Modernization of the Django admin

[Antoliny0919]

Code of Conduct

• I agree to follow Django's Code of Conduct

Feature Description

Modernizing the Django admin.
I feel that improvements to the UI/UX are needed.

Problem

Improving the UI/UX of the Django admin.
The backend of the Django admin is also built with older code and has many template tags centered around ChangeList, which I think can be somewhat problematic. However, I believe that what’s more needed is the improvement of the UI/UX rather than the backend.
(The exact scope of modernization may need to be discussed.)

Request or proposal

proposal

Additional Details

Among third party Django admin packages, there are already some that have excellent themes.
As an example, here are some packages that personally have designs I prefer.

• https://github.com/unfoldadmin/django-unfold
• https://github.com/farridav/django-jazzmin

Implementation Suggestions

I feel that making partial UI/UX improvements within the Core is somewhat difficult. While there are good approaches available, in practice it is challenging due to additional work, such as slightly modifying the structure. Moreover, improving only part of the UI can actually feel awkward.

I wonder if it would be possible for Django to provide a public separate package with a UI/UX enhanced admin theme. I believe such a Django admin theme package should prioritize compatibility so that it does not break existing customizations of the Django admin template when applied. It should also adopt the most “Django-like” design in terms of colors, fonts, and overall style.

Finally, it should consider accessibility standards that is, it should be designed for everyone, including those in minority groups.

Merging this into Core is not strictly necessary, and it would be fine to merge later if it gains popularity and poses no significant issues. Therefore, I think it would be possible to offer an improved UI/UX theme without any pressure.

[github-actions]

Thank you Antoliny0919 for sharing your idea! We have a lot of them so please be patient. You can see the current queue here.

Community instructions

For commenters, please use the emoji reactions on the issue to express support, and/or concern easily. Please use the comments to ask questions or contribute knowledge about the idea. It is unhelpful to post comments of "I'd love this" or "What's the state of this?"

Reaction Guide

• 👍 This is something I would use
• 👎 This is something that would cause problems for me or Django
• 😕 I’m indifferent to this
• 🎉 This is an easy win

[nanuxbe]

So I voted 👍 on this one. The UI/UX is important but there has also been talks of rewriting the admin in more modern Django, This has been on the Steering Council radar but we are also aware that such an endeavor would most likely need to be a sponsored feature

[Antoliny0919]

So I voted 👍 on this one. The UI/UX is important but there has also been talks of rewriting the admin in more modern Django, This has been on the Steering Council radar but we are also aware that such an endeavor would most likely need to be a sponsored feature

Rewriting it would be an enormous undertaking. I think I’ve seen discussions before about separating the admin. The proposed transition to Playwright could also happen alongside this. However, if this work goes beyond UI/UX and involves a full rewrite, I also feel it would definitely be impossible without someone sponsoring it. Even if it takes time, it seems likely that one or two main maintainers would be needed consistently. Still, it would be great if individual contributors could also participate in this project, since there are contributors who, even without compensation, are willing to invest a lot of time and have a strong attachment to the admin.
I’m not sure how much help I can be, but I’m ready to participate even without compensation. 😁

[knyghty]

I don't think it needs sponsoring per se but it definitely would help.

This is something I've spent some time (but not much energy) thinking about. I think as Django's killer app (well, one of them, there's also the ORM), the admin should live in core. I also think it needs a rewrite. What I'm most in favour of at the moment is making a newadmin in the vein of newforms for those old enough to remember this. We could consider it experimental. We could consider it outside of core to begin with and bring it in when stable. All options, depending how those other issues go. But I don't think there would be much interest in working on this if there wasn't a plan to merge it before that work started.

We would then need to think about if the interface should be compatible (I suspect this would be ideal). And if there needs to be feature parity (I suspect not, but somewhere close).

Renewing the UI is nice, but part of me thinks that some parts of that are really difficult because of the rather ancient code that's supporting it.

[Antoliny0919]

I am in favor of a rewrite. Personally, I think it would be fine to start by adding a separate repository under the Django organization (it could also be moved there after some initial progress). However, there are still many concerns.

When it comes to the question of how this work should proceed in order to be successfully completed, I’m not sure how to answer yet. I think someone will need to take on significant responsibility for it to reach completion. Because of this, I believe discussions about sponsorship are inevitable.

If a rewrite is undertaken, I would definitely prioritize compatibility. I also think it’s reasonable for the functionality to remain the same as the existing admin, with perhaps some minor additions.

Renewing the UI is nice, but part of me thinks that some parts of that are really difficult because of the rather ancient code that's supporting it.

I believe anyone who has tried working on the Django admin UI/UX would strongly relate to that last point.

I hope this work does not get delayed, as I believe it is something that eventually needs to be done 😁

[nanuxbe]

If a rewrite is undertaken, I would definitely prioritize compatibility. I also think it’s reasonable for the functionality to remain the same as the existing admin, with perhaps some minor additions.

I don't agree with this fully. A rewrite will introduce breaking changes for the simple reason that there are some ugly/undocumented parts in the current admin. The point of a rewrite would partially be to get rid of those things and therefore be incompatible. If "compatibility" is put forth, someone will inevitably come forward saying the new admin breaks their code because they have been monkey-patching this obscure private method somewhere for years and now it's gone.

I do agree that we need high-level compatibility (unless there is a good reason against that). Things like top-level public parameters to the ModelAdmin class maybe. But even there, this would prevent some much-appreciated improvements (like in relationship to inlines positioning and ordering for example)

We could consider it outside of core to begin with and bring it in when stable. All options, depending how those other issues go. But I don't think there would be much interest in working on this if there wasn't a plan to merge it before that work started.

Even with interest in creating the code, there is very little point in debating this if there is no pre-existing agreement that the goal is to merge it into core. I keep pointing to django-admin2 that has existed for 12 years and is already very close to a django-esque way of doing an admin, yet it just went to be forgotten and left to decay. For such an undertaking, this should be avoided.

[fabiocaccamo]

Among third‑party Django admin projects, there is also django‑admin‑interface (which I developed and actively maintain).
It enhances the native Django admin UI without altering or rewriting any of its underlying logic.

It's been widely used by the community and in production for many years.

https://github.com/fabiocaccamo/django-admin-interface

[lukasvinclav]

@Antoliny0919 if there is something what I can do. Let me know.