User talk:Lustiger seth

See /archive001 for old threads.

Spam search

es

Hi!, I'd like to report you a possible bug on your tool. When I try to search on wp-es this message apppears:

« Content-type: text/html

got an error: not a valid file: http://es.wikipedia.org/wiki/MediaWiki:Spam-blacklist »

Best regards Dferg ^(T-ES) 12:14, 3 January 2009 (UTC)

thx! should be fixed now.

reason was: es-wiki doesn't use <pre>, but uses <pre class="..."> in its lists. -- seth 16:21, 3 January 2009 (UTC)

Thank you very much Dferg ^(T-ES) 21:13, 3 January 2009 (UTC)

Sorry, the tool still doesn't work. Dferg ^(T-ES) 11:50, 8 January 2009 (UTC)

Thanks again. You are right, every entry below -nude\.blogspot\.com is ignored. I'll have a look at that in a few hours. -- seth 12:25, 8 January 2009 (UTC)

es:MediaWiki talk:Spam-blacklist#wrong_syntax. I could write a work-around. But I guess, a simple revert is much more comfortable for all. -- seth 15:45, 8 January 2009 (UTC)

 Done Dferg ^(T-ES) 16:12, 8 January 2009 (UTC)

ru

I got an error trying to check ruwiki:

got an error: Reference to nonexistent group in regex; marked by <-- HERE in m/https?://+[a-z0-9_.-]*(?:\1 <-- HERE host\.in/st\.exe)/ at ./grep_regexp_from_url.cgi line 280.

 — Mike.lifeguard | ^@en.wb 00:25, 8 January 2009 (UTC)

[1] -- seth 10:37, 8 January 2009 (UTC)

default=meta

Any chance you can change the default to check only Meta? Normally that's all the data we need - one can of course request further data from the tool as required. Thanks.  — Mike.lifeguard | ^@en.wb 02:07, 12 January 2009 (UTC)

At the moment the default is the only way to get some info about XBotLink. SO I better attach XBotLink to one of the languages/projects. Should I add it to "meta" or to "w:en"? -- seth 09:55, 12 January 2009 (UTC)

It's part of enwiki, I'd say.  — Mike.lifeguard | ^@en.wb 05:46, 29 January 2009 (UTC)

done. -- seth 14:29, 29 January 2009 (UTC)

caching

Another: Can you have the tool cache any pages it fetches live (when a user requests a "purge")? Then, when replicated data is older than the cached version, you can use the cached data (and still offer to fetch newer data). When replicated data is newer than the cache, throw it away, since it's useless. This avoids fetching live data on Meta's spam blacklist multiple times unnecessarily.

Just thinking now... the toolserver replicated database doesn't contain page text, I thought... so where are you getting the contents of the page?  — Mike.lifeguard | ^@en.wb 20:43, 9 February 2009 (UTC)

I'm caching all data already. I use the mirror function of LWP::UserAgent, see [2] (search for "mirror").

I'm using the SBL-urls like http://meta.wikimedia.org/wiki/Spam_blacklist. -- seth 02:27, 10 February 2009 (UTC)

trim

Could you have it trim leading/trailing whitespace from input?  — Mike.lifeguard | ^@en.wb 15:56, 20 March 2009 (UTC)

done. -- seth 02:10, 21 March 2009 (UTC)

Encoding

When checking your tool against ruwiki ([3]), I get something like "спам в ст. Аска Лэнгли Сорью, сайт содержит материалы, нарушающие АП. altes" - could you double-check you're reading the page text with the right encoding?  — Mike.lifeguard | ^@en.wb 23:59, 31 March 2009 (UTC)

Should be fixed now:

print $cgi->header();

→ print $cgi->header(-charset=>'utf-8');

-- seth 21:51, 1 April 2009 (UTC)

proper escaping?

