Having written most of the udpconnect code in LimeWire, the basic idea works fairly simply. The downloader starts pinging the desired uploader with UDP SYN messages. At the same time it uses what we call a PushProxy in Gnutella to tell the uploader to start doing the same thing. So then both computers are sending UDP SYNs. This makes the NAT/firewall open up to this traffic and the LimeWire hosts on both end respond to the UDP SYNs with UDP ACKs in order to identify their connection ID.
After receiving the ACKs, the connections can send UDP data messages in both direction safely. The only trick is you need to ensure that a message is sent every so often so the NAT/firewall doesn't close. If nothing else is sent, a special KEEPALIVE message is sent. Beyond this, the communication is somewhat similar to TCP with a FIN message shutting things down at the end.
You wouldn't be using the internet and your nice, small, efficient computer if it wasn't for the space program to send men to the moon. That's what drove the explosion of technology we see today. Duh!
Having worked in the mission analysis group at Telesat, they do like the cheaper solutions. As the other comment points out, they do also tend to like heavy satellites with lots of gyros and fuel.
They have used Ariane before which makes it easier to plan the achievement and maintenance of a geosynchronous orbit. They also tend to build 2 satellites for launch with the assumption that at least one will be successful which can mitigate any failure rate - cheaper than insurance.
By the way when I was there, their technology was much more sophisticated than other such companies. They could do with a staff of 5 what it took 50 US military planners to do.
Microsoft is definitely interested in dominating the content business. However, unlike cable they already have this entire industry in the palms of their hands.
By holding out hope of DRM and locking down content, media companies are very favorably disposed towards Microsoft. Buying a competitor and competing directly is too large a risk for Microsoft to seriously consider.
Then again, they want to get into recurring revenue streams of different types so eventually they need to successfully leverage their desktop dominance directly into the content business. To date, those attempts have mostly failed.
Totally agree. Sun's boxes are now only applicable for the super-highend and for the mission critical high uptime price insensitive. Sun will continue to make good money here. Their hardware business needs to target mainframe and supercomputer class users. These are the only areas where high profit per unit will arise.
It has always been possible for Sun to build a good software business. That remains a possibility but the odds are shifting against them (other than with StarOffice). Imagine if they had ever built a compelling suite of Java tools or applications on top of their APIs. To some extent, they have tried. They probably should have bought BEA Weblogic back during the bubble.
Sun will definitely still be around. They still have possibilities but I think they are aiming too high perhaps right now. Their dedicated customers still have value even as they decide to buy more Linux boxes.
Thankfully, Linux has prevented Microsoft from dominating the low end server market like they wanted. From the low end, Microsoft was then going to push into the high end. Can you imagine how much Microsoft would be making off this area if Linux had not stopped them?
On the desktop, we just bought 5 rather powerful developer PCs for $600 each. Of that price, $150 was for Windows 2000 (not XP thank you). 25% of the price is a rather large part of the cost. The decision between Linux and 2000 was pretty close on these boxes and getting closer all the time. Pretty soon, developer workstations could well be all Linux with OpenOffice and the like. I think the competitive threat to Microsoft will soon restrict their desktop and office pricing.
Humans are use to light in the visible spectrum so I wouldn't worry about a light bulb. The Sun puts out a lot of nasty stuff that our atmosphere and magnetic belt protect us from.
If we are not use to the specific spectrum in play, we can't guarantee how cellular biology will react.
It's not the absolute wattage that worries me although I believe that signal strength is governed by an inverse square law so the closer a source is, the greater its local effect.
Is anybody afraid of all this radiation that we are surrounding ourselves with? Sitting in front of a computer all day is bad enough but everywhere you go there is more spectrum in use. I love the idea of a wireless world but I hope our kids don't come out looking blue.
I've seen reports that cell phones can noticeably heat up the users brain (seriously!). Anything concrete yet on the overall effect?
Thanks. You learn something new every day. That is what is nice about unix style commands. And, yes p is for printing the current top of stack not =.
You know I've read most man pages a fair number of times but there is always something to be missed. I should have known in this case since I always thought it was a glaring shortcoming but I've never seen the K used in practice.
As far as I know, the unique feature of HP calculators is that they use reverse polish notation by default so I expect that is what this one uses.
RPN is similar to stacking languages where the variables are pushed onto the stack first and then the operators are applied to the appropriate stack entries. So, a + b x c should be entered as b c x a + and (a + b) x c should be entered as a b + c x
Similar to other stacking languages like Postscript and SmallTalk.
