[00:19:50] I always think of the telepathic gorilla #dirtyhippie [01:37:32] gn8 folks [03:19:48] Anyone around wanna help me reproduce a bug? [03:21:08] You need to be autoconfirmed on en.wiktionary . It won't take long. [03:24:23] would being a steward screw things up? [03:25:11] jdelanoy: I think not [03:25:18] ok [03:25:18] No, it wouldn't. [03:25:21] what's up [03:25:47] ok, one sec: I'm gonna unprotect my user page, delete my talkpage, then ask you to move my userpage to my talkpage and see if I get notiication [03:25:50] hang on [03:27:17] jdelanoy: ok can you more [[wikt:en:user:msh210]] to [[wikt:en:user talk:msh210]] please? [03:27:32] wait, can even autoconfirmed users do that? [03:27:35] s/more/move/ [03:27:40] I think they can [03:27:51] listgrouprights says so [03:28:18] moverootuserpage is separate from movepage iirc [03:28:42] ok [03:29:05] do you want me to leave a redirect? [03:29:06] ok [03:29:07] thank you [03:29:18] msh210: ok, I moved it [03:29:22] thanks I'll undo it. [03:29:28] ok [03:29:33] i did not get notification "you have new messages" fyi [03:30:00] ah [03:30:18] that's [[bugzilla:24937]] [03:30:57] do you want me to comment there, or will you? [03:31:33] I'l. [03:31:38] k [03:34:13] er, sorry. I just called you 'he', jdelanoy, on bugzilla, without actually knowing if you're a 'he'. [03:34:40] I am [03:34:48] but even if I was not, it would be fine :P [03:36:07] k. thanks again for your help. have a good timezone [03:36:22] heh [03:36:25] :) [06:23:15] new staffer? BJones (WMF) [06:23:36] sDrewthedoff: based on OTRS wiki, i would say yes [06:23:48] k, thanks [06:24:59] oh, and created by Philippe [06:25:13] that is a bit of a giveaway [07:01:27] sDrewthedoff: 22 07:00:18 < snitch> [[Template:Staff and contractors]]; Philippe (WMF); /* {{{Fundraiser|Fundraising}}} */ update; https://wikimediafoundation.org/w/index.php?diff=87777&oldid=87753 [07:04:02] goooood morning! [07:04:14] what's wrong with full text search (pages that contain...) on da.wiktionary? [07:05:13] I get no hits, even for very old content [07:07:19] same on sv.wikt and en.wikt, e.g. http://en.wiktionary.org/w/index.php?search=fireworks&title=Special%3ASearch&fulltext=1 [07:07:45] but it works just fine on en.wikipedia [07:09:51] the index might be refreshing [07:10:01] or the indexer on that cluster has crashed again [07:20:17] could that be a simple status check, that the search result page could inform the user of? [07:20:47] LA2: there are several bugs filed about it [07:20:51] now the search result page says all systems are okay, but the word was not found [07:21:24] LA2: did it consistently not return any result? and did you have to wait about 10 s or not [07:23:06] LA2: if (1) then https://bugzilla.wikimedia.org/show_bug.cgi?id=45073, if not-(1) and (2) then https://bugzilla.wikimedia.org/show_bug.cgi?id=42423 [07:25:21] you can click that link and see [07:25:29] ok, lol [07:30:07] I just restarted one of the search servers in that pool [07:30:18] in a few minutes it should clear up (for now) [07:32:12] LA2: that's why I didn't click; apergos is too quick at fixing stuff ;) so I wasn't sure it was still relevant [07:32:24] :-) [07:32:49] LA2: and your suggestion is filed as https://bugzilla.wikimedia.org/show_bug.cgi?id=43544 [07:38:13] good [07:38:35] now I get search results! [07:45:02] excellent [08:40:08] is it normal that every item I open at wikidata waits maaaaaaaaaaany seconds for bits (after meta and Commons)? [08:41:12] probably not, since most of bits is meant to be heavily cached [08:41:34] which browser are you using? it's developer console(/s) may be able to give you a few hints [09:04:04] p858snake: firefox [09:06:10] I havn't used firefox as a main browser for ages, I believe from memory you need to download firebug and it can graph resource calls and all that on loads [09:06:25] no downloads required [09:06:40] ctrl-shift-k (or cmd-shift-k) and then load the page agian [09:06:43] again* [09:06:51] that should tell you all you need to know [09:12:06] can anyone tell me why the edit pane keeps scrolling up when in code editor mode? [09:37:29] do you know of any widget that adds a comment on a talk page, without editing the whole section? [09:38:08] sorry, gadget [12:46:50] Hello people, I am currently importing a French Wikipédia dump into MySQL without having installed Wikimedia. Just to play with, no real