Completely Unrealistic Estimates

I have referred to “changes that would just take 5 minutes to make” before. There must indeed be cases where the fix really is that simple: I remember one case where the problem was a misplaced decimal point. Most of the time, however, the players saying that have no idea what they are talking about. And the developers want to hit them. If the fix really were that simple, it probably would have been done by now, unless the suggestion is also some stupid balance- or game-breaking idea. No, fixing the AI is not just a matter of telling it not to jump off the cliff. That is a complex behavior to teach it, and teaching cliff-avoidance might break half a dozen other things.

Do feel free to continue browbeating developers for having hard-to-modify spaghetti code. Documentation may seem optional when you need to release in six weeks, but six months after release you may not still have the guy who remembers what half the variables mean. The more flexibility you can build in at the start, the better your reaction time will be later.

: Zubon

4 thoughts on “Completely Unrealistic Estimates”

  1. Back in my EQ days, I loved to quest. I was the Questaholic. Due to playing on Test, I had more than average access to the devs, and I used to give them slack about fixing things that “should be easy!” One time I remember asking one of the devs once if I could help them fix all the broken, or dead end quests. After all, I knew where they dead ended…it’s be easy, right? No, because it was a 5 year old database, that had so many cross references, and the person who built it left years ago, and it’s had so many other people come and go, leaving misc ties to this and that in it that they’d basically given up on even trying to figure out what goes where. The time committment was just too much.

    Another example I remember testing part of the Paladin 1.5 epic quest. It requires you to solo a mob in one of the deserts (Oasis? NRo? It’s been years). My job was to find all the exploits. And I did…there were tons of them. One of the most annoying things, for the dev, was to contain the quest mob. Apparently EQ relies on X,Y, Z coords so much that he couldn’t simply mark a spot and say “when the mob gets here, it summons you back” or whatever. No, he had to physically go to that spot, then get the coords, then *type* them into another system to code that in, and repeat for the rest of the line. Hours it took. I truly felt sorry for him, and the rest of them. Nothing but respect for devs, honestly. Even though I reserve my right to give em flak =)

  2. I would give them slack for EQ1 code seeing as it was really one of the first major 3D MMOs. Nowadays, developers should know better, that MMO worlds need to have massive developing tools supporting them. They need to be able to change the entire scope of the world and they need to be accessible from at least an intermediate programmer’s level. I’m not saying the tools should be simple to use, because that would most likely destroy the ability to change things on a minute scale, but they DO need to be reasonably easily modifiable.

    That being said, I’d take suggestions from a general playerbase the same day I’d take suggestions from a talking squirrel. Dedicated testers are REQUIRED nowadays. The only time you break the developer/player wall is when the entire playerbase is rioting (ala SWG incident).

  3. Spaghetti Code FTW!

    Think of it as protection against corporate espionage…

  4. massive developing tools
    I’d like that even in IT.
    There’s plenty of IDEs, but whenever I change jobs I have to learn new ones even for the same languages, and most of them suck anyway.

    It’s even more fun when Marketing or Sales says “I told them it was easy and you could do it, it sure looks easy, just one little change, right? Oh we need this week or we’ll lose the sale.” :)

Comments are closed.