Slashdot Mirror
Bram Cohen's Response to Microsoft's Avalanche
An anonymous reader writes "Bram Cohen has reduced Microsoft's proposed file-sharing application--codenamed Avalanche--to vaporware, dubbing its paper on the subject as "complete garbage". "I'd like to clarify that Avalanche is vapourware," Cohen said. "It isn't a product which you can use or test with, it's a bunch of proposed algorithms. There isn't even a fleshed-out network protocol. The 'experiments' they've done are simulations.""
Sounds like Windows.
I cannot believe that we still do not use the Coral Cache and Bittorrent to prevent crashing poor servers. I mean, now we're just going to end up slashdotting that first link, uh, slashdot...
Forget it.
A NYC lawyer blogs. http://www.chuangblog.com/
Sounds like it, and the first salvos have gone back and forth... having read both, I have to give the points in the first round to Bram. Microsoft won't find him so easy to push around, methinks.
"It isn't a product which you can use or test with, it's a bunch of proposed algorithms. There isn't even a fleshed-out network protocol. The 'experiments' they've done are simulations."
Doesn't all software start off this way?
Unfortunately, I am not Wil Wheaton
"It isn't a product which you can use or test with, it's a bunch of proposed algorithms. There isn't even a fleshed-out network protocol. The 'experiments' they've done are simulations."
Another interesting project from the Microsoft team then? Looks like they've made an entire department, dedicated to making ideas about things that they could make, but never intend to do.. Nice going
Scully: Should we arrest David Copperfield?
Mulder: Yes we should, but not for this.
isn't that all that Microsoft makes anyways?
Go to the w3.org and put Slashdot.org through the validator.
Hi Bram, just to let you know that Microsoft Software Development Life Cycle works a 'little' different. We market it first, then release it, then develop it and then design it. And maybe, just maybe we might make it work if enough people are using it.
Hope that helps.
Sincerely,
Microsoft
PS: Note that I used maybe twice.
Free XBox, PS2
The question is: is it GOOD vaporware? Can the proposed algorithms deliver the results we want? That Microsoft has the manpower to turn it into real software is a given.
The illegal we do immediately. The unconstitutional takes a little longer.
--Henry Kissinger
According to Microsoft simulations performed just before this one, the Philadelphia Eagles are the defending Super Bowl Champions and Keith Richards has been dead for 30 years.
Does anybody have the torrent for the app?
But, yeah, like he said. Avalanche isn't supposed to take over the world. It isn't a product, and it doesn't exist in source code form.
Whoever corrects a mocker invites insult;
whoever rebukes a wicked man incurs abuse.
--Proverbs 9:7
This really shouldn't come as any surprise...after all, Microsoft's goal here wasn't to actually come out with a product, but to create the illusion of one. Microsoft will design a P2P system if and when they're good and ready...until then, Avalanche serves as a satisfactory decoy.
____
~ |rip/\/\aster /\/\onkey
Definition from dictionary.com for those like me who dont know what vapourware means. /vay'pr-weir/ (UK "vapourware") Products announcedfar in advance of any release (which may or may not actuallytake place). The term came from Atari users and was later applied by Infoworld to Microsoft's continuous lying about Microsoft Windows.
What's Adobe got to do with this?
Call me crazy, but what does this prove? What is the point to bashing this?
I can find no purpose to bashing a research paper ( per cohen ), especially from Cohen.
This is MS bashing, pure and simple. I'm sick of it. I'm not a fan of MS, I think they and their products are questionable at best, but needlessly bashing them instead of understanding their strengths is a fools' errand.
Mod me down with all of your hatred and your journey towards the dark side will be complete!
In other words, intentionally or not, the simulation is completely rigged against BitTorrent.
1) They don't want any other software to function at all, but since they know that a lot of users are wanting this, they want to jump on the P2P bandwagon and say "oh yeah, we have that". I guess they didn't want to buy any existing P2P apps, because of the bad publicity. So, the next best thing is "innovation" (i.e copying other's source code and changing the names of functions and authors).
2) put some DMCA crap in it and say, oh yeah, it is like totally unbreakable. 3) Profit!
I guess microsoft is just doing research, so they can patent their inventions. Those patents can than be used to make (other) fileswapping/p2p programs illegal due to patent infringement.
That's Bram Cohen, author of Bittorrent.
You mean like Longhorn?
Mercy was given to me by Christ...I must give the same to others.
"Unfortunately, [the paper] is actually one of the better academic papers on BitTorrent, because it makes some attempt, however feeble, to do an apples to apples comparison," he said.
Then Mr. Cohen patted Microsoft on the head, gave it a cookie, and sent it outside to play.
I want to drag this out as long as possible. Bring me my protractor.
Sharing of illegal material keeps P2P networks going. Obviously MS can't have illegal material on anything with their name on it. How does MS propose to turn this into a profit-generating venture?
Another pointless article. Troll me, but the fact is that this is addressing something that is behind MS's closed doors.
Ah, you mean like this research paper that Cohen is criticizing.
Or perhaps you are referring to these completely unfounded claims (from TFA):
The developer said Microsoft had completely misunderstood the way BitTorrent operated. The paper quotes "the tit-for-tat approach used in the BitTorrent network" as an inspiration for parts of Avalanche's own operation. Under the approach, a peer-to-peer client will not upload any content to another client unless it has also received a certain amount of content in return.
Cohen said, however, this was a waste of time and had been discarded long ago.
"I can't fathom how they came up with this," he wrote. "Researching either the source code or the documentation on the BitTorrent Web site would have shown that the real choking algorithms work nothing like this."
"Either they just heard 'tit-for-tat' and just made this up, or they for some odd reason dredged up BitTorrent 1.0 and read the source of that." BitTorrent is currently at version 4.0.2.
Cohen went on to say that the 'tit-for-tat' approach was used when BitTorrent was still being developed, but that the first real-world test with only six connected machines showed that it did not work well.
Yup, that's a guy bashing closed doors alright.
I hear there's rumors on the Slashdots
I think you confused marketers with researcher. :)
They are marketing their IDEA, not the actual software at this point. Sure, what they have done is research and simulations, which is obviously just one part of their software development cycle.
Microsoft has a huge amount of resources that they can and probably will pour into the p2p projects they are working on. It is foolish to mouth off and bash their development procedure, treating it as something other than it is. Microsoft has a strong track record of eliminating its competition by integrating products into its OS. Dont be too suprised if you see Avalanche as part of Longhorn.
"If we hit that bullseye, the rest of the dominos will fall like a house of cards. Checkmate." -Zapp Brannigan
"The term came from Atari users and was later applied by Infoworld to Microsoft's continuous lying about Microsoft Windows."
;)
You are a hoot!. You quote the dictionary and then add in your own terms. I looked it up myself and could not find anything in refernce to lying and Microsoft. hmmm. Perhaps it's your browser misinterpretting the definitions found there?
Are we supposed to take this "article" any more seriously than a pro-Microsoft "study" by some Microsoft shill?
Slashdot is turning into the Jerry Springer of tech news.
"My P2P architecture is soooo much better than your P2P architecture, bitch!"
"Oh no you didn't!"
So, when will we get to see the topless crack-whore midgets?
... does this have the obvious meaning of Microsoft 'burying' its competition? heh
BT is relatively new, I am sure within a few years some serious inadequacies will be found which will make this research from Microsoft more significant.
BT is NOT relatively new - in fact, it's relatively old, and there HAVE been a few years for any "serious inadequacies" to surface. What has happened in those years is that users of other P2P networks have flocked to BT by the millions, simply because it works much better at delivering maximal bandwidth for highly sought-after files.
