Oh you're funny. Pulse still sucks. It took forever to get any sound out of my Thinkpad - pulse just doesn't like it. And streaming audio over bluetooth? Forget it. On my MacBook I just have to click a button and it works.

The last Debian update I ran included systemd. I haven't managed to boot it since - I'm stuck in the recovery shell.

I honestly had a look at this discussion to see if Gnome 3 is finally ready to actually use, and has stopped pissing people off. Clearly not. I'll stick to XFCE. It manages windows, provides a usable environment, and doesn't have strong opinions about how I should work. Fancy that?

I work in Magnetic Resonance Imaging (MRI) research. The main machine I use cost around $10M, excluding building and installation costs. I'd love to keep science as a hobby - I could focus on the bits I enjoy, without worrying about all the local politics, etc. - but I simply could not afford the tools. Science has changed since the days of the gentleman scientist. Things are a lot more complicated, as the depth of our understanding slowly increases. Joining a hackerspace is very cool, but it really isn't bleeding edge science.

So the "no beta" link goes right back to the front page. I can have the front page without beta, or read the stories? I'm off somewhere else.

From what I've seen of the copy editing in ebooks, this has already been happening for some time. Or they're just badly checked. I can't wait to see the first science books coming through with random additions in the equations - that will be helpful.

OSX just works. Windows and Linux are both behind, in different ways.I happen to like tinkering, so Linux is attractive to me. I don't really understand why anyone would still choose to use Windows (I am forced to use it, for some critical software I can't get for the other platforms).

Moving back from running only Linux for a few years to running both Linux and Windows, I was genuinely surprised how bad the Windows driver experience is. Windows 7 usually needs a round or two of searching for the driver download (the CD in the box is always some way out of date), installation, remove crapware, reboot, remove more crapware. Using older devices under Windows is often impossible. Usually I just plug stuff into my Linux machine and it works. Being able to type apt-get update && apt-get dist-upgrade is magic. Windows is so far behind here. Funny, when the target audience (less computerate users) would benefit the most.

I think Eton might disagree with you. Maybe you're a geologist or something, but it's been running since 1440 - that's quite a long run. I'm not saying I like it, but it clearly works.

That's great until you have to use a different keyboard layout. Around here (CH) the keyboard may be EN-US, EN-GB, CH-FR, CH-DE or even FR-FR (which is just stupid). Y's, Z's and punctuation are best avoided.

I never intended to say we shouldn't try. What I meant to say is the approach of "let's simulate a whole brain" is a bit of a party trick and not a whole lot of use. I simply don't believe that making huge simulations with a very big computer is the answer. Simulating bits of brain is interesting. The true art in simulating - well, anything - is in the abstraction. Climate modelling is pretty cool, and often wrong. That doesn't mean it is of no use - the 'wrong' results are often as interesting as the right ones. I just have the feeling that the BB project is the equivalent of trying to model the dynamics of every raindrop, and hoping to show global warming. It's the wrong approach to a good question.

It's an interesting idea, but suggests a far greater understanding of how the brain works than we actually have. How can you abstract the important parts, when we have no idea what they are? We're still trying to figure out the many, highly complex biochemical pathways. Maybe that explain why, even though the project employs a full-time science writer, it never seems to publish very much?

The Bodleian Library in Oxford has operated a similar underground system for many years, although I think they use minions rather than robots to find the books. I always found it a rather empty and disappointing experience to be so close to so many books, but to only handle the particular one I had requested.

The new library at EPFL in Switzerland is much better. They have a fancy building above ground, some of which houses books. But most of them are kept in stacks underground, so they're tightly packed but still accessible.

+1 for guided random searches (a.k.a. following the shelf where you found a good one).

It's fast. Not the calculation speed (it's horrible on my old calc), but the speed of typing stuff in. I have an old TI-60 that I've been using since school, and I use it daily. I can hammer out numbers quickly with one hand, while holding a 'scope probe, soldering iron, or whatever with the other. I have a calculator app on my phone (RealCalc) - it's handy when I'm not near a real calculator. But in the time it's taken you to start your calculator emulator, I've been around three or four iterations of capacitance/inductance/resonant frequency calculations.

Maybe he's being blocked because it's not his access point?

It's all nice and dandy that you want a bunch of high-end professional equipment, but what do you actually want to do with your lab? Analogue? Digital? RF? Do you want some mechanical capabilities (drilling boxes, etching/machining PCBs, CNC, 3D printing, etc)? Do you need a microscope for really small stuff?
Rather than getting all excited about the shiny new toys, start with what you want to do. Then figure out what you need/want to help you do this. That's a question we can help with.

I agree that hardware is hard. But radio hams have been building hardware, and sharing designs, for longer than software has existed. A large part of their success is about the mindset. A professional RF engineer will demand a certain set of instruments to make their job possible. A ham will either find a way to make the tests with cheaper equipment, or find a way to build the instrument first (see, for example, the various homebrew network analyser projects). Partly this means relaxing design specs to make a project more likely to work. Partly it means recognising that building one of something is different to building 10,000 - if your project takes time spent at the bench tweaking individual components, well, that's part of the game.

If you come from a professional background, open source hardware looks impossible. But if you add a little more ingenuity, and pick your projects carefully, it's entirely possible and can produce some very impressive projects.