Maybe the percentage is a bit higher (or worse, even lower), but it’s totally pathetic. If this number were even slightly increased, we’d see tens of thousands of new residents streaming into Second Life every week, many of whom would stay and unleash their new creativity into the world. It would be a more vibrant place, with more creations and for business owners, more potential customers. It would be a Very Good Thing.
What’s happening to fix this problem? We’ve all seen the Lab’s experiments such as alternative approaches to orient new arrivals, better default avatars and more recently the Linden Homes program. All of these and others I haven’t mentioned are directed at the problem of 99% resident failure. I believe it is the most critical issue facing the Lab, since if it continues as is, the resident population won’t grow and the Lab may eventually face the prospect of closing SL.
Another key strategy pursued by the Lab for retaining new residents is an improvement to the viewer, and I suspect the introduction of Viewer 2.0 will generate very significant controversy into the SL blogosphere. Many people will be confused by different interfaces and truly unhappy they must learn something new.
But it must be done.
I believe one of the biggest impediments to retaining new residents is the staggering complexity of the viewer. As a student of clean design, I am astonished at the unbelievable amount of buttons and controls that face every user each time they fire up a viewer. The numerous controls are hidden within layers of menus, dialog boxes and bars, seemingly designed to be never found by humans.
There are so many controls it must be totally overwhelming to almost everyone. Maybe even 99% of population. Hmm.
But how complex is it? I wondered about this and finally decided to find out myself. I launched a recent version of Snowglobe and laboriously counted up all the controls within this viewer. My counts are probably off a bit, as my eyes were fading due to the humongous counts and deep layers one must go to see them, but the counts are representative of virtually every SL viewer. Not including “close boxes”, inventory items and any personally-generated controls (e.g. gestures, etc.) I found the following:
- Menu Items: 504
- Ticky Boxes: 200
- Text/Image Entry Areas: 201
- Adjustment Sliders: 282
- Tabs: 66
- Radio Buttons: 41
- Regular Buttons: 414
Grand Total: 1,708
Let’s step back and think about this for a moment: One thousand, seven hundred and eight control items. ONE THOUSAND SEVEN HUNDRED EIGHT! That’s This Many:
XXXXXXXXXX XXXXXXXXXX XXXXXXXXXX XXXXXXXXXX XXXXXXXXXX XXXXXXXXXX XXXXXXXXXX XXXXXXXXXX XXXXXXXXXX XXXXXXXXXX XXXXXXXXXX XXXXXXXXXX XXXXXXXXXX XXXXXXXXXX XXXXXXXXXX XXXXXXXXXX XXXXXXXXXX XXXXXXXXXX XXXXXXXXXX XXXXXXXXXX XXXXXXXXXX XXXXXXXXXX XXXXXXXXXX XXXXXXXXXX XXXXXXXXXX XXXXXXXXXX XXXXXXXXXX XXXXXXXXXX XXXXXXXXXX XXXXXXXXXX XXXXXXXXXX XXXXXXXXXX XXXXXXXXXX XXXXXXXXXX XXXXXXXXXX XXXXXXXXXX XXXXXXXXXX XXXXXXXXXX XXXXXXXXXX XXXXXXXXXX XXXXXXXXXX XXXXXXXXXX XXXXXXXXXX XXXXXXXXXX XXXXXXXXXX XXXXXXXXXX XXXXXXXXXX XXXXXXXXXX XXXXXXXXXX XXXXXXXXXX XXXXXXXXXX XXXXXXXXXX XXXXXXXXXX XXXXXXXXXX XXXXXXXXXX XXXXXXXXXX XXXXXXXXXX XXXXXXXXXX XXXXXXXXXX XXXXXXXXXX XXXXXXXXXX XXXXXXXXXX XXXXXXXXXX XXXXXXXXXX XXXXXXXXXX XXXXXXXXXX XXXXXXXXXX XXXXXXXXXX XXXXXXXXXX XXXXXXXXXX XXXXXXXXXX XXXXXXXXXX XXXXXXXXXX XXXXXXXXXX XXXXXXXXXX XXXXXXXXXX XXXXXXXXXX XXXXXXXXXX XXXXXXXXXX XXXXXXXXXX XXXXXXXXXX XXXXXXXXXX XXXXXXXXXX XXXXXXXXXX XXXXXXXXXX XXXXXXXXXX XXXXXXXXXX XXXXXXXXXX XXXXXXXXXX XXXXXXXXXX XXXXXXXXXX XXXXXXXXXX XXXXXXXXXX XXXXXXXXXX XXXXXXXXXX XXXXXXXXXX XXXXXXXXXX XXXXXXXXXX XXXXXXXXXX XXXXXXXXXX XXXXXXXXXX XXXXXXXXXX XXXXXXXXXX