usage. [12:49:42] (it takes a lot of room on the hard disk) [12:52:03] MathsPoetry: only the last revisions I suppose? [12:57:39] yes, only one revision per page - even though, the bzipped2 XML file is 4 gigs, and in mySQL it will probably be around 30 gigs [12:58:14] "even so" [13:06:12] * apergos lurks [13:12:41] so uh, what's the general rule on edits-per-second for bots these days? [13:14:10] https://www.mediawiki.org/wiki/API:Etiquette [13:14:22] checking [13:14:46] https://meta.wikimedia.org/wiki/User-Agent_policy [13:14:48] also doesn't say [13:15:11] i see a bot making up to five edits per second [13:15:12] sorry, I don't know, but if you find out, please add it to that first page [13:15:18] I'm also interested. [13:15:23] is that bad? [13:15:42] closedmouth- Last I've heard it's generally "Use maxlag and send your requests serially, not in parallel." Also, go slower if you are requesting old revisions or doing something else that's particularly hard on the servers. [13:18:03] Sometimes when I need to run on a lot of articles (like, all the articles of the French Wiktionary, i.e. >2,300,000 articles), I decrease the threshold to 2s. [13:18:34] Even so, it takes 2 months to run on all articles :P [13:29:45] hello, i deployed a fresh wikimedia instance on my server. How can i test mailing system, i created a test user and verification mail never came to his address, i can't find anything in maillog on recipient side so i presume it broke right on sending. Deployment is made by apt install, on debian 6 [13:30:18] BotaniCar|2: #mediawiki is probably where you want to go - this is for discussion of Wikipedia and other Wikimedia Foundation sites [13:30:31] sorry - the names are confusing [13:30:37] thank you [13:52:09] hello guys. What is the bandwidth-limit for rsync between virgina and amsterdam? (I don't want disturb anything) [13:53:16] DaBPunkt: apergos is on interrupts duty this week (per the /topic in #wikimedia-operations) so I hereby ping apergos to ask ^ [13:53:26] eh? [13:53:46] * apergos wonders what our network link is between the dcs [13:53:57] 10gb? lemme see if I can find out [13:54:05] oh amsterdam [13:54:13] gah no idea [13:54:28] any specific host? [13:55:07] bast1001 to the TS. I set 16MB/s limit at the moment [13:55:37] and I use ipv& [13:55:39] 6 [13:57:25] I guess we have a gigabit off that host but it's our bastion host, so we can't starve folsk out [13:58:46] the transfer will need ~120 minutes. I will be here if a problem occures [13:58:47] well hmm you could try upping it to 50 and see how that looks [13:59:27] we are mostly not doing bulk transfers over there, so you'll probably be ok (but sticking around is good) [14:00:42] DaBPunkt: whenever you have a question, you can check in #wikimedia-operations (the /topic) to see who's on interrupts rotation ("rt duty") and you can ping them and they'll answer your questions or get them answered -- you may have already known that but I just wanted to make sure [14:00:44] normaly I do these kind of transfers during my night with a low bandwidth-limit (2 or 3MBs). But the transfere tonight failed [14:01:06] sumanah: no, I didn't know. Thanks for the info [14:01:26] the person on rt duty might not be online [14:01:26] timezone [14:01:27] but otherwise yes [14:01:59] apergos: would you mind if I reminded wikitech-l/the toolserver list/etc. of the interrupts rotation? [14:02:14] no, why would I? [14:02:22] also if you happen to know who's going to be on next week we could add it to http://wikitech.wikimedia.org/view/Interrupts_Rotation [14:02:29] it's all the other people who will shortly be on rotation who will feel the crunch :-D [14:02:37] um I don't but lemme see if I can find it [14:03:03] apergos: I could speculate as to possible farfetched reasons why someone might mind, like if y'all were about to restructure it [14:03:17] no name yet [14:03:28] so you can't add them [14:03:32] are these things being decided week-by-week? [14:03:42] "I have time this week, sure"? [14:03:47] no but there just isn't annyone listed after me for whatever reaon [14:03:48] reason [14:07:47] apergos: ok, I figured out a reason why maybe I shouldn't remind public lists of the interrupts rotation; it was designed more for RT ticket followup than for general first-point-of-contact IRC interruptability (per engineering emails from Peter) [14:08:02] well that's true [14:17:17] * sumanah emails a