i think all linux distros will come with torrent by default within a year or so... give them time, but once it truly stabalizes and gets that settled in feeling, ditros will FLOCK... then all free OSs will follow. BSDs specifically. Mac os x will not be far behind... it would be nice if Mozilla or Safari could just have a torrent plugin to handle the downloading of files.
the sad part is, there are those people yet they still use AIM(or similar) which already holds your buddy list on their servers... AOL behind the scenes can already connect all these dots to 30 million or something people(+aim users, ofcourse), this would just give OTHER people the ability to see the 'friend of a friend' graphically... hell, just have AOL put an 'leave me outa this BS' checkbox in preferences and do it anyway.
They have the largest IM/ISP audience and it is perfect to overwelm this new trend.:)
It is sad, the 15" (17" too maybe) powerbooks have been plagued with problems... the 12", which i use right now, have been pretty much free of them as far as i know.... Hope Apple does the right thing and fixes the 15" line in a major way.
Rendezvous is Apple's implementation of the open standard of zeroconf... Linux could just as easily have a compatable equivalent if someone coded it. Apple employees the main(or one of the main) developers/writers/originators of the standard, which is why they have adopted and used this super-awesome protocol so quickly.
To switch form Windows to Linux, you save X number of dollars in licenses, that X, to save money immediately, needs to go to any training(minimal on a locked-down workstation) configuring (get a linux guru and make images of a solid secure locked-down workstation and any server stuff, etc). Now, to save money, you need to do training and config for X-1.
Basicly, what i am trying to say is that you can take the X number of dollars and spend 90% of it on doing a custom support package from a developer, configure your workstations, or whatever, and still have a 10% INITIAL savings on making the move.... Generally, the X (if you are in a moderate to large company) is gonna be big enough that you can do a lot and still have a 10% savings from the get go.
wait a minute, this doesn't make sense... An Apple prediction on a future product that is actually FEASIBLE and a good idea? Amazing. My hat goes off to you on this one. A Xserve Music Server geared toward indy record labels is probably the best non-apple apple product i have ever heard of...
No kidding, mode this guy to +5 and email Steve, I think you have a pretty good idea here.
i say ipv6 (with the idea in my head of zeroconf closely alongside) instead of dhcp+ipv4+private addressing just because the added security and decentralization of zeroconf gives you(Security comes from ipv6, that is). There shouldn't be a DHCP server, that makes a single node critical to the network's functionality... this network should use zeroconf, or some auto-negotiating protocol to self determine addresses, not rely on a single node's services.
The nodes should cache, and use that cache the same way(in principal) that freenet would use cache systems, minus the extreme security concerns taken... Just request a certain page and all nodes would check if they have it, just like searching for a.mp3 file, you would search for a.html file... not all that difficult at all. Also, having a fraction of a given site's contents in a cache could be a problem... 2 ideas would be some parsing software to strip the html hyperlink code from pages that don't have the linked file locally. OR 2, the ability to use another node as a bridge to the internet, if needed.
I could search node A's web cache, read on something i like, but once i reach a hyperlink that can't be resolved, it gets passed to the access point(node B) or the network bridge node C provides to the rest of the net.... Each node will have different functions and cache, if you request a/. page, and then a link to the ad on that page and the node you first used didn't have it, it would ask other nodes, one or more of those nodes might have the given page cached, OR posess a active link to the internet.
sounds like ideal ad-hoc network...
on
Mobile Wifi Backpack
·
· Score: 4, Interesting
Get a linux distro tuned just for this ad-hoc network... setup a nice cache server and ipv6 and let it accept everyone... all the users woul get in vacinity would supply their own address(ipv6, again) and would setup a wireless p2p ad-hoc network... the more users using a cache system the more % of the internet it holds...
Agreed, bugs are inherit to a large program (or OS or set of programs, etc)... Two things make the difference.
1. The style and process taken in coding and trying to catch bugs before they are released into a final product.
2. The manner in which bugs are corrected and patched once discovered....
Apple and the Open Source Community are both using systems AS or More complex than Windows (accomplishing the same level of tasks, atleast). Yet the way Microsoft handles bugs compared to Apple or the OSC is hugely different. The OSC has a turnaround on a discovered bugs that is quite high. A critical level bug may be patched in a matter of hours. Apple's policies are nearly as good, where most important bugs are handled in a matter of several days, to, a couple weeks(the more important the quicker, ofcourse)... Microsoft however, has left(and still leaves) critical remote bugs unpatched after a matter of years, relying more on the publicitiy to the masses of a bug rather than the severity of it... When Microsoft is in a hurry, there turnaround is much closer to the level of Apple, releasing a patch in a few weeks... However, Microsoft has repeatedly taken months, years and even ignored critical bugs in the OS.
the ibook's G4 chip may be what you are referring to. It is not the same chip as what runs the same clockspeed 12" powerbook, rather, a modified G3 chip with Altivec included. Adding Altivec gave it the base criteria to be a 'G4' chip and while the same clockspeed, i believe the powerbook chip equivalent is still a bit faster. I could be wrong though.
it is all about the right size. my sis's boyfriend(quite the freebsd guy, actually) has a 15" tibook that is nice, but for me just too big. He is the opposite(himself a pretty tall big guy), he thinks the 12" is too small... Go to a mac store and play around with them, the 12" ibook is pretty close but slightly larger than the 12" powerbook.
