I have shown you who the creator is and that they have given permision to use it and I will reupload it again as I am not ignoring any policy but rather following it. It is a publicity photo as it was tagged and I have shown a source for it's creator and where they have allowed it to be used here. If you feel the need to be a
dick and delete it at least have the freaking common courtesy to remove the red link and caption, which you haven't done this time or the last time and left the article looking awful. Seriously, it is people such as yourself who are here just to be an impediment and delete things so they feel they can have power when what they are deleting is obviously genuine that have soured my opinion of this project. Seriously, threatening me with a block for uploading a fair use image with a source of it's creator saying the photo can be used here. You must feel like a real big shot.--
E tac21:06, 15 March 2007 (UTC)reply
I would have assumed good faith until you accused me of blatant disreguard of policy and threatened to block me. Maybe it is you who should practice what you preach and not make threats and accusations in the first place. An image doesn't have to be FREE use if the person who took the image releases rights to it for use here, which they did. Did you bother looking into the sources that I provided? By the way you still didn't clean up the mess your deletion made so I will do that now.--
E tac06:43, 16 March 2007 (UTC)reply
You left me a message regarding a change to the above page. I changed "****" to "shit", which I believe is perfectly legitimate. This is supposed to be a serious academic reference source and using asterisks for minor swear words is unnecessary.
Hi - it would have been
User:MartinBot which left the message, and that bot is seperate from me, and runs unsuervised. I can only apologise for the inconvenience and suggest that you re-enter your edit. Martinp2317:11, 16 March 2007 (UTC)reply
Re: I am no newbie to wikipedia snowolf. I was simply requesting it here as this is the bots owners page.
Retiono Virginian
Retiono - the message wasn't directed to you, but to Khukri. Snowolf: I'va been informally admin coaching Khukri for a while now. Martinp2317:21, 16 March 2007 (UTC)reply
Miscellaneous, Science, Humanities, Mathematics, Computing & Language are apparently all fine, so it looks like there were only major issues with the Help and Entertainment desks--VectorPotentialTalk00:36, 17 March 2007 (UTC)reply
Thanks for all your hard efforts! As these two desks are less monitored than the others, I suspect that the bot missed an archival on them dur to a connection error or something, causing errors to propagate through over the next few days. The bot will run as normal tonight, and hopefully all will work well :). Thanks again, Martinp2319:57, 17 March 2007 (UTC)reply
Hmm...
Just reverted some vandalism on the article
Althea, which your bot quickly reverted back. Maybe it was because I used the word "vandalism" in the edit summary. That's clever, cause as we know, every self-respecting vandal will make clear his intentions to vandalise the article in the edit summary.
89.243.75.13500:41, 17 March 2007 (UTC)reply
It wasn't your edit summary, it was the page length, the bot reverts edits that make pages shorter, this unfortunately means that if a vandal adds several kB worth junk into a page, the bot will restore it--VectorPotentialTalk00:44, 17 March 2007 (UTC)reply
Why are you reverting back to the old British Museum History page - it was a sample, the page needs deletion. The information is already in the British Museum page.
We need to have the bot flag removed. MedBot never had it, as many people have bot edits hidden on their watchlists. We want to be sure MedBot1's edits run without a flag. ^demon[omg plz]13:30, 18 March 2007 (UTC)reply
Hi, I have a DotNetWikiBot and I need to revert changes I made to many pages. Do you know of a good way to do this? Thanks, Yoni
Yonidebest14:04, 18 March 2007 (UTC)reply
In what context would you be reverting the changes? What I mean is, will it be an always-on antivandalism bot, like MartinBot, or just a one time run reverting all of the changes by a single user? Both are perfectly doable, with the second being much easier. Martinp2314:07, 18 March 2007 (UTC)reply
Well, I made a bot that fixes spelling mistakes on he.wiki. But it turns out that two of my regex were messed up and thus the replacements were incorrcet. I have a list of the articles with the above problem, and I need to undo the bot's edit. I am completely lost - any help will be helpfull :-)
Yonidebest14:11, 18 March 2007 (UTC)reply
Something like: query.php?what=usercontribs&uctop=only&titles=User:Yourbotname&uclimit=(however many bad edits there are) will get you a list of the user's bad contributions which are still the most recent edit. If you retrieve the corresponding oldid for each page title from this list, you can then run query.php?what=revisions&revids=(the old id from earlier)&rvcontent to get the content of the revision before your bot edited, which you can then save back to the page, so reverting your bot.
I hope this makes sense - I'm not sure if there's an integrated funtion for this in DotNetWikiBot, but the above should work well if you just throw the results into some sort of array (something in System.Collections which uses a key/value pair would work well, with the oldid as the the key and the page title as the value). If you have problems doing this, I can code it myself for you, but it may take a while (I need to get an IDE installed on this system, for a start :)). Martinp2314:31, 18 March 2007 (UTC)reply
Thanks. Problem is that I dont wish to revert all my bots contributions. It made thousands of edit, and not all of them are problematic. I have a list of the problematic titles in a txt file. Perhaps there is a way to use a similar query page to find out the revid of each edit the bot made and then use the "undo" to revert the bot's edit:
'/?title=Title&action=edit&undo=2810979'
but how can I find out what is the revid of each edit that my bot did? Is there a query page that I can get this from?
