Campbell Barton demonstrates an early version of tabs for the Blender user interface. Remember this is still very much work in progress! I like what I'm seeing so far: the pinning feature is clever and I'm impressed by how it handles add-ons. What do you think?
Campbell having code fun, Blender UI mafia reviews: tabs! #b3d pic.twitter.com/7B5MpTLBuT
— Ton Roosendaal (@tonroosendaal) December 6, 2013
We need to see the case where the set of tabs is long enough to require scrolling as opposed to the content needing scrolling. How will the two interact? At the moment the tabs don't scroll with the body. Will they be scrolled independently or will they latch on to the content when they need their own scroll?
Good point. In addition, it would be good if you could rearrange the tabs so they make most sense to your workflow.
.....and be able to create custom tabs. (and possibly nest into categories)
I would've expected they work like tabs in most IDEs: collapse the least used tabs into a menu.
I could imagine they shrink in size and shorten their names like in most Browsers:
For example It would minimize "Animation" to "Anim…"
Yep, this is a TODO for the current tabs patch (shouldn't be so hard to resolve).
...a way to close tabs would be useful too (with MMB click or whatever).
keep the good work!! :)
the Material Preview image could have a Pin icon too. ;)
Possibly start a new column of tabs once one is full? That way, we won't be losing half the labels off the tabs and it'll allow the user to open quite a few tabs without much space lost, seeing as the tabs are quite narrow.
I'm also a fan of being able to close tabs (without having to remove the plugin). For this to work well though, it'll have to be pretty clear to the user how they add them back again. This way, any tab, even non-add-ons, can be tucked away if someone doesn't use them, but can be easily brought up again when needed.
Possibly a right-click menu that lists the tabs with checkboxes. If that's also the only way to close a tab, users will already know how to bring back a tab they hid.
So happy to see this — Hate it when I acrtivate two or more addons and it feels like that I have to scroll down kilometres… :)
Same here. I hate that drawback of addons. If not for that, I would have a lot more activated.
I'm working on a script called "Quick-On," a quick-access addon menu to solve that very issue.
Short description: Quickly access a menu of your favorite addons that you want around handy but don't want activated on your UI by default. It's an addon to control other addons.
Toggle the menu with the key, by default, and a menu pops up. Think of it as an addon palette.
Its concept is proving a bit tricky to translate into a product, but I'm trying to find some more time in my hectic life to just sit down, expand upon its basic concept, and just finish doing it.
I do have ideas I want to go with it, such as enabling favorite groups of addons as menu presets, to save time flipping through and activating addons.
By the way, I don't care if anyone beats me to this idea--I just need it!
Very nice ! Good job.
Now this is exactly the kind of smart, simple, minor UI evolutions we need. It's a simple thing but it'll help!
Interesting, I like that but... we already have tabs in a different form, see the properties editor header, or any radio button for that matter - a tab is nothing more than a radio button after all. So why not keep the current visual design ?
Because for panels like these, it allows customization by the user/addon creators. Not to mention, many of the actions in the toolbars deal with a specific set of uses per view. For example, it would likely be difficult to find 6 different images for the tabs that all deal with mesh manipulation and are easily distinguishable. Text is simply so much more descriptive and definitive. This being said, I DO believe Blender should have a unified UI style, but if there is a way to keep it user-friendly AND also easily modifiable, it has not yet been found.
I didn't mean to suggest images in lieu of text, no, text is fine I agree. I was rather targeting the button design itself.
There's one crucial missing tab - CUSTOM! If it were possible to add (multiple?) Custom tabs, we could drag on to the tab only the panels we use most often, and not have to search through the entire tab tree. Possible?
Agree, custom tabs could be a nice addition to this already impressive tabs system.
It would be even better if the buttons within the tabs were customizable, in which case there wont be any need for tabs and remove some options like Grab, Scale, Rotate. It is generally encouraged to use the shortcuts, and I expect everybody to eventually use the common ones G,S,R, giving the experienced users the option to de-clutter the tool shelf.
Maybe there can be an additional tab within the user prefs called 'Tools' where all the tools are available and you can select the ones you want to display in the tool shelf.
