Rethinking the ACP home page

[Arantor]

If you're staying on topic in a forum you're kind of doing it wrong

But more seriously, I'm pretty adamant that the correct way forward is to not try to do things as mods, only to redo them if they become core, but to just fix core.

I think it's possible to build something cool into the ACP that is useful for people and makes the ACP landing page more useful - it's just that what that 'useful' is will vary for different people. Solving the notifications of updates problem is trivial - SMF *already does this* given that the upgrade message appears in the package manager as well as the ACP home page. (I think the solution is awful internally, but that's a completely different discussion.)

Thus the suggestion of making the home page more modular so people can add/remove the things they care about - as part of a larger strategy to make functionality modular and exposed to people in places and ways that make sense for them. Give them the tools to help their community; we're never going to know how everyone uses their community, so let's give them the tools to shape it how they see fit.

(This of course builds onto the strategy outlined in previous topics relating to pages and blocks.)

[FrizzleFried]

Should there already be a 2.2 dedicated sub-forum for things like this?  I mean,  I'd think that once a version shipped,  the a sub-forum for the next version should be created to start collecting ideas for the next "big" update?

EDIT: Wait... maybe that is what "Feature Requests" is intended for?  That said... it sure sounds like "Feature Request" is for the current version... but it seems SMF team doesn't want to add 'features' to 2.1... so perhaps a title change to "Feature Requests for 2.2" or in some way incorporate 2.2 in there would be in order?

[Arantor]

Feature Requests is 'things for a future version', it always has been. And there's a gulf between whether a feature might be 10 lines of code vs 10,000 lines of code but still be eligible for 'the next version'.

[Kindred]

yes, what Arantor said.

The problem is that 2.1 took so long (without getting into the reasons) that we did release some feature updates in the 2.0 branch and some people are ultra focused on exceptions from the past rather than looking at the future.

So, doing that again is not what the plan is going forward.
2.1.x are point releases to correct bugs, security and support
2.2 is the next release branch.

(and no, support does not include changing things that WORK AS INTENDED, even if you don't like the way it works)

There IS a roadmap/plan. The Devs have indicated do not feel comfortable sharing that publicly yet (and it is their call on what/when things like that get shared). I understand Arantor's frustration -- but, as a developer, I'm sure he also understands their point of view at the moment)

[Arantor]

but, as a developer, I'm sure he also understands their point of view at the moment

No, not really.

We're four months in from 2.1's release with not even a *hint* of what's to come. Not even to rule things out for the next version, to indicate 'this is not in line with our general direction'.

Especially since there's no indication how far off said roadmap is - it might be next week, it might be next year. And then you've actually got to build it.

At this pace, I'm betting it would be quicker to get from scratch to SMF 1.0 again that it would to get from 2.1 to next version - and that was ~18 months. Of which 4 months has been spent coming up with a plan.

So, no, I really don't understand the point of view. I'm also not convinced I want to be here by then anyway.


EDIT: The other choice is opening the roadmap up sooner rather than later with 'here's what we're thinking of, what do you think' and involving the community as soon as possible. This approach absolutely favours a quick iterative release cycle. If you're not going for that, though, a roadmap that is published is favourable - but I would argue that there needs to be more engagement with the community EITHER WAY.

From the sidelines, the silence is utterly deafening.


EDIT 2: Engaging with the community was the best thing to come out of Wedge. Every experiment, every thought, every idea whether implemented or not, we talked about it. If it didn't go down well, we didn't include it. If it did, we refined it.

[Sir Osis of Liver]

@Sir Osis of Liver, you deeply misunderstand my intent here.

No, I do pretty much understand, it's just the frustration factor of having tried to contribute some suggestions going back to alpha, all of which were rebuffed or ignored, and seeing some of those same things now being reconsidered when it's too late to do anything for 2.1.   There's also the time factor of being old and fading fast.

As for your original ACP, the only thing that was really wrong with it was retaining a lot of the 2.0 clutter and placing it above the new cpanel style menu, pushing the menu offscreen.  Wasn't kidding when I posted that didn't know it was there until a couple versions into beta.  There was no reason to scroll down looking for it because the dropdown menus were still across the top.  I prefer sidebars, but the cpanel menu is also a good idea, and something most admins are familiar with.  If you clean up and reorganize the page, something like I posted above, it works well.

