Rendezvous is great and all, but I doubt it will ever consume more of the poster's time than did Cheshire's most important contribution: the network tank game Bolo
*g*
I have a DEC 3000/400 (no "station") that I got virtually for free a couple years back. I ran netbsd (which has much better hardware support on turbochannel machines than alpha linux, plus it's not linux::ducks::) on it for a long time, it was a web, name, ftp server, you get the picture. 150 or more days uptime, only interrupted by power outages, and it ran in a closet so must have been at least 80 degrees F in there continuously. When I went to move it, I was puzzled at the sticky grey goo underneath the machine until I realized it had melted its plastic feet!
It's a great machine, incredibly reliable, unfortunately the days of these beasts being useful are past I think. It's just so cheap to get an x86 (or in my case an iBook with a dead screen) machine to replace them which is faster, cooler, more energy efficient, and quieter.
Of course the coolness factor of running this old workhorse still appeals to me, perhaps when I get a house with a basement (alleviating the noise and heat) I'll set it up once more.
The article mentions something about being able to monitor bandwidth. If this new method is going to be doing a version of NAT itself, then wouldn't the fairest way be just to charge people for bandwidth (or like cell phones, a certain amount per month and then charge for anything in excess of that). I mean, that's where the cost (or at least the problem with people sharing broadband) is...right? An extra node for a device by itself costs nothing, just a few rules in the router's memory. If it was an actual IP address I could see paying the $4.95, those are getting scarce.
This would probably be more useful if some IDE company picked it up for Windows. There's a lot more buffer overflow problems there (a consequence of much more software). And Windows developers love their IDE's (shudder). Maybe Microsoft could use it to secure their OSen (that should get me some karma here;-) !
I look at it as a grammar checker, though. If you know what you're doing, it's not hard to design software that doesn't have these problems in the first place. You don't need a type checker to write good code.
Does internet 2 use IPv6? v4 is getting very limiting.
IMO the only way to get v6 adopted is (was?) to build a new internet. One of those chicken-and-egg problems, no incentive to upgrade the routers because the endpoints don't use it yet, endpoints don't use it because the routers can't route it.
The distros I've used (Yellow Dog and linuxppc) have included hacky booters that will allow you to dual boot. You can screw with Open Firmware too which is a more elegant but frustratingly undocumented way to dual boot; I once had my G4 set up to boot into linux if you didn't hit the space bar within 10 seconds of it booting. I ganked somebody's Forth script to pull that off. Can't remember where I got it though, that was 2 years ago.
Of course, you can run Mac-On-Linux which runs a lot of mac programs nearly perfectly and without noticable performance loss, without need to boot into MacOS.
NetBSD is great for a lot of platforms; it's running right now on my alpha, decstation, and i386. But for macppc it's just not there compared to linux. The documentation is even worse than with other major netbsd platforms (netbsd documentation is pretty bad/outdated). It _still_ comes with XFree 3.3.x (yes version 4 is in the package system), so no easy accelerated video if you've got a Radeon. Plus to boot it you have to muck around in OF trying to figure out the right settings (which if you have a machine built after 1996 aren't included in the install docs), compared to Yellow Dog's hacky-but-easy-to-use booter.
Now, I prefer netbsd to anything else (except OS X on my mac) but it needs a lot of work on macppc (mainly documentation-wise). I just wish I were up to the challenge.
This doesn't just apply to Microsoft. I worry that apple is moving in the same direction with iTunes and iDVD and all that. Just as WMP hurts Real, Disc Burner/Disk Copy in OS X 10.1 hurts Roxio. Of course, Apple isn't as huge as Microsoft, but they have a monopoly on Macs!
I think things that aren't part of the OS (media players, web browsers, instant messengers) should not ship with the default install. Fine to make them free, just include them on a separate CD or custom install, just something so the user consciously has to decide which product he wants to install and use.
Slashdot Mirror
There should be a moderation for +1, Your Mom Joke
It was not Glass's error. Reggio rearranged the images after the score was written. It's in the special feature on the DVD.
