I wouldn't really call it a mouse, but it's a good idea to decouple the cursor on a touch-screen from the point of finger contact, for visibility reasons. Somehow, I don't seem to have this problem with the iPhone, it clicks the right links almost all the time, but it used to drive me nuts on Android. Maybe the screen is physically more precise on the iPhone, but this is definitely solving a pain point for Android devices. The gesture system is just the icing on the cake, it makes interaction so much smoother.
Hi Tekahera, how would you explain that to your grandma?. We decided to go colloquial and massive. Its a bet but certainly has arguments to say its a multitouch mouse.
There was a really interesting panel at SXSW this year regarding the 'Future of Touch User Interface' by two guys from Microsoft (http://schedule.sxsw.com/events/event_IAP5984). I think trying to create a mouse interface on a touch device is almost a regression, not a step forward. I'd much rather see improved touch w/ voice integration, better multitouch, etc.
Its great to see the article on the future touch interface maerek.
As leader of RoamTouch and creator of its vision I can tell that we did not decide to simulate a mouse at all. We saw the picture of the future (five coming years) in 2006, the household and we are not improvising. Today PadKite is a mouse and playful kite, its also the hook for a new type of integration that aims to run on top of other applications and video, grow not only on the tablet and smart phone but also to grow to be remote controlled in a living room scenario (coming decade arena). PadKite is the proxy or bridge between today and the future.
In my opinion the future of computer is tactile. It is the most effective way to interact with a computer with the less energy consumption. The future is not in voice and body movement recognition technology. I do not imagine a working day (6-8 hs non stop) jumping or speaking, even moving arms to do 3D stuff. Its all about energy consumption and the most effective way to use our brains are with out hands, and millions of years have passed for that to be in such way.
PadKite and the GestureKit are an approach to start thinking about the big picture to come. It will be fun!.
I think it's a very interesting concept. My only concern is using two fingers for most of functions. Personally for me, it's not practical. Also, as the video mentioned, our fingers are big and clunky, so an added finger would block even more of the screen from the user. But I love the idea and the effort put towards solving this problem.
We infact we implemented Padkite to be controlled with a single finger but we left it aside cause multitouch gave more power. The problem was that the selection gestures where not that good with single finger cause you need to coordinate. What did work very well was the navigation experience and basic stuff.
Do you think we should include an optional single finger model on the settings?. It works like this: Drag, lift finger to select below and do a gesture with same finger. All in one hand.
What do you think?. Let me know man, I am intrigued.
Oh that is nice. You know I thought the name cuase its playful and in a way you are controlling something that is flying. But the most important thing is that when you dont use it you leave it aside flying like a kite while you continue working.
When you leave it aside you can do stuff as you already know, click, pan zoom. But when you bring it in you fly.
BTW, the kite concept will be enhanced when you control it remotely in the near future.
I really like the idea, but it seems from the video that the cursor is really quite a distance from the users finger.
I agree that we may want a small degree of improvement, but I think it is important to keep the simplicity of touch.
Can you maybe give some insight into why you decided to have such a large distance from the users finger to the cursor? as opposed to just having a cursor be an pointy extension to our fingers?
BTW, the video is a spot and not meant to be accurate but highly communicative.
We did that to make it wide opened to all sizes. We thought that the user was the one who should decide, therefore the opened distance. After using the mouse a couple of times, the user should adjust the finger to the best distance according to his/her size. Now a very big finger fit and a small one too. In the future you will be able to shrink the circle or enlarge it and tune it best.
On the other hand there is a visible area factor that made us leave that amount of size avalable to see. For example, when you are selecting content, you need to know what is below, the same when you are just viewing the different elements below the cursor.
A cursor in a point extension of your finger is a good solution but cannot work alone, its not enought to create a real solution. You need to execute, select, mark, etc. We could have stopped there like for example this project did http://www.youtube.com/watch?v=kkoFlDArYks. They found the offset but stopped there and didn't follow that great discovery, they kept on searching. We, on the other hand focused on that point, the the two regions that now are the mouse buttons allowed us to work on selection.
I couldn't have asked for a better answer. Thanks for explaining that. I assumed you wouldn't have gone through all the work without dealing with that issue, and was curious behind the reasoning.
It looks fiddly for constant use, but if it were toggleable, provided it's quick to invoke (and difficult to accidentally invoke) it'd be great for the situations where small links are a pain. Note that in this case it has to be less effort to invoke than a zoom action.
I'm not sure how much people hate animated gifs but you might want to consider adding a larger one, because I didn't get how the basic operation of the mouse cursor worked until I watched the video (the descriptions and static graphics didn't really explain much to me).
Also maybe make it a bit more obvious that it's android-only on the front page.
Superb feedback man, great info and very valuable. Taking lot from it.
You know, our most difficult challenge today apart from overcoming the technical issues of compiling in all versions and fixing all bugs, its the learning curve.
The user needs to get to the basic operations at glance and I agree with you that we should show that at front. Paradoxically we thought the video was the best, however an image can also speak million words. I consider your feedback.
Do you have a hammer in your hand fuzzythinker?. Cause I think you hit the nail on the head.
Definitely our logo and symbol '^' means to fly and that is what the app gives, power to fly. Invoking it on a friendly scenario for you with that gesture and using it would be really cool right?.
Many good ideas guys. Lots of stuff to consider today.
I can't count how many times a day I misclick on my Galaxy S or on my old G1, despite my skeleton fingers. You must be right, as I've never had that problem on any generation iPhone.
What if this is a solution in need of a more interesting problem?
I don't remember who said this, but hidden gestures in a multitouch device have the exact same learning curve as command line: they are equally magical, efficient, and completely insane in the eyes of people who haven't learned them yet.
That's not to say that the command line is hard (it's not), or that Bash is arbitrary and random (it has definite syntax and conventions).
But you're betting that people will use these new gestures to do the things they're already doing. Maybe they're actually more interesting for drawing or something?
Very interesting conversation, waited years for having them. Its good to see smart guys around.
Gestures are old, Palm graffiti did a lot already but was a different approach, they worked alone, not for the service of other gestures like ours. We are bringing in new gestures and overcoming the imagery gesture paradigm. First the selection gestures that we are "inventing" product of the offset pointer and the two button areas. And second type the graffiti type, which are a set of three to six gestures that the user will learn if he wants.
The novelty of what we are doing (quite a lot in terms of interaction) is that our graffiti type (we call it our GestureKit, in fact is a brand) works in conjunction and coordination with the selection gestures. It is an ongoing process of doing three movements, drag-select-gesture which makes it very fast and productive. The user could skip the latter but once the selection gesture finger is lifted up and the content selected is very handy to do a last touch down and make a gesture to command, for example the "S" for search on Google. It could be whatever really.
Try it and tell me what you think please. These are not just command line gesture, they belong to an interactive system that all parts together make things much faster. The users know already to do that by other means, we are offering to do it faster.
Attempted to test it, but it crashed after I clicked "Skip" on the tutorial. I am able to see a white screen with the cursor for about 2 seconds before it closes. Using an Evo 4G running Cyanogenmod 7 RC2 (2.3).
I agree with zachrose, but that doesn't mean there isn't a market for power users who will appreciate this work.
My concern is that the primary multi-touch gestures choosen are two handed. Unlike pinch, stretch, and swipe, holding onw finger in place while you begin another gesture require more coordination than most people can do with one hand. In my mind, this leaves it out of the realm of phone-sized devices on the go. It probably works great when working on a table or desk, which means it is really the mouse for a tablet environment. In that market, you may have a real winner.
Years of work worth reading your feedback, appreciate very much.
I am very interested what you say here:
"My concern is that the primary multi-touch gestures chosen are two handed.". I think that ultimately there will have to be a standard for handed, but an evolved set of gestures. I am working on that, you know why?
I strongly believe that there has to be a balance in terms of handed gestures to execute on what has been selected (our innovation, do a gesture on a selection). There will be a time when gestures have to evolve to cope with demanding tasks at home with multiple touch devices in almost all glass/crystal/plastic surfaces, so there is going to be plenty of need. New types of gestures that can be learned and reproduced fast. Nor the palm graffiti type nor the imagery type (pinch, stretch, and swipe). New types.
Something is in the middle that can be worldwide interpreted and very easy to execute, that is the key. As an example we created the Twitter peak ">" gesture cause is fast and symbolic. That creation process has to be applied many more times again for new gestures with an implicit action and a symbol. Also the key can be to synthesize the gesture to a minimal expression.
Also, once the user has done the gesture four times, the fifth goes on its own. So its all about learning curve. We humans have a great ability to remember mechanical movements. I would love to have a more consistent scientific background. Any doctor around?.
This is the way I think we can overcome the real problem and concern you are pointing out and that I agree 100%. Of course tablet will be a superb place to run PadKite.
"As an example we created the Twitter peak ">" gesture cause is fast and symbolic."
That is a problem though. It is neither fast or symbolic if you do not know it already or you have forgotten it.
What if this app had been out a long time ago and f was the gesture for Friendster?
What about a single gesture that brings up a list of actions, a further click then allows the user to post to facebook, tweet, email, or, Google it, etc.
This looks cool but why would people use this instead of pinch,zoom,click?
I feel the whole point of touch interfaces is to move away from using a mouse. Instead of mobile touchscreen devices replicating desktop gestures, websites will move towards designs more suitable for the new dominant platforms. OnSwipe is a case in point.
The kite metaphor is interesting though. If this multimouse idea doesn't fly (pun?), you could turn it into a kite wars game. Use proximity sensors (it's the in thing) to pick up fights with people near you. Sharing photos with strangers is not for me, but I will pick up a good kite fight challenge any time.
The point is we want people to continue doing pinch, zoom and click so that padkite is used only when needed. In fact outside the circle you can do all that you already do.
It is not padkite OR the other thing, its padkite PLUS the other thing, which is very handy. What is good has to be kept.
Ok I just watched all of your demo videos again. Like others have pointed out, this is for power users. The basic usage is complex enough, you have a TON of other features.
A good demo video of someone using an app in an everyday situation, clearly showing them holding the phone in one hand (not keeping it on a table), will really help showcase padkite's power.
You've clearly built some amazing tech, but you need to educate people on how to use it. You should seriously consider making a game which uses these gestures, so people learn the gestures while having fun.
Absolutely right about the video with the showcase. We are already working on a new script for it. These are the things that we will take care of before we launch. Thanks lot for the comment.
35 comments
[ 3.0 ms ] story [ 83.0 ms ] threadCool comment, appreciate much.
As leader of RoamTouch and creator of its vision I can tell that we did not decide to simulate a mouse at all. We saw the picture of the future (five coming years) in 2006, the household and we are not improvising. Today PadKite is a mouse and playful kite, its also the hook for a new type of integration that aims to run on top of other applications and video, grow not only on the tablet and smart phone but also to grow to be remote controlled in a living room scenario (coming decade arena). PadKite is the proxy or bridge between today and the future.
In my opinion the future of computer is tactile. It is the most effective way to interact with a computer with the less energy consumption. The future is not in voice and body movement recognition technology. I do not imagine a working day (6-8 hs non stop) jumping or speaking, even moving arms to do 3D stuff. Its all about energy consumption and the most effective way to use our brains are with out hands, and millions of years have passed for that to be in such way.
PadKite and the GestureKit are an approach to start thinking about the big picture to come. It will be fun!.
And like someone else already pointed out, one handed control is the holy grail here...
We infact we implemented Padkite to be controlled with a single finger but we left it aside cause multitouch gave more power. The problem was that the selection gestures where not that good with single finger cause you need to coordinate. What did work very well was the navigation experience and basic stuff.
Do you think we should include an optional single finger model on the settings?. It works like this: Drag, lift finger to select below and do a gesture with same finger. All in one hand.
What do you think?. Let me know man, I am intrigued.
When you leave it aside you can do stuff as you already know, click, pan zoom. But when you bring it in you fly.
BTW, the kite concept will be enhanced when you control it remotely in the near future.
I agree that we may want a small degree of improvement, but I think it is important to keep the simplicity of touch.
Can you maybe give some insight into why you decided to have such a large distance from the users finger to the cursor? as opposed to just having a cursor be an pointy extension to our fingers?
BTW, the video is a spot and not meant to be accurate but highly communicative.
We did that to make it wide opened to all sizes. We thought that the user was the one who should decide, therefore the opened distance. After using the mouse a couple of times, the user should adjust the finger to the best distance according to his/her size. Now a very big finger fit and a small one too. In the future you will be able to shrink the circle or enlarge it and tune it best.
On the other hand there is a visible area factor that made us leave that amount of size avalable to see. For example, when you are selecting content, you need to know what is below, the same when you are just viewing the different elements below the cursor.
A cursor in a point extension of your finger is a good solution but cannot work alone, its not enought to create a real solution. You need to execute, select, mark, etc. We could have stopped there like for example this project did http://www.youtube.com/watch?v=kkoFlDArYks. They found the offset but stopped there and didn't follow that great discovery, they kept on searching. We, on the other hand focused on that point, the the two regions that now are the mouse buttons allowed us to work on selection.
Hope I answered your question.
Cheers.
I'm not sure how much people hate animated gifs but you might want to consider adding a larger one, because I didn't get how the basic operation of the mouse cursor worked until I watched the video (the descriptions and static graphics didn't really explain much to me).
Also maybe make it a bit more obvious that it's android-only on the front page.
Superb feedback man, great info and very valuable. Taking lot from it.
You know, our most difficult challenge today apart from overcoming the technical issues of compiling in all versions and fixing all bugs, its the learning curve.
The user needs to get to the basic operations at glance and I agree with you that we should show that at front. Paradoxically we thought the video was the best, however an image can also speak million words. I consider your feedback.
Please forward me any more idea you might have.
Cheers.
Definitely our logo and symbol '^' means to fly and that is what the app gives, power to fly. Invoking it on a friendly scenario for you with that gesture and using it would be really cool right?.
Many good ideas guys. Lots of stuff to consider today.
Excellent!. Lets fly!. Kind regards.
Maybe PadKite is a solution to sub-par touch sensor performance on some Android devices, as shown here:
http://www.gadgetsdna.com/how-do-the-smartphones-rank-in-ter...
I don't remember who said this, but hidden gestures in a multitouch device have the exact same learning curve as command line: they are equally magical, efficient, and completely insane in the eyes of people who haven't learned them yet.
That's not to say that the command line is hard (it's not), or that Bash is arbitrary and random (it has definite syntax and conventions).
But you're betting that people will use these new gestures to do the things they're already doing. Maybe they're actually more interesting for drawing or something?
Very interesting conversation, waited years for having them. Its good to see smart guys around.
Gestures are old, Palm graffiti did a lot already but was a different approach, they worked alone, not for the service of other gestures like ours. We are bringing in new gestures and overcoming the imagery gesture paradigm. First the selection gestures that we are "inventing" product of the offset pointer and the two button areas. And second type the graffiti type, which are a set of three to six gestures that the user will learn if he wants.
The novelty of what we are doing (quite a lot in terms of interaction) is that our graffiti type (we call it our GestureKit, in fact is a brand) works in conjunction and coordination with the selection gestures. It is an ongoing process of doing three movements, drag-select-gesture which makes it very fast and productive. The user could skip the latter but once the selection gesture finger is lifted up and the content selected is very handy to do a last touch down and make a gesture to command, for example the "S" for search on Google. It could be whatever really.
