Wikipedia talk:The Wikipedia Library

Newspapers.com made temporarily unavailable

Tracked in Phabricator
Task T322916

I presume there's a trouble ticket on Phabricator out for this already if the TWL site was changed to remove the Newspapers.com button and replace it with "Temporarily unavailable"? Does anyone know the ticket number so we can track how things are going with it? Silverseren^C 20:05, 6 July 2024 (UTC)

Related, if the decision is that there isn't a technical fix, but that we're reverting to individual accounts, I'd be willing to volunteer as an approver for this database, or whatever the term may be for behind the scenes of the Library. -- Zanimum (talk) 20:24, 8 July 2024 (UTC)

I notice that one of the comments in the trouble ticket is that one cannot log in via Ancestry. I'm not sure how that is a problem. Typically one logs in through the Library interface, and then signs on to one's account to create clippings. But the other items on the trouble ticket are serious. Figureofnine (talk • contribs) 23:12, 8 July 2024 (UTC)

Figureofnine, where can one find this trouble ticket? Jweiss11 (talk) 18:50, 9 July 2024 (UTC)

If you go to the Wikilibrary page, it's right there next to "temporarily unavailable." https://phabricator.wikimedia.org/T322916 Figureofnine (talk • contribs) 19:20, 9 July 2024 (UTC)

Silver seren T322916 Figureofnine (talk • contribs) 15:45, 10 July 2024 (UTC)

Thanks, Figureofnine, I've linked the tracker for it at the top of this section. Unfortunately, it appears despite multiple people bringing their concerns to the ticket over the past few weeks, no one on whatever Phabricator group responsible for investigating such things has shown up or left any sort of comment. There appears to be no action being taken (at least none commented in the ticket) trying to fix this going on three weeks now. Which is concerning. Silverseren^C 21:24, 13 July 2024 (UTC)

I imagine it may be due to summer vacations and so on. Figureofnine (talk • contribs) 15:27, 14 July 2024 (UTC)

Copying over a workaround from pburka:

Super janky workaround: if you have a free account, you can search, but you'll be prompted to pay (or start a free trial) to actually see the articles associated with the results. But, as far as I can tell, only search is broken in the WP Library proxy. So search using your free account, click on a search result, then edit the URL to replace "www.newspapers.com" with "www-newspapers-com.wikipedialibrary.idm.oclc.org". e.g. this link is paywalled:

www.newspapers.com/image/124073618/?match=1&terms=%22gordon+simpson%22+%22le+mans%22

, but this one works:

www-newspapers-com.wikipedialibrary.idm.oclc.org/image/124073618/?match=1&terms=%22gordon+simpson%22+%22le+mans%22

. It's slow and clunky, but better than nothing. pburka (talk) 20:49, 7 July 2024 (UTC)

Works for me so far. Firefangledfeathers (talk / contribs) 18:07, 14 July 2024 (UTC)

Thanks. I used that workaround in researching an article and was remiss in not posting here. Figureofnine (talk • contribs) 12:15, 15 July 2024 (UTC)

This workaround does not work for me. I can provide details but they appear to be the same as for the others. Coretheapple (talk) 16:23, 22 July 2024 (UTC)

Correction to my last comment. It does work for me in that it allows me to see an article for citation pruposes. However I cannot create a clipping. Coretheapple (talk) 14:51, 27 July 2024 (UTC)

Is there any method, or enwp page to watch, to be notified of when Newspapers is working again? The bug-tracking site is too opaque for me to navigate, else I would, there. — Fourthords | =Λ= | 21:13, 15 July 2024 (UTC)

At press time, a similar issue (re: proxy access) is affecting Wiley Online Library as well (as officially stated on their front page). --Slgrandson (How's my egg-throwing coleslaw?) 20:47, 19 July 2024 (UTC)

Related, Ancestry is now also not working. I click the link in the library, the OCLC URL loads, but it says "log in" at the top. -- Zanimum (talk) 11:13, 24 July 2024 (UTC)

I just had no problem with Ancestry, at least not logging on. It seems that the problems with these databases vary from computer to computer and account to account. Coretheapple (talk) 15:24, 24 July 2024 (UTC)

Still unavailable as of 7/31?

