[02:45:30] bblack: Thanks! [03:27:58] I have a question: Does the software always trust the X-Forwarded-For header? [03:29:04] By "trusting" I mean using its value as the editor's IP address in place of the origin IP. [04:04:33] Zhaofeng_Li: no, only whitelisted trusted ones: https://meta.wikimedia.org/wiki/XFF_project [04:05:05] legoktm: Oh, that's good to hear. [12:22:13] https://codex.wordpress.org/Using_Your_Browser_to_Diagnose_JavaScript_Errors is great, at last we can have a guide on how to report javaScript bugs [13:21:33] Zhaofeng_Li, i wanted to holla at ya a little bit about the possibility of converting all the junk citations over to real citations with {{cite}} templates. do you think it's possible for refill to at least have an optional mode that starts trying to guess how to parse those and convert them into variables? [13:22:52] Zhaofeng_Li, for example, could you automatically parse Chicago style or whatever other typical parameters? and if not, then that fact alone just should seal their coffins [13:23:34] i'm sure you realize that there's a legion of idiots who are forcing us to have to eventually convert inestimable mountains of garbage data [13:24:25] dtm: I can try, but it could be hellishly buggy, as there are so many different formats in use. [13:25:12] "DVD Playback." Emedia Professional 12.7 (1999): 20. Business Source Complete. Web. 23 July 2013. <-- that should be parsed into ---> {{cite journal | magazine=Emedia Professional | volume=12 |issue=7 | date=1999 | page=20}} and strip out the "Business Source Complete". so ideally, refill would know the names of the sources to know whether they're a magazine or web site etc [13:25:30] Zhaofeng_Li, yes and buggy is okay because this is in the category of data forensics [13:26:03] so just giving us a start would mean that you've eliminated half the typing and AT LEAST given us the minimal variables for us to manually populate [13:26:07] or correct [13:26:35] is the source citation i just gave in any sort of .... *gasp*... standard? [13:26:41] is that Chicago MoS or something? [13:27:00] it's obviously garbage but is it at all standard? [13:27:36] is there a sequence of arguments at least? [13:28:19] I just took a look at some of the formats, and found many of them wrap the title in quotes. That's a start. [13:28:26] Zhaofeng_Li, speaking of MoS, you know that the stylization of "reFill" violates wikipedia's own MoS, right? :) [13:28:41] i mean we literally can't call it that. [13:28:55] Yeah, sometimes you break the rules for style :P [13:29:04] well.. no :) [13:29:18] And technically it isn't bound by the naming conventions of MoS. [13:29:24] how [13:29:42] oh... yeah, if we just dont talk about it. easy! [13:29:52] never name it on wikipedia. [13:30:21] User pages are not articles :P [13:30:26] uh huh. [13:31:03] And there is also https://en.wikipedia.org/wiki/reCAPTCHA [13:31:04] not a good idea! that's only just barely above the idea of literally having stolen the same name for a completely different product lol [13:31:12] ^_^ [13:32:15] WP:OTHERSTUFFEXISTS [13:32:27] the article you named is presently wrong. [13:33:05] fyi [13:33:33] if there are different MoSes out there that specify exact formats in exact orders, perhaps you could have a different checkbox or a pulldown for the user to select which parser to use [13:34:26] dtm: Yeah, perhaps. [13:35:22] Zhaofeng_Li, do you happen to know if that's the case? do those MoSes specify an order? yeah maybe they have the title= in quotes but .... i dunno. [13:36:57] * Zhaofeng_Li doesn't know, either [13:39:40] i think in a perfect world (one where we still have to fix infinite amounts of garbage citations but by using a perfect tool, lol), we would have a GUI for remapping. so it would have a bunch of little GUI target buttons for the standard options like title=, magazine=, work=, date=, etc and i could use the mouse to highlight a piece of source text and then drag it onto the drop target for a given option. so in that example, I'd highlight "1999" and dr [13:39:40] ag it onto the "date=" button and the tool would append "| date=1999" to the target string. [13:39:47] amirite [13:40:07] wouldn't that be some web 2.0 magic that NeXTStep would be proud of? [13:40:22] wouldn't that be the ultimate way to do it [13:40:46] dropzones, that's it [13:41:12] a source text selection helper and a bunch of optional dropzones [13:41:39] dtm: Neat idea. [13:41:49] i mean hey these are the forensic concepts that are used in OCR and whatever [13:41:54] the "better than nothing" [13:41:58] for mass automation [13:42:17] coz i'm serious, we will be revolutionizing the whole data store and drag the userbase kicking and screaming. [13:42:23] out of dead trees [13:42:41] what in the hell is wrong with anyone who cites internet content using print syntax? wtf [13:42:46] i see it constantly [13:43:12] [[Tech]]; 46.53.178.225; [none]; https://meta.wikimedia.org/w/index.php?diff=14982293&oldid=14928753&rcid=7100426 [13:44:10] [[Tech]]; Tegel; Reverted changes by [[Special:Contributions/46.53.178.225|46.53.178.225]] ([[User talk:46.53.178.225|talk]]) to last version by Stryn; https://meta.wikimedia.org/w/index.php?diff=14982305&oldid=14982293&rcid=7100428 [13:44:13] Zhaofeng_Li, the source parser could do its attempt to help identify source strings. it'd parse for whatever delimiters, like spaces and quotes, and parentheses, so that they could all be optionally preselected. so i dont have to select each one; i can just drag it. [13:44:46] Zhaofeng_Li, and that's if i already didn't like its opinion of its suggested output in the first place [13:45:12] i can decide whether the suggested output is already correct. or, if not, whether it'd be more work to correct that or to start over by drag n drop. [13:45:27] Zhaofeng_Li, would that be ridiculous to code? [13:45:54] it seems like you'd want someone who's really done drag n drop GUI design in javascript [13:46:15] or else your time would be a lot more useful elsewhere perhaps [13:47:19] Actually, I'm thinking about adding a "manual mode", just like the original Reflinks. [13:48:26] The tool processes the citations, and editors are allowed to edit the fixed references one by one. [14:26:14] Zhaofeng_Li, that sounds easier huh ;) [14:26:27] Zhaofeng_Li, did anything i said sound feasible? [14:26:35] i'd be glad to document it if that helps [15:07:00] dtm: I'll give it a try later. [15:39:43] Zhaofeng_Li, you are the man [17:13:28] Fun https://lists.wikimedia.org/pipermail/discovery/2015-December/000589.html [17:16:06] Wut? :o [17:16:27] * Niharika is not subscribed to Discovery list! \o/ [17:20:58] [[Tech]]; Username Violation; /* Please advice. Those shitty browsers ( Mozilla chrom and Explorer ) issuing message "Can not load page" when I trying to open WIKIPEDIA */ typos; https://meta.wikimedia.org/w/index.php?diff=14984857&oldid=14982305&rcid=7100966 [17:28:33] [[Tech]]; Wiki13; Reverted changes by [[Special:Contributions/Username Violation|Username Violation]] ([[User talk:Username Violation|talk]]) to last version by Tegel; https://meta.wikimedia.org/w/index.php?diff=14984946&oldid=14984857&rcid=7100973 [18:34:05] Nemo_bis, lol!!!