I had a customer who thought I was crazy for suggesting that all they really needed was chroot for their particular task. In the face of VMs people don't realize that simplicity often works better.
Xen's biggest obstacle right now is KVM. I am no VM expert, but I've been impressed with how well KVM runs, supporting non-VM-aware versions of Microsoft Windows among other things. It's really fun to put that Windows screen on the face of someone's iPad and watch them freak out when they see it's not a screenshot, somehow their iPad got Windows 7 installed on it!
Although I have not installed CyanogenMod on my Nexus 4, as I have on my Asus Transformer Infinity tf700, the option is available and I will probably eventually do so. I am installing nightlies every other day on the Transformer. I have the option not to use Google's services since I have control over the OS. IMO Google is selling the unit at parts cost, that's why it's from the Play store rather than another retailer.
Obviously, not being locked in is always considered in my choice of hardware.
I bought an LG / Google Nexus 4 a while back. They're less than half the price of other top-end smartphones, unlocked and with no contract. I put a Platinumtel SIM in it with the $10 for 60 days GSM plan, and set it to restrict background data. The network is T-Mobile. After a month I'm still on the first $10, having of course made extensive use of wifi.
As far as I can tell, I have all of the smartphone benefits without much of the cost.
If I understand it correctly, the defective sensor is sensitive at a spectral peak that is different from the value in normal individuals. It is through mutations to this sensor that color vision evolves. Theoretically, some rare women have four operative spectral bands rather than three.
Well, the only implementation so far is floating-point. I suspect the fixed-point performance would be a lot better if we ever get that written. It can keep up with real-time on a Raspberry Pi or an old Atom laptop.
Correct me if I am wrong, but the technology we are talking about here is merely splicing some reconstucted sequences into existing human cells.
You don't have to synthesize the entire cell. Only the nuclear and mitochondrial chromosomes matter. If you can replace the ones in a normal cell, what you have after division is the primitive cell reborn. You have to do this to a lot of cells, and grow them for a while, to get one without significant damage.
There is more than one Microsoft, and of course Microsoft's decisions may change over time. Remember that this is the company behind the overly-patented filesystem on SDXC cards.
It is because I have been working on Codec2, a perceptual compression project, for some years now. To say that we throw away data would be an understatement.
WebRTC has Opus, the Open Source audio codec that can outperform MP3 and pretty much any audio codec*. It does seem that the proprietary OS industry will do anything they can to stop open codecs from being net standards.
* Anything but FLAC and Codec2 (because FLAC doesn't compress and Codec2 is voice-only and ultra-low-bandwidth).
No one location is safe from natural or human disasters. So, I'd rather hear that they were going to a more distributed architecture and that they'd be able to sustain a complete loss of one data center.
I went to Usenix conferences in 1981 when they were really the only conferences about this technology. These days there are many conferences covering that sort of system, and many of them have a better focus.
We can definitely see SSB on our waterfall. And it interferes with us somewhat. This isn't like the ultra-narrow slow digital modes like WSPR or PSK31 where you might not have to care about another station on the frequency. So, I think our operators would generally avoid it.
I have a surface mount workstation at home. I've been cheering on Chris Testa, whose mobile SDR design is mostly surface mount. He did run into problems with a module socket that was awful to solder.
We have had other spread-spectrum rule-makings lately.
If you want to experiment with it on HF you should apply for a Special Testing Authority. The problem is, as you obviously already know, getting the existing HF operators to live with it. I didn't specifically address SS, but the bands with wide bandwidth or "Not Specified" can obviously handle it. I thought the document was ambitious enough without talking about spread spectrum.
CSMA doesn't work that well on HF. Sometimes you should be able to share a frequency with distant operators even though they fade in and out, etc. However, it makes sense if you are doing automatic link establishment.
I want to have my opinions heard. And having notoriety helps. So I do not shy from self-promotion. But the point here is to get people interested in what I wrote, rather than just me.
It might be different if the purpose was just to sell my stuff. This is a non-profit activity.
The 97.309(a) change I suggest effects all international communication. Not just HF but satellite. I hope they go for my standard of "publicly documented in sufficient detail that a knowledgeable programmer can implement a program to encode and decode".
irst the proposal claims to promote "paperwork reduction" while installing a whole new crazy array of complicated regs ON TOP OF the existing overall rules for reasonable and prudent and good engineering practice and emergency traffic priority or WTF the exact phrases.
You may have been confused by the stuff in the right-hand side of the big table. That's all existing FCC rules. I just moved them to where they'd be seen, instead of having them live in a list of footnotes as in the current Part 97.
If we are going to have unattended traffic, there needs to at least be a ham-adminstered band-plan to keep it in a subband. Nobody wins if unattended traffic is a big HF band user in non-emergency operations.
We are getting a lot of new hams since we got rid of the Morse Code requirement (which I evangelized for years). There are more hams today in the US than there ever have been!
