UI Improvement Ideas for LL! =)

[Greene Paine]

What is wrong with the v2.0 - 3.0 UI?

Not really all that much as it turns out. v2 was a huge improvement in functionality but just looked really bad. It was kind of like having a new baby brother that you can't wait to tickle and hold, that is until you see how ugly he is and suddenly find yourself looking for hand sanitizer instead.

That and it sort of castrated your biggest supporters, but we'll get to that. First off is making v3 pretty again!

What colors do highly successful UIs use?

1) Skype: Smooth gradients of whites and blues

2) Apple: Smooth gradients of whites

3) Google+: Solid bright colors

3) Google Chrome: Solid bright colors

Can you think of any dark depressing UI's out there that have had nearly the market success that these products have had? Even your website makes the user feel like their peering into a dark closet at the funny people inside. (Doesn't your closet have funny people inside?)

What else do these four examples have that makes them successful?

1) Highly uncluttered

2) UI reveals complexity where utterly needed (think of how Google hides all links until you move the mouse)

3) They get the hell out of your way (the best UI is the UI no one notices)

4) They are all highly polished (this means things fail gracefully, bugs are squished and not just bandaged, and all workflows are enjoyable)

What can LL learn from Microsoft?

1) When launching the .net platform, why did Microsoft not just ditch Visual Basic when C# had all the exact same functionality? A: Because they recognized the importance of the VB culture and not ditching that following. Satisfy the v1.0 fans with a few check boxes in the options to give them what they want!

What can LL learn from Blizzard?

1) Blizzard allows UI mods so that users can add in the features they want. !!!BUT!!! They have a policy that mods should not be -needed- for the enjoyment of their games. Thus when a particular mod becomes so useful it really should be part of core functionality .... they IMPLIMENT IT in the official UI! (It's about time for a UI animation override)

Third Party Viewers should NOT be -needed- to enjoy SL either as a newbie or for more advanced workflows. They should be regulated to the role of idea generators and solutions for extreme use cases. If your contracts are stopping you from integrating the innovations found in TPVs, then GET NEW CONTRACTS. Each morning at LL should start with the UI team using TPVs and reading up on new ideas while they drink their morning coffee. Blizzard does.

What can LL learn from Google Chrome?

1) Get out of the users way! There is a cool feature in Firestorm where the upper menu bar is transparent, thus increasing the view port. Things like these add more than you know to the enjoyment of the UI.

Other ideas may include: Auto hiding menus or UI components until the user moves the mouse near them (think start bar in Windows), marking all components resizable (think Skype sidebar), and avoid losing tons of space to menu bars (think of Google Chrome folding everything into ONE bar at the top)

Pop quiz: How many sides of the screen does the SL UI currently muddle up? A: Three!

2) Tabs are awesome! This is a more long range thing, but if you really want to push the 'SL as a browser' mentality, then why not allow a user to be in multiple places at the same time? Freeze rendering on tabs not in use to reduce computer lag of course.

But this will increase server load right? .... Then buy more servers. Look at Google Instant. They increased their server load HUGELY for something they felt they needed to have the ultimate search experience. Time for hard choices.

What can LL learn from Blue Mars -FAILING-?

1) That collaborative creation is half the magic! Take Charlar Linden, give him a team of people, and lock him inside for a few months until you can create mesh IN WORLD. We don't need a new blender here. We need something like an in world Google Sketchup, perhaps even simpler.

Requirements of in world mesh creation: Ability to convert prims to mesh, merging of prims, editing of vertexes, basic editing functions like extrude and perhaps a few binary modifiers like smooth. THATS IT!

I can't stress this enough. Imagine you're in SL building a space station with your buddies. You've learned how to work with prims already, but now you've pressed that little 'convert to mesh' button and now you're moving vertexes all over the place! Your friends are laughing at you and you're all having an amazing time.

Let Blender be used for polished professional content, let in world mesh editing be used for fast and sloppy creation that with some work can actually look better than pure prims alone.

What can LL learn from Photoshop?

1) Photoshop and many other programs have a method of selecting 'work flow presets' that allows people to get up and running with the software without making a million changes. Basic mode sort of touches on this, but I feel that making the interface highly customizable and then creating a few presets would have a much better effect.

Some preset ideas: Builder - increased floating point precision on build dialogs, more alignment tools, 'convert to mesh' button, more feedback text for things like Land Impact and ARC etc., Simple - Super simple UI layout, very few options, keep the eye focused on chatting and exploring, Roleplayer, Gamer, Texturer, Scripter, etc...

 

I have many more ideas, and there are of course other things that need to be done (such as ending the 'what do I look like to you' discussion), but I believe that these changes together with LL's awesome new focus on Avatar profiles might just get this place back into the exponential curve.

P.S. When teleports fail, I should not fall though the damn floor!

P.S.S. Teleports should not fail.

P.S.S.S. Adult content is IMPORTANT!!! As adults we express ourselfs in adult ways. I think this is the 'other half of the magic' that makes SL so enjoyable. And yes this can very well include sex and sexual expression in things ranging from art to roleplay to inworld games.

