I see. Yes. I would posit that the right way to have handled this would be to ascribe a power to any retired president to have any law or prior law passed to be be brought before the supreme court as a constitutional case. The court would elect to hear or not hear the case as per normal.
Jack of all trades. This isn't so bad, you just need to start interviewing and communicating that. Lots of companies are looking for people who can fill multiple roles. It's a common matter of discussing at the company I work (BAE Systems). Also, someone who knows a lot about many fields at work is a good management candidate. Hope that's not insulting, har har.:)
The person you are responding to needs to understand that the constitutionality of something is really only something that can be reduced to reality in a court. You said that. You also said it's "not right," but I don't think anything could be righter. Such as it is, judicial process is the only workable system of impartiality that we have. The only better process exists over yonder, in some perfect fantasy world.
As for our Constitution itself, it's been breaking at the seams for years. At least for the "parts we don't like." It took an Amendment to make Alcohol illegal, and everyone knew that. No Amendment for federal drug laws, though, eh? They're all Unconsitutional on their face. The States can do that, but not the Feds.
I am not free to own slaves. Am I restricted, or is everyone else more free?
Following this thread (and seeing the later rephrasing also), I would state that there are important definitional matters at hand here. I regard a theoretical someone complaining about not being able to own slaves as not speaking about any kind of "freedom" at all, because to me there is no real freedom that is at the expense of other human beings -- by definition. I should imagine there are cases where even this view breaks down and ceases to make sense, but I wouldn't say many.
In the few that I can think of, I'm reminded of an old quote that said something like "all government is evil, but sometimes some evil is necessary." Taxes for the common defense, where the one man is in some ways losing his "freedom" to not pay taxes, and so forth. Pretty arguable though, as he is "free" to live in a very free country like Somalia if he so elects.
I would have to agree. As an author of (small) GPL'd works, I'd certainly update. I can see some of the virtue of the spirited discussions around certain DRM terms in the GPL v3 and what not, but when the dust settles, I'll be swinging v3, without a doubt. The GPL is the community nuclear weapon used in opposition of some of the most important intellectual problems of today (in particular, our currently utterly broken patent system here in the US, which is utterly rife with abuse, particularly in the software space, but not to forget the current usurpation of fair use provisions of copyright law, so and so forth).
I don't think you understand. They're not saying that if they (the FSF) later to decide to switch to a modified license, you must use this license, rather they are saying that if they make the switch to a later version, you may use that license. I believe this is legal, unlike the opposite, which is most often illegal in English Common Law countries.
I'm pretty sure that FSF's software is distributed under the "GPL v2 or later". Which means that any commercial fork would have to keep using "GPL v2 or later". FSF would then be free to take any changes to the commercial fork, apply them to the FSF version and then release it under the GPL v3.
Certainly. But this is only bad to the collective maintaining of the v2 fork insofar as they cannot integrate code into the v2 code that comes from a v3 provenance.
I'd be amazed if your machine is down to 20W at idle though...
More modern boxes have a sort of deep sleep mode. Monitor off, disks spin down, CPU underclocks, just about everything down into zombie like status. Strike the keyboard and it spins back up, lots faster than booting, next best thing to power off. Good compromise, see? You do have to configure it, but it's great. It comes back on line very fast, and better yet I have my prior session state. Much better than a reboot.
Jesus Christ. My machine, a dual core athlon, draws about 118W when I'm on it. Like right now, I just looked under my desk at the energy meter. Granted, I'm not playing a game, just reading slash, but sheeesssh. When I walk away from the box and all the power saving features go on, it's like 20W.
Curiously, the article doesn't mention any specific problems. I'm racking my tiny brain right now to think of some problem that isn't desired by the RIAA.
The support phone call. Each and every one of them is like tick after tick after depressing tick of red ink.
[i]Here's a crazy novel idea for you: Why just not stop wasting so much energy in the first place?[/i]
A coworker of mine was going on and on about some new electric car that is coming out in a lotus like body. He was fawning over the lightweight composites and other clever weight reductions, like engines in the wheels so that there doesn't have to be the whole cam-shafty thingy.
What occurred to me here was that one of the best features of this vehicle is that it is lighter. That's definitionally energy-conservative. Neat stuff.
Well. VMWare has issues with IO latency. One has to watch for that, not try to virtualize everything. But. You say "Virtualization bad" for "CPU intensive," and I cannot agree with that. SPECint2006 and SPECfp2006, as well as rates are within 5% of hard metal for ESX. I've run the tests myself. Old school "CPU intensive" applications are a non-conversation in in virtualizaation today.
It's the network IO and network latency that will kill you if you don't know what you're doing. VMWare has known issues in that area (although they must break through these entirely or 10GE will never work properly in VMWare). One can work around these issues, however I'd simply say it's a Best Practice to simply plan to "not virtualize everything." I'd say target 65%-ish of your compute infrastructure in preplanning and base your real decisions on an actual analysis.
