I like this. I'd like to see more of Plan 9 in general use. Since I started reading about it I've thought it would be cool if a big company did for it roughly what Apple did for Unix with OS X, preferably in a more open-source manner.
This is great. I've been wanting my phone to be a node on my grid for a long time. I'm glad some people got motivated and actually hacked it all together. The dream of ubiquitous 9p/styx lives!
Here's a short video I threw together demonstrating the phone. It's pretty spur of the moment, and the video quality is low, but you may be able to get an idea of how it works: http://www.youtube.com/watch?v=dF_-jQc53jw
Sorry I'm late getting back on this. Since I did a lot of the low-level work getting Inferno running, I suppose I have as much knowledge about it as anyone ;)
For your first question, that depends entirely on whether or not the Android command-line tools allow you to set up an ad-hoc wireless network. If so, then yes. If not, then no.
As to the second question... I have never tried it, but you should at least be able to import /dev/draw from another device and do stuff with that.
Why on earth didn't anyone seriously think about Inferno for mobile devices? How much polish could've gone into Inferno given all the work on things like Maemo, Meego and Android...
Many people did, there are even some other early efforts, I am aware of some efforts that never went public.
Thing is, even though many people thought about this, very few actually worked on it. With open source talk/work ratio is around 50 (just made that number up).
Wow! Can it run acme? I'm wondering because "treats the touchscreen like a one-button mouse" If so, how is the right and middle click used? Anyway utterly cool stuff!
I don't know how well it can run acme, but the acme in plan9port works with touch gestures quite well. Maybe the same gestures could be implemented in Inferno as well.
Hi, I worked on this project, so let me jump in on this. It can run acme, but at this time we don't have a way to do right and middle clicks. I didn't make it a priority, because the accuracy you can get in selecting text on these capacitive screens is just not very good--I don't think Acme would be a very pleasant experience. It worked fine on my old iPaq, but that had a resistive screen and a stylus.
Indeed. I was just thinking I could try it on the device my neighbor has: A very cheap android 2.2 tablet with a resistive screen and pen, no multitouch. Never mind, it can hardly run android apps.
The [Octopus project](http://lsub.org/ls/octopus.html) has an interesting take on gestures. When right-clicking, it pops up a circular menu where the direction one goes in determines the action. See http://lockerz.com/s/139588614 That might be something for touch tablets.
I daily use AcmeSAC on Mac & Win7 with a trackpad (and MagicPrefs for three finger middle click) and with a Wacom tablet and pen. Very nice for editing without RSI problems.
Well, if you can get accurate-enough touch events, you could always go into emu/Android/screen.c and hack around with the pointer event handling code, make it check the state of some of the physical buttons--thus, if you're holding (say) the Home button when you click, it does a left click. Since you have a pen, you should be able to hit the regular window control widgets to minimize/close, rather than relying on our physical button shortcuts.
I used to be a huge Inferno fan, but with Go around (Rob and Ken both worked on Inferno, Rob was even responsible for the name), I wonder how relevant it is this days.
Go has a much more vibrant community and development is blindly fast. Given Go's dramatic memory efficiency advantages over Java I would hope Google will push it on Andriod soon.
Go code can already run on Android, but building Android apps in Go is another matter, but with Brad Fitzpatrick moving from the Android to the Go team recently, I'm optimistic.
I LOVE so much that #b2g and inferno are using the nice Linux Android base to quickly bootstrap new mobile platforms. Get the benefit of a large existing install base, choice of devices, etc.
I'll be interested to see if the other Plan9 inspired projects find a home in a similar circumstance. There's a surprising number of tools being fostered by Google and Googlers (in their 20% time) that harken back to P9.