O, and upgrade your ram thirdparty. Apple charges too much for builtin ram upgrades, and Mac OS X very much takes advantage of all available ram(though runs just fine on 256m, from my experience).
g5 ibooks will be a ways off. g5 powerbooks might be demo in july(june?) maybe... likely it will be imac g5's first, and at the same time, or the next expo we will see the large powerbooks get g5s(15/17"). the 12" pbook will be last, then the ibooks and a long time from now the emacs.
that is my guess.
NOTE: i bought my 12" aluminum 1ghz powerbook in january, love it to death, would use it 24/7 if at all possible, and it is worth a few extra bucks over an ibook. Also, i have noticed i get a bit above 'average' laptop battery, i can listen to music and do light stuff for ~4 hours, maybe a bit longer if i do energy saving stuff(dim the screen). that is a stretch, i normally get a solid 3-3.5 hours out of it though, which is 30-45 minutes longer than pc laptop users i have talked to.
I have heard of work being done to make a modtorrent for apache... any file that is requested enough automatically turns into a torrent. Also, why not have slashdot post a local cache of the site... and for those who say that the site might lose advertising: have a cache.txt file in the root directory like robots.txt to determine if you want to be cached or not, etc...
yes, why on EARTH would you bundle a media player into the OS such that it is impossible to remove... is it tied into the kernel?
But it doesn't matter, removing the media player is nice, but won't fix or change anything. Removing apps that are inaapropriately bundled is a start, but it is only a bandage to fix the real problem: protocols, documentation and standards... Have MS publish and document ALL the APIs in windows. Publish in detail the workings of the Office file formats(for the current, the next version and the past couple versions AT LEAST).
To anyone who says publishing APIs is a security risk: they are available to developers, Microsoft just doesn't tell you what they are or how to properly use them... It isn't impossible to use them, Microsoft uses them.
THEN the EU should require that all official documents be published in a format in a free and open format... one that can be accessed(read/write) on multiple platforms and the specifications are available for a compotent programmer could work properly with those files... So 15 years from now we(they) arn't stuck with gigs of MS Office 97.doc files that they can't read.
For another insight at the protection of trade secrets: The big-end (like with Dell or HP) license contracts Microsoft has are hidden by calling them 'trade secrets'. So say, if Microsoft put a clause in the contract that said any computer sold, wether it has windows installed or not, must be sold a windows license(to make it impossible to get a big name computer without an MS tax) no one could tell the press, because the contracts are a 'trade secret'.
Yes, the LoTR AI is cool, but it took many dozens+ of attempts and editing to get anywhere near the battle quality as they did in the movies. It was not uncommon for some groups to run AWAY from battle, due to their programming, rather than join their comrades.
"Gibson cast some of the blame on the packet-based nature of Internet Protocol, which was not designed for foolproof delivery of messages. The protocol cannot guarantee delivery of e-mail, for instance."
Yes, it was. TCP guarentees delivery of frames. (something like UDP does not, however)
"As a result, DARPA wants to fund development of new protocols or enhancements to the existing IP that will allow nodes, such as computers, to automatically sign on to networks in their vicinity."
IPv6 + ZeroConf(Apple uses it as rondevous) == GOOD! ZeroConf allows for a lot of automatic stuff, discovery of servers, dhcp, assigning it's own ip, tell other node's of what it can do('hey, i have a printer i can share with everyone!'), etc... This is why everyone LOVES Apple's rondevous(spelling?), it is just damn cool and very, very easy.
Redhat, for instance, ups the X.0 every time it breaks compatability. So that is why they jump from 8.0-> 9.0, for example. Whatever critical package(X Server or something) that breaks compat. with the old version get a whole new X.0 version.
"10 years from now will we have SuSE 24.2?" Yes, and debian will still be on kernel v. 2.4.x!"
