Whats Happening with UI Development?

Has the UI/Usability team activity pretty much fizzled/been killed off (not prioritized)? Is the “low hanging fruit” sitting untouched? I think this is what was feared once it started… that the momentum was only a temporary event and now we are back to the old “business as usual” attitudes.

Any honest feed back on whats going on with that? Emphasis on honesty.

To answer your question, not much and its not being prioritized.
And theres not that much low hanging fruit remaining.

Though Im not sure why you expect dishonesty…?
Being open means - if you don’t see plans to work in an area. Chances are high that there are none.
(not near term anyway).

Julian Severin has been working on a few UI related development projects as of late (some of his stuff is even in review I think). Though I’m guessing the main thing that is wanted is the new keymap (which unfortunately hasn’t seen much work as Johnathon W never seems to find the time).

That’s a shame, kind of goes against the initial premise that the UI/UX development would be given some serious attention after the BlenderCon which had a heavy UI/UX discussion. Why do you not prioritize user experience and usability?

I’d be curious to see what you or others think of the lack of low hanging fruit remaining and if that should mean low priority on bigger developments.

> Though Im not sure why you expect dishonesty...? > Being open means - if you don't see plans to work in an area. Chances are high that there are none. > (not near term anyway).

Not everyone is as forthcoming with such information. It seems more common to deflect or suggest otherwise just to shut the line of inquiry up. To your and Jonathan W.'s credit, neither of you seem to have a habit of doing that. The request for honesty means I’d rather only have someone answer it if they can be honest and direct. Thanks for the answer.

@SaintHaven, OpenGL viewport refactor and OpenSubDiv projects are getting priority at the moment.

UI is complicated & divisive topic, where people with best intentions can strongly disagree… so its not as simple as just having having developer time to devote.

Having said that, Julian is doing some great work on the UI, I don’t want to detract from that. If you check logs for past few releases, UI has seen some real improvements too. But I suspect you are thinking of kinds of improvements we would see from a team working full-time over months (not a few devs making improvements when they find the time).

Both worth pursuing, no doubt about that… as long as the OpenGL improvements can carry over to the next gen APIs coming out.

On the UI being complicated and divisive. It doesnt need to be, wasnt that the purpose of creating a UI team with module owners. They can take feedback but ultimately they should go through with their design otherwise nothing will happen. Ton also said in the re-defining blender blog that a team would be assembled which can tackle the UI issues via a UI team. So if thats not a priority for that team, it… from the outside looking in, feels as though theres not much of a dedicated team to begin with. Does the team list need to be updated?

Having said that, Julian is doing some great work on the UI, I don’t want to detract from that. If you check logs for past few releases, UI has seen some real improvements too. But I suspect you are thinking of kinds of improvements we would see from a team working full-time over months (not a few devs making improvements when they find the time).

And that is certainly a good thing… but yes I was thinking more on the team aspect as is advertised on the developer.blender.org site as well as what came from the october 2013 discussions with the BF. The wiki still lists at the very least 2.75 having some big targets but it appears thats not even close to being a reality… I would then naturally have to question how hard they are trying to meet those milestones.

I know its probably annoying to constantly be questioned on the subject, especially if some are mere volunteers… but sometimes nothing happens until theres a bit of pressure to follow through with the previously posted goals. ( http://wiki.blender.org/index.php/Dev:Doc/Projects/UI )

Well, at the moment I don’t now if anyones trying to meet any UI milestones, (feel free to correct me if am wrong), maybe a dev works on one or two of them.

We have this happen a bit, where enthusiastic people get together and plan to make big changes, sometimes it works, other times it energy/interest in the project fades.

Updating the team list only helps if the new members are able to complete tasks previously left unfinished, define new goals etc… (at the moment this doesn’t seem likely since we don’t have new people active in this area).


regarding: http://wiki.blender.org/index.php/Dev:Doc/Projects/UI
I don’t agree with these previously posted goals, and have commented before that this list was more of a wish list than realistic development targets.
The main issue with this page is a bunch of goals were posted and nobody was responsible for completing them before the next release.
Though it seems the list was cleaned up a bit since then and is a little more reasonable in its scope.

Anyway, I don’t like to diminish the work of the guys who make ambitious plans even if they don’t always succeed. But as something for the community to expect to be completed… this gets a bit tricky to balance.

I agree on that, also for the keymap, I doubt it will be as good as people think. There are simply to many possible workflows here to make all happy. I think only a way to drag and drop operators in UI would make sense like in Libreoffice or Firefox. But it’s something user can fix with a bit of python learning, so it’s nice to have but not impossible wihtout. What isn’t fixable by artists is consistency (making tools available in all relevant contexts and with all relevant object types and with comparable behaviors). Blender may be weird sometime for new users, but when you can reuse what you learn in many contexts inside Blender, it’s ok. Is there or could we create a list of inconsistencies? Thank you for starting this already anyway with your fixes.

