Actually, it does have major scalability problems...please refer to some of the numerous papers published on exactly this topic.
The supernode/ultrapeer addition to the protocol was meant to address some of the scalability issues of the previous version of the protocol. However, the fragile nature of the original overlay network is only made worse since ultrapeers are explicitly meant to be highly connected nodes.
The problem with Gnutella is that it can't do search efficiently due to the the broadcast nature of forwarding query messages. Research such as Random Walkers(SIGCOMM 02), and Attenuated Bloom Filters (IEEE INFOCOM 02) have tried to tackle the problem of scalable and fast search in unstructured P2P networks like Gnutella.
Sorry, that "solution" is not viable. That algorithm generates overlay networks with Power-Law node degree distributions. This means that the few super nodes have a very large number of connections while the vast majority of nodes have very few (maybe only 1). This means that the although the network can survive random failures, a coordinated attack on the most highly connected nodes (less than 3% of the total) effectively shatters the connectivity of the overlay network. This has been studied in several papers.
There are several P2P research projects that are looking at building reliabale and scalable P2P systems.
Take a look at Tapestry, and Chord (and read some of the papers) to understand the issues involved in providing scalable and high performance P2P services. Not only is scalable search and overlay graph connectivity an issue, but also node failure and short session times of P2P nodes.
Additionally, when you actually handle the issue of downloading data, building application-lvel multicast trees to distribute the data efficiently on a large scale is not easy. Two papers from SOSP '03 SplitStream, and Bullet address that issue.
I hate to be antagonistic, but there are several problems with the method you propose. That model requires that all the attributes that one would wish to search for be embedded in the filename when it really should be meta-data for the document. Also, your queries would become more obtrusive since you would also have to do:
% find . -name "*sophie*me*.img"
if you are searching for more than 2 attributes, this becomes a pain. Furthermore, the mechanism you describe would also retrived a document titled "sophie_in_mexico.img". So, simple regular expression pattern matching does not suffice for that type of search.
I would love to see good multi-platform render clients (nodes) for blender.
Also, being able to render to Renderman would be nice...current tools that support this are weak at best.
The interface, although weird at first, is perfectly fine. I have used 3DS Max, Lightwave, Softimage, and Blender and they ALL have different interfaces which take a little getting used to.
Weird MS Visual-C for-loop bug
on
Pet Bugs?
·
· Score: 1
About five years ago I was using MS Visual C and needed to perform some calculations which required very large for loops (>100,000 iterations)...the problem was that after the loop finished, what ever variable I was using inside the loop to hold the value of my calculation would get assigned to the last value of the loop iterator....but only for loops which had 100,000+ iterations...it didn't do this for small loops( 10 or 100 or even 10,000 iterations).
I ran it through a debugger and right before it would break out of the loop my variable had the correct value...but as soon as it left the loop, it would get assigned the last value of the iterator.
ex>
for(i=0; i 100000; i++) { somevariable = some computation }
You can't sell the code and give it away at the same time.
Sure you can....If you are the original copyright holder you can do whatever you want with it, including selling the code to some 3rd party with no open source requirements AND releasing a version of it under GPL or other open source license.
IANAL, but I think that is perfectly acceptable under GPL. NaN could, for example, release a version under the GPL and then turn around and sell the code base to anyone granting them whatever rights they wanted. But after that point, the two code bases basically become independent. Meaning that any developments made by the 3rd party after buying the code will not be required to be released under the GPL. And any developments made by the OS community can't be incorporated into the non-free code without honoring the requirements of the GPL.
It is worthwhile to note that Open Inventor is an Object-Oriented Toolkit for developing OpenGL applications. The Object-Oriented distinction is imortant as OpenGL is mostly Procedural.
Scenes are trees of nodes, where each node is either a renderable object, or a modifier for its subnodes. Scenes are composed for rendering by traversing the tree and examining each node for its contribution to the final rendered scene.
The unfortunate part of your argument is that you assume that if a child is born severely handicapped in some way...maybe to poor parents...then that child has NO expectation to lead a productive life and contribute to a better quality of life for themselves, their family, and/or humanity in general. No one argues that these children have a HUGE disadvantage and, statistically speaking, may not have a good chance at a good life. The problem is, statistics are just numbers and can't account for individual people. And so, to argue as to whether or not is wrong to kill a child because they have no [statistical] expectation to contribute to humanity pre-supposes that the child WILL NOT contribute to society...and that, dear friend, is a very dangerous thing because you are judging the longterm worth of a person and unluess you are clairvoyant, you can o not know how any individaul person will turn out.
