Implementation Plan: Machine-generated tags on the frontend #4302

obulat · 2024-05-10T15:53:08Z

Due date:

2024-05-24

Assigned reviewers

Both of you were chosen for your familiarity with the project and the frontend. Input from @fcoveram is also necessary from the design point of view: whether the proposed approach is feasible and good for the UX.

Description

Resolves #4039

Current round

This discussion is following the Openverse decision-making process. Information
about this process can be found
on the Openverse documentation site.
Requested reviewers or participants will be following this process. If you are
being asked to give input on a specific detail, you do not need to familiarise
yourself with the process and follow it.

This discussion is currently in the Decision round.

The deadline for review of this round is 2024-05-28

github-actions · 2024-05-10T16:00:04Z

Full-stack documentation: https://docs.openverse.org/_preview/4302

Please note that GitHub pages takes a little time to deploy newly pushed code, if the links above don't work or you see old versions, wait 5 minutes and try again.

You can check the GitHub pages deployment action list to see the current status of the deployments.

New files ➕:

https://docs.openverse.org/_preview/4302/projects/proposals/rekognition_data/20240510-implementation_plan_machine_generated_tags_frontend.html

AetherUnbound

Thanks so much for codifying this all into a document! I support the suggested approach broadly, just a few clarification pieces before a decision is made here.

AetherUnbound · 2024-05-10T17:25:13Z

...s/proposals/rekognition_data/20240510-implementation_plan_machine_generated_tags_frontend.md

+source and the generated tags. However, each section individually will still be
+normalized to prevent duplicates.


Ahh, this is a great way to handle that!

AetherUnbound · 2024-05-10T17:25:28Z

...s/proposals/rekognition_data/20240510-implementation_plan_machine_generated_tags_frontend.md

+there could be more tag sections. The tags in each subsection will be ordered by
+their accuracy.


Great call, I really like this approach.

What is the need behind this decision? I don’t envision users needing to distinguish between providers to make a better search decision. Showing the provider assumes that users know the difference between them, like how data is processed and therefore the output.

I would go with something simpler for this release, and once we reach users when #3559 begins, inquiry how users relate to machine-generated metadata during their searches.

This makes a lot of sense, @fcoveram. I'll update the IP to use alternative 1 as the selected choice.

I think it'd be best not to sub-divide by provider; later, we could follow-up and display that information some other way in the future.

AetherUnbound · 2024-05-10T17:34:02Z

...s/proposals/rekognition_data/20240510-implementation_plan_machine_generated_tags_frontend.md

+No infrastructure changes will be necessary, we will only update the single
+result pages of the Nuxt frontend.
+
+## Alternatives


Thank you for including the alternatives!

AetherUnbound · 2024-05-10T17:34:13Z

...s/proposals/rekognition_data/20240510-implementation_plan_machine_generated_tags_frontend.md

+
+The tags will be displayed in two sections: "Source tags" and "Generated tags".
+
+For the machine-generated tags, we want to add a link to the page that describes


It looks like this is missing a bit of information at the end.

Thank you for pointing this out. This was actually extra that should have been removed.

...s/proposals/rekognition_data/20240510-implementation_plan_machine_generated_tags_frontend.md

AetherUnbound · 2024-05-10T17:51:30Z

...s/proposals/rekognition_data/20240510-implementation_plan_machine_generated_tags_frontend.md

+## Risks
+
+<!-- What risks are we taking with this solution? Are there risks that once taken can’t be undone?-->
+
+## Prior art
+
+<!-- Include links to documents and resources that you used when coming up with your solution. Credit people who have contributed to the solution that you wish to acknowledge. -->


I think if we don't have any information for these sections, we can remove them.

zackkrida · 2024-05-14T22:14:41Z

@obulat I see there are a few places with missing text still, I can review this more deeply after those are addressed. I am also positive about the general approach outlined so far :)

obulat · 2024-05-15T18:29:28Z

@obulat I see there are a few places with missing text still, I can review this more deeply after those are addressed. I am also positive about the general approach outlined so far :)

@zackkrida, I updated the proposal, and it's ready for review

zackkrida · 2024-05-17T10:55:18Z

I will review this today.