I do think its quite important to seperate tools and its corresponding options, maybe via seperate dockers...?
Having the buttons within tabs customizable is tricky - Button layouts are defined in Python, so when a user updates Blender which buttons to use? - the ones they moved around or the new ones that come with Blender? (in fact we have this problem already with keymaps).
The toolbar is for tools already, where to show persistent tool options is something we should resolve - currently this is mostly handled by operator options, but there are some options like X-Mirror which have to be able to be accessed all the time.
Currently any panel can define their own tab, so with custom (user defined panels), we can have custom tabs. But this is another project.
This meand we could RMB over a panel's title to move that panel to another tab?
That could be useful...
No plans on this - not for built-in panels at least, parhaps for user defined panels it can work though.
Very useful feature and well implemented but instead of the 'pin' why
not use Shift/Ctrl for multiple selection? It seems a little bit more
natural that way.
In fact this was what I was going to do in the beginning, but after discussion with Johnathan Williamson - he wasn't much a fan of the idea, we agreed pinning is probably better.
The problem with Ctrl+Clicking on a tab is you instantly get all panels from both tabs (and likely have to scroll), if you only wanted one panel its not very convenient. Its also not so discoverable.
So I prefer pinning for now, note that we could have both, but at the moment I don't think its such a good idea.
I understand the issue was the length of the menu and I agree with the pinning but I first thought of some alternatives to adding the pin, this is why I considered the multiple selection option as in the no1 example.
I was also thinking about being able to rearrange the items in the menu as a plus, in case of a longer vertical menu, or just to make it a little more customizable, like in my second example.
The screen resolution was another aspect I considered, at 1080p scrolling is not a big issue, at least for me, so this is why pinning wasn't my first choice, nevertheless I think it is a very useful feature and the interface is more flexible this way.
Placing the pin on the right might be a little confusing as the handle that used to be there doesn't have a high contrast and it's not that visible. I think a better placement is like in the 3rd example as the handle is still in the same place. A similar example is the pin next to the rendering layer. Also by placing the pin in the same position the interface is kept quite consistent.
The reason I put the pin on the right is that it overlaps the text sometimes, if its on the left if can cover the text, whereas when the grab graphic overlaps the text you can still read it (This is weak I know - perhaps pin can be made less contrasting so yo can read when text overlaps too).
I think the pins simply need to be smaller, their position appears good to me.
pinning is better than ctrl+clicking but if the rationale for pinning is to allow the user to have access to panels that they want to keep visible, it could lead to situations where they have to do a lot of scrolling and clicking through tabs and pinning before they can get going... and if they need to alternate between two or more different pin setups, they're out of luck. and you could be out of luck between sessions in blender.
in the root of the thread, i propose an idea of a central repository of tabs (more like drop down menus) along the top and using the left tabs as customizable shelves... with users able to pull off items they don't want, add others and even make custom tabs.
Probably panels will be split up more then they are currently so pinning a single panel shouldn't cause too much scrolling. Further, I like to see how much users end up running into issues. Maybe potential problems end up not being an issue in practice.
Spot on. I was just reading the following article:
Postmortem of a Venture-backed Startup
Check the part under 'Listening to your users: False positives'
Which can be justification to just ignore everybody. Microsoft had probably adopted exactly this policy as they designed windows 8. Just put everything as false positive and you don't have to listen to anybody. Must be nice.
If you're taking it to an extreme then, yes. But I do encounter this in my work as a community manager quite often, too - there's always a vocal minority that's saying they 'absolutely need' a certain feature, and then once it's available, nobody uses it.
The proper way to approach this is to first gauge the real size of the audience that would require a certain function, and then weigh it against other features and their audience size/impact.
The issue at hand is the pin functionality. All I'm saying is that if people can find situations where pinning some panels in a certain way or order is desirable, it's not hard to imagine people wanting to RETURN to that setup later after doing some other stuff. does that really sound like something that would be very rare to you? I'm thinking about a way to do that. Without that, when they want to come back to that configuration after doing another task with different pins or when starting a new session, they have to go and pin everything back again manually.
Finally, when people seek to nullify the "vocal minority", understand that the term refers to anyone who speaks. There is no notion of a "vocal majority". It's the distinction between those who speak and those (in the majority) who remain silent. If those who speak are too be marginalized, may as well not have comments then.
What you suggest may be useful (some kind state storage for panel [pin+location]),
However a lot of whats being suggested (infact almost everything), including this - can be easily added later - as in, the current design supports it without big changes.
You have to appreciate too that you can loose time developing features which come more from speculation (guessing what may be useful), often once you have the feature running - you end up with a different set of issued then you might expect.
So even if I were to write in a bunch of extra functionality later, I wouldn't be doing anything different now. Also, theres a lot of value in sitting with a user and watching how they use Blender, at that point you can have much better insights and are in a better position to make decissions IMHO.
"What you suggest may be useful (some kind state storage for panel [pin+location]),"
or... a customizable shelf system where you drag in and out panels as you please - which actually is a somewhat different system with a somewhat different purpose (which is why i bring it up now)... but may be of more robust use in a production setting.
( - frankly, having a pin "save state" seems inelegant [going to end up with a tab full of pin states] vs. custom shelves/tabs and it feels like a "bolt-on"... which, again, is why i mention the customizable shelf paradigm now)
personally, i think it would be handy to have both - with pins being completely ephemeral and not "re-usable" across sessions.
"You have to appreciate too that we can loose time developing features which come more from speculation (guessing what may be useful)"
that's true. you can guess wrong. but essentially, ALL new features come from speculation. right? you didn't convene a gigantic forum of blender users to come up with tabs and pins in other words.... my appeal is to YOU AS A USER and YOU AS A DESIGNER... you as the person who "guessed" to come up with pins and tabs in the first place.
that being the case, it's unnecessary to deflect to principles like "agile development". with or without agile development, there has to be someone who's vetting initial ideas for implementation and guiding the development path to begin with.
as it pertains to this part of dev, that person is you. the burden on you as developer and designer is to GUESS RIGHT. and it is important not to defer every decision to committee and to have a solid, intentional design vision at the head.
and if you, the design head, feel this is not worth pursuing at this time, then that's all you have to say.
However the reason this may not end up being a problem is...
- If you want an animation tool, you click on the animation tab, users just have to accept this for the mostpart (if we end up adding this in), pinning is an exceptional case IMHO.
- If you really access a single tool all the time, then probably you use the key binding or bind a key (if it doesn't have a default binding).
- If you really have a special need where blenders toolbar fails to meet your needs, I rather make custom toolbars work well (which you can put into whatever category you like - including your own categories).
Maybe we're wrong - someone could explain a situation where all this falls in a heap - and in that case maybe we need to go back to the drawing board and figure out a totally different solution. However at the moment I didnt change my mind from my previous statement - existing system easily extends to be customized, so lets first see how basics work, then make more advanced where needed.
( Note, I wrote up the patch but didnt really do much design in this case - Check http://developer.blender.org/T37601 )
Also the kinds of points you are making would probably be better raised there, since this is where decisions are being made.
cool. point taken and rock on.
good job on the implementation.
i think blender sould get more icons less text in the toolbar and other areas.anyway blender rocks
Yes, small icons are more readable and save space better then vertical text.
I don't see icons being advantageous. Text simply spells it out and is much more intelligent to me. Open up any application you've never worked with, especially a sophisticated one like a 3D modeler, and tell me how intuitive the icons are.
For first beginner's steps text is better, over against for fast professional workflow are better icons with tooltips. User is beginner for short time.
Now this is a more sane UI idea. This actually makes sense very well does not radically change too much and seems to have plenty of room to grow.
This looks good.. Like a lot of people I love the layout of Blender, but small developments like this, allowing the layout to basically stay the same, but adding to its ease of use are great. Great work guys!
It'd be nice if you can optionally just bring a subpanel right into the 3D Viewport itself. Sometimes I don't need even most of the toolbar, and I just wish I could just pin certain subpanels into a fullscreen 3D Viewport, that go transparent and out of the way while you're not using them.
They'd be optionally collapsible in two ways: as a bar (as we normally see them collapsed) and further down as a "+" (just like how the toolbar and properties sidebars are done, except they'd be placed where you want them. It'd reduce the need of UI clutter and scrolling to just have the subpanels you need right there.