[Arantor]

How small is your screen that you had to scroll to see it?

I also can't speak for what happened around here after I left the dev team because after that my life somewhat upended itself so any feedback that got ignored was... out of my control entirely.

But I hear what you're saying. I feel in much the same boat - that trying to actually sort out the right path is simply a waste of time.

Mind you, you're still missing my point - 'too late for 2.1' is 'must be in 2.2 which needs to be soon'. However I take your point that this doesn't seem even remotely soon.

[Sir Osis of Liver]

I have a 23" monitor running at reduced resolution (1280x720), makes everything larger (poor vision).  This is what I see -

You cannot view this attachment.

If I rearrange it I see this -

You cannot view this attachment.

 

[Arantor]

Ahhh, I see what you mean.

What I will do is show you the *first* version of the design that was more than a full year before it made it into SMF (this was February 2012). The blocks were already gone in my world. I will also posit that, at the time, the whole PHP/MySQL version thing was so much less of a drama that it just didn't matter nearly so much to find this out in any hurry (and the other things that are on the support block... really don't come up that often)

As a bonus, here's also a second screenshot of an evolution I did that had a few 'you seem to be new, do you want to x' tasks, that you could hide.

Can't believe this was more than 10 years ago now. How time flies.

[Sir Osis of Liver]

That's pretty much what I thought when I first saw it, if you're going to develop a completely new menu system it should be featured prominently on the page, and get rid of the dropmenus, they're no longer needed.   Believe I posted some code on team board that did basically that, but no one ever looked at it.

[Arantor]

The dropmenus were there to prevent you having to go back to the front page every time, but also Nao was a huge fan of using them to drill down to whichever page he wanted.

Honestly, going back through my old screenshots on Wedge was a weird trip, seeing all the ideas I'd had back in the day, not all of which made it into production code.

[Antechinus]

The Wedge shots are a saner arrangement. Although they do still suffer from the text being less conspicuous than the icons, and centre-aligned. Something like cPanel's presentation is easier to scan, IMO.

The problem with colourful icons is that they may look cute, but they draw the eye, and nobody will actually remember all of the icons and what functions they relate to. So, although they may generate favourable reactions if you put it out there and everyone goes "Oh cool, look at the icons!" (which they will do, just because they haven't seen it before) when it comes to actually using it people will have to look past the icons and read the text to figure out where they want to go.

So, making it easier for people to read the text is going to make it more usable, in practice, regardless of anyone's knee-jerk reaction to seeing lots of pretty icons.

[Arantor]

cPanel looked like that in 2012 though (complete with even more unmemorable icons, and centered text). So did macOS (complete with more memorable icons and centered text). Those were the design cues I took from which I've never made a secret of

I dunno about the assessment of usability though. I think in hindsight I'm at the point where I don't really look at *either* consciously and go off some combination of vaguely recognising the combination (similar to people recognising avatars + usernames as a unified thing) and vaguely off muscle memory. Which is why I'm thinking what seemed like a good idea back then isn't now.

[Sir Osis of Liver]

Have you seen the new cpanel ui?  They went from this -

You cannot view this attachment.

To this -

You cannot view this attachment.

 

[Arantor]

I remember when Paper Lantern was first announced (2013? 2014? 2014 blog post introducing it) and I remember that having more punchy colours and centralised text.

Honestly, I remember cPanel looking like the attachment. As I said, when I made this, it was very much in line with what was contemporary.

I'm not responsible for people not making it better in the 9-10 years since I made it and even I agree it could be done better now!

[Oldiesmann]

Since we're talking about control panels, here's how Plesk does things. Ironically, Plesk owns cPanel (and has for years) but the two projects are completely separate from one another. Note that some of the items here are plugins (Immunify, PageSpeed, Softaculous), but you get the point.

You cannot view this attachment.

[live627]

Here's WordPress. I really like how the menu section expands when active, revealing more options.

You cannot view this attachment.

I have a 23" monitor running at reduced resolution (1280x720), makes everything larger (poor vision). 

I do a similar thing, except do DPI scaling at full resolution... keeps text sharp. Too blurry/blocky when downscaled.