Slashdot Mirror
I had a customer who thought I was crazy for suggesting that all they really needed was chroot for their particular task. In the face of VMs people don't realize that simplicity often works better.
Xen's biggest obstacle right now is KVM. I am no VM expert, but I've been impressed with how well KVM runs, supporting non-VM-aware versions of Microsoft Windows among other things. It's really fun to put that Windows screen on the face of someone's iPad and watch them freak out when they see it's not a screenshot, somehow their iPad got Windows 7 installed on it!
Although I have not installed CyanogenMod on my Nexus 4, as I have on my Asus Transformer Infinity tf700, the option is available and I will probably eventually do so. I am installing nightlies every other day on the Transformer. I have the option not to use Google's services since I have control over the OS. IMO Google is selling the unit at parts cost, that's why it's from the Play store rather than another retailer. Obviously, not being locked in is always considered in my choice of hardware.
As far as I can tell, I have all of the smartphone benefits without much of the cost.
Actually, if you look at Fox over the past few weeks, they've run a series of anti-solar articles. I guess somebody wants to sell more oil.
By the way, I've been told by doctors for at least 20 years that a magenta tint sometimes helps. This isn't really new art.
If I understand it correctly, the defective sensor is sensitive at a spectral peak that is different from the value in normal individuals. It is through mutations to this sensor that color vision evolves. Theoretically, some rare women have four operative spectral bands rather than three.
Well, the only implementation so far is floating-point. I suspect the fixed-point performance would be a lot better if we ever get that written. It can keep up with real-time on a Raspberry Pi or an old Atom laptop.
You don't have to synthesize the entire cell. Only the nuclear and mitochondrial chromosomes matter. If you can replace the ones in a normal cell, what you have after division is the primitive cell reborn. You have to do this to a lot of cells, and grow them for a while, to get one without significant damage.
There is more than one Microsoft, and of course Microsoft's decisions may change over time. Remember that this is the company behind the overly-patented filesystem on SDXC cards.
It is because I have been working on Codec2, a perceptual compression project, for some years now. To say that we throw away data would be an understatement.
Yes. I am out of the habit of thinking of lossless as "real" compression, but I seem to be in the minority.
* Anything but FLAC and Codec2 (because FLAC doesn't compress and Codec2 is voice-only and ultra-low-bandwidth).
No one location is safe from natural or human disasters. So, I'd rather hear that they were going to a more distributed architecture and that they'd be able to sustain a complete loss of one data center.
Is this a laptop camera on the desk? Whose idea was it to shoot up the poor woman's nose?
I went to Usenix conferences in 1981 when they were really the only conferences about this technology. These days there are many conferences covering that sort of system, and many of them have a better focus.
We can definitely see SSB on our waterfall. And it interferes with us somewhat. This isn't like the ultra-narrow slow digital modes like WSPR or PSK31 where you might not have to care about another station on the frequency. So, I think our operators would generally avoid it.
I have a surface mount workstation at home. I've been cheering on Chris Testa, whose mobile SDR design is mostly surface mount. He did run into problems with a module socket that was awful to solder.
Oops. Thanks for the typo.
We have had other spread-spectrum rule-makings lately. If you want to experiment with it on HF you should apply for a Special Testing Authority. The problem is, as you obviously already know, getting the existing HF operators to live with it. I didn't specifically address SS, but the bands with wide bandwidth or "Not Specified" can obviously handle it. I thought the document was ambitious enough without talking about spread spectrum.
CSMA doesn't work that well on HF. Sometimes you should be able to share a frequency with distant operators even though they fade in and out, etc. However, it makes sense if you are doing automatic link establishment.
Maybe he looked up Continental code instead of the International Morse code. I'm not parsing it either.
Hi Roscoe,
I want to have my opinions heard. And having notoriety helps. So I do not shy from self-promotion. But the point here is to get people interested in what I wrote, rather than just me.
It might be different if the purpose was just to sell my stuff. This is a non-profit activity.
I am indeed asking for a ton of stuff that ARRL did not. I support what ARRL asks for, but the don't ask for enough to explicitly authorize FreeDV.
The 97.309(a) change I suggest effects all international communication. Not just HF but satellite. I hope they go for my standard of "publicly documented in sufficient detail that a knowledgeable programmer can implement a program to encode and decode".
You may have been confused by the stuff in the right-hand side of the big table. That's all existing FCC rules. I just moved them to where they'd be seen, instead of having them live in a list of footnotes as in the current Part 97.
If we are going to have unattended traffic, there needs to at least be a ham-adminstered band-plan to keep it in a subband. Nobody wins if unattended traffic is a big HF band user in non-emergency operations.
Thanks
Bruce
We are getting a lot of new hams since we got rid of the Morse Code requirement (which I evangelized for years). There are more hams today in the US than there ever have been!