The illegal we do immediately. The unconstitutional takes a little longer.
--Henry Kissinger
A long time ago in the 1930's, a british fellow named Alan Turing came up with some ideas for a fantastical machine that could automatically compute mathematical formula. Unfortunately for those whose lives the "computer" might have changed, he was immediately shouted down by the trolls of Slashdot shouting "Where's the implementation?!"
Announce early, announce often.
Microsoft, on numerous occasions, has indicated that they have a(n) [ insert competitors product ] killer... with a cool code name and features that look very appealing. We find out months/years later that their product either doesn't materialize or doesn't deliver on the original specifications. Sure the 2.0 or 3.0 version might, but my point is, they fend off competition by using vaporware.
I'm tired of it... I'm moving back to my TRS-80
Longhorn is just one prime example. I wonder how many people didn't consider switching over to Linux/?nix/OS X/etc. etc. because of the overly hyped features of Longhorn... which now are disappearing left and right.
It takes years to make something like bittorrent, but it takes days for a marketing team to come up with a flashy code name and feature list.
Perhaps their research algorithms could find their way into something real.
Microsoft might just release it as an API, without releasing an easy to use client of their own. That way they don't promote piracy, but they can reduce the bandwidth needed for software updates and assist adware producers in delivering full motion full screen ads to unsuspecting users.
Bit torrent is 4 years old. They have been testing various configurations for the last four years.
I think this guy knows a bit more about how an effiecent torrent is going to work.
MSFT is once again playing catch up. In a few years they are going to end up duplicating the entire effort of bram just ot make a closed source version of the software, which will then fizzle out because msft won't make clients for anything other than windows. Yet Torrents can be had for any OS.
i thought once I was found, but it was only a dream.
The rest of the screed we can only react to instinctively, okay -- but the one piece of relatively concrete evidence, that code name, shapes that reaction pretty early on, doesn't it?
"Avalanche," as a name for a product or project, would be just about the worst possible choice. As a P2P tool that would imply bandwidth problems and the potential for a single point of failure. Does someone somewhere have a positive feeling about this name? Coming from blinkin' Microsoft?
I'm trying to think of a product I would happily buy with the name "Avalanche" on it. A quick Google doesn't get me much. Team names.
"Fundamentalism" isn't about divine morality. It's about human authority.
So let me get this straight you are siding with qualified and certified researchers of the creators of the biggest virus spreader in the world. Instead of an opensource programmer who has does nothing but help the internet. get your head out of your ass.
IMAGE VERIFICATION IS EVIL!
Good. The mud has officially been slung. We are in for a hell of a fight, it seems.
The "Avalance is vaporware" vibe is a true one, but let's give Microsoft a chance for a real-world test before we cast our lots. Not completely dismissing the paper demonstrates Cohen in a more rational and less infuriated moment, and is fortunate that he did so, as industry leaders who dismiss competition get burned all too often. This is not to defend the test model in the slightest, which is junk and atypical of typical Bittorent usage as Cohen rightly points out.
The Avalanche paper is a start. Microsoft will need to finish, refine, and check their facts about the product with which they are competing. The idea of building a file without all the pieces reeks of difficult implementation, for example.... that's one protocol I would love to see come into reality. Bittorent will need to flex and build upon the established track record of the protocol, and innovate on top of that. Decentralized trackers were a good step.
The Crimson Dragon
I just took a look at the MS paper and they propose using "network coding" to eliminate block scarcity. The idea is that each node transmits out a linear combination of the blocks they've downloaded, along with the coefficients used. When enough of these are gathered, the system can be solved and the file reproduced. This actually seems useful and is not "complete garbage".
Yeah, I know most of the posts here are bashing Cohen for bashing microsoft (and I was agreeing with them until I decided to RTFA - the summary is not a good one). Bram isn't randomly attacking them for having a vaporware product, he is specifically pointing out the many mistakes that they make in their paper, where they compare Bittorrent to their proposed algorithms. It seems that they made too many mistakes to make their research paper valid, so their simulations are crap... RTFA!
The AACS key is NOT 0xF606EEFD628B1CA427BEA93A9CA9773F
Or perhaps you are referring to these completely unfounded claims (from TFA):
I agree, he has a right to comment on that.
Yup, that's a guy bashing closed doors alright.
But to call the Avalanche research vapourware because he hasn't seen an implementation *is* bashing MS's closed doors.
I read the article too. Perhaps you put weight on papers and supposed conversations that they do not deserve. Cohen constantly quotes conversations that he has had, with no documentation to back it up. He points to papers. You point to slides and a couple of paragraphs of text that even MS states are preliminary presentation materials for a project proposal. You are being ridiculous on two points. One is trying to state how awful a product is, before it is a product, before it is even being coded to become a product, by your own documentation. In any other arena, you would be laughed out the door for that alone. The second is that you are challenging MS's own statements about something they admittedly are not working on currently. What is the point? Just to pick nits? Is your hatred of MS so deep that you can't see facts anymore?
Show me some real code, from the real source, and I will stop making fun of you. Otherwise you are making it far to easy.
I know I am resonding to a possibly trolling article, but I try to be rational.
Yes, the original article is a bit blunt, but we should ask, is this bad ? It seems to me that he pointed out why the Microsof work is "useless" and therefore the point should stand, even if not "politely" put.
Responding to the post, my opinion is that big corporations do not have a monoply on intelligence and therefore it is possible that something of "not high quality" is produced. Would it be strange if it happened in this particular case ?
On the inadequacies of Bit Torrent, why is Microsoft keen to find inadequacies in other people products ? My idea is that this is a way to prepare for future Microsoft products that are "better" since they do not have those issues. (They will have many others "problems", but that is another story, not to be told)
What Cohen and other /.-heads don't get is
that Microsoft Research
is not Microsoft. Sure, it's owned by
the evil Redmond company, but it's the
research arm.
.NET, but pure 100% theoretical
language theory research.)
Basically, it's one of the coolest places to work. It is not a product development arm of microsoft. They are one of the largest employers of theoretical computer scientists around. (Theory compsci jobs are hard to get in industry--usually only labs hire them, or academia.) If theoretical CS is what you do, the MSR (and not MS) is even better to work at than Google. (One of the other fields MSR dominates in is compiler research--not VC++ or
So, of course the proposes program is just a bunch of simulations. That's what they do, moron!
Sounds like Bran either needs to read more or adjust his ADHD medication.
Classic mistake. He has now given MS credibility in this market that they would never have gotten on their own.
Research papers that are released, even by MS, usually aren't intended to get the attention of the broader market. They're intended to put forth ideas and let a few select people know what's going on inside the heads of MS engineers. That, and to raise the ire of slashdotters who don't read them and choose to automatically assume that MS persues every single research idea as a major project.
Look at it this way - MS can't afford not to be looking into the area of filesharing, because it's obviously something that their customers really, really want. There hasn't been any announcement of any product, there's just a whitepaper with no details. The not-so-sinister truth is that this research paper is just evidence that they are starting to think about the problem, not a representation of an imminent product offering.
-- Truth goes out the door when rumor comes innuendo. -- Groucho Marx
Damn the arrogance of that Einstein.
A man more right than his neighbors already constitutes a majority of one. -- Thoreau
KFG
Since there's no way that Microsoft could fast track this project, put about a thousand engineers on it, have a working beta by Christmas and have a full client downloading onto 50 million Windows XP machines from WindowsUpdate.com before good old Bram could say "What's WinAmp?"
