While the salon article was informative, it is helplessly single-dimensional. University CS research groups release many of their code, and many are easily available online:
The thing is that it's not an insignificant effort to put software into an acceptably releasable state, and perhaps the key is to find a way to motivate developers to spend more time doing this.
I even know firsthand that many research groups even go as far as giving free tutorials and support for the software that they develop. And this is done by faculty members and students who are hopelessly overwhelmed by tenure requirements, doctorate qualifying exams, passing classes, writing papers, solving research problems, and trying to graduate.
So before you diss this group of people, don't forget to TALK TO THEM first. Many students and profs are more than happy to share their work. At least it makes us feel a little less useless.
I'm surprised that nobody has mentioned that it is already possible (and on top of that, for free) to run Linux on top of an installed Windows OS.
Check out Cygwin, which started as a Bash shell for windows, and now has advanced API's, etc. It even comes with X11, GCC, SSH, SSHD, Perl, etc. Now I have Win2k on my desktop, running SSHD so that I can login remotely and run scientific simulation jobs, access my files, etc. From here you can even run a CVS server.
In my opinion this is the way to go: have Windows and MS take care of hardware compatibility, and have Linux focus on software and usability issues. No matter how much I hate to admit it, having an underlying OS with good hardware support is bliss. Most OEMs ship their PCs with Windows installed anyway. With Cygwin, all you need to do is install Cygwin.
Sure, you can argue that it's not really Linux since it depends on Cygwin's API and everything, but with more support from the free software community this can make a very nice system that is very easy to install (since you don't have to fight to get your kernel to support your brand new graphics card), and has good UNIX-like features.
Oh, and they're available for free, with GPL and everything.
Re:They must be real.....
on
Apple PDA?
·
· Score: 1
...only Apple and movie sites which movie has carefully inserted Apple products. i'm still trying to spot the strategically placed Apple products on LOTR...
to many of us in academia, holiday season means that we're free from doing work for other people, and finally having some time to do work that we want to do but haven't had the chance to do yet. finally... free from classes so that i can do research!
there already ARE some ways to use the power outlet as a way to do surveillance. more details are available here: Nonintrusive Appliance Load Monitoring.
This is an excellent point, but you should realize
that the big companies have put lots of money
and resources into the research, development,
and standardization of 3G. Until they get that
money back, they will continue to push 3G, unless
there is an obvious point of disavowing any
prior involvement. And there is not such point
in the horizon yet.
It's the same as in 802.11b. Now that they've
spent all the effort to R&D & standardize it,
they'll try to take in as much profit before
rolling out the next generation technology.
This is a good example of big corporations betting
on creating a new technology that will create a
demand for itself. When 2G was coming out, the
stench of money was not nearly as strong. And now
for 3G it's almost overwhelming. Everybody jumps
into the bandwagon thinking that if they didn't,
they'd be dead in the water.
At this point it's a good idea to reflect on the
lesson we should learn from the European 3G
spectrum auction 1-2 years ago. The first market
to be opened was sold for a ridiculous price,
and the later markets had to cancel the auctions
because hardly anybody was bidding anymore! It's
also interesting to note that people don't talk
about 3G anymore in conferences and symposiums,
they talk about 4G now. Maybe after all it's a
good idea to learn the lessons from 3G and move on
to the next thing.
Several readers have brought up a good point of
the uncertainty of market demand for 3G services.
While there are attractive applications (think
wireless porn;-) it's still not clear whether
people would be willing to pay the price for the
services.
The last point to mention is the difference in
the "gadget model" between the Asian consumers
and American consumers. The Asian consumers warm
up more to small, specialized gadgets, which can
then be enriched with additional features. OTOH,
the American consumers tend to start from the most
powerful laptop they can lug on their SUVs and
try to add mobility to the features that they
already have. Perhaps after all 3G is the merging
point of the two approaches?
I don't think the statement that the state-of-the-art hasn't advanced in the past 3 years is very inaccurate and cynical. You should separate the curriculum from the real research and development that's going on in EECS. I agree that the curriculum has to be revamped, and schools like MIT, Berkeley and Rice are reworking their curriculum. The difficulty here is integrating the newer ideas into the curriculum while maintaining a strong foundation of learning.
There are more applications of Smartdust at http://basics.eecs.berkeley.edu/sensorwebs.
The most salient feature is that the small size and capability for independent operations allow Smartdust motes to be used just about anywhere, to enhance/complement existing tools.
Slashdot Mirror
While the salon article was informative, it is helplessly single-dimensional. University CS research groups release many of their code, and many are easily available online:
http://iceberg.cs.berkeley.edu/
http://tinyos.millennium.berkeley.edu/
http://nms.lcs.mit.edu/
and many more.
