***masoudd_ is now known as masoudd
<paroneayea>Honestly, that's why I'm just not really interested in that part of the software stack. I'm more interested in the software that runs on top of base Android, augmenting the user experience. Basically the level that Google apps run at. Focusing on that will do a lot to get us to a state where we can actually compete in the market with Google. Because right now the options are either "live without a billion nice features" or "install <paroneayea>Google apps and give up all the things that make freedom-respecting software useful." <paroneayea>the other thing was part of someone's blogpost about android. <atheia>Ah accidental blogposts! It's why I stopped using IRC in Emacs. <atheia>Way too easy to post in the wrong buffer :-/ <atheia>Ugh, and when I wrote blogposts I meant pastes. That's a weird thing to mix up. <civodul>it's great that the FSF talks about Guile :-) <holomorph>ACTION wonders what the other 14 letters are in e14n <paroneayea>I think it's meant to look like it spells "evan" if you see the logo <paroneayea>I don't know hat "exclusive nation" would mean, but I guess it doesn't sound like a term you'd use to describe a federated system <paroneayea>distributed scheme by Jonathan Rees and Richard Kelsey <amz3>fsf is interested by guile, weird <davexunit>amz3: I used to work there and got everyone hooked on it, of course ;) <paroneayea>wingo: did you ever look at Kali scheme? seems pretty interesting <wingo>paroneayea: i just checked that link <wingo>i suspect that a distributed system should build on a concurrent system <wingo>(and that a distributed system is a harder problem!) <wingo>i also want to understand what lasp is <wingo>i don't know whether replicating erlang is really where we need to be; but i don't know much :) <OrangeShark>amz3: the other day I committed some stuff for guile-git. I have added finalizers to the repository pointers and it should be something we apply to the rest of the types. <OrangeShark>amz3: I was wondering if we should somehow just roll the finalizers into the pointer->type wrappers when defining new types <OrangeShark>so it would be (define-libgit2-type type finalizer-for-type) <wingo>why do we have cancel-thread. <wingo>why does it use pthread_cancel(). <wingo>it should just queue an asynchronous interrupt on that thread and throw a guile exception (or prompt abort) from the interrupt <wingo>pthread_cancel seems exceedingly dangerous <wingo>civodul: wow that is lovely :) <wingo>never thought about redefining arithmetic operators to trace an operation <civodul>i'm seeing if i can package it to give it a try <wingo>i think i will deprecate cancel-thread or something like that