The truth is, it doesn't much make sense that property values would be outstripping the income increases.
I know out here in California, a very strong factor in local property value increases was foreign investment. This particularly applied to "estate" property. A lot of wealthy Chinese were buying mansions out here as a hedge against economic changes in China. Or so it's been claimed.
Anyway, when they refer to the "new economy," they are often talking about forces like foreign investment propping up local markets.
"Symmetric reference system"; C# can pass value types by reference. The caller and the callee must agree and both label it "ref". This makes the semantics clear to the code reader, who is given an immediate signal that his value type may be changed as a consequence of the invocation of the method.
Non jagged arrays have much higher performance than jagged arrays due to memory adjacency issues. That's why every single high-performance-computing library in the world uses them.
"abort case goto". Contrary to what your computer professor told you, there are a few special use cases for "goto" that are not only "bad," but rather are "good." C# supports them.
Writing "override" is helpful, as it (like the ref case above), makes clear the intent and signals to future readers the semantics of the situation.
Swing is slow. Everyone knows that. That's why people mistakenly label the entire Java suite as "slow," because Swing is godawful slow.
Anyway. I agree that there are some better things in Java. "Default package visibility" is top on my list. I can see what you mean about class loading.
As for the "no multiple inheritance" issue I agree... I poisonously agree!:) Every argument I've heard against it is compiler authors whining about how hard it is to do. Well, whine, whine, whine. Works fine in Python and other dynamic languages, these guys need to grab a clue.
BTW, in my op, I mixed and matched the terms.NET with C#. That was a mispeak. It is only purely C# (the language) that I find better than Java (the language). JDK > SDK. And so forth. In fact, until the (recent) SDK 2.0, the SDK was an embarrassment. You should have seen the container classes! Hello, anyone home?
I suppose if I were to summarize in a nutshell why I like C# better than Java, I would say that it appears to me that the author was more concerned with the practical job of day to day program than he was concerned with maintain some kind of zealous adherence to "purity".
What you fail to fully comprehend is that the very wealthy enjoy the ability to decide whether or not their cash flow is expressed as "income" at all. Be that as it may, I don't think sales taxes are the best solution. Property taxes are. They cannot be dodged in any meaningful way, and the rich tend to hold onto gobbles and gobbles of property.
It's in fact most likely that the drives will fail at closely clustered periods of time. I.e., if drives are failing, there's a SIGNIFICANTLY non zero chance that others will be failing really soon.
I really would like to know what parts of C# you find well done.
Delegates (old style) Delegates (new style, with closures) Symmetric reference system... both practical AND clear. Autoboxing. Properties Jagged and non jagged arrays. generics structs (yes sometimes the stack really is better) enums sensible support of abort-case (fallout case) goto (Hejlsberg actually read Djikstra... gasp!) P/Invoke.. hate it if you like, but it's very practical
Some things I don't like:
Access semantics. Java is better, with default package visibility (default package visibility strikes me as "correct").
The format system, which is the worst one I've ever used. Why do people keep reinventing printf?
The spurious changing of keywords from Java for change's sake. Annoying.
For everyone whose last memory of Java is 10 years ago, you need to look at it again. Aside from basic single platform desktop applications,.NET is where Java was 10 years ago.
Your comments about J2EE are fair. Microsoft simply does not have that level of an ecology in that space.
For my hobby time, I use.NET to write fat desktop clients and raw server code for which J2EE or SOA would be of no interest. Down in the fundaments, C# was a rethink of Java, not just a copy, and it improved on a lot of things which Java got wrong. Some of them are pretty simple. Like the brain-dead com.company.package.subpackage carpal-tunnel crap. And so forth. C# is more modern than Java and has had a chance to learn from Java's mistakes. This plainly shows.
BTW, about Microsoft copied Java litany: it's tired, old, maladroit, boring. Give it up.
Slashdot Mirror
Mod -1, D&D DORK :)
I don't think so. Wherein the v3 license does it attempt to make assertions about agreements made under terms of the v2 license?
C//
I see. Yes. I would posit that the right way to have handled this would be to ascribe a power to any retired president to have any law or prior law passed to be be brought before the supreme court as a constitutional case. The court would elect to hear or not hear the case as per normal.
C//
Jack of all trades. This isn't so bad, you just need to start interviewing and communicating that. Lots of companies are looking for people who can fill multiple roles. It's a common matter of discussing at the company I work (BAE Systems). Also, someone who knows a lot about many fields at work is a good management candidate. Hope that's not insulting, har har. :)
C//
The person you are responding to needs to understand that the constitutionality of something is really only something that can be reduced to reality in a court. You said that. You also said it's "not right," but I don't think anything could be righter. Such as it is, judicial process is the only workable system of impartiality that we have. The only better process exists over yonder, in some perfect fantasy world.
