| 02:13.07 | *** join/#bzflag Patlabor221 (~JeffM@2605:e000:7d02:c700:f58f:d157:58f3:cc8c) |
| 03:17.08 | *** join/#bzflag smoooth (~smoooth@cpe-071-074-004-042.ec.res.rr.com) |
| 03:41.14 | *** join/#bzflag zlsa (~zlsa@static-205-157-147-198.ded.unwiredbb.net) |
| 05:44.55 | Flash | lurks impatiently |
| 05:49.30 | allejo | lurks disgruntledly |
| 06:05.00 | kierra | guilty guilty guilty!!!! |
| 06:05.08 | kierra | off with your heads! |
| 06:05.25 | DTRemenak | lurks gruntledly |
| 06:07.27 | kierra | hmmm, is that the smoooth who used to play in gu league? |
| 11:17.32 | *** join/#bzflag I_Died_Once (~I_Died_On@unaffiliated/idiedonce/x-1828535) |
| 11:47.27 | *** join/#bzflag I_Died_Once (~I_Died_On@unaffiliated/idiedonce/x-1828535) |
| 12:01.31 | *** join/#bzflag infobot (ibot@rikers.org) |
| 12:01.31 | *** topic/#bzflag is http://BZFlag.org || https://www.ohloh.net/p/bzflag || http://wiki.BZFlag.org/Getting_Help || Channel Logs: http://ibot.rikers.org/%23bzflag/ || 2.4.2 Current version http://goo.gl/PI9KFD |
| 12:01.32 | *** mode/#bzflag [+o infobot] by ChanServ |
| 14:20.25 | *** join/#bzflag zlsa (~zlsa@static-205-157-147-198.ded.unwiredbb.net) |
| 17:08.37 | *** join/#bzflag Patlabor221 (~JeffM@12.167.61.2) |
| 17:26.43 | *** join/#bzflag zuii (~akos@BC06BD4B.catv.pool.telekom.hu) |
| 17:49.33 | *** join/#bzflag DTRemenak|RDP (~DTRemenak@about/essy/CrazyCoder/DTRemenak) |
| 17:49.33 | *** mode/#bzflag [+v DTRemenak|RDP] by ChanServ |
| 17:56.19 | *** join/#bzflag gnu-sense (~gnu-sense@c-98-223-179-205.hsd1.in.comcast.net) |
| 18:02.39 | *** join/#bzflag WarPig_ (~j-mo@207.235.53.82) |
| 18:40.55 | *** join/#bzflag zlsa (~zlsa@static-205-157-147-198.ded.unwiredbb.net) |
| 18:50.32 | *** join/#bzflag zuii (~akos@BC06BD4B.catv.pool.telekom.hu) |
| 19:07.18 | *** join/#bzflag gnu-sense (~gnu-sense@c-98-223-179-205.hsd1.in.comcast.net) |
| 19:15.54 | *** join/#bzflag zlsa (~zlsa@static-205-157-147-198.ded.unwiredbb.net) |
| 19:27.23 | WarPig_ | bug: If output from /banlist or /viewreports list is too long, it borks up the client requiring reconnect. |
| 19:41.30 | gnu-sense | seems like these commands should paginate somehow, or be able to show recent bans, anyway |
| 19:41.56 | kierra | That happens on bztank servers also |
| 19:43.54 | *** join/#bzflag veronicas (81c20849@gateway/web/freenode/ip.129.194.8.73) |
| 19:59.45 | WarPig | In any case, it should display only as many, most recent, entries as it can without borking up the client. |
| 20:01.02 | veronicas | Hello, I am a student at University of Geneva doing a master thesis on open source video games. I'm trying to contact someone (a developer) who could answer to my short survey about localization techniques used, what language you use for the game, translation process, etc. HELP PLEASE! |
| 20:08.21 | kierra | Erotica |
| 20:08.30 | kierra | Oops |
| 20:08.37 | WarPig | :D |
| 20:08.43 | kierra | Veronica |
| 20:09.01 | kierra | I hate autocomplete |
| 20:09.36 | *** join/#bzflag zlsa (~zlsa@static-205-157-147-198.ded.unwiredbb.net) |
| 20:12.17 | kierra | :p WP |
| 20:13.48 | Patlabor221 | veronicas, what do you need to know? |
| 20:14.05 | Patlabor221 | WarPig, yes it has to do with buffering, it's a problem with messages that send large blocks of data |
| 20:14.23 | Patlabor221 | and now we know what kierra reads |
| 20:14.32 | Patlabor221 | how many shades are you up to? |
| 20:15.05 | kierra | is innocent |
| 20:15.08 | veronicas | hehehe |
