Yes, but the thing to realize about optical switches is that the switching time is on the order of tens of *milliseconds* (as opposed to nanoseconds), because you have to physically move a mirror in order to change the path. It's totally different from an electronic switch in that you're not switching on a packet-by-packet basis.
There are several killer apps for this kind of technology: one is setting up dedicated channels between a server and a client, EG so that you can download a 2GB movie in a couple of seconds. Another is dynamic allocation of channels on the backbone - eg if an ISP gets slashdotted, additional fiber channels could be brought up to the backbone provider or other peers. Finally, you can use it to switch a particular circuit over to an alternate route when a backhoe cuts the fiber, without having to have all the fiber terminated at routers on each end - just move the physical link in between.
so do we have any proof that if a human baby is given no "input", they will grow up to be intelligent?
Obviously not, but that isn't the point at all. A human has the CAPACITY to absorb and process input. Alice can't do anything unless explicity programmed. That's why it's not AI.
You can put a couple gigs of RAM on your motherboard, where the bandwidth to the CPU is at least 10x the fastest SCSI interface. Run any modern OS on there, and all of that main memory is going to be used a filesystem cache. Voila: all the benefits of a RAM disk (fast seek, throughput) and none of the drawbacks (no need for a separate disk backup).
If what you want is a TRUE ram disk, i.e. not backed by magnetic storage at all, then you can do this in Linux or FreeBSD by setting aside a chunk of main RAM as a file system. I don't know if you can do that in OSX or Windows...
If you start with, say, 20 lbs of supersonic projectile, and then you zap it with a laser, you still have 20 lbs of something moving with about the same average velocity as before.
Find a tile floor. Drop a 16lb bowling ball and see what happens. Now drop 16lbs of marbles weighing 1oz each. Any difference?
... but Ogg isn't going to make any major headway until the embedded decoder vendors (Crystal, Micronas, ST) start supporting it. Two things need to happen: one, the Vorbis folks need to get the codec to run on these smaller DSPs with a free reference implementation, and two, the DSP vendors need to be convinced that it's worth the precious ROM space to fit another codec in there.
Ogg just came to the party WAY too late. It is up against a massive chicken-and-egg problem if it wants to supplant MP3. Nobody's using Ogg because it's not supported, and nobody's supporting it because nobody wants it. The advantages of Ogg (slightly better quality, free) are massively outweighed by the ubiquity of MP3. Like 'em of not, Fraunhofer did a fantastic job with the original codec, and it's going to take something with a massive improvement in quality/compression/cost to supplant it. Ogg is better, but not "better enough".
So lets get off the anti-Asian rants and show a bit of consistency. Either both are bad, or both are a chance for artists to reach an audience they otherwise would not.
uh - many different people post on slashdot. Why do you expect a consensus?
He said every problem on their list was present more than once, on top of the distance being 50% outside their max window for IDSL (which would have been a whopping 144kbps anyway).
Something sounds wrong here.... you should be able to get IDSL through any central office where the CLEC provides the service, regardless of distance, because it used good old fashioned "repeatered" ISDN circuits. that's really the whole point of IDSL. This was definitely true three years ago, at least with Covad/SBC - I know because I was owned an ISP back then, and I installed many, many Covad IDSL circuits for people who were too far away for broadband DSL.
You should call your ILEC and ask if they can install a regular BRI ISDN line. If they say yes, then someone is jerking your chain. Depending on what state you live in, the ILEC may even be *required* to service you with ISDN. At least you'll be able to tell the DSL provider that line distance/quality is definitely not the problem...
BTW, I am currently using Speakeasy/Covad IDSL at home, because I'm too far from the CO for broadband DSL, and cable modem is not available. In terms of $/bps, it sucks, but speakeasy is very reliable and their service is top-notch.
Researchers have created a new way to encrypt information in a digital image and extract it later without any distortion or loss of information.
So, if I can add some information to an image without any loss of information in the original, then I don't see any reason why I couldn't use this technique repeatedly, ad inifinitum, on the resulting image. Therefore, they have created a way to turn any one of my pr0n jpegs into an unlimited storage device.
This really changes everything we thought we knew about computer science and information theory. What an incredible discovery!
I mean jeeze, only 4 or 5 out of 13 survived according to the WashPost.
I'd say this just goes to show how reliable the root name servers are. I didn't notice any dns problems yesterday. In fact, I don't remember any root name server problems since the infamous alternic takeover.
Hurricane Electric supports spamming by not kicking them off their network.
