Your comment about regulation is nonsense, there is too much importation from China to inspect and regulate, it's impossible.
No, it's not. It's expensive. And perhaps we should do exactly that. It would raise the price of imports while providing jobs, which would stimulate an increase in manufacturing domestically. That would be a win-win-win for the US.
I did say at least through all versions of XP, and as one commenter pointed out, this continued through Server 2003, which makes sense since XP64 (based on Server 2003) still exhibited this extremely annoying behavior.
I haven't used any MS products for my main desktop in over 5 years, so I can't comment on Vista or W7, except that my exposure to W7 and Server 2008 R2 was more than a little underwhelming on the architecture and performance sides. IOW, nothing had changed, or rather, some things had gotten worse.
NT 4 moved a lot of user space Windows GDI functionality (as defined by Win 95/98/ME)
Did Win9x even have the concept of "user space" and "kernel space" as such?
Of course not, but these functions or their predecessors originally resided in user mode in the original version of NT. On that time's hardware, it was slower than molasses and only ran acceptably for office type applications.
This is one of the reasons why opening a 10MB networked file or attachment in Outlook causes your entire machine to lockup until it's done
That's bullshit. It doesn't "lockup your entire machine" - at best it would lock up Outlook, and that would be because it'd use a synchronous file API (open or read) for something that's on the network, from a UI thread. This doesn't have anything to do with GDI, or userspace/kernelspace distinctions.
You haven't tried this experiment on the appropriate software/hardware then. Downloading that through Outlook will lock up every other program that depends upon the GDI layer. This leaves any open Command Prompts responsive, and the Task Manager. You can't even open a new program from Task Manager because that also queues up in the GDI layer. From an open command prompt, you can run/launch anything that is not GDI bound. Alt-Tab still works, kind of. This particular pain is best experienced on a congested network with a large server load where a 10MB download can take 5 minutes or more.
And before you go calling BS again, not all GUI programs utilized the GDI layer. Besides Task Manager, PMMail was one, ported from OS/2, it followed OS/2 conventions on I/O, utilizing separate threads for input/output and network connectivity.
Actually, IIRC, it was Win NT 3.1 that had the initial full security model you ascribe to Win NT 3.5. Win NT 3.5 had already slid a good portion of the way down the slippery slope of Ring 0 code, including some of the graphics drivers. (Again, IIRC, it's been a while)
NT 4 moved a lot of user space Windows GDI functionality (as defined by Win 95/98/ME) into a kernel mode GDI API, which is single threaded btw, that persisted at least through all versions Windows XP, if not beyond. (This is one of the reasons why opening a 10MB networked file or attachment in Outlook causes your entire machine to lockup until it's done)
This was in contrast to OS/2, which continued to follow the original design criteria, and hence was perceived to be slower on the same hardware as NT 4 for single tasks, although multi-tasking was much faster on OS/2. I mention this because NT's original basis was the OS/2 criteria, which was then mutated to be able to support the Win 95/98/ME gaming solutions.
What you're describing applies to all human beings not just geeks. Geeks are by definition MORE rational than the average person because their interest is in trying to understand the equipment they're using. So that argument is weak, trumps nothing and does not change my mind.
Seriously? You really think so? I've met "geeks" who are nothing more than tinkerers. They don't strive to do much of anything other than put the pieces in configurations that please them.
As for the argument about 3rd party software don't be so disingenuous. You know perfectly well that to get full functionality from an Apple product you must run iTunes - all other options are partial solutions - and this is by Apple's own design. You're the sort of person that would tell a guy to RTFM if he was having trouble: Not at all helpful unless the docs are available and comprehensible at his level.
An ipod need never see iTunes to be fully usable and useful. Unless, of course, you wish to purchase items on iTunes and play them on the iPod?
I've rarely seen geeks be generally rational about anything they are passionate about. They just happen to sometimes have some actual arguments to support their view instead of being based on feelings alone.
So does my generic statement trump yours?
As for Apple products, some cost more, some less (I know - shocking!) And honestly, I cannot get over the large sentiment against iTunes on here. Don't like it, don't run it. There are other options, even if you own an iPod. iPad, or an iPhone. You just won't get updates, nor have backups. Choices.
I'd argue lossy barely makes sense for your portable device with the size of the things these days, it makes no sense anywhere other than PR or trial bits for people to see whether they might like a song. Unless, of course, you're listening to the strains of Britney Spears and company, in which case the lossier the better.
Not FUD about flac, half my music is flac and my NanoTouch (a present) won't play them - unless you know of a way of getting flac to work?