That is true and I agree with you in full. However, the intent of my statement (which I regret I regret I did not convey it well) was not that users or developers have more or less of a right to defend the freedom of the software they use/develop. Indeed, I do consider the type of user you describe as "one in the same..."; I tend to be both a user of free software and a developer...I could not be developing the free software I am working on right now if it were not for a multitude of open source(free) programs.
In any case, the point of my statement was that there is a a very vocal segment of the Linux/free software community that simply jump at the chance of attacking anything that even smells like it might not be free software. I may go out on a limb and say that I believe a great deal of those people to be more interested in the free beer aspect rather than the free speech. My statement, again I regret not being more clear, was meant to point out that when we have a group of people that do not know about or care about the development of the system, just care that they can get it for free, and simply start yelling and screaming GPL this or GPL that....well it just sounds silly to me because I continually get the feeling that these people haven't even read the article that was posted or know exactly what the facts are, or even read the GPL before they start crying bloody murder.
I fully agree...I believe that the collective voice of our community is one its greatest assets. My comment on jumping the gun is simply to state that it is ok to be loud, it is even ok to be wrong...but not because we were too lazy to actually find out what the facts were.
Maybe some of the people that started yelling so loudly about this should have found out what exactly did Corel mean. People in this forum have a tendency to start crying foul without ever knowing the facts. I am not defending Corel or any other company or person, just please, before you start going off about this or that open source topic, find out for your selves what the facts are.
I would wager that the overwhelming majority of the people yelling bloody murder have not ever written a single line of open source code.
The phrase You got mail is ok because it states that in the past you received mail. In fact You got mail is similar to You received mail. The phrase you have gotten is not a phrase of possesion, but a phrase stating that in the past you recieved mail. The correct way to express possesion is with "have". "Get" and "Got" express the present and past actions of retrieval, not possesion.
That may be all good if the woman was using a legitimate Trademark like "AOL", but the phrase You've got mail has been used by postal carriers for years before AOL ever existed, at any rate, didn't AOL lose a lawsuit to AT&T over the same thing. It seems that AOL has realized that it can't win against the big players like AT&T and has decided to pick on someone little instead.
...IBM is too much of a control freak to allow something as important as an OS to be developped by someone else...
Maybe you don't know IBM as well as you thought. How do you think Bill Gates got to where he is today....That's right remember the whole mess from years ago when MS was developing DOS and their deal with IBM?
...IBM is too much of a control freak to allow something as important as an OS to be developped by someone else...
Maybe you don't know IBM as well as you thought. How do you think Bill Gates got to where he is today....That's right remember the whole mess from years ago when MS was developing DOS and their deal with IBM?
If this story is true....then it is a very interesting step for MS. But not necessarily an outrageous step. MS is riding the Windows wave... Just like 10 or so years before it rode the wave a a couple of little programs known as MS-DOS, MS-Excel and MS-Word.
If you think about it....you realize that MS goes on cycles....that is, their plans for their next big thing are in the works, but will not be implemented until, say, 2005...approx 10 years after win 95. Maybe it will follow the way of Java & Jini and have a MS alternative to Object Oriented, Distributed, network computing...just like they did with GUIs and Apple's Mac.
What ever it is, they can stand to do this because EVERYONE uses windows...specifically MS windows...so, if they open source windows, they will still maintain control of windows until their next big thing comes around. It may not be that big of a stretch to think that MS will release source code because at 35 Million lines of code for windows 2000, the only people who can really benefit from the release are the hardware vendors which write device drivers....something MS does not do.... and allow then access to specific hardware level areas of code to optimize the drivers. since these companies already are writng these drivers, it does not hurt MS to do this because the companies will compete with each other and not with MS. they already allow certain companies access to their source code under NDAs (Non Disclosure Agreements for those not in the know). this way they can offer the same thing and "look good" in the process it will not benefit software companies that offer products that compete with windows because the license will PROBABLY include clauses that require any application which makes use of the open software to release the modified source under the MS open source license..... which would then allow MS to take advantage of that in their own product.
Slashdot Mirror
I just wrote a java program to do just that...without printing each value to the console and it takes 0.218 seconds.
Actually, it does have major scalability problems...please refer to some of the numerous papers published on exactly this topic.
