Famo.us is now open source
github.com
github.com
The closed nature of their 'open source' project can be frustrating at times. Having said that and despite all the hype, famo.us is sitting in a pretty unique position in the html mobile web scene and frankly I'm more excited about it than any of the alternatives (sencha, jquery mobile, etc).
It's early days, I wouldn't write it off yet...
> Famo.us is built around a neat idea: by directly using the CSS matrix3d transform in combination with the window.requestAnimationFrame function, you can describe the complete layout and animation of your app in a way that’s hardware accelerated with consistent performance.
> It’s a stroke of genius, but in order to implement that simple idea you need a sophisticated math library to help translate your app’s UI into the series of matrix transformations that get pushed to the GPU. This library is Famo.us.
Demos - http://demo.famo.us/
Codepen Examples - http://www.codepen.io/befamous/
edit: Also found this - http://famo.us/about
I would say "promises" instead of "allows". It is still experimental.
I feel like the engine reminds me more of the slickness and responsiveness of a Window Mobile 8 interface as opposed to an iOS interface with how smooth some of the scrolling is.
https://github.com/Famous/examples
I'm even more confused now. I looked at their other repos and there is a smattering of information to be found in "Guides". None of the demo links work except for the mobile page that someone posted below.
There's nothing on the actual website except a login and when i register I'm "76882 in line".
If I try to access the documentation (http://famo.us/docs), I'm told I need to be a developer and I can't access it.
A resounding meh, I've wasted 10 minutes trying to work out how this thing works and what I can do with it. Some other kind folks have provided CodePen links, but back to work for me.
Your spam-bot needs work; real humans don't emit things like this.
Never mind that, I'll just put my name at the end of this new 74k+ long waitlist. I'm sure this whole thing is just a really long on-boarding process.
Spotted a typo too, unviersity, can't seem to get back the source that I spotted it in, perhaps they're updating as we type.
There is quite a bunch of free engines already out there: Babylon.js, Copperlicht, CubicVR, TheeJS, ...
The native engine builders: Epic, Crytek and Unity are also adding some more pressure on the game sector in this field with their announcements to also move into the web and alter their price schemes.
Some projects, thanks to VCs gets too much marketing that put main players under their shadows. MeteorJS is the same.
Derby is the closest thing to it but Meteor has more capabilities and has a better development team plus a lot larger developer community
- Nobody knew about Meteor before their HN launch[0]
- Famous appeared at conferences, TechCrunch Disrupt and others with demos of periodic table[1] (which was a ported example from ThreeJS)
- Meteor launched on HN, people loved it, and only then they raised VC and announced it [2]
- Famous announced their funding and sometimes dropped news of various partnerships with "hardware partners" they cannot disclosure. [3]
- Meteor was open-sourced and available with documentation since the launch. They changed their licensing from GPLv2 to MIT later but it all was open[4].
- Famous is open-sourcing their 1-commit repo after 2 years of private beta and now their documentation is in private beta as well.
[0]: https://news.ycombinator.com/item?id=3824908
[1]: http://gigaom.com/2012/11/25/famous/
[2]: https://www.meteor.com/blog/2012/07/25/meteors-new-112-milli...
[3]: http://us5.campaign-archive1.com/?u=4656ba2b0a364690c8530bc1...
[4]: https://www.meteor.com/blog/2012/04/20/mit-license-http-requ...
https://github.com/meteor/meteor/commit/d69c2d1f198ab08a26a2... https://github.com/facebook/react/commit/75897c2dcd1dd3a6ca4... https://github.com/angular/angular.js/commit/c9c176a53b1632c...
We had a reason for doing so. From now on all future development will be out in the open. The docs specifically were supposed to be made available without logging in, but we were short on time. They nicely designed docs will be made available tomorrow. In the meantime, there are guides available in the https://github.com/famous/guides repo and all the source is JSdoc-ed. Check out the code and try making something with it.
disclaimer: I work for famo.us.
Now, where is the documentation? The readme from the repo points to a docs folder that does not exist, and links to the online versions of documentation lead me to login-only pages... registering only puts me in line for access.
