| 01:00.09 | *** join/#bzflag BulletCatcher (~bc@bzflag/developer/BulletCatcher) |
| 01:04.41 | blast007 | Think I just stumbled upon a way to improve the server list sorting to give more servers a chance at activity. We've discussed previously about making observers not count towards the sorting, and I started with that. |
| 01:05.42 | blast007 | But there were still a lot of servers that start with a number that show up after ones with players, so poor xpert.jumpingcrab.com:5154 is gonna show up at the end of the list all the time |
| 01:07.39 | blast007 | We had normal players weighted to 1000 in our sorting with observers valued at 1 point. So, in the constructor for ServerItem class (each server list line has a ServerItem) I initialized a random number from 0 to 500, and then use that value when calculating the sort factor. |
| 01:08.52 | blast007 | so now each server list access it randomizes the sort order while still maintaining the primary sort on player counts (and servers with the same number of players will also appear in a random order with one another) |
| 01:09.04 | blast007 | does that sound sane? |
| 01:10.49 | blast007 | I'd probably push it to 2.4 so we can include that in the next 2.4 release |
| 01:22.51 | allejo | MUFFIN BUTTON |
| 01:30.51 | *** join/#bzflag reece (uid35886@gateway/web/irccloud.com/x-ftridsajaajugevv) |
| 02:40.34 | *** join/#bzflag JeffM (~JeffM@cpe-23-240-144-5.socal.res.rr.com) |
| 02:42.01 | delusional | Last is a coveted spot. They WANT to be the last on the list. |
| 03:04.21 | *** join/#bzflag brad` (~brad@unaffiliated/brad) |
| 03:06.25 | *** join/#bzflag AAA_awright (~aaa@2001:41d0:1:d632::1) |
| 03:06.26 | *** join/#bzflag brad (~brad@unaffiliated/brad) |
| 03:14.58 | *** join/#bzflag TD-Linux (~thomas@66-188-206-105.dhcp.roch.mn.charter.com) |
| 03:15.08 | TD-Linux | oh my what could possibly be happening here http://a.pomf.se/neymou.png |
| 03:16.20 | Flash | better question? |
| 03:19.53 | TD-Linux | Flash, it was rhetorical. |
| 03:21.58 | allejo | And I gave you a rhetorical answer |
| 03:22.32 | blast007 | hehe |
| 05:24.44 | *** join/#bzflag TD-Linux (~thomas@about/essy/indecisive/TD-Linux) |
| 05:36.19 | *** join/#bzflag Cobra_Fast (~coprah@wtwrp.de) |
| 06:02.20 | *** join/#bzflag Gnurdux (~Michael@edgerton-seven-fourty.mit.edu) |
| 07:15.42 | *** join/#bzflag Snake12534 (~Snake1253@c-50-156-82-76.hsd1.ca.comcast.net) |
| 07:48.02 | *** join/#bzflag I_Died_Once (~I_Died_On@unaffiliated/idiedonce/x-1828535) |
| 10:47.45 | *** join/#bzflag catalyst (~catalyst@allegro/user/angelchild) |
| 11:47.28 | *** join/#bzflag _lmnop1115 (~lmnop@91-65-161-106-dynip.superkabel.de) |
| 12:01.11 | *** join/#bzflag ruskie (ruskie@sourcemage/mage/ruskie) |
| 12:05.40 | *** join/#bzflag ruskie (ruskie@sourcemage/mage/ruskie) |
| 12:08.15 | *** join/#bzflag reece (uid35921@gateway/web/irccloud.com/x-lsaqpkkdccmcbujr) |
| 12:14.21 | *** join/#bzflag Nitroxis (~n@wtwrp.de) |
| 12:29.22 | *** join/#bzflag thot (~t@wtwrp.de) |
| 13:02.35 | *** join/#bzflag I_Died_Once (~I_Died_On@unaffiliated/idiedonce/x-1828535) |
| 15:48.24 | *** join/#bzflag _lmnop1115 (~lmnop@91-65-161-106-dynip.superkabel.de) |
| 17:38.54 | *** join/#bzflag JeffM (~JeffM@cpe-23-240-144-5.socal.res.rr.com) |
| 18:17.10 | *** join/#bzflag catalyst (~catalyst@allegro/user/angelchild) |
| 18:51.17 | *** join/#bzflag JeffM (~JeffM@cpe-23-240-144-5.socal.res.rr.com) |
| 18:55.42 | blast007 | would like some feedback before I go ahead with this change to the server list sorting: http://static.bzexcess.com/server_list_sorting.txt |
| 18:56.10 | blast007 | (and to clarify, it adds the random value to the sort factor) |
| 18:56.41 | blast007 | possible downsides is that it might make it more difficult to browse the list if you know what you're looking for |
| 19:01.55 | BulletCatcher | That change ought to make the server list more fair. |
| 19:01.56 | BulletCatcher | The search function is there for those who want a specific map. |
| 19:02.48 | BulletCatcher | Perhaps "favorite" maps could be given sort priority within a given occupancy level. |
| 19:03.32 | JeffM | do you think that taking the last join/score time into account would help? put servers with actual activity at the top |