Yes I know, the above scenario is complete BS and all of you would keep using Bittorrent anyway. The point is that, regardless what you think about them, MS is still a very large company with a very large amount of clout and a staggering amount of resources at their disposal. If I'm Bram I don't think I would be waving a red hankerchief at that particular bull.
2) The paper also assumed that each client would only try to connect to 4 peers. Bram says that 30-50 is more realistic.
3) In spite of the poor comparison, the ideas might be useful.
The actual blog entry
I don't see the point of flaming Microsoft here, they've done nothing wrong or evil. It seems Cohen is unfuriated because they claim to outrank bittorrent based on simulation and he goes on the old "real life is not simulation" blah blah rant. Having rtfrp from Microsoft I do find that there are many clever ideas, especially the linear combination part which also has the particularity to provide some kind of stealth on the data beeing transmitted. If Microsoft's paper doesn't give enough credit to bittorrent who cares ? ( ok Cohen does ). The important thing is to profit from the possible enhancments they offer. Come on, if google was the one proposing a swarming ft technology, everybody would be like whaaaa... As much as I hate Microsoft business methods and products I don't think ms researchers are evil at all.
\u262D = \u5350
Windows Media has been prepared for several years as a leading format for use in pay-to-view downloaded material. Microsoft even developed Windows media centre to run TV-connected PCs.
What's missing is the distribution technology. Even with modern 8mbit DSL / Cable connections, an HTTP or FTP download of a 900mb movie file is very expensive for the company hosting the software and files. However, if each set-top-box or WMC PC has a secure file-sharing system preinstalled, then most of the upload bandwidth can be shared among people who have already downladed the same file.
Consumers will hate it - especially as upload bandwidth is often slow and overall bandwidth capped - but the media distributors will love it to bits.
I have been a user for about 10 years. This ends Feb 2014. The site's been ruined. I'm off. Dice, FU
""Avalanche," as a name... ...would be just about the worst possible choice. "
Since when has MS ever started a project and released the code under the project name? Longhorn? Whistler? etc. etc.
Show me where you can find a single MS product that is for sale under the project name it was serviced under within the walls of MS. You can't, because that isn't the way they do things there . I am not sticking up for MS, just trying to remind you of the facts. It's just not the way they do business. Good bad or indifferent, your arguement holds no merit in light of history and current facts.
The 'experiments' they've done are simulations.
I've been saying that about Windows. My friends won't listen.
Avalance is what happens when a lot of drag and dropping is going on.
It takes a man to suffer ignorance and smile
Be yourself no matter what they say
RTFA! It's still there people! But if you haven't RTFA and want to spout off something please keep in mind the following: He invented Bittorrent, and his biggest problem is this "research" is based on massive misunderstandings of how Bittorrent actually works.
So, lay off! :)
You are not the customer.
...prior art - even a borderline retarded patent examiner (unfortunately the norm from some granted patnets I've come across) will have heard about p2p.
I am NaN
Oh the irony... Read the article and then we'll let you play again...
The AACS key is NOT 0xF606EEFD628B1CA427BEA93A9CA9773F
"Avalanche," as a name for a product or project, would be just about the worst possible choice. As a P2P tool that would imply bandwidth problems and the potential for a single point of failure.
Um... Would it? Why? I don't see these implications at all. Quite the opposite, really. An avalance is (in popular imagination, anyway) started by a small cause and quickly develops into an unstoppable mass of snow. Just like a single limited-bandwidth uploader of a popular file to a P2P network can result in many Terabytes of data being moved.
Quite similar to the "Torrent" part of BT.
The illegal we do immediately. The unconstitutional takes a little longer.
--Henry Kissinger
BitTorrent Web site would have shown that the real choking algorithms work nothing like this.
Illegal file swapping uses asside, one of the advantages that I had expected BitTorrent to provide was increased download speed. I expected that the use of multiple download sources would allow me to saturate my connection rather than be at the mercy of a single FTP server's connection. What I have experience repeatedly, version after version, with BitTorrent has been quite different. I have experienced slow connections, at least much slower than I would expect and in some cases, slower that a single FTP server. Now we see the problem. It is because of the "choking alogrithm".
Now, don't even bother offering your comments on configuration of BitTorrent or the firewall, I know what I am doing and I know that the configurations were fully optimized. The fact is that BitTorrent consistently under performs because of its "choking algorithm".
This choking algorithm may be beneficial in the sense of increasing the total connections that a seed will accept but it robs the system of its performance benefits. So what is the advantage then? Provide slow connections to the largest possible audience or provide better performance? The answer seems to be the former, which is very advantageous to the illegal P2P crowd but doesn't really help the legitimate users. There's little or no benefit in being able to connect to an ISO download when it will take a week to complete the download. It is better than waiting a week for a connection and then downloading the ISO is 15 minutes!
But, Cohen doesn't want to here this stuff. In fact when presented with these ideas, he reacts like a mental patient! It is because of Cohen and his arrogance or mental disorders that MS Avalanche will emerge from its vaporware as a vastly superior system. It will of course also include DRM so it won't help the P2P crowd very much.
Vapourware is the act of annoucing features that you don't intend to ship in order to harm the release of a competing product.
"Yeah, Microsoft's version will have all that and more in a couple of months" is enough to put people off buying an application temporarily, and by the time Microsoft's version is released, they've forgotten about the competition.
Example: "Windows 95 will be uncrashable because of 32-bit memory protection". How many people would have switched from DOS+Windows 3.1 to something like OS/2 if they had known it would take another six years for Microsoft to release a halfway stable desktop OS?
This is not vapourware because announcing a couple of slightly improved algorithms is not an attempt to kill Bittorrent.
Avalanche Peppermint Schnapps.
Tastes and smells just like mouthwash, which makes it very easy to hide in, say, a bottle of mouthwash. Not bad when mixed though.
Since when is Dictionary.com an accepted standard for any language. It is not. Go back to the books my friend. Here I will even help you. Lose the conjecture and the blogish references from you arguements and you might learn how to win.
n ary&va=vaporware&x=0&y=0
http://www.m-w.com/cgi-bin/dictionary?book=Dictio
http://www.bartleby.com/61/64/V0026400.html
Just a couple of standards accepted by most colleges and schools in our country. Wiki's do not count, sorry. You cannot use sources that take user submissions as source material and call it fact. There has to be certifiable work behind it to prove it. Dictionary.com does not have neither the work nor the certifiable material. Just their own thoughts mixed in with the submissions of users. That doesn't make it fact, lol. Just as the sources for the article under discussion are flakey, misrepresented and full of conjecture, so are your references. Perhaps you think one bad reference can be supported by another?
Anyway, the global idea was to add some Par2 volumes to the files, so that someone close to 100% would be able to reconstitute the missing bits, and then have a 100% copy to offer on the P2P network faster.
Wasn't that a good idea ? Of course, Par2 volumes mean downloading some more bits. Thus, to be advantageous, downloading these supplementary bits has to globally lead to a faster download than the classical bittorent protocol.
I am not a specialist, but that would be interesting to test, and I do not think this is that difficult to implement.
Don't think of it as Avalanche v. Bittorrent.
Think of it as Microsoft v. MPAA/RIAA
There's no place like ~/
I just took a look at Cohen's article and he responds exactly to that proposal and elaborates why it does no good and can actually be bad.
RTFA
The implication is that Microsoft is doing a lot of talk with little follow through.
Yes, I know what it means.