At [4], it seems there may be missing escaping or something - the regex fragment isn't grabbed from the blacklist properly (or isn't shown properly at any rate).  — Mike.lifeguard | ^@en.wb 19:56, 2 June 2009 (UTC)

should be fixed now. -- seth 20:38, 4 June 2009 (UTC)

XLinkBot log not defined

While your spamlists search does include listings on http://en.wikipedia.org/wiki/User:XLinkBot/RevertList, it does not show the log entries. I get this;

"no log defined for this project. ask there if you want this script to use the log. "

The log is located here. I don't know what we did before the spamlists search tool. I've placed a Note on XLinkBot here. thanks seth.--Hu12 15:42, 28 August 2009 (UTC)

thx for that hint. i'll cope with that in perhaps 5 days, because i'll be away from my code for that time. meanwhile perhaps you could "clean up" the log so that it looks similar to the other logs. otherwise i'll do that, too. -- seth 18:12, 28 August 2009 (UTC)

Cleaned up the log...If more need to be done let me know. thanks--Hu12 16:36, 29 August 2009 (UTC)

thx! should be fixed now. -- seth 20:15, 2 September 2009 (UTC)

sbl

User:COIBot/XWiki/mountainzones.com

Could you take a look - affects dewiki mainly.  — Mike.lifeguard | ^@en.wb 00:08, 23 March 2009 (UTC)

Yes, it's already been discussed at de:WP:SBL#mountainzones.com. It's not considered as spam at de-wiki. Well, actually I considered it to be spam, but some reputable users requested un-blacklisting. -- seth 12:57, 27 March 2009 (UTC)

Blacklisting

Dear Lustiger Seth,

It was brought to our attention that all the references to our websites have vanished from Wikipedia. A short check has shown that our domain, 888.com, was blacklisted due to conflict of interest issues. We wish to emphasize that we never had any intension of doing something which is against Wikipedia’s guidelines. We are a worldwide, leading company in our field, which has been operating for over a decade. We will appreciate any feedback that you could give us, as well as steps we need to take in order to improve our status and be removed from the blacklist. We would like to make sure this situation does not repeat itself, in the interest of all parties involved.

Looking forward for your reply.

Thank you and best regards, Oris 12:21, 25 March 2009 (UTC)

Hi!

Links to that domain were added by different users, see m:user:COIBot/XWiki/888.com. Afaics those links are no advantage to wikipedia articles in sense of en:WP:EL (or de:WP:EL a.s.o.). Is there any article where a link to that domain would be useful? -- seth 15:54, 27 March 2009 (UTC)

Hi Seth,

Thank you for your prompt reply! We carefully read the instructions written in en:WP:EL and went over the list of references to 888.com m:user:COIBot/XWiki/888.com. In this list we found several references which are very relevant to the articles and can definitely enrich their content. For instance, it is only natural and straightforward to add external links to the company’s main websites which are mentioned within the article on 888 Holdings.

Furthermore, in some cases the external links to 888.com were inserted as we are the official sponsors of personas and teams such as Shane Warne (see au.888.com/shane-warne/), Michael Keiner (see de.888.com/sponsorship/de/michael_keiner.htm), Jeff Fenech (see www.888.com/sponsorship/en/jeff-fenech.htm) & Sevilla FC (see es.888.com/sponsorship/es/sevilla.htm). In each of these cases, relevant content which contributes to the article’s theme, was added. Though I agree with you, It might have been even more appropriate to reference these pages under “References” or “Notes”.

We will greatly appreciate any feedback you may have on the cases noted above or on any other steps we should take in order to follow Wikipedia’s guidelines.

Best Regards, Oris 09:41, 31 March 2009 (UTC)

At 888 Holdings there is a link to the company already.

I'm not sure about the sponsor links. I guess, it is better, if you choose one of those articles and ask on its talk page, whether a link to 888.com is wanted or unwanted. -- seth 10:20, 31 March 2009 (UTC)

Dear Seth,

We contacted you several months ago regarding the blacklisting of 888.com in Wikipedia. As a result we took some measures regarding the guidelines and instructions you provided us with. We now have only one focal point which is allowed to suggest content to Wikipedia according to its guidelines and in an effort to really enrich it with high quality, relevant content. We will be grateful if you could reconsider our removal from Wikipedia blacklist.