Evidence?
SPEWS does not keep records of who the offending customers were, who they talked to at the ISP, what the resolution was, and so on. If you read the report, it's just a huge list of IP address. I could not find any indication on google or usenet that he.net is spam-friendly. The onus should be on the spam cops to show that an ISP is irresponsible before blacklisting them, and anyone using a blacklist should insist on at least this level of accountability.
Besides, switching ISPs is not easy or cheap. I've got a lot of equipment and telco circuits installed there. I've pressured them to get this resolved, but they tell me that they have strict policy of disconnecting spammers, and they've already done everything they can. Now SPEWS needs to confirm that the spammers are no longer there, and lift the block.
Take a look on comp.net.abuse.email and read about the many admins who are complaining about SPEWS. The problem with SPEWS is that they often block large ranges of IP addresses as a punitive measure against ISPs they don't like - willfully blocking legitimate mail in this way seems awfully ironic. I realize that anyone can choose whether or not they want to filter with SPEWS, but the problem is that they don't tell you about this policy. Every once in a while I'll get an email from someone and my reply will bounce back because they're blocking me. I'll contact them from another account and explain the situation, and these people are unanimously surprised and pissed that SPEWS is doing this.
They recently blacklisted a huge swath of IP addresses - hundreds of class Cs, deliberately blocking not just spammers but thousands of IP addresses on neighboring subnets. Sadly, my little block of 64 IPs was included. So I went on the mailing list (SPEWS will not respond to inquiries) and suggested than an error had been made. My IP was coming up as a "confirmed source of spam" in spamasassin and other tools. I was immediately bombarded by a bunch of leet little fucks telling me it was my fault for choosing the wrong ISP, and I need to switch.
Fuck SPEWS. I like my ISP, and I could find no evidence of them being spam-friendly. In fact, SPEWS keeps almost zero documentation. The just block whatever the hell they want, and they're accountable to no one.
Please don't filter with SPEWS unless you want to lose contact with a good chunk of legit mail servers which have deliberately been blacklisted!!!
IDE drives are fast in a single user/workstation environment. As a file server for thousands of people sharing an array of drives? I'm sure the output was solid for a single user when they benched it... looks like/. is letting them know what multiple users do to IDE. 'Overhead of SCSI controller'... Methinks they do not know how SCSI works. The folks who share this box will suffer.
Methinks it's been a LONG time since you've read up on IDE vs SCSI, and me also thinks you dont have the first clue about how a filesystem works. Yes, there was a time when IDE drives were way slower, mainly because the bus could only have one outstanding request at a time. IDE has since advanced to support tagged command queuing and faster data rates, closing the gap with all but the most horrendously expensive flavors of SCSI. Really, the bottleneck is spindle and seek speed - both IDE and SCSI are plenty fast now.
The only thing SCSI really has going for it is daisy-chainability and support for lots of drives on one port. HOWEVER there are some really killer things you can do with IDE now. In my web server I'm using the promise RM8000 subsystem: a terabyte of RAID5 storage for about $3500 including the drives IIRC. Try doing that with SCSI drives!
Anyway.... you suggest that this server is slashdotted because it's disk-bound. Serving the exact same page over and over again. Uh huh. Go read up on any modern file system, then figure out how long it takes to send a 100KB web page to 250,000 people over a DSL line, and then tell me where you think the problem lies.
In-car TV & DVD already exist so what makes this any more of a distraction?
Here in California, these things are illegal unless they're designed to only operate when the engine is off. Anybody will install one for you, but you have to go in yourself and change the wiring to make it think that the ignition is always off.
Unfortunately, for the past couple decades, management has been setting its own wages without much imput from the owners (the stock holders who care about profit).
That's the owners' fault, not the managment! We're all out for #1 - the problem is that many businesses are created where the exectives' personal agenda is at odds with the needs of the company. In nearly all of these cases, management either a) did not hold a sufficient stake in the company to care about its success, or b) DID own a good chunk, but never believed in the business to begin with, so they wanted to cash out ASAP.
Cash messes with your head. Few people can keep their cool with a lot of dough, so it's important that CEOs are compensated in stock to the greatest extent possible. Then, when the company brings in money either through investments or, heaven forfend, SALES, the executives will be incented to spend the money on growing the business rather than enriching themselves short-term.
The best managers/CEOs are those who would rather spend the company's money on growing their business than on their own paychecks - ironically, they're the ones who end up with the most personal wealth in the long run. You just have to think of wealth as a consequence of your hard work, rather than an immediate goal.