The only way it would be possible to do this, through the vaious Wikimedia APIs, would be to get the user contribs list (as I mentioned above). This gives the revid, time and page, and the option &rvcomments will give the edit comment for each of the contribs. If there is aspecific time when the bad edits started, and a time when they ended, then you could pass these to the query (ucstart= and ucend=) to get only the contribs between those dates, and so limit your query to the bad edits. Alternatively, if there's no specific time, you can just get the full list of contribs (with oldids and titles) and get the code to remove the unaffected pages from this list, though in this way you risk removing some edits which need reverting. Could you give me a link to the contribs page of the bot so that I can see what I'm talking about :)? Martinp23
and get everything I need. I can check the time of the contributions and set the start and end times as you said.
Here is a link to my bot's contributions. One last question I think... Is there a way to immediately save an "undo" link?
Yonidebest15:05, 18 March 2007 (UTC)reply
On that last question, I'm not sure :(. The way I'd do it, to avoid the undo feature (simply because I don't know how to use it) would be to get the revision text for each old ID as described above (using query.php) and then save that. If you need to use the undo feature, you may be able to get more help at
WP:BOWN. Martinp2315:33, 18 March 2007 (UTC)reply
Well, considering I rather not revert to the old ID because edits to the page after the bot edited it will be deleted, I rather use the undo function. Problem is that it doesnt save the page. I'll check that notice board, and if no one knows how to use it, I'll return to you to get the final touches on the old ID idea. Thanks!
Yonidebest15:39, 18 March 2007 (UTC)reply
Yes - that's going to be a problem, and makes the automatic undo feature very attractive for this. If it can't easily be done, then the command "&uctop=only" in the query string will only show contribs which are still the newest on their respective articles. Martinp2315:52, 18 March 2007 (UTC)reply
Looks like it isn't possible to save a page this way. Perhaps it is possible to get the text in the loaded undo page and save that? I don't know how to manage this...
Yonidebest17:23, 18 March 2007 (UTC)reply
I'm not sure on this either :( - have you tried intalling popups into your bot's js and using the autosave feature? It might just work. Martinp2317:28, 18 March 2007 (UTC)reply
Yes, I did try that. But it doesnt work. I check the js and the autosave is for reverting, not undoing. About my previous question: Perhaps it is possible to load the undo page and then save it, just as the DotNetWikiBot loads and save this normal articles. The only problem is that I have no Idea how to change DotNetWikiBot so that it will do this =/
Yonidebest17:39, 18 March 2007 (UTC)reply
Though I don't know whether it's providing an undone version or simply ignoring the &undo part. Probably something to test with. Martinp2317:55, 18 March 2007 (UTC)reply
No - it doesn't provide an undone version :(. Perhaps there'll be someone at
WP:BOWN who'll have some knowledge of how to do this properly - using the conventional method it's impossible to achieve a result (I think). Martinp2318:01, 18 March 2007 (UTC)reply
OK, Thanks for your help. I gave it another try and it seems to work ok. I opened another option to load a page in DotNetWikiBot:
public void LoadUndo(string undo)
{
Uri res = new Uri(site.site + site.indexPath + "index.php?title=" + title + "&action=edit&undo=" + undo);
Bot.InitWebClient();
try {
text = Bot.wc.DownloadString(res);
Regex rx = new Regex("id=\"wpTextbox1\" rows='25'\ncols='80' >(.*)</textarea>", RegexOptions.Singleline);
MatchCollection matches = rx.Matches(text);
foreach (Match match in matches) // although there should be only one result anyway
text = match.Groups[1].Value;
}
...
The query.php only outputs a set limit of results with each query - you can increase that with uclimit=500 (that's the max), to get
[2]. There should also be a way to get the subsequent pages of contribs, though I can't find the documentation right now. Give me 10 mins. Martinp2319:50, 18 March 2007 (UTC)reply
OK, this is slightly different, but is well documented [/w/api.php?action=query&list=usercontribs&ucuser=Yonidebot&ucstart=20070315000000&ucend=(put the timespamp of the last "bad" edit here)&uclimit=500] gives you a list. At the start of that is
<usercontribs ucstart="20070313155223" />
If you get this value, it will allow you to go to the next page of results by inputting that ucstart time in the query. I hope this works out - Martinp2320:09, 18 March 2007 (UTC)reply
Hi Martin. I am
User:RayTomes Advocate. He is interested in creating a Category:Cycles which has been deleted twice previously. I advised him to set it up again and I would back him. He has now been blocked by
User:Ruud Koot. I have requested that Ray be unblocked; however, Ruud is not cooperating. Your advice would be very welcome. Regards
SilkTork17:44, 18 March 2007 (UTC)reply
Hi - in policy, Ruud's block is sound, as RayTomes had re-created a deleted page twice. On the other hand, however, I'd say that one week could be a little excessive, especially without a recent warning. It would be wise to bring this matter up at
WP:ANI, and mention that you will advise Ray against re-creating the category in future, without favourable community consensus (by going to the Village Pump or something), as I would be unwilling to remove Ruud's block without contact from him first. Martinp2318:26, 18 March 2007 (UTC)reply
That was typed before I read the ANI post. To be honest, you need to tell Ray that consensus has spoken, and Wikipedia runs on consensus, so this cannot be unilaterally overturned. Think of your role as advocate not as someone who needs to get their advocee's way, through good or bad means, but as an adviser, who should tell their advocee what they're doing wrong. Thanks, Martinp2318:30, 18 March 2007 (UTC)reply
Yes. I am beginning to appreciate I have given Ray the wrong advise. I didn't research enough, and I didn't talk to enough people. Thanks.