As for our Constitution itself, it's been breaking at the seams for years. At least for the "parts we don't like." It took an Amendment to make Alcohol illegal, and everyone knew that. No Amendment for federal drug laws, though, eh? They're all Unconsitutional on their face. The States can do that, but not the Feds.
C//
- You can't write a compiler
True.
- You can't debug C/C++ programs
Totally false.
- You don't really know why buffer overflows are bad
I'd say you "don't know, practically, how they go about being bad."
- You don't really understand memory management and what the heap and stack really are
Crock.
- You don't really know why threads are different than processes
Bull.
- You can't write a device driver
True.
- You don't know any computer architecture at any depth that matters
Crap.
- You won't ever understand garbage collection
Lie.
- You don't know how your CPU works
Hmmm. Maybe true.
- You won't think the movies with "hacking" in them are as funny as the rest of us do.
Well, at least you ended with "funny".
C//
I am not free to own slaves. Am I restricted, or is everyone else more free?
Following this thread (and seeing the later rephrasing also), I would state that there are important definitional matters at hand here. I regard a theoretical someone complaining about not being able to own slaves as not speaking about any kind of "freedom" at all, because to me there is no real freedom that is at the expense of other human beings -- by definition. I should imagine there are cases where even this view breaks down and ceases to make sense, but I wouldn't say many.
In the few that I can think of, I'm reminded of an old quote that said something like "all government is evil, but sometimes some evil is necessary." Taxes for the common defense, where the one man is in some ways losing his "freedom" to not pay taxes, and so forth. Pretty arguable though, as he is "free" to live in a very free country like Somalia if he so elects.
C//
I would have to agree. As an author of (small) GPL'd works, I'd certainly update. I can see some of the virtue of the spirited discussions around certain DRM terms in the GPL v3 and what not, but when the dust settles, I'll be swinging v3, without a doubt. The GPL is the community nuclear weapon used in opposition of some of the most important intellectual problems of today (in particular, our currently utterly broken patent system here in the US, which is utterly rife with abuse, particularly in the software space, but not to forget the current usurpation of fair use provisions of copyright law, so and so forth).
C//
A retrospective licence? How free is that?
I don't think you understand. They're not saying that if they (the FSF) later to decide to switch to a modified license, you must use this license, rather they are saying that if they make the switch to a later version, you may use that license. I believe this is legal, unlike the opposite, which is most often illegal in English Common Law countries.
C//
I'm pretty sure that FSF's software is distributed under the "GPL v2 or later". Which means that any commercial fork would have to keep using "GPL v2 or later". FSF would then be free to take any changes to the commercial fork, apply them to the FSF version and then release it under the GPL v3.
Certainly. But this is only bad to the collective maintaining of the v2 fork insofar as they cannot integrate code into the v2 code that comes from a v3 provenance.
C//
I'd be amazed if your machine is down to 20W at idle though...
More modern boxes have a sort of deep sleep mode. Monitor off, disks spin down, CPU underclocks, just about everything down into zombie like status. Strike the keyboard and it spins back up, lots faster than booting, next best thing to power off. Good compromise, see? You do have to configure it, but it's great. It comes back on line very fast, and better yet I have my prior session state. Much better than a reboot.
Joe.
My machine: a twin Xeon, draws 140W at idle....
Jesus Christ. My machine, a dual core athlon, draws about 118W when I'm on it. Like right now, I just looked under my desk at the energy meter. Granted, I'm not playing a game, just reading slash, but sheeesssh. When I walk away from the box and all the power saving features go on, it's like 20W.
C//
Curiously, the article doesn't mention any specific problems. I'm racking my tiny brain right now to think of some problem that isn't desired by the RIAA.
The support phone call. Each and every one of them is like tick after tick after depressing tick of red ink.
C//
[i]Here's a crazy novel idea for you: Why just not stop wasting so much energy in the first place?[/i]
A coworker of mine was going on and on about some new electric car that is coming out in a lotus like body. He was fawning over the lightweight composites and other clever weight reductions, like engines in the wheels so that there doesn't have to be the whole cam-shafty thingy.
What occurred to me here was that one of the best features of this vehicle is that it is lighter. That's definitionally energy-conservative. Neat stuff.
C//
Perhaps the op meant the "spelling and grammer police".
Grammar Nazis seem to be "loosing" the war against spelling idiocy. To (SIC!) bad Slashdot isn't edited.
This is an intentional joke, yes?
C//
I also recall Microsoft doing some mandatory virtualization elements in the DirectX 10 driver spec, although I'm not sure who benefits here. :)
VMWare HA *reboots* the VM on another blade. Not only is there down time, but all TCP connections are dropped, just as is in any blade.
The purpose of HA is to avoid having to reimagt new hardware in response to a hardware failure. It's good, yes. But pretty grossly grained.
