Benefit of snapshots of journal article listing webpage

Gurdas_Sandhu

When I create journal article items, from a database or journal website, using the browser bar utility, Zotero will also attach a snapshot of the page. Is there any merit in saving webpage snapshot for journal articles? I am thinking of deleting all such snapshots in my library (over 1,000), but want to know if I am overlooking some future benefit of keeping the snapshots. Note, I am talking ONLY about snapshots for journal articles.

I have disabled auto PDF download for various reasons but let snapshots remain because for some webpages I prefer having a snapshot since there is no guarantee that page will be around next year. For journal articles, the way I see it, a snapshot of the webpage where the particular article is listed is almost meaningless because the articles are referenced through volumes, series, and/or DOIs.

adamsmith

No compelling reason, no -- there may be keywords on the page, the abstract may not have gotten picked up by Zotero and a number of things like that, but there's generally often no compelling reason to get snapshots. Generally I'd be inclined to reduce the number of snapshots we attach, especially where we're already attaching PDFs, but that takes some work.

Gurdas_Sandhu

I agree with that, adamsmith. Since I am careful about keywords and abstract when the item is created, I see no benefit from keeping the snapshot. Will start the deletion process.

I understand there is no automated or semi-automated way to delete snapshots for journal articles and I have to expand each article and use the "Move Item to Trash" from right-click menu, right? Can I assign the keyboard shortcut to avoid the right-click and select step?

dstillman

You can search for "Snapshot" (or similar) and press Ctrl/Cmd-A to Select All, which will only select the matched (black) items, and then move the items to the trash at once.

Gurdas_Sandhu

That helps, though over 70% of the snapshots have names same as article name (and do not have the word "snapshot"). The phrase "full text" grabs another 20%.

I tried an advanced search for "Item Type is Journal Article" and "Attachment File Type is Web Page", but cannot find a way to show only the child items.

adamsmith

if you do that search with all checkboxes unchecked and then create a saved search, only the attachments will actually be included in the saved search (i.e. black) and you can use select all --> delete as described by Dan.

Gurdas_Sandhu

I am not getting that same behavior. With all checkboxes unchecked, I see no results. If I check the "include parent and child items..." checkbox, then I see everything.

dstillman

Don't use Item Type is Journal Article — that selects just the parents.

Gurdas_Sandhu

I seem to be missing something here. If I do not specify item type journal article, then won't I end up selecting all items that have web page attachment? I want to delete web page attachments ONLY for journal articles.

dstillman

Oh, if you want to limit to journal articles, here's a way to do it:

1) Create one saved search with [Item Type] [is] [Journal Article] and "Include parent and child items of matching items" checked.

2) Create another saved search with three conditions: [Saved Search] [is] [name of first search], [Attachment File Type] [is] [Web Page], and [Title] [does not contain] ["link"].

The second search should then match only snapshots (and links, which you may actually want to keep) for only journal articles, and you can use Select All and move all the snapshots to the trash at once (perhaps after manually deselecting some, such as links).

Gurdas_Sandhu

I can't seem to replicate that behavior. For the second saved search, if all check boxes are unchecked, I get no results. If the parent/child box is checked, the results are same as the first search. What is the third line item of the second search meant to do?

FYI, 99% of my journal article items have a link to a PDF of the article. I do not directly attach the PDF file.

dstillman

Sorry, meant to say [does not contain] ["link"] instead of [is not] for the third condition of the second search, but you can also just leave that condition out — it's just meant to exclude some web link attachments.

But if you're getting no results, I think you're doing something wrong. I have this working.

Gurdas_Sandhu

Hmmm... I can't get this to work. I created a test condition, the results of which are puzzling. Removed the third line item from the second search for simplicity. Created two versions of the second search. In version 1, [Attachment File Type] [is] [Web Page] and parent/child is checked. In version 2, [Attachment File Type] [is not] [Web Page], parent/child box status does not matter. Both searches return the exact same items. For version 1, I picked an item that has a webpage attachment and sent the attachment to trash. Yet, it continue to show in the search. What could I be doing wrong?

dstillman

Can you take screenshots of the saved search edit windows for both searches — the two I described — and upload them somewhere (e.g., Dropbox, imgur.com) and post links here?

Gurdas_Sandhu

I put the screenshots in a PDF. See:

https://drive.google.com/file/d/0B9roN3pm2S3hcWpXZjJrQ0pidVU/view?usp=sharing

dstillman

Yeah, that's wrong. "Include parent and child items of matching items" should be checked on the first search, not the second.

dstillman

The second search will only ever select from the matches (i.e., black items) in the first search, so you need the parent/child box on the parent to be selected so that both parents and children are matched. The parent/child box shouldn't be selected on the second because you're trying to select only the children.

Gurdas_Sandhu

What was I thinking/seeing !!?? Quite an embarrassing case of not following simple directions.

Fixed the error and it is working now. Thanks, Dan!

Gurdas_Sandhu

Is there a way to narrow down the results to only those items that also have a link to a PDF as the child. I mean to delete webpages only for those journal article items for which I have already download and attached a PDF. Note, I only attach link to the PDF on my HDD.

May be I can first find items with PDF link attachment and assign a tag and then include a test for tag in the second search?

Edit: I already have a tag assigned to all items for which a PDF is not yet available. May be I can use that?

adamsmith

you'd have to experiment a bit with the searches, but generally using tag -- is not -- (and then your tag) should work. I think you need to set it as an additional condition on the first search and then it should work fine.