For the non-physicists of us wandering around Imperial, any chance we could get to see some of the cool experiments? Just for kicks and giggles quite frankly (*far* too down the CS route to considering switching allegiance), but might it be possible anyway?
Computer Science ain't dead yet (although it may smell that way at times), you just need to have a degree program that's worthwhile. Crappy places churning out more idiots hoping to make a fast quid tend to die off at these times, but the better ones survive.
Only if we also get a button that says "your idea is heavy on PR fluff, contains zero real information and it was in fact a waste of my time to RTFA. Therefore, I would like one of your dollars to compensate for my wasted time"
Given VIA's history of "rapid with press releases, incredibly slow with actual boards" (NanoITX anyone?), any bets on how long before we see available boards? I'm betting mid-2007 personally...
802.15.4 is the PHY, Packet and a basic MAC layer. ZigBee is a routing layer on top of 802.15.4.
Right now, TinyOS has limited 802.15.4 support (certainly for the CC2420 radio that the MicaZ uses), but no-one AFAIK has implemented ZigBee support yet. I'm thinking that a more sensible option is to use 802.15.4 to actually do packet level transmission, but then use another routing layer (shameless plug) to do node-to-node transmission. How to do this is to have 802.15.4 nodes from RandomCompany doing the actual sensing in combination with something like the MicaZ - the RandomCompany nodes just see a 802.15.4 network, and the MicaZ nodes can then encapsulate the data from the sensors into a TinyOS packet which can then be routed around as per normal. This sort of heterogeneous network (specialist purpose 802.15.4 sensor nodes w/o TinyOS + general purpose computing/routing nodes w/ TinyOS) looks like one of the interesting directions that things are going.
ZigBee is designed towards the home automation market, *not* wireless sensors. I could name half a dozen more stable and power efficient routing algorithms for WSNs (Wireless Sensor Networks) than this beacon crap they came up with. 802.15.4 on the other hand, is being grabbed onto thoroughly, and that's because the advantages of having a stable PHY and Packet layer for WSNs (as opposed to the current situation where various nodes even from the same people can't communicate) is enormous. The combination of a general purpose computing node + 802.15.4 hardware is a damn good idea, and one that's liable to survive for at least a while. I've spent the last few weeks messing around with a TinyOS node with 802.15.4 and porting our MAC layer onto it, and it's been a lot easier than most platforms.
(On this note, don't buy the MaxStream 802.15.4 chips, because they're non-conformant - got an official line on that from one of their engineers. They're building a 802.15.4-like proprietary protocol.)
So? They still have a problematic human-rights record, and it's getting *worse* not better. Most of the major injustices (Guantanamo Bay for example) didn't exist 5 years ago, and so far every new break of international law isn't getting fixed later on. The biggest problem is that the USA is getting away with a lot of this stuff, mainly because it's the 400 pound gorilla right now.
Replacing the USA's role in the internet with an impartial UN-backed body would also avoid the problem of global choices being made here according to the opinions of a vocal minority of US citizens, as opposed to by the bloody committee that's meant to decide these matters.
Quoting the article: Turning the Internet over to countries with problematic human-rights records, muted free-speech laws, and questionable taxation practices
He was speaking to El Reg after an article on Reuters quoted him as saying that anyone who patented software would be prevented from using free software. Greve says this is not quite what he was getting at:
"The basic idea is that if someone uses software patents against a Free Software program under the GPL, he might lose the right to distribute that particular software, to use it for their products. We have no interest in restricting the way people can use and develop software."
Yes, but can they do it repeatedly in the same way to lots of different people. Betting that no matter how much you pay them, they won't be able to hit everyone *exactly* the same way...
Given that the new Administrator is a politicial nominee, and given that he's the one hiring/firing here, people being shifted by a politicial nominee might well have political reasons for their change in state.
Of course, this might not be the case, but given the current US administration's track record of replacing good people with duckspeakers, it's a good idea to check their backgrounds.
AFAIK, a lot of this gets done by a variety of American companies, who are quite happy providing and customizing their filtering software for anyone willing to pay up. Unlike cryptographic software, there aren't any restrictions on the export of filtering software, and the continual efforts of users to get around the software provide a steady revenue stream.
VMWare creates a virtual machine for your OS to run in. Advantage: provided it's simulation is good, everything that runs on the real hardware runs in the virtual machine Disadvantage: that compatability comes at a significant runtime cost, which makes VMWare mainly used only for testing purposes, not for running multiple OSes for general work.
Hypervisors (like Xen) and what MS is claiming (I will believe this when we see it given the list of things they've dropped) use a technique called paravirtualisation to vastly reduce the speed problems. However, this requires support from the host OS. The Xen performance page (http://www.cl.cam.ac.uk/Research/SRG/netos/xen/pe rformance.html) describes this better than I could.
IANAL, and also not any sort of licensing expert, but the really iffy bit looks like
Only Red Hat, Inc. may make changes or additions to the list of Approved Interfaces.