VMotion can move your VM over to another blade without losing connections (actual 'downtime' is planned downtime.
Hope this helps.
C//
Well. VMWare has issues with IO latency. One has to watch for that, not try to virtualize everything. But. You say "Virtualization bad" for "CPU intensive," and I cannot agree with that. SPECint2006 and SPECfp2006, as well as rates are within 5% of hard metal for ESX. I've run the tests myself. Old school "CPU intensive" applications are a non-conversation in in virtualizaation today.
It's the network IO and network latency that will kill you if you don't know what you're doing. VMWare has known issues in that area (although they must break through these entirely or 10GE will never work properly in VMWare). One can work around these issues, however I'd simply say it's a Best Practice to simply plan to "not virtualize everything." I'd say target 65%-ish of your compute infrastructure in preplanning and base your real decisions on an actual analysis.
C//
The truth is, it doesn't much make sense that property values would be outstripping the income increases.
I know out here in California, a very strong factor in local property value increases was foreign investment. This particularly applied to "estate" property. A lot of wealthy Chinese were buying mansions out here as a hedge against economic changes in China. Or so it's been claimed.
Anyway, when they refer to the "new economy," they are often talking about forces like foreign investment propping up local markets.
C//
"Symmetric reference system"; C# can pass value types by reference. The caller and the callee must agree and both label it "ref". This makes the semantics clear to the code reader, who is given an immediate signal that his value type may be changed as a consequence of the invocation of the method.
:) Every argument I've heard against it is compiler authors whining about how hard it is to do. Well, whine, whine, whine. Works fine in Python and other dynamic languages, these guys need to grab a clue.
.NET with C#. That was a mispeak. It is only purely C# (the language) that I find better than Java (the language). JDK > SDK. And so forth. In fact, until the (recent) SDK 2.0, the SDK was an embarrassment. You should have seen the container classes! Hello, anyone home?
Non jagged arrays have much higher performance than jagged arrays due to memory adjacency issues. That's why every single high-performance-computing library in the world uses them.
"abort case goto". Contrary to what your computer professor told you, there are a few special use cases for "goto" that are not only "bad," but rather are "good." C# supports them.
Writing "override" is helpful, as it (like the ref case above), makes clear the intent and signals to future readers the semantics of the situation.
Swing is slow. Everyone knows that. That's why people mistakenly label the entire Java suite as "slow," because Swing is godawful slow.
Anyway. I agree that there are some better things in Java. "Default package visibility" is top on my list. I can see what you mean about class loading.
As for the "no multiple inheritance" issue I agree... I poisonously agree!
BTW, in my op, I mixed and matched the terms
I suppose if I were to summarize in a nutshell why I like C# better than Java, I would say that it appears to me that the author was more concerned with the practical job of day to day program than he was concerned with maintain some kind of zealous adherence to "purity".
Technological zealotry always turns me off.
C//
What you fail to fully comprehend is that the very wealthy enjoy the ability to decide whether or not their cash flow is expressed as "income" at all. Be that as it may, I don't think sales taxes are the best solution. Property taxes are. They cannot be dodged in any meaningful way, and the rich tend to hold onto gobbles and gobbles of property.
C//
It's in fact most likely that the drives will fail at closely clustered periods of time. I.e., if drives are failing, there's a SIGNIFICANTLY non zero chance that others will be failing really soon.
C//
I really would like to know what parts of C# you find well done.
Delegates (old style)
Delegates (new style, with closures)
Symmetric reference system... both practical AND clear.
Autoboxing.
Properties
Jagged and non jagged arrays.
generics
structs (yes sometimes the stack really is better)
enums
sensible support of abort-case (fallout case) goto (Hejlsberg actually read Djikstra... gasp!)
P/Invoke.. hate it if you like, but it's very practical
Some things I don't like:
Access semantics. Java is better, with default package visibility (default package visibility strikes me as "correct").
The format system, which is the worst one I've ever used. Why do people keep reinventing printf?
The spurious changing of keywords from Java for change's sake. Annoying.
For everyone whose last memory of Java is 10 years ago, you need to look at it again. Aside from basic single platform desktop applications, .NET is where Java was 10 years ago.
.NET to write fat desktop clients and raw server code for which J2EE or SOA would be of no interest. Down in the fundaments, C# was a rethink of Java, not just a copy, and it improved on a lot of things which Java got wrong. Some of them are pretty simple. Like the brain-dead com.company.package.subpackage carpal-tunnel crap. And so forth. C# is more modern than Java and has had a chance to learn from Java's mistakes. This plainly shows.
Your comments about J2EE are fair. Microsoft simply does not have that level of an ecology in that space.
For my hobby time, I use
BTW, about Microsoft copied Java litany: it's tired, old, maladroit, boring. Give it up.
C//