Maybe they should've said that in the original news post?
They should have said that. I think all those renaming of projects are stupid. And I still wonder why they had to rename the projects. It is not like that BIOS vendor invented the word.
However, nothing should force me to receive SMTP connections from you, if I don't want them.
Of course not. But if you want to recieve your email, you have to accept SMTP connections from anywhere. If you block SMTP connections you'll end up loosing email. As long as it is only your own incomming mail you block it is acceptable. But an ISP deciding what email their customers are allowed to receive is not acceptable.
If you are dial up or home dsl you should not be talking diectly to smtp servers anyway you should be sending mail through your provider.
Bullshit. Nothing should prevent you from running an SMTP server on your own computer if that is the way you want to do it. I can accept that an ISP watches out for open relays on their customers computers and block SMTP in that case. But that is also about the only blocking I can accept. If you pay for an internet connection you should expect to get an internet connection, not a connection limited to 90% (or less) of the functionality.
It all started with DSL, but the telcom companies squeezed things till
the development on Cable started.
Really? In the city where I live internet over cable was available years
before DSL. But only one cable provider implemented it, they did however
offer amazing speeds of up to 4mbit/sec the first few years. Then they
started a commercial campaign stating that they were simply the best provider,
which was true at that time. They no longer are the best. Their prices have
increased while their speed have dropped. They still claim they are the best
however.
I don't know about you, but the less government controlled crap that comes into my house the better!
I'd rather have something government controlled than something controlled by a company with monopoly on communication infrastructure. And keep in mind that the control should only be over the lowest possible levels below IP. The control should only ensure that line and bandwidth is available. On top of the physical line any ISP can implement the necesarry range of protocols including: ARP, IP, DHCP, DNS. And you will get competition like never before, because sudenly anybody with a 100-1000mbit/sec backbone connection and three highend PCs can become an ISP. The backbone itself can to large extents be implemented using the same network. Of course those who worry about security is going to encrypt their communication, that can happen below the IP layer or above (possibly both).
That'll give them SPARC Solaris, x86 Solaris and x86 Linux available to mix and match in the B1600 shelf.
Unless x86 Solaris is modified to include x86-64 support it is not going to fully utilize the CPU. But I guess that is an amount of changes Sun is willing to make. What I find more interesting is that this new CPU will make it easier for Microsoft to get Windows running on the box. We all know that Sun and Microsoft are not the best of friends, and Sun probably don't like to see their boxes running Windows. But how about Microsoft, do they want Windows to run on Sun hardware?
Turing-completeness. This guarantees you bigO efficiency
I agree that bigO efficiency do matter, and you can do that in any half-decent language. But you cannot do efficient programs on a Turing-machine, you will often add a linear factor to the time usage by implementing it on a Turing-machine.
Implementing bubblesort is more complex and expensive than calling Arrays.sort().
Reminds me about the first book I read about assembly code. It started explaining that performance was one of the reasons to write assembly. The first code example was an implementation of bubblesort. (Duh!)
Telling a business "Can't you just switch" or "If you don't like it, do use it" is completely naive.
Nobody said you can switch for free. But neither can you stay for free. And notice that Microsoft has done whatever they can to make the cost of switching as high as possible. The price is now so high that they can also start increasing the price for anybody to stay with their software. What I think is that the sooner you leave microsoft the cheaper it will be.
Other systems don't try to make it difficult to switch. I believe it is much simpler to switch between Linux, BSD, AIX, IRIX, Solaris, HPUX, and probably other systems.
The only reason I switched is because less and less new software is being done on Gnome 1.4 and more and more bug fixes and updated show up as ports to Gnome 2.
I'm considering to upgrade because I'm pretty tired of Gnome crashing all the time. I'm running RH7.3 with all upgrades, and the crash is 100% reproducible. RedHat knows about the bug but doesn't want to fix it because they think the bug is no in RH8. I have also had problems with the i810 drm driver causing kernel panics. With a minor modification of my kernel I have reduced the amount of crashes, and now I have been running for 11 days without kernel crashes. I don't want to spend a lot of time tracing this bug since I know a new drm system is available, yet another reason to upgrade.
perhaps you could "cross the streams" and safely mix them?
In fact I have recently been working on some
code doing something similar to that. Unfortunately the time complexity is quadratic, and I'm afraid it cannot be done any faster. If you want to work with megabytes of random bytes, quadratic time complexity is going to be a pain.
I think the RFC is flawed. Evil content come in so many flavours that a subtype is simply not enough to identify it. Instead a new top-level media type should have been defined so we could have evil/virus, evil/DoS, and evil/spam content-types.
