Archive for the 'omgbots!' Category

Working on it…

I know I’ve been out of touch lately, for a number of reasons (some of the pretty good ones, too *smirk*), but I was reading the daily Coding Horror blog entry this morning and it contained a little observation so obvious and powerful (yet frequently ignored) that I had to immediately act on it :

And yet over and over, I’m gathering stories that point to the fact that programmers do not want to write code out in the open. Programmers don’t want their peers to see mistakes or failures. They want to work privately, in a cave, then spring "perfect" code on their community, as if no mistakes had ever been made.

One of the unfortunate consequences of this trait, which I exhibit just as much as (and maybe more than) any other developer I’ve ever met, is that for very long periods of time it will appear that I am doing absolutely nothing with regard to C:SI, while the truth may be that I’m just in "dark mode".

While the aforementioned blog post has more relevance to professional software development than to in-world scripting, it at the very least demonstrates to me that I could do a much better job of communicating what I’m doing "behind the scenes" rather than giving the appearance of doing nothing.

So on that note :

I’ve been reviewing the comments on my using bots for testing post, and compiling those into a more formal (in the sense of ‘do these steps’) list of tests to be performed, as well as trying to understand the whole ‘what makes a balanced weapon’ subject, and am beginning the process of working on updating the bot code. 

This will likely be a long-term ongoing project, but if it’s at all interesting to people I could post about my progress here.  I’ve also been considering posting periodically about how I implement various things in the bot code in case anyone else is interested in the subject, but I’m not certain I want to do something like that until I get some issues worked out (which is, of course, the whole point of the subject mentioned in the first paragraph, lol).

In any case, I just wanted to let people know that I’m recovering just fine, and *am* working on stuff, even though it probably doesn’t look that way :)

Popularity: 54% [?]

OMGBOTS! - Using bots for unit testing

It’s been one of those days…  On top of still dealing with the stress of Albertsons.com losing my order and having all of my grocery money locked up and unavailable for a week, I’m still working on my RL project even though I’ve been at it since about 5:00 this morning.

And yet, a conversation I had with Ayame recently keeps coming back to me, even while I’m working.  She’s got some ideas and some projects she’s working on which utilize bots built on libsecondlife code which will, I believe, be a very nice benefit for the C:SI community.  I’m going to be assisting in any way that I can, and I’m quite excited about the idea.

Thinking about her ideas keeps bringing me around to the idea of how I could do so many things with bots to make my own life easier, unit tests being a prime example.

I’m still very much trying to come up with more objective methods to determine whether a weapon is balanced, and I can think of several ways that automated testing via a bot running a predefined script of actions can assist in that process.

Unfortunately, my schedule being what it is, I haven’t really had time to think on this too deeply.  The technical details of how to implement the tests are not the big challenge, really.  The big challenge, as I see it, is coming up with a valid and reasonably complete list of tests that should be performed for all C:SI weapons.

I will continue to think on this topic as time allows, and should have some time this weekend to really dig into the topic, but I wanted to post this now in order to elicit feedback from the community as well.

I know that many C:SI members are software devs, and even many who are not have repeatedly shown that they have a pretty solid understanding of ways to test the weapons for balance and "correctness", so I’m asking anyone and everyone who has ideas to go ahead and submit them.

So whatcha think, got some good ideas for ways to objectively test C:SI weapons for balance and correct function?  I’d love to hear your thoughts!

Popularity: 49% [?]

Clan Information

In an ongoing attempt to make the Combat: Samurai Island system more comprehensive, compelling, and open we have started working on a few projects that require, among other things, the ability to programmatically discover who is in each official C:SI clan.

Unfortunately, there is no consistency on how the various clan groups are set up in Second Life®, with the result that as far as I can tell, a few clans are so small that they are in danger of being disbanded :

  1. Crescent Moon - 34 members
  2. Shadow Samurai - 20 members
  3. Ako Roshi - 35 members
  4. Kyousei Clan and Dojo - 5 members
  5. Vanguard Samurai - 63 members
  6. Order of the Fox - 3 members
  7. Yakuza Samurai School - 4 members
  8. Tendai-shu - 72 members
  9. Ancients of Eternity - 4 members
  10. Samurai Giri - 25 members
  11. The Oda Clan - 15 members
  12. Samurai of Daoism - 453 members

Now, I know that some of these numbers are incorrect, most likely because the people who set up the SL™ Groups have the "Members are visible" checkbox unchecked:

MembersVisible

For now we’ll use what information we have, but in the future it’s possible that there will be problems caused by not being able to know who’s in which clans, so I’d like to personally ask clan leaders to make clan members visible.

Popularity: 29% [?]

Gutsy Gibbon Broke My Bots

Damnit.  I may be the only person I know that has trouble EVERY SINGLE TIME they update an existing Ubuntu installation.  Earlier this week I did a dist-upgrade on the machine that I’ve been using to host my bots, naively believing all of the blogs I’d read gushing about how this new version was the "best evar" and how the update was so painless.

You’d think I would have learned by now.  While this update didn’t completely hose my machine like previous updates have, it take a great deal of time and in the end left me with no way to run my bots from that machine.  I don’t know at this point if the problem lies in the version of Mono (or if the Mono version was even different) or whether it’s a more fundamental problem, but this version has a SIGABRT from an issue somewhere in the appearance code.

The bots were working before I did the dist-upgrade, and were not working after, but I suppose it’s also remotely possible that the problem is related to Second Life® rather than Ubuntu.  If that’s the case, I should be able to get the latest version of libsecondlife and ensure that the bots are using the latest SL™ protocols, but I REALLY hate to do that because the libsl guys have no idea of what it means to have a stable interface; They change the public API of the libsl code on a daily basis, and this makes it incredibly difficult and time-consuming for anyone to maintain a non-trivial project dependant on libsecondlife. 

The libsl team have done a pretty fantastic job of putting together a codebase that you can use to build custom viewers, bots, SL™ <-> RL integration, etc, and are to be commended for all of their hard work, but I often get the feeling that the majority of contributors have not spent a lot of time deep in the trenches of a large commercial software project where different teams maintain different subsystems and critical libraries.

Okay, enough ranting for now.  I had hoped to have the bots ready for public usage at the same time as the new C:SI update, and have been testing them in secret for the last several weeks on a few of the sims where my wife and I have large chunks of land, but this setback will most likely mean that the bots will be delayed.

Popularity: 27% [?]



Bad Behavior has blocked 158 access attempts in the last 7 days.

Close
E-mail It