It's okay for the owners to be hands-off, as long as they've set up the business properly to begin with.
Steve Jobs said, ""If you legally acquire music, you need to have the right to manage it on all other devices that you own."
And at the same time, Apple implements a form of DRM by crippling the iPod - you can copy songs onto it, but you can't copy them off. I'm an Apple fan, but this kind of BS undermines my faith in them.
This sounds interesting. When you say "in concert", do you mean that the music output is perfectly synchronized on each device? So I can have several of these things around the house, and not get nasty echoing effects when I play the same tunes out of all of them at once? I had assumed that the various buffering at different stages along the way would make this impossible.
That is correct. Of course there are practical limitations due to network and OS latency on the server, but it gets very close to perfect sync. When the stream initially starts up, it is synchronized to within 2ms (that's roughly two feet at the speed of sound). So as you walk from room to room, you won't hear any reverb/echo effect. This effect only becomes noticeable to the untrained ear at about 75ms, and annoying at about 100ms. At 2ms, it's probably less than the delay between your two speakers!
After multiple players have been running for a long time (about 12-24 hrs) they may drift out of sync a little. However, it resyncs every time you switch tracks, so drift is not a problem unless you're using this for an unusual application where you need perfect sync with a continuous stream, for days on end.
In case you were wondering, the UI for this is also pretty itneresting. When multiple SliMP3s are grouped together, they all share the same playback queue. However, music browsing and player-specific things like brightness/volume can all be controlled independently. When you hit fast forward/stop/pause etc, all the players repsond in unison.
Why not send the decoded ogg over the wire in raw format?
We're working with Micronas to get a firmware image for the decoder chip that will let it do this (basically, a "null codec"). No luck yet, though the rest of the system is fast enough to handle the data rate.
Slashdot Mirror
Yes, but the thing to realize about optical switches is that the switching time is on the order of tens of *milliseconds* (as opposed to nanoseconds), because you have to physically move a mirror in order to change the path. It's totally different from an electronic switch in that you're not switching on a packet-by-packet basis.
There are several killer apps for this kind of technology: one is setting up dedicated channels between a server and a client, EG so that you can download a 2GB movie in a couple of seconds. Another is dynamic allocation of channels on the backbone - eg if an ISP gets slashdotted, additional fiber channels could be brought up to the backbone provider or other peers. Finally, you can use it to switch a particular circuit over to an alternate route when a backhoe cuts the fiber, without having to have all the fiber terminated at routers on each end - just move the physical link in between.
so do we have any proof that if a human baby is given no "input", they will grow up to be intelligent?
Obviously not, but that isn't the point at all. A human has the CAPACITY to absorb and process input. Alice can't do anything unless explicity programmed. That's why it's not AI.
Fine, I should have RTFA, but the exact same argument holds.
And here's why:
You can put a couple gigs of RAM on your motherboard, where the bandwidth to the CPU is at least 10x the fastest SCSI interface. Run any modern OS on there, and all of that main memory is going to be used a filesystem cache. Voila: all the benefits of a RAM disk (fast seek, throughput) and none of the drawbacks (no need for a separate disk backup).
If what you want is a TRUE ram disk, i.e. not backed by magnetic storage at all, then you can do this in Linux or FreeBSD by setting aside a chunk of main RAM as a file system. I don't know if you can do that in OSX or Windows...
But a RAM disk on the SCSI bus? What's the point?
Dan Quayle reads slashdot?
If you start with, say, 20 lbs of supersonic projectile, and then you zap it with a laser, you still have 20 lbs of something moving with about the same average velocity as before.
:)
Find a tile floor. Drop a 16lb bowling ball and see what happens. Now drop 16lbs of marbles weighing 1oz each. Any difference?
Obviously, you're not a golfer.
I guess it's ime to plug a few tantalum caps into 110VAC and see what happens... :)
... but Ogg isn't going to make any major headway until the embedded decoder vendors (Crystal, Micronas, ST) start supporting it. Two things need to happen: one, the Vorbis folks need to get the codec to run on these smaller DSPs with a free reference implementation, and two, the DSP vendors need to be convinced that it's worth the precious ROM space to fit another codec in there.
Ogg just came to the party WAY too late. It is up against a massive chicken-and-egg problem if it wants to supplant MP3. Nobody's using Ogg because it's not supported, and nobody's supporting it because nobody wants it. The advantages of Ogg (slightly better quality, free) are massively outweighed by the ubiquity of MP3. Like 'em of not, Fraunhofer did a fantastic job with the original codec, and it's going to take something with a massive improvement in quality/compression/cost to supplant it. Ogg is better, but not "better enough".
