[06:05] *** Turing_i left irc: Quit: Leaving [08:09] *** Turingi joined #setiquest. [10:56] *** Turingi left irc: Read error: Connection reset by peer [15:13] *** Turingi joined #setiquest. [16:52] *** sigblips joined #setiquest. [17:52] *** nigra joined #setiquest. [18:01] Welcome to the weekly setiQuest community meeting. [18:01] jrseti won't be here today. [18:02] Hello sigblips. Anything on the agenda today? [18:03] I'd like to talk about the forum's spam problem and then we can talk about SETI Live. Anything you'd like to talk about today? [18:04] Sounds like a plan. [18:06] This morning I woke up to 82 spam messages. 81 of them were created by a single spammer from China over the course of 20 minutes. [18:07] It took my an hour to clean up this mess. What a complete waste of time. I've had enough. [18:07] anything to do with "puma metal racing shoes"? Saw that listed as a topic twice on the main page. [18:08] Nasty. Anything that can be done about it? [18:08] Yes, that was 81 of the spams. [18:08] I didn't want to do this but I'm switching the setiQuest forum to a moderation mode. [18:08] so, any posts have to be moderated/approved first? [18:09] Yes. This will be Game Over for the spammers but it will also be a slight pain for legit users. [18:10] won't the moderation be just as much trouble for the moderator? or can it be handled more easily than deleting posts after the fact? [18:11] The spammers won't bother spamming because they'll know that there is zero chance of their spam being approved. Well at least that's the theory. [18:12] ah, I get it. [18:13] I've been running my moderated baudline Google Group http://groups.google.com/group/baudline for about 8 years or so and spam is not an issue. [18:14] I hope the moderation works here because the other option is shutting down the setiQuest forum. [18:15] Is spam an issue over on the SETI Live talk pages? [18:16] not a problem really (knock on wood) but maybe it's because or the way it's structured. Don't think you can use a bot on it. [18:17] I don't think it will be a major inconvenience for users. Discussions aren't usually on a rapid frequency. [18:18] I guess it'll depend on the frequency of moderation checks by a moderator. [18:18] Still, not a desirable transition, but it's pretty obviously necessary. [18:19] He have a CAPTCHA and I think that stops the bots. I think all of the setiQuest forum spamming is done manually. The source is mostly China and a few poorer Eastern European countries. Spam is a billion dollar worldwide industry. ): [18:19] We have a CAPTHCA ... [18:20] Oh yeah, CAPTHCA... forgot about that. [18:21] We joked about two years ago that the CAPTCHAs were too easy and instead of basic math they should be calculus problems! That would keep spammers away. [18:22] hehe. That would keep a lot of we engineers who are a bit rusty out of it as well :-) [18:22] Anyway we'll know in about a week or two how well the forum moderation is working. [18:23] Cool [18:23] We have new poll called "What is your favorite setiQuest data signal source?" Vote at http://setiquest.org/content/what-your-favorite-setiquest-data-signal-source [18:23] And discuss at http://setiquest.org/forum/topic/what-your-favorite-setiquest-data-signal-source [18:24] That's it for me. What is going on at SETI Live? [18:25] Well, inching toward upgrading the site with the improvements... [18:26] Just a few loose ends and we'll be able to switch over with an improved classification interface, tutorial and followups actually working... [18:27] I really want to do this in the next few days, but we'll see... [18:28] Well that sounds good. I like the new date and frequency labels. They're simple but informative. I think this will help connect people with what the data means a bit more. [18:30] Latest followup testing on the main site was on saturday (I think). I triggered a followup through manually marking the same waterfall repeatedly with special test code. very realistic for normal operation... [18:30] we're still working on confirming we can still go beyond the first followup as we have with simpler, less realistic testing... [18:31] Yes, the time will also be accurate finally. [18:32] I think the loading speed will be the biggest difference for most users. [18:33] At what level in the automated followup process do red lights start flashing, sirens go off, and a computer sends Jill Tarter an instant message? [18:34] people will also be more informed about what's going on at the moment - how long we have to classify as a group as well as a suggested time for individuals based on how many waterfalls left in the batch and how many users are actively marking signals. [18:36] Hehe, actually, it's a total of ten waterfalls that have to be marked (or not marked for an off-target measurement) consistently and an email goes out at SI to certain people. Don't know if Jill gets one, but I assume so :-). I've triggered 2 I think during testing. [18:36] We're still debating at exactly what people should be aware of at what point in the process... [18:37] Part of it will probably have to do with how often it happens. [18:38] Yes, we don't know how often a followup will be triggered yet, but if it's fairly rare, I'd sure like people to know the moment it happens. [18:40] The simulations we've put up for people to mark along with the samples of single-beams people like you and others have collected, and the marking database should help us figure out how often it might happen. [18:41] Anxious to get the site improvements up there so we can move on to improving the signal gallery help, define the data access API and start on some data analysis applications for the site. [18:42] That sounds like a lot of work. Do you know how long SETI Live has funding for? setiQuest ran out of funding last year ... [18:44] That ATA hibernation really wrecked havoc on the project. [18:47] Yeah, that was a real blow. I really don't know what the situation for ongoing SETILive support is. zooniverse may keep it maintained at low effort like their other projects. I think that will be settled in the next few months. [18:50] To finish on the subject of followups and what to tell users and when. We have to be concerned about biasing, especially with off observations. We'll err on the side of keeping folks informed, however. [18:52] That's all I've got, I guess. [18:54] Yeah, I can see how biasing might be a problem. A group of SETI Live users could conspire to keep the ATA doing a followup on the same spot, over and over. The only real harm would be some wasted observing time, so it doesn't seem like a huge problem. [18:56] No harm other than waking Jill Tarter up in the middle of the night of course. (: [18:58] hehe... There are ways to prevent people conspiring to trigger followups. Not perfect, and could waste some of that ATA's time, but but would allow Jill to sleep through the night at least. [19:00] I'm sure Jill would be happy to know that people are thinking about these potential problems. [19:02] Talking about Jill, I wonder how her Google Zeitgeist presentation went? [19:02] Good question. Haven't heard. [19:04] Gotta sign off. Oh, When are you going to switch to moderated mode on the forum? Have you decided yet? [19:05] Sometime today. I'm still trying to find the setting again in the Drupal control panel. The Drupal administration pages are total interface disaster. [19:06] Well, good luck. Thanks for moderating the chat today. See you next week, sigblips. [19:06] Well it's noon here in California. Till next week. [19:07] *** nigra left irc: Quit: nigra [00:00] --- Wed Oct 24 2012