|I'm surrounded by idiots...|
I'm surrounded by idiots...
Nov. 10th, 2004 @ 11:54 am
When you're a small company, or a small group, you're under the radar. You can do things that you wouldn't be able to do if you were a larger organization. You could set up servers randomly, for no apparent reason. You can do deployments without official standards because, well, you're the only one doing it and what you're doing works for you. You can have single servers hosting things because, well, it's not the end of the world if it goes down and hey, I can fix it.
At some point though, as you succeed in that strategy, you need to change. You need to do things differently because suddenly a lot of people rely on the services you offer, you have more people doing things that can't read your mind, and you need to think about putting in decent architectures so that as things grow you need to do less rework with the older stuff.
Fundamentally, "Architecture" is about planning ahead as best you can for things that will change and making decisions in the original work that allow them to change with as little rework as possible, and even putting things in place that allow things you don't expect to change to flex in response to unpredictable events. It's about investing a dime to prevent yourself from having to part with a dollar later.
Right now I'm fighting with people who have owned everything in their world. They have single points of failure, lack of structure, and a serious desire to not change the things they've done so that they can grow into services everyone can use. In my opinion, they've used their freedom to ignore architecture and planning... to allow things to go forward because it was the path of least resistance. Well, now it's time to pay the piper, and they don't want to.
Fucking idiots. Do it right the first time, and you won't go through this crap again, will you.
|Date:||November 10th, 2004 01:12 pm (UTC)|| |
Wow. You've just eloquently described my entire frustration with work. Being constantly under time pressue with not enough resources, there is never any thought to planning or architecture. Thus, everything is a hack upon a hack upon a hack. Sure, we're the first to get anything working, but in the long run, we are constantly plagued by stuff not working and countless wasted hours trying to make simple things work. Oh, the only way to do that is to stand on your head and reboot while singing I'm a little teapot.
Heh. We had 3.5 hours of enterprise-wide downtime this morning. I sat on a conference call while everyone dithered and whined and played the blame game. I ran traces and pings and figured out the problem was a single route on the network through which everything flows -- A SINGLE POINT OF FAILURE with a single router handling it. Fools.
Ugh. My boss came and put his hand on my shoulder and told me to shut up because I was ranting. It took two minutes to fix when the guy who wasn't supposed to be on the call told them to reboot a router.
I feel your pain, and it's not my role to fix it here. Were it my role, I would likely end up with dead people in my wake.
|Date:||November 10th, 2004 09:39 pm (UTC)|| |
In another life . . .
I was a QA engineer for Texas Instruments for 3 1/2 years before law school. After four years in the Strategic Air Command I was checklist oriented. We had to have a plan. The optics division at TI was a small, but profitable little enterprise. Some science, some physics, and some black magic. As our volume grew we had to start writing procedures and then getting them by a customer / government audit.
The worst program was something called IRADS that went into the F-117 stealth fighter. Those guys didn't want any rules at all and hid behind their BLACK PROGRAM status (Oooh!). I didn't have the right industrial clearance to do an audit - I was only confidential at TI at that time.
But I had a Top Secret with the Air Force through my reserve activities. Came back one day with a bright, shiny SECRET level badge courtesy of two phone calls and a helpful secretary (bless you, Theresa).
An hour later some guys were explaining things, while standing on their heads, while singing I'm a Little Teapot.
|Date:||November 11th, 2004 02:06 am (UTC)|| |
well, sometimes people can't do it right the first time because they don't know any better.
sometimes people can't do it right the 2nd time because they're just doing the opposite of what didn't work the first time.
the third time's a charm, however.
|Top of Page
||Powered by LiveJournal.com|