I do: Max It can losslessly convert your FLAC into ALAC. Granted, this program is old, but it will work and preserve your tag data. I converted a few hundred albums with it recently. I can also convert it back into FLAC if I like, with no losses. The beauty of lossless codecs. All it takes is a little time.
BTW, Max, being relatively old, does have a quirk every now and then. It's free, so I can't really complain since it does do the job.
What are you talking about? Windows 7 is by far the best OS Microsoft has put out...
That still doesn't mean it's any good.
Yes, I have used it, and it's big brother 2008 R2 as well. Coded system software for it. Neither is secure in any form whatsoever no matter what MS's PR machine wishes you to think. The core problem is the fact that MS chose a run model with maximum privileges that can be throttled down, instead of minimum privileges that can be escalated via authentication, for example. Then, of course, there's the issue with dynamic code injection into system DLLs no matter what your process privs and then executing that code as "root" from the injecting process. That was actually the suggested work around for the extremely short sighted (IMNSHO) MS action in 2008 R2 of removing any meaningful security token manipulation.
No matter how pretty the outside (lipstick) the OS is still underwhelming underneath (pig).
And just a side note - OSX has stagnated? 7 major revisions in 11 years changing architecture and bringing new features while maintaining a reasonable memory and disk footprint vs 3 from MS, each requiring an order of magnitude increase in memory and disk space, one being a complete failure by any measure (Vista)? Ok, I'll admit there wasn't quite an order of magnitude increase from Vista to 7, but it did increase.
If states were left to do some of these things, which they were up until several of these agencies were founded, you'd still have mud paths in places like Louisiana or Montana instead of interstates.
Of course, he's a kitten, and he probably just sees the motions anyway, but did you really think I'd had a baby? Here, a slashdot poster?
Apparently not a very alert/.er. Only slow cats can see something resembling an image, the rest see a moving dot of light (for those with CRTs) or just a flickering stutter for slower LCDs.
There's other links out there for those that thought their "bright" animal watched TV. It certainly explained why the cat was never bothered by silent cats on TV, but had the shock of his life walking around the corner and seeing himself in the new full length mirror. Poof! Twice his normal size.:)
The other advantage is that it will seamlessly work with your OS. I have a Dell U2411, and it's always brighter than I want it to be and does not have autodim capabilities, or at least none that I've discovered as of yet.
Oh, there most certainly should be a flamebait moderation. There are definite meritless posts with the sole intention of causing further flamewars or merely being inflamatory and nothing else. (Most of us don't respond to those)
However, those moderators who use flamebait on a post that most definitely should not be flamebait should be forever removed from the moderator pool (IOW, use that thing carefully, it can cut both ways.) Merely my opinion, it's as valid as any other.
Actually, I'd argue that JDK 8 isn't that important right now. People are still adjusting to JDK 7, and fixing the issues with JDK 7 should have a higher priority than doing all new stuff for JDK 8.
I'd say based on my experience that's there's still a sizable group out there that's having trouble adjusting to JDK 5 or 6. JDK 7 isn't even on the horizon yet.
JDK 7 has some serious goodness in it, for my needs anyways. And I'm fine if there's less bloat. You may see my improvements as bloat, and what I think of as bloat your core needs. Depends upon your viewpoint.
Microsoft gained a lead on Java? With what? I've done a year in.NET recently, and while it has some ok features, it's still got a long way to go to match the capabilities and dev environment provided by even Java 6 and Eclipse. And never mind the CF underneath the covers that MS calls security....
Compared to C++ or C, Java is slow. How slow depends on the workload. Anywhere from a few percent to a large multiple, the latter being fairly easy to trigger by hitting the JIT just the way it doesn't like.
If you care about performance or latency, Java is the wrong language. But it's faster than Python.
It depends upon how it's written, or are you unaware of the tests showing Java beating C++ by a large margin? Granted, they're probably all biased, intentionally or not, by the authors own capabilities, but even so, Java is not "slow", especially not in a server platform. It's also a whole lot easier to program robustly, especially by the majority of "CS Majors" currently being cranked out of schools. (That's another topic entirely) Note that there are seeming classes of problems that C++ would be faster in such as iterating across arrays of arrays of arrays, but a decent programmer can eliminate 95% of the performance hit with a proper iteration algorithm, and the remaining 5% can be removed by removing bounds checking in the JVM, bringing the operations in line with a C++ compiled runtime.
I actually read the articles. Gasp! Turns out that new techniques allow for seeing the planets in the old data. ie, new data reduction/filtering methodologies improve the results coming out of the old pictures.
1) this allows for a second source for confirmation of any suspected planets 2) it allows for viewing orbits immediately across a multi-year span 3) the validation of the new techniques allows for all the Hubble data to be analyzed anew with potentially much different results than the first time around.
Think of this as looking at a drop of pond water with your eye, a magnifying glass, and then a series of pictures taken through the magnifying glass at different angles and putting together a 3-D representation of the drop in a computer.
Open drive, remove platters, place platter on an open vise, place chisel in middle, hit with hammer. Repeat.
At this point, only the most dedicated folks will even try to get data off of the platters, if you want to be sure, break them into multiple pieces and dispose of separately, or, get a large fresnel lens and create yourself a nice solar powered smelter. Molten platters won't tell anyone anything, other than you went out of your way to destroy something.
So you're really saying don't mix 2 types of load balancers in a cluster, and that type of thing? That would be insanity in production systems.
Honestly, while I have seen this in the real world, I often wonder if those responsible for it are qualified to even touch a keyboard or a network cable. For a production rollout, I spec homogenous layers, as it removes "configuration hell" from the equation. I only have to worry about single sets of configuration per equipment type in the layers, and removes an entire set of potential complicating factors from any problem resolution I need to perform. This is merely common sense, and the only ones who don't follow it are either entirely incompetent or are intending to milk the target for non-productive work for the foreseeable future (ie, "job-security").
[1] It's time to stop fooling ourselves -- the OS is not an application, and it's not a video game. The purpose of the OS is to load applications and manage resources, not to keep us entertained.
If I had mod points today, this would get you a bump all by itself. OSes should stick to that core purpose and do it well. The interface it displays to do so may change for various reasons, such as the single full screen presentation on a 4" phone for example, as that use case wouldn't be well served by having multiple windows floating around at least for the majority of the populace. But otherwise, it should be a lean mean set of APIs for accessing resources.
Slashdot Mirror
Your comment about regulation is nonsense, there is too much importation from China to inspect and regulate, it's impossible.
No, it's not. It's expensive. And perhaps we should do exactly that. It would raise the price of imports while providing jobs, which would stimulate an increase in manufacturing domestically. That would be a win-win-win for the US.
I did say at least through all versions of XP, and as one commenter pointed out, this continued through Server 2003, which makes sense since XP64 (based on Server 2003) still exhibited this extremely annoying behavior.
I haven't used any MS products for my main desktop in over 5 years, so I can't comment on Vista or W7, except that my exposure to W7 and Server 2008 R2 was more than a little underwhelming on the architecture and performance sides. IOW, nothing had changed, or rather, some things had gotten worse.
NT 4 moved a lot of user space Windows GDI functionality (as defined by Win 95/98/ME)
Did Win9x even have the concept of "user space" and "kernel space" as such?
Of course not, but these functions or their predecessors originally resided in user mode in the original version of NT. On that time's hardware, it was slower than molasses and only ran acceptably for office type applications.
This is one of the reasons why opening a 10MB networked file or attachment in Outlook causes your entire machine to lockup until it's done
That's bullshit. It doesn't "lockup your entire machine" - at best it would lock up Outlook, and that would be because it'd use a synchronous file API (open or read) for something that's on the network, from a UI thread. This doesn't have anything to do with GDI, or userspace/kernelspace distinctions.
You haven't tried this experiment on the appropriate software/hardware then. Downloading that through Outlook will lock up every other program that depends upon the GDI layer. This leaves any open Command Prompts responsive, and the Task Manager. You can't even open a new program from Task Manager because that also queues up in the GDI layer. From an open command prompt, you can run/launch anything that is not GDI bound. Alt-Tab still works, kind of. This particular pain is best experienced on a congested network with a large server load where a 10MB download can take 5 minutes or more.
And before you go calling BS again, not all GUI programs utilized the GDI layer. Besides Task Manager, PMMail was one, ported from OS/2, it followed OS/2 conventions on I/O, utilizing separate threads for input/output and network connectivity.
NT 4 was, arguably, the first version of Windows to really enjoy any sort of success in the server room.
Only for the first 42 days or 2^20 page outs....
The graphics drivers I mentioned had all their utilities directly interacting at Ring 0, instead of running in User Mode.
Actually, IIRC, it was Win NT 3.1 that had the initial full security model you ascribe to Win NT 3.5. Win NT 3.5 had already slid a good portion of the way down the slippery slope of Ring 0 code, including some of the graphics drivers. (Again, IIRC, it's been a while)
NT 4 moved a lot of user space Windows GDI functionality (as defined by Win 95/98/ME) into a kernel mode GDI API, which is single threaded btw, that persisted at least through all versions Windows XP, if not beyond. (This is one of the reasons why opening a 10MB networked file or attachment in Outlook causes your entire machine to lockup until it's done)
This was in contrast to OS/2, which continued to follow the original design criteria, and hence was perceived to be slower on the same hardware as NT 4 for single tasks, although multi-tasking was much faster on OS/2. I mention this because NT's original basis was the OS/2 criteria, which was then mutated to be able to support the Win 95/98/ME gaming solutions.
What you're describing applies to all human beings not just geeks. Geeks are by definition MORE rational than the average person because their interest is in trying to understand the equipment they're using. So that argument is weak, trumps nothing and does not change my mind.
Seriously? You really think so? I've met "geeks" who are nothing more than tinkerers. They don't strive to do much of anything other than put the pieces in configurations that please them.
As for the argument about 3rd party software don't be so disingenuous. You know perfectly well that to get full functionality from an Apple product you must run iTunes - all other options are partial solutions - and this is by Apple's own design. You're the sort of person that would tell a guy to RTFM if he was having trouble: Not at all helpful unless the docs are available and comprehensible at his level.
An ipod need never see iTunes to be fully usable and useful. Unless, of course, you wish to purchase items on iTunes and play them on the iPod?
I've rarely seen geeks be generally rational about anything they are passionate about. They just happen to sometimes have some actual arguments to support their view instead of being based on feelings alone.
So does my generic statement trump yours?
As for Apple products, some cost more, some less (I know - shocking!) And honestly, I cannot get over the large sentiment against iTunes on here. Don't like it, don't run it. There are other options, even if you own an iPod. iPad, or an iPhone. You just won't get updates, nor have backups. Choices.
I'd argue lossy barely makes sense for your portable device with the size of the things these days, it makes no sense anywhere other than PR or trial bits for people to see whether they might like a song. Unless, of course, you're listening to the strains of Britney Spears and company, in which case the lossier the better.
Not FUD about flac, half my music is flac and my NanoTouch (a present) won't play them - unless you know of a way of getting flac to work?
I do: Max It can losslessly convert your FLAC into ALAC. Granted, this program is old, but it will work and preserve your tag data. I converted a few hundred albums with it recently. I can also convert it back into FLAC if I like, with no losses. The beauty of lossless codecs. All it takes is a little time.
BTW, Max, being relatively old, does have a quirk every now and then. It's free, so I can't really complain since it does do the job.
For a funny mod point.... sigh.
What are you talking about? Windows 7 is by far the best OS Microsoft has put out ...
That still doesn't mean it's any good.
Yes, I have used it, and it's big brother 2008 R2 as well. Coded system software for it. Neither is secure in any form whatsoever no matter what MS's PR machine wishes you to think. The core problem is the fact that MS chose a run model with maximum privileges that can be throttled down, instead of minimum privileges that can be escalated via authentication, for example. Then, of course, there's the issue with dynamic code injection into system DLLs no matter what your process privs and then executing that code as "root" from the injecting process. That was actually the suggested work around for the extremely short sighted (IMNSHO) MS action in 2008 R2 of removing any meaningful security token manipulation.
No matter how pretty the outside (lipstick) the OS is still underwhelming underneath (pig).
And just a side note - OSX has stagnated? 7 major revisions in 11 years changing architecture and bringing new features while maintaining a reasonable memory and disk footprint vs 3 from MS, each requiring an order of magnitude increase in memory and disk space, one being a complete failure by any measure (Vista)? Ok, I'll admit there wasn't quite an order of magnitude increase from Vista to 7, but it did increase.
Interstates weren't built solely or even mainly for the residents local usage.
If states were left to do some of these things, which they were up until several of these agencies were founded, you'd still have mud paths in places like Louisiana or Montana instead of interstates.
Of course, he's a kitten, and he probably just sees the motions anyway, but did you really think I'd had a baby? Here, a slashdot poster?
Apparently not a very alert /.er. Only slow cats can see something resembling an image, the rest see a moving dot of light (for those with CRTs) or just a flickering stutter for slower LCDs.
There's other links out there for those that thought their "bright" animal watched TV. It certainly explained why the cat was never bothered by silent cats on TV, but had the shock of his life walking around the corner and seeing himself in the new full length mirror. Poof! Twice his normal size. :)
$1100 retail in the US.
