So I am not the only one who thinks this way? The moment I saw it in the headline, I cringed, but resolved not to comment about it because it was offtopic.
Now that the topic has been broached, I can say: Amen brother.
That's true, but I wasn't addressing that fact in particular, just that.NET is not completely free of DCOM necessities. For some of us, not having a simple pipe protocol like JDWP is a big deal, especially if you take into account that older microsoft development platforms used to have similar venues.
I thought I would bring it up. I don't understand your ire. I thought it was an interesting regression.
Just wanted to remind that it is not completely free of DCOM tie-ins (eg, you can get 100% functionality without DCOM). The above restriction can be moderately annoying problem. I don't think density has anything to do with it.
"That's unfortunate, because.NET does not require DCOM at all."
That is something of a lie. Have you ever tried, say, remote debugging? Or, perhaps more to the point, did you ever wonder WHY microsoft doesn't support a TCP/IP pipe-type remote debugging protcol (as they did with their native code) for.NET?
You have to install a service on the other machine to remote debug! Isn't that a hoot?
For a time I was going to write a JDWP implementation for.NET, but upon further research I came upon a blog written by a.NET developer who says, in paraphrase: stay away from the debugging classes. They're huge, buggy, and so undocumented that they don't even appear on MSDN!...tune in for version 2.0
Seriously; You can't be blamed for being a monopoly when you are a barely emergent and first supplier to a given market...at least, not reasonably so, methinks.
Although I am also somewhat baffled, I am inclined to believe that they mean that they are putting everyone in the "wheel" group, that is, people who can obtain root and change settings. Generally speaking, without being in wheel you cannot do these things.
PAM for linux has a setting that mimics this behavior.
Games. At least, that's why my windows partition still exists. I also have an NTFS formatted removable drive that I need to write to now and then, and to do that I need to use Windows. The reason that drive is NTFS is for the benefit of carrying it around to most of the other computers in the world that grok NTFS.
I'm not so sure about that. It seems with the amount of money Microsoft is willing to chew through (and still can't wrap its mouth around, as seen in the extremely unusual huge dividend a few quarters ago) they probably just look at market penetration numbers, and if they go up, celebrate the impending ownership of another market.
Of course it is. When you buy a different version of windows, you are buying compile flags--and ONLY compile flags when it comes to kernel options like how many CPUs are supported. It's an excellent sham.
It has been in my experience that I generally have a tougher time with hardware under FreeBSD. Actually, that is the reason why I swithced over the other way.
If I were to run a server though, I'd still probably run FreeBSD.
...Like Hoary Hedgehog and Warty before it, it is simply the name of the branch. Don't get too tangled up--it all turns into Ubuntu when ready for public abuse.
(Although the repositories still contain their namesakes, eg warty or hoary or...grumpy, I guess)
...if you want to waste unnecessary amounts of power. My machine "to use" isn't quite that busy most nights, thank you.
(Although I will leave it on when I go elsewhere so that I can shell in)
I may be talking out of my behind depending on how Linux does disk scheduling, but when it comes to multiplexing I/O from magnetic disk, multiplexing (mostly) sequentual reads can slow down the overall process, as 'context swaps' (to use a metaphor) cost you a few ms, which can be really damaging if multiple processes want to talk to the disk.
Of course, there are so many parameters that go into this dependent on what daemon is being started, so it may be a good idea for the boot process to attempt to be a little bit smarter and use more information about what to parallelize as to keep the CPU busy and not waiting for I/O.
Not that I care much; how often do I boot in a day?
This is, by Microsoft's standards, a relatively cheap way to keep application writers from targeting *NIX platforms. No move is too excessive, because an operating system without applications will never become a serious competitor. If a country's population moves into the computer age accustomed to and expecting applications for the *NIX platform, then Windows will lose the big card of application availability and have to compete on technical merits instead.
It's much easier to simply prevent those applications from becoming major in the first place.
Slashdot Mirror
Until a software developer sees it fit to mmap that entire DVD 8GB DVD on their hard drive. Or anything like that.
Since the advent of Overnet and Kad-DHT overlay networks-this is not quite fact anymore.
Wait, I'm not entirely clear on who does what...but the FCC rates video games? I thought that the game industry was self-policed by the ESRB
Because the best thing we could do to bolster adoption of superior operating systems is intentionally cripple hardware support.
Cut off my other leg already.
So I am not the only one who thinks this way? The moment I saw it in the headline, I cringed, but resolved not to comment about it because it was offtopic.