I haven't had access to newspapers.com for 10 days and counting now ("There was an error while loading the form. Please contact customer support"). I can't tell from this talk page or WT:Newspapers.com what the latest is. Clearing cookies nor changing internet browsers fixes anything. What is going on and when is this going to be resolved??? SportsGuy789 (talk) 23:15, 31 July 2024 (UTC)

Based on the Phabricator ticket, there haven't been any updates at all besides one "we're working on it" from a week ago. Which isn't all that comforting. Silverseren^C 23:33, 31 July 2024 (UTC)

I can still access it through this link - https://www-newspapers-com.wikipedialibrary.idm.oclc.org/ - I can't log in, save or clip anything, but the search function still works for me, and when I find an article I need to use as a ref, I click on the Print/Download button (which still works for me) and I save the image as a .jpg, which allows me to read/research the article offline for the content I need to cite, and then I change the wikipedialibrary.idm.oclc.org URL to - https://www.newspapers.com/image/668904663/ - which I can then use in {{cite news}} with {{subscription required}}. It's a clunky workaround, but until they get it fixed, it works for me. Isaidnoway (talk) 15:56, 1 August 2024 (UTC)

Apologies for the ongoing issues here. I've just posted an update at T322916#10036209 on what the current situation and plan going forward is. Please let me know if you have any questions. Samwalton9 (WMF) (talk) 16:58, 1 August 2024 (UTC)

Thanks so much for keeping an eye on this, as well for keeping us updated! — Fourthords | =Λ= | 00:27, 2 August 2024 (UTC)

Nope, search function doesn't work for me using that link. Samwalton9 thanks for the update. Looks pretty grim. Coretheapple (talk) 21:45, 1 August 2024 (UTC)

Have you tried using the method mentioned previously in this topic? That is, to search through newspapers.com (you need to log in through newspapers.com first, as it seems) and once you try to open a result and get the "start a 7 day free trial" window replace newspapers.com in your browser's address bar with www-newspapers-com.wikipedialibrary.idm.oclc.org. AstonishingTunesAdmirer 連絡 22:11, 1 August 2024 (UTC)

worth a try! Coretheapple (talk) 16:20, 2 August 2024 (UTC)

So, clipping worked previously for me, on July 24 to be exact. But I guess cookies expired or something, because now when I tried helping @Cunard over at WP:RX it no longer worked. I've found a solution but it's a little bit technical: once you log in through newspapers.com, you need to make sure the TWL proxy tab has cookies from newspapers.com. On Firefox, you can access them by pressing Ctrl+Shift+I → Storage → Cookies and then clicking the plus sign button to add a new one; on Chrome, it's Ctrl+Shift+I → Application → Cookies and double clicking the table to add a new one. Not sure yet which ones are required, but after I copied ncom-session, ncom-subscription, user_data and user_token clipping through the TWL proxy worked again. Keep in mind that you shouldn't share values of these cookies as that will allow others to use your newspapers.com account (probably not that big of an issue if it's a free account but still). AstonishingTunesAdmirer 連絡 10:58, 2 August 2024 (UTC)

Copying the cookies worked for me so that I could log in to newspapers.com on www-newspapers-com.wikipedialibrary.idm.oclc.org. Thank you so much for sharing this, @AstonishingTunesAdmirer! Pinging @SL93 so you know about this. Cunard (talk) 11:48, 2 August 2024 (UTC)

Worked for me too. Thanks :) PARAKANYAA (talk) 12:15, 2 August 2024 (UTC)

I'll give that a try. Regarding your previous suggestion---I will try that too. Coretheapple (talk) 16:19, 2 August 2024 (UTC)

I was doing some genealogical research through some old Missouri newspapers and found access here. --Kansas Bear (talk) 16:24, 2 August 2024 (UTC)

Might be because I'm using a laptop, but Ctrl+Shift+I brings up Newspapers.com's info panel on the righthand side of the screen (which opens when pressing Ctrl+I). For anyone else experiencing this, you can just right-click to bring up Inspect/Inspect Element. From there you can access the Storage/Application and Cookies tabs. Thank you @AstonishingTunesAdmirer! —Matthew  / (talk) 03:45, 6 August 2024 (UTC)

AstonishingTunesAdmirer, I'm trying to follow your instructions above. I'm using Chrome. I can view the cookies on each website via Ctrl+Shift+I, etc, but I can't figure out how copy cookies from one website to another. I need to copy four cookies (ncom-session, ncom-subscription, user_data and user_token) from newspapers.com to www-newspapers-com.wikipedialibrary.idm.oclc.org, correct? Do I just copy each cookie value? I'm not seeing how to double click the table to add cookies. Thanks, Jweiss11 (talk) 05:26, 6 August 2024 (UTC)

Yes, you need to copy the 4 cookies you listed from newspapers.com to www-newspapers-com.wikipedialibrary.idm.oclc.org. Initially, www-newspapers-com.wikipedialibrary.idm.oclc.org doesn't have some of these cookies, so you need to create them yourself. On Chrome, you need to double click in there to create a cookie and then enter its name (on Firefox, you need to click the plus button in top right corner). Once you create a cookie, you need to copy its value from newspapers.com, which is done by double clicking on it in Value column → Ctrl+C → double click the same cookie in Value column over at www-newspapers-com.wikipedialibrary.idm.oclc.org tab → Ctrl+V (for whatever reason Chrome won't let me paste via context menu, so need to use hotkeys). AstonishingTunesAdmirer 連絡 09:55, 6 August 2024 (UTC)

@AstonishingTunesAdmirer, many thanks for posting about this. I'd note though that the only one I needed to copy was user_token as the other 3 existed already under the proxy url's cookies (and user_token was the only one of those 4 that showed under my newspapers.com cookie list). It actually wasn't that tedious a task but I do hope a proper resolution is forthcoming (i.e. back the old direct account approach). Bungle ^{(talk • contribs)} 18:18, 6 August 2024 (UTC)

Yeah, second time I did it I only copied user_data and user_token and it worked, so you are probably right, but I am not sure if it checks for the other ones later on. It is a little bit tedious because it logged me out after a couple of days, but it's better than nothing while we wait for a proper fix (although the announcement over at Phabricator doesn't sound too promising). AstonishingTunesAdmirer 連絡 18:36, 6 August 2024 (UTC)

AstonishingTunesAdmirer, thanks for all your help on this. What I was missing before was scrolling to the bottom of the table of cookies at oclc.org and then double clicking on the blank row at the bottom to add more data. I copied over just user_data and user_token from newspapers.com and it worked! Jweiss11 (talk) 23:25, 6 August 2024 (UTC)

@AstonishingTunesAdmirer thank you for the cookie fix, very much appreciated. I did find I only needed to copy user_token. Kind regards, Goldsztajn (talk) 23:22, 14 August 2024 (UTC)

Bookmarklet for toggling between Newspapers.com and OCLC

Until the Newspapers.com logon issue is completely resolved (major thanks to @Firefangledfeathers and @AstonishingTunesAdmirer for finding workarounds for searching and making clippings), it's going to be necessary to switch back and forth between newspapers.com and newspapers-com.wikipedialibrary.idm.oclc.org versions of pages pretty often. I've been using a bookmarklet to automate the process. To use it, create a bookmark, then edit the URL and paste in the code below. Clicking it while on a Newspapers.com page will change the active page to the OCLC equivalent, and vice versa.

javascript:var url = window.location.toString(); var np = 'www.newspapers.com';var oclc = 'www-newspapers-com.wikipedialibrary.idm.oclc.org'; if (url.includes(np)) {var newurl = url.replace(np, oclc)} else {var newurl = url.replace(oclc, np)}; window.location = newurl;

Nick Number (talk) 17:45, 2 August 2024 (UTC)