XXXXXXXXXX XXXXXXXXXX XXXXXXXXXX XXXXXXXXXX XXXXXXXXXX XXXXXXXXXX XXXXXXXXXX XXXXXXXXXX XXXXXXXXXX XXXXXXXXXX XXXXXXXXXX XXXXXXXXXX XXXXXXXXXX XXXXXXXXXX XXXXXXXXXX XXXXXXXXXX XXXXXXXXXX XXXXXXXXXX XXXXXXXXXX XXXXXXXXXX XXXXXXXXXX XXXXXXXXXX XXXXXXXXXX XXXXXXXXXX XXXXXXXXXX XXXXXXXXXX XXXXXXXXXX XXXXXXXXXX XXXXXXXXXX XXXXXXXXXX XXXXXXXXXX XXXXXXXXXX XXXXXXXXXX XXXXXXXXXX XXXXXXXXXX XXXXXXXXXX XXXXXXXXXX XXXXXXXXXX XXXXXXXXXX XXXXXXXXXX XXXXXXXXXX XXXXXXXXXX XXXXXXXXXX XXXXXXXXXX XXXXXXXXXX XXXXXXXXXX XXXXXXXXXX XXXXXXXXXX XXXXXXXXXX XXXXXXXXXX XXXXXXXXXX XXXXXXXXXX XXXXXXXXXX XXXXXXXXXX XXXXXXXXXX XXXXXXXXXX XXXXXXXXXX XXXXXXXXXX XXXXXXXXXX XXXXXXXXXX XXXXXXXXXX XXXXXXXXXX XXXXXXXXXX XXXXXXXXXX XXXXXXXXXX XXXXXXXXXX XXXXXXXXXX XXXXXXXX
There are more than 500 menu items alone. Do you know what they all are? Do you know what they do? Do you know where to find each of them? How long would it take to explain them to someone? Can you even know 100 of them?
Let’s compare this to other relatively complex things you must operate: your car, for example, probably has less than 100 knobs you can turn (without going under the hood). My browser appears to have less than 200 items. I suspect the flight control panel of a Boeing 747 has less than 1,708 buttons - and there’s serious training courses required to operate one of those.
And it gets much worse. If you enable the Advanced Menu in the SL browser, my totals went to this:
- Menu Items: 2523
- Ticky Boxes: 213
- Text/Image Entry Areas: 213
- Adjustment Sliders: 282
- Tabs: 72
- Radio Buttons: 44
- Regular Buttons: 445
Grand Total: 3,792
Many people enable the advanced menu in order to find a few evidently critical features, but must suffer through an extra two thousand controls. Not good.
And it’s even more complex on the third party open source viewers, which unfortunately tend to add features, not delete them. I can imagine the totals on the Emerald Viewer would be quite a bit higher than the above.
Many of these items are used for various personal customizations, and the remainder is quite a bit smaller. I wonder what this implies about avatar needs?
These astonishing totals no doubt cause many new residents to freak out. Wouldn’t you, if suddenly faced with a 1,708 item control panel? That’s like having a four-foot square sheet covered with knobs every square inch (or a 6 foot square panel if using the advanced menus) pushed in your face. I suspect that very few people have the perseverance, technical skill and motivation to make their way through the 1,708 controls in order to determine where the 36 key functions they’d actually use are located. Maybe, say, one percent?
That’s why Viewer 2.0 is so critical. That’s why Linden Lab has been silent on viewer progress lately - they are putting lots of effort into creating this new viewer.
I’m hoping Viewer 2.0 will dramatically simplify the interface - but allow experienced users to “turn on” more complex features when they need to. I’m hoping Viewer 2.0 organizes the controls in a contextual manner, whereby the appropriate controls are visible depending on the situation. I’m hoping Viewer 2.0 enables many more new residents to successfully join our virtual world.
Some may wish for more new features in Viewer 2.0, but I don’t think that’s what it’s about. Instead it should be about Fewer Features and better presentation of them. Viewer 2.0 is not about you. It’s about all the people that aren’t here.
Yet.
virtual business, building virtual products and exploring the virtual world.
17 comments:
Video Toaster on the Amiga ran one heck of a lot of TV stations a ways back. Nothing like turning on the tube to see the good old Amiga workbench after a crash.
