GUI paper example

[V1ntage]

I wanted to share the following excellent paper with you. It was made by William Reynish and is used as the backbone for the new Blender 2.5 GUI.

http://download.blender.org/documentati ... ers_ui.pdf

I'm currently working on my portfolio and the website analysis. When the website blueprints are done I will be creating a similar analysis for the CA GUI. I'm absolutely convinced it's in our best interest to analyse the current GUI in every possible way to enhance the user's experience.

[Neddie]

Finally, an exemplar. Here I was, reinventing the wheel.

[Regret]

The only thing useful from that pdf is the structure of the document, and even that is arguable.

[Gota]

I thought there were several interesting points in the document.
A lot of it talks about some of the mistakes done in the old blender UI.
In spring the UI is so basic and lacking it hasn't reached the part where it can be polished yet since it lacks basic commands and features.

A UI that includes all of the available commands and everything would be nice as a start.

[V1ntage]

I highly doubt the approach for GUI analysis in this document is not usable We better create something similar

[Regret]

I highly doubt the approach for GUI analysis in this document is not usable We better create something similar

RTS vs modeling program. Yep I can see the applications.

[Licho]

I think you suffer from overengineering vintage :) By the time you finish analyzing stuff for website or GUI, we can have functional website and GUI in line with what your documents eventually describe..

[V1ntage]

Regret, there's no need to use irony. I understand if you don't see the relevance in looking for reference material, but I do.

Licho, it may be you will have finished most development. And yes there's a lot of overlap. It's also the reason why I asked you if it's possible to branch out. I happen to have a vision. If you like to use the results of my work, be my guest. If not, there's no loss.

[Neddie]

If it was going to be done under the current model, it would already have been completed under the current model.

[V1ntage]

I don't understand what you mean neddie. Can you please explain?

[Neddie]

That was in response to Licho's point, where by the time you would be done planning it would be complete.

[V1ntage]

That could be so, but I doubt the results will be the same. My planning also involves altering game design on a fundamental level.

Iterating very fast by concentrating on results has its benefits, but so has planning and analysis. I happen not to believe in quality results for large projects without planning, while you seem to think the opposite. It's a different approach and mindset.

Either way, nothing holds you back to go ahead and move on. I'm just working on what I believe is right in my personal vision. Use it when you see anything you like. If you don't like it, then why bother?

[Gota]

Vintage you can bet ill use it in SA....
I afking crave for a good newbie friendly UI and will help you in any way possible.
I want to use RedUI but regret is completely impossible to cooperate with to the extant he wont add stuff unless he uses them himself...

I'll test and express my opinion :) I have had some ideas at how the UI can drastically change to become more intuitive and faster to use and maybe a small % of it can help you.

[MidKnight]

erm, he's working on a website.

[jK]

erm using Blender as a good GUI example is just blasphemy ...

[Argh]

erm using Blender as a good GUI example is just blasphemy ...

+1000. Worst. UI. Ever.
OK, so maybe Kai's Power Tools was even worse. Maybe.


Anyhow, my only serious thoughts on this topic are that:

1. Yes, it is very important to plan out a UI's symbolic and logical behavior ahead of time, but you have to keep in mind, this is an art, not a science. While we know a lot about UI theory these days, in the end, it's down to presentation and plain ol' visual polish.

2. It is more helpful to do that discussion with pictures before code gets committed.

3. It's easiest if you plan the entire UI out as a single module, whether it will be flexible in its configuration, with moveable / dockable elements or not. A UI is a whole, not a collection of parts.

One of the biggest problems with CA's UI is that it doesn't have that complete focus on a final outcome, imo. I feel that it's too ad-hoc, and isn't focused on the gamer- it feels like a collection of coder experiments, not a professionally designed UI.

4. You don't need a giant design document to have this discussion, in a collectively-planned and collectively-executed game design like CA. That just encourages lengthy-but-futile talk-talk from the talkers, not a lot of action from the doers.

What you need is to have simple design proposals, done by visual people, and then discuss them until the layouts are approved, and then execute the code and art assets. Then the artists know exactly what specifications they're expected to reach, and the coders know exactly what the business case is.

There really ain't a lot of room for talk-talk, when developing these things. This isn't a philosophy discussion, it's a fairly serious engineering / art goal. If you can't do either, you probably don't have anything useful to say, frankly, other than whether you like / dislike a given proposal, which everybody is entitled to.

5. You will never build a UI that pleases everybody. Keep the focus on potential users of your game. Keep it simple. Be conservative, unless there is a compelling game-design reason to try something really radical.

Never forget- there are literally thousands of videogames out each year, built by other people who are also tempted to do something whacky. The vast majority of the time, it just makes learning the UI harder, or totally confuses users, i.e.:

"Let's see... uh, do I stroke that leaf to brighten the screen gamma, or was it a double-click on that small robotic dog?"

If that kind of thing ever happens, your UI probably has a problem, which is why the giant corporations that build most video games have such stringent QA for these things.

[V1ntage]

MidKnight: I'm currently working on an analysis of the website, but my planning reaches a lot further than webdesign alone.

jK: If you find Blender's GUI blasphemy, then that is just your opinion. I find that mindlessly stating an opinion without explaining the reasoning behind it not constructive at all. At least, you're not helping me to understand your vision without expressing the idea that lies behind it.

jK and Argh: It gets even more interesting when it becomes clear that this thread is not about Blender at all, nor about the paper itself, but rather the constructive approach used in this paper to enhance an already existing GUI.



I'll repeat myself once more: If you like it, use it. If you don't like it, why bother?

[jK]

jK: If you find Blender's GUI blasphemy, then that is just your opinion. I find that mindlessly stating an opinion without explaining the reasoning behind it not constructive at all. At least, you're not helping me to understand your vision without expressing the idea that lies behind it.

trust me 90% of all ppl who tried Blender, think the same.
It's always the same, GIMP tries to defend their GUI the same way AND WE KNOW IT SUCKS.

PS: Blender doesn't need a document how small stuff needs to be changed (-> how free space can be optimized LOL), it needs a totally new GUI design ...

[Neddie]

Well, the paper isn't about praising the GUI, it is about outlining what seems right and what seems wrong, then explaining how to fix the problems.

Everybody knows that OS, Indie and free projects have TERRIBLE interfaces. Some developers think their failure to provide a useful interface is somehow a way to weed out incapable users. Others think that their interface is excellent, having absolutely no grasp of anything outside their chosen realm - usually programming. Still more go for form over function with triangular and circular control panels, all sorts of mad wankery.

I would like to note for the record that V1ntage's work is more of an inspiration for my tired organizing, researching and advertising mind than the sum of many mod or game development teams. I'm posting in his threads mostly because they interest me, just like how I try to pin down TheSleepless and see where he is going with his work.

[Pxtl]

I think Blender gets a bad rap for the GUI - I mean, it's counter-intuitive for no good reason, but once you get used to it then it doesn't seem much worse than 3DSMax. All of the big general-purpose 3D modelling/animation apps have nightmarishly complex GUIs. Blender's unconventional approach only makes it marginally worse.