Access to BNA

The subscription account I originally had expired earlier this year and then the library option became unavailable with no accounts available. I have since observed that the BNA may have been added to the library bundle via the oclc proxy based on info here but I cannot work out how to actually access this? If I type the url in manually it says unauthorised and I can't see any other previous discussions. Is BNA now unavailable again? Bungle ^{(talk • contribs)} 08:18, 7 August 2024 (UTC)

@Samwalton9 (WMF): are you able to shed any light here? Bungle ^{(talk • contribs)} 17:07, 12 August 2024 (UTC)

@Bungle Hi, sorry for the delay in responding - I was out at Wikimania last week. Unfortunately I think BNA is going to be a non-starter for us at the moment. We tried the voucher code route but, per previous communication we sent out, some users were being charged money and this risk was unavoidable. We tried proxy but, per my note here, these manual proxy configurations have turned out to be fragile, complicated, and very time consuming. As such I don't think we're going to be able to proceed with BNA. We'll put out some communication soon to confirm. Samwalton9 (WMF) (talk) 10:16, 13 August 2024 (UTC)

Thanks for the response. That's a real shame, as for the short time it was available it proved to be invaluable in citing some British-specific content which isn't available on the more international newspaper archives sites. Bungle ^{(talk • contribs)} 19:03, 17 August 2024 (UTC)

Springer

What's availble from em? SerialNumber54129 16:33, 8 August 2024 (UTC)

Nature plus most of Springerlink. Nikkimaria (talk) 23:51, 14 August 2024 (UTC)

Search errors on Newspapers.com

I know this has been mentioned a few times (including by myself a month back), but is there any workaround or solution to this persistent and frequent error message? "There was a problem loading the search results. Please try again. If the problem persists, please contact us." Without the ability to search the site 95% of the time, Newspapers.com is rendered completely unusable, and even the contact form spits out an error message whenever I try to submit it. I do see a lot of discussion on the numerous other issues the service is experiencing at the moment, but this issue in particular feels a bit overlooked and it has been the bane of my existence the past two months. Some of the proposed fixes involve logging into Newspapers.com directly, which isn't an option for me since I'm entirely reliant on the Wikipedia proxy. I'm holding out hope that there is some kind of ongoing dialogue between Wikipedia and Newspapers.com to find a fix. NGPezz (talk) 02:30, 19 August 2024 (UTC)

So, I'll break down the method I've been using as explained by others above, but I'll try to explain it in more general and understandable detail. Hopefully. Note that you'll probably have to use this method every day you're going to login, but it should last the entire day.

First, make sure you're logged into The Wikipedia Library. Next, in two separate tabs, open up the TWL Newspapers.com and then open up the OG Newspapers.com. Make sure you're logged into your Newspapers.com account on the latter. What we'll be doing is making it so the TWL version, which doesn't let you log in right now due to the errors, does recognize you as logged in. Since that's the one that lets you search and make clippings and such.

Then, for both Newspapers tabs, right click and then select "Inspect". This option is probably at the bottom of your right click menu, you may need to scroll down. It's below View Page Source. Now, there's probably a lot of text and menus and chaos on your screen. Just look at the top of the page and you should see a menu bar. You're likely currently on the "Elements" option. Six options to the right of that should be "Application". Click on that. Make sure you're doing this on both separate Newspapers tabs.

This should open up a menu on the left when you click on "Application". Under the "Storage" list, there should be an option named "Cookies". If that option isn't expanded, click the little arrow to the left of it, which should then let you click the indented option below it, which is probably something like "http://www.newspapers." and it probably runs off the screen at that point. Click that.

Finally, we're at the meat of what we want. This is the list of remembered cookies for the website. On your general Newspapers.com one, there's four that we care about, "ncom-session", "ncom-subscription", "user_data", and "user_token". These cookies are what let you be logged in to the general Newspapers.com site. If you look at the Cookies list on the TWL Newspapers tab, you'll likely see only "ncom-session" on that one.