question to Peter et al. to follow up. [14:31:16] closedmouth, what bot? [14:32:43] I'd say depends on its usefulness. if it's an antivandalism bot, it's probably ok to revert as fast as it can (well, if reverts are for a good reason). if it does some minor thing like general fixes, I'd say it should slow down [16:09:37] MaxSem: addbot [16:30:09] DaBPunkt: hey [16:30:16] DaBPunkt: still around ? [16:30:19] yes [16:30:31] DaBPunkt: so there's no actual connection between the two dc's, it goes out via the intertubes [16:30:38] that's why we have 2 as's [16:31:02] so it's just limited by the network interfaces at both sides [16:31:29] since bast1001 is shared, i'd probably limit it at 600 mbit (which with tcp and the latency you probably won't get anyways) [16:31:36] ok, tnx. The transfer is done BTW [16:31:41] oh haha [16:31:43] well for next time [16:32:23] 13.12MB/s was the average speed says rsync [16:36:11] sigh, why in the world does the network world use bit and the rest use byte [16:36:24] i blame everyone :) [16:36:27] but yay [16:37:45] LeslieCarr: thanks for weighingin, 600 mbit comes out to how many MB in the end? [16:37:55] 600/8 [16:38:02] 75mbyte [16:38:06] with some overhead [16:38:10] yeah [16:38:17] ok so maybe 50 wasn't too far off [16:38:40] excelt that he didn't get anything close to that anyhow [16:39:08] are you better btw? [16:39:18] (no I don't have any tasks to dump on you :-P) [16:39:20] ehe [16:39:27] not better :( [16:39:29] ugh [16:39:36] this is like the ultra mega cold from outer space [16:39:41] well, my fever is gone today [16:39:42] so that's good [16:39:49] yes, that's the first step [16:39:52] get some super soup [16:40:00] or ultra mega soup even [16:40:06] LeslieCarr: Get well soon [16:40:21] :) [16:40:29] i have a half order of veggie pho in the fridge [16:40:33] perilla makes decent veggie pho [16:40:43] i just need to have them send over more than 2 next time [16:41:01] kibibytes and mebibits :p [16:41:15] I swear I read that as "veggie php" and then tried to figure out what the heck that would look like [16:41:42] :) [16:42:01] mutante: Don't get me started. "byte" isn't a SI unit. :-) [16:42:19] Hell, byte isn't a unit full stop. [16:42:41] * Coren grumbles something about the young 'uns that keep saying byte when then mean octet [16:43:10] hehe “Kibibyte” is a multiple of the unit byte for quantities of digital information. [16:43:17] http://en.wikipedia.org/wiki/Kibibyte [16:45:04] mutante: Yeah, you'll not win me over with that. A "byte" isn't eight bits except by happenstance in the context of a specific CPU, and 'kibi' is a monstrosity invented by people who can't keep domain-specific terminology straight. :-P [16:45:16] LeslieCarr: what is "pho"? [16:45:40] it's yummy vietnamese soup [16:45:50] http://en.wikipedia.org/wiki/Pho [16:45:52] "Byte", for instance, is entirely meaningless over a network. There isn't a machine it could be the smallest addresable unit /of/. :-) [16:46:18] Coren: AFAIK a "byte" is always 8 bits. What you mean is a "Word" [16:46:40] * mutante leans back :) [16:47:01] DaBPunkt: An 'octet' is always 8 bits. [16:47:01] DaBPunkt is right [16:47:02] DaBPunkt: Nope. A byte is the smallest addressable unit. A word is the "natural" addressable unit. 8 bit is an octet; and I've worked on computers with 9 bit bytes and 36 bit bytes. [16:47:16] in the modern era, a byte is 8 bits [16:47:24] in other weirdness blecch [16:47:34] octets are so clearly 8 bits [16:47:39] LeslieCarr: Yes, modern architectures almost all have bytes that are octets. [16:47:40] :-P :-P [16:48:08] though now i am learning about the history of computing [16:48:17] When I get around to building a CPU for my new operating system, I'm definitely making it use non-octet bytes just to confuse people. [16:48:27] LeslieCarr: That doesn't make a byte meaningful in a context devoid of a CPU, nor its meaning equivalent with "8 bits". :-) [16:48:38] http://en.wikipedia.org/wiki/Octet_%28computing%29 [16:49:04] only when dealing with pedantic people [16:49:13] LeslieCarr: Why do you think RFCs always say "octet" when describing protocols? :-) [16:49:14] we work on wikipedia [16:49:20] we are the pedants pf pedantry [16:49:50] layer 8 ,eh 9, problem [16:49:55] haha [16:50:03] LeslieCarr: Since I have a computer, next to my desk, that is both operational /and/ doesn't have 8-bit bytes, it's hardly pedant. :-) [16:50:15] ah [16:50:19] what kind of computer ? [16:50:39] PDP-8 [16:50:41] and does it use punchcards ? [16:50:52] wow, still working ? [16:51:09] I worked on one system with 60-bit words [16:51:12] nice that they called it -8 :) [16:51:13] LeslieCarr: Last time I checked, though I haven't tried bootstrapping it in 3-4 years. [16:51:15] Now I know why WP sometimes so slow… ;) [16:51:15] oh the pdps [16:51:20] god bless em [16:52:01] LeslieCarr: It's memorabilia rather than working equipment, really. :-) [16:52:08] Coren: http://www.vandermark.ch/pdp8/local/emulator/PDP8.jnlp [16:52:59] http://www.pdp8online.com/ [16:52:59] hehe [16:55:25] wolfgang42: Oh, wow, he has a 8/e in INCREDIBLE shape! [16:55:57] Mine's the much cheaper 8/f, and in nowhere that nice conditions. I wonder if he redid the front paint to have it look this nice. [16:58:05] The grandaddies of the Internet. [16:58:15] * Coren still mourns Digital. [16:59:20] I'm over it [16:59:29] My fave box ever was my trusty AlphaStation. [17:00:06] I did not ever have a personal dec box [17:00:13] http://en.wikipedia.org/wiki/Commodore_128#CP.2FM_Mode [17:00:29] first personal system was a trash-80 [17:01:46] apergos: It was a Tsunami with two EV67. A powerhouse *monster* at the time I got in an auction from a business closed for not paying their taxes. [17:02:07] sweet deal [17:03:58] apergos: It remained my primary desktop well into the 2000s. [17:04:29] Paid $300 for it. It had to be worth at least 20 times that. [17:04:45] well I strted with trash [17:04:50] Nobody else at the auction bid on it because "it wasn't pc compatible". :-) [17:05:10] and the next system after that was an att 3b...1 I think (borrowed from the university) [17:05:13] horrible thing [17:05:28] meh, the 3b4 was okay, if you could survive AIX. :-) [17:05:35] couldn't [17:05:37] :-D [17:05:53] but it was still better than no system [17:06:27] My very first computer was a Southest S-09 done in wirewrap from bootleg plans. :-) [17:06:43] Southwest* [17:07:26] dunno that one [17:07:29] (It helps to have grown up with an EE uncle who adored me and was willing to bankroll my learning the trade) :-) [17:07:50] http://www.digibarn.com/collections/systems/swtpc-6800/index.html [17:08:25] It talks a bit about the S-09 there. It was a very nice 6809 box in the days of CP/M hegemony. [17:09:05] my very first computer was http://en.wikipedia.org/wiki/IBM_Personal_System/2 [17:09:26] it had just come out! [17:10:12] LeslieCarr: Newbie! :-) [17:10:29] trivia, which programming language is this? [17:10:30] hehe [17:10:32] fd – Move the turtle forward by steps. [17:10:39] Logo [17:10:43] oh i loved the turtle! [17:10:43] 10 points :) [17:11:32] And that one? [17:11:45] 10 0 DO I . LOOP CR [17:12:23] (Hiint: the 10 isn't a line number) [17:13:03] (Hint 2: the code outputs: "0 1 2 3 4 5 6 7 8 9" [17:13:10] BTTZ! [17:13:13] Forth! [17:13:24] heh, no idea. one more: [17:13:28] CAN HAS STDIO? [17:13:35] LOLCAT! [17:13:39] bingo:) [17:13:49] One more: [17:13:52] ++++++++++[>+++++++>++++++++++>+++>+<<<<-]>++.>+.+++++++..+++.>++.<<+++++++++++++++.>.+++.------.--------.>+.>. [17:13:56] Brainfuck! [17:14:02] Ding ding ding! [17:15:11] Then again, I'd expect everyone here knows of Brainfuck, befunge and Intercal. :-) [18:34:45] I'm having a bit of trouble connecting to en.wp over HTTPS, getting " The OCSP response is not yet valid (contains a date in the future). (Error code: sec_error_ocsp_future_response)" [18:34:49] * sumanah investigates [18:35:22] it's fine over HTTP [18:35:41] hmm, and over https when not logged in (in Epiphany) [18:37:14] ok, so HTTPS works in Epiphany but not Firefox. weird [18:39:14] There was an ocsp fix that just got pushed to ubuntu today... [18:45:31] can anyone replicate this? [18:47:09] sumanah: it sounds like it might be related to a wrong date on the local computer .. [18:47:16] ok [18:48:24] ok! yes. Fixed my time and it works now [18:48:27] glad I didn't file a bug [18:48:29] thank you [18:48:41] :) [20:03:56] at which time (PST/SF) are Tech talks on Thursdays normally? [20:06:36] andre__: I feel like it's 12:30pm but I could be wrong [20:07:00] I believe so [20:07:29] cannot find it in any Google calendar either, and not via Google. Hmm. [20:07:43] thanks [20:07:55] * andre__ trying to set up meetings in SF outside of the office :) [22:21:41] hi [22:22:04] where is a langs names for box what gives?