Most desktop apps at least support going through arts, esd or some other software mixer so while its kindof a crappy solution its not that much of an issue.
I tried playing TuxRacer with sound through one of those daemons. But I found that the sound would be delayed. It might have been less than one second of delay, but it was obviously far too much and was actually very confusing.
800x600 instead of native 1024x768 and it looks about as fuzzy. I think they do it for accessability reasons
I'd expect a better result from using 512x384. But it is not easy getting a computer produce that resolution. I tried on a laptop running RH7.3 without any luck.
I'd much rather have a *real* 1024x768 console.
So would I. And then a nice large 12x24 pixels font so we get 85*32 chars on the screen. Just a few more chars but at a 50% better resolution than we are used to.
"weak" prime - composite numbers which falsely pass those tests.
If you use Miller-Rabin it can be proven that for any composite number at least 75% of the possible values for the random choice will reveal that it is a composite number. The actual probability can depend on the composite number, but no composite number can give an error probability above one fourth.
In other words if your n-bit prime candidate can pass the Miller-Rabin test n times, the error probability is negligible. Of course this means you need a large number of random bits. I wonder if there are ways you can safely do the test with a smaller amount of random bits.
Doesn't this make it illegal to sell "007: Agent Under Fire" since that is effectively the circumvention device?
Good point. I think you might be right about that. It is after all that game which contains the signed code opening the possibility to execute arbitrary code on the box. This case actually demonstrates a major weaknes in the security model. Basing your entire security on signing code is unsafe, for any large amount of code you will never be able to review it enough. This leads to the risk of signing code with flaws.
What we have seen is a flawed design which was accidentially broken by an insider. But it might not be a DMCA violation, because it is not a security model introduced to prevent unauthorized copying. (Even if microsoft claims so, it is not.) It is a security model designed to control who is allowed to develop code for the architecture.
Slashdot Mirror
Maybe they should've said that in the original news post?
They should have said that. I think all those renaming of projects are stupid. And I still wonder why they had to rename the projects. It is not like that BIOS vendor invented the word.
Just gotta whip out one of your spare accounts to post with after you moderate
Multiple accounts is only for the schizophrenic, good if you want to argue with yourself. I can imagine that leading to some interesting flames.
Oh all those X-words. And none of them really has anything to do with X.
I could type ~70 WPM.
70 words per minute? That is fast! Are you sure you didn't mean 70 characters per minute?
However, nothing should force me to receive SMTP connections from you, if I don't want them.
Of course not. But if you want to recieve your email, you have to accept SMTP connections from anywhere. If you block SMTP connections you'll end up loosing email. As long as it is only your own incomming mail you block it is acceptable. But an ISP deciding what email their customers are allowed to receive is not acceptable.
If you are dial up or home dsl you should not be talking diectly to smtp servers anyway you should be sending mail through your provider.
Bullshit. Nothing should prevent you from running an SMTP server on your own computer if that is the way you want to do it. I can accept that an ISP watches out for open relays on their customers computers and block SMTP in that case. But that is also about the only blocking I can accept. If you pay for an internet connection you should expect to get an internet connection, not a connection limited to 90% (or less) of the functionality.
As an antispam measure I've blocked AOL, hotmail and Yahoo for a while
.com to prevent spam. Unfortunately that doesn't catch all spam anymore.
Some years ago I just blocked
Someone get Cogentco to come to Utah. Now *that's* what I consider "ultra high-speed internet!"
That is what I consider "ultra high-price internet"!
What they advertise as "ultra high speed internet access"
You know you should be skeptical when they advertise with such words instead of an actual speed.
It all started with DSL, but the telcom companies squeezed things till the development on Cable started.
Really? In the city where I live internet over cable was available years before DSL. But only one cable provider implemented it, they did however offer amazing speeds of up to 4mbit/sec the first few years. Then they started a commercial campaign stating that they were simply the best provider, which was true at that time. They no longer are the best. Their prices have increased while their speed have dropped. They still claim they are the best however.
I don't know about you, but the less government controlled crap that comes into my house the better!
I'd rather have something government controlled than something controlled by a company with monopoly on communication infrastructure. And keep in mind that the control should only be over the lowest possible levels below IP. The control should only ensure that line and bandwidth is available. On top of the physical line any ISP can implement the necesarry range of protocols including: ARP, IP, DHCP, DNS. And you will get competition like never before, because sudenly anybody with a 100-1000mbit/sec backbone connection and three highend PCs can become an ISP. The backbone itself can to large extents be implemented using the same network. Of course those who worry about security is going to encrypt their communication, that can happen below the IP layer or above (possibly both).