Just putting that idea out there. ;)
I absolutely love this idea, this stops me from having to scroll down all the time and it helps me find the tool I need much easier and quicker, sometimes al spend up to half an hour looking for that tool because it is hidden in amongst every other tool which makes it hard to see, something this addresses very well.
Oh and yeah another cool idea which is not even remotely linked in any way to this is that it would be cool to have the sculpt mode and the vertex paint mode combined so that you can either just sculpt the model, just paint the model or you could sculpt and paint the model at the same time.
My thinking is that this would enable different workflows that blender at the moment does not, and one of those work flows is in conjunction with another video that was posted on here of someone painting like you would in photoshop inside blender using the uv room, well I had the idea what if you could paint like that but in the 3d view port of which you currently can, but what if you could paint in the d view port similarily like a traditional artist would paint in the real world with thin and thick paint, the sculpting mode would give the impression that you are painting onto the paper or canvass with real paint and give the paint an impression of depth, like it has a volume, the paint mode would just allow you to paint on the paper or canvass at the same time.
But anyways back to reality, these tabs don't just have to used on the left menu, but they could be used in other areas as well to declutter things.
I dislike this refurbish. I'd rather prefer more shortcuts. Console should be for hardcore users. We aren't script kiddies, we are serious 3d artists, aren't we !?
maybe you are, but maybe many others would be and, not knowing misterious shortcuts, they prefer to start with visible menu and buttons.
However please take a look at Maya, C4D, 3DS... interface.
Not sure I see your point, just hide the toolbar if you don't need it.
But we are now at the point we basically run out of keys so not _every_ tool can have its own key shortcuts. Also - While addons can define key shortcuts - if every addon does this it will quickly conflict with other addons and become untenable.
Yep, I have a Logitech G-13 hooked up for complex custom combinations. Anyone else use Ctrl+Shift+Alt+C+Numpad 1 (Make a claw with your hand) over and over when modeling? Hehe.
I think being able to colorize the tabs would help to quickly differentiate between them. Or at least have different colors for the default and add-on tabs. The more customizations is possible, the better. Personally I never use things like the "transform" options (because I really can't see why you'd ever want to use eg. translate in the toolbar over just pressing "G", especially considering that your mouse position is crucial when it comes to using those tools, making the toolbar even less useful for such operations since your mouse is far to the left of the screen when you start using them). Generally a nice idea though, I love seeing good UI changes :)
i think andrew price is the man
Love it already. The pin was one of the things I kept thinking would help avoid a lot of unnecessary scrolling . And tabs are a great solution all around.
Being able to drag out floating tabs would make this perfect.
Guys, it looks cool! Better than before!
Great job Campbell! I can see this UI pattern spread to other areas of the interface as well.
I would like to see icons instead of text like in properties panel - it would be easier in navigation
How about one long scroll and instead of tabs there will be anchor points?
And when you scroll each anchor point(that can look like the tabs you have now) will highlight on the proper position of the scroll.
This way, you can scroll or click the anchors depending on what you like.
This could be made to work, but may be a bit hard to do well, some anchor points may be close and you would have to make sure the text is aligned well.
Also - there is the issue if some panel at the top draws more/less buttons (say you have a mesh selected, then change to an armature). --- it could mess up scrolling, (OK- this can be resolved with smart code too, just ensure with changed panels the UI wont jump).
Thinking about this further... With tabs at a glance you can tell that you have the "Animation" tab, and then you know what panels to check for (their order and where they are located).
Imagine if you didn't have tabs in a web browser but just one long page to scroll... (an extreme comparison and not quite the same)...
Sure its possible to divide up the pages into sections of the scrollbar, but at some point its nicer to select a context and know what to expect (IMHO).
The scrolling system works well when you need 2 or 3 tabs that are close to one another so you don't need to go all the way to the tabs panel and switch a tab every time you can just scroll a bit between them and use the anchores for long scrolls.
I don't code much and I never touched blender's code so I don't know if it's hard to do or not but I think of it as a chain of tabs grouped into a single panel.