Now that the topic has been broached, I can say:
Amen brother.
I should mention that is precisely the blog that mentioned that the .NET debugging interface classes were large, undocumented, and buggy.
That's true, but I wasn't addressing that fact in particular, just that .NET is not completely free of DCOM necessities. For some of us, not having a simple pipe protocol like JDWP is a big deal, especially if you take into account that older microsoft development platforms used to have similar venues.
I thought I would bring it up. I don't understand your ire. I thought it was an interesting regression.
Just wanted to remind that it is not completely free of DCOM tie-ins (eg, you can get 100% functionality without DCOM). The above restriction can be moderately annoying problem. I don't think density has anything to do with it.
"That's unfortunate, because .NET does not require DCOM at all."
.NET?
.NET, but upon further research I came upon a blog written by a .NET developer who says, in paraphrase: stay away from the debugging classes. They're huge, buggy, and so undocumented that they don't even appear on MSDN!...tune in for version 2.0
That is something of a lie. Have you ever tried, say, remote debugging? Or, perhaps more to the point, did you ever wonder WHY microsoft doesn't support a TCP/IP pipe-type remote debugging protcol (as they did with their native code) for
You have to install a service on the other machine to remote debug! Isn't that a hoot?
For a time I was going to write a JDWP implementation for
Seriously; You can't be blamed for being a monopoly when you are a barely emergent and first supplier to a given market...at least, not reasonably so, methinks.
man setfacl
man getfacl
there if you need them, but I find the POSIX permissions good enough for my purposes.
Although I am also somewhat baffled, I am inclined to believe that they mean that they are putting everyone in the "wheel" group, that is, people who can obtain root and change settings. Generally speaking, without being in wheel you cannot do these things.
PAM for linux has a setting that mimics this behavior.
Games. At least, that's why my windows partition still exists. I also have an NTFS formatted removable drive that I need to write to now and then, and to do that I need to use Windows. The reason that drive is NTFS is for the benefit of carrying it around to most of the other computers in the world that grok NTFS.
I'm not so sure about that. It seems with the amount of money Microsoft is willing to chew through (and still can't wrap its mouth around, as seen in the extremely unusual huge dividend a few quarters ago) they probably just look at market penetration numbers, and if they go up, celebrate the impending ownership of another market.
Also some people mod these ways to give out karma points as opposed to the funny modifier, which does not.
Of course it is. When you buy a different version of windows, you are buying compile flags--and ONLY compile flags when it comes to kernel options like how many CPUs are supported. It's an excellent sham.
It has been in my experience that I generally have a tougher time with hardware under FreeBSD. Actually, that is the reason why I swithced over the other way.
If I were to run a server though, I'd still probably run FreeBSD.
...Like Hoary Hedgehog and Warty before it, it is simply the name of the branch. Don't get too tangled up--it all turns into Ubuntu when ready for public abuse.
(Although the repositories still contain their namesakes, eg warty or hoary or...grumpy, I guess)
Opera costs money.
Opera is lagging behind yet again. It all depends on one's metrics, no?
Your daughter must like pain. Or have nothing better to do.
...if you want to waste unnecessary amounts of power. My machine "to use" isn't quite that busy most nights, thank you.
(Although I will leave it on when I go elsewhere so that I can shell in)
I may be talking out of my behind depending on how Linux does disk scheduling, but when it comes to multiplexing I/O from magnetic disk, multiplexing (mostly) sequentual reads can slow down the overall process, as 'context swaps' (to use a metaphor) cost you a few ms, which can be really damaging if multiple processes want to talk to the disk.
Of course, there are so many parameters that go into this dependent on what daemon is being started, so it may be a good idea for the boot process to attempt to be a little bit smarter and use more information about what to parallelize as to keep the CPU busy and not waiting for I/O.
Not that I care much; how often do I boot in a day?
I have found the implementation at http://www.audioscrobbler.com great for finding new music, no P2P attached.
This is, by Microsoft's standards, a relatively cheap way to keep application writers from targeting *NIX platforms. No move is too excessive, because an operating system without applications will never become a serious competitor. If a country's population moves into the computer age accustomed to and expecting applications for the *NIX platform, then Windows will lose the big card of application availability and have to compete on technical merits instead.
It's much easier to simply prevent those applications from becoming major in the first place.
I have something square, but it's on its side...a diamond, I think. And not in the middle, but to the right....
Silly kids.