Of course UI and UX are working together!
UX is the whole thing. User EXPERIENCE. The UI is a part of this. You interface with a program via the interface (hence the name). How easy, intuitive and pleasant this is is called experience.
So a UX designer decides how the user interface works while the UI designer decides how the user interface looks. This is a very collaborative process, and the two design teams tend to work closely together. As the UX team is working out the flow of the app, how all of the buttons navigate you through your tasks, and how the interface efficiently serves up the information user’s need, the UI team is working on how all of these interface elements will appear on screen.
This question is a little bit more complex than what you may think.
This is a convention that seems obvious : to warn user when there is a potential lost of user’s work.
Everybody discovering the tool will expect that.
But when you are frequently using a feature ; and each time, a prompt asks a confirmation that you know and accept consequences about a potential data lost : the warning becomes an annoyance, slowing down a lot, progression of the work.
That is a proof that UX is dynamic and may evolve through time involved in practice of software.
Annoying confirmation prompts that are only needed once to inform user and then, are becoming useless is a different topic than Fake User concept.
User can make a conscious choice to create a Fake User when he wants data to be preserved.
But that does not mean that people defending that principle are defensive. They may sincerely think that is providing a better UX, relatively to what their use of software is, after having been warned.
That is seen as an automatic purging of unnecessary, invalidated attempts, tests.
In practice, Blender just saves what was validated by user.
The opposite way would be to force user to consciously delete everything, he does not validate.
So, that is a really subjective topic.
If you create a library of data that may not be attached to a user (exposed datablock), you will prefer to have preservation by default and a conscious deletion.
But if you are just creating lots of variations and finally only want to preserve satisfying iteration that will be sticked to a user (exposed datablock), you will prefer to have deletion of most iterations by default and only a conscious preservation of your best work.
It is hard to make beginners understanding that non-discoverable, uncommon behaviors of Blender can be seen as a more effective preference, by people with a little bit more of practice, for valuable reasons.
Because this undeniable that has a bad impact on beginner experience.
And that looks very weird to say that can have a good impact for other advanced people.
But this is true.
Blender currently simply has no security setup to even let users know that it is going to delete some potentially vital datablocks. It just does it. And except for a quick flash in the statusbar it doesn’t even tell new users it is going to do so.
I’ve been told that this has been the case for techical reasons in earlier versions because Blender didn’t have any way to eliminate datablocks in a running session. So it had to delete unused datablocks on exit. That is not the case any more. It’s a legacy behaviour.
“But it slows advanced users down to always have a popup”.
So give popups a checkmark ‘don’t display this message in the future - I know what I am doing’. If needed these ‘skipable popups’ could be reset in the options. Doesn’t slow you down any more.
But instead of trying to find a way that caters to advanced users (what qualifies an advanced user anyways?) and every other user alike we have discussions telling people that it’s a good thing.
Again: Blender eliminates datablocks that may have taken a very long time to create without warning. They are gone. And it’s easy to lose them. It’s not even like the user has to consciously mark a block to be deletable on exit. They are gone by default.
Sure - there may be blocks where it’s a good thing. At least for Materials and Actions it is not. For these things there is a purge button. You want to create variations for testing - make it a conscience decission for the user to create a variation that will be deleted on exit so that they know this is going to happen.
There are many ways this could be made in a way that it’s still as fast for those who want to work that way and everybody else.
(edit) Sorry, I am going to rant on on this topic a little more, now:
This is pretty much the core of what I meant by “being told it’s better that way”, BTW.
Someone pointing out a UX flaw will rarely start a discussion about how it might be improved so that the fast core functionality remains the same for all the users who want it that way but improve it for the vast majority of people who may need a different approach. Instead of acknowledging that it might actually be bad for most users the backlash will always be a long wided lecture on why it is good to have it this way. others will chime in. Thread goes up in flames. Case closed.
I’ve been using Blender through all UI iterations now. The first book I read about Blender still had instructions on how to buy the program. And I remember these discussions have been the case in the Blender community forever. And it’s seriously the single. most. disappointing. thing.
Especially since after every change of a terrible UX Blender grew in its user base. It actually in most cases made Blender better! Yet along the way users were (and still are) rarely met with the same open mindset like if they would ask for the solution of a problem. Usually they are, in fact, told something along the lines like being not advanced enough. It’s often not even cosidered that there could be a better solution while having the functionality remain intact.