I think it's something worth thinking of because it can be a modular system that can work on other parts of blender.
Lots of programs use tabbing the wrong way and it's always a pain to see good tools being hidden from the users.
I don't really agree that this is hiding, any more then having modifiers in their own section for the properties editor is hiding.
If you are a modeler and never animate - it should be nice that you can leave more relevent panels open and ignore the ones you don't use. Same goes for physics (which probably a lot of users wont use). Also, used to use a 3D app this split tools up this way and found it worked quite well.
Wow nifity! This has got me fired up! Just as Apple took the lead with the mouse, why can't Blender take the lead with haptic technology! Something to day dream about! :)!
This is cool stuff! I found interesting to be able to underline a leter of each tab to quick acces it with shortcut. Like edge tool or face tools in 3d view port. And pin/unpin with ctrl and the respective shortcut.
It will be a great time saver
Since these tabs run vertically, we can see that it's running out of room quickly. I think Maya shelf concepts are the best. You can keep the default tools tabs or create you own tab from any menu or shortcuts. They also run horizontally, which has more space, especially for wide screens.
at first I thought that they were going to change completely to that. it looks like they're well integrated with the design so great job! :)
Thumbs up for me. On a side note, being an avid dog lover gotta get a kick out of the calm and professional demeanor of our orator while his pooch is scratching at their collar in the background.
I am very happy to see that work happens at the UI. I am at the same time very unhappy what kind of work it is. Instead of fixing the mess in the toolbars you introduce tabs to clutter the mess even more. This is a typical open source workaround. Let`s create a mess. Then let`s categorize the mess to make it look less messy. Dunno if this is really the right direction. Wouldn`t it be better to start with cleaning up the mess first? Just by removing of all the empty and wasted space in the toolbars you could avoid lots of scrolling already.
If you want to show tools from a wide range of areas - even packing buttons only goes so far, and not far enough IMHO.
What could work, is extend modes - so have a rigid-body-mode, animation-mode (like we have for editing and paint already) - Then each mode can have its own toolbar and keymap.
... that said - even mesh editmode along is already scrolling quite a bit, and we are leaving out tools simply because there are already too many tools there.
We didn't agree on how to split up tools - The categories in the video were just an example, lets see how it progresses.
That`s of course true. But as told, you can reduce the need of scrolling quite a bit by removing empty space. That should be the starting point then to categorize things. You may categorize things a bit different then :)
Agree space should be used optimally, but this gets to become a matter of taste too (how much space around buttons is too much, how small a font is too small... etc).
The main reason to keep buttons one per row, is you cant assume word lengths are the same with translations enabled, 2.4x packed buttons assuming English words, but this failed baddly for translations.
What`s so bad about shortened or cut off terms? Do you really need to display the full name? Make the maximum button length fixed or give it at least a maximum allowed length. You can display the full toolname in the tooltip then in case it is really too long.
And does the full names really need to be so cryptic and long? Like the button called Duplicate Objects? Duplicate would be enough here. It is already in the Objects category. Or even Dupli. Or the RulerProtractor Button. Ruler would be enough.
Here we are back at cleaning up the area first. What you need is a word that tells you that a special tool is behind a specific button. A catchword that is easy to remember. Not the full description of the tool. You could also call it Steve or Henry. The description what the tool does can be found in the tooltip then.
Anyways. I have also seen text buttons toolbars that are very compact, and uses nevertheless the full tool name. 3D Coat is a good example here. Maybe worth a look to get some inspiration.
What`s problematic is of course that you mix the toolbuttons with their settings in one toolbar. And a slider needs a defined width to work. So you cannot really make the toolbar small. Which leads to lots of wasted space. That`s why i would love to see the settings separated from the tools. Like Photoshop, where you have your buttons to activate a tool at the left side, and the settings at the right side then.
I personally am a big fan of icons. They all have the same size. And are
much faster to recognize and to find than text buttons where you need to read the text first. You read two or three times over the tooltips, and then you should have memorized the icon. Or make it modal. Allow pro users to switch to icons only. This would free lots of space.