This means that only Red Hat, Inc. gets to make changes to the list of Approved Interfaces. Since the file that contains the list of Approved Interfaces is also distributed under the GNU GPL, it's perfectly fine to make changes to that file. We just wanted to make sure that if you do make changes to that file that the additional rights granted through this exception do not apply.
I think there's some unresolved issues there, namely that (for the sake of argument) Red Hat goes out of business, the software becomes significantly restricted as no-one else would be allowed to extend the Approved Interfaces list (without getting permissions from the now disappeared major copyright holder i.e. Red Hat), and so the development of new interfaces for plugins to non-GPL'ed code becomes impossible. Ok, this means we just stick to GPL plugins, but still...
"A feature will be implemented if and only if there is a developer who wants to implement it"
Of course. For those of us not living in magic pixie land, a user doesn't just request a feature, wave a wand over the CVS tree and a feature appears out of thin air... However, there are various ways to get a developer to do a feature
1) The developer thinks of the idea and does it 2) You are paying the developer's salary and want something done (or have persuaded those paying said salary) 3) The developer is persuaded this is a good idea. Ways include: providing example cases for why this feature would be good, adding items to the Bugzilla about the feature (if you can't be bothered doing that, don't ask others to put their time in building things for you), *talking* to the relevant developer and explaining any questions they may have about your requested feature, etc.
Commercial software only does 1) and 2). Be happy you've got 3) as well.
Given Via's history of announcing Mini/Nano-itx boards as "shipping now" and not shipping for anything up to a year or so (anyone actually seen commercial supplies of standard generic 1-processor Nano-ITX boards yet?), does anyone know if this is *actually* shipping? Mini-itx.com doesn't have it, neither does epiacenter.com or linitx.com. I'll believe "shipping now" when someone actually has it...
Have you actually *tried* to get this free access? As opposed to getting IBM to do it themselves, they've offloaded this onto the hapless University of Portland, who appear to currently be completely unable to give out accounts. Possibly the/. effect has hammered their mail server into the ground, but I'm not seeing any sign of the "you have signed up" mail for one of those accounts. Tried setting up another account a few hours ago, and still no dice.
Somewhat. I'm using a Debian system with the Ubuntu package lines added to my apt.conf. At the moment the Ubuntu packages mostly use the same names as Debian ones, so you can vaguely mix and match. I use Ubuntu to get more up-to-date things (some parts of Gnome 2.8 aren't packaged yet for Debian for example). However, I wouldn't recommend this for a Debian newbie, as the best way to do this is with apt-pinning.
Now there's a first. Not even Google has ever directly supported Mozilla - the Google toolbar from Google is IE only. And this one now has a Mozilla search plugin link on the front page. Kudos.
Slashdot Mirror
For the non-physicists of us wandering around Imperial, any chance we could get to see some of the cool experiments? Just for kicks and giggles quite frankly (*far* too down the CS route to considering switching allegiance), but might it be possible anyway?
Yes, but legally speaking you need a Windows license before you can download/use IE...
Computer Science ain't dead yet (although it may smell that way at times), you just need to have a degree program that's worthwhile. Crappy places churning out more idiots hoping to make a fast quid tend to die off at these times, but the better ones survive.
Only if we also get a button that says "your idea is heavy on PR fluff, contains zero real information and it was in fact a waste of my time to RTFA. Therefore, I would like one of your dollars to compensate for my wasted time"
Given VIA's history of "rapid with press releases, incredibly slow with actual boards" (NanoITX anyone?), any bets on how long before we see available boards? I'm betting mid-2007 personally...
802.15.4 is the PHY, Packet and a basic MAC layer. ZigBee is a routing layer on top of 802.15.4.
Right now, TinyOS has limited 802.15.4 support (certainly for the CC2420 radio that the MicaZ uses), but no-one AFAIK has implemented ZigBee support yet. I'm thinking that a more sensible option is to use 802.15.4 to actually do packet level transmission, but then use another routing layer (shameless plug) to do node-to-node transmission. How to do this is to have 802.15.4 nodes from RandomCompany doing the actual sensing in combination with something like the MicaZ - the RandomCompany nodes just see a 802.15.4 network, and the MicaZ nodes can then encapsulate the data from the sensors into a TinyOS packet which can then be routed around as per normal. This sort of heterogeneous network (specialist purpose 802.15.4 sensor nodes w/o TinyOS + general purpose computing/routing nodes w/ TinyOS) looks like one of the interesting directions that things are going.
ZigBee is designed towards the home automation market, *not* wireless sensors. I could name half a dozen more stable and power efficient routing algorithms for WSNs (Wireless Sensor Networks) than this beacon crap they came up with. 802.15.4 on the other hand, is being grabbed onto thoroughly, and that's because the advantages of having a stable PHY and Packet layer for WSNs (as opposed to the current situation where various nodes even from the same people can't communicate) is enormous. The combination of a general purpose computing node + 802.15.4 hardware is a damn good idea, and one that's liable to survive for at least a while. I've spent the last few weeks messing around with a TinyOS node with 802.15.4 and porting our MAC layer onto it, and it's been a lot easier than most platforms.
