Monday, March 01, 2010

Google Coder

In another life, I was a programmer and a pretty mediocre one. However, having learned from people who really knew what they were doing, I have appreciation for and do recognize good code when I see it. It is a lot like not having the talent to be a become a musician but enough training to be an aficionado.

This article on how to hire good programmers (and weed out the fakes, liars and duds) attracted an assortment of interesting comments. One of them is by a self-professed "Google" coder and I am completely able to relate to his perspective.Often, I find myself in a position where I need something quick and dirty built as a proof of concept for something I am trying to pitch to a client. There is no one in the IT shop that is willing or able to help out. The red tape is enough to strangulate.

Under the circumstances, I have sought out open source software that met my needs and cobbled together something which a lot of help from Google advance search. On an unrelated note - I just don't get Bing - nothing seems to make sense when I type in a search term. The results are so far off the mark that I don't know where to go from there. But I digress. The fact is that my crudely constructed demo apps do end up working reasonably well and gets the job done.

Every once in a while, someone will like it enough to want to keep it around until IT is able to replace the functionality it provides with something enterprise standard. Often that never ends up happening because the cost of doing stuff the "right" way is a little too high and budgets simply don't allow such luxuries these days.

When I look at it from short term standpoint, it seems like a competent Google "coder" like the commentator Steve or myself get the job done quickly even if not in the best way possible. If the better, alternative way is so hard and expensive that no one wants to take it on, it begs the question what is "good" worth.

Obsolescence is inevitable in this line of work. Whether you throw the best architecture and design talent at something and create a high class product or get someone who can Google search their way around to standing something up that works, both will be replaced by something better soon enough. There is a lot to be said for standards in systems and process, one version of truth when it comes to data but that is hardly the reality on the ground in large IT shops.

Over the years, I have tried to encourage business customers with some inclination for technology to try and be self sufficient. If their problem calls for a bicycle there is really no need to build a space shuttle. More often than not they can rig up a bicycle by themselves - Google is their friend too.

No comments: