Web apps these days consist nearly entirely of dynamic content invisible to googlebot. If you try to make your page visible on the web, this is really a problem. But think twice before adding invisible div's or alike in order to achieve proper seach results: Google might as well ban you (since they don't check whether or not the keywords you name in your invisible divs do in fact relate to the page's purpose or contents).
OTOH, there's nothing wrong at all with having static content that is only displayed to people who do not have javascript active. In fact, this is positively encouraged. AFAIK, google does index the content of tags.
The operating system on a multiple-core machine can split up the processes but one process can only run on one core unless it has been written in a multi-threaded fashion.
In parallel processing general each machine is running one part of a program, thus one program, and unless that program is multi-threaded as well as parallel then it can only use one core per node on a cluster.
Though, someone who writes multi-threaded parallel applications should be held in high esteem! I don't know any such coders.
Have you considered that if you run two copies of the process on each node, it will use both cores?
I'm considering building a vbox (linux,xp,vista) but I'm not sure what I would gain over my current setup of three shuttles linked via a kvm...other than the electric bill.
I lose graphics acceleration, except for which ever OS acts as host. Anything else lost?
A little performance, again on everything but your host.
You gain flexibility and peek power: need more RAM in one of those three machines? Just shut one of the others down and change the VM config. Want a fast processor? No need to split your budget between three different machines now; whichever you're using gets the processor power.
OK, so it's only for Macs so far. But that's a step in the right direction...
VMGL -- this one won't work for Windows guests, but can be used for Linux guests. A similar approach could definitely work for Windows guests, but you'd need to write a DirectX-compatible driver that translates the DX API calls into paravirtualized OpenGL API calls. Tricky, but I imagine possible.
I can think of a situation where one machine has an opengl window open and you go to the other machine to play an FPS, what will happen?
At least as of late last year, neither Xen nor VMware supported 3d acceleration. Although there was some discussion of it on the Xen mailing list, and with the release of Vista I suspect it has become somewhat more of a priority, so perhaps we can expect to see it soon. I think the primary problem was that writing 3d card drivers for Windows is actually pretty tricky.
I suspect what will happen when support exists is that it will emulate a 3d card using calls to opengl. This should mean that it will work happily.
Only just. My PII-400 cannot play DVDs properly; the audio quickly drifts out of sync with the video (which I'm informed is a symptom of a too-slow processor, even though it sounds like a bug in the playback software).
Say you have 4 VMs of Windows 2003 running. All of the pages of memory that are the same (say, core kernel pages and the like) get mapped to the same physical page. The guest operating systems never know. You can effectively scoop up a lot of extra memory if you have a lot of systems running the same software. All of those common libraries and Windows/Linux processes are only paid for once in memory.
Does this actually work? I know it's theoretically possible, but I understand that in practice it's quite hard to achieve. Do all of the current VM systems actually achieve it?
Why do I need to run 4 instances of Windows Server 2003? It's a modern, pretty bullet-proof OS with protected memory. Why can't I run all the apps in your 4 instances on a single instance?
Because you may not be able to run your apps in a single instance. They may be operated by different individuals who for security reasons are not allowed to access each others' data. They may all need to bind to the same TCP port, and aren't multihomed server aware.
Because separating them into 4 OS instances gives you better managability. You can split the memory on the server in a fixed ratio between them, preventing one application with a memory leak negatively effecting the others. You can specify CPU usage ratios with a system much more flexible than Windows' thread priority scheme. If you decide you need to move one application to a different machine, you can move it across without needing to reinstall. With some systems, you can move it without needing to reboot: the move will be completely transparent.
(And get 100% of my CPU).
CPU overhead on a modern virtualized system is pretty low. You'll hardly notice it. I/O performance is a little worse, but even that isn't particularly bad.
Also, make sure to try OpenVZ before you try Xen. If you are virtualizing all Linux machines, then VZ is IMO a better choice.
I've not used OpenVZ (or Virtuozzo), but I've spent a while using an alternative system based on the same principles (OpenVSD) and I have to say that the approach is not without its disadvantages, particularly in terms of software that is incompatible due to there not being a real root account available. It also doesn't isolate your virtual servers' memory requirements from each other like a VM solution would do, which cuts both ways: it means you automatically get an across-the-board optimal memory allocation, but it means that a misbehaving server can affect all the servers running on the same host.
Of course it didn't help that OpenVSD was an unstable pile of **** that caused servers to completely fail whenever they run out of space on/tmp, requiring a repair by the host machine's root user.
Another alternative (frequently overlooked these days) is User Mode Linux. Again, there are plusses and minuses.
Cygwin isn't a Linux emulator. It's a development environment for Windows based on POSIX-compliant open-source software. It includes the capability of developing native windows applications.
I suspect he was pressured to put fake credentials in because people were requesting that editors have credentials.
Most people who've spent any time analysing the situation come to the conclusion that he adopted the credentials in order to able to browbeat anyone who disagreed with him on his pet topic into submission using an "I know better than you do" attitude. When clearly he didn't know better.
This is a feeling I have had for quite a while now, though it's not usually a very popular view so I'm forced to keep it quiet.
Why forced to keep it quiet? I mean, I've been a fairly vocal critic of Stallman for years now, and I haven't been visited by the FSF Thought Police yet. Wait, who's that at the door...
*long delay*
What was I saying? Oh yes. Stallman's right. I was wrong. I know it now. I love Stallman...
Getting the tools to do so is many times easier (faster, cheaper, less confusion, etc.) for me on Linux than on Windows. A colleague recently suggested I try quantlib. He also mentioned that they require Boost which can be a real pain in the ass to get compiled and installed on an XP machine. I went home and installed both of these libraries in 10s of seconds with Synaptic.
When I needed Boost on Windows it was incredibly easy. I ran cygwin setup, opened the 'libraries' branch, clicked 'boost' and then OK. It downloaded & installed automatically.
Re:Just learn about Turkish government
on
Turkey Censors YouTube
·
· Score: 4, Insightful
Mustafa Kemal Ataturk replaced something. Why don't you check out what he replaced. You will understand VERY clearly why he's a saint in Turkey.
I don't care what he did. Winston Churchill did great things when he lead Britain to defeat the Nazis, yet I can call him a fat drunken slob without fear of recrimination if I so desire. That's what freedom's about: not having to care about offending people just because they did something important.
Emphasis is obviously mine. Bascially, Turkey could argue that their laws are not restrictive to free speech, and that their laws only "protect the morals [and] reputation" of the citizens of its country. (Both past and present.)
Protection of morals has been used as a get-out in the past; it is how censorship of pornography and/or "hate speech" is typically justified. I just don't see how it would apply in this case, as (if I read the summary correctly) nothing even approximately moral is involved. The issue is purely political.
I believe the protection of the reputation of others has been interpreted by the court in the past as only applying to those who are still alive, which would rule it out in this case.
Other EU countries can try to make a stink about it, but I seriously doubt that anyone is going to push Turkey too hard.
Of course, anybody can petition the European Court of Human Rights these days. I'm not sure how much profit YouTube will be losing here, but I'd be unsurprised if they weren't at least considering doing so right now.
Thinking about it I find it a bit ironic that the country I live in, Denmark (member of the EU) have done exactly the same thing with allofmp3 that Turkey now did to youtube, yet Denmark is probably quite well known for their so called defense of freedom of speech, latest demonstrated by some drawings of Muhammed.
Of course there are differences between the right to free speech and the right to distribute unlicensed music.
Anyhow I kind of understand if the Turkish negotiators will think of the EU as talking with two tongues if they start telling them that Turkey can't do this kind of thing if they want to be part of the EU.
Frankly, Turkey is already a signatory to the European Convention on Human Rights, which effectively grants Google/YouTube/anyone who was distributing a video via YouTube the right to compensation from the Turkish government via the European Court of Human Rights over this. EU membership is irrelevant, largely.
Roaming profiles can (and by default do) copy your ENTIRE user profile, so it does work
Yes, OK, it works. Barely. The problem is that it stores temporary cache information in this directory, which should be stored in the 'Local Settings' directory so that it *isn't* copied. This resulted in login times on my network of in excess of five minutes after I'd been using it for a few weeks.
(many other F/OSS apps do the same & they work with our roaming profiles).
I've had exactly the same problem with GIMP before now; it stores its tile cache outside of the local directory, causing it to be copied across the network during login. Not exactly sensible behaviour.
I just don't understand the following: the pulse/no pulse thing is itself a bit of information.
Because of the way the experiment is set up, the pulse has to arrive; you can predict that it will arrive because of previous things that have happened. Basically, as I understand the experiment, a sequence of short pulses of light are sent down the chamber, with known gaps between them. The 'faster than light' wave results from the phase motion of these normal speed light waves. By the time it starts propogating, you can already tell that it will do so from observations you can make at the end of its run.
but it is certainly up and running, so not only is a port in progress, it looks like it is even usable already.
Having recently tried to use it, I'd say no. There are several major issues:
* Redraws are nightmarishly slow (admittedly this could be because I'm using an old PC, but I haven't seen any application redraw this slowly before). * Initial configuration doesn't seem to work entirely correctly: if you need to change between SSL modes for an IMAP connection, you have to restart the program, but nothing tells you this. This may or may not be a Windows-only issue, I don't know. * It stores its files in a subdirectory called ".evolution" of your user profile directory, not your application data or local settings directory. If you're using roaming profiles, this just plain won't work.
Slashdot Mirror
Web apps these days consist nearly entirely of dynamic content invisible to googlebot. If you try to make your page visible on the web, this is really a problem. But think twice before adding invisible div's or alike in order to achieve proper seach results: Google might as well ban you (since they don't check whether or not the keywords you name in your invisible divs do in fact relate to the page's purpose or contents).
OTOH, there's nothing wrong at all with having static content that is only displayed to people who do not have javascript active. In fact, this is positively encouraged. AFAIK, google does index the content of tags.
That's cute, you think the current British government gives a flying fuck about protests, or indeed, what "the people" think.
That'd be the government that has banned protests outside parliament, right?
The operating system on a multiple-core machine can split up the processes but one process can only run on one core unless it has been written in a multi-threaded fashion.
In parallel processing general each machine is running one part of a program, thus one program, and unless that program is multi-threaded as well as parallel then it can only use one core per node on a cluster.
Though, someone who writes multi-threaded parallel applications should be held in high esteem! I don't know any such coders.
Have you considered that if you run two copies of the process on each node, it will use both cores?
I'm considering building a vbox (linux,xp,vista) but I'm not sure what I would gain over my current setup of three shuttles linked via a kvm...other than the electric bill.
I lose graphics acceleration, except for which ever OS acts as host. Anything else lost?
A little performance, again on everything but your host.
You gain flexibility and peek power: need more RAM in one of those three machines? Just shut one of the others down and change the VM config. Want a fast processor? No need to split your budget between three different machines now; whichever you're using gets the processor power.
And lo and behold, there is movement:
VMWare Fusion Beta 2 comes with "Experimental 3D Acceleration"
OK, so it's only for Macs so far. But that's a step in the right direction...
VMGL -- this one won't work for Windows guests, but can be used for Linux guests. A similar approach could definitely work for Windows guests, but you'd need to write a DirectX-compatible driver that translates the DX API calls into paravirtualized OpenGL API calls. Tricky, but I imagine possible.
I can think of a situation where one machine has an opengl window open and you go to the other machine to play an FPS, what will happen?
At least as of late last year, neither Xen nor VMware supported 3d acceleration. Although there was some discussion of it on the Xen mailing list, and with the release of Vista I suspect it has become somewhat more of a priority, so perhaps we can expect to see it soon. I think the primary problem was that writing 3d card drivers for Windows is actually pretty tricky.
I suspect what will happen when support exists is that it will emulate a 3d card using calls to opengl. This should mean that it will work happily.
a DVD plays fine on my 500 MHZ machine.
Only just. My PII-400 cannot play DVDs properly; the audio quickly drifts out of sync with the video (which I'm informed is a symptom of a too-slow processor, even though it sounds like a bug in the playback software).
Say you have 4 VMs of Windows 2003 running. All of the pages of memory that are the same (say, core kernel pages and the like) get mapped to the same physical page. The guest operating systems never know. You can effectively scoop up a lot of extra memory if you have a lot of systems running the same software. All of those common libraries and Windows/Linux processes are only paid for once in memory.
Does this actually work? I know it's theoretically possible, but I understand that in practice it's quite hard to achieve. Do all of the current VM systems actually achieve it?
Why do I need to run 4 instances of Windows Server 2003? It's a modern, pretty bullet-proof OS with protected memory. Why can't I run all the apps in your 4 instances on a single instance?
Because you may not be able to run your apps in a single instance. They may be operated by different individuals who for security reasons are not allowed to access each others' data. They may all need to bind to the same TCP port, and aren't multihomed server aware.
Because separating them into 4 OS instances gives you better managability. You can split the memory on the server in a fixed ratio between them, preventing one application with a memory leak negatively effecting the others. You can specify CPU usage ratios with a system much more flexible than Windows' thread priority scheme. If you decide you need to move one application to a different machine, you can move it across without needing to reinstall. With some systems, you can move it without needing to reboot: the move will be completely transparent.
(And get 100% of my CPU).
CPU overhead on a modern virtualized system is pretty low. You'll hardly notice it. I/O performance is a little worse, but even that isn't particularly bad.
to sum up tfa: poor planning and execution are the cause of problems.
You missed one: proprietary software licenses cause legal difficulties sometimes, too.
Also, make sure to try OpenVZ before you try Xen. If you are virtualizing all Linux machines, then VZ is IMO a better choice.
/tmp, requiring a repair by the host machine's root user.
I've not used OpenVZ (or Virtuozzo), but I've spent a while using an alternative system based on the same principles (OpenVSD) and I have to say that the approach is not without its disadvantages, particularly in terms of software that is incompatible due to there not being a real root account available. It also doesn't isolate your virtual servers' memory requirements from each other like a VM solution would do, which cuts both ways: it means you automatically get an across-the-board optimal memory allocation, but it means that a misbehaving server can affect all the servers running on the same host.
Of course it didn't help that OpenVSD was an unstable pile of **** that caused servers to completely fail whenever they run out of space on
Another alternative (frequently overlooked these days) is User Mode Linux. Again, there are plusses and minuses.
Cygwin isn't a Linux emulator. It's a development environment for Windows based on POSIX-compliant open-source software. It includes the capability of developing native windows applications.
There's no reason not to use it.
The funny thing is that EssJay is now notable enough to have his own Wikipedia entry.
Which has already been considered for deletion twice. I doubt it'll last long.
I suspect he was pressured to put fake credentials in because people were requesting that editors have credentials.
Most people who've spent any time analysing the situation come to the conclusion that he adopted the credentials in order to able to browbeat anyone who disagreed with him on his pet topic into submission using an "I know better than you do" attitude. When clearly he didn't know better.
This is a feeling I have had for quite a while now, though it's not usually a very popular view so I'm forced to keep it quiet.
Why forced to keep it quiet? I mean, I've been a fairly vocal critic of Stallman for years now, and I haven't been visited by the FSF Thought Police yet. Wait, who's that at the door...
*long delay*
What was I saying? Oh yes. Stallman's right. I was wrong. I know it now. I love Stallman...
Getting the tools to do so is many times easier (faster, cheaper, less confusion, etc.) for me on Linux than on Windows. A colleague recently suggested I try quantlib. He also mentioned that they require Boost which can be a real pain in the ass to get compiled and installed on an XP machine. I went home and installed both of these libraries in 10s of seconds with Synaptic.
When I needed Boost on Windows it was incredibly easy. I ran cygwin setup, opened the 'libraries' branch, clicked 'boost' and then OK. It downloaded & installed automatically.
Mustafa Kemal Ataturk replaced something. Why don't you check out what he replaced. You will understand VERY clearly why he's a saint in Turkey.
I don't care what he did. Winston Churchill did great things when he lead Britain to defeat the Nazis, yet I can call him a fat drunken slob without fear of recrimination if I so desire. That's what freedom's about: not having to care about offending people just because they did something important.
Emphasis is obviously mine. Bascially, Turkey could argue that their laws are not restrictive to free speech, and that their laws only "protect the morals [and] reputation" of the citizens of its country. (Both past and present.)
Protection of morals has been used as a get-out in the past; it is how censorship of pornography and/or "hate speech" is typically justified. I just don't see how it would apply in this case, as (if I read the summary correctly) nothing even approximately moral is involved. The issue is purely political.
I believe the protection of the reputation of others has been interpreted by the court in the past as only applying to those who are still alive, which would rule it out in this case.
Other EU countries can try to make a stink about it, but I seriously doubt that anyone is going to push Turkey too hard.
Of course, anybody can petition the European Court of Human Rights these days. I'm not sure how much profit YouTube will be losing here, but I'd be unsurprised if they weren't at least considering doing so right now.
Thinking about it I find it a bit ironic that the country I live in, Denmark (member of the EU) have done exactly the same thing with allofmp3 that Turkey now did to youtube, yet Denmark is probably quite well known for their so called defense of freedom of speech, latest demonstrated by some drawings of Muhammed.
Of course there are differences between the right to free speech and the right to distribute unlicensed music.
Anyhow I kind of understand if the Turkish negotiators will think of the EU as talking with two tongues if they start telling them that Turkey can't do this kind of thing if they want to be part of the EU.
Frankly, Turkey is already a signatory to the European Convention on Human Rights, which effectively grants Google/YouTube/anyone who was distributing a video via YouTube the right to compensation from the Turkish government via the European Court of Human Rights over this. EU membership is irrelevant, largely.
Your point? That's how settings are saved in every other operating system.
My point is that when porting software to a new operating system, changing its behaviour so that it respects local platform conventions is desirable.
Roaming profiles can (and by default do) copy your ENTIRE user profile, so it does work
Yes, OK, it works. Barely. The problem is that it stores temporary cache information in this directory, which should be stored in the 'Local Settings' directory so that it *isn't* copied. This resulted in login times on my network of in excess of five minutes after I'd been using it for a few weeks.
(many other F/OSS apps do the same & they work with our roaming profiles).
I've had exactly the same problem with GIMP before now; it stores its tile cache outside of the local directory, causing it to be copied across the network during login. Not exactly sensible behaviour.
I just don't understand the following: the pulse/no pulse thing is itself a bit of information.
Because of the way the experiment is set up, the pulse has to arrive; you can predict that it will arrive because of previous things that have happened. Basically, as I understand the experiment, a sequence of short pulses of light are sent down the chamber, with known gaps between them. The 'faster than light' wave results from the phase motion of these normal speed light waves. By the time it starts propogating, you can already tell that it will do so from observations you can make at the end of its run.
Exactly. It's exactly the same experiment we've seen time & time again, and it's meaningless because no information is transmitted.
but it is certainly up and running, so not only is a port in progress, it looks like it is even usable already.
Having recently tried to use it, I'd say no. There are several major issues:
* Redraws are nightmarishly slow (admittedly this could be because I'm using an old PC, but I haven't seen any application redraw this slowly before).
* Initial configuration doesn't seem to work entirely correctly: if you need to change between SSL modes for an IMAP connection, you have to restart the program, but nothing tells you this. This may or may not be a Windows-only issue, I don't know.
* It stores its files in a subdirectory called ".evolution" of your user profile directory, not your application data or local settings directory. If you're using roaming profiles, this just plain won't work.
Isn't that what SuSE OpenExchange is?