SilkTork19:42, 18 March 2007 (UTC)reply
Bot message tone
Although I've seen bot messages for many moons, it struck me as inappropriate to leave a standard message for
this edit. As a CVU veteran I feel the message/warning shouldn't be apologetic in tone. -
RoyBoy80022:01, 18 March 2007 (UTC)reply
The problem is that AVB makes roughly 1 error per minute, and makes thousands of reverts per hour, and we don't want to chase away good users, the bot has no way of telling the difference--VectorPotentialTalk22:07, 18 March 2007 (UTC)reply
Beat me to it again, Vector :). RoyBoy - what Vector says is correct (though I'd like to think that the error frequency was much lower :)) - we need to keep the warnings at a low level for the first offence noted by the bot, and of course it can't tell how malicious an act of vandalism is (afterall, it is only a few lines of python). Martinp2322:12, 18 March 2007 (UTC)reply
That bot reverted my legitimate edit in Aaliyah's article.
bleach anime correction
hi,
i edited bleach anime's "bleach media and material's" page cuz the song for the fourth opening theme was "TONIGHT, TONIGHT, TONIGHT"
, not "TONIGHT TONIGHT TONIGHT" but the bot of yours penalized me...this bot scares me...=P —The preceding
unsigned comment was added by
69.231.134.158 (
talk)
04:09, 19 March 2007 (UTC).reply
The statement "If the bot reverted a legitimate edit, please accept my humble creator's apologies" will only serve to embolden this type of vandal, and lead to further ongoing vandalism. WP is not well served by encouraging blatant vandals. A human user reverting this vandalism would have left a message more like {{uw-vandalism3}} (or at the very least {{uw-vandalism2}}) for this particular offense.
Please consider modifying your bot's code to leave much stronger messages for these types of cases. I assume there is a lookup table of common obscenities and offensive phrases which your bot can use to identify blatant vandals, and then leave a more effective anti-vandalism message. The milder message should be saved for less blatant cases.
Except, that the bot also tends to revert people in cases where obscenities are appropriate, for instance, we have an article on
obscenities, virtually any edit to that page would set off the bot, do you really want to warn off people who are making good edits?--VectorPotentialTalk12:15, 19 March 2007 (UTC)reply
The number of articles on WP where obscenities are appropriate is miniscule, probably less than 1/100,000 of the articles for sure (< 20 on the entire en:WP). So the risk of giving out too strongly-worded a message in less than 0.001% of the cases is certainly an acceptable error rate, and much better than extending "humble apologies" to thousands of blatant, filthy-mouthed vandals. I know we should assume good faith, but once we see actual vandalism in progress (usually on multiple articles in a row) that has to stop, and stern warnings (or blocks) are needed. That's just how I feel, thanks for your response. --
Seattle Skier(talk)12:40, 19 March 2007 (UTC)reply
This probably won't ever happen. Your statement above "blatant, filthy-mouthed vandals" is would be a dire abuse of
WP:AGF and civility, which we must show even towards vandals. The risk of driving a way a good faith contributor (even one) is too insurrmountable to allow us to change the message, when there's no evidence thawt the current message even remotely encourages vandals to continue. Sorry :) Martinp2312:52, 19 March 2007 (UTC)reply
Martin, you are welcome to run your bot as you wish, but please don't accuse me of a dire abuse of
WP:AGF, which currently states the following: "This guideline does not require that editors continue to assume good faith in the presence of evidence to the contrary. Actions inconsistent with good faith include constant vandalism." So I am in full compliance with both the letter and the spirit of that policy. Please read the WP policies next time before wrongly accusing others of violating them. My words also do not cross the line of violating
WP:CIV, which states that "incivility is roughly defined as personally targeted behavior that causes an atmosphere of greater conflict and stress." I have not personally abused any vandals (ever), and the {{uw-vandalism}} templates I suggested do not violate civility either. I have also been very calm and reasonable in my messages posted above, and have thanked you for your efforts in maintaining this useful bot. Still, extending "humble apologies" to vandals is too much in my opinion; I wish you would reconsider the phrasing of that sentence, even if you have no desire to place appropriate and justifiable anti-vandalism templates. Thanks. --
Seattle Skier(talk)15:37, 19 March 2007 (UTC)reply
Sorry about that misunderstanding. I meant for the comment to be interpreted that if the bot were to treat everyone as you just categorised vandals, we would be likely to drive away a nimber of potential contributors. I have no problem with your expression on my talk -page here (and, indeed, I often share it), but to apply that sort of logic to the bot messages would lead to a huge number of problems and further complaints (even if we look for profane words and phrases, they often have a place in many of our articles (maybe as high as 1/1000), so we could have problems). My honest feeling is that the uw warnings would probably be more appropriate, especially given the annoying header that AVB seems to add to warnings, and will be looking at implementing something of the kind in the future, though it is important, in my mind, to make the distinction between this being a bot editing compared to normal human edits - I hope you see what I'm getting at here, and with the bot being much more prone to error, some sort of recognition of this is appropriate (perhaps drop the "humble", saying something like "please accept my creator's apologies", with some wording before that to say when it applies. I'm happy for any suggestions, but as I've said, it is impossible for me to make the bot give a different warning based on the context of the vandalism. Martinp2317:12, 19 March 2007 (UTC)reply
OK, thank you very much for clearing that up, since I did think you were directing that statement at me, and so I reacted strongly. And thanks also for clarifying some of the complications involved in running a bot and the extra caution needed in placing warnings, versus a human editor. I would be happy if you dropped the word "humble", since that's the one word in the whole message that really caught my eye as being far too apologetic, and led me to post here. Well, goodbye for now,
Seattle Skier(talk)21:28, 19 March 2007 (UTC)reply
Certainly - I've definately been meaning to do so over the past few days, and will take a look tonight - as you may have seen from above, I've been quite swamped with work :) Martinp2312:53, 19 March 2007 (UTC)reply
Hi, I'm not sure if this is the right way to do this, but can I make a request for blocking of the IP Address 65.39.114.238. Your bot has reverted an edit by this address to
Aztec society and the address already has numerous other vandalism instances (take a look:
[3]). Thanks for the good work done by the bot!