But i digress. I start to end in the whole UI discussion again, heh. My point was that it would be a better starting point to clean up the current toolbars first before categorizing them and putting them into tabs :)
@disqus_eFyKp8U9zc:disqus - This is a discussion I rather not enter, of course all options can be explored - we are well aware that buttons can be made more compact without the need to be inspired by other apps (this stuff is really obvious). - I've done tests on my own system but Im just a coder right? :), and let others do the design.
However you are conflating topics - improving efficiency of layout is good weather you have tabs or not. - I see it mostly as an unrelated topic. Mainly because its not as if the scrolling problem goes away just because we get a bit more efficient button packing.
You mention using the name "Dupli" -- this is confusing since blender has Dupli's already in the form of Dupli-Verts, Dupli-Frames, Dupli-Groups, etc.
Anyway - Lots of changes can for sure be made here, and I dont disagree with you entirely...
This is the problem we have with UI design, lots of good suggestions, but not many devs - we need to tackle them but can't do all at once.
Yeah, we users are vampires. We want everything. And we want it now :D
I am well aware of the lack of developers. I wish i could help. But i don`t know C nor Python :(
yes, this is true. I myself as a translator face this problem everyday, trying to find right/short words to fit in button's space, but that's not always possible.
also, on the same topic, images (icons) have the advantage of being independent of language for deciding the layout of a UI, so in certain places they could be a good solution (as the Preperences editor has shown).
Compared to the current setup, with collapsible panels, this mostly allows for quicker navigation I think?
To really enjoy the faster navigation, the tabs would have to stay in place and be easily findable. Thus, no scrolling. Darkening disabled tabs does not seems such a good idea in that regard, as the disabled tabs are more important than the active tab which needs no further clicking.
Keeping the tabs from scrolling out seems most important. This could work well if there are categories as opposed to a new tab for every plugin or function.
I am very happy to see this type of thing coming to Blender. I only use a fraction of the power available partly because of the interface.
I like the idea that someone suggested before about having the ability to choose between text buttons and icon buttons, that I think would satisfy both types of people, people who learn via reading and people who learn via visual basically people who like to read text and people who like to watch pictures.
How ever I am aware that to make this work, when you change to icon buttons, the whole ui would have to be changed to more efficiently make use of the more space provided since the buttons would be a lot smaller, so in effect you would have to have to different ui designs and that would take a lot of developer time, time which could be spent improving the tools or other areas.
unfortunately as some has already mentioned, we do want everything and expect it as quickly as possible.
but to be honest I am happy with blender at the moment and any improvement even if its not on my wish list is still welcome. if anything, this just proves that they are listening to us and trying o come up with ways to improve blender for everyone.
and as i end I have to agree that the 2.70 series is going to bring some very nice additions to blender, unfortunately I am slightly impatient lol and here I thought I was going to get it for Christmas, oh well I guess 2.69 will have to do for know.
Its nice,but general the toolbar tabs is on the top，not on the left ,sometimes on the left not convenient and differentiate . In cinema 4d,the toolbar tabs is on the top,and if you want select many tabs ,you can press the Shift key select others you want ,or press Shift key drag mouse on many tabs,I am think its very useful .
And I hope its can appeared in blender ,hope that ,good job!
its nice,but general the toolbar tabs is on the top，not on the left ,sometimes on the left not convenient and differentiate . In cinema 4d,the toolbar tabs is on the top,and if you want select many tabs ,you can press the Shift key select others you want ,or press Shift key drag mouse on many tabs,I am think its very useful .
And I hope its can appeared in blender ,hope that ,good job!
its useful,maybe need sometime to adapt，hope its could much more powerful and useful
It seems the source of the scrolling problem is that each column of panels is only allowed to be one panel wide. If all/any buttons panels could have the option to be made double-width, with two columns of controls side by side, any panel with lots of controls could use more screen width and so wouldn't need to be scrolled at all. Also, why can't the T and N panels be made available like any other buttons panels, so I can place them somewhere on my second monitor and have them permanently open?
Still should use icons instead of text on those new tabs...
While icons can certainly serve the benefit of speedy recognition, sometimes one common problem with icon is image ambiguity. And it's not always just an issue with new users--even long-time users can sometimes be intimidated to press an icon button without taking time to reassure they can press it.