The other advantage is that it will seamlessly work with your OS. I have a Dell U2411, and it's always brighter than I want it to be and does not have autodim capabilities, or at least none that I've discovered as of yet.
Oh, there most certainly should be a flamebait moderation. There are definite meritless posts with the sole intention of causing further flamewars or merely being inflamatory and nothing else. (Most of us don't respond to those)
However, those moderators who use flamebait on a post that most definitely should not be flamebait should be forever removed from the moderator pool (IOW, use that thing carefully, it can cut both ways.) Merely my opinion, it's as valid as any other.
Actually, I'd argue that JDK 8 isn't that important right now. People are still adjusting to JDK 7, and fixing the issues with JDK 7 should have a higher priority than doing all new stuff for JDK 8.
I'd say based on my experience that's there's still a sizable group out there that's having trouble adjusting to JDK 5 or 6. JDK 7 isn't even on the horizon yet.
JDK 7 has some serious goodness in it, for my needs anyways. And I'm fine if there's less bloat. You may see my improvements as bloat, and what I think of as bloat your core needs. Depends upon your viewpoint.
Microsoft gained a lead on Java? With what? I've done a year in .NET recently, and while it has some ok features, it's still got a long way to go to match the capabilities and dev environment provided by even Java 6 and Eclipse. And never mind the CF underneath the covers that MS calls security....
Compared to C++ or C, Java is slow. How slow depends on the workload. Anywhere from a few percent to a large multiple, the latter being fairly easy to trigger by hitting the JIT just the way it doesn't like.
If you care about performance or latency, Java is the wrong language. But it's faster than Python.
It depends upon how it's written, or are you unaware of the tests showing Java beating C++ by a large margin? Granted, they're probably all biased, intentionally or not, by the authors own capabilities, but even so, Java is not "slow", especially not in a server platform. It's also a whole lot easier to program robustly, especially by the majority of "CS Majors" currently being cranked out of schools. (That's another topic entirely) Note that there are seeming classes of problems that C++ would be faster in such as iterating across arrays of arrays of arrays, but a decent programmer can eliminate 95% of the performance hit with a proper iteration algorithm, and the remaining 5% can be removed by removing bounds checking in the JVM, bringing the operations in line with a C++ compiled runtime.
MicroBenchmark with Java being the best until custom memory handling was introduced into C++
Old Bencmark with some number crunching occurring
Just a couple of supporting links. If you want more, google is your friend.
I actually read the articles. Gasp! Turns out that new techniques allow for seeing the planets in the old data. ie, new data reduction/filtering methodologies improve the results coming out of the old pictures.
1) this allows for a second source for confirmation of any suspected planets
2) it allows for viewing orbits immediately across a multi-year span
3) the validation of the new techniques allows for all the Hubble data to be analyzed anew with potentially much different results than the first time around.
Think of this as looking at a drop of pond water with your eye, a magnifying glass, and then a series of pictures taken through the magnifying glass at different angles and putting together a 3-D representation of the drop in a computer.
Open drive, remove platters, place platter on an open vise, place chisel in middle, hit with hammer. Repeat.
At this point, only the most dedicated folks will even try to get data off of the platters, if you want to be sure, break them into multiple pieces and dispose of separately, or, get a large fresnel lens and create yourself a nice solar powered smelter. Molten platters won't tell anyone anything, other than you went out of your way to destroy something.
So you're really saying don't mix 2 types of load balancers in a cluster, and that type of thing? That would be insanity in production systems.
Honestly, while I have seen this in the real world, I often wonder if those responsible for it are qualified to even touch a keyboard or a network cable. For a production rollout, I spec homogenous layers, as it removes "configuration hell" from the equation. I only have to worry about single sets of configuration per equipment type in the layers, and removes an entire set of potential complicating factors from any problem resolution I need to perform. This is merely common sense, and the only ones who don't follow it are either entirely incompetent or are intending to milk the target for non-productive work for the foreseeable future (ie, "job-security").
I would agree with homogeneous equipment in specific layers. Homogenous throughout? Not so much, especially when going through firewalls.
[1] It's time to stop fooling ourselves -- the OS is not an application, and it's not a video game. The purpose of the OS is to load applications and manage resources, not to keep us entertained.
If I had mod points today, this would get you a bump all by itself. OSes should stick to that core purpose and do it well. The interface it displays to do so may change for various reasons, such as the single full screen presentation on a 4" phone for example, as that use case wouldn't be well served by having multiple windows floating around at least for the majority of the populace. But otherwise, it should be a lean mean set of APIs for accessing resources.