Shrub of power20:06, 21 March 2007 (UTC)reply
Lines got croosed, I see :). If you feel that an IP needs blocking, the best place to report is to
WP:AIV when the vandalism is in process. As the user hasn't had the appropriate full range of warnings, I regret that I can't block on sight.
Fixed - note that you yourself cannot delete and article by removing the content from it. You need to ask an admin to do it - see instructions
here. Martinp2322:29, 22 March 2007 (UTC)reply
We've opened up an official review period to attempt to add you to BAG
here. It will end in 1 week (give or take). --
RM02:02, 22 March 2007 (UTC)reply
Hello, MDP23! Thank you for reverting
vandalism to Wikipedia. After you revert, I would recommend also warning the users whose edits you revert on their talk pages with an appropriate
template or custom message. This will serve to direct new users towards the
sandbox, educate them about Wikipedia, and a stern warning to a vandal may prevent him or her from vandalizing again. Thanks! zero»03:30, 22 March 2007 (UTC)reply
I know :) - as you may be aware, one somtimes doesn't have the time to apply the appropriate warning, especially when (as is often the case for me) you spot a bit of vandlaism on your watchlist just before you go out! Can I ask exactly which edit you were referring to? I have, as you may know, knowlegde of the appropriate warning guidelines, but sometimes it can be impractical to follow them, though I do try to adhere. The thing is - the only vandalism I have recently reverted is to MartinBot's user page, and I tend to block when that occurs (personal attacks). When I'm actively on RCPatrol, I always warn religiously, however if I'm making a casual revert, it's often difficult (in any case, admins should look through the user's contribs before blocking - not just the vandalism warnings (from experience)). Martinp2322:08, 22 March 2007 (UTC)reply
I know this is probably small but on the Help desk, it placed the March 21st almost at the end of the days so iot only captured 2 comments, before moving onto March 22nd so most of the March 21st comments are in the March 20th archive.
Simply south17:31, 22 March 2007 (UTC)reply
Toolserver was down at 0:00 (UTC) when the bot usually runs, so it was switched on in the middle of the next day, hence the discrepancy in the date headlines, it was a minor issue and had been corrected by hand on all of the reference desks, only I forgot all about the help desk. It has now been corrected as well--VectorPotentialTalk17:56, 22 March 2007 (UTC)reply
Your bot made an incorrect revert of my edit of
Chord Progression. I was slightly changing the example of a 12-bar blues chord progression to agree with the
Two-Bar-Blues article and to be correct. thought i'd bring it to your attention. I'm User: Yakob, if you have more questions. I'm kind of lazy about logging in. —The preceding
unsigned comment was added by
71.229.243.243 (
talk)
01:53, 23 March 2007 (UTC).reply
There is a link error to your bot's warnings. This portion:
"...Click here for frequently asked questions about the bot and this warning... "
links to Tawkerbot2/FAQ instead of User:MartinBot/FAQ (you should have created that subpage from your bot.) I believe you must have copied AntiVandalBot's source from
yours. zero»03:29, 23 March 2007 (UTC)reply
Yes - this is quite intentional and is by no means an error :). TB2 no longer runs, with its responsibilities now being moved to
AVB and MartinBot, who run the same software as it. As the software is the same, it srickes me a perfectly appropriate that the FAQ page in in the same place - do you not agree? Martinp2317:52, 23 March 2007 (UTC)reply
By the way, it is in my "plan" to move that TB2 FAQ to MartinBot's space, and chagne the name references, but it's time that's the issue :) Martinp2321:10, 23 March 2007 (UTC)reply
I proudly award you this barnstar for the ultimate anti-vandalbot. Martinbot. For creating such a fast, accurate and hardworking program, which wastes the vandals time and keeps wikipedia going. Well done. Certainly catches and reverts faster than me!
