Wikipedia talk:Newspapers.com

Publisher Extra no longer available

Back when access was still account based, we would get access to all Publisher Extra features and publications. Is there a way to access this using the proxy method, or is it something that is no longer available? Yeeno (talk) 10:16, 17 December 2023 (UTC)

I am having the same problem. I hope the limitation is only temporary. Eddie Blick (talk) 02:41, 4 January 2024 (UTC)

Hmm, works for me. I was able to browse the 03 Jul 1978 Boston Globe, which I believe is a Publisher Extra feature. Maybe try tossing the cookies in your browser? See also links at the bottom of Wikipedia:Newspapers.com#Using_the_"Clipping"_function. Cxbrx (talk) 03:21, 4 January 2024 (UTC)

@Teblick @Cxbrx: I was able to access my page while logged in by replacing newspage in the url with image. For example, if I access https://www-newspapers-com.wikipedialibrary.idm.oclc.org/newspage/436990449/, I run into a paywall, but https://www-newspapers-com.wikipedialibrary.idm.oclc.org/image/436990449/ works fine. Yeeno (talk) 03:45, 4 January 2024 (UTC)

Thanks, @Yeeno! I will try that technique. Eddie Blick (talk) 03:50, 4 January 2024 (UTC)

This is happening to me now, as of last night (so 5 April 2024). Accessing as usual through TWL, I can't see any search results, they all take me to the "Upgrade to a Publisher Extra Subscription to view this page" popup. I think I can see my own old clippings fine. Anyone else? Penny Richards (talk) 14:23, 5 April 2024 (UTC)

Update: got it working in another browser. Probably a cache problem. Penny Richards (talk) 20:54, 5 April 2024 (UTC)

Lots of links to newspapers.com /image/ that are not readable

I'm not sure what to do about all the links to www.newspapers.com articles that require a non-WP newspapers.com account and are to the entire page, not a clip. Today it looks like there are 20,696 links like this.

The issue is that the text referred to by these links is not readable by most readers because a non-WP newspapers.com account is required. Specifically, using the WP:LIBRARY for access will not work as the URL is to www.newspapers.com and not to the proxy. For details, see Updating URLs that start with https://www.newspapers.com/image

There was a bit of discussion above in issue converting clipping to citation, but I feel this is somewhat different issue.

I was trying to clean these up by hand, but perhaps the preview facility could be adjusted so that links to https://www-newspapers-com.wikipedialibrary.idm.oclc.org/image or https://www.newspapers.com/image would prompt a warning that is similar to the warning that is shown when there is a typo in the access-date etc.? I suppose I could submit a feature request for this feature.

Does anyone have any other ideas? Cxbrx (talk) 21:22, 3 January 2024 (UTC)

The only fix I know of for these is to manually create clippings. I do it when I see them, but 20K is going to take a while. Regarding a warning, using paywalled links in references isn't prohibited, but it would good if there were a prompt to add a |url-access=subscription parameter, and perhaps a polite suggestion to create a clipping. Nick Number (talk) 04:31, 4 January 2024 (UTC)

Couldn't a bot clip the entire page and replace the reference link with that? Ideally we would want the specific article clipped, but it at least gets the source accessible to more readers. It seems like it wouldn't be too hard so long as there's not a limit on number of clippings. We would probably need to consider whether Newspapers.com would be okay with a large number of automated clipping. Hey man im josh (talk) 05:10, 4 January 2024 (UTC)

Interesting idea to clip the entire article. I'm not sure if newspapers.com would go for it. Another issue is that the resolution for full page clippings means that it is sometimes difficult to read the article. Over at User_talk:WeirdNAnnoyed#Links_to_newspapers.com_articles_that_require_a_WP_account, User:Bsoyka, author of the User:BsoykaBot proposed eventually modifying the bot to send email to editors who add non-clipped links. I'm not sure about this as I agree that paywalled links are not prohibited. I think it might be best to nudge users in real time when they are adding the link. Some users will not want to be bothered with creating a newspapers.com account and doing the clipping, but others who use newspapers.com regularly perhaps will eventually create an account and start clipping. This issue is probably not specific to newspapers.com and might affect the WP:LIBRARY proxy for newspaperarchive.com or any other proxy that allows creation of publicly readable links. Cxbrx (talk) 06:09, 4 January 2024 (UTC)

Some of these links may also be added by users who don't have access. Most pages have the OCR text available even without access to the page image itself, so they may be adding the citation based on that. Perhaps the first step would be to have it show as a CS1 message (in edit mode only), with a tracking category to allow others to find and fix these. Pi.1415926535 (talk) 06:37, 4 January 2024 (UTC)

