[00:31:21] halfak: I have a much better outline for https://docs.google.com/document/d/1g2_mHPg92CgK1iY3s1TYPVAI-ioiAc7AzudhMD_eLhw/edit# [00:31:26] Still working on it, of course... [00:43:22] o/ [08:43:10] 10Scoring-platform-team, 10Wikimedia-Site-requests, 10Community-consensus-needed: Enable RecentChanges patrolling on English Wikipedia - https://phabricator.wikimedia.org/T143791#3988340 (10Krinkle) >>! In T143791#2581143, @MusikAnimal wrote: > Is there a permission that will prevent a user's edits from bein... [10:20:18] (03CR) 10Zfilipin: Create Selenium UI tests for ORES damaging and good faith filters. (034 comments) [extensions/ORES] - 10https://gerrit.wikimedia.org/r/402863 (https://phabricator.wikimedia.org/T184451) (owner: 10Etonkovidova) [11:26:40] 10Scoring-platform-team, 10ORES, 10editquality-modeling, 10Collaboration-Team-Triage (Collab-Team-This-Quarter), and 4 others: Enable ORES filters for svwiki - https://phabricator.wikimedia.org/T174560#3988639 (10Lokal_Profil) @Halfak @Trizek-WMF @Catrope Are there any blockers for deploying this other tha... [11:44:10] (03PS1) 10Zfilipin: ESLint parserOptions is redundant when env is set to es6 [extensions/ORES] - 10https://gerrit.wikimedia.org/r/413144 (https://phabricator.wikimedia.org/T182421) [12:35:48] (03CR) 10WMDE-Fisch: [C: 032] ESLint parserOptions is redundant when env is set to es6 [extensions/ORES] - 10https://gerrit.wikimedia.org/r/413144 (https://phabricator.wikimedia.org/T182421) (owner: 10Zfilipin) [12:41:18] (03Merged) 10jenkins-bot: ESLint parserOptions is redundant when env is set to es6 [extensions/ORES] - 10https://gerrit.wikimedia.org/r/413144 (https://phabricator.wikimedia.org/T182421) (owner: 10Zfilipin) [12:42:34] (03CR) 10jenkins-bot: ESLint parserOptions is redundant when env is set to es6 [extensions/ORES] - 10https://gerrit.wikimedia.org/r/413144 (https://phabricator.wikimedia.org/T182421) (owner: 10Zfilipin) [13:12:00] 10Scoring-platform-team, 10Wikimedia-Site-requests, 10Community-consensus-needed: Enable RecentChanges patrolling on English Wikipedia - https://phabricator.wikimedia.org/T143791#3988982 (10Ladsgroup) Just wanted to add that old ORES filters even exclude patrolled edits but the new filters don't (which I tho... [13:34:41] hey there [14:14:58] halfak: Can you add me to the JADE meeting tonight? [14:15:28] awight, yes. Done. [14:15:32] :) [14:15:40] Was meaning to ask because it was outside of hours and forgot :| [14:15:52] I accidentally calendar-spotted [14:16:22] halfak: lmk when you want to chat about the JADE connector in ORES [14:19:37] Sure. Am still in breakfast mode ^_^ [14:19:49] Maybe in 45 mins so I can get through a round of emails? [14:21:10] Staff meeting at that time, maybe afterwards? [14:26:14] Amir1: Feel like some CR today? [14:35:30] Oh yeah. Good point. [14:35:36] Probably afterwards. [15:01:15] akosiaris: Does today mark the death of ORES-on-scb? [15:11:51] awight: unfortunately no. I got dragged into 3 different rabbit holes since this morning [15:12:01] no problem! [15:12:06] I 'd rather not do it now and postpone for tomorrow morning [15:12:20] EU morning that is [15:32:21] +1 I like it [15:56:31] halfak: have some time? txt / voice? [15:56:54] yes. Voice? [15:57:05] cool [15:57:32] Call when ready [16:43:19] 10Scoring-platform-team (Current), 10ORES: Deploy ORES mid-April - https://phabricator.wikimedia.org/T162892#3989746 (10Pchelolo) [16:43:25] 10Scoring-platform-team (Current), 10ChangeProp, 10ORES, 10Services (done): Change ORES rules to send all events to new "/precache" endpoint - https://phabricator.wikimedia.org/T158437#3989743 (10Pchelolo) 05Open>03Resolved The change has been deployed and seems to work fine. Resolving. [16:54:56] halfak: you don’t have to deal with this, but I’m on a whole new outline: https://docs.google.com/document/d/1g2_mHPg92CgK1iY3s1TYPVAI-ioiAc7AzudhMD_eLhw/edit# [16:55:33] awight, can you get this into a task and ping comms for a review/discussion about it? [16:55:44] They'll value your brief outline :) [16:55:50] hehe for sure [16:56:08] Feel free to drop any comments on the substance, as well. [16:56:20] Uh oh [16:56:35] precache deployment may have been a failure, or metrics changed in a way I wasn’t expecting [16:59:10] AIUI, volume should be the same. [17:09:40] 10Scoring-platform-team (Current), 10ORES: Deploy ORES mid-April - https://phabricator.wikimedia.org/T162892#3989891 (10awight) [17:09:45] 10Scoring-platform-team (Current), 10ChangeProp, 10ORES, 10Patch-For-Review, 10Services (done): Change ORES rules to send all events to new "/precache" endpoint - https://phabricator.wikimedia.org/T158437#3989889 (10awight) 05Resolved>03Open This was failing quietly, > [2018-02-21T17:04:21] [pid: 253... [17:10:00] Looks like it was real simple, GET vs POST [17:11:11] 10Scoring-platform-team (Current), 10ORES: Deploy ORES mid-April - https://phabricator.wikimedia.org/T162892#3178444 (10awight) [17:11:14] 10Scoring-platform-team (Current), 10ChangeProp, 10ORES, 10Patch-For-Review, 10Services (done): Change ORES rules to send all events to new "/precache" endpoint - https://phabricator.wikimedia.org/T158437#3989897 (10awight) [17:19:42] 10Scoring-platform-team (Current), 10ORES: New precache endpoint isn't reporting its metrics correctly - https://phabricator.wikimedia.org/T187914#3989969 (10awight) p:05Triage>03Normal [17:21:14] 10Scoring-platform-team (Current), 10ORES: New precache endpoint isn't reporting its metrics correctly - https://phabricator.wikimedia.org/T187914#3989989 (10awight) scoring_system.py, line 48: ``` if not request.precache: self.metrics_collector.scores_request(request, duration) els... [17:25:04] halfak: ^ [17:25:40] ^ ah OR [17:25:45] *PR [17:25:47] grr [17:26:57] 10Scoring-platform-team (Current), 10ORES: New precache endpoint isn't reporting its metrics correctly - https://phabricator.wikimedia.org/T187914#3990006 (10awight) https://github.com/wiki-ai/ores/pull/246 [17:27:11] 10Scoring-platform-team (Current), 10ORES: New precache endpoint isn't reporting its metrics correctly - https://phabricator.wikimedia.org/T187914#3989969 (10awight) a:03awight [17:28:04] merged [17:28:06] awight, ^ [17:28:48] I could wait 3.5hr for our deployment window... [17:29:09] wiki-ai/ores#926 (precache - 76816d9 : Adam Wight): The build passed. https://travis-ci.org/wiki-ai/ores/builds/344424602 [17:48:00] halfak: Check out Grafana for the last 3hr [17:48:34] The extra backlog for the span we missed should have been filled in within 15min, I think. [17:48:50] It is possible that the new precaching code causes 3x the number of scoring requests? [17:49:35] I also don’t understand why we were only seeing changeprop hitting eqiad, previously. [18:08:24] halfak: You said you had looked at rowiki and it had no ORES? I see the filters here, https://ro.wikipedia.org/wiki/Special:Schimbări_recente?hidebots=1&hidecategorization=1&hideWikibase=1&limit=50&days=7&damaging__likelygood_color=c2&urlversion=2&uselang=en [18:10:23] 10Scoring-platform-team (Current), 10Global-Collaboration, 10MediaWiki-extensions-ORES: [Epic] Audit of pending ORES GUI deployments - https://phabricator.wikimedia.org/T187836#3990181 (10awight) [18:10:25] 10Scoring-platform-team, 10MediaWiki-extensions-ORES: Deploy ORES Review Tool on Romanian Wikipedia - https://phabricator.wikimedia.org/T156518#3990178 (10awight) 05Open>03Resolved a:03awight Config exists and it behaves correctly, https://ro.wikipedia.org/wiki/Special:Schimbări_recente?hidebots=1&hideca... [18:10:54] 10Scoring-platform-team (Current), 10Global-Collaboration, 