The paper is from the IEEE Infocom in March. It's three months old. Sure, if it were a year old you could call it vaporware but I think it's a little early to call names just yet.
Sounds like the real issue here that Cohen is scared to death at the notion of MS challenging BT.
Maybe make your product better in response, instead of this childish outburst?
Nawww, that'd be silly...
Ooh. So Cohen's got an ego... He's arrogant... He's an asshole...
So what?
You don't have to live with the guy. You don't have to work with the guy. All you do is use his software. So why the fuck do you care what he's like as a person? If he spends his spare time sucker punching old ladies, what does it matter to you?
This is not first time M$ is trying to steal other's ideas, and create FUD about original product. They have tried it with Office, LAN servers, Internet Explorer, SQL server, Instant Messenger, Java, IP (4 and 6 both) and recently iPod and now BT.
I can understand Bram's fear, if he thinks M$ is after his ideas to steal them and kill him.
hilarious
Jeeeez. Is someone a little defensive of his turf? (It's not even like Cohen invented the idea of peer-to-peer software delivery.)
Microsoft Research, and many other research labs and universities, publish papers on "vaporware" every day. Only, this is not vaporware because it is not supposed to be a product--even if some news media who don't know the difference between Microsoft and Microsoft Research make that mistake, Bram and others should know better. On the other hand, research on algorithms is fundamental to the development of the next generation of products, because no amount of pure coding can make the kinds of technological leaps that are necessary. To that end, it behooves us not to bash it, or at least only to evaluate it based on what it is.
A Microsoft employee using LATEX?. This is bad advertizing for MS Office. Well, the other author of the paper is from a University, he has an excuse for using LATEX.
At least the presentation is in PowerPoint.
Its not vapourware, in fact its going be desinged to use WinFS.
> "Either they just heard 'tit-for-tat' and just made this up, or they for some odd reason dredged up BitTorrent 1.0 and read the source of that."
... see this is why corporations HAVE corporate communications departments, to keep people from jumping the gun with embarrassing announcements like this. Though, did MS announce it, or did some random doofus just dig it up from MS research and start screaming like a nutter about it?
Considering that the only up-to-date documentation of the protocol IS the source, one is not surprised. The only paper of Bram's that really details the protocol refers to version 1.0.
Still, you'd imagine Microsoft would have a fellow or two who can read C or python (is the reference implementation still in python?) Contamination might be an issue, but you'd also imagine they could just cleanroom another implementation of their own prototype if they were serious about productizing it.
Nope
I am no longer wasting my time with slashdot
How is this offtopic? 1) the article summary is not very useful 2) that is his picture 3) it lets people know who he really is.
I would say Bram really knows his algorithms, his dad taught me CIS 435: Data Structures and Algorithms at NJIT. He was a good teacher.
- my userid is lower than yours
Add PARS to the Bittorrent protocol. From what I can tell, that is the only new thing proposed for Avalanche, except for DRM. Steal their good idea(s) for your own and get it into production first. Patent it, and then sue them if they put theirs into production.
This is from Microsoft Research, not from Microsoft. Although they share a name, and some of the same goals, they are different entities. It really has nothing to do with MS Engineers.
Fair enough..... But microsoft does like to be a part of anything that gets big.... Bram is always working on bittorrent, improoving new releases.... adding new features.... etc. Its a bit annoying that microsoft buts in by making a client that is supposidly meant to be a bit more efficient than bittorrent..... If they are going to do something like this... do something original.... instead of just copying someones idea then trying to make it slightly better. If i were bram, i would be annoyed as well.... especially if they started shipping this new client with the next release of windows
Have any of you ever given thought to the fact that MS isn't stupid. If I were MS, I might release bogus specs on a vaporware product so others might develop something similar. I then develop a superior product and then eclipse everyone else.
Despite MS having bug-ridden software that is the target of hackers and jokes, they are still a damned powerful company, and like it or not, will be likely be around long after you and I are dead and gone.
He's allowed to be ballsy on this issue: it's a direct competitor to a product that he pioneered.
If someone challenged Matthew Lesko (the crazy Riddler wannabe) on his government money techniques by publishing a summary of the contents of their new book, you can be sure as hell that Matthew Lesko is going to read that summary and evaluate it - afterall, he could integrate some ideas into his book since such material discovery and reporting is basically journalism. On the other hand, he could turn around and blast the newcomer for attempting to make people see greener pastures where there are none.
Bram has every right to comment on that paper in any way he sees fit, and I'm not quoting the USA First Amendment, either.
Colin Dean Go a year without DRM
Maybe Microsoft could make better products instead of planning on the patent system giving them free reign to crush competitors?
Nawwww, that'd be silly...
This is from Microsoft Research, not from Microsoft. Although they share a name, and some of the same goals, they are different entities. It really has nothing to do with MS Engineers.
Except presumably MS own all the IP, and can take the ideas and code if they want. (For example, MSR implemented Microsoft's IPV6 stack that is shipped with XP and 2003.)
And presumably MS steer MSR's research, and MSR are given more time to come up with these ideas than MS engineers assigned to a shipping product.
So...he often flys off the handle with comments that normal people might not make...but then again...Theo from OpenBSD makes similar types of comments about stuff and he IS'NT Autistic so it makes me wonder....
The only thing he's critiquing is the methodology of the comparison to Bittorrent. But then, the paper's primary goal was to introduce the Avalanche technique. From my read of it, the authors were providing the Bittorrent comparison as a means of introducing some context to the paper, NOT as a head to head comparison. Adter all, Avalanche is by no means mature yet, so such a comparison can't exist yet. The Avalanche technique is very insightful. It is mathematically elegant, and looks to be a very effective way to solve the slowing of downloads near completion (especially when there are few seeds left). This is a well known problem. The authors of the Avalanche paper were very forthcoming about the costs of Avalanche: there is a significant processor cost to their technique, as they have to invert a bunch of linear equations, whereas Bittorrent chunks are all orthogonal by definition. Asd far as I can tell, the Bram article is a reactionary zeaolt's rant and contains no useful critique (see my first paragraph). The methodology that Bram is criticizing, while valid, has nothing to do with the (commonly recognized) problem that Avalanche can solve. Whether tit-for-tat is used or some other scoring algorithm has nothing to od with the "last few pieces" problem that Avalanche solves. If Bram had picked the "locally rarest" methof to attack, then there might have been some hope of the Avalanche paper's conclusions being tossed out. That not being the case, I'd have to say it looks like Bram is unintentionally giving the nod to Avalanche.
argumentum ad fallacium: Fallacy of defining a fallacy which allows one to dismiss the argument in question.
The more M$ fails at jumping into a technology just because they can the better. "Cool! File sharing! Let's do that!" They've already fscked up the web browsing. Hopefully filesharing has been spared some of their antics. There's enough to deal with already with MPAA and RIAA.
Has Mr. Cohen ever read research papers before? If so, and his response is always this way, then he must be a very angry man! :) Research papers are often just algorithms and simulations. Especially when it comes to large scale networking research. Welcome to the world of academia, where it's not about the code or real life performance, but how your stuff holds up to mathematical analysis.
Microsoft's biggest problem is end-user resistance to all their crap.
:-) and it may be a bust because nobody will give a shit.
There are still '3.11 systems out there.
There are still '95 systems out there.
There are still '98 out there.
There are still NT systems out there.
There are still Win2k systems out there.
If they couldn't coerce the OEMs into bundling the latest crap, it probably wouldn't 'sell' at all. People PAY Apple $129 directly for each version of OS X. I wonder how many people PAY Microsoft directly?