According to your recommendation, our focal point tried to add some relevant content within one of the talk pages but it seems like the editor relied more on the fact 888.com being blacklisted than on relevance of the content itself. See here.

We are currently in a process of uploading valuable, interesting content to our websites. We feel that Wikipedia readers might find this content attractive and relevant. You are more than welcome to review it yourself at: de.888.com/magazine/de/ or de.888.com/magazine/en/.

Best regards,Oris 14:15, 25 August 2009 (UTC)

Hi!

It's not true, that the blacklisting itself was the main reason, see [5].

You can tell me a specific 888 web page and a specific wikipedia article, and I'll will ask on the article's talk page, whether the authors would like to add the link there. (Of course you can ask by yourself, if your German is good enough or if you want to ask in English there.) In the English wikipedia there was one answerer only, user:2005, and seems to have a similar opinion as I have, i.e., one link to the main page should be enough. -- seth 20:12, 26 August 2009 (UTC)

Hi Seth,

Thanks for your patience and willingness to help!

888.com sponsors Dr. Michael Keiner (a professional German poker player) and keeps track on all of his main achievements in the following URL: de.888.com/sponsorship/de/michael_keiner.htm. I have noticed that Dr. Michael Keiner's Wikipedia page doesn't have any updates on his more recent achievements in the 2009 WSOP.

If you too believe that this information is relevant and contributes to the Wiki's article, I would be more than happy to get your help in addressing the relevant Wiki editor.

Thanks!! Oris 13:17, 22 September 2009 (UTC)

Hi!

I asked at de:talk:Michael_Keiner for some opinions concerning the link. Let's wait a few days for some answers. -- seth 11:49, 23 September 2009 (UTC)

Thanks! Oris 05:48, 24 September 2009 (UTC)

Hi Seth,

I have read the responses at de:talk:Michael_Keiner and they seem very peculiar to me since I believe that Wikipedia has a need for a good and reliable information with relevant and trusted references on and off the web. Maybe I didn't understand the editors' remarks correctly and so I would appreciate it a lot if you could help me understand this issue better.

I also wanted to thank you a lot for all your efforts - I really appreciate your help and it is not taken for granted!! Oris 08:04, 28 October 2009 (UTC)

Hi!

I'll try to translate and summarize the comments:

I asked at de:talk:Michael_Keiner: The article doesn't say anything about Keiner's achievements at 2009 WSOP. His sponsor's website gives some information. That domain is on our blacklist right now, so I'm interested in opinions on that link, for making a dicision about whitelisting it.

User GiordanoBruno doesn't see any reason for linking the webpage. He says, it would be better to copy the information to the wikipedia article, if the information was important enough. There wouldn't be anything special about that page.

User He3nry agrees with GiordanoBruno.

217.86.39.170 sees the Webpage as a reliable source and adds that every evidence is better than an unreferenced information. 217.86.39.170 claims that the wikipedia article contained a lot of numbers which should be proved by references. So in this user's opinion the link is good as a reference, but not as a regular external link (in that point 217.86.39.170 agrees with GiordanoBruno).

User Oberfoerster doesn't see the benefit of this external link as a reference, because he could not find the numbers from the wikipedia article in the 888-page.

My English is not very good, but I guess, I could outline the main points.

All in all I don't see that the link really is wanted there. -- seth 21:59, 28 October 2009 (UTC)

Lofoten.info

Hi, in case you haven't noticed, Talk:Spam_blacklist#lofoten.info. As I am writing there, I don't quite understand why this was added to global blacklist, but I guess I'm missing something ;) Best regards, Finn Rindahl 19:04, 8 November 2009 (UTC)

Thx! I will answer there. -- seth 23:26, 8 November 2009 (UTC)

Question

Is it abusive?. Thanks, ---Dferg 22:12, 22 October 2009 (UTC)

Hi!