10MediaWiki-extensions-ORES: [Epic] Audit of pending ORES GUI deployments - https://phabricator.wikimedia.org/T187836#3987540 (10awight) [18:10:56] 10Scoring-platform-team, 10MediaWiki-extensions-ORES: Deploy ORES Review Tool for Albanian Wikipedia - https://phabricator.wikimedia.org/T163008#3990185 (10awight) 05Open>03Resolved a:03awight Confirmed working: https://sq.wikipedia.org/wiki/Speciale:NdryshimeSëFundmi?hidebots=1&hidecategorization=1&hide... [18:11:47] awight, you're right. Not sure how I missed that [18:13:08] I think it’s cos you checked during a meeting :p [18:13:24] Do give Grafana a look. [18:13:27] I’m concerned. [18:14:04] We might have to revert the change-prop rule hitting new precaching [18:14:30] Hey hey [18:15:07] :) [18:15:20] Okay, let's see what I need to work on today [18:15:49] I was at the middle of cawiki edit quality campaign, let me finish that [18:16:02] anything I should be aware of? halfak awight ? [18:16:09] o/ Amir1 [18:16:13] Amir1: That I’m dying for CR [18:16:19] https://gerrit.wikimedia.org/r/#/c/412907/ [18:16:46] It's 1k lines /o\ [18:17:03] Amir1: ino [18:17:08] Amir1: What happened is... [18:17:20] I had about 8 nicely isolated patches lined up in GitHub [18:17:36] then got mirrored over to gerrit [18:17:36] Meanwhile, I spun up CI [18:17:44] so… every patch suddenly had to pass CI in order to be merged. [18:18:00] oh neat [18:18:21] I made all the CI fixes at the end, and discovered it would be prohibitively annoying to backport them into the patches where I sinned in the first place [18:18:24] aaargh [18:18:26] so I squashed. [18:19:43] I see [18:19:57] If you see any obvious places to cut, I’m happy to tease this into a few separate patches again, but you’ll see that it’s pretty clean code, and most of the lines are tests [18:21:16] yeah, I have lots of nitpicks. I will make follow ups for that [18:21:29] :) [18:22:03] I’m also happy to pick nits in WIP followups for easier review, then squash. either way [18:22:51] (03CR) 10Ladsgroup: [C: 032] Implement validations, write tests [extensions/JADE] - 10https://gerrit.wikimedia.org/r/412907 (owner: 10Awight) [18:23:00] O_o [18:23:10] You must be using Wakanda tech [18:24:39] (03Merged) 10jenkins-bot: Implement validations, write tests [extensions/JADE] - 10https://gerrit.wikimedia.org/r/412907 (owner: 10Awight) [18:25:06] (03CR) 10jenkins-bot: Implement validations, write tests [extensions/JADE] - 10https://gerrit.wikimedia.org/r/412907 (owner: 10Awight) [18:26:36] 10Scoring-platform-team (Current), 10ChangeProp, 10ORES, 10Services (done): Change ORES rules to send all events to new "/precache" endpoint - https://phabricator.wikimedia.org/T158437#3990250 (10Pchelolo) 05Open>03Resolved Ok, after the second deploy everything seems to be fine. Resolving. Again. [18:28:59] halfak: Donno if you saw my question about how new precaching works? [18:30:11] I didn't [18:31:54] halfak: We flipped the switch, but scoring volume jump to 3x and isn’t settling. [18:31:58] Please check Grafana. [18:32:05] I think we might have to revert and debug. [18:32:07] In meeting [18:32:13] Can't help debug right now [18:32:43] No worries. I’m jumping into meetings too. [18:33:09] I don't see the 3x jump [18:33:39] Oh. One note is that our old changeprop config was horribly out of date, [18:34:08] Old style ended at 16:40, there was a bug for a while, then we got the correct rule deployed at 17:15 [18:34:12] halfak: ok, noted [18:34:13] https://grafana.wikimedia.org/dashboard/db/ores?refresh=1m&panelId=1&fullscreen&orgId=1&from=now-6h&to=now-1m [18:34:19] That does look wrong though [18:34:26] That’s a metrics issue [18:34:46] Just use “all scores returned” until I can deploy the metrics fix [18:39:14] Amir1: Is this true? > We now have browser CI to check that Ext:ORES hasn't