UX means: Keep the functionality but make it actually pleasant and easy for everybody.
And I think what you fail to see in this is that from a well designed UX not only the majority of users would benefit but also advaced users alike. Because a good UX is designed in a way that scales along the complete spectrum of users. No matter where they are.
I believe what @chippwalters is saying in his proposal can be summarized on: the BF should have UX guidelines and a dedicated UX designer and/or UX training for the current devs so that they don’t just add features and place them wherever they can, but place them with proper thought and planning considering the possible use cases. And to know those use cases there’s a need for surveys in order to have tangible data and not just guesses. We need to know what kind of users we have, and how many of those kind of users, and what role those different kinds of users play into the final design of Blender.
Unused “data objects” are deleted in any software when you close your file, or never is saved, or directly you don’t have access to that info. I don’t know if actually still happening, but in 3dsmax happened the same, to delete a unused material you need to save and open the file again. The material was not saved in the file if it is not assigned in somepart or in the material editor grid.
And yes, it can be improved, but many things like this are expected to be looked at when the asset manager is implemented. Just like sculpting brush management is pure crap from a user’s perspective.
But honestly the problem of databocks complaints I read to advanced users, who already know how the program works, not new users. New users don’t care if unused material disappears. They care if they click and the program responds as they expect. I have taught blender to a hundred people, I have never encountered that question, nor problems with datablocks. But I have encountered hundreds of weird faces when people have to deal with something like this
This shit have been the 40% of questions that I received when I teach blender.
I think it’s a real UX problem what the newbies are always asking me. Not something that I’ve rarely been asked and when they ask, I answer with a simple “yes, it’s deleted when you close it, like in max, it’s not saved by default, you have to hit it here” That doesn’t seem to me to be a serious problem, it seems to me to be something that can be improved but that is also as simple as adapting to it takes half a minute.
Every program has its peculiarities, you think it’s easy to face the pie menus? almost all the users I know who have ended up in Maya hate it. Same to Modo, Zbrush,… But by far the easiest program to learn nowadays is blender because its interface is very well designed in general.
Something like this has been in the works for months. The problem is the same as always, when one talks about wanting some “guidelines” at the end what he wants is “his guidelines” which usually means ignoring the blender design and putting the one he thinks is good… and what is good? yes!! you are correct! the last program that somebody used. And some people ask for maya, other for C4D, SketchUp and of course, iPad, iOS,…
Can anyone tell me in blender today what inconsistencies there are between parts of the program in UX? Except Houdini blender has a very consistent user experience.
I don’t deny that. I am not saying that the status quo is a satisfying choice.
I am denying the fact that reaching a good UX that scales along complete spectrum of users is something that can be build in one year.
That is globally you and chip who are saying that you just have to ask to one expert and problems will be magically solved.
No, there were years of discussion before each iteration of interface to improve blender.
And if after each one, there are still things to improve, it is because manpower and time dedicated to those tasks was not sufficient.
What bothers me is that people are constantly pointing same issues like they were unknown issues.
I am posting in this thread, links that contains ,each, a dozens of design tasks pages to prove that those issues have already been addressed.
If you are thinking, problem is collecting feedback : you are wrong.
If you think that no guideline was given : you are wrong.
If you are thinking that a UX designer could help the team to have a better order of priorities, I agree with you.
But because of the amount of work that represents, we are still at years of a good UX in the whole spectrum of the software.
Some parts are currently improved.
Some can’t be improved because of lack of people to work on it.
And during that period, that may be a minus damage to preserve status quo for those parts, instead of starting something, keeping it half-finished to work on other stuff and then, finishing the process years later.
Yes, now, we have a button to purge orphan data and it is time to reevaluate Fake User concept.
Yes, some prompts are optional but not all, yet.
If currently, Blender is at an in-between state. It is not because people are not agreeing on what to do.
It is because passing from one state to another corresponds to work time.
I’m remembering when blender was going to have two interfaces, one for newbies and one for professional users. And the reality gave a very strong punch, and after more than a year it stayed in a toolshelf, context menu and change the theme to dark.
I believe you’ve made your points well known and have successfully played the contrarian, posting in 7 of the last 11 posts on my thread.
