touch demo for tomorrow on hacks
Christopher Blizzard
I'm going to use Marcio's demo on hacks tomorrow. Marcio, what do you
have to get it up and running? Jay Patel wanted to know.
--Chris
Marcio Galli
Felipe (cc:ed ) prepared a build for us to play at the event, using the
Mozilla HP laptop, with the latest multitouch gesture support he worked on
- https://wiki.mozilla.org/Firefox/Projects/Multitouch_Polish.
BTW Sarah also had another of these multitouch machines running the right
Firefox build.
/\/\
PS: This week I am stuck but I wanted to do an update release to the drawing
app as soon as I can touch the multitouch bulid computer for another round..
I also mentioned at Felipe's talk on multitouch support, about a need for
Mozilla to help developers - some kinda of emulation way so that a
developer can do stuff without buying the computer.
On Mon, Oct 12, 2009 at 6:04 PM, Christopher Blizzard
<bliz...@mozilla.com>wrote:
> _______________________________________________
> evangelism mailing list
> evang...@lists.mozilla.org
> https://lists.mozilla.org/listinfo/evangelism
>
--
Taboca - Laboratório Web
www.taboca.com
+55 16 8213 3210
+55 16 3374 9853 ( fax )
Christopher Blizzard
would be, uhh, awesome. Think about spinning a 3D model or something
even more complicated.
--Chris
Al MacDonald
Lets' put something 3D multi-touch out that shows how Firefox could be
used on interactive tables and other modern touch devices. It would be
fun to make some kind of simple musical puzzle game that uses <audio>
tags, but before that we can make a Hello World 3D cube that responds to
touch.
Felipe Gomes wrote:
> Yeah, that'd be awesome! I don't know much about webgl yet but it'd
> be great to collaborate with Al and Vlad to build some cool demos!
>
> Felipe
Christopher Blizzard
> OK, time to go get Win7!
Oh, yeah, you have that touch laptop from HP, right? No idea what the
driver support looks like. And it sounds like you need some stuff
that's not in-tree quite yet as well? But sounds like Felipe and others
are all over it.
>
> Lets' put something 3D multi-touch out that shows how Firefox could be
> used on interactive tables and other modern touch devices. It would be
> fun to make some kind of simple musical puzzle game that uses <audio>
> tags, but before that we can make a Hello World 3D cube that responds
> to touch.
>
Man, that would be fun.
--Chirs
Al MacDonald
What are the specs/model number for your multi-touch device? I am
thinking of getting the same setup as you and working from there.
Al
Al MacDonald
Multi-touch emulation is something I would be very interested in. I am
currently building a simulated event-based DOM for the Canvas, I was
thinking of adding multiple cursor support for the future. If any
simulated event hooks have been exposed to Javascript, I would be very
excited to test these out.
Kind regards,
Al
Marcio Galli
Yes - the emulation layer as a means for others to experiece certainly is
going to help for many other projects. Since I coded the very few lines in
the drawing app area, I should put some priority for maintenance to that one
for now. Aside from that, I wanted to contribute with some hack
experimentation where people could do the whole system at home. So the
emulation/simulation layer is a component but would not be eveything. One of
the things I did before is the CamCanvas API - very early stage of a Camera
( Flash ) to Canvas via JavaScript - it works all right - probably not too
fast. My feeling is that I should be able to hack in a detection experience
to use the camera as a means to get input that can later be sent as touch
events - plug into this emulation simulation layer thing..
/\/\arcio
--
Al MacDonald
> Hey Al, the computer is a HP TouchSmart TX2, and the specs are 2.2GHz
> AMD Turion X2 processor and ATi Radeon HD 3200 graphics, 4GB Ram.
> It's not reaally developer-focused, more like consumer-grade, but
> works well for me.
>
> I'm really interesting in knowing more about your work of simulating
> events for the Canvas. As me and Marcio brainstormed last week we
> noticed how useful would be to have a simulation layer for multitouch
> events as well as a means to allow more people to experiment and
> develop using these events without actually having the multi-touch
> device. So I'm curious in what's the format that you're using, what's
> the library like.
>
> We were initially imagining some sort of add-on or library that could:
> - generate touch events programmatically if you don't have access at
> all to a touchscreen (like defining a bezier curve or path to follow,
> circle movements, etc)
> - record events from an actual device and reproduce then later as a
> way to create a test framework or using the real-life recorded events
> on a non-touchscreen computer
>
>
> Felipe
>
>
> On Tue, Oct 13, 2009 at 11:29 AM, Al MacDonald
> <dire...@hyper-metrix.com <mailto:dire...@hyper-metrix.com>> wrote:
>
> Hi Marcio, Felipe,
>
> Multi-touch emulation is something I would be very interested in.
> I am currently building a simulated event-based DOM for the
> Canvas, I was thinking of adding multiple cursor support for the
> future. If any simulated event hooks have been exposed to
> Javascript, I would be very excited to test these out.
>
> Kind regards,
>
> Al
>
>
It would be neat to create a small script, which creates multiple new
cursor objects, each with the ability to record it's position in
JavaScript spacetime. Pair this with a nice controller to scale,
translate, rotate gestures around in spacetime, and it could be a
powerful tool to aid UI development and testing in Open Web interfaces.
Then the engine could be extended to measure differentials between an
input buffer of say... 50 frames, and an n-tree of pre-recorded buffers
containing gestures. This way you could also play back the gesture
buffers to trigger real multitouch events within the realms of
abstracted JavaScript event listeners.
The greater the cancellation difference between the last 50 frames and
the checked gesture... the less likely it is you have a match. I wonder
if each gesture would need a separate accuracy threshold, I bet it
would! So the engine would also need a way of simulating alternate
magnitudes of "random" inaccuracy.
Imagine you could push touch events over to another Window through an
IFrame also running the simulator, and you could receive and run
multitouch events from a alien source, letting people work
collaboratively in the same HTML document... better... the same Window
object. ( I have to jump in and coin this one 'Teletouch" )
Then you could have rack of computers in test-mode, firing inbound test
gestures and bouncing out the subsequent state-change information back
to the server for analysis. Or... you could just gaze in wonder at all
the little screens being controlled from the touch-hardware of a single
device. :)
It could be a good fit for something I am working on right now called
the ECanvas. You can take a look at GitHub if you like:
http://github.com/F1LT3R/ecanvas
Al