Slashdot Mirror
i think all linux distros will come with torrent by default within a year or so... give them time, but once it truly stabalizes and gets that settled in feeling, ditros will FLOCK... then all free OSs will follow. BSDs specifically. Mac os x will not be far behind... it would be nice if Mozilla or Safari could just have a torrent plugin to handle the downloading of files.
stayed there? most people think they still ARE there, for some uninformed reason.
the sad part is, there are those people yet they still use AIM(or similar) which already holds your buddy list on their servers... AOL behind the scenes can already connect all these dots to 30 million or something people(+aim users, ofcourse), this would just give OTHER people the ability to see the 'friend of a friend' graphically... hell, just have AOL put an 'leave me outa this BS' checkbox in preferences and do it anyway.
:)
They have the largest IM/ISP audience and it is perfect to overwelm this new trend.
why doesn't AOL just add a social network plug to aim?
not just when they go home, when they go to lunch... you can use it as a screensaver as well:). This system is pretty damn cool.
that must be a tricked out powerbook....
It is sad, the 15" (17" too maybe) powerbooks have been plagued with problems... the 12", which i use right now, have been pretty much free of them as far as i know.... Hope Apple does the right thing and fixes the 15" line in a major way.
Rendezvous is Apple's implementation of the open standard of zeroconf... Linux could just as easily have a compatable equivalent if someone coded it. Apple employees the main(or one of the main) developers/writers/originators of the standard, which is why they have adopted and used this super-awesome protocol so quickly.
i would be happy to buy a couple and re-sell them (+ shipping and handling ;) fees) to my European neighbors... $300 Mini iPod anyone?
i think of this as the 90% rule...
To switch form Windows to Linux, you save X number of dollars in licenses, that X, to save money immediately, needs to go to any training(minimal on a locked-down workstation) configuring (get a linux guru and make images of a solid secure locked-down workstation and any server stuff, etc). Now, to save money, you need to do training and config for X-1.
Basicly, what i am trying to say is that you can take the X number of dollars and spend 90% of it on doing a custom support package from a developer, configure your workstations, or whatever, and still have a 10% INITIAL savings on making the move.... Generally, the X (if you are in a moderate to large company) is gonna be big enough that you can do a lot and still have a 10% savings from the get go.
wait a minute, this doesn't make sense... An Apple prediction on a future product that is actually FEASIBLE and a good idea? Amazing. My hat goes off to you on this one. A Xserve Music Server geared toward indy record labels is probably the best non-apple apple product i have ever heard of...
No kidding, mode this guy to +5 and email Steve, I think you have a pretty good idea here.
i say ipv6 (with the idea in my head of zeroconf closely alongside) instead of dhcp+ipv4+private addressing just because the added security and decentralization of zeroconf gives you(Security comes from ipv6, that is). There shouldn't be a DHCP server, that makes a single node critical to the network's functionality... this network should use zeroconf, or some auto-negotiating protocol to self determine addresses, not rely on a single node's services.
.mp3 file, you would search for a .html file... not all that difficult at all. Also, having a fraction of a given site's contents in a cache could be a problem... 2 ideas would be some parsing software to strip the html hyperlink code from pages that don't have the linked file locally. OR 2, the ability to use another node as a bridge to the internet, if needed.
/. page, and then a link to the ad on that page and the node you first used didn't have it, it would ask other nodes, one or more of those nodes might have the given page cached, OR posess a active link to the internet.
The nodes should cache, and use that cache the same way(in principal) that freenet would use cache systems, minus the extreme security concerns taken... Just request a certain page and all nodes would check if they have it, just like searching for a
I could search node A's web cache, read on something i like, but once i reach a hyperlink that can't be resolved, it gets passed to the access point(node B) or the network bridge node C provides to the rest of the net.... Each node will have different functions and cache, if you request a
Get a linux distro tuned just for this ad-hoc network... setup a nice cache server and ipv6 and let it accept everyone... all the users woul get in vacinity would supply their own address(ipv6, again) and would setup a wireless p2p ad-hoc network... the more users using a cache system the more % of the internet it holds...
Agreed, bugs are inherit to a large program (or OS or set of programs, etc)... Two things make the difference.
1. The style and process taken in coding and trying to catch bugs before they are released into a final product.
2. The manner in which bugs are corrected and patched once discovered....
Apple and the Open Source Community are both using systems AS or More complex than Windows (accomplishing the same level of tasks, atleast). Yet the way Microsoft handles bugs compared to Apple or the OSC is hugely different. The OSC has a turnaround on a discovered bugs that is quite high. A critical level bug may be patched in a matter of hours. Apple's policies are nearly as good, where most important bugs are handled in a matter of several days, to, a couple weeks(the more important the quicker, ofcourse)... Microsoft however, has left(and still leaves) critical remote bugs unpatched after a matter of years, relying more on the publicitiy to the masses of a bug rather than the severity of it... When Microsoft is in a hurry, there turnaround is much closer to the level of Apple, releasing a patch in a few weeks... However, Microsoft has repeatedly taken months, years and even ignored critical bugs in the OS.
the ibook's G4 chip may be what you are referring to. It is not the same chip as what runs the same clockspeed 12" powerbook, rather, a modified G3 chip with Altivec included. Adding Altivec gave it the base criteria to be a 'G4' chip and while the same clockspeed, i believe the powerbook chip equivalent is still a bit faster. I could be wrong though.
