How do I get better web page screenshots?

lmrqrt

Hi Zotero community,

I've been trying to add web page entries of a database to a group library, including proper screenshots. Unfortunately, my screenshots always end up looking like (I'm no expert, but what I think of as) a basic HTML web page.

I've compared results with my colleague, whose screenshots do retain the look of the original web page. We both use the most recent version of Firefox plus the Zotero connector.

Here are two screenshots (of screenshots) to compare: https://seafile.unistra.fr/d/fc01fbd6b0b44402be93/

We both use institutional logins to the WBIS database, but via different libraries. I'm suspecting this is somehow the reason why mine look so bad, but I would love to understand why.

Is there anything I can do to get proper screenshots?

Thank you!

dstillman

Make sure you have Zotero open when you save and Zotero is listed as available in the Zotero Connector preferences. You'd get snapshots without auxiliary files and styling if you save directly to your online library instead of the Zotero app.

lmrqrt

Both are the case. I don't even have saving to zotero.org authorized in the preferences...

lmrqrt

I've now tried saving a variety of web pages – for some newspaper sites, the connector simply snapshots the printable version, for some, the screenshot's a "page not found", so there seems something wrong more generally with my snapshotting

dstillman

Can you provide some example URLs where you're not getting a proper snapshot?

What happens if you try to save this forum page?

lmrqrt

This forum page comes out perfectly fine.

Here are some examples of it not working:

https://www.nytimes.com/2019/02/13/us/politics/manafort-mueller.html?action=click&module=Top Stories&pgtype=Homepage

For the NY Times, I can either get a newspaper article item with metadata but no snapshot, or a web page item with a broken snapshot ("page not found")

https://www.sueddeutsche.de/kultur/erinnerungskultur-deutschlands-imperativ-1.3866258

This German news site automatically just gives me the printable version of any article as a snapshot

https://www.royalacademy.org.uk/article/magazine-the-tale-of-russias-revolutionary-ceramics

Here the connector tries to make a snapshot, but fails.

Many other web pages work fine, though...

lmrqrt

Update: I get an error by the connector when I try saving this page to Zotero: https://wbis.degruyter.com/login
It does save to Zotero, but without snapshot.

My Debug ID is D1908815117.

dstillman

https://www.nytimes.com/2019/02/13/us/politics/manafort-mueller.html?action=click&module=Top Stories&pgtype=Homepage

For the NY Times, I can either get a newspaper article item with metadata but no snapshot, or a web page item with a broken snapshot ("page not found")

NYT recently started doing something weird with their pages that breaks snapshots, so we disabled snapshot saving in the translator for now. (When you override that by using the webpage option, you get the broken snapshot that we disabled.) Zotero uses the same save functionality as Firefox, and you'll see the same thing if you try to save a complete webpage from Firefox (or Chrome, for that matter).

https://www.sueddeutsche.de/kultur/erinnerungskultur-deutschlands-imperativ-1.3866258

This German news site automatically just gives me the printable version of any article as a snapshot

The Sueddeutsche translator saves the print-friendly page by design. Triggering the print dialog is certainly non-ideal, though.

https://www.royalacademy.org.uk/article/magazine-the-tale-of-russias-revolutionary-ceramics

https://wbis.degruyter.com/login

These two both time out for me too. I've created an issue to track those, though we have a more general change to saving planned that would address these, the print-dialog problem, and most other issues with JavaScript-heavy pages.

But to be clear, these wouldn't behave any differently on anyone else's computer. I actually meant examples where other people are getting different results, which was your original report.

lmrqrt

Oh ok, thank you for the extensive answer! Good to know that these are not part of a specific technical issue on my side.

My original issue was with this database:

https://wbis.degruyter.com/login


I did suspect that my bad snapshots on this database have something to do with the institutional access via my library I am using (this is what the URL looks like for me: https://wbis-1degruyter-1com-1dvg49rj90496.emedia1.bsb-muenchen.de/index), but my colleague also gets access via their library (albeit a different one!), and their snapshots look fine.