Retiono Virginian17:01, 24 March 2007 (UTC)reply
Is there any document which exists which discusses how to interact with the IRC channels in Wikipedia? I did some searching but didn't find anything. --
D00:31, 23 March 2007 (UTC)reply
Not that I know of, I'm afraid not. It may be worth asking around on the various IRC channels (eg #wikimedia-toolserver perhaps) for advice. If you understand C#, I can send you some code to look at for an idea of how to do it. Martinp2319:08, 25 March 2007 (UTC)reply
OK - it seems to have got in a strange sort of edit conflict (because of the way it works). You were perfectly correct to remvoe the warning :). Thanks, Martinp2319:09, 25 March 2007 (UTC)reply
Thanks for sorting it - it seems that some of the date headers were in the incorrect format, hence the bot grabbed far too much text. Martinp2319:10, 25 March 2007 (UTC)reply
I would like to thank you for your support in my recent RFA. As you may or may not be aware, it passed with approximately 99% support. I ensure you that I will use the tools well, and if I ever disappoint you, I am
open to recall. If you ever need anything, don't hesitate to leave me a note on my talkpage. Thanks again, ^demon[omg plz]20:24, 25 March 2007 (UTC)reply
Now that you're an AWB developer (and BAG member, gosh what a busy week!), any interest in adding the wiki tools you've written in C# (or any dotnet language) to the AWB SVN repository? We might be able to recycle or integrate code, help out with each other's tools etc. Just a thought. --
kingboyk14:54, 26 March 2007 (UTC)reply
Very busy week! I'll take a look at what's useful in my code and add to SVN either tomorrow, Weds, or next week :) (and even busier week IRL coming up!). I have some concerns about releasing the code of NPWatcher for now, but the issues should (hopefully) die down soon. Thanks, Martinp2318:24, 26 March 2007 (UTC)reply
Hey Martin, I've tweaked the {{DRV top}} template in a way that it automatically adds a (closed) marker behind the title in the subect line. This might make the job easier for the bot to recognize when all discussions ar closed. It's diescribed in detail at the bottom of
WT:DRV. Take care,
trialsanderrors09:31, 27 March 2007 (UTC)reply
Hi - sorry for the time it's taking with this bot - I got a little bogged down in real life :). I'll get it up and running on my return - thanks, Martinp2323:30, 28 March 2007 (UTC)reply
Do you have any knowledge of/experience with VB.net? I have a resolute bug or two in my plugin's logging feature, and could do with a second pair of eyes looking at it. --
kingboyk23:56, 27 March 2007 (UTC)reply
I may not be of much use, I'm afraid. I can find my way around VB.NET, but C# is my true love, and the lack of { } in VB.NET scares me :). I will, however, be happy to take a look on my return, if the problem persists. Martinp2323:26, 28 March 2007 (UTC)reply
I will be happy to help you on my return, however for the next week or so it might be good for you to try out
AutoWikiBrowser in manual mode for things like typo fixing, etc. Thanks, Martinp2323:27, 28 March 2007 (UTC)reply
MartinBot source
Hi! I'm wondering if MartinBot's source is available, and can it be customized to work on other wiki project? I'm running a bot on toolserver too, and that would be used as AntiVandalBot for bswiki. --
Emx21:12, 28 March 2007 (UTC)reply
Hi - at some point I'll be working on internationalising MartinBot's source, and would welcome your help if you can offer it! I'll get in contact on my return, but feel free to leave messages here/drop me an email. Thanks, Martinp2323:29, 28 March 2007 (UTC)reply
I have a request to make regarding MartinBot, and how it works with pgkbot on #vandalism-en-wp. Would it remotely possible to not display diffs on #vandalism-en-wp that MartinBot is in the middle of reverting? This would be a bit difficult if pgkbot's source is not open, although the theoretical possibility of it is exciting—mainly in that more subtle, and truly damaging, vandalism can be caught. Then again, we can have a channel for MartinBot to report IPs so that other people can warn the users with an appropriate level. It just gets complicated, though. MartinBot... kind of screws with normal vandal-fighting. I can't say whether its net effect is positive or negative. (I know it's just a clone, but no matter.) So, would this synchronization be possible?
GracenotesT §
01:47, 29 March 2007 (UTC)reply
For the time being, MartinBot has effectively taken over from AVB. The best person to ask about this is pgk, who is (obviously) more familiar with the pgkbot code than I, but I will look into it on my return (I'm editing from the airport now). Martinp2306:27, 29 March 2007 (UTC)reply
re: meadiation
yes i am aweare of that and no offense taken. i applied for modreatiorship (i know thats not a word) and it says
Currently unassigned mediation requests can be found at: Wikipedia:Requests for mediation/Open Tasks. Applicants are encouraged to select an unassigned mediation and volunteer to mediate for it; this allows the committee to review the candidate's performance, and provides the candidate an opportunity to experience mediation firsthand. —The preceding
unsigned comment was added by
Razor romance (
talk •
contribs)
14:17, 29 March 2007 (UTC).reply
MartinBot
Hi there, your bot (Martinbot) recently reverted my edit to
Subhoodz, it was a copyright violation which had not been blanked but had a tag on it, I blanked the page per
WP:CSD and said in my edit summary that I had blanked it in good reason because it was a copyvio, the bot reverted it and warned me and I'm not a vandal so maybe you could set the bot so that it does not revert edits if they are marked with {{copyvio}} as it was a legitamite edit and do you mind if I remove the warning and explain it there? Cheers! Tellyaddict16:25, 29 March 2007 (UTC)reply
Please spread the word about Wikipedia Weekly, we're trying to spread the word so that people know about the project, we've got some cool guests lined up and it makes it much more fun if people tune in! Feel free to post to the mailing lists too.... apparently not many people know about us.... yet
You are receiving this message because you are listed on
Wikipedia:WikiProject WikipediaWeekly/delivery - if you do not wish to receive such notifications please remove yourself from the list.