Note that this should be the same as the desktop "dc" command on linux except that dc is only good for integers (and you need to enter "=" to see the result). I routinely add a few extra significant digits when using dc so that I can deal with fractional values. Works nicely occasionally when doing command line scripting.
We found that posting our contact email addresses on a well known website was definitely the worst thing to do. There are some very aggressive email harvesters out there that just eat up website content and easily parse out the email addresses. Using some simple javascript tricks to assemble and display your email address piece by piece will defeat the current generation of harvesters.
Some of our old email accounts are now firmly planted in the email lists that these companies sell to each other and will "be in play" forever. Having received numerous offers to assemble and sell email lists (which we will never do), I know a little about these companies. Once your email is known by one of the big players, it will be sold to others in units of thousands for as little as pennies but sometimes up to a buck per thousand.
Re:Gnutella is Scalable, Alive, Well and Growing
on
Gnutella at One Year
·
· Score: 1
The network grew too large for the original LimeWire crawling algorithm. If we were still taking 30 minutes to do the crawl, the numbers would actually be higher. Our current crawl takes about 6 minutes.
Gnutella is Scalable, Alive, Well and Growing
on
Gnutella at One Year
·
· Score: 1
Gnutella is scalable. That "article by the napster backend guy showed that it's dead in the water" was some interesting mathematics but totally WRONG. When that article was written, there were around 6,000 concurrent users on Gnutella - now there are 25,000 at any given time. The network should have already blown up! See for yourself: http://www.limewire.com/hostcount.htm#rolling
There is a very active Gnutella Developer Forum where all the true Gnutella developers from all the major clients have been working to improve the protocol and network for months. They have made great progress and will continue to.
Gnutella was a simple protocol and idea with staying power for the long term. There will be more power and surprises in store.
It is interesting that given these constraints, Gnutella file sharing clients will still work. You will be able to make outgoing requests (and you can use any port) which is all that the protocol needs. Others will be able to download from you by pushing a request to you and having you open the outgoing connection. This is all built in to the Gnutella clients.
These types of actions will however degrade the Gnutella network if everyone adopts them. To some extent, there is a slight shortage of available "Incoming" connections. The network relies on those able to accept incoming connections to do so rather than make outgoing ones. Both types of connections work the same regardless. See Gnutelliums
Slashdot Mirror
Having written most of the udpconnect code in LimeWire, the basic idea works fairly simply. The downloader starts pinging the desired uploader with UDP SYN messages. At the same time it uses what we call a PushProxy in Gnutella to tell the uploader to start doing the same thing. So then both computers are sending UDP SYNs. This makes the NAT/firewall open up to this traffic and the LimeWire hosts on both end respond to the UDP SYNs with UDP ACKs in order to identify their connection ID.
After receiving the ACKs, the connections can send UDP data messages in both direction safely. The only trick is you need to ensure that a message is sent every so often so the NAT/firewall doesn't close. If nothing else is sent, a special KEEPALIVE message is sent. Beyond this, the communication is somewhat similar to TCP with a FIN message shutting things down at the end.
If you are not joking, you are a total moron!
You wouldn't be using the internet and your nice, small, efficient computer if it wasn't for the space program to send men to the moon. That's what drove the explosion of technology we see today. Duh!
Having worked in the mission analysis group at Telesat, they do like the cheaper solutions. As the other comment points out, they do also tend to like heavy satellites with lots of gyros and fuel.
They have used Ariane before which makes it easier to plan the achievement and maintenance of a geosynchronous orbit. They also tend to build 2 satellites for launch with the assumption that at least one will be successful which can mitigate any failure rate - cheaper than insurance.
By the way when I was there, their technology was much more sophisticated than other such companies. They could do with a staff of 5 what it took 50 US military planners to do.
LimeWire is spyware/bundled software free.
LimeWire has always been a supporter of an open protocol and open source (limewire.org). Bad Kazaa. Good LimeWire.
This is great news! The importance can not be understated w.r.t. software developer's rights.
Having submitted an amicus brief in the case on behalf of Gnutella developers, we hope that it helped.
http://www9.limewire.com/amicus_brief.pdf
Microsoft is definitely interested in dominating the content business. However, unlike cable they already have this entire industry in the palms of their hands.
By holding out hope of DRM and locking down content, media companies are very favorably disposed towards Microsoft. Buying a competitor and competing directly is too large a risk for Microsoft to seriously consider.
Then again, they want to get into recurring revenue streams of different types so eventually they need to successfully leverage their desktop dominance directly into the content business. To date, those attempts have mostly failed.