Have a look at what I always thought was the worst UI in the universe until Visual Basic 3 hit the shelves.
http://www.digibarn.com/collections/systems/amiga2500/Image30.jpg
@ann - OMG, that is truly butt-ugly! However, it has FAR less controls than any SL Viewer and therefore a higher percentage of people could conceivably learn how to use it than the SL Viewer.
That's a lot of hard work, counting.
Now can somebody do a similar count in in the text-editor Microsoft word or similar? ;)
/Speedmaster bing
Hoping Viewer 2.0 goes some way to solving the problem. I'd like to see right now the ability to preconfigure the UI from command line switches and secondlife:// urls or slurls. Eg - turn off toolbar, or a selection of configurations that you could choose between in the same way. Of course no one would agree on what should go in each of those configurations...
For the past two years and a few months I've taken an active role in working with new residents. Counting...I recently did a rough estimate on how many nooblets I've worked with, the ones I"ve:
shown how to open boxes,
how to use camera,
how to set home,
change time of day,
use IM,
dressed,....
that number exceeds 1000.
A further estimated 2 to 3 hundred I've shown how to rezz a prim, move it, texture it, very basic stuff.
I can almost always cover the basics in under an hour... and while I find it personally rewarding...It's obvious that a *one on one* new resident experience just won't scale.
While a more simplified viewer might well help, I really hope the Lab makes all the current features available much as the Advanced Menue is today.
I must admit that a *dumbing down* concerns me. I came to Secondlife at 65 years old unable to even copy/paste...a Secondlife friend from Bangladesh really taught me the basics of not only Secondlife but PC use in general, {from 12,000 miles away}. I hate to use the old saw that if I can anyone can. The thing is if it's dumbed down...won't we lose the creativity?...will Secondlife just turn into a giant chatroom? I will stay as long as the lights are on...I just don't want to lose the flavor I found in early 2007...and still see.
"It's not about you, it's about the people who aren't here yet" renders as "It's not about the people who already pay for this now, it's about as-yet imaginary people who might expand Linden Research income someday."
Great post ArminasX! The first thing I did when I joined the Lab was to count the top-level menu items. I came up with a similar figure. In Viewer 2.0, I can tell you that the top level menus will have a fraction of the menu items you'll find in 1.23...even fewer than Firefox (according to my quick tally)!
We are not eliminating functionality, but we are re-organizing it. The menus in 1.23 are a bit like a game of "Find the thing that doesn't belong." Here's one...in the View menu (arguably the worst) you have a cluster that includes Local Chat, Communicate, Inventory, Active Speakers, Mute List. What is Inventory doing in that cluster? You got me.
Communicate controls are found in three different menu areas in 1.23 and some choices are duplicated.
This is an enormously complex redesign effort. The new viewer -- which will be out in public beta soon -- is the first major step in the journey to a natural and intuitive experience.
We're reorganizing and in doing so simplifying. I don't see it as dumbing down. In fact, I think we are doing the opposite. We are letting good and smart interface design do some of the hard work Residents have had to do to climb the learning curve in the past. The learning curve in SL is too steep today. It is a barrier to creativity and innovation. Imagine the great creative spirits we've lost because they were stymied by the interface in their first days/weeks/months.
I use Viewer 2.0 now and the transition was easy. Most everything is there and it's where you'd intuitively expect it to be. It's not perfect, and there is still work to do, but it is a big step forward.
Soon we'll have Viewer 2.0 out. By mid year you'll see a materially improved "first hour" experience for new Residents. We are going for "Delight." :)
Happy New Year!
M Linden
*coughs and spills her coffee*. Gah! I never thought it would be that bad!
Thanks for pointing this out, ArminasX, and thanks to M for explaining that the ordeal is almost over...
I would prefer less lag less crashing and more fps.
I think there is one feature Viewer 2.0 seriously needs if that "2.0" is going to be taken seriously.
The appearance editor needs to show avatar height. It needs to be right there, for every resident to see when they are editing their avatar.
It seems like a tiny detail, but the lack of this content creation tool (and make no mistake, that's exactly what it is, a content creation tool) has had far reaching consequences for SL.
If people had known the height of their avatars from the beginning, you would see a lot more variety in avatar sizes in SL, rather than nearly everyone scrunched into the tallest extreme end of the scale.
People would have been building more efficiently for years now, as making double-triple sized buildings has resulted in people wasting lots of area and space. Sims would be, for all intents and purposes, twice as large. (Avatar size and camera placement are what create your sense of scale. To a 5'9" avatar with an eye level camera a 10x10 room is as spacious as a 20x20m room appears to an 8' avatar with the default SL camera.)
