Luskin's lawsuit is grounded on defamation. I can defame you and be sued for it without publishing anything; e.g., I can just go stand in a crowded public place and tell everyone you are a child molester.
Sorry, I think your first assessment is right. There are few new insights here. The phenomenon described in the article sounds roughly like the formation, "mitosis," and migration of bubbles in a lava lamp. Okay, you can call these things cells. That's somewhat reasonable. But the researcher said, "the emergence of such spheres seems likely to be a prerequisite for biochemical evolution." That sounds like serious pop science quakery to me. It is only correct with the loosest interpretation of "prerequisite," "bio," and/or "evolution," and even then it's highly misleading.
Another well-known periodical, the Wall Street Journal was quite cordial when these kidscracked the Journal's session authentication scheme. I can see how exposed SSNs and address books could spook a company a lot more than a cracked online subscription system, but it's still a disparity worth keeping in mind if you're one of those folks who's keen on voting with his consumer dollars...
The standard Microsoft weenie excuse for instability in the past has been "it's the drivers!"
Why is this a "weenie" excuse? The same is true in Linux. Even after you scale the bug counts to compensate for the fact that drivers account for 70% of the code in Linux, the error rate for driver code is still vastly greater than the error rate for the rest of the kernel. This is because (in both the Windows and Linux cases) drivers are written by people who aren't as familiar with kernel hacking, and are therefore more likely to make mistakes (such as cut-and-paste errors from copying boilerplate code from another driver).
GoToMyPC now supports one-time passwords, so their users now have no excuse if they get their passwords sniffed on public terminals.
By the way, you all should be using one-time passwords on public terminals, too. If you run Linux, install the S/Key PAM module. FreeBSD supports OTPs out of the box.
They also are all symetrical along a vertical axis.
Well, that's inherent to their
traditional production process. Since Microsoft's inkblots are computer-generated, they could conceivably make them all wacky and assymmetric, but the traditional Rorschach inkblot will always be symmetric about the paper fold.
Although CORBA now provides the underlying technology for things such as J2EE, it is largely gone as far as a standalone technology.
Uh, excuse me? Do you have *any* evidence to back this up?
CORBA, for instance, is designed for interlanguage operability, and is at once heavyweight and feature-deprived as a result, whereas J2EE remoteobjecttechnologies are very Java-specific. CORBA, for instance, uses distributed reference counting for garbarge collection, RMI uses leases.
If by "underlying technology," you mean that CORBA offers naming, lookup, and invocation, and J2EE does too, then, uh, ok. I wouldn't say that McDonald's is the underlying technology for my mother's house, just because they both have kitchens and serve food.
A key to Linux's success was that its license (dubbed a "copyleft" instead of a "copyright") required users also to give away their own modifications to the software for free.
Uh, no, sorry Mr. Law Professor, it doesn't. Not if they're not distributing it.
No, they do actually want to run Portage on embedded devices. From the "Reasons for Forking" document:
There are numerous outstanding issues with Portage that have been documented as part of the embedded project, and even Nick Jones, the Portage architect, has been considering the possibility of a rewrite for quite some time.
Further, the current implementation language (Python) is not well-suited for many embedded systems; a rewrite in a lower level language (e.g. C/C++) will eventually be required to reach all targeted platforms.
...you never hear the adult industry complaining about p2p. Perhaps they have modified their business model so that p2p sharing has only limited negative effect (or maybe even a positive effect).
Even a modest 1.5MBit/s divx is far from perfect when watching it being transfered from a (fast) file server to a (fast) computer. It's ok for short clips, but watching a movie is out of question.
I occasionally stream 1.4-1.5 Mbps movies from my workstation at work to my laptop at home over the Internet (cable modem) and then a wireless LAN. I'd say it's definitely not "out of the question;" it works fine. Occasionally (maybe once during a movie) I'll hit network congestion (or, more commonly, router glitches on the server side) for a few seconds. I also stream all the music I listen to, though, and audio streams experience similar interruptions with the same frequency.
I don't know what a "fast" file server and "fast" computer has to do with it; my laptop uses less than 20% CPU to play the movie, my workstation uses essentially no CPU, and the bandwidth requirements are a fraction of the server's disk bandwidth.
Yet, I can move about 600kByte/s data on the 11MBit/s wireless LAN.
So, I can watch an entire movie streaming over the Internet and wireless, but you can't watch a 1.5 Mbps stream over a 4.8 Mbps LAN (~5 Mbps is, incidentally, about what I get out of my wireless LAN too)? I'd say something is going wrong. You might expect to get a packet dropped here and there, but there's no way mplayer (or whatev) should drain its buffer. Maybe you're in a part of your house far from the AP, and you're actually getting well below peak bandwidth??
Well, first you have to consider that DOS, for the most part, is 100% assembly; some of the tools aren't but the OS itself? Assembly. It just isn't complex enough to warrant being written in a compiled language like C; and back when DOS was used rather than Windows, the speed hit (and 'bloat') from compiled languages was considered a 'big deal' back then. Remember compilers were also quite a bit less advanced back then..
I feel compelled to point out that C was invented expressly to write an operating system for a not-too-burly minicomputer (burlier than the first hardware MS-DOS ran on, but obviously not burlier than what DOS ran on for most of its lifetime).
Try about $350 for a dedicated T1 (not counting telco charges) with no bandwidth cap.