The supernode/ultrapeer addition to the protocol was meant to address some of the scalability issues of the previous version of the protocol. However, the fragile nature of the original overlay network is only made worse since ultrapeers are explicitly meant to be highly connected nodes.
The problem with Gnutella is that it can't do search efficiently due to the the broadcast nature of forwarding query messages. Research such as Random Walkers(SIGCOMM 02), and Attenuated Bloom Filters (IEEE INFOCOM 02) have tried to tackle the problem of scalable and fast search in unstructured P2P networks like Gnutella.
Sorry, that "solution" is not viable. That algorithm generates overlay networks with Power-Law node degree distributions. This means that the few super nodes have a very large number of connections while the vast majority of nodes have very few (maybe only 1). This means that the although the network can survive random failures, a coordinated attack on the most highly connected nodes (less than 3% of the total) effectively shatters the connectivity of the overlay network. This has been studied in several papers.
There are several P2P research projects that are looking at building reliabale and scalable P2P systems.
Take a look at Tapestry, and Chord (and read some of the papers) to understand the issues involved in providing scalable and high performance P2P services. Not only is scalable search and overlay graph connectivity an issue, but also node failure and short session times of P2P nodes.
Additionally, when you actually handle the issue of downloading data, building application-lvel multicast trees to distribute the data efficiently on a large scale is not easy. Two papers from SOSP '03 SplitStream, and Bullet address that issue.
Check out last year's SOSP proceedings for research on OS and hardware support for protected memory spaces.
% find . -name "*sophie*me*.img"
if you are searching for more than 2 attributes, this becomes a pain. Furthermore, the mechanism you describe would also retrived a document titled "sophie_in_mexico.img". So, simple regular expression pattern matching does not suffice for that type of search.
Check out A Logic Filesystem from this year's Usenix conference. It is a filesystem with precisely these capabilities.
I hate to split hairs...but a full version of Lightwave 7.5 (the newest version) is only $1,595 and NOT $5,000.
That being said, I own and use Lightwave and I agree, the interface takes some getting used to.
I would love to see good multi-platform render clients (nodes) for blender.
Also, being able to render to Renderman would be nice...current tools that support this are weak at best.
The interface, although weird at first, is perfectly fine. I have used 3DS Max, Lightwave, Softimage, and Blender and they ALL have different interfaces which take a little getting used to.
About five years ago I was using MS Visual C and needed to perform some calculations which required very large for loops (>100,000 iterations)...the problem was that after the loop finished, what ever variable I was using inside the loop to hold the value of my calculation would get assigned to the last value of the loop iterator....but only for loops which had 100,000+ iterations...it didn't do this for small loops( 10 or 100 or even 10,000 iterations).
I ran it through a debugger and right before it would break out of the loop my variable had the correct value...but as soon as it left the loop, it would get assigned the last value of the iterator.
ex>
for(i=0; i 100000; i++)
{
somevariable = some computation
}
somevariable=i!!!!
Weird.....
You can't sell the code and give it away at the same time.
Sure you can....If you are the original copyright holder you can do whatever you want with it, including selling the code to some 3rd party with no open source requirements AND releasing a version of it under GPL or other open source license.
IANAL, but I think that is perfectly acceptable under GPL. NaN could, for example, release a version under the GPL and then turn around and sell the code base to anyone granting them whatever rights they wanted. But after that point, the two code bases basically become independent. Meaning that any developments made by the 3rd party after buying the code will not be required to be released under the GPL. And any developments made by the OS community can't be incorporated into the non-free code without honoring the requirements of the GPL.
I think Watson and Crick did Acid and later envisioned the DNA double helix....for whatever that is worth...
It is worthwhile to note that Open Inventor is an Object-Oriented Toolkit for developing OpenGL applications. The Object-Oriented distinction is imortant as OpenGL is mostly Procedural. Scenes are trees of nodes, where each node is either a renderable object, or a modifier for its subnodes. Scenes are composed for rendering by traversing the tree and examining each node for its contribution to the final rendered scene.
The unfortunate part of your argument is that you assume that if a child is born severely handicapped in some way...maybe to poor parents...then that child has NO expectation to lead a productive life and contribute to a better quality of life for themselves, their family, and/or humanity in general. No one argues that these children have a HUGE disadvantage and, statistically speaking, may not have a good chance at a good life. The problem is, statistics are just numbers and can't account for individual people. And so, to argue as to whether or not is wrong to kill a child because they have no [statistical] expectation to contribute to humanity pre-supposes that the child WILL NOT contribute to society...and that, dear friend, is a very dangerous thing because you are judging the longterm worth of a person and unluess you are clairvoyant, you can o not know how any individaul person will turn out.