Animations would work more consistently between avatars, since they could be designed with specific heights in mind.
Beyond just the indisputable benefits, finally (it's been over 6 years!)we would have a firm acknowledgement on the part of Linden Lab that they do understand the needs of creating content within the SL platform, and are finally working to address those needs.
The moral boost in that cannot be understated. Lacking such a basic and essential feature shows a fundamental lack of understanding which LL simply cannot afford if they want to seriously expand SL's appeal to a broader audience. Continuing that detached approach into Viewer 2.0 would signal "more of the same" to people who've stuck it out in SL for years now.
I can respect the need for more simplicity in the viewer, but speaking as a content creator who's been involved in SL since 2005, this is just one essential feature SL has gone without for far too long. Fixing the many problems caused by it will take time, so the sooner the base issue is addressed, the sooner SL can benefit.
I don't think it's about the upside total possible menus, etc... it's about helping the new person find the ones to learn the basic skills in SL.
I doubt I use 10% of the functionality of MS Word, for example, but when I needed it, I found what I needed. That connection of "what do you want to do?" to -> "here is how you do it" is the missing link, in my opinion.
Although I agree on simplifying the viewer, I also feel that the advanced options should still be available for power users. Don't kill off your dedicated user base in favor of new users.
Don't forget the debug settings as well (under the advanced menu). There are tons and tons to muck around in there.
Yeah, I agree with keeping the advanced items - it's just that they are not required for new users and many others, too. Keep them hidden until the right context arises, then surface them for convenient use.
I did count the debug settings. OMG. Don't ask.
Firstly, virtual worlds are complex places, and simplifying the interface can go only so far. I have my doubts whether newbie confusion owing to viewer complexity is the main cause of the low retention figures. I bet it's more related to there being no obvious fun things for newbies to do in SL, plus the lag on mid-to-low end equipment.
Secondly, there's no doubt that the LL viewer could be better organized, but it's a mistake to think that this is important for newbies only.
In the same way that we don't keep our sewing kit, hardware toolbox, cutlery drawer, and food cupboards all open in front of us all the time, so it is in virtual worlds --- we don't do all the possible activities simultaneously. The viewer's menu looks confusing because it quite naturally contains entries for all possible actions. It's a mistake to think that better partitioning will fix the problem for newbies. While it will help a little because some of the current menu assignments are clearly daft, even newbies want to do various different activities throughout the day, so the complexity will remain. Partitioning the menu into Beginner and Advanced sections doesn't reflect reality --- it's really a continuum of expertise.
It should also be said that designing an interface for the primary purpose of newbie retention is completely missing the point, and as usual, ignoring the residents. Everybody deserves a good UI, not just newbies.
Ultimately there's no solution to this problem other than a customizable interface, with a default simple setting for beginners. That would improve everyone's daily experience in SL. If the design of Viewer2.0 was indeed focused specifically on designing for newbie retention, then I have to say that this was a remarkably myopic policy decision.
*just shakes his head*
The viewer needs customizable, hot swappable menus so advanced users can set up the UI the way they want and switch to a default or help mode setup when it is necessary to explain or fix a problem for a new resident and then switch back again.
This ensures that both parties are seeing the exact same screen and help can be administered more efficiently.
Once a new resident has become more familiar with Second Life, they can create their own menus and place the tools they are most comfortable with within reach without exposing anything they do not need.
This also serves to enhance the experience of the more seasoned resident as they can put all of the advanced tools they need within easy reach for different tasks. For instance, a menu specifically customized for managing land or moderating a sandbox. A custom menu that puts only the necessary tools at hand for making a machinima, one for building, one for scripting and texturing etc. All of them easily swappable on the fly.
Hosing the interface for everyone right across the board with no ability to rectify it and make it useful to them is bad.
Go to http://inflife.net/ and get the SL Pop Viewer! Best viewer I have EVER used! You’ll be VERY happy with it, and you’ll never have to deal with those pesky slow logins and load times. The SL Pop Viewer completely blows away all other viewers by far!!! Go get it now!
Although this thread is old, obviously, some people still look for such things - like me.
Viewer issues are always a concern, but i Don't think that's the biggest turnoff for new residents. Most people expect to have to learn new controls for a "game". I think the bigger issue is what people see when they first log in.
Their avatar is goofy looking, it's wearing ugly clothes, the area has bland buildings, and they walk like a duck. In an age of slick graphics designed to impress, the noob areas are BORING.
There are tons of ways to give a better first impression and to get experienced users to volunteer to help out the noobs. As far as I can tell NONE of them are used though.
Post a Comment