So lets get off the anti-Asian rants and show a bit of consistency. Either both are bad, or both are a chance for artists to reach an audience they otherwise would not.
uh - many different people post on slashdot. Why do you expect a consensus?
If they only enforce the laws for domestic films, then what is going to stop the average Chinaman from going to pirate an American movie instead?
Dude, Chinaman isn't the proper nomenclature; Asian-American, please.
Does anyone know whose chip they're using for the video decoder?
www.waterhouse.com doesn't work for beans in anything except IE, and the site has always been very cluttered and slow.
He said every problem on their list was present more than once, on top of the distance being 50% outside their max window for IDSL (which would have been a whopping 144kbps anyway).
Something sounds wrong here.... you should be able to get IDSL through any central office where the CLEC provides the service, regardless of distance, because it used good old fashioned "repeatered" ISDN circuits. that's really the whole point of IDSL. This was definitely true three years ago, at least with Covad/SBC - I know because I was owned an ISP back then, and I installed many, many Covad IDSL circuits for people who were too far away for broadband DSL.
You should call your ILEC and ask if they can install a regular BRI ISDN line. If they say yes, then someone is jerking your chain. Depending on what state you live in, the ILEC may even be *required* to service you with ISDN. At least you'll be able to tell the DSL provider that line distance/quality is definitely not the problem...
BTW, I am currently using Speakeasy/Covad IDSL at home, because I'm too far from the CO for broadband DSL, and cable modem is not available. In terms of $/bps, it sucks, but speakeasy is very reliable and their service is top-notch.
Researchers have created a new way to encrypt information in a digital image and extract it later without any distortion or loss of information.
So, if I can add some information to an image without any loss of information in the original, then I don't see any reason why I couldn't use this technique repeatedly, ad inifinitum, on the resulting image. Therefore, they have created a way to turn any one of my pr0n jpegs into an unlimited storage device.
This really changes everything we thought we knew about computer science and information theory. What an incredible discovery!
I mean jeeze, only 4 or 5 out of 13 survived according to the WashPost.
I'd say this just goes to show how reliable the root name servers are. I didn't notice any dns problems yesterday. In fact, I don't remember any root name server problems since the infamous alternic takeover.
my bad.. I meant he.net
Hurricane Electric supports spamming by not kicking them off their network.
Evidence?
SPEWS does not keep records of who the offending customers were, who they talked to at the ISP, what the resolution was, and so on. If you read the report, it's just a huge list of IP address. I could not find any indication on google or usenet that he.net is spam-friendly. The onus should be on the spam cops to show that an ISP is irresponsible before blacklisting them, and anyone using a blacklist should insist on at least this level of accountability.
Besides, switching ISPs is not easy or cheap. I've got a lot of equipment and telco circuits installed there. I've pressured them to get this resolved, but they tell me that they have strict policy of disconnecting spammers, and they've already done everything they can. Now SPEWS needs to confirm that the spammers are no longer there, and lift the block.
Take a look on comp.net.abuse.email and read about the many admins who are complaining about SPEWS. The problem with SPEWS is that they often block large ranges of IP addresses as a punitive measure against ISPs they don't like - willfully blocking legitimate mail in this way seems awfully ironic. I realize that anyone can choose whether or not they want to filter with SPEWS, but the problem is that they don't tell you about this policy. Every once in a while I'll get an email from someone and my reply will bounce back because they're blocking me. I'll contact them from another account and explain the situation, and these people are unanimously surprised and pissed that SPEWS is doing this.
They recently blacklisted a huge swath of IP addresses - hundreds of class Cs, deliberately blocking not just spammers but thousands of IP addresses on neighboring subnets. Sadly, my little block of 64 IPs was included. So I went on the mailing list (SPEWS will not respond to inquiries) and suggested than an error had been made. My IP was coming up as a "confirmed source of spam" in spamasassin and other tools. I was immediately bombarded by a bunch of leet little fucks telling me it was my fault for choosing the wrong ISP, and I need to switch.
Fuck SPEWS. I like my ISP, and I could find no evidence of them being spam-friendly. In fact, SPEWS keeps almost zero documentation. The just block whatever the hell they want, and they're accountable to no one.
Please don't filter with SPEWS unless you want to lose contact with a good chunk of legit mail servers which have deliberately been blacklisted!!!