| 19:03.41 | JeffM | or is that too much a popularity contest? |
| 19:05.06 | blast007 | I think that again would just help bury some servers |
| 19:05.12 | JeffM | probably |
| 19:05.24 | JeffM | adding some random weights would not be bad |
| 19:05.30 | BulletCatcher | It depends on whether our goal is to make it easy to find the maps people want, or to make it more fair for all. |
| 19:05.44 | BulletCatcher | ... for all servers. |
| 19:05.56 | JeffM | different people want ether of those things |
| 19:05.59 | blast007 | until we can sort by ping times, I think this might be an option |
| 19:06.18 | JeffM | finding maps people want means searching and we can improve our searching tool (like let them search by map name) |
| 19:06.20 | blast007 | I just don't like it sorting alphabetically by hostname ;) |
| 19:06.46 | JeffM | some people don't care about a specific map and want a sugestion, that's where the random part can come in |
| 19:06.51 | BulletCatcher | Searching already looks at both map title and host name. |
| 19:07.11 | JeffM | no it looks at the description, it is up to the server owner to put the map name in there |
| 19:07.23 | JeffM | we have no concept of "map name" |
| 19:07.29 | blast007 | more metadata inside the map file's 'world' section |
| 19:07.33 | JeffM | yeah |
| 19:07.41 | blast007 | map name, author, etc |
| 19:07.42 | BulletCatcher | That's what I meant when I said "title". |
| 19:07.54 | JeffM | I've seen people call the same map diferent names :) |
| 19:08.08 | JeffM | how about using the random thing to show 3 "join now" servers |
| 19:08.09 | BulletCatcher | Keywords are something for the future. |
| 19:08.14 | JeffM | and the rest of the list is the same |
| 19:08.42 | JeffM | like other games have "instant action" |
| 19:09.38 | JeffM | if you want to search you search, if you want to be suprized, instant it. If you want to play somewhere you played before, we should have a list of recent servers and favoriates |
| 19:09.40 | BulletCatcher | There isn't much action when you're the first/only player on a map. :-) |
| 19:10.01 | JeffM | thats why I'd say it should randomize between servers with players |
| 19:10.07 | BulletCatcher | Got it. |
| 19:10.09 | JeffM | and if it doesn't find any, don't show that option |
| 19:10.39 | JeffM | you could always have the other side of the random "Random ghost town" :) |
| 19:11.40 | JeffM | but I think both of you are right, there are people who want to use the list in ether way |
| 19:12.11 | JeffM | the main list should be sortable by a number of factors |
| 19:12.26 | JeffM | but we should also make it easy to jump in if you just want to play any game |
| 19:12.51 | JeffM | new players favor the later, with older players the former |
| 19:14.22 | JeffM | "Free Online Multiplayer Action" is the first tagline that is shown on our website :) |
| 19:16.30 | JeffM | goes to find eggs and bacon |
| 19:19.04 | BulletCatcher | Would meeting this complex set of goals be any easier if we were to first bring in the 2008 GSoC enhanced server list? |
| 19:21.21 | blast007 | that's the eventual goal, yes |
| 19:25.38 | BulletCatcher | It seems like the enhanced server listing ought to be compatible with 2.4, but my memory of it is dim. |
| 19:26.24 | BulletCatcher | r18517 touches a lot of files. |
| 19:27.23 | BulletCatcher | Unnecessary file changes reverted in r18521. :-) |
| 19:40.38 | *** join/#bzflag apeman (~Victor@222-155-41-238.jetstream.xtra.co.nz) |
| 19:50.03 | *** part/#bzflag catalyst (~catalyst@allegro/user/angelchild) |
| 20:06.53 | *** join/#bzflag Pimpinella (~frank@gondolin.pimpi.org) |
| 20:19.27 | *** join/#bzflag cods (~cods@rsbac/developer/cods) |
| 20:29.09 | *** join/#bzflag JeffM (~JeffM@cpe-23-240-144-5.socal.res.rr.com) |
| 21:32.17 | allejo | there sohuld be a way to register help messages for custom slash commands |
| 21:32.19 | allejo | should* |
| 21:34.47 | *** join/#bzflag unclelightning (~chatzilla@pool-64-222-240-10.port.east.myfairpoint.net) |
| 21:44.25 | TD-Linux | if anyone is curious |
| 21:44.27 | TD-Linux | https://github.com/tdaede/blender-bzw |