What you want to do is copy the second column, the "Value" one next to the names I just mentioned, and add them into your TWL Newspapers list of Cookies. You might need to double click to make it allow you. For the "ncom-session" one, just copy paste that value over from the general Newspapers.com tab. For the other three, you'll likely have to add the name as well first. So scroll down to the bottom of the Cookies list until you see an open line and right click there and select the option "Add new". Then add in one by one the three names I mentioned above other than "ncom-session", hitting enter after each one and copy-pasting the related Value for each one from the general Newspapers.com tab.

After you have all four names and values entered into their own separate lines on the TWL Newspapers page, X out of the Inspect screen with the button on the top right and refresh your page. It should now show you as logged in on the TWL Newspapers tab (you can close the general Newspapers.com tab now, you don't need it anymore).

That's all the details I have on the mechanism to get the TWL Newspapers website to work. Sounds like a lot, but once you're done it a couple times, it gets pretty easy to do. Annoying you have to do it every day to get the website to work as it used to, but hey, it does work once you do it. Clipping ability and all. I've been using it for a little over a week now. It'll do until we get an actual proper fix. Let me know if you have any questions or if I was too obtuse in any of my explanation. Silverseren^C 03:44, 19 August 2024 (UTC)

Oh, one more thing. When you make a clipping with your account and try to go to it, it'll take you to an error page. Just change the "-newspapers-com.wikipedialibrary.idm.oclc.org" part of the URL to "newspapers.com" to be able to view and use the clipping. Another annoying thing to have to do, but it works. Silverseren^C 03:49, 19 August 2024 (UTC)

Thank you, but none of this advice is applicable to the search issue I'm experiencing, which is why I created a separate discussion. "Make sure you're logged into your Newspapers.com account on the latter" does not help, because I don't have a Newspapers.com account. I only use the Wikipedia proxy. I've already tried removing any and all cookies related to the proxy or the site, it does not make a difference. Logging in is not the problem for me, clipping is not the problem for me, my concern is entirely related to the error message which renders the search function unusable. For what it's worth I tried again this morning, it functioned for about half an hour before the error popped up once more. NGPezz (talk) 15:33, 19 August 2024 (UTC)

Ah, I misunderstood what you originally wrote. Is there any particular reason why you can't just make an account on Newspapers.com, NGPezz? It's free and quick to do so and then my method above will work. Otherwise, the entire issue is with the proxy access, which doesn't seem like it's going to be fixed for quite some time. You seem to be having a unique version of the errors, but the general problem does include the search function acting erratically. Silverseren^C 20:20, 19 August 2024 (UTC)

And I'm fairly certain it will work if you do so. Since the error message is "There was a problem loading the search results. Please try again. If the problem persists, please contact us", right? That's the one I'm seeing right now because I had an old tab logged into the TWL Newspapers, but I haven't done the method above yet today, so I don't have access. Once I do use the method, I'll be able to use it just fine. Silverseren^C 20:34, 19 August 2024 (UTC)

Thanks. My problem is not entirely unique, a few other users voiced similar concerns in an earlier discussion here on June 27th. Correct me if I'm wrong, but isn't Newspapers.com a subscription-based service? They advertise a free 7 day trial, after which they would start charging the credit card I used to register. NGPezz (talk) 23:31, 19 August 2024 (UTC)

That's only if you pick a subscription. You can made an account for free, no credit card required, just an email. You just won't be able to access anything with that account (which is unimportant because you'll be using the TWL version to access things, you just need an account login for my method to work). Silverseren^C 01:13, 20 August 2024 (UTC)

Thank you very much! I did not know that a free account was all it took to use the search function without issue. And thank you for the step-by-step process on how to login (and thus effectively search) in the proxy as well, at least on a daily basis. I did quickly run into the deficiency inherent to a free account on the base site (not being able to view the entire page which hosts the details in question). Fortunately the bookmark toggling function created by Nick Number is exceedingly useful for avoiding this problem, so that I can seamlessly search without the proxy and switch the proxy on whenever I need to view a page. I'll leave another comment if further problems come up, but for now your advice has worked like a charm! NGPezz (talk) 02:46, 20 August 2024 (UTC)