Longhorn may be everything to everyone (as if
They'll 'sell' it the same way they've 'sold' every other system they put together, they'll bundle it with new hardware (and MIS shops may trash it because they 'roll their own' and need to control exactly what the user sees,) but that only works if the hardware churns as fast as it did before.
(The old versions are there because the hardware still works, doing what its supposed to and nothing more.)
MSBPodcast.com The opinions expressed here are my own. If you don't like 'em... Think up your own stuff.
As they're cutting features left and right, you expect them to build, integrate and test something that's just been released as a *proposed algorithm* into Longhorn at *this* stage of development? Are you on crack?
Bram may be right about Microsoft's paper, but he would have had more credibility if he had taken the high road.
Quotes like "The lack of any concrete numbers at all shows the typical academic hand-wavy 'our asymptotic is good, we don't need to worry about reality' approach" certainly don't earn him much respect from academics in system programming research who work very hard, thankyou very much, to ensure that their results are realistic. He has turned a simple observation about the paper (they neglected certain overheads) into a bigoted rant (academics are foolish). Not cool.
Patrick Doyle
I mod down every jackass who puts his moderation policy in his sig. Oh, wait a sec....
At the same time, I really doubt that the largest software company on the planet, that has at least some of the brightest software engineers in the world, is going to have a hard time figuring out BitTorrent works. Even if they don't get it now somehow, I figure they'll work it out. Besides, aren't we talking about something that's not even in development yet?
I don't respond to AC's.
> So, of course the proposes program is just a bunch of simulations. That's what they do, moron!
They are simulating BitTorrent entirely wrong (RTFA!), so the value of the simulation is almost equal to zero.
- sigs are for wimps.
The developer said Microsoft had completely misunderstood the way BitTorrent operated. The paper quotes "the tit-for-tat approach used in the BitTorrent network" as an inspiration for parts of Avalanche's own operation. Under the approach, a peer-to-peer client will not upload any content to another client unless it has also received a certain amount of content in return.
Someone may want to update this paper http://www.bittorrent.com/bittorrentecon.pdf that begins:
Gotta love those "qualified and certified researchers" from MS. That statement kind of reminds me of "certified MS professionals".
Personally, I'll take results over certifications anyday. Can you show me one single thing out of Microsoft Research which has been bigger than Bittorent in the past 4 years? I can't think of one single thing.
The best way to predict the future is to create it. - Peter Drucker.
Actually, even as a research paper, the paper falls rather short. There has been prior work on P2P using forward error correction (FEC). FEC also means that you don't have to get every single chunk, but that you can reconstruct missing chunks from data you already have.
The authors should have demonstrated that their approach is better than FEC-based P2P protocols, but instead, they only compared it to simple P2P protocols. So, their protocol may actually not be better than the state of the art at all, and may actually be harder to implement in practice.
...than their promise to have a Linux version of Windows Media Player running. That piece of vaporware nothingness is over 2 years old (my blog entry on it is dated 4/30/03).
so having "experiments" is still more than they have for some products they've announced over the years...
"But remember, most lynch mobs aren't this nice." (H.Simpson)
-- Joe
I have to wonder why Microsoft would be interested in P2P technology. But perhaps they are following the precedent of the BBC's Interactive Media Player in finding efficient ways to distribute their own product. For example, Windows Software Update might work very well as a P2P network. Also, with their DRM attached to all files on the network, and the P2P protocol built into Windows Media Player, it would then be very easy for them (or their affiliates) to open up an online video store offering very fast downloads. However, there's also nothing to stop anyone who doesn't have an inbuilt bias against open source from doing those exact same things today. So I'd expect Apple to beat them to the punch - again.
--- Yx3 = Delilah ---
I work at a corporate F100 R&D lab, most of what gets written up is either
-prescriptive stuff that you'd like adopted
-things you built that you'd like the world to know about.
For a corporate group, a paper is only a half-success, depending on the ranking mechanism. A Popular paper is good, but not as good as getting into shipping product. And there MSR have the same problem I have -the gulf between research code and production stuff. Actually, their problem is worse, they have to go through the MS lifecycle, whereas our codebase is now open source (http://smartfrog.org/) so that we can have stuff in users hands in real time.
Summary: the presence of an MSR paper on its own is meaningless.
returning to MS and P2P, note that MS own groove, which has an excellent P2P filesystem, though one that will forever be windows only. They already do P2P products, they just are not as common as say, Exchange Server.
Aren't we busy trying to change the patent system from "first to invent" to "first to file"?
It doesn't matter if Microsoft was first at inventing X. All that matters is that they are the first to file for a patent. (Costing tens of thousands of dollars for each application filed.)
For suitable values of X...
- Avalanche (ala BitTorrent)
- Office documents stored as XML files inside a Zip file
- Using XML to serialize objects
- etc.
The lesson here is that even if BitTorrent was four years earlier than Microsoft, the fact that BitTorrent cannot spend tens of thousands to file a patent still gives Microsoft the opportunity to be first to file for a patent. Being first to file is what matters in the future.I'll see your senator, and I'll raise you two judges.
I met Matthew Lesko. He was at my friend's wedding. He's not crazy, but he did wear his suit. :)
Unless mankind redesigns itself
Yup. Bram's blog makes it clear that MS's approach to P2P is to reverse engineer an outdated version of his protocol. The result will probably be similar to MS's effort to reverse engineer the Macintosh: Windows 3.1 -- meaning that it will be buggy, slower and vastly more popular.
Keep reading Bram's blog. How far do you get before it starts going over your head? The dude has skillz that dust 98% of the wannabes here on Slashdot.
And as for motives, in my experience with autistics, it's common for those with Asperger's Syndrome to be quite guileless. They speak and act without consideration for other's "feelings". As a result they are more frank and honest than most people are comfortable with. Sorta like if Mr. Spock insults your work. He's not doing it to hurt you, or out of jealousy, he's saying it because it is the most logical observation.
As I recall, the paper claimed a 20-30% speedup compared to plain P2P protocols. But there are other, known ways of speeding up plain P2P protocols. Since people haven't even bothered with those in real P2P implementation, it seems implausible that Microsoft's protocol translates into a practical advantage anybody cares about.
When you first configure files to publish with Microsoft P2P software, Clippy(TM) will hop out and say "Your MP3 files are not optimized for transition - converting to WMA now..."
Later, when some incoming MPG files have arrived, Rover will trot out, and lift his leg on the MPG icon and magically stream it into a WMV for you.
And, of course the whole file shareing experience will be fun and safe because it will only talk to other Longhorn machines on which the integrity of the files have been verified...
This issue is a bit more complicated than you think.
his understanding of their coding scheme. He states:
"The central idea here is basically 'Let's apply error correcting codes to BitTorrent'"
and even admits that perhaps they are not strictly error correcting codes. But he does not understand simple linear algebra apparently. Reading the paper shows a useful method to generate new blocks without specifically being sent the data. And mathematically it does work, and the algorithms are very efficient computationally. He complains that they did not do rough calculations and says this is a flaw, but he shows no "rough calculations" to show their method won't work.
Anyone with some knowledge of solving linear systems and how efficient it is over finite fields will see that the proposed method is indeed fast and useful.
Cohen should rethink his rant.
Imagine what a wonderful selling point it would be, both to the copyright industry and to congress.
If Avalanche is DRM'ed, then it would only exchange content that is not known to be infringing. The RIAA could sleep safe at night, knowing that they are one step closer to their goal of extracting a payment from anyone who even overhears any of their music -- because the neighbors have it turned up too loud.
