Wikipedia:Link rot/URL change requests/Archives/2019/November

This is an archive of past discussions about Wikipedia:Link rot. Do not edit the contents of this page. If you wish to start a new discussion or revive an old one, please do so on the current main page.

clydesite.co.uk

(Discussion excerpted from WP:VPT#Bot to replace highjacked links)

Is there a bot to disable hijacked links in refs/ELs that may now be unfit/malicious? At Wikipedia:Teahouse#Hijacked site with possible malware, Lyndaship has reported that clydesite.co.uk has moved to clydeships.co.uk, and the new www.clydesite.co.uk host may be malicious (certainly misleading). The layout of the new site has changed, so it's not an easy replace of the hostname (though maybe editors could help construct a translation table?). Meanwhile, disabling the links would probably be reader-friendly. Is there a bot or some other means to handle this? —[AlanM1(talk)]— 23:07, 5 November 2019 (UTC)

@AlanM1: Agree on malicious domain: http://www.clydesite.co.uk/clydebuilt/viewship.asp?id=2467 it is a sketchy domain. Normally would set the domain to Blacklisted in the IABot interface and run IABot and archive them all (since we don't know the new URLs second option is archive). But they will also require adding |url-status=usurped, to suppress displaying the malware link, which is custom coding. WaybackMedic can do it. Post/move this request to WP:URLREQ. Might also edit filter request. -- GreenC 01:44, 6 November 2019 (UTC)

After further research, I've found there does not seem to be an obvious mapping between the old and new sites. Some content (like the www.clydesite.co.uk/articles/... links) appears to be missing completely. However, mappings for most of the links, pointing to entries in a ship database, might be possible to get from the website authors, or maybe by a simple scraping I can throw together. Archive.org does have archives for the ones I spot-checked, as expected, so the question becomes do we urgently mark all of the links usurped because they are now potentially mailicious and point the archive params to archive.org, or wait a few days to see if I can come up with a translation map and then run a bot to just substitute the new URLs. —[AlanM1(talk)]— 19:32, 6 November 2019 (UTC)

For the cite template usage, add |url-status=usurped, |archive-url= and |archive-date= set to the appropriate (last before access-date?) archive.
For the external link usage, replace with the appropriate archive URL (right?)

for the 545 links I've found. I should be able to provide a list of articles and links tomorrow. —[AlanM1(talk)]— 19:32, 6 November 2019 (UTC)

User:AlanM1/Clydesite now has a list of the articles and links extracted from this search. —[AlanM1(talk)]— 19:43, 6 November 2019 (UTC)

@AlanM1: thanks for the list. Bot ran and..

Converted 581 links to archive.org and 7 to archive.today - there were also existing archive URLs with |url-status= converted to "usurped"
User:GreenC/data/Clyde has a few remaining links that need attention - can you convert these to the new site?
Converted 44 links on Commons [1] though some of these it added a {{dead link}} meaning it couldn't find an archive and would need manual switch to the new site.
A few remain in search [2] that might need manual removal or converted to archives or switch to the new site.
The domain is "Blacklisted" in IABot database, meaning iabot will treat all found cases as dead and not import any new into the database.

-- GreenC 01:26, 7 November 2019 (UTC)

@GreenC: Thanks again for the quick resolution! I'll have a look and write back, probably 2019-11-08 (PST). Should I still request an edit filter? —[AlanM1(talk)]— 03:26, 8 November 2019 (UTC)

Looks like Lyndaship handled item 2. I went through the articles at item 4, which had clydesite.co.uk in the cite's |publisher=, changing them to Clyde-built Ship Database instead. The only remaining mention is at Scottish inventions and discoveries, where it's part of the title (the site had a periodical called "Clydesite magazine"). —[AlanM1(talk)]— 23:20, 8 November 2019 (UTC)

@AlanM1: For citations to entries in the ship database, it might be preferable to switch them over to use {{cite ship register}} so that the link scheme can be easily updated in the future if needed. A similar things was done when Det Norse Veritas moved from dnv.com to dnvgl.org (in that case, AWB was used to change {{cite web}} to a subst of an invocation of Module:Sandbox/Ahecht/Cite DNV, which parsed the parameters and spat out a properly formatted {{cite ship register}} template). --Ahecht (TALK
PAGE) 21:40, 11 November 2019 (UTC)

@Ahecht: Thanks. In this case, unfortunately, they did a complete site redesign, and the ID numbers changed, so we (well, mostly GreenC

) had to just use the archived versions of the pages. Note there was also some urgency because of a concern over the new host at the old site being potentially malicious. I generally agree, though, and used the {{Cite ship register}} a couple times on the edge cases. It would have been nice, if we could have generated a map of old to new IDs, to switch to using the template. That could still be done, I suppose. Perhaps we can get the site operators to contribute such a list if they have an easier way to do it than we would? I'll note, too, that one of the ones I did manually had lost an important detail (the keel-laid date) from the entry in the transition, so I would have preferred to keep the archived site in that case. I don't know the reason or the extent to which this may have happened with other entries. —[AlanM1(talk)]— 01:07, 12 November 2019 (UTC)

Add that custom templates create link rot in practice. When a site changes its scheme, invariably some URLs are not migrated to the new scheme, sometimes a lot of them are left behind. This is always the case, it has been in every request made here, because a change in scheme means they moved to a new server/software/owner and didn't migrate everything (by choice or accident). Those URLs require archive URLs. However, since the URLs are housed in a custom template (which may or may not have support for archive URLs), it requires custom programming specific to that template, and my bot indeed no bot has the code to deal with thousands of custom templates on Wikipedia. Stick with CS1|2 when possible it is more secure. -- GreenC 06:43, 12 November 2019 (UTC)

wilhelmgustloff.com

The above mentioned domain has been usurped. I have archived all the existing links on the English wikipedia but notifying here in case anything else needs to be done. Lyndaship (talk) 11:07, 13 November 2019 (UTC)

Ok, blacklisted in the IABot database. -- GreenC 14:16, 13 November 2019 (UTC)

plimsollshipdata.org

This site is used on about 1500 cites but no longer hosts the data. The information appears to be mostly now available at plimsoll.southampton.gov.uk with similar urls. However as they are just photos of the pages from Lloyds Register of Shipping to save any urls which haven't been moved across could they be linked to the archive copies of the original website? Lyndaship (talk) 16:00, 14 November 2019 (UTC)

HighBeam Research

Hi there. HighBeam has been shut down and no longer works. There are numerous links to highbeam with https while there's only a handful with http. Some of these links have already been archived with Wayback machine, but I'm not sure how many still need an archived link. Thanks! --MrLinkinPark333 (talk) 23:38, 25 November 2019 (UTC)

I did Highbeam a while back. Any remaining should be marked with a {{dead link}} if there is no archive available. There could be some missed because the bot back then was less sophisticated. If you see some not archived or not with a dead link template, let me know. -- GreenC 00:33, 26 November 2019 (UTC)