webrecorder / browsertrix

Browsertrix is the hosted, high-fidelity, browser-based crawling service from Webrecorder designed to make web archiving easier and more accessible for all!
https://webrecorder.net/browsertrix
GNU Affero General Public License v3.0
201 stars 35 forks source link

fix: Archived item navigation improvements #2062

Closed SuaYoo closed 2 months ago

SuaYoo commented 2 months ago

Resolves https://github.com/webrecorder/browsertrix/issues/2056

Changes

Screenshots

Page Image/video
Crawl detail Screenshot 2024-09-03 at 3 40 33 PM
Crawl detail - Crawl Settings Screenshot 2024-09-03 at 3 54 28 PM
Upload detail Screenshot 2024-09-03 at 3 41 41 PM
Crawl detail - From collection Screenshot 2024-09-03 at 3 41 08 PM
Shrinks99 commented 2 months ago

The back to crawl button on the QA page is a great addition. Similar to how I don't think we should use the menu actions for navigation when possible, I think we should move this to the overview section along with the workflow name and a link listed the same way we list collections in the metadata section. This isn't a completely holistic take — we use the settings gear buttons in the actions section of our panels and that functions as a link — but this feels more like navigating to a different section than configuring the current item to me.

ikreymer commented 2 months ago

Is there a reason to have the 'Viewing Workflow Crawl' label?

Screenshot 2024-09-01 at 8 27 14 PM

Wondering if it may add more confusion to have that

SuaYoo commented 2 months ago

Similar to how I don't think we should use the menu actions for navigation when possible, I think we should move this to the overview section along with the workflow name and a link listed the same way we list collections in the metadata section. This isn't a completely holistic take — we use the settings gear buttons in the actions section of our panels and that functions as a link — but this feels more like navigating to a different section than configuring the current item to me.

IMO the "In Collections" has always felt buried in archived item details. Having it in the metadata section seems slightly inaccurate, since information on the workflow or collection is more significant than metadata descriptors attached to the crawl.

What if we renamed "Crawl Settings" to "Workflow" and showed a primary action button in that section to "Edit Workflow"?

Shrinks99 commented 2 months ago

What if we renamed "Crawl Settings" to "Workflow" and showed a primary action button in that section to "Edit Workflow"?

Adding the edit workflow button there seems like a good addition to me!

I would be against changing the name though. The communication difference is that these are the workflow settings that the crawl ran with, not the current workflow settings. We could probably further reinforce that with a bit of descriptive text below the title... Should maybe come up with some standards for adding that in other places! I know you've already started with a few of them!

Shrinks99 commented 2 months ago

From today's call:

The path bar implementation is good but Archived Items living cannonically at two different paths (under both crawl workflows and archived items) isn't terribly consistent. Instead, we will double down on the crawl workflow → archived item parent/child relationship with crawls living within crawl workflows. Uploaded archived items will continue to live under /items.

Screenshot 2024-09-09 145055

SuaYoo commented 2 months ago

New approach makes sense! Only nitpick is with the crawl breadcrumb showing the finished date, should we show the crawl id? Crawls / Finished at 08/30/24, 1:01 AM (sched-18cf78aa-abc-28750080) or Crawls / sched-18cf78aa-abc-28750080 (Finished at 08/30/24, 1:01 AM) so that there's a way to relate the crawl id, which users might want for API, etc..

The crawls are currently identified by name and finished (created in archived items) date. I think it'd make sense to show a short ID if it's also how we identify the crawl in a list, otherwise I think it'd be fairly meaningless to the majority of non-API users.