Oskar, I appreciate your candid response =) You raise good points..... --- Oskar Sandberg <oskar at freenetproject.org> wrote: > On Thu, Nov 21, 2002 at 08:22:46AM -0800, Tyler > Riddle wrote: > > Ugh, I really hate personal attacks (as they make > my > > argument look weak) so I'll temper that with real > > world examples of why you are wrong as well. > > > > First of all, I have seen you make some completley > > hairbrain assumptions on this list in the past > > (personaly, your hit/miss ratio with me is about > 1/5) > > and I am definitly going to call this one > hairbrain to > > the tune of 100%. > > If you are aware that personal attacks make you look > dumb, then why the > hell go straight into one? Edgar has been around > here a lot longer than > you have, and while everybody is wrong some of time, > he has a lot more > credibility too. If you can't deal with having your > ideas shot down then > STFU already.
I don't care if my idea is shot down, really - if the patch makes it into the code or not maters not to me, I'll still use it and it will make my life easier. I've seen Edgar poop on other perfectly good ideas in the past with out proper cause. The reason I decided to go half flamage on the email is because I wanted Edgar (and everyone else who is interested) to understand why I am throwing his opinion in the garbage regarding my patch. Your opinion I do respect as you raise good points to the majority of the questions/problems you reply to. > <> > > Assuming that my internet connection is failing > when I > > get RNF is retarded. Do you honestly expect that > > someone who can atleast code a little java is not > able > > to detect the difference between a working > internet > > connection and a broken one? I allmost feel > insulted, > > or atleast I would if I did not consider you an > > allmost total retard. I assure you that when I get > RNF > > my internet connection is working as I get > messages to > > the tune of > > Regardless of whether it is your Internet connection > or not, and RNF is > an error condition that SHOULD NOT happen in > practice when using > freenet. As it is, it does happen. We need to find > out what the cause is > - maybe too many nodes are overloaded, maybe > something is going wrong > inside your node, maybe we are just lousy coders, > maybe it is something > else. Can you explain which problem that causes RNFs > flooding the > network with requests is going to solve? > > You say that with the repeating request code, if you > go to bed at night > and leave it running, it will find the data. Well, > the network is not > very large at the moment, by most guesses smaller > than a thousand > permanent nodes. It doesn't take a genius to see > that your long night of > hammering away at the network probably managed to > get your query to hit > every single permanent node. Querying every single > node will make you > find the data! Why on earth didn't we think of > that!! > I suspect that as freenet gets older the health of the network will degrade more, not less. This is simply because more transient users will show up and more bandwidth consuming applications will be invented that place more and more strain on the network. I predict that the number of nodes will not be able to compensate and we will have to deal with such issues as we do now, only worse. It is important that freenet is able to deal with situations that are extremly negative (such as network flooding) as well as when things are all happy. Internaly, the node will (probably) allways have some kind of timeout and give up in the face of problems and that means there will allways be a place for something externaly that kicks it over for another try. > If you have a network of 1000 nodes and routing is > not working, then you > can solve that by flooding. If you have a network of > 100,000 nodes and > routing is not working, then good luck trying to > flood them all. If > this network is to have any future, we need the > routing to work. Short > sighted patches to work around the symptoms of it > failing by abandoning > our goals are not the right way forward. > > Get a clue before you start flaming. > I agree that fixing routing is the beter solution and I understand completley that the request most likely reaches nearly every node on the network before it gets fulfilled (in this case). Just because that is the ultimate solution does not mean this solution does not work now and won't continue to help things in the future. In my opinion, this patch gives freenet the ability to operate under extremly adverse conditions which brings it closer to it's design goals. I understand that the last thing a sick network needs is more trafic clogging things up but freenet is really a free-for-all anarchy style network where anything goes. If you think my patch is bad wait untill things get more mature and people start dropping in their own nodes from a completely different code base that will use methods that seem tame compared to this one. So in summary, I see freenet as having no choice but to operate under the worst set of conditions possible as the network will probably wind up there anyway. Even after every other tactic possible has been thought up, implemented, and tested, you can allways keep clicking the retry button. A good analogy is nuclear weapons - the only way to fight them is to make sure you have more then your enemy. The network will probably get reduced to a request war anyway. You can either sit on the side lines looking at your RNF and DNF page and not get your content or jump in and start fighting and screw the network over just a little more. I leave it up to the collective judgement of the devs to figure out which one they want the majority of the network to do. > -- > > Oskar Sandberg > oskar at freenetproject.org > > _______________________________________________ > devl mailing list > devl at freenetproject.org > http://hawk.freenetproject.org/cgi-bin/mailman/listinfo/devl ===== AIM:rllybites Y! Messenger:triddle_1999 __________________________________________________ Do you Yahoo!? Yahoo! Mail Plus ? Powerful. Affordable. Sign up now. http://mailplus.yahoo.com _______________________________________________ devl mailing list devl at freenetproject.org http://hawk.freenetproject.org/cgi-bin/mailman/listinfo/devl