The thing is that it's not an insignificant effort to put software into an acceptably releasable state, and perhaps the key is to find a way to motivate developers to spend more time doing this.
I even know firsthand that many research groups even go as far as giving free tutorials and support for the software that they develop. And this is done by faculty members and students who are hopelessly overwhelmed by tenure requirements, doctorate qualifying exams, passing classes, writing papers, solving research problems, and trying to graduate.
So before you diss this group of people, don't forget to TALK TO THEM first. Many students and profs are more than happy to share their work. At least it makes us feel a little less useless.
I'm surprised that nobody has mentioned that it is already possible (and on top of that, for free) to run Linux on top of an installed Windows OS.
Check out Cygwin, which started as a Bash shell for windows, and now has advanced API's, etc. It even comes with X11, GCC, SSH, SSHD, Perl, etc. Now I have Win2k on my desktop, running SSHD so that I can login remotely and run scientific simulation jobs, access my files, etc. From here you can even run a CVS server.
In my opinion this is the way to go: have Windows and MS take care of hardware compatibility, and have Linux focus on software and usability issues. No matter how much I hate to admit it, having an underlying OS with good hardware support is bliss. Most OEMs ship their PCs with Windows installed anyway. With Cygwin, all you need to do is install Cygwin.
Sure, you can argue that it's not really Linux since it depends on Cygwin's API and everything, but with more support from the free software community this can make a very nice system that is very easy to install (since you don't have to fight to get your kernel to support your brand new graphics card), and has good UNIX-like features.
Oh, and they're available for free, with GPL and everything.
...only Apple and movie sites which movie has carefully inserted Apple products. i'm still trying to spot the strategically placed Apple products on LOTR...
to many of us in academia, holiday season means that we're free from doing work for other people, and finally having some time to do work that we want to do but haven't had the chance to do yet. finally... free from classes so that i can do research!
there already ARE some ways to use the power outlet as a way to do surveillance. more details are available here: Nonintrusive Appliance Load Monitoring.
This is an excellent point, but you should realize
that the big companies have put lots of money
and resources into the research, development,
and standardization of 3G. Until they get that
money back, they will continue to push 3G, unless
there is an obvious point of disavowing any
prior involvement. And there is not such point
in the horizon yet.
It's the same as in 802.11b. Now that they've
spent all the effort to R&D & standardize it,
they'll try to take in as much profit before
rolling out the next generation technology.
This is a good example of big corporations betting
;-) it's still not clear whether
on creating a new technology that will create a
demand for itself. When 2G was coming out, the
stench of money was not nearly as strong. And now
for 3G it's almost overwhelming. Everybody jumps
into the bandwagon thinking that if they didn't,
they'd be dead in the water.
At this point it's a good idea to reflect on the
lesson we should learn from the European 3G
spectrum auction 1-2 years ago. The first market
to be opened was sold for a ridiculous price,
and the later markets had to cancel the auctions
because hardly anybody was bidding anymore! It's
also interesting to note that people don't talk
about 3G anymore in conferences and symposiums,
they talk about 4G now. Maybe after all it's a
good idea to learn the lessons from 3G and move on
to the next thing.
Several readers have brought up a good point of
the uncertainty of market demand for 3G services.
While there are attractive applications (think
wireless porn
people would be willing to pay the price for the
services.
The last point to mention is the difference in
the "gadget model" between the Asian consumers
and American consumers. The Asian consumers warm
up more to small, specialized gadgets, which can
then be enriched with additional features. OTOH,
the American consumers tend to start from the most
powerful laptop they can lug on their SUVs and
try to add mobility to the features that they
already have. Perhaps after all 3G is the merging
point of the two approaches?
i can already know all their deepest secrets. why waste time getting superficial information in person? ;-)
I don't think the statement that the state-of-the-art hasn't advanced in the past 3 years is very inaccurate and cynical. You should separate the curriculum from the real research and development that's going on in EECS. I agree that the curriculum has to be revamped, and schools like MIT, Berkeley and Rice are reworking their curriculum. The difficulty here is integrating the newer ideas into the curriculum while maintaining a strong foundation of learning.
There are more applications of Smartdust at http://basics.eecs.berkeley.edu/sensorwebs.
The most salient feature is that the small size and capability for independent operations allow Smartdust motes to be used just about anywhere, to enhance/complement existing tools.
How is entropy the dual to the Gaussian distribution? Entropy is defined for any probability distribution, not just Gaussian.