That is true and I agree with you in full. However, the intent of my statement (which I regret I regret I did not convey it well) was not that users or developers have more or less of a right to defend the freedom of the software they use/develop. Indeed, I do consider the type of user you describe as "one in the same..."; I tend to be both a user of free software and a developer...I could not be developing the free software I am working on right now if it were not for a multitude of open source(free) programs.
In any case, the point of my statement was that there is a a very vocal segment of the Linux/free software community that simply jump at the chance of attacking anything that even smells like it might not be free software. I may go out on a limb and say that I believe a great deal of those people to be more interested in the free beer aspect rather than the free speech. My statement, again I regret not being more clear, was meant to point out that when we have a group of people that do not know about or care about the development of the system, just care that they can get it for free, and simply start yelling and screaming GPL this or GPL that....well it just sounds silly to me because I continually get the feeling that these people haven't even read the article that was posted or know exactly what the facts are, or even read the GPL before they start crying bloody murder.
that's all
...............
I fully agree...I believe that the collective voice of our community is one its greatest assets. My comment on jumping the gun is simply to state that it is ok to be loud, it is even ok to be wrong...but not because we were too lazy to actually find out what the facts were.
Maybe some of the people that started yelling so loudly about this should have found out what exactly did Corel mean. People in this forum have a tendency to start crying foul without ever knowing the facts. I am not defending Corel or any other company or person, just please, before you start going off about this or that open source topic, find out for your selves what the facts are.
I would wager that the overwhelming majority of the people yelling bloody murder have not ever written a single line of open source code.
ok....
Flame Suit on........
The phrase You got mail is ok because it states that in the past you received mail. In fact You got mail is similar to You received mail. The phrase you have gotten is not a phrase of possesion, but a phrase stating that in the past you recieved mail. The correct way to express possesion is with "have". "Get" and "Got" express the present and past actions of retrieval, not possesion.
That may be all good if the woman was using a legitimate Trademark like "AOL", but the phrase You've got mail has been used by postal carriers for years before AOL ever existed, at any rate, didn't AOL lose a lawsuit to AT&T over the same thing. It seems that AOL has realized that it can't win against the big players like AT&T and has decided to pick on someone little instead.
Go away you troll....
Java's syntax is similar to C++ but its architecture is much more similar to smalltalk.
...IBM is too much of a control freak to allow something as important as an OS to be developped by someone else...
Maybe you don't know IBM as well as you thought. How do you think Bill Gates got to where he is today....That's right remember the whole mess from years ago when MS was developing DOS and their deal with IBM?
...IBM is too much of a control freak to allow something as important as an OS to be developped by someone else...
Maybe you don't know IBM as well as you thought. How do you think Bill Gates got to where he is today....That's right remember the whole mess from years ago when MS was developing DOS and their deal with IBM?
If this story is true....then it is a very interesting step for MS.
But not necessarily an outrageous step. MS is riding the Windows wave...
Just like 10 or so years before it rode the wave a a couple of little programs known as MS-DOS, MS-Excel and MS-Word.
If you think about it....you realize that MS goes on cycles....that is, their plans for their next big thing are in the works, but will not be implemented until, say, 2005...approx 10 years after win 95. Maybe it will follow the way of Java & Jini and have a MS alternative to Object Oriented, Distributed, network computing...just like they did with GUIs and Apple's Mac.
What ever it is, they can stand to do this because EVERYONE uses windows...specifically MS windows...so, if they open source windows, they will still maintain control of windows until their next big thing comes around.
It may not be that big of a stretch to think that MS will release source code because at 35 Million lines of code for windows 2000,
the only people who can really benefit from the release are the hardware vendors which write device drivers....something MS does not do....
and allow then access to specific hardware level areas of code to optimize the drivers. since these companies already
are writng these drivers, it does not hurt MS to do this because the companies will compete with each other and not with MS.
they already allow certain companies access to their source code under NDAs (Non Disclosure Agreements for those not in the know).
this way they can offer the same thing and "look good" in the process
it will not benefit software companies that offer products that compete with windows because the license will PROBABLY include clauses
that require any application which makes use of the open software to release the modified source under the MS open source license.....
which would then allow MS to take advantage of that in their own product.
HMMM.............