Could the warning added to user talk pages be a little more like the level 1 user warnings, especially when it is the first thing on a user talk page (ie when creating the talk page)? It would be even better if it automatically determined a level of warning to use.
That way we humans can more easily decide to add a true user warning template, or not, directly underneath the bot's warning.
Mark Hurd02:03, 2 April 2007 (UTC)reply
For fixing up where I seem to have stopped half way through a bot approval, leaving the archival and listing stages. Quite a bit on my mind at the moment :) Martinp2321:55, 4 April 2007 (UTC)reply
I noticed RefDeskBot hasn't archived the Help Desk in several days, the page is starting to push 338 kilobytes--VectorPotentialTalk16:58, 5 April 2007 (UTC)reply
I've just noticed the same thing (in fact, drawn by your edit summary referencing RDB :)). I'll try to get it back up to date tonight (UTC) or tomorrow. Thanks, Martinp2317:02, 5 April 2007 (UTC)reply
Although the current (release/stable) version doesn't use the ExpandTemplate feature for user warnings automation, it still does (by default) for all substs, so the error could still be prevalent... What do you think? I have no idea how many/few people use the subst feature, but the fact that ExpandTemplate is the default option does concern me. Thanks, Martinp2322:46, 5 April 2007 (UTC)reply
You think we should release a new version, you mean? I'm fine with that, but I'm not sure if others are ready (i.e. they might be half-way done with some features). Probably better to ask at
WP:AWB/DEV. —METS501 (
talk)01:57, 6 April 2007 (UTC)reply
It looks like RefDeskBot lost the March 29/30 sections of the Science refdesk while archiving it. In
this edit dated 00:01, 5 April 2007 RefDeskBot removes the sections for March 29, March 30 and April 3 (the sections for April 1 and April 2 had been previously archived for some reason), but
only creates an archive for April 3. I've no idea what might've caused it, but I figured you'd like to know (and maybe might be able to fix it more easily than by creating the archives manually). —
Ilmari Karonen (
talk)
18:47, 6 April 2007 (UTC)reply
It doesn't seem like RefDeskBot was malfunctioning, April 1st through April 3rd all seem to be archived properly, and the correct number of days seem to be transcluded to that desk. Additionally all archives for the month of April seem to be accounted for--VectorPotentialTalk19:13, 6 April 2007 (UTC)reply
This does seem strange - I suspect that it missed the March 31 archival, so didn't put a header down (or the header went missing at some point), and so it was unable to archive properly for a couple of days later. This seems to be the bot's way of fixing problems like this, but this can be (and is) a pain to fix. Martinp2319:20, 6 April 2007 (UTC)reply
Correction, it seems to only function for months preceding March, but produces totally random results when used on dates after that. I can't begin to explain how that's possible. It's probably best to run RefDeskBot as is, and see what happens--VectorPotentialTalk19:32, 6 April 2007 (UTC)reply
Hopefully the HTML Tidy issue will resolve itself, making further changes to the template unnecessary--VectorPotentialTalk19:54, 6 April 2007 (UTC)reply
I need some advice
hey! I probably would've filed a regular request; but the issue is, well it's about enforcing fundamental wikipedia policies/guidelines in a wikipedia in another language.
I probably should be filing requests or whatever on that particular wikipedia, but the place is legally void (or something).
I need some counseling.
ok the background is as follows. (you can just skip it)
the language uses latin alphabet, but there's this rule that proper nouns should be transliterated (but there is no clear, standardized, general definition to what extent or how it is done so there is a large number of usually dubious systems or actually just lists) (if the language used IPA characters, everything would be near perfect, but it's doesn't)
I made this wonderful inclusionist proposal-to make this sort of huge reference list of those proper nouns that are transliterated without any ambiguity for reference when someone's invented a new transliteration because he'd know better! and all other more obscure designations (that you can find no reference about in local literature) would be written as they are originally leaving no place for ambiguity.
from a total of ~10 admins, i guess, two engaded in discussion (if you call that a discussion) one proposed that we (editors in that particular wiki) invent some sort of our transliteration for the ISO tables and the second one ruled that spelling original is 100% unacceptable (he didn't source it and frankly i think he couldn't) and proposed that we replicate the pronancation, as we feel it should be done. he didn't say anything in response to my arguments that everything must be backed-up with literature (even if it's in a foreign language as it usually is) or that you don't use wikipedia to publish your thoughts .
then i illustrated my proposal with examples then he repeated what he had already said that we should invent article headlines ourselves.
he didn't source a single one of his claims and in my article on
Kyūshū now there are at least 3 diff. spellings of the word all over the text
[7] (thats just an example).
