Steve you made me laugh to fall off my seat... not far to fall I realise now. Hope that goddess crown is a tall one (do they get crowns?) for the newly shortened moi, and thank you for the honor Sir Summit! Julia Rossi (talk) 22:25, 22 January 2008 (UTC) : ))[reply]
Yes. The funny thing is, based on the logging the archiving script prints as it runs, I suspected it might have made that very mistake last night, but when I went and looked, all seemed fine. But I now see that's because User:Metalcore424 jumped in and fixed it so quickly I didn't even get a chance to see the wrong version! —Steve Summit (talk) 03:29, 25 January 2008 (UTC)[reply]
The Help desk archives have had error links to the preceding and following month since August. I have just manually fixed it for August to February, for example [1]. PrimeHunter (talk) 16:59, 3 March 2008 (UTC)[reply]
I noticed that a few days ago and neglected to follow through. I had no idea the glitch went that far back, though. Thanks; I'll look into it. —Steve Summit (talk) 17:24, 3 March 2008 (UTC)[reply]
Okay. I see the problem. It only has to create that page once a month, of course, and for the Help Desk, it looks like it has never worked. :-( But happily tomorrow is the day to see if my newly-implemented fix for the problem works; I won't have to wait long. —Steve Summit (talk) 05:21, 4 March 2008 (UTC)[reply]
Possibly, although in general it's not a good choice, because it's quite narrowly tailored for both the formatting and the archiving structure of the existing Reference Desks and the (very similar) Help Desk. There are quite a few other archiving bots that are much more general, I think.
Actually, we do not keep archives. Once an article is translated adequately, its entry on PNT is summarily removed as a closed case. Same thing when an article listed at PNT ends up being deleted (speedy or otherwise) before its translation is completed. It's only the date header that needs to be inserted on a daily basis. We need the date header so that we know when it is time to nominate an article for deletion after the delay for translation has expired. --Blanchardb-Me•MyEars•MyMouth-timed 02:06, 4 April 2008 (UTC)[reply]
Oh! Okay. Date headers it's good at. I'm not sure what the best way will be of ensuring that it doesn't get confused, though, by that second, monthly section you've got at the bottom of the page. Lemme think about it. —Steve Summit (talk) 02:42, 4 April 2008 (UTC)[reply]
Actually, it is pretty straightforward: the new header should be inserted right before ==Translated pages that could still use some cleanup==. --Blanchardb-Me•MyEars•MyMouth-timed 01:08, 7 April 2008 (UTC)[reply]
Looks like it's working (modulo maybe an extra blank line).
Couple things you should know:
It's semiautomated (not fully automated), meaning that it runs only when I manually invoke it. Sometimes it's more like my midnight than UTC midnight, and some days (for example, if I'm on vacation) it doesn't run at all. But that's okay, because
It's designed to do as much work as it has to; it doesn't assume it runs every night. If it notices that there's a date header from a day or two ago missing, it'll insert as many as it has to. But that's potentially a mild problem for you, because
If you delete a header 'cos everything for that day is dealt with, the bot will painstakingly reinsert it.
For now, at least, I'm afraid you're going to have to live with #3, and I hope it's not too big an issue for you. (After three days, you can delete an empty date header with impunity, because that's as far back as the bot looks.) —Steve Summit (talk) 00:45, 11 April 2008 (UTC)[reply]
I have the very vaguest of memories of visiting a pylon website which suggested that different designs of pylons (for different voltages), were given, IIRC, animal names. But, of course I cannot find the site, and in any event do doubt my memory. But hey, it's a plausible-ish explanation of the question. --Tagishsimon(talk)01:26, 16 May 2008 (UTC)[reply]
I found this[2] and under code words here[3] seems they're short cuts to otherwise technical descriptions of wire etc weights, sizes and composition. Julia Rossi (talk) 01:50, 16 May 2008 (UTC)[reply]
AfD guidance
Hi Steve, I've been supporting a new article against deletion and was wondering if you could tell me when such an article is likely to be out of quarantine once it's been improved and met criteria etc because its creator would like it to go into noms for DYK on time, or do you have to wait for someone to announce the outcome in their own sweet time? Thanks for any advice, Julia Rossi (talk) 23:18, 15 May 2008 (UTC)[reply]
There's a defined, formal "quarantine" period now? Gad, I knew AfD had gotten alarmingly officious, but I hadn't heard of that wrinkle...