AetherUnbound

Excellent, no blocking objections! This is so exciting!

AetherUnbound · 2024-05-17T17:52:10Z

...s/proposals/rekognition_data/20240510-implementation_plan_machine_generated_tags_frontend.md

+and the accuracy of the tag without any additional actions. We need to make sure
+that the tag name and its accuracy are separated (such as with an invisible,
+`sr-only`, dot) to make the screen readers add a pause between them.


Great call-out!

Since we're deciding not to go with this approach, do you mind changing the language (specifically We need to make sure...) to indicate that, if we were to choose this approach down the line, this is a consideration we'd need to take?

AetherUnbound · 2024-05-17T17:52:27Z

...s/proposals/rekognition_data/20240510-implementation_plan_machine_generated_tags_frontend.md

+<!-- Choose two people at your discretion who make sense to review this based on their existing expertise. Check in to make sure folks aren't currently reviewing more than one other proposal or RFC. -->
+
+- [ ] @zackkrida
+- [ ] @AetherUnbound


Suggested change

- [ ] @AetherUnbound

- [x] @AetherUnbound

openverse-bot · 2024-05-20T00:00:07Z

Based on the medium urgency of this PR, the following reviewers are being gently reminded to review this PR:

@zackkrida
This reminder is being automatically generated due to the urgency configuration.

Excluding weekend¹ days, this PR was ready for review 5 day(s) ago. PRs labelled with medium urgency are expected to be reviewed within 4 weekday(s)².

@obulat, if this PR is not ready for a review, please draft it to prevent reviewers from getting further unnecessary pings.

Specifically, Saturday and Sunday. ↩
For the purpose of these reminders we treat Monday - Friday as weekdays. Please note that the operation that generates these reminders runs at midnight UTC on Monday - Friday. This means that depending on your timezone, you may be pinged outside of the expected range. ↩

fcoveram · 2024-05-20T09:08:56Z

...s/proposals/rekognition_data/20240510-implementation_plan_machine_generated_tags_frontend.md

+### Add "About tags" description page
+
+We should add a page to explain the users the difference between where the
+Openverse tags come from, the difference between the source and generated tags,
+and a short description of the machine-generated tag providers. This page should
+be linked from the Generated tags section.


I agree with adding content to explain tag are generated, but I disagree with adding a 10th page in the content header.

The current header is quite long, and adding an extra link in the current set doesn't provide enough information about what "tags" is. Instead, I would address the content need as proposed here for #3564.

The page could be very simple: reuse the current grid and page style, and add a fixed TOC on the left to jump between sections. Since the section will be linked from the search page, the information is introduced with more context about what Openverse is about and how relates to sources and content.

I wrote "similar to the sensitive content page" several times, and deleted it. I guess I should have left it in.

What I meant here was we should have a page explaining what the machine-generated tags are, but not add it to the header. The sensitive media page is just like that: you can only navigate it by typing the address in the URL bar, or by opening a sensitive media item and clicking on a link from that page. Do you think this would work, @fcoveram ?

What I meant here was we should have a page explaining what the machine-generated tags are, but not add it to the header.

"Add a page" resonated with me as adding it in the header as it's the only component where we add pages. But you're right, it doesn't mention the header at all.

...s/proposals/rekognition_data/20240510-implementation_plan_machine_generated_tags_frontend.md

fcoveram · 2024-05-20T09:24:16Z

Very clear document. I left comments related to showing how tags are generated.

In short, I agree with the content need and embrace the "giving all the details" approach, but I think we can introduce the info with more context more smoothly.

zackkrida · 2024-05-20T16:35:16Z

...s/proposals/rekognition_data/20240510-implementation_plan_machine_generated_tags_frontend.md

+slot in the `VTag` component will enable us to display the additional accuracy
+information.
+
+### Add "About tags" description page


Good 👍 at some point in the future we'll probably need to start thinking about the page hierarchy on openverse.org and what should be on docs.openverse.org or openverse.org. Also, if we thinks docs.openverse.org is appropriate for users or only contributors and API consumers.

zackkrida

Agreed we should skip disambiguating providers from each other on first pass; otherwise LGTM.