Try it and tell me what you think please. These are not just command line gesture, they belong to an interactive system that all parts together make things much faster. The users know already to do that by other means, we are offering to do it faster.
Thanks lot.
Apologize for that. Please let me know your email and I will come to you with the proper build.
Mine is jose.vigil ad roamtouch dot com.
Thanks.
My concern is that the primary multi-touch gestures choosen are two handed. Unlike pinch, stretch, and swipe, holding onw finger in place while you begin another gesture require more coordination than most people can do with one hand. In my mind, this leaves it out of the realm of phone-sized devices on the go. It probably works great when working on a table or desk, which means it is really the mouse for a tablet environment. In that market, you may have a real winner.
Years of work worth reading your feedback, appreciate very much.
I am very interested what you say here: "My concern is that the primary multi-touch gestures chosen are two handed.". I think that ultimately there will have to be a standard for handed, but an evolved set of gestures. I am working on that, you know why?
I strongly believe that there has to be a balance in terms of handed gestures to execute on what has been selected (our innovation, do a gesture on a selection). There will be a time when gestures have to evolve to cope with demanding tasks at home with multiple touch devices in almost all glass/crystal/plastic surfaces, so there is going to be plenty of need. New types of gestures that can be learned and reproduced fast. Nor the palm graffiti type nor the imagery type (pinch, stretch, and swipe). New types.
Something is in the middle that can be worldwide interpreted and very easy to execute, that is the key. As an example we created the Twitter peak ">" gesture cause is fast and symbolic. That creation process has to be applied many more times again for new gestures with an implicit action and a symbol. Also the key can be to synthesize the gesture to a minimal expression.
Also, once the user has done the gesture four times, the fifth goes on its own. So its all about learning curve. We humans have a great ability to remember mechanical movements. I would love to have a more consistent scientific background. Any doctor around?.
This is the way I think we can overcome the real problem and concern you are pointing out and that I agree 100%. Of course tablet will be a superb place to run PadKite.
Great feedback, really.
That is a problem though. It is neither fast or symbolic if you do not know it already or you have forgotten it.
What if this app had been out a long time ago and f was the gesture for Friendster?
What about a single gesture that brings up a list of actions, a further click then allows the user to post to facebook, tweet, email, or, Google it, etc.
I feel the whole point of touch interfaces is to move away from using a mouse. Instead of mobile touchscreen devices replicating desktop gestures, websites will move towards designs more suitable for the new dominant platforms. OnSwipe is a case in point.
The kite metaphor is interesting though. If this multimouse idea doesn't fly (pun?), you could turn it into a kite wars game. Use proximity sensors (it's the in thing) to pick up fights with people near you. Sharing photos with strangers is not for me, but I will pick up a good kite fight challenge any time.
Good luck!
The point is we want people to continue doing pinch, zoom and click so that padkite is used only when needed. In fact outside the circle you can do all that you already do.
It is not padkite OR the other thing, its padkite PLUS the other thing, which is very handy. What is good has to be kept.
But if you want to fly.....you just invoke it.
Thanks very much.
A good demo video of someone using an app in an everyday situation, clearly showing them holding the phone in one hand (not keeping it on a table), will really help showcase padkite's power.
You've clearly built some amazing tech, but you need to educate people on how to use it. You should seriously consider making a game which uses these gestures, so people learn the gestures while having fun.
The Wii primed people for the Kinect.
Absolutely right about the video with the showcase. We are already working on a new script for it. These are the things that we will take care of before we launch. Thanks lot for the comment.
As regarding the game, I have already thought that but its a very drafted idea yet. I think this type of HTML game could help http://mrdoob.com/projects/chromeexperiments/ball_pool/
We will also put more tiny links on the home page to pay with in order to both enhance the concept and provide useful data to the user and small ads.
What do you think?.
Superb feedback.
Please keep us posted. Never fall off the radar, make some noise. Break a leg.