I’ve been actively coding for 15 years now, ever since I’ve discovered the Interwebs. If you count in the lectures I took for ZX Spectrum 48 during my elementary school, this range stretches to more than 20 years, but that was more fun than business. In this time, I’ve developed more than 50 different pieces of software, mostly business-oriented information systems, majority of them successfully implemented. But I still don’t know shit about programming, about different environments and technologies, methodologies and other technical fetishes. And I’ve accepted the fact this situation won’t change any time soon, if ever.
The first screen of the Treasure Island Dizzy, ZX Spectrum version (Photo credit: Wikipedia)
When I see people like Swizec doing their touring machines in hundred-something bytes in JavaScript, I sadly have to admit that’s something I’m not able to reproduce. While I am truly impressed, I know it would take me ages to understand what’s happening, so I don’t even try. Too hard core for me. In my world, the essence lies somewhere else.
I see technology mostly irrelevant. It’s just the means that helps you achieve something bigger. Transformation of real-life processes into abstract information processes, developing software thatdoes what it’s supposed to do. That’s the goal, and it doesn’t really matter how yo get to it. In this context, I see the database as the most important part of every information system. If the data model is done properly, with integrities and proper relationships, the code doesn’t stand a chance against it. It can’t fuck it up, it can’t do much more or less than what it’s supposed to do. It just makes it awesome or not.
But even if I see myself as a system (database) architect, that doesn’t mean I don’t like to play. I’ve developed a few interesting prototypes in the past years, but they were all on a “higher” level. Like having an abstract super-entity that all other objects are derived from. Or having a meta-engine that serves complex applications in a similar fashion than a CMS serves a website. Or having everything in a system defined by EntityID – InstanceID (eg. User – Grega or Company – Neolab), allowing secondary requirements, such as attaching files or adding notes, trivial. Or having amini-feed similar to Facebook’s to document every action in a system.
But this are all abstract things, that don’t have that much to do with programming, even if they are coded. They are prototypes, serving as proof of concept, that are written poorly, and then usually rewritten by someone else who knows what he/she is doing, and from that point on, I mostly don’t understand how they are working anymore. And I really don’t mind. I try to discover the boundaries of what can be done, and since technology I work with supports that, I don’t even try to learn new ones or study the advanced details of those I use (asp.net (c#) + MS SQL, if you’re interested).
I think the real magic happens when profiles like Swizec and I get together (we haven’t yet, though). That’s when the synergies kick in. And it’s situations like these that make me perfectly aware that I don’t need to know much about programming to code well. And I can freely say that I will always suck at programming, and it’s ok. Black boxes happen all the time, so I have to let go to go further. Writing my crappy code that does awesome shit even though you would probably laugh at it.
Grega Stritar is an entrepreneur, software architect, blogger and geek based in Ljubljana, Slovenia. He shares his thoughts on stritar.net.
Pingback: Блог команды FlexyMind о разработке игр, мобильных платформах и т. д.