I'm afraid there might be problems in his (their) willingness to come to a consensus. as well I'm concerned that some people there might perceive adminship as some sort of irrevocable status or whatever, granting them right to reject other peoples opinions on the basis of the unquestionable righteousness of their possibly biased (unsourced) views.
So the question is actually where do go if admins in charge of a local wikipedia abuse their privileges? or just refuse to engage in a discussion bouncing off everything I say with unsourced claims what they think is right. I'd be really glad if you could give me some advice on how are proposals made or how is admin misconduct treated on local wikipedias or refer me somewhere
okyea
actually I think I just changed my stance on this issue,
[8]
(it's a comment someone made regarding wikipedias in smaller languages, his point is that encuclopedicness in smaller language wikipedias is not that important which I actually totally agree with, a good point on this issue)
okyea
Bug with the Martinbot inadvertently restored a vandalism of Batman... page is correct now but reporting the bug. Please leave a message on my talk page if you need more details.
Sorry for the mistake on the bot's part - that are occasionally made, but it is impractical to try to make the bot perfect :) (would take hours to check each diff!). Martinp2323:24, 8 April 2007 (UTC)reply
New Article Bot
Hi, Martin:
Sorry for being dense but can you point out where the bot was discontinued from the trials of NPP? I was under impression that
my proposal for limiting bot's functionality was not opposed by anybody.
Is refdeskbot still archiving the help desk? According to the bot's contribs, it doesn't look like it archived it since March 30.
Mr.Z-mantalk¢23:09, 6 April 2007 (UTC)reply
On the MartinBotIII page it says "Substs or deletes templates based on consensus at WP:TFD". Does that mean this is/will be an admin bot? --
TeckWizParlateContribs@(Lets go Yankees!)
18:17, 7 April 2007 (UTC)reply
No :) - it just removes the template instances from pages, then an admin deletes. My wording there is definately dodgy, however I'm not sure how to express it better (and keep it short). Martinp2318:25, 7 April 2007 (UTC)reply
I think what you just said was short enough. I see you just got your bot approved. I was thinking of expanding my bot to cover the same things you just got approved for. I would need another approval, but how do you use the page
WP:SUBST? I would guess you use the "Links on page" in AWB to make the list, but the page also has the templates not to subst. --
TeckWizParlateContribs@(Lets go Yankees!)
18:31, 7 April 2007 (UTC)reply
The only way I've found to do it is to maunally build the list by repeatedly seaching for transclusions of the different templates I want to subst, and then filter namespaces and remove duplicates. I'd like to find a better way of doing it, but have no idea... Martinp2319:23, 7 April 2007 (UTC)reply
Really an amazing tool. I discovered one minor bug: it (sometimes?) doesn't place line breaks between successive talk page warnings,
[9] which leaves them jumbled together in one paragraph. Cheers,
Postdlf00:44, 8 April 2007 (UTC)reply
My bot was clearly approved for trial by another BAG member, and you just on your own initiative chose to flatly deny the request? How exactly does that work? --Rschen7754 (
talk -
contribs)
23:01, 8 April 2007 (UTC)reply
It was a duplicate request, as noted there. Unless there's a compelling reason for these two bots to do the same task, the request shall, for good reason, remain denied. I have already explained the technical position on this to Vishwin60, who's bot is capable of carrying out the whole task on its own - more than one bot just creates problems and wastes time. The only reason I didn't reject Vishwin's request was that it had been filed before yours, so common sense prevailed. As the BAG member currently on your request has gone on Wikibreak, I saw no reason not to close it in the same manner as I'm sure he would, if he had been here. Thanks, Martinp2323:07, 8 April 2007 (UTC)reply
SatyrBot and LocateMe
As you may have seen on my talk page and on the bot's talk page, it is no longer involved in that project and won't be. As for WP:EA, I will be sure to go through the Bot Approval process before starting any work on that, assuming there's community consensus to make the change. Thanks for your comments and checking on it. -- SatyrTN (
talk |
contribs)22:44, 9 April 2007 (UTC)reply
Special note to spamlist users: Apologies for the formatting issues in previous issues. This only recently became a problem due to a change in
HTML Tidy; however, I am to blame on this issue. Sorry, and all messages from this one forward should be fine (I hope!) -Ral315
Only a few of these missing dates have been archived (by hand?). If you look at
Special:Contributions/RefDeskBot, it is apparent that there were problems; on problematic days either the full cycle of desks was not completed, or there were fewer than 3 edits per desk, or both. Perhaps you should take a peek at that page every now and then to make sure all is up and well. Also – I realize that is a lot of work – perhaps the bot could be made more robust; in particular, ideally it should check each time it is run when the last successfully completed action was, and perform its duty for all later days, not just the current one. --
LambiamTalk13:34, 10 April 2007 (UTC)reply
I see that it missed out running on the 9th, and so couldn't do the 10th - this is something I'll rectify tonight, bringing the bot back up to date. Now, on the issue of missing archives - these are not usually caused by the bot's actions - for example, there have been occurances of users removing the date headers, which screws things up. Similarly, if auser does a manual archive of a day in the middle of the desk, leaving the transclusion in place, the bot is broken, as it can't find the date header. Basically, all that the bot needs is the right number of transclusions at the top of the desk, date headers intact and in the correct format and the ability to edit. When these things go missing, bad stuff happens, as is the case with *any* such Wikipedia bot which runs off an area where users edit. Martinp2313:48, 10 April 2007 (UTC)reply
The RefDeskBot, in producing archive indices, turns a topic header like "==Password retrival==" into "assword retrival", throwing away the first significant character of the topic title; see e.g. #56
here.
