On the contrary.

Especially BECAUSE NASA should focus its resources on those visionary missions does it make sense to shift to commercial partners for the initial launch part. Why? Because launching by itself is not all that interesting. There will never be a vehicle that will launch from Earth and fly to Mars in one go. The only sensible solution for manned deep-space missions is to develop pure space-vehicles, that never touch own on a solid surface. If NASA can simply purchase seats to LEO on a commercial launcher for its astronauts, this frees up a tremendous amount of resources it now spends on launchers to work on those deep-space vehicles.

Of course, this assumes there will be something to replace the moon-landing portion of Constellation. Since NASA is actually expected to get an increased budget in the next few years, despite a general budget-freeze in many departments, I have some hopes for this. They have to spend the money on something, after all.

I'm sure proportional fonts can be read faster than monospace fonts, but this is absolutely and completely irrelevant. In no way, shape or form is reading speed something that slows down developers.

The reason it's not widely used yet, and why it will not be widely used by individuals (a requirement for ENUM to become useful, IMHO), is because when registering an ENUM number costs EUR 45 signup fee + EUR 1.79/month, according to the website of one of the registrars that is currently contracted to sell ENUM numbers in the Netherlands.

I this on Facebook, and I can NOT confirm these claims. Nothing prevents me from setting the Search result privacy setting to "everyone" while setting my "About Me" privacy setting to "Friends Only", for example

Printers and scanners are excellent sources of scavenged parts for robot projects. They contain motors, stepper-motors, pulleys, belts, gears and all sorts of other useful bits.

The couple of times I looked into Gears, the main feature touted by Google was the ability to use your web apps when you're not connected to the internet. This was reason enough for me not to spend a lot of energy on Gears, as in practice, in this day and age, I never find a computer that is NOT connected to the internet.

So in short, I've never had a need for Gears.

Good point.

Even worse, contacts do not remain at the same spot on the retina. They move about a little bit with each eye-movement and blink. This is ok for a simple lens as long as the actual pupil remains fully covered, but for a screen it would be catastrophic. Imagine your monitor slamming down when you blink, and then slowly work its way back up (which is what a contact does).

I'm glad I'm not the only one wondering about this

Can anyone shed some light on how the optics of a contact-lens display would work? After all, when all is said and done, this is going to be a display that is not simply "close" to your eye, but ON TOP of it, and I don't know about others, but my eyes are unable to focus on anything closer than 5 cm (2") or so.

There are mirror/lens systems in VR-helmets and those fancy spectacle-like, wearable displays that create a virtual display some distance away from the viewer, but I don't see how that could be replicated in a contact lens.

It's slow, but bearable, and unlike your EPIA 5000 it's i686.

Yeah, the CPU is the reason I want to switch to a Jetway or Intel Atom board. The 5000 suffices for what it currently does, but _everything_ is CPU-bound. Now that I'm thinking of also running Asterisk on my home-server, it's really becoming a problem.

Still, if low power-usage is your overriding consideration, and you can live with some limitations on what you can run, the EPIA 5000 is a winner.

Second that. My home server runs FreeNAS on an EPIA 5000. Including a gigabit ethernet card, 4-port SATA card and four 1 TB drives, this system draws about 35 watts. When the drives spin down, power usage drops to
One downside is that the EPIA 5000 is too light-weight to do software RAID (even JBOD), which I found out the hard way (by losing data!), so I am now running the HD's as plain, separate partitions.

This is very much like the FEI Phenom microscope, which my company wrote the software for. The Phenom has been on the market for several years already. It's cheaper too, if I'm not mistaken.

It's of course a myth that products today are specifically engineered to fail right after the warranty expires. In this particular case, the more likely explanation is that the Russians simply accept a larger risk-factor than the other partners. Due to the politics involved, organizations like NASA and ESA are simply not allowed to fail, and so they will rather scuttle a module than squeeze the longest life possible out of it.

That, and the fact that NASA and ESA simply don't have the funds available to continue operating the ISS after 2015 or so.

This would be interesting for the treatment of tumors as well. A number of low-power X-ray laser beams converging on a tumor would create a very small, and precisely controllable, spot of high-intensity x-ray radiation.

While this is the first "hard" X-Ray laser, there have been lasers producing (softer) x-rays far longer.

When I was in college studying physics, we went on a field-trip to the FOM institute near Utrecht, the Netherlands. Even back then, they had a FEL operational that they told could produce coherent light in a large range of wavelengths. If I remember correctly, this range extended into the soft x-ray range.