broken Special:RecentChanges [18:39:30] yup [18:50:42] awight: "Wakanda tech", yes I do [18:50:48] * Amir1 puts his sunglasses on [18:50:52] lol [19:06:22] halfak: I’m unclear why the precache rules are like this. They seem to just connect “edit” to every model. Can we get rid of the precache config entirely, or flip it to “opt-out”? [19:06:41] Not all models are "edit" [19:06:48] Some are "bot_edit" [19:06:56] or "non-bot_edit" [19:06:58] or whatever [19:07:53] Some models we'll probably want to run on "page_create". [19:08:17] OK we can talk later. Seems like we should have defaults [19:08:29] Also, change-prop is currently matching on non-bot-edits [19:08:39] for all wikis [19:10:20] On the bright side, the ores config is more up-to-date than what was in change-prop. [19:10:42] Nothing in the diff that explains the increased volume, though. [19:11:37] A few wikis went from [reverted] to [damaging, goodfaith] [19:12:11] But that could only have doubled the load at maximum. [19:12:23] More wikis getting scored? [19:12:30] Also maybe it's scoring bot edits in wikidata. [19:12:33] That's a lot of edits. [19:12:34] aah [19:12:40] tinx: more wikis [19:12:43] jinx [19:13:01] no, I was saying the change-prop config had previously filtered out bot edits on all wikis, so I carried that over. [19:13:30] Now I see that we should have dropped that matching condition. [19:15:31] 10Scoring-platform-team (Current), 10ChangeProp, 10ORES: Drop "non bot" condition from ORES changeprop rules - https://phabricator.wikimedia.org/T187927#3990435 (10awight) p:05Triage>03Normal [19:18:06] 10Scoring-platform-team (Current), 10ORES: Investigate 3x scoring load after ORES changeprop migrated to new precache endpoint - https://phabricator.wikimedia.org/T187928#3990452 (10awight) p:05Triage>03Normal [19:18:29] Additional wikis satisfies most of my curiosity. [19:19:59] Let’s wait to remove the non-bot restriction until we’re on newer hardware... [19:20:10] scb1002 is handling this load on its own [19:21:08] 10Scoring-platform-team (Current), 10JADE: Build prototype JADE extension - https://phabricator.wikimedia.org/T187216#3968157 (10awight) https://gerrit.wikimedia.org/r/#/c/412907/ [19:21:53] Showcase starts in 9 minutes. [19:23:34] See #wikimedia-research [19:26:24] wiki-ai/editquality#125 (cawiki_campaign - 3f5b39f : Amir Sarabadani): The build passed. https://travis-ci.org/wiki-ai/editquality/builds/344475811 [19:35:10] halfak: awight ^ [19:35:34] Showcase in progress. I'm gonna be distracted for a little over an hour ^_^ [19:35:41] https://www.youtube.com/watch?v=fpmRWCE7F_I [19:35:54] I present second -- so, in about 20 minutes. [19:42:12] (03PS1) 10Awight: Add coverage annotations [extensions/JADE] - 10https://gerrit.wikimedia.org/r/413228 [19:43:17] Amir1: Vanity patch, https://gerrit.wikimedia.org/r/413228 [19:43:35] (03CR) 10Ladsgroup: [C: 032] Add coverage annotations [extensions/JADE] - 10https://gerrit.wikimedia.org/r/413228 (owner: 10Awight) [19:43:57] :P [19:46:57] (03Merged) 10jenkins-bot: Add coverage annotations [extensions/JADE] - 10https://gerrit.wikimedia.org/r/413228 (owner: 10Awight) [19:47:23] (03CR) 10jenkins-bot: Add coverage annotations [extensions/JADE] - 10https://gerrit.wikimedia.org/r/413228 (owner: 10Awight) [19:56:05] 10Scoring-platform-team (Current), 10User-Ladsgroup: Edit quality campaign for Catalan Wikipedia - https://phabricator.wikimedia.org/T187771#3985183 (10Ladsgroup) It's uploaded, now you can label in http://labels.wmflabs.org/ui/cawiki/ [20:05:52] Amir1: aww, the coverage report isn’t budging. [20:08:31] 10Scoring-platform-team (Current), 10ORES: Investigate 3x scoring load after ORES changeprop migrated to new precache endpoint - https://phabricator.wikimedia.org/T187928#3990710 (10awight) Two things we can do to