| 20:16.23 | Patlabor221 | we just use a text file for string replacment for localization, it's very simple |
| 20:16.29 | veronicas | My survey is about localization of the game. I need someone to answer my survey (so that it becomes official information for my thesis). questions range from how translator are recruited to localization techniques and programming languages used. |
| 20:16.41 | Patlabor221 | we don't do any of those things |
| 20:16.57 | Patlabor221 | people submit translation files |
| 20:16.58 | Patlabor221 | that's it |
| 20:17.04 | veronicas | In that case I would need someone to write it in my survey. |
| 20:17.21 | Patlabor221 | what personal information is collected? |
| 20:17.22 | veronicas | are those translation files .po files? |
| 20:17.27 | Patlabor221 | yeah |
| 20:17.42 | veronicas | that's what interests me for my thesis! |
| 20:17.59 | Patlabor221 | it's just a text file with string replacements |
| 20:18.09 | Patlabor221 | if someone wants to trasnlate it, they do it and submit it as a patch |
| 20:18.18 | Patlabor221 | it is very informal |
| 20:18.30 | veronicas | about personal information, i ask where do you live and if you have had an academic course of l10n, but in case you don't want to answer you can write "--" and it's fine. |
| 20:18.55 | Patlabor221 | then I can answer your server, I am one of the project adminstrators for bzflag |
| 20:19.05 | veronicas | that information about age or location is not relevant for my study but it's there because it came with the model from university. |
| 20:19.30 | Patlabor221 | this is graduate or undergrad? |
| 20:19.57 | veronicas | graduate, i'm doing my master thesis (so that i can become a grad) :) |
| 20:20.07 | Patlabor221 | cool |
| 20:20.54 | veronicas | it' also cool to learn from scratch... i didn't know much of the open source community and the files used... and coding |
| 20:21.13 | Patlabor221 | different projects reuse different amounts of code |
| 20:21.26 | Patlabor221 | we just do a simple string lookup before display |
| 20:21.33 | veronicas | i know it stays very theoretical for me, but it's interesting. |
| 20:21.35 | Patlabor221 | it's been in the codebase for a very long time |
| 20:22.05 | blast007 | WarPig: the reason it borks the client is actually because it kicks the client because the send queue gets too large |
| 20:24.05 | Patlabor221 | buffers! |
| 20:24.49 | veronicas | aha! that's why they kick me out! |
| 20:24.52 | veronicas | :) |
| 20:25.04 | Patlabor221 | we do a very bad job of network buffer managment |
| 20:25.17 | Patlabor221 | it is very easy to flood out a client |
| 20:27.28 | blast007 | veronicas: also note that our .po files aren't necessarily proper gettext translation files. we have a custom parser and a format that resembles gettext files. |
| 20:28.24 | veronicas | ok, i'll write it down. :) |
| 20:29.58 | *** join/#bzflag Patlabor221 (~JeffM@12.167.61.2) |
| 20:33.17 | blast007 | Debian freeze today - next stable release will have 2.4.2. Time to release 2.6! ;) |
| 20:39.48 | WarPig | blast007: Awesome! Understanding the problem is key to FIXING IT. |
| 20:39.49 | WarPig | n:D |
| 20:40.38 | blast007 | ~n |
| 20:40.38 | infobot | i heard n is the disease bzflag players get when they play too much. There is no cure. |
| 20:41.11 | blast007 | 2.99.x may have already prevented that from happening |
| 20:42.43 | Patlabor221 | it did |
| 20:42.48 | Patlabor221 | except it didn't work :) |