Google's view pdf as html to the rescue. :)
it is all about the right size. my sis's boyfriend(quite the freebsd guy, actually) has a 15" tibook that is nice, but for me just too big. He is the opposite(himself a pretty tall big guy), he thinks the 12" is too small... Go to a mac store and play around with them, the 12" ibook is pretty close but slightly larger than the 12" powerbook.
O, and upgrade your ram thirdparty. Apple charges too much for builtin ram upgrades, and Mac OS X very much takes advantage of all available ram(though runs just fine on 256m, from my experience).
Then why did you use quotation marks?
g5 ibooks will be a ways off. g5 powerbooks might be demo in july(june?) maybe... likely it will be imac g5's first, and at the same time, or the next expo we will see the large powerbooks get g5s(15/17"). the 12" pbook will be last, then the ibooks and a long time from now the emacs.
that is my guess.
NOTE: i bought my 12" aluminum 1ghz powerbook in january, love it to death, would use it 24/7 if at all possible, and it is worth a few extra bucks over an ibook. Also, i have noticed i get a bit above 'average' laptop battery, i can listen to music and do light stuff for ~4 hours, maybe a bit longer if i do energy saving stuff(dim the screen). that is a stretch, i normally get a solid 3-3.5 hours out of it though, which is 30-45 minutes longer than pc laptop users i have talked to.
I have heard of work being done to make a modtorrent for apache... any file that is requested enough automatically turns into a torrent. Also, why not have slashdot post a local cache of the site... and for those who say that the site might lose advertising: have a cache.txt file in the root directory like robots.txt to determine if you want to be cached or not, etc...
yes, why on EARTH would you bundle a media player into the OS such that it is impossible to remove... is it tied into the kernel?
.doc files that they can't read.
But it doesn't matter, removing the media player is nice, but won't fix or change anything. Removing apps that are inaapropriately bundled is a start, but it is only a bandage to fix the real problem: protocols, documentation and standards... Have MS publish and document ALL the APIs in windows. Publish in detail the workings of the Office file formats(for the current, the next version and the past couple versions AT LEAST).
To anyone who says publishing APIs is a security risk: they are available to developers, Microsoft just doesn't tell you what they are or how to properly use them... It isn't impossible to use them, Microsoft uses them.
THEN the EU should require that all official documents be published in a format in a free and open format... one that can be accessed(read/write) on multiple platforms and the specifications are available for a compotent programmer could work properly with those files... So 15 years from now we(they) arn't stuck with gigs of MS Office 97
For another insight at the protection of trade secrets: The big-end (like with Dell or HP) license contracts Microsoft has are hidden by calling them 'trade secrets'. So say, if Microsoft put a clause in the contract that said any computer sold, wether it has windows installed or not, must be sold a windows license(to make it impossible to get a big name computer without an MS tax) no one could tell the press, because the contracts are a 'trade secret'.
Yes, the LoTR AI is cool, but it took many dozens+ of attempts and editing to get anywhere near the battle quality as they did in the movies. It was not uncommon for some groups to run AWAY from battle, due to their programming, rather than join their comrades.
"Gibson cast some of the blame on the packet-based nature of Internet Protocol, which was not designed for foolproof delivery of messages. The protocol cannot guarantee delivery of e-mail, for instance."
Yes, it was. TCP guarentees delivery of frames. (something like UDP does not, however)
"As a result, DARPA wants to fund development of new protocols or enhancements to the existing IP that will allow nodes, such as computers, to automatically sign on to networks in their vicinity."
IPv6 + ZeroConf(Apple uses it as rondevous) == GOOD! ZeroConf allows for a lot of automatic stuff, discovery of servers, dhcp, assigning it's own ip, tell other node's of what it can do('hey, i have a printer i can share with everyone!'), etc... This is why everyone LOVES Apple's rondevous(spelling?), it is just damn cool and very, very easy.
This is the US Military, we have much cooler, bigger, more expensive things to blow the crap out of cds than shotguns. :)
Redhat, for instance, ups the X.0 every time it breaks compatability. So that is why they jump from 8.0-> 9.0, for example. Whatever critical package(X Server or something) that breaks compat. with the old version get a whole new X.0 version.
"10 years from now will we have SuSE 24.2?"
Yes, and debian will still be on kernel v. 2.4.x!"
(i'm mostly kidding...)