Congress could now pass a law outlawing other types of P2P. Why allow all this unauthorized stealing of files when Avalanche can ensure that "safe" Internet sharing.
In fact, they could do one better. Make Avalanche only exchange DRM'ed content; only share content that has been authorized to be shared -- by a registered (and paid) DRM content holder member.
Now they've stopped the sharing of, say, Linux ISO images, and reduced that 35 % and growing network bandwidth, and stopped piracy, at the same time!
And...
3. Profit
I'll see your senator, and I'll raise you two judges.
I recall that a past article discussed Brams Asperger's/Autism tendencies. The presence of this trait in anyone would lead me to believe that what he states is pretty much legitimate. Folks with Aspergers/HP Autism don't typically react defensively. That of course and the fact that he is a decent, intelligent person and could pretty much be regarded as the leading authority on this type of technology.
What does that expression refer to, anyway: To the invention of bread (and of knives), or to the concept of pre-sliced (i.e. before you buy it) bread? I always found that expression a bit funny, especially since pre-sliced bread is (IMHO) a terrible idea! Makes the bread dry and tasteless... Atgeirr
anybody - Microsoft has done this for years and will still continue to do it.
Bram mentions that a malicious user would be able to poison an Avalanche based system by uploading bogus data. BitTorrent gets around this by precomputing checksums on each chunk of data and storing them in the .torrent file. Why would this approach not work for Avalanche? You would have to precompute the hashes on the FEC codes instead, but I don't see where it is fundamentally different than BitTorrent.
I read the internet for the articles.
Microsoft's standard operating practice is to release vaporware of a product long before any such products exists. They used to call it slideware because the product existed only on powerpoint slides. The idea is to choke off the air supply (ie: revenue) of any competiting company by giving people an excuse to wait for the Microsoft product.
There has also been some discussion about all this n the p2p-hackers mailing list.
Research papers that are released, even by MS, usually aren't intended to get the attention of the broader market. They're intended to put forth ideas and let a few select people know what's going on inside the heads of MS engineers.
I think you're right. If the original paper is just a research paper, then there's no particular point in claiming that it's another vaporware campaign.
On the other hand, Cohen also offered explicit reasons why it was a misleading research paper. In particular, the analysis of BitTorrent was simply wrong, both in explaining the design (tit-for-tat, says MS, like hell, says the author) and in guessing the typical environment (4 to 6 connections, guesses MS, more like dozens, says Cohen).
So, even if we ignore the apparently spurious claims of vaporware, Cohen's article has real criticisms of the paper.
Phiwum's law: anyone that names an obvious law after himself and then puts it in his own sig is just pathetic.
Given that Microsoft keeps peeling pieces out of Longhorn, I find it difficult to believe they will integrate an as-yet unwritten P2P service into Longhorn, and still ship this decade.
Exactly how is this troll?
Nintendo's 1080:Avalanche is one of the best snowboarding games available on any platform. But it didn't do particularly huge business, so clearly some people agree with you.
"I Know You Are But What Am I?"
Did I ever state that my language was strictly per a dictionary? Never. I don't recall stating that I was a scholar or making a speech that needed it's spelling and grammer checked. To even suggest that what someone states is illegitimate because of spelling and or word usage only points to the fact that you haven't an argument to support your cause yourself. It's a childish way to divert the point at hand. Get over it.
They are doing something original. They're doing network coding, so each node can get useful information from more nodes than is the case with bittorrent. Bram Cohen is a blowhard.
Try reading the correspondence of the Royal Mathematical Society during the conflict between Leibnitz and Newton.
Bram probably does understand the strengths Microsoft has. He's adressing several specific questions that people have been asking him.
I'm not sure what it is that makes Slashdot commentators react so poorly to anything which is posted and which is not an authoritative and correct article on an established set of facts. Is it that geeks really think the world works in such a way, that the truth is there, and a simple dispassionate summary can always be written?
Bram's *job* is to be the architect of a certain P2P solution. This is an established role that he has. As an authority in the field, it is his responsibility to comment on new developments in the field.
In this case, misreporting has held that Microsoft is just about to release a new P2P app that blows his out of the water. These reports talk about it the way one would talk about the release of a new game next month. Such is not the case, ergo, it's appropriate for him to point out his competing product is here now, and that the information about his product in these reports is erroneous.
"But microsoft does like to be a part of anything that gets big"
Ah ha! I agree with you on that point. Tremendous insight there. They like very much to move into proven arenas. Perhaps that is a smart way to spend money. Spend it where it is already proven that there is a market, rather than making one yourself. Proof is in several area's. Office, to compete with Word Perfect, IE, to compete with Netscape, even DOS to compete with CPM/Unix and Digital Research. Have they ever made anythign truely original? I can't think of one. Yet they still make boat loads of profit. Perhaps their business tack is what makes them good rather than their ability to produce good original products.
"Folks with Aspergers/HP Autism don't typically react defensively."
Where did you get this from?
Actually, someone with Asperger's tend to be like this more than someone without it. Just think about it, Aspies feel awkward socially, wouldn't you logically expect them to be defensive? It is true that aspies value logic more than someone without Asperger's, but Asperger's is not about not having emotions, it is about the difficulty of expressing them.
It takes a man to suffer ignorance and smile
Be yourself no matter what they say
If Cohen can't figure out where MS got the idea that BT was still tit-for-tat, he should read his own FAQ on his own web page, or his own design documentation on his own web page. If he's not going to keep that information current, he should just take it down or quit criticising people for the misperceptions he is helping foster. Professionals don't say that the source code is the documentation, that's just lame.
Anyway, MS has been developing a semi-P2P protocol for a while now called BITS, I'm suprised no one has mentioned it yet.
i think that if we scratch the surface a little deeper here, m$ has been introduced to a fast and inexpensive way to deliver media with DRM to consumers.
think about it: m$ wants to sell you music and movies, but they don't want to invest huge amounts of money in servers all over the place, they want a way to deliver media with DRM, and you want faster downloads. so, enter avalanche. you deliver microsoft's content, and you don't complain because you get can download content very quickly. (you might complain, but not Joe WindowsUser)
i bet you'll see something like this built in behind the scenes in windows media player. maybe it'll integrate with their music store, and maybe they're working on a video store.
"If the algorithms are faulty I don't see how the derived software is going to turn out much better"
You don't get it do you. It was presentation material. That's all. It wasn't algorithms used in an existing product. It was simply presented to fetter out exactly the issues you are mentioning, prior to any production work. If you think that all potential products have to have dead on, stone cold, absolutely rock solid foundations prior to the presentation of an idea, then I fear you would not have any software at all. You use Bit Torrent as a standard. That's fine. But, I am sure if you went back and looked at their ideas prior to coding, they also would have a great number of things, at that time, that wouldn't hold water today. My only point is that judging a product by it's pre-production notes, is silly. I can't think of any project in which the outcome was exactly the same as the picture that was painted for it prior to work. Everything changes as production starts and continues. This never even got off the ground and everyone shakes a fist a MS for it. Perhaps we should use the same standard for judging Linux or NASA or even our own governments. They all are dismal failures when using those same standards of judgement. Just be fair in your doling out of blame and the standards of how you measure it. That's all. I am in no way suggesting that MS is right, it's just the process of blaming them that seems so wrong to me.
Along with the new command line tool. the new search paradigm, the new filesystem and Sherlock!
I have nothing to hide. So, why are you spying on me?
The question is: Will it be pulled from Longhorn?
(Yes, this is a joke, morons.)
Richard Steven Hack - This sig is TOO GODDAMN SHORT TO DO ANYTHING USEFUL WITH! MORONS!
Would a MS P2P program actually help the current P2P market (which is under attack by **AA etc)? Since it will most likely be distributed with Windows and everyone will use it, could the backing of MS to P2P make it much harder to attack P2P stuff? Maybe they should ask Bram's advice, and in turn he should help because it can only help P2P (though not necessarily BT).
They are marketing their IDEA, not the actual software at this point. Sure, what they have done is research and simulations, which is obviously just one part of their software development cycle.
Really, what have they actually used from Microsoft R&D? Bob? Precious little is the answer.
No, the way the product cycle works is that other people do the R&D and code, which then Microsfot aquire or duplicate. The job of Microsoft R&D is to keep smart people out of other companies where they might do Microsoft some harm.
Now this is closer to that model since people are happily using Bittorrent already, but it seems like Microsoft can't even copy things properly anymore and instead have to add new layers of useless complexity to them.
"There is more worth loving than we have strength to love." - Brian Jay Stanley
He's calling it vaporware for two reasons:
1) It doesn't exist yet, early or not, which is a point other articles on the subject might not have emphasized before they started talking about a "competitor" to BT.
2) As stated, it's not going to work as far as he can tell, which is the second point other articles on the subject might not have emphasized before they started talking about a "competitor" to BT.
In other words, vaporware does not depend on how old the vaporware is. Microsoft can announce vaporware five minutes from now based on an idea still in in someone's head, with no papers or anything, and it's still vaporware until you get to at least an alpha test product.
Richard Steven Hack - This sig is TOO GODDAMN SHORT TO DO ANYTHING USEFUL WITH! MORONS!
There are two things that I immediately see could come from this.
:(
1) Avalanche is launched and is a success
Coupled with the 'marketting' force of MS, distributors will start using Avalanche over BitTorrent because it's less likely to get them sued. BitTorrent will be viewed as the 'evil pirate upstart' version of MS's newest 'invention'.
2) Avalanche is canned
MS will put enough marketting spin on it to say that they have discovered this method of delivery unable to support 'secure, legal distribution' and the rest of the world will have to agree, or be sued. BitTorrent will be viewed as the 'evil pirate upstart' that can only be used for evil because MS have 'proved' it.
Either way, given MS's trustworthiness record, it's not looking good
Didn't this guy go to work for valve and make thier distributed downloader for Steam? I had steam for 4 months before I could actually use it and it took an ISP change to make it work. For all his smacktalking maybe he could go back and make THAT thing more than vaporware. Don't hate on the Microsoft people for having the resources to run simulations. They are trying to make sure their stuff works before they ship (hasn't that what you have been asking for?).
The MS researchers are typical academics. Bram pointed out some excellent points including that tit-for-tat has been ripped out of the BT code for 5 years now simply because in the real world it doesn't work, even on his little 6 machine network. He devised better methodologies. Also the researchers are assuming too much consistency about the internet and assume that each peer is out for the total good of the network rather then being selfish and out for their own good. The researchers also assumed that only 4-6 peers would be needed and thats what they compared BT using (BT uses 30-50). I'm a computer science major and deal with this type of academic bullshit all the time. These PHDs go "Oh look my algorithm runs in 100000000*n time, that is O(n)! I have a great linear algorithm!" and they completely ignore facts like the network on average will be 10 to 20 million people and that in such a case the constant (100000000) outweights the benefits of the algorithm. An exponential function might work better in such a case but I've seen professors just outright refuse to acknowledge such things simply because the constant is a constant and so it can be ignored. Unfortunately, in the real world, constants are constant and can't be ignored. Academics have a wierd way of thinking of things, they like things to work out perfectly, everything has to be a mathematical absolute. The network simulation the researches used had everything ideal and consistent. Please show me a large network where such is true.
Regards,
Steve
Yeah, I'm pretty sure that was a joke, and you totally missed it.
"Researching either the source code or the documentation on the BitTorrent Web site would have shown that the real choking algorithms work nothing like this."
This is Bram's comment. Is he hallucinating that the Web page documentation says this? If not, then he's correct to say it, regardless of what other sections of his documentation might say. Inconsistent documentation is not good, but obviously very common in either commercial or OSS software.
If Microsoft wants to do simulations based on BT concepts, they should obviously use the latest (stable) software build as their model and not some Web site FAQ. You'd think MS would know that.
Richard Steven Hack - This sig is TOO GODDAMN SHORT TO DO ANYTHING USEFUL WITH! MORONS!
Microsoft's servers send out countless GBs a day for Windows Update (and Office Update, may it be integrated!) so imagine if Microsoft could use BT-like methods to force end users to share the Updates? I would require Windows to bandwidth test, and integrate into Windows in such a way (svchost?) that it would cause other problems if blocked by a firewall...
Maybe someday I'll sign up for a name here.
lets slashdot livejournal!
WTF are you talking about? Why would "M$" need a "decoy"? Not only have you no idea of how and when and by whom research papers are posted, at Microsoft or elsewhere, but you didn't even bother to read it, did you? Don't let that stop you from sprouting your uninformed sheep-mentality opinions though. After all, it's "M$" we're talking about!
Avalanche is a *research* project. It's a *paper* with a simple prototype. Has MS announced plans to ship a product based on it? Is this a beta, or even an alpha?
No, and it was never presented as such. You people just want an excuse to piss on MS. You complain that MS doesn't "innovate" -- well, this is a research paper that claims to be a way to improve P2P networks. Sounds like innovation to me. It's not a product, and was NEVER presented as such!
Grow the phuck up.
The obligatory 'You must be new here' post.
i have my doubts about some (but not all!) of the research that people undertake in the academic world of computer engineering/systems. from what i have seen, some academics will rush to publish for the sake of having a paper, even if it means cutting corners on an otherwise interesting idea or application. while i won't say that this is the norm at the very best institutions in the world, things on the whole can seem less than perfect.
h tm
i imply nothing about the individuals in the paper that Bram attacks, since i haven't interacted with them firsthand. however, it might be interesting to note that the primary author is a grad student at Georgia Tech. according to his web page, his stint at MS research was just a ~6 month period, 2/04-6/04 & 7/04-8/04:
http://www.cc.gatech.edu/~gantsich/biography.htm
the call for papers for this 2005 conference set a deadline of 7/7/2004:
http://www.ieee-infocom.org/2005/call_for_papers.
this does not leave a huge block of time for one student to brush up on the research background, flesh up the practical aspects of the idea, implement (and validate?) a simulator, complete a preliminary set of data runs, and write a paper draft worthy of acceptance. let's not forget any downtime that might arise at the start of an internship (moving over the pond, getting acclimated, etc.).
here, i assume the not unrealistic situation where the official research scientist principally serves as a primary investigator. he brews the idea, perhaps working out some more theoretical aspects of the problem, and handles all the headaches related to funding/approval/propaganda. this entrusts a good deal of the grunt work to the student. i tend to see this sort of behavior in the ivory tower, but it is entirely likely that research in industry is much more balanced!
time should not be an excuse in any case, but it does raise an eyebrow toward the paper-happy nature of some research these days. you make the call on what you believe is reasonable concerning those flaws in methodology that Bram has so derided in his blog.
does anyone have a clue about the timing of the media's spin on things? The Register's article from the first slashdot posting is one of the first according to Google News...
Don't be silly! BT is very new.
xmodem is old!