It also picks up header-like strings in the middle of a line; see e.g. #27
here. The last example shows that the bot also disregards "<nowiki>" markup.
Using vi/grep syntax, a topic header must be of the form
^==\([^=].*\)==[ ]*$
in which leading and trailing spaces in the part matching \(...\) still must be stripped off to get the topic title. So the line must start with exactly two equal signs" "==", and must end with two or more equal signs, possibly followed by spaces. (Disclaimer: I have determined this experimentally; there may be subtleties in the wikimedia algorithms that I did not spot.) --
LambiamTalk08:27, 11 April 2007 (UTC)reply
Error: invalid time
I manually fixed many links botched up by "Error: invalid time" messages in archive headers (e.g.
this one), and many more problems. Apart from doing something about the errors, wouldn't it be a good idea if RefDeskBot produced a log of all problems and anomalous situations it encounters, for example on a page
User:RefDeskBot/Log, so that we can all inspect this? --
LambiamTalk07:16, 11 April 2007 (UTC)reply
It isn't RefDeskBot doing that - it's bad template syntax I'm afraid (or the syntax has recently been changed, and the bot doesn't know) - I will check. Martinp2309:05, 11 April 2007 (UTC)reply
{{HD Archive header}} hasn't been edited in over a month, so if there has been a change, it's been with the mediawiki syntax itself, not the template. Also, for whatever reason, it's now, as of the most recent HD archive, producing the correct date header, so it there was a problem, it's solved itself--VectorPotentialTalk01:34, 12 April 2007 (UTC)reply
I was the one who did the COI warning. Sorry I didn't catch the egregious edits - at the time I gave the COI warning, he had only done his first edit, which changed year of birth and the educational institution. They were factual but unsourced, for all I knew. Wish now that I had come back to visit a few minutes later. Oh, well. I'm wondering now if my COI warning prompted the second, more blatant false edit. Cheers,
RJASE1Talk01:27, 12 April 2007 (UTC)reply
AntiVandalbot was at one time also monitoring this page, but now it is not. Can you have it start monitoring the page again? Or, if it is not too much trouble, extract out the code for a seperate bot to monitor this page (and possibly
Wikipedia:Introduction 2 and
Wikipedia:Introduction 3.
User:Tawker left me a note saying that Martinbot should be handling the Intro text now - it doesn't seem to be working - can you take a look at it? TIA --
Trödel14:15, 3 April 2007 (UTC)reply
Hmm - it should be, but I suspect that there's a line of code commented out somewhere which I need to fix. I'll get it resolved ASAP. Martinp2323:17, 8 April 2007 (UTC)reply
I've been looking in MartinBot's code, and it should be working (but clearly isn't). I'll write another bot to do it now. Martinp2311:37, 12 April 2007 (UTC)reply
Probably a good idea. With some of the things people write there, a clean slate is often best. Maybe a time-based reset would be good too? Every 30 minutes or so, per Sandbot's old setup? --
Quiddity02:09, 13 April 2007 (UTC)reply
Agree - thank you very very much! Good idea re reset when the header is removed - too often junk ends up staying for a while. --
Trödel10:43, 14 April 2007 (UTC)reply
Sorry that it took so long! The bot is resting every 30 mins now, because it could be a while until the header text was vandalised. It'll (as of now) also rest whenever the header is removed, so we effectively have two reset methods now :) Martinp2311:19, 14 April 2007 (UTC)reply
'tis much appreciated! I can finally unwatchlist them again (If only we could watchlist just the talkpage, in a case like this!). Thanks again. --
Quiddity19:05, 14 April 2007 (UTC)reply
Thanks again! We're all busy and understand things take time in a volunteer organization
Dispute needs to be resolved on the Buffalo Soldier article
In reference to the
Buffalo Soldier article, the user
Beetstra has a bot called
Shadowbot which is automatically deleting external links which I am trying to provide on the site for additional credible sources of information. The user
Beetstra refuses to add the members.aol link
The Buffalo Soldier Story which is featured on the official Fort Leavenworth Buffalo Soldier Monument website:
http://garrison.leavenworth.army.mil/sites/about/Buffalo.asp therefore it should be credible to add on the Wiki site since the logic that the U.S. Army has approved it for their external link it should be on the Wiki site as well.
In addition this user has tried to add a linkfarm on the site, where all of the articles are valid and credible sources of information in reference to the article. --
Signaleer18:20, 16 April 2007 (UTC)reply
I'm afraid that Beetstra is quite correct in the removal of the aol link, per points one and eleven of "Links normally to be avoided", in
Wikipedia's policy on external links. The correct use of {{linkfarm}} is really editorial judgement, but it is suggested that it be used wherever there are a high number of external links. Links in a collection of the size shown at
Buffalo Soldier often duplicate content both between themselves and the article, so it is almost always preferable to trim the links section to just a few, factual, expert links. I hope that I've been able to make the policy clear to you, Martinp2318:34, 16 April 2007 (UTC)reply