I just thought the same. Thanks from me too, Campbell, for taking the time to answer user questions and interacting here on BA in general! It is much appreciated.

I agree, the overall tone of Blender-related discussion on this forum has seen a noticeable reduction in pessimism and cynicism ever since Campbell became more active in them. I tend to want to do what I can to fill in the blanks based on what I read on the sites run by the BF, but you can’t match the accuracy and detail of what you get from a dev.

The goal of the UI Team was (and is), amongst other things, to push through a lot of this divisiveness. But even then it’s very difficult when we simply haven’t had the time do it. It’d be one thing if there were several of us working on it full time, able to make decisions every day and keep pushing it along. In reality, though, those of us involved in the UI Team do our best to address issues and provide direction in our spare time, which is terribly hard to come by.

I don’t mean for that to sound like an excuse, it’s just the reality of it.

For what it’s worth, all of the goals that were posted there were all agreed upon by myself, Pablo, and at least one dev whom would be responsible (generally Julien). The intent had been for those goals to be ones that were realistic and ready to be completed, as they’d all been extensively discussed during several of the UI meetings.

But that’s really neither here or there, since we haven’t been able to effectively coordinate the time anyway.

Anyway, I don’t like to diminish the work of the guys who make ambitious plans even if they don’t always succeed. But as something for the community to expect to be completed… this gets a bit tricky to balance.

No worries; you’re not diminishing the plans :slight_smile: For my part, this has probably been the most difficult to balance. Balancing expectations when developing/planning in the open is challenging.

Its fine, just to note the first time I saw this site (http://wiki.blender.org/index.php/Dev:Doc/Projects/UI) was when a user here linked to it as an example of how “Blender developers are constantly failing to meet their own targets.” ~ words to that effect.

So it’s just an issue with planning and communicating, if a goal looks like its going to miss its target… remind devs and ask whats going on. Before each release - re-confirm planning … adjust it to be realistic. Its also fine not to have release targets if this ends up being the kind of work developers will do in spare time, just list tasks with some priorities, and cross them off as they’re finished.

So UI dev is in zombie mode or something.

Nope. There is much activity on, among others:

Absolute grid snapping - https://developer.blender.org/D910

Node automatic offset - https://developer.blender.org/D1373

Ctrl-click to open panels toggle - https://developer.blender.org/T45346 (this one needs a bit of user voice)

Better node drag&drop - https://developer.blender.org/D1393 (you can speak up here if you use nodes a lot)

I agree, UI is a divisive topic. Everyone has his/her own unique preferences. That’s why I’ve always been with the mindset that power should be given to the users by doing what needs to be done in the customization department, fix things that are broken. Then perhaps users can share or submit their default configuration, or “skins” as some people call them, and perhaps BF could just pick the most popular and sensible config and adopt it.

BF deciding on the default now is important for people who want to write manuals, books, make videos, etc. (You don’t teach Blender with a heavily customized version.)

I think the introduction of sticky keys slowed things down. It started to look like a massive change is required. In my case, I was only thinking about what can be done now based on the current customization capability like switching to LMB, remap the cursor key, etc., like I have done…

@blendDoodler related to your comments, https://developer.blender.org/T37518

@blendDoodler, I don’t see a connection between what you wrote and sticky keys? Could you describe a bit more what you mean?

You’re right, there’s no connection. It’s a side personal comment. http://blenderartists.org/forum/images/smilies/sago/smile.gif

What I’m wondering about is why are the devs still not making any decision to switch to LMB config default already based on what is already available? What are you guys afraid of?

I, for example, have long switched to LMB setup many months ago, although it’s only for the modeling and sculpting aspect. I model, I sculpt… I have zero problem with it. Zero, none, nada, 0… I can move an object, edge, vert freely in space using LMB drag (object and edit mode). My RMB is assigned to Dynamic spacebar menu (which is better than that silly cursor, which I have assigned to Ctrl/Alt/Shift LMB, see how I dislike that guy? http://blenderartists.org/forum/images/smilies/sago/smile.gif LMB double click is desselect, things like execute on release, etc.

/me feels like the only person on earth who’s happy with Blender’s UI.

I doubt it, Piotr. Thing is, I reckon that you & most others happy with the current UI will be fine with changes to the UI given time too.

Some people were apparently perfectly happy with the Blender UI pre-2.5. I doubt one could objectively state it was better than it is now. :wink: