The doctors can do it on-demand now (well more or less).

So say you have incredible pain due to some tumor they can target the pituitary gland with a gamma knife and after a few weeks the pain is relieved so that you don't need to many drugs to avoid pain.
Apparently this has been done since the 50s but before the gamma knife the side effects were pretty big. There does not seem to be any side effects with the gamma knife.

It's really great for people who benefit from it although apparently nobody really knows why this works.

I disagree with AT&T. I want maps, precise maps for 3G. I mean, I have EDGE on my Blackberry and it's very bad. So if I get a 3G service, I really want to be able to make sure that I won't have to fall back to EDGE...

I saw the ad and it really made me think it was cool and informative and I wanted to go online and check this out next time I shop for service. Unfortunately I expect the maps to be deceptive. T-Mobile's maps show reception in areas where I have been and where I would get cut all the time. So the maps are not really very good and can be tweaked.

I guess we need to do our own coverage maps with some Android application that checks coverage, records GPS information and aggregates that on a website. Independent maps are the only way to get anything remotely accurate.

How do you handle the lack of dimming? I have a bunch of sockets with a dimmer and am afraid to put a CFL in there. Can I disable the dimmer easily? Or is it OK to just push the button all the way every time?

    Thanks a lot,

You know, it's just one trade-off and it might not be much worse than linking C++ dynamically... Or Jitting Java byte code.

I have owned an Agenda VR3 and a Zaurus. I can tell you that there are many ways to wast CPU cycles and that in any case it takes effort to avoid them. The Agenda guys spent a lot of time trying to optimize the binaries and resolve issues caused by C++. They might have had an easier time coding in Python and avoiding C++. Or not. It's a complex issue, so it's hard to say even now after the dust has settled.

I don't think there is any silver bullet nor that Javascript, Dalvik or C++ are bad per se. What matters is the effort that the makers put into integrating the software and the hardware, testing, optimizing etc.

The Pre uses the V8 Javascript engine and it might be more efficient to jit a few tens of lines of Javascript than to link a C++ binary with all kinds of libraries (say stdc++, qt, X11 etc.).

Also, I remember that programming in GFA Basic and running the programs on my 8MHz Atari was quite fast actually. So why would it be such a mistake on Palm's part to use Javascript on a machine that is about 500 times faster (the fastest instruction took 4 clock cycles on the 68k)?

I think the Pre is really interesting as almost everything in it is Open Source (V8, Webkit, Linux etc.), based on standards (HTML, Javascript). Also people seem to be able to compare it with the iPhone without laughing so it must be a rather good integration of hardware and software too...

Anyway just my 2 cents and an occasion to fondly remember the gadgets of my youth ;-)

I am not that sure actually. It's not very well defined and different people use it differently, sometimes with a marketing agenda.
But it also conveys some property quite clearly:
  - cloud computing is not precisely located and you don't really care
  - it's not happening in your home
  - it's everywhere or almost
  - it's out of your control (others may access it without your knowledge etc.)
  - it can disappear and be unavailable anytime (just like real clouds ;-)

The previous terms were not bad either, but the market made them more precise. For instance, there was Grid computing: it's always a cluster inside a company that provides centralized computing power for embarassingly parallel problems. The original idea was that it would be like the electrical grid and you would just send your problems to the computing grid and they would go wherever there was an excess supply of computing power. Very cloud-like actually...

I don't think the term is useless and I actually think it's nice to change the buzzwords every so often...

It looks like there are Ruby books on scribd.com actually. It's probably one way to do what you just said.
And now I am waiting for other people to comment and list all the other ways that I did not know about ;-)

No need to have a working network connection either.
Here in New York City, cell phones seldom work in the subway, access is spotty on trains etc.
So this is a definite plus.

It is also supposed to work 12 months with the AAA batteries. That certainly beats any of my gadgets...

As a side note, I was reading the SkyMall catalog 2 days ago and they had dictionary-type devices for about 100$ to 300$. So this device is not badly priced either.

Finally, since it's from OpenMoko, it's probably quite hackable...

There are people running and developing Hurd:
'The GNU Hurd is under active development. Because of that, there is no stable version'
It's only the FSF that said that Linux temporary. Also, temporary solutions tend to last...

Hi,

how can it be a 1000$ cheaper? The Iphone's ATT plan is about 1000$ a year I believe (I don't have one so I never saw the bill...).
Could you elaborate?

    Thanks

Engadget has a comparison of all the Android devices to date and they all are very similar at the moment. I guess it's good news to some extent as it means the apps don't have to cope with different screen sizes or even different CPUs.
But at the same time, why bother with Dalvik if they were going to all target the same Qualcomm/ARM chip at the same frequency? They could have just used gcj and compiled to ARM...

Hopefully this is going to change and the manufacturers are going to come up with exciting Android handsets. I mean, good enough to get Nokia to lower the price of the N900 would be great!

This duplicates the iTunes syncing functionality. Sorry. Just removing a wire does not justify duplicating functionality in the Experience(TM).
Oh well, yes I have karma to burn.

Actually, you should have stuck with hire a lawyer. Apparently knowing about the patents you infringe upon makes your situation worse. So apparently you are better off not looking at existing patents.
I think this should really not exist as the main argument for patents is that they are a way to share knowledge, so anything that discourages reading them is counterproductive.

Maybe the older ones are "dead"