I agree that OO is not so good for databases, but it works well in OSes, device drivers, and realtime systems. You just need to know how to get good abstractions without sacrificing any performance, and that's not an easy skill to master.
OO is not all about classes and jump tables. For example, you can get polymorphism in C++ without using any virtual methods at all. If you disagree, then I think your view of what constitutes OO is quite limited, and I'm not surprised you think it's a "niche player".
I think all monopolies should become branches of congress. There are lots of positive effects this would have, but just to name a few: Companies would do their best to avoid becomming monopolies, or their board of directors would be replaced by elected congressmen. The monopolies would become subject to the Constitution ("Congress shall pass no law..."). Imagine how Microsoft would behave if they had to be competitive or risk becomming the Federal Ministry of Software.
This also covers your suggestion, because the Bells would already be a branch of the government.
Ok, you're right. I stand corrected. There's more to the quality of a program than the quality of its code. There are installation procedures, documentation, support, etc.
Actually, sometimes debugging with printfs is much better than gdb. Two situations occur to me:
The problem is nondeterministic. If the failure occurs only once in 100 runs, you probably won't see it when you run it in gdb, especially if it's timing-dependent (in which case gdb may stop it from occurring at all). However, with a proper debug trace, you can just grab the log file when the crash does occur, and do a lot of diagnosis from that (plus possibly the core file if you get one of those).
The program state and state transitions are too complex. For instance, try to debug a compiler without a good debug trace facility. It's just not very practical to walk complex IR data structures manually after various optimizations have been performed.
As for the father's sperm, I have no idea how the DNA in sperm is not "aged" like the father's own DNA. I'm certainly not an expert in that area, but I don't have to know *why* it's that way; I only need to know that it is, which I do happen to know.
It is what? Not aged? Ok, so let's consider how that happened. The testicles' DNA must age, and clearly they are capable of producing "young" DNA for sperm anyway, so again I say there's more to the story than your simplistic "old/young DNA" dichotomy.
(Or, if you want to claim that testicles' DNA doesn't age, then why not clone babies from testicle cells?)
Clearly there's more to genetics than you know.
You're not one of those Raelian nutcakes who claim they've solved the problem of cloning a baby from old DNA, are you?
I'm not claiming I know anything about cloning. I probably know less than you do. I'm just claiming that you don't know enough to say cloning is impossible. Your argument from your first post doesn't hold water, because if that argument were sufficient, then it would also rule out sexual reproduction.
You are speculating. There's nothing wrong with that, unless of course you try to claim that you are not speculating.
Re:What speculation?
on
Goodbye, Dolly
·
· Score: 2, Insightful
Ok, so as long as you're not doing any speculating, then how does regular reproduction work? Re-read everything you just wrote and imagine you're talking about normal sexual reproduction. From that point of view, it's simply 100% incorrect.
Easy there, cowboy. The checksums in this case are not meant to detect errors. They are meant to decide whether two blocks are different from each other without transferring either block across the network. There's a difference, though it may be subtle.
MD5 (or actually MD4) is a cryptographically strong checksum, so the chances of the checksum being the same for two different blocks is truly 2^-128. CRCs, on the other hand, were not designed for this, and so the probability of a collision is much higher.
We're not talking about detecting single-bit errors, sequences of bit errors, etc. that CRC is good at; we're talking about finding arbitrary differences between two blocks of data. It's not hard to construct two blocks with minor differences that have the same CRC, but only a brute-force search will find two blocks with the same MD4.
Slashdot Mirror
OO is not all about classes and jump tables. For example, you can get polymorphism in C++ without using any virtual methods at all. If you disagree, then I think your view of what constitutes OO is quite limited, and I'm not surprised you think it's a "niche player".
What if you had a 64-bit machine with a 1TB swap partition?
I think "can not" means that someone is unable to do something. Not that it is impossible in principle. For instance, I can not play the trombone.
And I suppose you made the leap from "can not" to "impossible"?
The only problem is that their snowflakes have five points.
This also covers your suggestion, because the Bells would already be a branch of the government.
You bet. Surely mankind can only spend money on one space mission at a time.
Damn, you're right. I stand corrected.
If we all started talking about "Microsofting" something, I'm sure you'd get a cease-and-desist, and you'd deserve it. The word is trademarked.
Not bad, until you try to port your software to Windows. :-)
Ok, you're right. I stand corrected. There's more to the quality of a program than the quality of its code. There are installation procedures, documentation, support, etc.
Now if you had said that low defect rates are not the only thing in the world, I would have agreed.
Don't forget to try IBM's.
(Disclaimer: I work for IBM on on their Java JIT compiler, and therefore cannot be trusted on this matter.)
I think you mean compu-hyper-global-mega-SPARC.
(Or, if you want to claim that testicles' DNA doesn't age, then why not clone babies from testicle cells?)
I'm not claiming I know anything about cloning. I probably know less than you do. I'm just claiming that you don't know enough to say cloning is impossible. Your argument from your first post doesn't hold water, because if that argument were sufficient, then it would also rule out sexual reproduction.You are speculating. There's nothing wrong with that, unless of course you try to claim that you are not speculating.
Clearly there's more to genetics than you know.
Ok, thanks. You clearly know more about this than I do.
It depends on what is the common case. If blocks usually do match, then the additional CRC would be pure overhead.
MD5 (or actually MD4) is a cryptographically strong checksum, so the chances of the checksum being the same for two different blocks is truly 2^-128. CRCs, on the other hand, were not designed for this, and so the probability of a collision is much higher.
We're not talking about detecting single-bit errors, sequences of bit errors, etc. that CRC is good at; we're talking about finding arbitrary differences between two blocks of data. It's not hard to construct two blocks with minor differences that have the same CRC, but only a brute-force search will find two blocks with the same MD4.
Think long and hard about why 6 connections from the same client to the same server gets the file in 1/6 the time, and you should have your answer.
Yes, you have successfully demonstrated that you got someone else's joke.
That, combined with phrases like "vehemently poking at his tuna salad", make this article a real winner all around.