Following up from part 1, here are lessons 6 through 8. I didn’t quite make it to 10, but at the end, you’ll see what I want to do about that.
Lesson #6: Remember everything else!
…working at an MMORPG is not only programming. You also have to do some advertising, public relations, dealing with the players, read/moderator the forums and the game, and many other things you don’t even think of before you actually have a working game full of players.
This is particularly nerve-wracking to me, because I’m going to need extra help once we launch. So I’ll be calling on responsible, reliable, and enthusiastic community members to help me with as much as they want.
Lesson #7: Do backups!
The good news was that he replaced that machine with a better one (Dual P3, 1.1GHZ). The bad news was that we had no backups whatsoever. So all the player files, the guild files, and pretty much all the dynamic data, plus the logs and ban list was gone.
And not just the game…
One nice July day, while I was at work, a script kiddie used an exploit to gain moderator access, and deleted most of our forums.
I’ve learned this one from running and building websites. Still, you can never assume anything about your hosting with regard to backups. You should do manual backups, too!
Lesson #8: When change comes, so will complaints
…usually whenever we change how something works, there are at least a few players complaining, threatening to leave the game, etc. In fact, whenever we are making something that was previously simple more complex, a lot of players take it almost as a personal insult.
I was thinking that the chance is pretty reasonable, and was hoping the players will enjoy this change, as it was designed to add some spice into a boring activity while reducing macroing as well. However, the intensity of the complaints was overwhelming. Someone went so far as removing one of our download mirrors. But on top of that, people were raging on our forums, threatening to quit, saying how much the game sucks now, and so on and so forth. Even players that were previously on our side whenever we made an unpopular change started to complain to me, saying they will quit playing if we don’t change it back.
Corroborated by Brian Green (via Raph Koster’s blog):
The basic player demand is, “Improve the game, but don’t change anything.”
This is pretty scary to me, because I know there will be large adjustments made to Lila Dreams as we add new features and really “figure out” the gameplay (because you never know until you have a full set of players in there if something is really going to work in the long term).
It’s made a little more drastic by the fact that the launch version of the game is not the whole game. But in order to launch something that is “whole enough” and fun, I will have to make concessions to design with the expectation that there will be possibly large changes after that to accommodate the post-launch features. I aim to minimize this, but it’s not an easy thing to do.
Lessons 9 and 10
You have a new quest! 
Since I really enjoy blog interaction, I figured we could continue this in comments. As players (and developers?), what are your #9 and #10 lessons?
What isn’t handled well or could be handled better in MMOs you play (or played) that we could learn from? What things have gone right that we could learn from?
Any takers?
I ran across a really interesting article from 2005: http://www.devmaster.net/articles/mmorpg-postmortem/part1.php. It’s a unique look into how things work behind the scenes of an indie/freeware MMO.
What’s more interesting is I could compare my plans with the author’s experience, and determine if I’m doing anything right. Luckily, most of what he covered I had accounted for. But I thought it is worth highlighting here, and maybe commenting with regard to where Lila Dreams’ development is at this very moment.
So this is part 1, the first five lessons.
* * *
Lesson #1: Nobody reads
…before adding the combat, even though we had a big, red bold text on the download page stating that the game has no combat, people were still downloading, and their fist question was, invariably, “How do I kill something?”
You can always be surprised by how true this is. I knew this from other game development experiences, but it’s easy to forget.
Lesson #2: All or nothing
Eternal Lands isn’t an AAA game, but most of the people expect the same quality from any game they download, even if it’s free or beta. Not surprisingly, 99% of the people that downloaded the game didn’t stay for more than, at most, one hour.
Corroborated by Nerfbat.
A game is only as strong as its weakest feature. Games are more often judged by their weaknesses than their strengths, just like anything else. Any incomplete feature or complete but crappy feature will leave a bad taste in players’ mouths.
If a feature isn’t up to snuff, leave it out until it’s really ready for the public. Of course, this means you can’t promise too much before you launch if you can’t get all those cool features implemented with sufficient quality, or players will be disappointed! 
I may have already crossed that line (that’s why I clammed up when I realized development would be delayed by the recent change in staff). But the silver lining is that this is an online game, so we can have constant updates and new features after launch. I hope that takes out some of the sting.
Lesson #3: People will cheat
Even in a freeware game. In any game. Even when it’s kinda pointless. I don’t get it, but we have to take measures against it anyway.
Some resources were placed in very convenient places; for example, there were some flowers right outside of the flower shops so a player could easily make a lot of money by harvesting flowers and selling them 10 meters away in the shop. Macroing is a common cheating method that automates a player’s actions by external macros that simulate specific actions.
Lesson #4: Always make tutorials optional
…a lot of the beginners didn’t feel like doing the tutorials either. They just wanted to “kill stuff”. Consequently, many weren’t able to leave the newbie island, because they didn’t read the NPC texts. In order to address this problem I had to implement a text command to skip the tutorial. Even though the first NPC did say that if you want to skip the tutorial just type #skip, this didn’t really help. For some reason, some people can’t even read small amounts of text. So, in a way, the newbie island was defeating it’s own purpose.
This is kind of a corollary to #1, and like #1 you can never get used to how much people will ignore your work on tutorials.
Lesson #5: Always have tutorials
I decided to just remove the tutorials, and let the beginners start on a larger island, which had “stuff to kill” (as they would expect) and didn’t require any tutorials to leave it.
Nevertheless, many people were now complaining that there is no tutorial… so I implemented a different tutorial on the starting island, which enables (but doesn’t require) you to get more familiar with the game and even earn you a few prizes. This worked much better than the first type of tutorial.
Of course there will be tutorials in Lila Dreams. This game is a lot different from a conventional MMO, and I view that as a liability as much as a feature. So, I’ll try to compensate by having plentiful, repeatable (but optional!) tutorials to ease players into things.
This is harder than it seems, because consider that the control scheme itself is even a little unconventional (although we’re trying to keep it as intuitive as possible). If a player can’t easily grasp the controls quickly, that’s a big deal. But I think we can make it fairly painless, and the control scheme is not crazy, it’s just not “click here and watch your avatar move to that spot.”
* * *
Next time, I’ll talk about the other five lessons.