The compilation benchmarks are not comparable as the compilers are different, not only in version number but in architecture! OS X ships with llvm-gcc, which is a different compiler from GCC. Think of it a LLVM pretending to be GCC (accepting GCC options, etc) for backward compatibility. This would explain the huge discrepancies between the results of the compilation benchmarks

Disk performance is another thorny issue. The Postmark benchmark shows Ubuntu 12.04 being 3x faster than OS X 10.8 (246 tps vs 80 tps), yet the postgresql database benchmark shows OS X to be 3x faster than Ubuntu. No explanation is even attempted. Why? Readers would like to know! How can OS X be faster at a database benchmark when a raw disk benchmark shows it to be a lot slower than Ubuntu?! Perhaps there's something screwy with the configuration of Postgres on Ubuntu? Does this mean that OS X is *THE* choice for hosting busy databases? My suspicion is that this is due to fsync (http://www.postgresql.org/docs/8.1/static/runtime-config-wal.html). If fsync is enabled, the database waits for the transaction log to be flushed to disk every time a transaction is committed. It's basically down to defaults, and who knows what the default values are for Postgresql on OS X vs Ubuntu?

The graphs raise far too many questions that are not addressed. Many of them should have raised warning flags, like the one about disk performance vs actual database performance. As such, the results are thoroughly suspect and no reasonable conclusions can be drawn. Pity, because they clearly have the kit just not the knowhow.

You're correct that a PhD doesn't guarantee anything. My personal experience of working in the software industry in the UK, after getting a PhD in computer science has been mixed. On one hand, employees still have the stupid mind set of looking for X years commercial experience. It didn't matter that I had spent 4 years writing lots of C++ code for complicated machine learning algorithms, and like most on /. had been programming from a very early age before going to university. It still counted as 0 years commercial experience at a lot of places. I gave up trying to figure that one out. A PhD isn't going to automatically give you a high paying job.

On the other hand, having a PhD can open doors. I've found out that clued up start-up founders are desperately keen on hiring PhDs. This isn't strictly down to the area of your research (though it helps obviously). A PhD says that you've spent years working on problems where the solution isn't well defined (buzz word here is "wicked problem"), you're self motivated (no need for management hand holding), you can work with plans that change, you're not fazed by failure and most importantly you persevere and finish the damn job. Big companies tend to be pretty "Meh" about these traits, but start ups know that these traits are absolutely vital to getting off the ground.

TL;DR version: The PhD may not help you in your career in well established organizations, but it may give you a better shot at working at start ups where the skills you picked up over the course of your PhD are better valued.

... is a dangerous thing! I can just see bosses putting more pressure on coders to "get the job done now!" and then failing to understand why code takes so long to be delivered.

When I was with them 2+ years ago, not only did they shape BitTorrent downloads they also shaped HTTP and streaming video downloads. I require bit torrent when downloading WoW client updates (don't use it for anything else as I don't have the time. See WoW ...). I noticed things speeded up when I disabled the Blizzard Downloader's P2P functionality. I've also noticed them throttling Steam downloads from about 5 - 9 pm, and they throttle video services that compete with their BT Vision package.

Avoid them like the plague.

So true. I remember trying to get to grips with MOO 3, failing hard and thinking there must be something wrong with me because games are supposed to be fun.

Define working :)

GTK+ apps look out of place on Windows, even more so on Mac. In addition to that, Qt just integrates a lot better into the native tool chain (e.g. Visual Studio, Xcode). Prior to being bought out by Nokia, Trolltech were charging $1500 per developer, per platform for Qt. And Trolltech were profitable! It is *that* good a toolkit. It's benefited immensely from being backed commercially and it shows.

Will this continue after Nokia bails? Will the pace of development slow, to the extent that it no longer integrates as well with new tool chains and platforms? That is an unknown and I really hate unknowns ...

Having used both, VCL has nothing on Qt. It also doesn't hurt that Qt is free and cross platform while VCL costs a fortune and has only recently gained OS X support. However now that Nokia appears intent on offloading Qt, I'd worry about Qt's long term future.

I can't comment on the others, but Canonical is most certainly NOT making money. They're burning far more money than what's coming in and given Shuttleworth's attitude (i.e. So what?) it doesn't look like they'll be profitable any time soon.

While profits might be anathema to some in the open source world, a business needs to break even if it's going to have long term survivability.

I would hope he was being sarcastic...

Yes. Personally, I prefer to just log in as the sudo user and then do "sudo su" to become root.

The default configuration of SSH is to only allow you to log in with your private key so you're still going to muck around with to allow password logins or get the root user a public/private key pair.

Do what I do. Reserve a micro instance on a 3 year plan. 1 time fee: $100 Monthly fee: $3.66 * 36 Total: $231.76 Averaged over 36 months: $6.44 That's slightly more expensive than some shared hosts, but the upshot is that you full access to a VM that's under your control to install whatever you please. I've got a blog, VPN and photo gallery and once you factor in bandwidth and Amazon S3 storage costs it's still under $7 a month.

The default Amazon EC2 linux instance exposes only SSH and logs you in as a user with sudo privileges.

If your instance has a significant amount of idle time, you're likely better off with an on-demand instance and powering it off when it's not in use.

From the linked document:

2003 just called. They want their Gentoo Ricers back.

Problem with that analogy is that you die within days if you don't have water, while even the most hardcore torrent buff will survive without access to torrents.