"Schwanz in Auge" just means "dick (penis) in eye". I would not block somebody with this name, if that user would be helpful. But there maybe admins at w:de which would block that user just because of the name. -- seth 18:54, 23 October 2009 (UTC)

Note

Hello Seth. Maybe you're interested in this thread. Regards, --dferg ^{☎ talk} 19:50, 6 September 2010 (UTC)

thx. :-) -- seth 22:12, 6 September 2010 (UTC)

Hi,

I had to edit the title blacklist - diff. I wish to block the word "Peidar" with as many variants as we can from any new account. It is portuguese strong language and it is not needed anywere. That word, along with others, are being user by an idiot vandalizing all PT projects. I would like to be sure the regexp I've added is not causing massive harm, etc. Thanks,

--dferg ^{☎ talk} 22:42, 7 October 2010 (UTC)

Hi!

Sorry for late answer. Is the meta abuse filter active by now? -- seth 20:40, 11 October 2010 (UTC)

Hi. On meta we have activated local abusefilter, but I do not know how it will help on this case since that vandal only hits pt.wiktionary/wikiquote/wikipedia, etc. projects. Regards, --dferg ^{☎ talk} 07:33, 12 October 2010 (UTC)

The way you did it (P[eèéêë][iìíïî]d[aàáâä]r) was actually ok. There is no really better possibility, cause there are no character classes like "characters that look like e". And even if there were such classes, the regexp could be easily circumvented by using context-sensitive whitespace-similar characters like "P_e_i_d_a_r", "PxExIxDxAxR" a.s.o. (afaik the last example would even trick the abuse filters normalize function)

So I guess that you have to modify the regexps "on demand", i.e., every time when the vandal modifies his behaviour. -- seth 09:11, 12 October 2010 (UTC)

Thank you for your advice. Since my knowledge of regexp is still very basic I prefered to ask you. I am sorry for any inconvenients I might have caused. Regards, --dferg ^{☎ talk} 06:36, 13 October 2010 (UTC)

Your questions are not inconvenient at all. :-) -- seth 20:30, 13 October 2010 (UTC)

Hello again Seth. I'd like to ask if the proposed regexp in this thread is OK. Feel free to handle it if you want to. Cheers, -- Dferg ^{☎ talk} 21:34, 29 July 2011 (UTC)

Notice of review of adminship

Hello,

In accordance with Meta:Administrators/Removal and because you have made fewer than ten logged actions over the past six months, your adminship is under review at Meta:Administrators/Removal/October 2011. If you would like to retain your adminship, please sign there before October 10, 2011. Kind regards, vvv^t 17:05, 3 October 2011 (UTC)

google spam

Hi Lustiger seth, could I ask you to have a look at Talk:Spam_blacklist#Google_redirect_spam - a lot of Google needs to be blocked as it can be used as a redirect site. Thanks. --Dirk Beetstra ^{T C} (en: U, T) 13:38, 24 October 2011 (UTC)

Could I ask you to have another look at the request, I am not happy with the current rule, which may have false positives (though I agree, if the link redirected to is blacklisted, it should already not work - there may be a bug there). Thanks! --Dirk Beetstra ^{T C} (en: U, T) 15:57, 1 November 2011 (UTC)

Sorry for my late replies. For the last two months I couldn't spend much time in wikipedia. I hope this will get a bit better now. -- seth 20:20, 1 November 2011 (UTC)

regex

Hi Seth, I need some we have a regex that kills \bxairforces\.(com|net|org)\b, but I would also like to include xairforce (without the s). I figured xairforce[s] would do the trick, but apparently not. How do I make this simple thing work? EdBever 19:46, 4 December 2011 (UTC)

Hi!