As mentioned above, I think it might be useful to at least mention to users in some way that this can be somewhat problematic. Just a bot task watching recent changes and sending a friendly reminder to anyone who adds unclipped links (asking them either to clip or at the very least add a url-access param). If they follow along with the instructions and start clipping, that's great. If not, no extra harm done, just more links to clean up. I'd be willing to take a shot at making such a bot, but I'd love to see if there's some consensus here first. As for automatically clipping articles, I'd personally doubt Newspapers.com would be on board with that plan, but maybe Samwalton9 (WMF) could look into it and reach out to them? Bsoyka (t • c • g) 15:52, 4 January 2024 (UTC)

Also, hate to be the bearer of bad news, but there appears to be another 9,000+ that need to be clipped here. Perhaps a bot task could replace these /newspaper URLs with /image too? See the section above. Bsoyka (t • c • g) 15:55, 4 January 2024 (UTC)

Ugh. Thanks for finding those. It would be nice to come up with a search regular expression that would find both the newspage and image urls, though I'm not sure if the search mechanism has support for complex regular expressions. I guess a tracking category could be useful, though I worry about yet another tracking category causing yet more clutter. Perhaps any notification mechanism would only work for editors who meet the WP:LIBRARY requirements for access to newspapers.com? I slightly prefer CS1 messages while editing during preview because the editor would likely have newspapers.com open via the proxy and could do the clip. A notification from a bot would likely happen after the editor moved on. I see the CS1 message as a nudge, and the after the fact bot notification as more of a nag. Cxbrx (talk) 17:08, 4 January 2024 (UTC)

I have access to Newspapers.com but via Wiki. I can try help creating the clippings when I have time. 9000 is a lot though. Timur9008 (talk) 17:22, 4 January 2024 (UTC)

I agree with the bot idea by Bsoyka and the tracking category, as well. Both a simple nudge to clip Newspapers.com with a link to a good example and the url-access param seem helpful to me. --Engineerchange (talk) 17:36, 4 January 2024 (UTC)

Additionally on the bot clipping the entire article idea provided by @Hey man im josh: I think the readability can be impacted when the page is too large. You can't zoom in and read text at a certain point when the clipping is too large. --Engineerchange (talk) 17:46, 4 January 2024 (UTC)

Agreed that clipping the entire page is probably not what we want. BTW - it looks like the newspage url is more useful than the image url as the newspage url includes OCR. Compare this newspage with this image page. In the near term it might be worth having a bot that replaces image with newspage. Cxbrx (talk) 02:08, 6 January 2024 (UTC)

I think the image page is more useful, particularly for TWL users. When you visit an image page with no account, you easily get to the newspage URL by clicking the X on the big popup. However, if you visit a newspage URL even with TWL access, you seem to need to manually change the URL to an image URL in order to see the actual page. (TL;DR: image -> newspage Y... newspage -> image N) Bsoyka (t • c • g) 03:15, 6 January 2024 (UTC)

Ok, I agree, I retract the above. The image urls are better than the newspage urls. Cxbrx (talk) 17:43, 6 January 2024 (UTC)

I just discovered that using /image with ?clipping_id, e.g. https://www.newspapers.com/image/20360056/?clipping_id=145167881, may allow unauthenticated users to see both a clipping and, by dragging and zooming, the rest of its page in a readable size. I don't know if this works for all papers, or whether it's unintended functionality which will be removed in the future, but it's worked in a private browser tab for all of the clippings I've tried so far. However these links don't appear to archive well, so probably shouldn't be widely used. Nick Number (talk) 14:59, 12 April 2024 (UTC)

CloudFlare issue

For those wondering, as of January 27, 2024, an issue is impacting access (T322916). It seems to be on their end. Tkbrett (✉) 14:16, 28 January 2024 (UTC)

This issue appears to be resolved now. Jweiss11 (talk) 22:52, 30 January 2024 (UTC)

Yes, it works again! I've been checking all day. Thanks to all who worked towards this fix. Penny Richards (talk) 23:11, 30 January 2024 (UTC)

I will add my thanks, also. I appreciate having access to Newspapers.com restored. It is a valuable source. Eddie Blick (talk) 02:45, 31 January 2024 (UTC)

https://www-newspapers-com.wikipedialibrary.idm.oclc.org/

Is it just me or does the site load much more slowly via the new URL through the Wikipedia Library? When I was using plain old Newspapers.com, it seemed to go much quicker and there were fewer errors. Now it seems I have to wait for search results and images to load and sometimes have to reload them after my first effort fails. I'm not sure if this is a me problem or if this is universal. Dennis C. Abrams (talk) 14:26, 20 March 2024 (UTC)

@Denniscabrams: I've never really noticed any significant issues, although it will inherently be slightly slower as your requests have to go through the Wikipedia Library as a sort of middleman before reaching the Newspapers.com servers. Bsoyka (t • c • g) 20:42, 20 March 2024 (UTC)