So I'm afraid I don't know.
(And does the quarantine specifically preclude DYK nominations? Oh, what a tangled web we weave...)
Thanks for your quick reply, appreciate it. The AfD is a scramble on that article (rather dour around the deletes) and I hate to draw the heat, but it's been vastly improved and since there's no telling, I'll go ahead and do the DYK nom. Cheers, Julia Rossi (talk) 01:14, 16 May 2008 (UTC)[reply]
Gotta love them rabid deletionists. Some editors define their contribution to the project as adding information to the encyclopedia, others (it seems) by aggressively removing stuff from it. 'Tis sad. —Steve Summit (talk) 01:21, 16 May 2008 (UTC)[reply]
Sad indeed. There's a scramble to kill Florence Scovel Shinn I added art credentials to way back when it was a full article, now it's a stub with a pile of references and fingers around its throat and a foot or two on its head. One more question Steve, I don't know where to put the dyk nom for the new one – on the day it was started (May 11) or the day it was finished being expanded (May 15), so more guidance please would be good, (hope you're still up!) Julia Rossi (talk) 01:40, 16 May 2008 (UTC)[reply]
("Still up"?! It's at least two hours before my normal bedtime -- whatever that means -- although I'm sleepy after getting three hours last night, so I'll probably crash soon.)
Reading their guidelines, it looks like the clock starts the day the article was created (or, with special dispensation, the day it was "expanded fivefold or more". So it sounds like May 11's the day.
Good luck. (Me, I've never had any with DYK -- I can do lots of things, but complying with that level of bureaucracy isn't one of them.) —Steve Summit (talk) 02:06, 16 May 2008 (UTC)[reply]
Thanks again and for the warning about b'crats. I also have a policy of avoiding shark-infested metaphysics articles which I forgot with Flo Shinn coz I was only thinking art and article improvement. Silly me. Sweet dreams and a good day in the morn, Julia Rossi (talk) 02:13, 16 May 2008 (UTC)[reply]
Steve, I sometimes get lucky with DYK -- maybe it's about the catchy line, I dunno, but if you have an article you want to try out DYK for (even ACSR), let me know and I'll nominate it, you never know –– best Julia Rossi (talk) 11:45, 16 May 2008 (UTC)[reply]
To be honest, when I said "I've never had any with DYK", there'd really only been the one time, at which point I hadn't realized that DYK facts had to be from new articles, such that the (relatively antique) one I'd proposed didn't stand a chance. But thanks for the offer all the same! —Steve Summit (talk) 02:08, 17 May 2008 (UTC)[reply]
I am such an idiot. When I read "try out DYK (or even ACSR)", I thought that "ACSR" was some other Main Page featurette I'd never heard of. (But eventually I figured it out.) —Steve Summit (talk) 03:17, 17 May 2008 (UTC)[reply]
Haha. A spinout -- my write-speak is sometimes idiomatic so maybe that didn't help, but when I went to your article ACSR mentioned in the question about code words above, I saw a stub and thought if you were going to expand it, there's a chance. : ) All good, Julia Rossi (talk) 08:07, 17 May 2008 (UTC)[reply]
Steve, I know nothing of bots and spam and technology in general. Do you think the spambot threats using multiple IP's can be adequately met by manual reversion alone? Alternatively, do you think it is possible that a (perhaps even minimal 1 min) protection might be sufficient to turn them off? I understand the points made my PPG and others, but I don't understand the mechanics enough to know whether 15 minute semi-protection is a valid conclusion. Maybe we need to see it happen first. In any event, your assessment of this spambot situation would be appreciated. ---Sluzzelintalk07:03, 15 June 2008 (UTC)[reply]
I don't think the 15 minute conclusion is valid at all. That's like saying if you unplug your computer from the Internet for fifteen minutes, the email spammers will all give up and go away and not spam you any more.
Legitimate bots (I've written a few) notice problems like that and stop. But a bot written with bad intent obviously possesses all the bad intentions put into it by its author, and is infinitely more patient. So there's no reason to think that after one minute, or 15 minutes, or one hour, or one day, the bot won't be there ready and waiting to start spamming again.
I don't know if manual reverting can adequately deal with a large-scale bot threat. Certainly if the bad bot is aggressive enough, manual reverting won't cut it. But that doesn't mean that protection is our only recourse, and I continue to hold that it's the wrong recourse. Other possibilities are autorevert bots and blacklists. —Steve Summit (talk) 14:10, 15 June 2008 (UTC)[reply]
Thanks Steve. In that case I wish we had an autorevert bot watching over the desks. I'm really not happy about the possible effects of this precedent, probably the main reason I'm opposed to allowing semi-protection. I've seen too much eagerness in other areas, to trust every admin with doing the right thing. I don't know which law this is, but the consensus of 99 admins to do nothing, doesn't mean a lot when 1% decides to be bold with administrative tools. I don't like stubbornly sticking to principles, but I'm concerned about protections happening even when it's clearly unnecessary and happening regularly. ---Sluzzelintalk03:54, 16 June 2008 (UTC)[reply]
I think there's plenty of hope to hold out for. There are "autorevert" bots out there, and the author of one of them has offered to help here. There's also another technical solution that's been proposed and may get implemented. For myself, I'm tempted to write precisely a "watch over the refdesks" bot myself. Furthermore, this may be the incident that goads me into seeking adminship myself, so that I'll be in a position to immediately adjust inappropriate protection and stuff. —Steve Summit (talk) 04:19, 16 June 2008 (UTC)[reply]
Because the bot is semiautomatic, which is to say, it runs only when I invoke it. Depending on my schedule, it can be any time during my evening (ET) when I run it, or occasionally not at all, in which case it catches up the next night. —Steve Summit (talk) 03:56, 27 June 2008 (UTC)[reply]
Hi Ummit, this is about as trivial of a niggling concern as they come. At least it may give you a chuckle.
Can you change Scsbot's edit summary to say "edited by bot" rather than "edited by robot"? Technically, you probably don't have a mechanical device typing at the keyboard (but send me the plans if you do). I've thought about this for a while, then I actually looked at Robot and the article also says "software agents ... are usually referred to as bots to differentiate". It's even referenced! (But the cite points to a dead link, and I found a few more reference problems - serves me right :). Just a suggestion, accept or ignore as you wish. Cheers! Franamax (talk) 03:46, 15 July 2008 (UTC)[reply]
You're probably right, although I confess that my initial reaction to "usually referred to as bots to differentiate" is "citation needed". I'll think about it, and if I end up leaving it like it is, please don't think I didn't appreciate the suggestion! —Steve Summit (talk) 01:20, 17 July 2008 (UTC)[reply]
Gahh. Please don't add any cite tags to that article! I lost a few hours of my life just trying to clean up the first dozen cites already there :) I also was able to confuse myself up real good looking for a definitive citation for that exact subject. The data is mixed, some call 'em bots, some call 'em robots, I guess I'm just expressing my personal preference.
I like Coneslayers comment though - I think I'll get one of those little birds for when I actually want to make coherent edits, after all, it worked for Homer J. Absolutely no problem if you leave it as is, it's already forgotten :) Franamax (talk) 03:04, 17 July 2008 (UTC)[reply]
Thanks! (Actually it was such a great trip that I wouldn't have minded staying longer, but at the same time, I'm glad to be back, too.) —Steve Summit (talk) 00:04, 29 July 2008 (UTC)[reply]
Mixed joys then. Hopefully the after effects will linger for you, who knows. Maybe you can revisit when not being Sir Summit, a pillar of the pedia, : ) Julia Rossi (talk) 07:55, 29 July 2008 (UTC)[reply]
Thanks much. I thought I put a {{prod}} tag on that page last night, but reviewing my edit history, it looks like I might have neglected to hit Submit, or something. —Steve Summit (talk) 00:04, 29 July 2008 (UTC)[reply]
Oh! Thanks. I had no idea. (The bot's been issuing an error message for the past week or two, but I misinterpreted it and didn't realize what was actually going on.) Fixed now. —Steve Summit (talk) 00:37, 24 September 2008 (UTC)[reply]
Just to let you know, I've solved the December 11th problem. If the bot is still holding on to the December 11 content you can purge it. See the RD talk page for details Nil Einne (talk) 12:04, 16 December 2008 (UTC)[reply]
Thanks much! —Steve Summit (talk) 12:39, 16 December 2008 (UTC)
_053.jpg)