing it to garage work.

So we don't compete against them, we specialize in the Internet, which they do not and will not. -- Dwight

Linda's Response: Thanks for the comments - you're right about who ends up owning it. But I'm afraid that even the Internet is not immune - see GM Drives Application Development Offshore.

However I'm not sure I understand your point about software being on the periphery and the impact of ASPs. Do you mean fewer people will develop code because it is easier to rent? If you get a minute, I'd be interested in some clarification. It's an intriguing idea.

about time someone said this and it is high time people started to dust down their copies of the decline and fall of the american programmer -- Philip

Linda's Response: And thanks for reminding me! I'd forgotten all about that book.

I read your article, "Deliver us from genius," with some interest. We're doing offshore work right now. I have three craftsmen and an apprentice on staff - the offshore team allows us to have them work on interesting, leading edge stuff in a scalable way while the offshore folks build out things we need, learning new things as we learn them and transfer the knowledge.

The big problem with an offshore company being CMM 4 or 5 certified is the impedence mismatch between them and us - we're using iterative techniques in our offshore efforts because:

1) We'er not CMM 4 or 5 and don't plan to be anytime soon, so we'll never feed them what they want, 2) We've always been willing to toss some amount of work here - they're cheaper so we can actually afford to toss more, and 3) They work for us, we're using iterative techniques, and they want to climb the software food chain (up from maintenance).

I have to get back to work.

Thanks and have fun! - Bob

Linda's Response: Thanks, Bob. Your arrangement is intriguing and it sounds like it is working for you. I'm sure that is every developer's dream - to work on the edge and let the rest go elsewhere.

I realize you are busy but you might consider letting me interview you some time for a future column or article about how to integrate the best of local and remote talent, staying on the edge without getting cut.

I concur whole heartedly! I have spent most of my career developing simple solution to major problems.

But then the people that got the big "bucks" and the quick promotions were the ones who built the "bleeding edge" systems costing 20 to 100 times what I developed. A major development effort with all kinds of fancy "bells and whistles" is so much more impressive. -- Steve

Linda's Response: Thanks, Steve. I know what you are saying. Maybe a point I missed is that it isn't just developers who want to pad their resumes, it's behavior that is actually rewarded by management. What's up with that, I wonder? Maybe now with the budget squeeze some economic sense might prevail. Or not...

I couldn't agree more. Great article. My motto is the simpler the better. "Elegant" code is just an alias for incomprehensible. I have clients like IBM, Union Carbide, Verizon Wireless and Cigna insurance using my applications. They all work. They all tell me how happy they are. I've become enamored of Visual Basic 6 and I like the cleanliness of VB.Net. I don't use a lot of comments but my variable names are "self documenting" and I can even understand my old code six months later. -- Frank

Linda's Response: I love it! A developer who admits they like VB - I have two developers, one who likes and uses it and the other who sneers at it ("real developers code in C"). Yet...the VB guy can put out 3X the functionality in the same amount of time. BUT - don't tell anyone I said so, or the office atmosphere could be a little tense.

Bravo! -- Jim

Linda's Response: Thanks, Tad. I can tell you are a man of few words... I'll bet your code is consise, too! :-)

Jim's Response: Thanks Linda.

I'm actually TOO verbose sometimes, but I guessed you were a busy woman and didn't have time for a bunch of dribble.

At any rate, as a 20-year IT person, I think we as a community of professionals have to realize that the computer is not our playground, it is to the business what a tractor is to a farmer. If you find the farmer merrily driving his tractor around the field with no plow attached to it, he ain't gonna grow much corn. Sooner than later, he will be without the tractor!

The IT community, over the years has adopted that attitude that makes them more enamored with the technology (as you expressed) than delivering the goods. We've given business no option than to drive costs out and those who are capable of delivering the goods are going to and they will be the survivors.

Personally speaking, I am OK with foreign workers taking over the programming tasks, much like I was OK with them taking over textile manufacturing, we (Americans) have a long history of discovering things and allowing others to take over the mundane things once we have them all figured out and we move on to 'bigger and better' efforts.

Will there be short term pain? You betcha! But will we survive, prosper and move on to a new height, it's a sure thing.

Thanks for telling it like it is. We need to hear more of that kind of stuff.

Yours was an interesting read. Keep it comin'. -- Jim

Linda's Response: I like your optimistic approach. It gives me hope...

Editor's Note: Please post your own feedback in our discussion forum

Copyright © 2002 IDG Communications, Inc.

  
Shop Tech Products at Amazon