| 20:43.08 | Patlabor221 | WarPig, both the client and server need to keep buffers of unprocessed/unsent messages |
| 20:43.37 | Patlabor221 | then you don't have to fit a message into a single packet |
| 20:44.15 | Patlabor221 | optimaly this buffer would live in a networking thread so it could be serviced outside of the game loop |
| 20:45.15 | Patlabor221 | and you'd have 2 buffers, pendingInbound, pendingOutbound |
| 20:45.20 | blast007 | message priorities might be useful as well |
| 20:46.55 | Patlabor221 | sure |
| 20:47.03 | Patlabor221 | that's why the real solution is to just swap to raknet |
| 20:47.05 | Patlabor221 | it does all that |
| 20:48.23 | kierra | ~Patlabor221++ |
| 20:48.40 | Patlabor221 | only sometimes kierra |
| 20:49.43 | kierra | Today you were (I overlooked the comment about my reading material) :) |
| 20:50.09 | Patlabor221 | there ain't nothing wrong with reading |
| 20:50.36 | Patlabor221 | well except maybe the series of horrible books that one was based on |
| 20:51.28 | Patlabor221 | blast007, that is if you are ok using facebook owned code ;) |
| 20:54.39 | blast007 | as long as it doesn't make use log in with Facebook Connect :) |
| 20:55.30 | blast007 | "Someone has been looking at your tank. Click here to see who!" |
| 20:56.00 | Patlabor221 | every flag is a selfie |
| 20:56.55 | Patlabor221 | but really RakNet is a very powerful and proven system, we should be able to drop it in with minimal changes. |
| 20:57.07 | Patlabor221 | the main thing it would do would be to break all external conneciton scripts |
| 20:57.08 | blast007 | Updated status: Just got genocide! |
| 20:57.20 | blast007 | 5 people liked this |
| 20:57.45 | Patlabor221 | so we'd have to change the list server, and strayer's stat site would stop updating |
| 20:57.46 | WarPig | :D |
| 20:57.51 | blast007 | and then what happens to our fun HTTP interface? |
| 20:57.59 | Patlabor221 | that can still exist |
| 20:58.07 | blast007 | oh, right, raknet is UDP only? |
| 20:58.13 | Patlabor221 | I'd have to convert it over to using it's own socket lib |
| 20:58.17 | Patlabor221 | but yeah |
| 20:58.23 | Patlabor221 | TCP would be for HTTP, UDP for game |
| 20:58.28 | Patlabor221 | goood clear seperation |
| 20:58.48 | Patlabor221 | we could take the current nethandler, streamline it and leave it for TCP |
| 20:59.15 | Patlabor221 | then for poll stats impelement a built in HTTP stat responce |
| 20:59.19 | Patlabor221 | in json or something |
| 20:59.21 | blast007 | then I guess here's another question... do we want to keep the HTTP server thing, or would we rather switch it to a local-only API that can be accessed by web apps running on proper web servers? |
| 20:59.38 | Patlabor221 | I don't know what people want |
| 20:59.50 | Patlabor221 | the local HTTP is very easy to set up |
| 20:59.57 | Patlabor221 | and it's nice that it runs on the same port as the game |
| 20:59.58 | blast007 | easy to attack too |
| 21:00.04 | Patlabor221 | yeah |
| 21:00.16 | Patlabor221 | and it makes combining multiple hosts harder |
| 21:00.22 | Patlabor221 | so it has pros and cons |
| 21:00.25 | blast007 | yep |
| 21:00.38 | blast007 | the main thing it has been used for thus far is fastmap |
| 21:00.46 | Patlabor221 | I'd ask you or someone like you to lay out what our "expected management infrastructure" would be like |
| 21:00.48 | blast007 | I don't know if any servers using other HTTP plugins |
| 21:00.53 | blast007 | know of* |
| 21:00.53 | Patlabor221 | then we can see what features best fit that goal |