IDE drives are fast in a single user/workstation environment. As a file server for thousands of people sharing an array of drives? I'm sure the output was solid for a single user when they benched it... looks like /. is letting them know what multiple users do to IDE. 'Overhead of SCSI controller'... Methinks they do not know how SCSI works. The folks who share this box will suffer.
Methinks it's been a LONG time since you've read up on IDE vs SCSI, and me also thinks you dont have the first clue about how a filesystem works. Yes, there was a time when IDE drives were way slower, mainly because the bus could only have one outstanding request at a time. IDE has since advanced to support tagged command queuing and faster data rates, closing the gap with all but the most horrendously expensive flavors of SCSI. Really, the bottleneck is spindle and seek speed - both IDE and SCSI are plenty fast now.
The only thing SCSI really has going for it is daisy-chainability and support for lots of drives on one port. HOWEVER there are some really killer things you can do with IDE now. In my web server I'm using the promise RM8000 subsystem: a terabyte of RAID5 storage for about $3500 including the drives IIRC. Try doing that with SCSI drives!
Anyway.... you suggest that this server is slashdotted because it's disk-bound. Serving the exact same page over and over again. Uh huh. Go read up on any modern file system, then figure out how long it takes to send a 100KB web page to 250,000 people over a DSL line, and then tell me where you think the problem lies.
In-car TV & DVD already exist so what makes this any more of a distraction?
Here in California, these things are illegal unless they're designed to only operate when the engine is off. Anybody will install one for you, but you have to go in yourself and change the wiring to make it think that the ignition is always off.
Unfortunately, for the past couple decades, management has been setting its own wages without much imput from the owners (the stock holders who care about profit).
That's the owners' fault, not the managment! We're all out for #1 - the problem is that many businesses are created where the exectives' personal agenda is at odds with the needs of the company. In nearly all of these cases, management either a) did not hold a sufficient stake in the company to care about its success, or b) DID own a good chunk, but never believed in the business to begin with, so they wanted to cash out ASAP.
Cash messes with your head. Few people can keep their cool with a lot of dough, so it's important that CEOs are compensated in stock to the greatest extent possible. Then, when the company brings in money either through investments or, heaven forfend, SALES, the executives will be incented to spend the money on growing the business rather than enriching themselves short-term.
The best managers/CEOs are those who would rather spend the company's money on growing their business than on their own paychecks - ironically, they're the ones who end up with the most personal wealth in the long run. You just have to think of wealth as a consequence of your hard work, rather than an immediate goal.
It's okay for the owners to be hands-off, as long as they've set up the business properly to begin with.
Steve Jobs said, ""If you legally acquire music, you need to have the right to manage it on all other devices that you own."
And at the same time, Apple implements a form of DRM by crippling the iPod - you can copy songs onto it, but you can't copy them off. I'm an Apple fan, but this kind of BS undermines my faith in them.
In the MBA world mature means the industry has stopped growing in sales volume (read $$$$),
That's compensated for inflation *and* population growth, right? I'm no MBA, but I don't see hown crackers and soap meet your definition otherwise....
This sounds interesting. When you say "in concert", do you mean
that the music output is perfectly synchronized on each device?
So I can have several of these things around the house, and not
get nasty echoing effects when I play the same tunes out of all of them at once? I had assumed that the various buffering at different stages along the way would make this impossible.
That is correct. Of course there are practical limitations due to network and OS latency on the server, but it gets very close to perfect sync. When the stream initially starts up, it is synchronized to within 2ms (that's roughly two feet at the speed of sound). So as you walk from room to room, you won't hear any reverb/echo effect. This effect only becomes noticeable to the untrained ear at about 75ms, and annoying at about 100ms. At 2ms, it's probably less than the delay between your two speakers!
After multiple players have been running for a long time (about 12-24 hrs) they may drift out of sync a little. However, it resyncs every time you switch tracks, so drift is not a problem unless you're using this for an unusual application where you need perfect sync with a continuous stream, for days on end.
In case you were wondering, the UI for this is also pretty itneresting. When multiple SliMP3s are grouped together, they all share the same playback queue. However, music browsing and player-specific things like brightness/volume can all be controlled independently. When you hit fast forward/stop/pause etc, all the players repsond in unison.
Why not send the decoded ogg over the wire in raw format?
We're working with Micronas to get a firmware image for the decoder chip that will let it do this (basically, a "null codec"). No luck yet, though the rest of the system is fast enough to handle the data rate.