No visuals, just a bunch of beta devs talking about how great everything looks..
Demos: http://demo.famo.us/
Codepen: http://www.codepen.io/befamous/
And yes, it works fast on chrome, dog slow on Firefox. Since when did we decide to collectively ignore Firefox users because we want to use some fancy shmancy JS framework so we can write our code using an "opinionated" approach? And I quote "opinionated" as that is precisely how the readme describes the "famou.us" way of building "rich, fast web applications".
Didn't famo.us show off some fancy spinning sphere thing a year or so back? I seem to recall that working wonderfully smoothly on my (Galaxy Nexus) cell phone. The demos all available now are terribly slow and jerky and really don't sell the tech in the slightest.
Flappy bird written in famo.us
https://github.com/Famous/guides
The biggest disappointment is what seems like less-than-rock-solid desktop browser support. Support for iOS and Android only with "Broader support for more browsers is coming"?? (After all the discussion of not releasing until everything is rock solid?) I noticed it with some of the initial demos that didn't allow you to rotate the element chart using your mouse, but it worked fine on an iPad. For all the talk of 3d, I don't see anything on par with a three.js in the source.
Seems like React + a physics engine...
http://m.infoworld.com/t/mobile-development/famous-were-buil...
Mouse support in desktop browsers will improve over time. Designing input to accommodate two fairly different inputs is pretty tricky and this isn't a solved problem anywhere yet. Sites/apps that work in both mobile touch and desktop browser settings currently all use mouse pointer interactions (basically just tap, double tap, swipe) Adding multi-touch interactions complicates things quite a bit. Since we're working from mobile towards desktops.
We're working on better desktop support ourselves and would love contributions from the community that help improve the desktop experience for everything.
How is this more useful than a language/platform rev?
My general point is genericX.js seems to get a lot of up votes here relative to other language/framework topics .
I also totally dig meteor as a huge step-up in how to build web applications. I think they solve a big problem that most people are still working on (react also looks interesting)... not having to duplicate your business logic and rendering code. I think that's a huge benefit to getting things done quickly. Plus you get real-time rendering of data changes in your app, largely for free.
So I wouldn't call these genericX.js libraries. I think they are enabling new opportunities and are worth the interest they're generating.
Facebook Paper: http://demo.famo.us/paper/
Yahoo Weather: http://disrupt.famo.us/1022e61cbff3fe1fc382a81a2d6e2396078a6...
- github repo is now public
- they're letting in the first 75k devs to famo.us community website tonight
The first 75k devs will get 3 invites.
The website contains what they're calling Famo.us University. Tutorials, forums, IRC, etc.
The explanation for letting in 75k devs at a time is to not overwhelm the community with new devs. They want to be responsive to the first batch and train them so the first batch can support other batches. It's pretty brilliant community management and a sane approach to being able to actually deliver on the hype over time.
The framework is opinionated in the sense that there are prescribed ways to build apps for the best results. However, it is at the same time very modular and those willing to work with the core modules in famous/core directly will find the interfaces low-level enough that they could explore alternative ways of building apps that represent their own opinions. If you're the kind of developer that can build with low-level primitives, by all means explore other directions, since we're curious to see what people create and how they use the framework in novel ways. For everyone else who just want to build a performant app with excellent UI interactivity for mobile browsers, we definitely express an opinion. We've been doing this for a while and know quite well what approaches work and which ones don't. So unless you're performance neckbeards like we are, you're better off building the apps according to our opinion (until you know enough to form your own).
disclaimer: I work for famous.
They also have a yeoman generator
EDIT: The "mystery" over a lot of this is just making me less interested at this point. Sorry, marketers.
Yeah it's really starting to like this is more of a marketing company than a technology company, trying to milk some publicity but doing a really bad job of it.
https://chrome.google.com/webstore/detail/beepe-phone/hcfncm...
A lot of progressbars for animation also don't include visual feedback (except a number changing).
Makes it too hard to find what is it really about, how good it really is, and whether it is for you or not.
The link gives a 404.
All their audience are developers who can basically get it off of github.