Not always, but in many cases, you can only use so many icons before you start to need text labels. But there's no ambiguity with "knife tool." Not that text labels can't sometimes be ambiguous, though, icons should probably be used selectively rather than exclusively, because text do a better job with being self-explanatory.
It's just like the situation with hotkeys--there are only so much functionality we can map to hotkeys before we start needing the drop-down menu. What's perhaps the best option, though, is simply just to allow users to use icons or text labels for the tabs.
Though, it should be noted that these tabs are still a work-in-progress, so we may just see something along my latter suggestion, by the end.
I still believe in the quest for the symbol without ambiguity. But if that takes a life to find it, then an improvement suggestion to the mentioned proposal and experimentation would be the use of titles abreviations and maybe horizontal text side tabs: Maybe it would save space for more tabs...
It looks a great workflow enhancement.
Will we be made aware of which test build it will become available?
Look cool ! i think it better than before that parts of toolbar is organized in tabs than separate line ^^ can't wait next official release!!
The name of the duplicate operator (Duplicate Objects) seems to be inconsistent.
Maybe its name should be just "Duplicate" or we should add the "objects" word to all other operators (Delete Objects, Join Objects, Translate Objects, Rotate Objects,...).
awsomeness, another way of doing this is by mimicin the tabs on the right side, my idea is to have icons instead of writing, this will mean that every addon in the world of blender would have an associated icon,
when the end user hovers over the icon a tool tip shows up displaying the title of the tab.
happy blending blender heads
This is great, I am looking forward to playing with your tab.
I wish show you a short video of a very old Atari CAD software: Dynacadd.
This program has a special feature: dynamic multi-level menu.
It was simple and effective.
This is nice feature. You should consider moving pin to the end of the vertical tab bar rather than having it in the open panel. This will reduce number of clicks needed for pinning closed tab from 2 to 1.
i would suggest that the tabs run HORIZONTALLY across the top (or even bottom) of the screen. it only makes sense - our screen are much longer horizontally than they are vertically so why try to use the smaller realestate when you just KNOW the amount of tabs can and will be huge.
and you can color code the tab and have a line or bit of frame with that color run out to the left panel which can also adopt the color of the tab if people really need to see that the tab above controls what is seen in the panel.
actually some ideas from maya and zbrush may also be helpful:
in zbrush, every single possible function of the software is available through a top menu and clicking on them unrolls everything (basically a pull down menu). the concept is that the top bar is the CENTRAL LIBRARY OF ALL THINGS. you can't delete anything from there, you can't rearrange... this is just the central database of everything the software can do. and no matter what you do or experiment with in interface re-arranging, central repository is sacrosanct... in other words, YOU CAN'T MESS IT UP. no matter how bad you screw things up, you can STILL access all functions from the top menu.
then, on the left (or right or bottom or below the tippy top), you have SHELVES, where you can pull items from the CENTRAL LIBRARY OF ALL THINGS and drop in your favorite and most used functions.
this way, you can kind of get the best of all worlds - EVERYTHING available along the top menus (and because there will be LOTS of entries, it requires and uses the horizontal realestate) and then reserving the smaller vertical space for shelves that don't encompass everything the software can do but addresses the most important and most used tools.
not only will users be able to customize the shelves on the left (and blender could and SHOULD come with certain shelves labeled and already populated as a default, basic workflow setup) and drag in or drag out stuff that they want that's specific to what they do, they can also CREATE THEIR OWN TABS (like custom shelves in maya) which they can name (maybe nothing more fancy than "CUSTOM1", "CUSTOM2" etc.) and populate as they please.
Great !!! Blender is already powerfull, but now with this kind of add-ons, it'll become more and more flexible than before !!! Good Job guys !!!
I think this is a great idea. I find it annoying to have to continually scroll through so many panels looking for the option I need.
Besides, there are features I don't use that often and it would be nice to be able to have them tucked away in a tab but I could easily bring it up when I need it.
I also vote for a future capability for customized tabs. I'd like to arrange the tools to match my particular workflow.
Thanks to all the Blender programmers for your great work. You have many thousands of people who appreciate Blender as an open source application.