This is AndersMi's Typepad Profile.
Join Typepad and start following AndersMi's activity
Join Now!
Already a member? Sign In
Recent Activity
And I have yet another different story. I've always enjoyed writing code. I started when I was 13, writing programs in basic as a hobby. Never been a real supergeek but in many years of intermittent experience I wrote some interesting and sometimes pretty complex stuff and experimented with various languages. Now I'm 38. I've graduated in computer science at university, so I have some formal background and I work as a developer. And for the first time, more and more often I find myself thinking that this job is not for me. The reason is that I have to deal every day with tons of incredibly complicated code that is for the most part completely useless. Absolutely trivial features that should be implemented in the matter of hours, end up requiring the effort of whole teams for days and weeks: to write the code, to fix the problems arising from this or that (usually unneeded) framework, to fix the bugs hidden in thousands of lines of boilerplate code. And all this stuff leaves no time at all for the better part of software development- which is, creating cool stuff that goes beyond the (most basic) expectations of the customers, doing more in nicer and better ways. And the cause of all this lies in a generation of developers that are pedantic as lawyers, driven by fear that anything may go wrong, blindly applying rules they don't understand and repeating over and over exactly the same catchphrases they've read in some blog or software architecture manual, no matter how many times reality proved them wrong. (One of my favourites: "we use this tool/ framework/ library because it's standard, so any developer knows it" but then complaining that the same tool/f/l "in itself is good, but almost nobody uses it in the right way"). It took me the last few weeks to implement minimal changes to a web application that provides trivial CRUD functionality on five or six simple entities: an application of six pages in all but with a total of ten thousands lines of code, and bugs everywhere, from the conceptual and domain level to the architecture, implementation and data model: but everything strictly written with dependency injection, assorted patterns, WCF and NHibernate, CQRS, and whatnot. What? Change company? Useless. The last two or three were the same. The messier and more bloated the application, the bigger the team that's needed to maintain it, and so the greater the importance of its architects/ team leaders inside the organization. Entire companies are built around applications that are barely able to deliver the minimum of functionality needed to run the business, with big dev teams absorbing disproportionate amounts of resources to keep piling crap over crap, but always following the so called "best practices" everybody is babbling about. And I don't see this thing coming to an end anywhere soon. There is a principle, that any bureaucratic organization tends to obfuscate its knowledge and to grow in size as much as possible; it seems to apply well to what's happening the big parts of the software development world. As a boy (and as an adult as well) I fancied programming as a field where skills and results have the maximum value, and buzzwords and authority principles are worthless. I'm afraid now that this is not the case anymore.
AndersMi is now following The Typepad Team
May 29, 2013