(On this note, don't buy the MaxStream 802.15.4 chips, because they're non-conformant - got an official line on that from one of their engineers. They're building a 802.15.4-like proprietary protocol.)
So? They still have a problematic human-rights record, and it's getting *worse* not better. Most of the major injustices (Guantanamo Bay for example) didn't exist 5 years ago, and so far every new break of international law isn't getting fixed later on. The biggest problem is that the USA is getting away with a lot of this stuff, mainly because it's the 400 pound gorilla right now.
Replacing the USA's role in the internet with an impartial UN-backed body would also avoid the problem of global choices being made here according to the opinions of a vocal minority of US citizens, as opposed to by the bloody committee that's meant to decide these matters.
I think you've already got a full set there.
Yes, but can they do it repeatedly in the same way to lots of different people. Betting that no matter how much you pay them, they won't be able to hit everyone *exactly* the same way...
Given that the new Administrator is a politicial nominee, and given that he's the one hiring/firing here, people being shifted by a politicial nominee might well have political reasons for their change in state.
Of course, this might not be the case, but given the current US administration's track record of replacing good people with duckspeakers, it's a good idea to check their backgrounds.
AFAIK, a lot of this gets done by a variety of American companies, who are quite happy providing and customizing their filtering software for anyone willing to pay up. Unlike cryptographic software, there aren't any restrictions on the export of filtering software, and the continual efforts of users to get around the software provide a steady revenue stream.
VMWare creates a virtual machine for your OS to run in.
e rformance.html) describes this better than I could.
Advantage: provided it's simulation is good, everything that runs on the real hardware runs in the virtual machine
Disadvantage: that compatability comes at a significant runtime cost, which makes VMWare mainly used only for testing purposes, not for running multiple OSes for general work.
Hypervisors (like Xen) and what MS is claiming (I will believe this when we see it given the list of things they've dropped) use a technique called paravirtualisation to vastly reduce the speed problems. However, this requires support from the host OS. The Xen performance page (http://www.cl.cam.ac.uk/Research/SRG/netos/xen/p
I think there's some unresolved issues there, namely that (for the sake of argument) Red Hat goes out of business, the software becomes significantly restricted as no-one else would be allowed to extend the Approved Interfaces list (without getting permissions from the now disappeared major copyright holder i.e. Red Hat), and so the development of new interfaces for plugins to non-GPL'ed code becomes impossible. Ok, this means we just stick to GPL plugins, but still...
I for one welcome our super-rodent masters.
/. staple, at least get it right.
</pedant>
If you're going to go with a
Either all of the above, or maybe just Yet Another Moron. It's hard to tell.
"A feature will be implemented if and only if there is a developer who wants to implement it"
Of course. For those of us not living in magic pixie land, a user doesn't just request a feature, wave a wand over the CVS tree and a feature appears out of thin air... However, there are various ways to get a developer to do a feature
1) The developer thinks of the idea and does it
2) You are paying the developer's salary and want something done (or have persuaded those paying said salary)
3) The developer is persuaded this is a good idea. Ways include: providing example cases for why this feature would be good, adding items to the Bugzilla about the feature (if you can't be bothered doing that, don't ask others to put their time in building things for you), *talking* to the relevant developer and explaining any questions they may have about your requested feature, etc.
Commercial software only does 1) and 2). Be happy you've got 3) as well.
Given Via's history of announcing Mini/Nano-itx boards as "shipping now" and not shipping for anything up to a year or so (anyone actually seen commercial supplies of standard generic 1-processor Nano-ITX boards yet?), does anyone know if this is *actually* shipping? Mini-itx.com doesn't have it, neither does epiacenter.com or linitx.com. I'll believe "shipping now" when someone actually has it...
ANT is a video aggregator for video blogs
Single platform, don't have the platform, unwilling to pay lots of money to get the platform. Next!
http://www.btefnet.net/index.php?show=19
Still haven't seen a thing. Tried again about an hour ago...
Have you actually *tried* to get this free access? As opposed to getting IBM to do it themselves, they've offloaded this onto the hapless University of Portland, who appear to currently be completely unable to give out accounts. Possibly the /. effect has hammered their mail server into the ground, but I'm not seeing any sign of the "you have signed up" mail for one of those accounts. Tried setting up another account a few hours ago, and still no dice.
Somewhat. I'm using a Debian system with the Ubuntu package lines added to my apt.conf. At the moment the Ubuntu packages mostly use the same names as Debian ones, so you can vaguely mix and match. I use Ubuntu to get more up-to-date things (some parts of Gnome 2.8 aren't packaged yet for Debian for example). However, I wouldn't recommend this for a Debian newbie, as the best way to do this is with apt-pinning.
Now there's a first. Not even Google has ever directly supported Mozilla - the Google toolbar from Google is IE only. And this one now has a Mozilla search plugin link on the front page. Kudos.