\bxairforces\.(?:com|net|org)\b → xairforces.com, xairforces.net and xairforces.org

\bxairforce[s]\.(?:com|net|org)\b → same as above

\bxairforces?\.(?:com|net|org)\b → xairforces.com, xairforces.net, xairforces.org, xairforce.com, xairforce.net, and xairforce.org

-- seth 20:47, 4 December 2011 (UTC)

OK, thanks! This guy keeps spamming the different domains. If he finds another TLD, I'll block \bxairforces?\b EdBever 20:53, 4 December 2011 (UTC)

Hi!

In that case maybe \bxairforces?\. would be better (less general). -- seth 21:46, 4 December 2011 (UTC)

protocol relative links from grep_regexp_from_url.cgi

Hi Lustiger seth,

With the tool when I am logged in with secure url for https://toolserver.org to assist with my secure login for WMF, I trip across protocol relative issues with the output from the regex tool

• the purge link is fine
• the links to lists: wikiname all give http:// rather than https:
• when I get a positive hit for meta when it displays the log links it shows a problematic double url of

  http://meta.wikimedia.org//meta.wikimedia.org/w/index.php?title=Spam_blacklist&diff=prev&oldid=3209951&diffonly=yes

  which I don't see if I get a hit on enWP

list: meta blacklist
    dogsex(?!posed)
        log entry: dogsex(?!posed) #: # lustiger_seth # modification; see request

Thanks. billinghurst sDrewth 01:41, 6 January 2012 (UTC)

Hi! Thanks for reporting. Should be fixed now. -- seth 09:53, 6 January 2012 (UTC)

Possible XSS vulnerability

One of your tools may have an XSS vulnerability: https://toolserver.org/~seth/grep_regexp_from_url.cgi?userdeflang=%3Cbig%3E%3Cbig%3E%3Cbig%3E%3Cbig%3EXSS?%3C/big%3E%3C/big%3E%3C/big%3E%3C/big%3E&url=bit.ly πr² (t • c) 03:07, 28 March 2012 (UTC)

Hi!

Many thanks! Should be fixed now. -- seth (talk) 20:09, 28 March 2012 (UTC)

Regexp check

Hi seth. I added this to the SBL based on Talk:Spam_blacklist#Cross-wiki_spammer. Is the pro-* regexp correct? Thanks in advance. Best regards. -- MarcoAurelio (talk) 15:07, 16 October 2012 (UTC)

I have changed it to \bpro-.*?\.ru\b. Thanks. -- MarcoAurelio (talk) 15:17, 16 October 2012 (UTC)

Hi! I'm using unsafe computers/connections right now, so I'm not logged in.

pro-* matches any string (or to be more precise: domain) that contains "pro", "pro-", "pro--", "pro---" snd so on.

\bpro-.*?\.ru\b matches any domain that contains "pro-" (where "pro-" must not be prefixed by any alphanumerical character) and that contains somewhere behind that an occurence of ".ru" (where ".ru" must not be followed by ny alphanumerical character), e.g., "www.pro-wikipedia.ru", "pro-.ru", "sub.some-pro-domain.ru". -- 41.182.45.198 09:34, 17 October 2012 (UTC) (not logged in -- seth (talk) 21:25, 2 November 2012 (UTC))

Thank you very much for your advice. -- MarcoAurelio (talk) 22:18, 2 November 2012 (UTC)

Fundraising translation feedback

Hey Lustiger seth, I have a bit of a request to ask from you. We pulled down our banners nearly a fortnight ago for what was a highly successful international fundraiser and brought the curtain down on last years fundraiser. This week however we will be changing payment processors and during the testing of the new system it would be useful to use the time productively on on testing banner text.

To help us out with this I wonder if you would be willing to help us improve our geman text using This Link

Simply follow the simple instructions on that page and if you have any questions feel free to contact me on my talk page.

We are going to run the test on tuesday so if you dont see this message till 24 hours after it was sent you can ignore me :) Many Thanks though.

Jseddon (WMF) (talk) 20:21, 28 April 2013 (UTC)