I remain active to answers queries related to CSBot, which I intent to keep operational as long as it is still useful and welcome.

I have changed the copyproblems under "Pukeuria" by replacing it with a link.

Turbonilla (talk) 15:41, 17 July 2009 (UTC)Turbonilla 17 Juli 2009[reply]

I copied the album track listing from another site, plus some text now deleted. It seems that the bot interprets a lift of a track listing as being inappropriate, yet it can't be--at least, I don't think so. I have removed the caution accordingly, which I hope is OK.

Dreadarthur (talk) 16:59, 16 July 2009 (UTC)[reply]

I've discovered an unpleasant form of vandalism, which I documented at WP:BOTREQ#Forged Billboard Links, but I'm getting no takers on building a bot to help me search it out. It struck me that CorenSearchBot might have the skeleton of similar code (i.e. performing a Google search based on text extracted from a Wikipage), and, if that's true, your mentioning it there might help inspire some eager young bot writer to borrow your code and build it for me. Basically, Billboard has a bug which allows you to move record chart positions around: if a song reached a position on any chart, vandals can massage the link to make it appear that it reached that position on any chart they want on any date they want. I have no idea how prevalent it is: I only noticed because they made a truly bizarre claim (a Latin pop star charting on the "Vietnam Hot 100"), and it was so bizarre that I didn't trust the chart position being served up, despite having "Billboard" stamped all over it.—Kww(talk) 03:11, 17 July 2009 (UTC)[reply]

Hmm. I'm not sure what a bot can do to help, from what I understand of the problem. Do you have a specific method in mind to figure out such constructed links from the genuine thing? — Coren ^(talk) 01:35, 19 July 2009 (UTC)[reply]

Perhaps I'm not being clear enough, and that's the reason I'm not getting responses

• use http://en.wikipedia.org/w/api.php?action=query&list=exturlusage&eunamespace=0&eulimit=max&euquery=*.www.billboard.com&format=yamlfm to get a list of links to Billboard.com
• For each link retrieved

• Is it of form http://www.billboard.com/bbcom/esearch/chart_display.jsp?cfi=n&cfgn=Singles or Albums&cfn=namestring&ci=n&cdi=n&cid=datestring

• extract ci=n field from link.
• search billboard.com for links including that string

• If no hits, then log as unverifiable link
• If hit, then compare cid and cfn fields

• if mismatch, log as forgery
• if all match, then link is good

As an optimization, it could keep a database of verified links, so that on subsequent runs it wouldn't have to query Google every time.—Kww(talk) 13:57, 19 July 2009 (UTC)[reply]

Some context that might help: it's the "ci=" field that identifies a unique chart. Every week that a new "Canadian Hot 100" is created, a new chart id is created, and all links to that chart will include that field, encoded as "ci=". The charts name should be the same in each link, thus "cfn=" should be the same for every link. The data should be the same in each link, thus "cid=" should be the same in each link. For correct referencing, people do a search on a song title, artist, or chart, and Billboard's server manufactures the link for them. Unfortunately, Billboard's server doesn't have an interface to say what the correct values of the various fields are for a given chart: they assume that once they give you the link, you will use it without tampering with it. What I'm proposing is that we search through Billboard's news stories and find their internal links to various charts, on the assumption that they don't vandalize their own links internal to their site. Random testing shows me that for about 75% of links that I check, I can find an internal Billboard link to verify it against.—Kww(talk) 14:56, 19 July 2009 (UTC)[reply]

Hi. CorenSearchBot has flagged HSD3B7 as a possible copyright violation. Some of the material in this article was taken from "Entrez Gene: HSD3B7". This material was attributed to this source and in addition the {{NLM content}} template was added to the article. Therefore I believe the warning was a false positive. Cheers. Boghog2 (talk) 21:16, 17 July 2009 (UTC)[reply]

recreating non-infringing elements of article.00pj (talk) 07:28, 17 July 2009 (UTC)[reply]

"If CorenSearchBot is in error: Simply note so on this article's discussion page" No I'd rather do it here.

I am in the process of removed a lot of incompatible copyleft material See User:Philip Baird Shearer/BCWs copyright issues

A simple read of the page Henry Mildmay shows that it can not possibly be copyright material. Further it has on the page both citations and in the References section:

Attribution

{{DNB}} "This article incorporates text from the Dictionary of National Biography (1885–1900), a publication now in the public domain."

Further the page it compares is http://www.pepysdiary.com/p/3799.php has written on it "This text was last fetched from this Wikipedia page (where you can edit it) on 19 Jul 2009, 9:07pm under the terms of the GFDL."

So why is the dumb bot putting a template on the top of the page? --PBS (talk) 21:50, 19 July 2009 (UTC)[reply]

Please remove the template from the top of Henry Mildmay. --PBS (talk) 21:52, 19 July 2009 (UTC)[reply]

You need to fix this. There is no copyright violation for this article. The info was taken from the findagrave website (bith/death dates and burial location) and the official citation on the Army Medal of Honor website per the inline citation. Additionally, stating that the information contained on the Home of Heroes website is in copyright violation is incorrect. They display the actual citation from of the Medal of Honor and that is not in copyright from them or anyone else. It is freely distributable as being a work of the US governent and I recommend you add it as an exception so I will stop getting these error messages. I don't mean to seem rude but I am getting a little tired of the bot leaving these messages for the Home of Heroes website when the Home of Heroes website, regardless of what they may claim, does not hold the patent or copyright for this info. Please add the home of heroes website as an exception so that it will not continue to generate. There is no info on the home of heroes site that is not already on the AMOH site. All the bot is doing in this case is wasting time for me, it and you. Thanks.--Kumioko (talk) 23:04, 19 July 2009 (UTC)[reply]