Follow Slashdot blog updates by subscribing to our blog RSS feed

 



Forgot your password?
typodupeerror

Comment Re:Fake (Score 2) 206

You want your mirror large in order to resolve small angles and small objects. The smallest angle you can resolve is lambda/D where lambda is the wavelength of light you are using (400nm for near-UV blue) and D is the diameter of your mirror or lens.

So suppose you have satellite in 100km orbit around the moon with a 2.4meter aperture (like Hubble) using 400nm light. Then the smallest angle you can resolve is 0.034 arcseconds and you cannot resolve features smaller than 16mm. If you use red light (600nm) then you cannot resolve features smaller than an inch.

Comment Re:But...but... (Score 3, Insightful) 206

You are oversimplifying things. Yes, slow moving charged particles (such as electrons or Helium nuclei) can be easily shielded. However, fast moving particles are much harder to shield against as they create showers of new particles (of lower energy) upon collision.

The spectrum of these particles extends way up - scientists are busily observing particles with energies on EeV scale (roughly what a moving golfball has), though these are quite rare.

Neutral particles, like gamma rays, can only be shielded by a bulk material - the penetration depth depends on density.

Lastly, we have we have direct visual observation of cosmic rays by astronauts on Apollo missions and ISS.

In summary - being in space is kinda like being on a battlefield - if your general did not screw up the chance of being hit by an artillery shell is quite small. But this does not mean it cannot happen.

Comment Re:The problem is Ballmer (Score 1) 407

I am glad you liked it.

One other way to look at this is that if you try to maximize some function describing performance this decreases the uncertainty in function value (as the first derivative is 0) at the cost of increasing uncertainty in function parameters (i.e. everything else).

And unlike share price risk is good deal harder to quantify..

Comment Re:Bot! (Score 4, Insightful) 85

Pretty sure the article was auto-generated by a buzzwordifier:

Panguite (IMA 2010-057), (Ti4+,Sc,Al,Mg,Zr,Ca)1.8O3, is a new titania, occurring as fine-grained crystals with Ti-rich davisite in an ultra-refractory inclusion within an amoeboid olivine inclusion from the Allende CV3 carbonaceous chondrite.

Doesn't mainstream PC tech use the least abusive field-related babble when compared to medicine and legalese?

For once we have a line of scientific discussion and you are complaining ?

Comment Re:One of those little skills (Score 1) 321

There is a trick for soldering fine pitch cases like MSOP or TQFP: just soak it all nice in solder without worrying too much about bridges. You want to make sure that enough solder gets between the pins and pads. Then use copper braid to soak all the extra solder off. In a way, you are just giving your part a localized solder bath. Make sure to use large tip for your iron - at least a few mm or a knife.

My god! Why did I never think of that?

Heh :) I don't know who invented this first. I found this trick in a manual for Chinese workers after spending a few hours reading all about making reliable SMT connections. Most manuals describe "ball of solder" technique where you use a small iron tip to carry a ball of solder to each pin and then make sure that pin and pad wick the solder in. Rather tedious, and I had to retouch connections often.

Comment Re:One of those little skills (Score 4, Informative) 321

There is a trick for soldering fine pitch cases like MSOP or TQFP: just soak it all nice in solder without worrying too much about bridges. You want to make sure that enough solder gets between the pins and pads. Then use copper braid to soak all the extra solder off. In a way, you are just giving your part a localized solder bath. Make sure to use large tip for your iron - at least a few mm or a knife.

Comment Re:thoughtful recs that all require more NIH fundi (Score 2) 226

While this is also true, the current system is completely unsustainable unless the funding basically increases exponentially, which is never going to happen. The problem is that for each faculty (each lab), you typically have ~4 postdocs and ~4 PhD students at a time... so after 5 years, you've gone from needing 1 faculty position to 5. If they each get jobs, after another 5 years you're up to 25 positions... unless funding (and, equally as importantly, university positions/space) is going to increase exponentially, it eventually falls apart.

It's exactly the same training problem as other fields (law, medicine) in that you're constantly training more people than there are current positions... except that in those fields if you really can't find a position, you can go open your own practice. In biomedicine, that's nearly impossible - any serious research lab is going to require a significant amount of funding and resources that you basically can't get outside the university/grant system, and it's very difficult to do a biomedical startup without having a prototype already existing (since it's biology, and the failure rate is high simply because we don't understand enough about most systems yet to know what will work and what won't without actually testing it).

There is a flaw in your argument - the population of United States is growing much more slowly. So at some point everyone will be trained. Wouldn't that be nice ?

Comment Re:Geiger (Score 3, Informative) 133

Why don't they call a 'radiation detector' by its name? It's a Geiger Counter. Way to make a name for something fall out of common usage...

There is not much description in the article, but I don't think it is a Geiger tube, as that requires high voltages and is fairly bulky. This is probably some sort of silicon detector.

Slashdot Top Deals

Q: How many IBM CPU's does it take to execute a job? A: Four; three to hold it down, and one to rip its head off.

Working...