[Einsman Schlegel]

I wondered the past few months..

Is the servers based on JavaServer Faces TM - or based on Oracle?

[leliel Mirihi]

---

Greene Paine wrote:

Can you think of any dark depressing UI's out there that have had nearly the market success that these products have had? Even your website makes the user feel like their peering into a dark closet at the funny people inside. (Doesn't your closet have funny people inside?)

---

Not that I'm disagreeing with you, but yes actually. Winamp, Lightroom, iPhoto, Logic, Pro Tools and Reason to name a few. Three of those are made by Apple by the way.

[Randall Ahren]

You have some excellent ideas. I think the length of your OP may have exceeded the input buffer of most readers. What is morale like at LL, does anybody know? Are most employees just hanging on hoping they can survive through the great depression or do they view themselves as working for a leading edge company producing a transformative product?

I don't see anyone that actually cares about their product and customers could output such consistently poor quality. Who made the display names decision, can't they roll that back? What about the basic/advanced mode feature, can't they fix that? Why are these forums so user unfriendly? Can't they use the forum to facilitate the inworld experience? 

The poor quality and decision making indicates to me a product being produced by people who don't use it.

 

[Tristizia Demonista]

Your color suggestions are certainly interesting, but, as almost everything in the world, highly dependent on personal taste.

Not everyone likes the Appledesign, you know (Or Chrome for that matter)

Also i wonder, if these would make a UI successful, why is Phoenix the most used Viewer? (Their default color scheme is everything BUT nice to look at.) 

I prefer to be able to customize the colorschemes for the software I use! 

 

Otherwise, i agree, that some streamlining is of great importance. I just don't think that LL is even able to pay the same amount of money that Apple pays it's Designers, or Google. Those are slightly bigger companies.  

[Greene Paine]

"Your color suggestions are certainly interesting, but, as almost everything in the world, highly dependent on personal taste."

I could not agree more. The idea of having 'skins' is that there should be multiple choices. I am merely suggesting that the default skin for new users should have a light, happy, 'simple' vibe going on. In reality there should be at least 2 different skins to select from with an import system for adding more.

To the person who pointed out the dark UI's that are popular ... I agree there are some popular dark UIs, but I pointed out the ones that I did because they stole the thunder from other much more popular services with their simplistic designs.

Google Chrome for example entered into a highly crowded and competitive browser market and somehow managed to pull a huge following, AND to slowly GROW its user base. Some of this has been attributed to its 'simplistic' UI.

Google homepage of course beat out sites like AOL and Yahoo by having not only a much better product, but by having a very simple 'clean' homepage. Nothing says clean like a white background.

Skype has beaten out many other VOIP services and gained huge popularity.

Apple came back from the verge of failure with its bright colorful Macs and then later with its sleek iPods and iPhones.

I suppose a dark color scheme is not inherently bad, but I think that white does a better job of conveying a 'clean and simple' vibe, where as black conveys an 'exclusive and expensive' vibe, which SL is neither.

[Erwin Solo]

Not one word about the Pie?  I thought the reason people didn't like V2/3 was because the scalable right-click drop-down menu didn't allow them to click their mouse as many times as if they had a Pie. 

Also, you are suggesting that V2/3 be more like other successful user interfaces.  But, the primary complaint I hear about V2/3 is that it did choose to be more like other successful interfaces--instead of being more like the V1 Pie.

[Erwin Solo]

---

Tristizia Demonista wrote:... Also i wonder, if these would make a UI successful, why is Phoenix the most used Viewer? (Their default color scheme is everything BUT nice to look at.) ... 

---

Indeed, that is the heart of the issue.  Phoenix succeeded because it was more like V1.  V2/3 got a lot of criticism because it was more like other successful user interfaces, but less like V1.

It would seem that making V2/3 more like other popular software, and therefore less like V1, would decrease V2/3 acceptance rather than increase it.

My understanding of the situation is that to quiet and calm those who complain about V2/3, LL must revert to V1 and never change anything at all ever again.  The alternative is to move forward and let the complainers complain.  There is no middle ground.

[Greene Paine]

Actually from talking to the v1 crowd, they arn't quite as unreasonable as one may think. I believe the primary reason v2 got such a bad wrap is because it muddled up workflows, and sort of subtly told all exsisting customers that they wern't important. That the 'mainstream' market was many times bigger than they would ever be. While this is likely true from a mathmathic standpoint, the exsisting customers are what's keeping this boat floating. =)

[Hitomi Tiponi]

An excellent post Greene.  As you so clearly point out the problems with Viewer 2/3 UI come down to:

• poor design
• poor understanding of workflows

There was also a pretence that it was a good UI right from the start back in April 2010 - so treating your users like idiots is not a good move either.  Now we have a UI that is at least tolersable, but sadly many of the great features in the Viewer e.g. outfits are not presented very well.

Viewer 3 will never be a great UI - but hopefully work by LL will help to improve it substantially as long as they learn from their mistakes and address the root problems rather than just tampering around the edges.