| 21:45.47 | allejo | out of sheer curiousity, is line 106 supposed to have a path hard coded? |
| 21:46.39 | Flash | Supposed to? One would hope not |
| 21:46.58 | Flash | Of course, __main__ scripts are often test fixtures and don't matter |
| 21:47.01 | TD-Linux | allejo, that's only for testing. |
| 21:47.21 | allejo | ahh makes sense |
| 21:47.24 | allejo | is clueless about python |
| 21:47.42 | TD-Linux | I can make it just a relative path I suppose too. but in blender you'll get a pretty file save dialog. |
| 21:47.50 | Flash | but if you wanted to change it to 'os.path.join(sys.env['HOME'], 'wheel.txt')' it wouldn't be bad. |
| 21:48.10 | TD-Linux | what's amazing to me is that I got a decent object exporter in so few lines of code inside blender |
| 21:48.42 | TD-Linux | Flash, I'll actually just change it to "test.bzw" or something so it writes to the working directory of blender. it'll never be executed if the script is installed as an addon |
| 21:49.09 | allejo | I kinda thought it'd take more lines of code to achieve this |
| 21:49.18 | allejo | nice work :D |
| 21:49.51 | Flash | TD-Linux: that's yours? |
| 21:50.01 | TD-Linux | Flash, yeah just wrote it. |
| 21:50.35 | Flash | you might consider making export a method on the map object (assuming there is one) for better encapsulation. |
| 21:50.50 | TD-Linux | blender 2.6+ is very nice compared to older versions |
| 21:51.00 | TD-Linux | Flash, it's already namespaced by being in its own .py file |
| 21:51.40 | Flash | Encapsulation goes beyond namespacing. It has to do with separation of concerns, and limiting required knowledge of implementation. |
| 21:52.51 | TD-Linux | Flash, true. but those functions are pretty stateless (except for the file pointer I guess). |
| 21:53.31 | Flash | Stateless, yes, but they know a lot about the composition of "object" which would more appropriately be encapsulated to object's definition. |
| 21:53.45 | TD-Linux | anyway I'm planning on adding correct parenting transforms, then materials and textures at least yet |
| 21:54.10 | TD-Linux | blender has a new "custom properties" feature in 2.6+ that would actually allow blender to be a 100% feature complete bzw editor if I desired to go that far |
| 21:55.12 | Flash | I'd also encourage you to use print (in Python 2.x, see "from __future__ import print_function" to get the Python 3.0 print syntax) to have simpler syntax and (possibly) better file portability. |
| 21:56.16 | TD-Linux | Flash, blender is python 3 only |
| 21:56.22 | Flash | cool |
| 21:56.32 | Flash | so why f.write instead of print? |
| 21:56.56 | TD-Linux | because I want to write to a file? |
| 21:57.04 | Flash | print lets you do that |
| 21:57.12 | Flash | print(foo, file=f) |
| 21:57.27 | TD-Linux | that's way longer :P |
| 21:57.45 | Flash | but clearer, and more maintainable |
| 21:58.35 | Flash | here's an example that would be shorter: |
| 21:59.40 | Flash | print("vertex", " ".join([str(c) for c in v.co]), file=f) |
| 22:00.37 | Flash | granted, that isn't because of the print, but the list comprehension and the join |
| 22:01.11 | Flash | but it's more 'pythonic', and the next guy should understand it |
| 22:01.36 | Flash | or (I think this should work) |
| 22:01.49 | Flash | print("vertex", *v.co, file=f) |
| 22:03.36 | Flash | The explicit conversions to string are indicative of insufficient encapsulation/design. Note that print() will perform such conversions for you, because it's intrinsic in the requirements of printing. |
| 22:07.29 | *** join/#bzflag Snake12534 (~Snake1253@c-50-156-82-76.hsd1.ca.comcast.net) |
| 22:12.44 | TD-Linux | Flash, yeah I'll switch to list comprehensions at some point probably. I'm currently trying to figure out how to order transforms when using parenting in blender |
| 22:13.26 | Flash | btw, did you know about math.degrees instead of doing the conversion manually? |
| 22:19.04 | *** join/#bzflag plastix- (~plastix@107.170.226.100) |
| 22:19.45 | *** join/#bzflag rodgort (~rodgort@192.241.198.49) |
| 22:23.31 | *** join/#bzflag MonsieurApple (~MonsieurA@unaffiliated/mrapple) |
| 22:50.10 | TD-Linux | Flash, good to know :) |
| 22:50.23 | TD-Linux | Flash, are you interested in helping with this tool? |