Now that it's been a couple days, I can confidently mention that the search error still often occurs within the proxy even in a logged-in state. Fortunately the ability to search is still working perfectly outside the proxy, so I'm satisfied. I'm tagging some other users who have run into the same problem, so that they know there's a decent workaround now: PK-WIKI, Connormah, Zanimum. NGPezz (talk) 17:10, 22 August 2024 (UTC)

Bless you for this. I got newspapers.com to work which I needed for some work I've been doing. Thanks for taking the time. Jessamyn (my talk page) 19:59, 21 August 2024 (UTC)

User:Silver seren, I appreciate you. I tried, but I got lost about halfway through; I didn't see a ncom-subscription cookie, and what follows was probably over my head anyway. I wish someone would just fix this: of all the kinds of databases we could use, this is possibly the most useful one. Drmies (talk) 22:21, 23 August 2024 (UTC)

I've found since, after doing some testing, that you just need to copy over user_data and user_token with their values. The other two don't seem to matter. Just your user login cookies, which those both are. See if you can try again with just those two? I believe in you, Drmies! Just right click -> Inspect -> Application tab -> Cookies option on the left -> Copy over user_data and user_token. Silverseren^C 22:28, 23 August 2024 (UTC)

Per User:Goldsztajn above, just copying user_token is enough. I've been doing that lately. However, I now have to redo this procedure daily, not sure if it's related. AstonishingTunesAdmirer 連絡 09:55, 24 August 2024 (UTC)

For the time being perhaps we bookmark a shortened version of this workaround at the top of the page? Or maybe just a link to the discussion when it is archived? Regards, Goldsztajn (talk) 11:35, 24 August 2024 (UTC)

Agree. I think we should also note that a newspapers.com account itself is free. The free version allows you to search, but to see pages you need a subscription. And what we do here allows us to see pages without the subscription, using the TWL proxy (which is currently broken and needs this fix to work).

We should also mention the bookmarklet above (thanks User:Nick Number!). Search seems to be broken in the TWL proxy. So, once you fix cookies and see your account name in the top right corner when you open the proxy, you go back to newspapers.com, search there, click on search a result, get the message saying that you need to subscribe, and then you use the bookmarklet to switch to the TWL proxy version of the page. AstonishingTunesAdmirer 連絡 14:21, 24 August 2024 (UTC)

Getting page numbers from Oxford Academic

Is it just me or is there now no way to get page numbers from Oxford Academic? I no longer see any option to download a pdf or view page numbers in the browser. It's a lot less useful if I can't provide a verifiable cite. (t · c) buidhe 06:05, 22 August 2024 (UTC)

It's possibly something with Oxford Academic; I have access through a university library and see the same problem for books, although journals appears to be working. Regards, Goldsztajn (talk) 07:28, 22 August 2024 (UTC)

I see no option to download an entire book, but chapters seem to have a button to download a PDF that indicates the page numbers in the margin. XabqEfdg (talk) 22:58, 22 August 2024 (UTC)

XabqEfdg where are you finding the download button? I cannot see it, for example for this chapter. (t · c) buidhe 03:06, 23 August 2024 (UTC)

When I am at the top of the page, I see a row of buttons in the middle of the screen just above the "Abstract" heading. The leftmost button has a PDF icon and the text "PDF". I can also find it if I search the page (i.e., CTRL+F) for "PDF". XabqEfdg (talk) 03:44, 23 August 2024 (UTC)

Weird. I tried a few different chapters, the pdf button is showing up for some of them but not others. (t · c) buidhe 05:16, 23 August 2024 (UTC)

Oxford are pretty precious all over. They state—even for books where everything can be downloaded—that 'Oxford Academic does not intend for entire eBooks to be downloaded'. So although you can, you shouldn't... for what we are about to receive, etc. SerialNumber54129 14:01, 23 August 2024 (UTC)