Perhaps you would be kind enough to let others have an opinion without having to argue with them? If not, perhaps you would have the grace to start you own thread on this or any other subject where you can continue your comments?
I am interested in hearing more about what others think. Thanks for your contributions.
It is generally a bad idea to base design decisions on feelings alone, because feelings can easily be manipulated and they can change on a monthly, on a weekly, or even on a daily basis.
Instead of figuring out if a tool or feature feels bad to use, look at the tool or feature from an objective standpoint (ie. the data that shows actual sore points such as stability issues, performance issues, excessive clicks when used right, general bugginess, ect…). Is the performance not optimal? Can the same task be done with half the clicks? Are certain features overly obscure and/or relatively useless for production-quality results? Let the questions be driven by data rather than being dependent on whether the user is just having a bad day in general.
I think you might need to take a step back. The tone of your replies is defeating whatever cause you are fighting for, and undermining your message. It’s very difficult to win people over by either directly calling them ignorant or indirectly suggesting as much. Even like-minded people will be pushed away from your cause.
As such, I would suggest closing this line of discussion and moving to a more structured proposal. A concise introduction, some presentation of evidence (no handwaving, no anecdotes on a small sample size) and then draw a couple of conclusions that could form the basis of some initial work (don’t fix the entire world, start with painting a wall in a house).
Additionally, rather than asking someone else to own the solution to the problem you perceive, take ownership of it and lead. Essentially, use the marketplace of ideas to persuade and win, with the application of some elbow grease.
I’m also fairly new to Blender (2.79), and I do think I understand what Chipp talks about. I also see discussions in new features that are very developer heavy, and I’m a developer myself so I get that, but there should be someone more heavily on the side of the user weighing in when a feature gets added in my opinion.
I am all for things being mathematically correct, I’m all for features being effective and low on clutter, but there are so many things in Blender that are great as they are feature wise, but could just have a “Usable default” as a UI default. The simplest example of that is render settings. Having for instance presets.
Also there could be base setups of many things, it would cost developers very little, but it won’t be user friendly unless someone speaks for the user or tests on the users. And there’s many types of us now for sure. There are so many types of people being interested in Blender now and starting to use it, it would be a great time to really step it up.
I don’t think the UX is bad in general, a bit small by default, to this day I still spend too much time looking at an icon before knowing it’s the tab I want in the right side panel.
I think you could even simplify many parts of it to have a few settings + advanced. And presets, because some settings probably are set exactly the same by someone hundreds of times.
I am happy to help lead if there is a desire to accomplish something. That is the purpose of asking the question, Should Blender establish User Experience as a core value?
I’m not at all interested in moving forward if there is no clear decision by Blender and it’s management to do so. For me to proceed, this commitment should be made first, and that is the reason for asking the question.
The word “ignorant” is mentioned only two times in this thread-- by you and Alberto-- and both times you incorrectly attribute it to me.
All I ask is people be given a chance to discuss without the continual threat of rebuke. Certainly, robust and critical thought and conversation is important-- but when it becomes a constant challenge to every point of view, people will (and have) stopped contributing.
What you may see as arrogant, are my efforts to managemy thread on the subject. I do not think it impolite to ask those who have spoken, given their view, and continually reinforce it over and over and over-- to please let others speak. And if they wish to start their own thread on the subject, I will happily do the same.
In fact, if you have strong feelings on the matter, you should also create your own thread. I do not wish this thread to become some sort of flame war about whose side you are on.
That is certainly an approach, but not the one I am interested in. My experience tells me my time is worth more than trying to convince folks who are not committed into seeing the light of “design thinking.”
Never, in my 40+ years of consulting have I seen a “bottom up” approach actually work with regard to design. Success has always been the opposite. People seek out design counsel when they know they need help. If no help is required, then there is not a need to proceed.
I agree wholeheartedly with what you say. I would also mention that UX is not always part of the development cycle. It is most effective when placed before development-- using professionals that are as smart and schooled in their craft as developers are in theirs.
All this talk about UX is getting me very excited for the future of Blender. If the Blender Foundation focuses hardcore on UX improvements, stuff like making it easier for new users to learn & reducing the amount of clicks expert users need per action, then Blender’s user base will probably increase significantly and will enjoy using the program even more than they already do.