Rendezvous is great and all, but I doubt it will ever consume more of the poster's time than did Cheshire's most important contribution: the network tank game Bolo *g*
I have a DEC 3000/400 (no "station") that I got virtually for free a couple years back. I ran netbsd (which has much better hardware support on turbochannel machines than alpha linux, plus it's not linux ::ducks::) on it for a long time, it was a web, name, ftp server, you get the picture. 150 or more days uptime, only interrupted by power outages, and it ran in a closet so must have been at least 80 degrees F in there continuously. When I went to move it, I was puzzled at the sticky grey goo underneath the machine until I realized it had melted its plastic feet!
It's a great machine, incredibly reliable, unfortunately the days of these beasts being useful are past I think. It's just so cheap to get an x86 (or in my case an iBook with a dead screen) machine to replace them which is faster, cooler, more energy efficient, and quieter.
Of course the coolness factor of running this old workhorse still appeals to me, perhaps when I get a house with a basement (alleviating the noise and heat) I'll set it up once more.
If only we had more than one moon...
The article mentions something about being able to monitor bandwidth. If this new method is going to be doing a version of NAT itself, then wouldn't the fairest way be just to charge people for bandwidth (or like cell phones, a certain amount per month and then charge for anything in excess of that). I mean, that's where the cost (or at least the problem with people sharing broadband) is...right? An extra node for a device by itself costs nothing, just a few rules in the router's memory. If it was an actual IP address I could see paying the $4.95, those are getting scarce.
But every time I play chess I'm forced to move my king even though I may not want to :-(
# = tic-tac-toe Eventually we'll have 7 vertical and 7 horizontal, it will be C-chess.
This would probably be more useful if some IDE company picked it up for Windows. There's a lot more buffer overflow problems there (a consequence of much more software). And Windows developers love their IDE's (shudder). Maybe Microsoft could use it to secure their OSen (that should get me some karma here ;-) !
I look at it as a grammar checker, though. If you know what you're doing, it's not hard to design software that doesn't have these problems in the first place. You don't need a type checker to write good code.
Does internet 2 use IPv6? v4 is getting very limiting.
IMO the only way to get v6 adopted is (was?) to build a new internet. One of those chicken-and-egg problems, no incentive to upgrade the routers because the endpoints don't use it yet, endpoints don't use it because the routers can't route it.
Has to have open firmware; so no ydl or linuxppc for your 5300. mklinux works for some older macs but I don't think the 5300 is one of them.
The distros I've used (Yellow Dog and linuxppc) have included hacky booters that will allow you to dual boot. You can screw with Open Firmware too which is a more elegant but frustratingly undocumented way to dual boot; I once had my G4 set up to boot into linux if you didn't hit the space bar within 10 seconds of it booting. I ganked somebody's Forth script to pull that off. Can't remember where I got it though, that was 2 years ago.
Of course, you can run Mac-On-Linux which runs a lot of mac programs nearly perfectly and without noticable performance loss, without need to boot into MacOS.
NetBSD is great for a lot of platforms; it's running right now on my alpha, decstation, and i386. But for macppc it's just not there compared to linux. The documentation is even worse than with other major netbsd platforms (netbsd documentation is pretty bad/outdated). It _still_ comes with XFree 3.3.x (yes version 4 is in the package system), so no easy accelerated video if you've got a Radeon. Plus to boot it you have to muck around in OF trying to figure out the right settings (which if you have a machine built after 1996 aren't included in the install docs), compared to Yellow Dog's hacky-but-easy-to-use booter.
Now, I prefer netbsd to anything else (except OS X on my mac) but it needs a lot of work on macppc (mainly documentation-wise). I just wish I were up to the challenge.
This obviously didn't tip her off, because after I had to spell it for her, she still asked for my address.
"Do I really need to give you my address to buy this?" I asked. She replied indignantly, "Well, we're supposed to ask!"
sigh
LEO is still free fall isn't it?
This doesn't just apply to Microsoft. I worry that apple is moving in the same direction with iTunes and iDVD and all that. Just as WMP hurts Real, Disc Burner/Disk Copy in OS X 10.1 hurts Roxio. Of course, Apple isn't as huge as Microsoft, but they have a monopoly on Macs!
I think things that aren't part of the OS (media players, web browsers, instant messengers) should not ship with the default install. Fine to make them free, just include them on a separate CD or custom install, just something so the user consciously has to decide which product he wants to install and use.
But maybe I've been using netbsd too much.
> monitors with resolutions of 3000dpi or more
Damn, that's high-res pr0n.
I'd settle for 300 dpi (150 is probably enough for photos)