...s/proposals/rekognition_data/20240510-implementation_plan_machine_generated_tags_frontend.md

obulat · 2024-05-21T15:10:05Z

Drafting to update

AetherUnbound

Two small notes, but no blocking objections!

AetherUnbound · 2024-05-23T21:23:38Z

...s/proposals/rekognition_data/20240510-implementation_plan_machine_generated_tags_frontend.md

+   page to display this data in full. As a bonus, if a tag is generated by
+   multiple providers, the user will see that different providers generate the
+   same tag, and will be able to see the difference in accuracy between the
+   providers.


Similar to point 1, do you mind adding a note here on why this was not an ideal choice right now?

AetherUnbound · 2024-05-23T21:25:06Z

...s/proposals/rekognition_data/20240510-implementation_plan_machine_generated_tags_frontend.md

+and the accuracy of the tag without any additional actions. We need to make sure
+that the tag name and its accuracy are separated (such as with an invisible,
+`sr-only`, dot) to make the screen readers add a pause between them.


Since we're deciding not to go with this approach, do you mind changing the language (specifically We need to make sure...) to indicate that, if we were to choose this approach down the line, this is a consideration we'd need to take?

zackkrida

Agreed with @AetherUnbound's suggestions. LGTM

Signed-off-by: Olga Bulat <obulat@gmail.com>

…lementation_plan_machine_generated_tags_frontend.md Co-authored-by: Madison Swain-Bowden <bowdenm@spu.edu>

Signed-off-by: Olga Bulat <obulat@gmail.com>

obulat requested a review from a team as a code owner May 10, 2024 15:53

obulat requested review from krysal, AetherUnbound and zackkrida and removed request for krysal May 10, 2024 15:53

github-actions bot added the 🧱 stack: documentation Related to Sphinx documentation label May 10, 2024

openverse-bot added 🟨 priority: medium Not blocking but should be addressed soon 🌟 goal: addition Addition of new feature 📄 aspect: text Concerns the textual material in the repository labels May 10, 2024

AetherUnbound reviewed May 10, 2024

View reviewed changes

obulat force-pushed the add/machine-generated-tags-frontend-ip branch from e735032 to 33af2a0 Compare May 15, 2024 18:28

AetherUnbound approved these changes May 17, 2024

View reviewed changes

fcoveram reviewed May 20, 2024

View reviewed changes

...s/proposals/rekognition_data/20240510-implementation_plan_machine_generated_tags_frontend.md Outdated Show resolved Hide resolved

zackkrida reviewed May 20, 2024

View reviewed changes

zackkrida approved these changes May 20, 2024

View reviewed changes

AetherUnbound added the 🧭 project: implementation plan An implementation plan for a project label May 20, 2024

fcoveram reviewed May 21, 2024

View reviewed changes

...s/proposals/rekognition_data/20240510-implementation_plan_machine_generated_tags_frontend.md Show resolved Hide resolved

obulat self-assigned this May 21, 2024

obulat marked this pull request as draft May 21, 2024 15:10

AetherUnbound mentioned this pull request May 21, 2024

Incorporate Rekognition data into the catalog #431

Open

4 tasks

obulat force-pushed the add/machine-generated-tags-frontend-ip branch 2 times, most recently from 7d38ec9 to 5323d54 Compare May 23, 2024 14:20

obulat force-pushed the add/machine-generated-tags-frontend-ip branch from 5323d54 to 6216619 Compare May 23, 2024 15:19

obulat marked this pull request as ready for review May 23, 2024 15:19

obulat requested review from AetherUnbound and zackkrida May 23, 2024 15:19

AetherUnbound approved these changes May 23, 2024

View reviewed changes

zackkrida reviewed May 23, 2024

View reviewed changes

obulat and others added 6 commits May 24, 2024 12:26

Add the IP for machine generated tags on the frontend

55b6570

Signed-off-by: Olga Bulat <obulat@gmail.com>

Update documentation/projects/proposals/rekognition_data/20240510-imp…

3bc6b24

…lementation_plan_machine_generated_tags_frontend.md Co-authored-by: Madison Swain-Bowden <bowdenm@spu.edu>