help: * Turn on debug logging for a bit, to see exactly what's being scored. * Write tests for /p... [20:10:40] awight: I can fix that [20:10:45] wait a sec [20:10:59] Any idea what’s going on? [20:12:53] yup [20:41:21] 10Scoring-platform-team (Current), 10ChangeProp, 10ORES: Drop "non bot" condition from ORES changeprop rules - https://phabricator.wikimedia.org/T187927#3990435 (10mobrovac) @awight are you computing scores for them? If not, it would be good to keep the condition in ChangeProp in order to reduce useless netw... [20:44:57] OMGLUNCH [20:45:47] halfak: ^ I don’t know the big picture yet. Is it correct that we want to expand the changeprop to precache all edits, bot or not, with the exception of wikidata which is marked “non bot edit” and will be filtered by the precache_map? [20:47:15] awight: yes, I think I was the person who disabled precaching in change prop [20:47:24] ? [20:47:30] It was still enabled AFAICT [20:47:55] The number is too big in some wikis that it has network traffic implications [20:48:00] e.g. https://gerrit.wikimedia.org/r/#/c/411087/ [20:48:16] Amir1: Ah I see, you disabled bot edits. [20:48:28] yup [20:48:54] even we just discard them, the request internally cost too much [20:49:08] (Think of global warming) [20:49:53] but seriously, looking at gangalia I/O stats of scb nodes, it was cut to around half just because of disabling bots [20:49:55] Only wikidata has the “nonbot_edit” filter in our config [20:50:11] hmm [20:50:19] okay so the plan is to never precache bot edits? [20:50:34] s/never/not at this moment/ [20:50:36] I think so, halfak|Lunch has the last call IMO [20:50:51] I'm OK with cutting out bot edits. [20:50:57] * halfak|Lunch lied and never leaves for lunch on time. [20:51:05] not just we discard it in precaching, also not even send it via change prop [20:51:11] "Urgent" emails are sometimes urgent [20:51:19] +1 for not even sending it. [20:51:19] :)))) [20:51:22] Oh wait. [20:51:23] Shoot. [20:51:30] We want it for the wp10 model. [20:51:37] Nothing wrong with bot edits for wp10 [20:52:08] How much are we saving by turning it off? [20:52:27] Would we also want bot edits for itemquality in wikidata? [20:52:44] Yes. [20:52:52] All of the article/item quality models should have bot edits [20:52:53] Hmmm [20:52:59] Those are the two largest wikis :) [20:53:06] Right [20:53:09] :| [20:53:39] totally random sample of enwiki: https://stats.wikimedia.org/EN/EditsRevertsEN.htm [20:53:47] It’s only 10% bot edits [20:57:17] So, enwiki doesn't matter. I think Wikidata is majority bot though [21:00:51] yup [21:01:54] (03PS1) 10Awight: Bump ORES submodule [services/ores/deploy] (stretch_conversion) - 10https://gerrit.wikimedia.org/r/413245 [21:02:35] (03PS1) 10Awight: Bump ORES submodule [services/ores/deploy] - 10https://gerrit.wikimedia.org/r/413246 [21:08:04] (03CR) 10Awight: [C: 032] Bump ORES submodule [services/ores/deploy] (stretch_conversion) - 10https://gerrit.wikimedia.org/r/413245 (owner: 10Awight) [21:08:06] (03CR) 10Awight: [C: 032] Bump ORES submodule [services/ores/deploy] - 10https://gerrit.wikimedia.org/r/413246 (owner: 10Awight) [21:09:05] Disconnected, just saw the backlog. [21:09:31] But we’re talking about enabling wikidata and enwiki’s bot edits in changeprop. I was using enwiki as an upper bound on bot activity on the runner-up wikis by volume. [21:10:21] If we’re going to be precaching bot edits on the two most botty wikis, do we really need to worry about <10% extra HTTP requests from changeprop for the other wikis? [21:10:35] s/most botty/bottiest/ [21:10:58] (03CR) 10Awight: [V: 032 C: 032] Bump ORES submodule [services/ores/deploy] (stretch_conversion) - 10https://gerrit.wikimedia.org/r/413245 (owner: 10Awight) [21:11:00] (03CR) 10Awight: [V: 032 C: 032] Bump ORES submodule [services/ores/deploy] - 10https://gerrit.wikimedia.org/r/413246 (owner: 10Awight) [21:19:15] halfak: Amir1 ^ Sorry for the net’ fail, just trying to finish up the bot/nonbot changeprop discussion. [21:20:26] awight: most other wikis doesn't matter for me. Wikidata is the case I really want to avoid and if it's not possible, let's be it [21:20:54] Amir1: But halfak was saying we should have itemquality enabled for bot edits, unless I misunderstood? [21:21:19] yes, which means it's not possible and I have no objections to that [21:21:33] I see. OK thanks for the confirmation! [21:25:42] 10Scoring-platform-team (Current), 10ChangeProp, 10ORES: Drop "non bot" condition from ORES changeprop rules - https://phabricator.wikimedia.org/T187927#3990950 (10awight) In IRC, we determined this is the right thing to do. We do want to compute scores on bot edits in enwiki (wp10 model) and wikidata (item... [21:33:18] PROBLEM - https://grafana.wikimedia.org/dashboard/db/ores grafana alert on einsteinium is CRITICAL: CRITICAL: https://grafana.wikimedia.org/dashboard/db/ores is alerting: 5xx rate (Change prop) alert. [21:38:08] awight: are you around for a fast review of the ores extension? [21:38:28] Amir1: In a meeting for 20min, but will stick around afterwards [21:38:34] cool [21:42:04] (03PS1) 10Ladsgroup: Add tests for new system of $wgOresModels [extensions/ORES] - 10https://gerrit.wikimedia.org/r/413259 (https://phabricator.wikimedia.org/T181335) [21:44:57] awight: halfak for when you have time: https://github.com/wiki-ai/editquality/pull/130 [21:44:59] awight: https://gerrit.wikimedia.org/r/413259 [21:45:04] :) [21:57:28] RECOVERY - https://grafana.wikimedia.org/dashboard/db/ores grafana alert on einsteinium is OK: OK: https://grafana.wikimedia.org/dashboard/db/ores is not alerting. [22:04:57] halfak: Amir1: metrics are fixed, so we can see the beast precaching has become. [22:07:50] cool [22:11:55] Amir1: hammered for time, can the review wait until tomorrow? [22:12:18] Two things: https://github.com/wiki-ai/editquality/pull/130 and https://gerrit.wikimedia.org/r/413259 [22:12:23] The second one is just adding tests :D [22:12:37] (which I found a bug while writing the tests and fixed it) [22:15:20] Amir1, why are you switching the argument order? [22:15:47] halfak: because of https://phabricator.wikimedia.org/T187742 [22:15:48] :P [22:16:31] Seems like sorting alphabetically isn't quite right. [22:17:14] halfak: if we don't give it a sort it practically changes it everytime we generate a Makefile [22:17:36] Can we be explicit about which file gets listed first? [22:17:54] we can define a list and a dictionary in the yaml config but that seems overkill to me [22:18:42] awight: I actually thought your comment mean, "give me the reviews" sorry for misunderstanding [22:18:47] have fun [22:19:28] https://github.com/wiki-ai/editquality/blob/master/templates/Makefile.j2#L78 [22:19:29] Amir1, ^ [22:20:20] the dictsort in the line above makes things problematic [22:20:51] There's only ever two files and they have predictable names, right? [22:21:15] if we change big_sample|dictsort to big_sample.items(), it messes up every thing [22:21:31] halfak: hmm, can't say for sure. I think fawiki has two campaigns [22:22:45] halfak: hmm, It seems we merge them later [22:22:49] so it seems fine [22:26:22] I think it's safer to be explicit than to rely on sorting in this case. [22:26:30] If possible :) [22:26:41] sure thing [22:27:00] Let me finish this PR and will get to fixing it probably tomorrow [22:27:11] halfak: is it okay, I think I've got headache [22:27:29] OK no worries. [22:57:17] 10Scoring-platform-team (Current), 10User-Ladsgroup: Fix or decide on edge cases of Makefile - https://phabricator.wikimedia.org/T186453#3991276 (10Ladsgroup) a:03Ladsgroup