It's a tiling window manager written in Common Lisp, which means I can change any of the behavior on the fly. I guess you can do the same with xmonad, but I don't know Haskell, I do know a bit of Lisp.

Unlike a lot of tiling window managers, it does user-defined "static" layouts. You start with a single frame covering the whole screen. You can split that vertically or horizontally. You can then split either of the new frames vertically or horizontally, and so on. When I used to use i3 and xmonad, I'd get annoyed when my windows kept moving around and resizing whenever I opened/closed a window. Now I have a few different virtual desktops with layouts I find useful. Browsing on one, with a big frame for the browser and some smaller ones on the sides for xterms. Programming has half the screen for emacs, the other half for an xterm or two.

It takes a second or two to start, because let's face it, modern Lisp implementations are slow to launch, but once it's running it's quite responsive. Memory footprint isn't tiny (we're running a full Common Lisp system here), but I'd wager it's still smaller than KDE, GNOME, or Unity.

One thing I really loved was that the default keybindings were good. The default keybindings in WMs like i3 often eat a bunch of bindings for other programs--I had a hell of a time using i3 and emacs at the same time, for instance. Most of the commands are prefixed with a Ctrl-T, so you'd hit "C-t c" to create a new terminal, sort of like doing an Emacs command (C-x C-s). Yes, I know you want Ctrl-T to open a new tab in your browser; to send a Ctrl-T, you just hit it twice (C-t C-t). You can also add keybindings to the running WM using a little snippet of Lisp code, and if you like it, stick that in your .stumpwm file so it works every time. I left most things default but added Alt- to switch between desktops.

The thing about scientific computing is that scientists like to write MPI and Fortran. They just love that shit. And they are traditionally really resistant to any new programming model. So when you tell them they need to start using XYZ instead of MPI so their programs can actually complete at exascale *before* hardware failure, they get unhappy and instead implement things like checkpoint/restore that takes 70% of the runtime. Source: I work in HPC.

I don't know what a Beowolf cluster is either. I am, however, familiar with Beowulf clusters, an overly fancy name for "a bunch of commodity computers running MPI". MegaDroid is not a Beowulf cluster, nor does it run on a Beowulf cluster.

A "virtual android device" is an instance of Android-x86 running in a KVM virtual machine, which takes advantage of hardware virtualization instructions to give better performance. You could take the same software and install it on a desktop PC, assuming Android-x86 includes the appropriate drivers.

Actually, each virtual machine has a VNC session running. And they run on 520 desktops with i7 processors and 12 GB of RAM each.

How many times are you going to post this exact thing in the thread?

At least at my school, it was clearly (and frequently) explained what plagiarism is and exactly what happens if they find you pulling that kind of stuff. Now, maybe if you come in as a 3rd year or something like that you might be able to miss all the talks, but I think any school would be extremely remiss if they neglected to give that sort of orientation to incoming foreign students.

As a personal anecdote, semi-related, my girlfriend is from India. She has a cousin back home who is paying somebody to attend college for her, so she can get a teaching degree. It blew my mind that yes, it's considered acceptable to do this--her whole family knows, and at least the ones over there don't see any problem with it.

Ah, cool. Apparently, they do have a VMWare product, and that's the one that I've ended up using. I hated it so much. If they have an alternative - Awesome. I know EC2 uses a very customized version of Xen.

Seems like VMware to me: http://www.rackspace.com/information/newsroom/pressreleases/rackspace-private-cloud-leverages-vmware-to-extend-enterprise-computing-on-demand/

RS = Rackspace? If so, they use VMWare, because it's ENTERPRISE!

What's wrong with AoE? It's high performance, low-overhead, and simple. We're looking at using KVM, or Xen. Xen PV has much better performance compared to our current ESXi cluster. I think our ESXi cluster is on FC + 1GE + HP Blades. We can get far better performance from Xen by using DAS and replicating it up to a AoE store in case of local drive failure. Drive latencies are 1/10th of what they are on FC, and we're able to get much better IOPs from using local SSDs. Our systems, and TCO is far lower cost by using commodity Dell hardware instead. Last company I was at had something like 500 TB aggregate storage, and 18 racks. We have over 100 Gbit/sec of aggregate bandwidth on our cluster. I'd say probably 25 gbit/sec of that was storage traffic. I'd like to see you try to meet that scale with a traditional FC SAN.

It should actually be easier than Android. MeeGo seems much more of a standard Linux system, and Inferno already supports Linux/arm systems. It's up to you to run Inferno as an app on top of the MeeGo graphical layers, or to strip things down to the kernel/utilities layer and build from there, like we did.

It would be far better to just make a NAND. You can build any logical function out of NAND gates.

I have helped write non-trivial programs in Go. It's quite pleasant. I think the big reason Google hasn't bothered to provide "adequate tooling" is the developers. That is, I think many of the creators and big users prefer to just work with a plain text editor rather than an IDE. There's a reasonably decent Emacs mode for Go, but it's readable enough without any syntax highlighting, and gofmt will fix your indentation and such for you. As for the debugger, well, there's http://blog.golang.org/2010/11/debugging-go-code-status-report.html (very old but even then they already had GDB support going), but as the first line says, "When it comes to debugging, nothing beats a few strategic print statements to inspect variables or a well-placed panic to obtain a stack trace", which is all I ever used to debug my Go code--and found it reasonably painless! Oh, and regarding generics--I'm not sure they'll ever go in. It is proposed CONSTANTLY on the mailing list, and Pike et al always indicate that they're not interested.