This is my latest startup. The current version (0.1) is a very early version, it is not intended for production usage yet. I am interested in opinions on the approach, and also would be happy to find investors.
Sorry man, this just goes against the grain of the web - the future is imo in using browser as a smart, hw accelerated runtime that is native on every platform, not in creating a dumb/slow proprietary emulation layer inside. Flash has already lost, and whatever good it had (fast accelerated video) has already been made native.
The problem with this is that the web is driven by commitees, so innovation is terribly slow. For example there is no real dynamic layout in web standards, which is crazy. (Look at my v0.1 GridPanel class and compare it to CSS layout...) I want to create a system in which people can create their own components, own layout managers, etc... This thing should not replace the web, but has a place in UI tech research and application development.
Also we already have a lot of browser UI toolkits that behave the same in modern browsers. I don't really feel this could bring anything new to the table.
A lot of the time in fact we don't want our UI to look the same everywhere, for example I would love my site to scale down and degrade gracefully on mobile or in other words: progressive enhancement.
Performance and legacy browsers still plague the canvas element, so I won't rely on that as the foundation of a UI technology.
And last but not least: this UI framework has to duplicate a lot of behavior that the browser already does, which I feel is both a waste of time and a waste of resources.
I don't want to look too negative about this, but I'm really skeptical that a UI framework like this would be better than what we already have.
the rendering sometimes fails too. for example, the button won't show until i click on them. and resizing the windows won't properly render the resulting windows.
Perhaps - I'm on OSX. Tried it in Chrome on the same system and it works fine, so it's definitely an OSX/Firefox issue. Ping me here if you make changes and need someone to retest...
If we call this 'flash': My dream is to create a better flash than flash. I do this in my spare time, so the features a I've put into this are very limited as of yet, but once I will have more time/resources I will try to do my best on things like accessibility.
IMO this is interesting as a technical exercise, and stops there.
Its dynamic layout, dynamic stylesheets, customizable
painting and conceptual simplicity beats modern
desktop/plugin/platform-locked technologies
That sounds like HTML+CSS+JS.
Bitmap drawing is nothing new. What are the expected advantages of creating a whole new renderer, graphics library, and UI kit inside an enviroment that already offers essentially the same?
15 comments
[ 4.4 ms ] story [ 43.3 ms ] threadIn the long run, I believe this is the wrong approach. Scenegraph-based UI can be easier to HW accelerate.
These are also dead or dying platforms. I don't see a compelling reason for another one to rise, even if it runs atop canvas.
http://blogs.gnome.org/alexl/2010/11/23/gtk3-vs-html5/
Also we already have a lot of browser UI toolkits that behave the same in modern browsers. I don't really feel this could bring anything new to the table.
A lot of the time in fact we don't want our UI to look the same everywhere, for example I would love my site to scale down and degrade gracefully on mobile or in other words: progressive enhancement.
Performance and legacy browsers still plague the canvas element, so I won't rely on that as the foundation of a UI technology.
And last but not least: this UI framework has to duplicate a lot of behavior that the browser already does, which I feel is both a waste of time and a waste of resources.
I don't want to look too negative about this, but I'm really skeptical that a UI framework like this would be better than what we already have.
the rendering sometimes fails too. for example, the button won't show until i click on them. and resizing the windows won't properly render the resulting windows.
firefox 8 also.
Bitmap drawing is nothing new. What are the expected advantages of creating a whole new renderer, graphics library, and UI kit inside an enviroment that already offers essentially the same?