Update documentation/projects/proposals/rekognition_data/20240510-imp…

0d978c9

…lementation_plan_machine_generated_tags_frontend.md Co-authored-by: Madison Swain-Bowden <bowdenm@spu.edu>

Add missing data

00d45a2

Signed-off-by: Olga Bulat <obulat@gmail.com>

Update the IP to NOT display tag accuracy/provider

48ffbaf

Signed-off-by: Olga Bulat <obulat@gmail.com>

Update the wording

7b7df91

Signed-off-by: Olga Bulat <obulat@gmail.com>

obulat force-pushed the add/machine-generated-tags-frontend-ip branch from 6216619 to 7b7df91 Compare May 24, 2024 11:24

Add approvals

44ba6e8

Signed-off-by: Olga Bulat <obulat@gmail.com>

obulat merged commit 87880a0 into main May 24, 2024
41 checks passed

obulat deleted the add/machine-generated-tags-frontend-ip branch May 24, 2024 11:39

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Implementation Plan: Machine-generated tags on the frontend #4302

Implementation Plan: Machine-generated tags on the frontend #4302

obulat commented May 10, 2024 •

edited

Loading

github-actions bot commented May 10, 2024

AetherUnbound left a comment

AetherUnbound May 10, 2024

AetherUnbound May 10, 2024

fcoveram May 20, 2024

obulat May 20, 2024

zackkrida May 20, 2024

AetherUnbound May 10, 2024

AetherUnbound May 10, 2024

obulat May 15, 2024

AetherUnbound May 10, 2024

zackkrida commented May 14, 2024

obulat commented May 15, 2024

zackkrida commented May 17, 2024

AetherUnbound left a comment

AetherUnbound May 17, 2024

AetherUnbound May 23, 2024

AetherUnbound May 17, 2024

openverse-bot commented May 20, 2024

fcoveram May 20, 2024

obulat May 20, 2024

fcoveram May 20, 2024 •

edited

Loading

fcoveram commented May 20, 2024

zackkrida May 20, 2024

zackkrida left a comment

obulat commented May 21, 2024

AetherUnbound left a comment •

edited by zackkrida

Loading

AetherUnbound May 23, 2024

AetherUnbound May 23, 2024

zackkrida left a comment

		source and the generated tags. However, each section individually will still be
		normalized to prevent duplicates.

		there could be more tag sections. The tags in each subsection will be ordered by
		their accuracy.


		The tags will be displayed in two sections: "Source tags" and "Generated tags".

		For the machine-generated tags, we want to add a link to the page that describes

Implementation Plan: Machine-generated tags on the frontend #4302

Implementation Plan: Machine-generated tags on the frontend #4302

Conversation

obulat commented May 10, 2024 • edited Loading

Due date:

Assigned reviewers

Description

Current round

github-actions bot commented May 10, 2024

AetherUnbound left a comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

zackkrida commented May 14, 2024

obulat commented May 15, 2024

zackkrida commented May 17, 2024

AetherUnbound left a comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

openverse-bot commented May 20, 2024

Footnotes

Choose a reason for hiding this comment

Choose a reason for hiding this comment

fcoveram May 20, 2024 • edited Loading

Choose a reason for hiding this comment

fcoveram commented May 20, 2024

Choose a reason for hiding this comment

zackkrida left a comment

Choose a reason for hiding this comment

obulat commented May 21, 2024

AetherUnbound left a comment • edited by zackkrida Loading

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

zackkrida left a comment

Choose a reason for hiding this comment

obulat commented May 10, 2024 •

edited

Loading

fcoveram May 20, 2024 •

edited

Loading

AetherUnbound left a comment •

edited by zackkrida

Loading