"I'd rather win in an ugly car than lose in a pretty car" - Jari Lahdenpera
I spent last summer researching the exact idea that Microsoft is pushing now, and never figured out how to make it faster than bittorrent in my simulations.
Microsoft used to understand that actually shipping software is a really important feature. But in the post 9/11 world, I guess the figure fear, uncertainty and doubt are more important.
How's the student "plaza" coming along? =)
-Matt
Duke '05
Yes, I experienced that too. But it wasn't one user, they had a block of machines flooding bad data with the express purpose of poisioning the torrent. But you can still get the file.
If you got the torrent from the pirate bay, they also listed the IP's to filter. After filtering those, I only had one bad IP left and Azurues automatically banned it.
After that addess was banned, I didn't have a single hash failure.
This is an interesting test. Everything is about money, so I suspect if the company responsible can get a significant reduction in number of succesful downloads, they will probably be marketing this "service" when the fall TV seasons starts.
Also there is the interesting possability that eventually they can "poison the earth", by moving ISPs and getting different blocks of IP addresses, they can get larger swaths of the IP space banned by P2P block lists, blocking more and more would be users.
Currently I note that the Peer Guardian list is at about 400Million addresses blocked. Or about 10% of the IP space.
Huh? Add a feature to Longhorn? They have been gutting features from Longhorn as fast as they can, and it still does not look like it will be released before 2007. Is this "threat" supposed to be even remotely threatening, or did I miss some sarcasm here?
P.S.
Considering the rate of adoption for XP, when will Longhorn reach critical mass, if ever? 2012?
Read Typically. And note that I didnt say they are emotionless merely that - as you say - their reactions tend to be more logic based than emotional. My experience has been with relatives and neighbors who have been diagnosed both with Aspergers and HPA. The actual experience of an HPA/Aspy(?)will remain a mystery to all but the one diagnosed. But I think its well accepted that they are less likely engage in discourse that is emotionally motivated. Additionally our interpretations of discourses with people who tend toward HPA and Aspergers tend to be colored - I still am taken aback by the bluntness of some comments and then realize that the comments weren't meant to be personal - merely accurate.
Since they stripped out every other feature from Longhorn, that would make Avalanche its only feature. Somehow I doubt they will disrupt the zen perfection of a null upgrade.
What we need is to have BitTorrent enabled in the Automatic Update client in Microsoft Windows.
Viruses you say? Windows update has previously been hacked and viruses have been installed onto it. So they sign the updates now. Besides, BT does checksums.
Imagine the bandwidth savings I could have on my WAN links if Automatic Updates (WSUS/BITS) was smart enough to download the patch once over the WAN link, then distribute it locally to other stations from that one copy.
Zenworks, SMS, Patchlink, UpdateExpert, etc could all benefit from this.
LMAO.
I guess 'finished' means something different to everyone. I loved BeOS, but it had a smaller feature set than most current OS do.
So we should see in in what, 2014?
"Live Free or Die." Don't like it? Then keep out of the USA
Thanks for being the only one to discuss this issue without becoming too defensive or, as Cohen does, going off on a psychotic tirade.
I can't speak for other P2P programs because I have never used any of them. I have used BitTorrent on many occasions because it promised much better transfer rates than swamped FTP sites. The idea of multiplexing a download sounds like it would be a tremendous performance boost and perhaps, for those unfortunate souls still stuck on dialup or X.25 connections, it is. But, I have yet to see significant increases in performance.
I suspect that you are absolutely correct that it is the asymetric nature of MOST broadband connections that is causing the choking algorithm to impair the overall download speeds. But, you have also confirmed my point that the choking algorithm is the problem since the BitTorrent concept would suggest that larger numbers of peers should equate to greater downoad speeds but that is not the case. Because of the choking algorithm it doesn't seem to make a difference if there are 20 peers or 20,000 peers, regardless of which is the case, download speeds are very disappointing.
What I can't understand is why there is so much venomous denial that this is happening. Surely asymetric broadband connections are the most common so it would seem that the algorithms would/should be adjusted to accomodate such connections but, that does not seem to be the case. I can't place the blame anywhere else but BitTorrent when half my bandwidth is sitting idle while there are more than enough peers to saturate the pipe several times over.
Simulations can get you through some of that, and it's much easier to do if you've got the resources of a Microsoft research lab as opposed to an individual with some help from his friends and friendly users. But it still takes a bunch of cycles, and it's hard to compress that into three months.
Bill Stewart
New Fast-Compression-only CPR http://preview.tinyurl.com/dy575ks
SDLC?? Most people switched to HDLC years ago, and usually use PPP now...
Bill Stewart
New Fast-Compression-only CPR http://preview.tinyurl.com/dy575ks
At this point Avalanche is a research project which would like to become a bigger research project. That doesn't mean it's a product that's ready for prime time, with friendly user interfaces etc. It's fundamentally a transport protocol, and the obvious way to handle DRM is as a separate layer, e.g. part of the user interface or the media player applications, rather than gumming up the transport bits with DRM.
Bill Stewart
New Fast-Compression-only CPR http://preview.tinyurl.com/dy575ks
MSR
All this is is bittorrent with all files encoded via parity archive (PAR). Nothing more. If people PAR'd their files and posted just the PAR files with the current bittorrent, you'd have EXACTLY what MSR is talking about. What retards. Nothing like reinventing the wheel, then bragging about it.
Sure was nice of Brahm to give them feed back on their initial paper. His advice on their research would have probably been worth tens of thousands of dollars if he were considered a consultant. Too bad he is about to be on the butt end of a large "embrace and extend" campaign by MS. It'll be interesting to see all of the security holes, back ends, and reverse engineerings of the Avalance though. Add DRM and Avalanche distributed TV and pretty soon all of us will have free television and movies.
http://research.microsoft.com/~pablo/avalanche.htm
looking at that URL.. (~pablo) isn't that apache?
Ah, the power of the bitchslap. (joking there /. editors)
I'm by no means a Microsoft fanboi, just saying that maybe Bram's pride has more to do with his disdain than anything else. Hey, I'd be upset if they said my best work was marginal at best.
Get your Unix fortune now!
but your honor the file sharing software is an intregral part of the operating system if you remove it windows wont work so i had to share all my files.
Didn't he write Dracula? :-)
Ok, everybody needs to RTFA again. Bram claims that nc is vulnerable to malicious clients (why? is he implying that he believes that they wouldnt use checksums at all?).... but the nc_contentdist paper addresses this (Obviously Bram hasnt even read the pdf). Other people claim that the central server would have to calculate the hashes of the encoded blocks , eliminating the advantage of having the clients do the encoding (nc -> networkcoding)... neither of these assertions are true.
/. news for trolls. stuff that splatters.
from pg 11:
"Another major concern in any content distribution scheme is the protection against malicious nodes. malicious node can introduce arbitrary blocks in the system and make the reconstruction of the original file impossible. When the nodes do not perform coding, the server can digitally sign the packets transmitted and, thus, protect against malicious users. Digitally signing is more difficult when rateless codes are used, but recently [29] demonstrated how homomorphic collision-resistant hash functions can be used to provide protection in that case. Similar schemes can be used to provide protection when network coding is in place [30]. In Avalanche we use special sets of secure hash functions that survive network coding operations and consume very little computational resources, as opposed to traditional homomorphic hashes."
hmmm. now homomorphic (distinct from homeomorphic) means same form and without either [29] or [30] i would guess that the homomorphic hash functions are invariant (in some way) under the encoding/decoding scheme.