Am i a Boiling Frog?
This is some writeup i wrote for our technical forum in Credit Suisse. Thought you’d be interested.
Rephrasing the adage in relation to the vi and the emacs, “Nobody uses two programming languages in a single sentence and gets away without getting involved in a war”. This article is definitely not supposed to moot a language war or a Babel Tower scene. This is just a “-verbose” of what is currently running in my mind and i am sure there will be agreements and disagreements to the thought.
Most of us here have been programming primarily in one single language for the past 5~ years and I personally feel that we are giving Larry Wall’s quote a whole different meaning – “The three chief virtues of a programmer are: Laziness, Impatience and Hubris” - We are lazy to do things the right way, we are impatient with deadlines and we are amazingly arrogant of the useless years of experience we have. The quote, not surprisingly, just meant this – http://www.hhhh.org/wiml/virtues.html. Talking about experience, a friend of mine (who had 4 years of IT experience) was asked an interesting question when he took up an interview – “What do you think your experience factor is? Meaning, Is your experience 4*1 year or 2*2 years or 1*4 years?”. The interviewer is basically asking him the amount of repetitive work he has done over the past 4 years and how much of new learning is involved.
When was the last time we read a technical book?
Most of us commute by train and I am sure most of us are sick seeing the same old trees and roads. Pick a book, any technical book this month and see the difference in your confidence.
Is our testing process systematic?
Diving into a fully blown Agile methodology or XP (http://en.wikipedia.org/wiki/Extreme_Programming) overnight is a difficult task. However, it is not difficult going for Test Driven Development (http://en.wikipedia.org/wiki/Test-driven_development) or incremental writing of JUnit/NUnit testcases to cover major portion of your code.
Are we writing code which are actually Object oriented or reusable?
This is tricky. We know that just because we use a Object Oriented language doesn’t make our code object oriented. For example, programming to interface and not implementation is something which was advocated in Gang of Four back in 1995 (http://www.c2.com/cgi/wiki?GangOfFour) but except for the forced coding to interface in EJBs, I personally didn’t see its usage anywhere else. And even more demotivating is our usage of Abstractions.
Do we have a proper code review mechanism?
“Given enough eyeballs, all bugs are shallow”. That was Eric Raymond in “The Cathedral and the Bazaar” (http://www.catb.org/~esr/writings/cathedral-bazaar/cathedral-bazaar/) talking about Linux and the Open Source movement. We definitely can’t/don’t do pair programming to have line-by-line reviews. However, it is always good to have a review, at the least for a major enhancement.
Are we taking the “shortcut” method to fix an issue instead of the “long but right” method?
Is our code a broken window? (http://www.artima.com/intv/fixit2.html).
Lastly, Are we becoming the Boiling Frog? (http://en.wikipedia.org/wiki/Boiling_frog)
What do you think?