View newspaper articles in the citation

I have previously used the following format:
https://www.newspapers.com/search/#lnd=1&query=%22Terrace%22&ymd=1929-01-06&t=11702:
for readers of articles to view the actual newspaper excerpt when clicking on the citation. This no longer works. Is there another way to achieve this outside of logging into that site by one means or another? DMBanks1 (talk) 22:47, 11 April 2024 (UTC)

I'm afraid I don't know what that syntax previously produced. What are you trying to accomplish beyond what is done by a clipping? Nick Number (talk) 14:42, 12 April 2024 (UTC)

@Nick Number: Up to a few days ago, this syntax produced an image similar to the one seen in a search when logged into the site via Wiki Library. It does not appear even possible now to search the site when not logged in. I am unfamiliar with creating clippings. Can this be done when logged in via Wiki Library? Can a small section of an article be clipped? Are there instructions to create a clipping? DMBanks1 (talk) 15:27, 12 April 2024 (UTC)

Basic instructions are at Wikipedia:Newspapers.com#Using the "Clipping" function. Nick Number (talk) 16:11, 12 April 2024 (UTC)

@Nick Number: When the "Welcome from Wikimedia Foundation" "Free Registration" screen appears at step 7, the message is
"There was an error while loading the form. Please contact customer support"
Is free access no longer available? DMBanks1 (talk) 17:22, 12 April 2024 (UTC)

I'd recommend creating a free account with email through their normal form first. Then at step 7, log in with that account instead of registering. Nick Number (talk) 17:58, 12 April 2024 (UTC)

@Nick Number: Register button on normal form is inactive, making registration impossible. DMBanks1 (talk) 19:27, 12 April 2024 (UTC)

I'm not able to reproduce that; I can at least enter an email address and password and click the Cloudflare box to get the Register button to light up. I'd recommend clearing your cache and/or trying a different browser. Nick Number (talk) 19:30, 12 April 2024 (UTC)

@Nick Number: Where is the Cloudflare box? DMBanks1 (talk) 19:40, 12 April 2024 (UTC)

For me it shows a "Verify you are human" box below Show Password after entering an email address. Nick Number (talk) 20:04, 12 April 2024 (UTC)

@Nick Number: The button took a few minutes to appear. Thank you for your patience. I think I've now created a clipping
https://www-newspapers-com.wikipedialibrary.idm.oclc.org/image/774906896/?match=1&clipping_id=145249056
from Victoria Daily Chronicle Mon, Dec 11, 1865 ·Page 3, but am not entirely sure. If not, what step have I missed? If so, how is this expressed as a regular https file, since I cannot figure out the wiki guideline? DMBanks1 (talk) 21:29, 12 April 2024 (UTC)

The publicly accessible URL (reached by clicking the View Clip button and then changing www-newspapers-com.wikipedialibrary.idm.oclc.org to www.newspapers.com) is https://www.newspapers.com/article/victoria-daily-chronicle-the-proposed-wa/145249056/. This process can be streamlined a bit, and archiving added, using the bookmarklet described here at the line starting with "My workaround". Nick Number (talk) 21:37, 12 April 2024 (UTC)

@Nick Number: Your help in navigating me through this exercise has been much appreciated. DMBanks1 (talk) 21:49, 12 April 2024 (UTC)

Date range searching

@Nick Number: Until yesterday morning, searching on a date range worked fine. Since then, searches produce varying numbers of searchable items followed by a consecutive list of page numbers, which are not buttons to anywhere else. Searches on specific dates still appear to work. Is the site experiencing chronic problems when accessed via Wikimedia? DMBanks1 (talk) 15:47, 23 April 2024 (UTC)

They seem to have changed their search sometime last week. I'm finding it less convenient since it no longer appears to support searching for a string and date in the same query. Maybe there are different ways to get to these functions now, but I'm unaware of them. Nick Number (talk) 17:17, 23 April 2024 (UTC)

I am finding different things on different days. 2 days ago, I had to change the date by sliding the bar in the right margin (a pain in the X, IMO), but then yesterday I was able to add a date range at the top. So that seems to support that they are making changes. SusunW (talk) 17:40, 23 April 2024 (UTC)

I can confirm that I've been seeing the same changes as SusunW. Pi.1415926535 (talk) 18:03, 23 April 2024 (UTC)

Extra Subscription error message

It probably has something to do with all the changes happening on the site, but today every search result ends up with the error "Upgrade to a Publisher Extra Subscription to view this page". I can neither see any article nor clip it as I can't get past the popup. SusunW (talk) 16:15, 25 April 2024 (UTC)

I haven't been able to access Newspapers.com articles since at least 24 April. I've been forced to use the California Digital Newspaper Database for my research, but that doesn't have all my sources and is limited to California.– Whaleyland (Talk • Contributions) 12:04, 27 April 2024 (UTC)