I don't really understand what point you're trying to make here. A T1 is 1.5 mpbs. The article and the post you're replying to are talking about 100 mpbs. $350 * 100 mpbs / 1.5 mpbs = $23k/month, which is significantly more than the $7500/month that was quoted in the post you're replying to.
I can get 300GB of bandwidth at a datacenter for $100/mo.
This is where I realize that this post is a troll, and feel silly responding to it...
On usenet, there's too many propigation problems anyways. Many of us miss posts done by ISP's within 10-15 class A netblocks.
Considering there are no ISPs in the world with 10-15 class A address blocks, it isn't surprising that you're not getting any posts from them.
Here's the current allocation of class A ipv4 address blocks. Note that only two organizations have more than 10 class A addresses. These are ARIN and APNIC, which are registries that do IP address space assignment.
For example, with one of these, you could implement a truly distributed DNS system that doesn't use hierarchy or centralization, and thus would be much more immune to DoS attacks than the current system.
Can someone please explain to me why "code size matters"?
Cost? Memory is cheap. Hard drive space is cheap.
Execution speed? Make your instruction cache bigger. Goes with the territory.
Download time? You'll find that RISC programs are more compressible than their CISC counterparts, so this shouldn't be much of a problem.
So, really, why is code size important? I'm sure there's something I'm missing here, but code size strikes me as something that was a lot more important "back in the day," when memory was more precious.
And a chaser: Nept points to this interesting Microsoft-funded.NET obfuscation project.
Why is this "interesting"? Java bytecode obfuscators have existed for years (23,000 matches on google). It's pretty much par for the course; Sun has been distributing a bytecode disassembler with the JDK since its early releases, after all. I wouldn't expect things to be any different with.NET.
This reminds me of the Jetsons, where all their meals were in pill form, and they used chemicals to give the pills food-like taste. I remember one episode in which George humorously chided his robotic maid, "You burned the toast, Rosie."
Slashdot Mirror
Luskin's lawsuit is grounded on defamation. I can defame you and be sued for it without publishing anything; e.g., I can just go stand in a crowded public place and tell everyone you are a child molester.
Sorry, I think your first assessment is right. There are few new insights here. The phenomenon described in the article sounds roughly like the formation, "mitosis," and migration of bubbles in a lava lamp. Okay, you can call these things cells. That's somewhat reasonable. But the researcher said, "the emergence of such spheres seems likely to be a prerequisite for biochemical evolution." That sounds like serious pop science quakery to me. It is only correct with the loosest interpretation of "prerequisite," "bio," and/or "evolution," and even then it's highly misleading.
Another well-known periodical, the Wall Street Journal was quite cordial when these kids cracked the Journal's session authentication scheme. I can see how exposed SSNs and address books could spook a company a lot more than a cracked online subscription system, but it's still a disparity worth keeping in mind if you're one of those folks who's keen on voting with his consumer dollars...
By the way, you all should be using one-time passwords on public terminals, too. If you run Linux, install the S/Key PAM module. FreeBSD supports OTPs out of the box.
Java provides interoperability with CORBA != CORBA is the "underlying technology" behind remote objects in Java.
If you want proof of this, consider the fact that support for IIOP and most all other CORBA support was added after RMI already existed.
you.retract || you.explain || you == mean_troll
CORBA, for instance, is designed for interlanguage operability, and is at once heavyweight and feature-deprived as a result, whereas J2EE remote object technologies are very Java-specific. CORBA, for instance, uses distributed reference counting for garbarge collection, RMI uses leases.
If by "underlying technology," you mean that CORBA offers naming, lookup, and invocation, and J2EE does too, then, uh, ok. I wouldn't say that McDonald's is the underlying technology for my mother's house, just because they both have kitchens and serve food.
I don't know what a "fast" file server and "fast" computer has to do with it; my laptop uses less than 20% CPU to play the movie, my workstation uses essentially no CPU, and the bandwidth requirements are a fraction of the server's disk bandwidth.
So, I can watch an entire movie streaming over the Internet and wireless, but you can't watch a 1.5 Mbps stream over a 4.8 Mbps LAN (~5 Mbps is, incidentally, about what I get out of my wireless LAN too)? I'd say something is going wrong. You might expect to get a packet dropped here and there, but there's no way mplayer (or whatev) should drain its buffer. Maybe you're in a part of your house far from the AP, and you're actually getting well below peak bandwidth??Nevermind; I'm an idiot. I misread your "within" as "with". Sorry.
Here's the current allocation of class A ipv4 address blocks. Note that only two organizations have more than 10 class A addresses. These are ARIN and APNIC, which are registries that do IP address space assignment.
Cost? Memory is cheap. Hard drive space is cheap.
Execution speed? Make your instruction cache bigger. Goes with the territory.
Download time? You'll find that RISC programs are more compressible than their CISC counterparts, so this shouldn't be much of a problem.
So, really, why is code size important? I'm sure there's something I'm missing here, but code size strikes me as something that was a lot more important "back in the day," when memory was more precious.
Letterman also owns it.
Global warfare with diplomacy sounds a bit like Eschaton. (Fingers crossed that that game doesn't meet a similar fate...)
This reminds me of the Jetsons, where all their meals were in pill form, and they used chemicals to give the pills food-like taste. I remember one episode in which George humorously chided his robotic maid, "You burned the toast, Rosie."
What miffs me (a little) is that they haven't made an installation guide available to non-paying customers since 7.2.