Totally agree. Sun's boxes are now only applicable for the super-highend and for the mission critical high uptime price insensitive. Sun will continue to make good money here. Their hardware business needs to target mainframe and supercomputer class users. These are the only areas where high profit per unit will arise.
It has always been possible for Sun to build a good software business. That remains a possibility but the odds are shifting against them (other than with StarOffice). Imagine if they had ever built a compelling suite of Java tools or applications on top of their APIs. To some extent, they have tried. They probably should have bought BEA Weblogic back during the bubble.
Sun will definitely still be around. They still have possibilities but I think they are aiming too high perhaps right now. Their dedicated customers still have value even as they decide to buy more Linux boxes.
Thankfully, Linux has prevented Microsoft from dominating the low end server market like they wanted. From the low end, Microsoft was then going to push into the high end. Can you imagine how much Microsoft would be making off this area if Linux had not stopped them?
On the desktop, we just bought 5 rather powerful developer PCs for $600 each. Of that price, $150 was for Windows 2000 (not XP thank you). 25% of the price is a rather large part of the cost. The decision between Linux and 2000 was pretty close on these boxes and getting closer all the time. Pretty soon, developer workstations could well be all Linux with OpenOffice and the like. I think the competitive threat to Microsoft will soon restrict their desktop and office pricing.
Funny, I've heard that before ...
Actually, I recall having "k" pointed out to me on another occasion when I introduced someone to dc.
I guess the trick to remembering something is to actually use it.
[root@www httpd]# dc
1
k
2.222
1.111
+
p
3.333
Great! Now I will remember.
Humans are use to light in the visible spectrum so I wouldn't worry about a light bulb. The Sun puts out a lot of nasty stuff that our atmosphere and magnetic belt protect us from.
If we are not use to the specific spectrum in play, we can't guarantee how cellular biology will react.
It's not the absolute wattage that worries me although I believe that signal strength is governed by an inverse square law so the closer a source is, the greater its local effect.
Is anybody afraid of all this radiation that we are surrounding ourselves with? Sitting in front of a computer all day is bad enough but everywhere you go there is more spectrum in use. I love the idea of a wireless world but I hope our kids don't come out looking blue.
I've seen reports that cell phones can noticeably heat up the users brain (seriously!). Anything concrete yet on the overall effect?
Thanks. You learn something new every day. That is what is nice about unix style commands. And, yes p is for printing the current top of stack not =.
You know I've read most man pages a fair number of times but there is always something to be missed. I should have known in this case since I always thought it was a glaring shortcoming but I've never seen the K used in practice.
As far as I know, the unique feature of HP calculators is that they use reverse polish notation by default so I expect that is what this one uses.
RPN is similar to stacking languages where the variables are pushed onto the stack first and then the operators are applied to the appropriate stack entries. So, a + b x c should be entered as b c x a + and (a + b) x c should be entered as a b + c x
Similar to other stacking languages like Postscript and SmallTalk.
Note that this should be the same as the desktop "dc" command on linux except that dc is only good for integers (and you need to enter "=" to see the result). I routinely add a few extra significant digits when using dc so that I can deal with fractional values. Works nicely occasionally when doing command line scripting.
We found that posting our contact email addresses on a well known website was definitely the worst thing to do. There are some very aggressive email harvesters out there that just eat up website content and easily parse out the email addresses. Using some simple javascript tricks to assemble and display your email address piece by piece will defeat the current generation of harvesters.
Some of our old email accounts are now firmly planted in the email lists that these companies sell to each other and will "be in play" forever. Having received numerous offers to assemble and sell email lists (which we will never do), I know a little about these companies. Once your email is known by one of the big players, it will be sold to others in units of thousands for as little as pennies but sometimes up to a buck per thousand.
The network grew too large for the original LimeWire crawling algorithm. If we were still taking 30 minutes to do the crawl, the numbers would actually be higher. Our current crawl takes about 6 minutes.
http://www.limewire.com/hostcount.htm#rolling
There is a very active Gnutella Developer Forum where all the true Gnutella developers from all the major clients have been working to improve the protocol and network for months. They have made great progress and will continue to.
Gnutella was a simple protocol and idea with staying power for the long term. There will be more power and surprises in store.
Uhh, Furi is a little out of date now. Perhaps you should try LimeWire: http://www.limewire.com/download
These types of actions will however degrade the Gnutella network if everyone adopts them. To some extent, there is a slight shortage of available "Incoming" connections. The network relies on those able to accept incoming connections to do so rather than make outgoing ones. Both types of connections work the same regardless. See Gnutelliums