| 21:01.07 | Patlabor221 | yeah and we can change how maps are sent by default |
| 21:01.13 | blast007 | yup |
| 21:01.32 | Patlabor221 | if we want to provide a php or python interface to the server and let management happen externaly that's fine |
| 21:01.54 | Patlabor221 | or maybe just json responces from localhost only |
| 21:02.07 | Patlabor221 | localhost/list of approved IPs |
| 21:02.17 | Patlabor221 | that all would need to be researched |
| 21:02.23 | blast007 | yep. |
| 21:02.40 | Patlabor221 | but whatever is decided, it's seperate from raknet |
| 21:02.45 | blast007 | it was a bit tricky with the more advanced HTTP plugins in 2.99.x since we had template files at whatnot |
| 21:02.51 | blast007 | and whatnot* |
| 21:03.12 | Patlabor221 | yeah, maybe it's best to just have those return the info that they filled into the templates as json |
| 21:03.22 | Patlabor221 | then someone build external sites to use that data |
| 21:03.33 | Patlabor221 | and move auth to IP range, not per user |
| 21:03.39 | Patlabor221 | then the external site can do it's own auth |
| 21:04.23 | blast007 | meaning what? |
| 21:04.24 | Patlabor221 | I kinda like the localhost/approved IP range |
| 21:04.49 | Patlabor221 | the actual info on the site, bans, chats, who's playing etc.. |
| 21:04.58 | Patlabor221 | return that via json |
| 21:05.04 | Patlabor221 | then a real site can format it |
| 21:05.08 | blast007 | "move auth to IP range, not per user" that, I mean |
| 21:05.17 | Patlabor221 | oh |
| 21:05.25 | Patlabor221 | make bzfs only accept TCP conenctions from localhost |
| 21:05.38 | Patlabor221 | and it takes a json request and gives back json responces |
| 21:05.46 | Patlabor221 | so we define an API there |
| 21:06.07 | Patlabor221 | then if you want web services, you run your own HTTP and write a site that hits bzfs for the json info |
| 21:06.19 | Patlabor221 | then bzfs doesn't have to authenticate the request |
| 21:06.26 | blast007 | for that, I was thinking of it just binding to 127.0.0.1 by default (and eventually ::1 for IPv6, but that's probably for later unless we think we could do that for 2.6) and optionally binding it to another address |
| 21:06.39 | Patlabor221 | yeah that's exactly what I was thinking |
| 21:06.42 | Patlabor221 | localhost by default |
| 21:06.49 | Patlabor221 | but you can provide a list of other acceptable IPs |
| 21:06.56 | Patlabor221 | in case your webserver is on another machine |
| 21:07.18 | Patlabor221 | but then bzfs is just an API host, not a user facing server |
| 21:07.18 | blast007 | if it was bound to 127.0.0.1, then it wouldn't respond on any other address, so they would have to re-bind it. |
| 21:07.25 | blast007 | yep |
| 21:07.37 | Patlabor221 | yeah that's the techincal way you'd limit it to that address ;) |
| 21:07.50 | blast007 | and firewalls could even be the limiter |
| 21:08.02 | Patlabor221 | part of the research |
| 21:08.07 | blast007 | yep |
| 21:08.19 | Patlabor221 | but that's the idea, limited scope connection for json API access |
| 21:09.34 | Patlabor221 | it would be nice if at the same time we did that change, we were able to povide a sample PHP site that did some of the webadmin stuff we want |
| 21:09.51 | Patlabor221 | so that it'd be easy for owner to hook up then customize if they wanted to |
| 21:10.07 | Patlabor221 | well not nessiarly php "Some accepted website language" |
| 22:53.24 | *** join/#bzflag zuii (~akos@BC06BD4B.catv.pool.telekom.hu) |