for now

Well, presumably most communication is two way or actionable. If not, then there is no reason for the communication in the first place.

If you think Kore is interesting, then also check out Tntnet <http://www.tntnet.org>. I've checked it out a few years ago and it felt good - stable, complete, easy to use etc.

In Python you can: https://docs.python.org/2/library/stdtypes.html#truth-value-... https://docs.python.org/2/library/functions.html#bool

I really wish it worked on Google App Engine. It would be cool to have abstraction layer, so there's no vendor lock-in with GAE.

Not trying to be awkward or anything, but why would you use GAE without using all the technologies that lock you in to it?

If you write a "vanilla" application which can run on GAE and other platforms without significant work, then I don't see how you'd get much benefit from GAE itself, but I might be missing something.

Why doesn't it?

+1

How is it different from ASE (http://code.google.com/p/android-scripting/)?

it looks like this is (going to be) for making self-contained android applications, packaged, and distributed through the market (possibly?). ASE as it stands requires ASE for your code to run, plus it has to download the Python interpreter separately. This project (a part of a multitouch multimedia framework project) is aiming to do this all a bit different and it seems embed the entire interpreter. i'll be excited for it!

ASE is a basically a scripting environment to automate tasks using python (or LUA, etc.) by running scripts.

You can write scripts and then launch them using the ASE.

This fails in a few ways, but the big ones are:

- You need the ASE installed to run your scripts. - You can't make an 'app' with it, because there's no ui.

This lets you build a stand alone .apk using python including a user interface (well, opengl surface anyway); very, very neat.

ASE (is now SL4A) has a Python interpreter module, but also has support for other languages like Beanshell, Lua, perl and Javascript. This project may be a tighter integration.

Choice is good.

What about performance? One of the reasons to use namedtuple is performance, esp. memory-wise (afaiu nametuple stores field names in class definition, which can save a lot of space compared to lists of dicts). I would expect recordtype to be better than dict but not as good as namedtuple. It would be nice if author posted some benchmarks…

It reminds me of Zope 2

but anonymous is not simply one group, it's an idea... it's very likely there's another independent group of people that orchestrated attacks against sony. they too have the right to call themselves as anonymous, as long as they are.

it's likely anonymous and sony where right first denying and then implicating anonymous' involvement in hacks.

It's possible that the one or more people behind this are a splinter group of Anonymous, who have been involved but in more secrecy for this. It's possible they're completely seperate but felt a connection to Anonymous and so decided to class themselves the same way. Or it's possible the hackers don't give a damn about Anonymous, but thought it was a great way to get LEOs looking into people who are known to be affiliated with Anonymous, and/or looking away from the actual hackers. If I were to steal credit card data in this way right now, I'd sure as hell pin the blame on Anonymous if I thought I could.