That'll give them SPARC Solaris, x86 Solaris and x86 Linux available to mix and match in the B1600 shelf.
Unless x86 Solaris is modified to include x86-64 support it is not going to fully utilize the CPU. But I guess that is an amount of changes Sun is willing to make. What I find more interesting is that this new CPU will make it easier for Microsoft to get Windows running on the box. We all know that Sun and Microsoft are not the best of friends, and Sun probably don't like to see their boxes running Windows. But how about Microsoft, do they want Windows to run on Sun hardware?
Turing-completeness. This guarantees you bigO efficiency
I agree that bigO efficiency do matter, and you can do that in any half-decent language. But you cannot do efficient programs on a Turing-machine, you will often add a linear factor to the time usage by implementing it on a Turing-machine.
Implementing bubblesort is more complex and expensive than calling Arrays.sort().
Reminds me about the first book I read about assembly code. It started explaining that performance was one of the reasons to write assembly. The first code example was an implementation of bubblesort. (Duh!)
Telling a business "Can't you just switch" or "If you don't like it, do use it" is completely naive.
Nobody said you can switch for free. But neither can you stay for free. And notice that Microsoft has done whatever they can to make the cost of switching as high as possible. The price is now so high that they can also start increasing the price for anybody to stay with their software. What I think is that the sooner you leave microsoft the cheaper it will be.
Other systems don't try to make it difficult to switch. I believe it is much simpler to switch between Linux, BSD, AIX, IRIX, Solaris, HPUX, and probably other systems.
they can unveil RedHat 10 on Oct 31
But will 2.6 be ready for production by that date?
The only reason I switched is because less and less new software is being done on Gnome 1.4 and more and more bug fixes and updated show up as ports to Gnome 2.
I'm considering to upgrade because I'm pretty tired of Gnome crashing all the time. I'm running RH7.3 with all upgrades, and the crash is 100% reproducible. RedHat knows about the bug but doesn't want to fix it because they think the bug is no in RH8. I have also had problems with the i810 drm driver causing kernel panics. With a minor modification of my kernel I have reduced the amount of crashes, and now I have been running for 11 days without kernel crashes. I don't want to spend a lot of time tracing this bug since I know a new drm system is available, yet another reason to upgrade.
perhaps you could "cross the streams" and safely mix them?
In fact I have recently been working on some code doing something similar to that. Unfortunately the time complexity is quadratic, and I'm afraid it cannot be done any faster. If you want to work with megabytes of random bytes, quadratic time complexity is going to be a pain.
Microsoft's got the Office .dlls already loaded on startup
Makes me want to say that I bet on Linux StarOffice starts faster than MicrosoftOffice.
if MS was running a search engine
They are.
Maybe you should try reading Slashdot instead of bitching ;-)
You could read Slashdot.org and start bitching about dupes. But that wouldn't be fair, since that one is mentioned four times: 1 2 3 4.
Content-Type: application/evil
I think the RFC is flawed. Evil content come in so many flavours that a subtype is simply not enough to identify it. Instead a new top-level media type should have been defined so we could have evil/virus, evil/DoS, and evil/spam content-types.
Most desktop apps at least support going through arts, esd or some other software mixer so while its kindof a crappy solution its not that much of an issue.
I tried playing TuxRacer with sound through one of those daemons. But I found that the sound would be delayed. It might have been less than one second of delay, but it was obviously far too much and was actually very confusing.
800x600 instead of native 1024x768 and it looks about as fuzzy. I think they do it for accessability reasons
I'd expect a better result from using 512x384. But it is not easy getting a computer produce that resolution. I tried on a laptop running RH7.3 without any luck.
I'd much rather have a *real* 1024x768 console.
So would I. And then a nice large 12x24 pixels font so we get 85*32 chars on the screen. Just a few more chars but at a 50% better resolution than we are used to.
"weak" prime - composite numbers which falsely pass those tests.
If you use Miller-Rabin it can be proven that for any composite number at least 75% of the possible values for the random choice will reveal that it is a composite number. The actual probability can depend on the composite number, but no composite number can give an error probability above one fourth. In other words if your n-bit prime candidate can pass the Miller-Rabin test n times, the error probability is negligible. Of course this means you need a large number of random bits. I wonder if there are ways you can safely do the test with a smaller amount of random bits.
Doesn't this make it illegal to sell "007: Agent Under Fire" since that is effectively the circumvention device?
Good point. I think you might be right about that. It is after all that game which contains the signed code opening the possibility to execute arbitrary code on the box. This case actually demonstrates a major weaknes in the security model. Basing your entire security on signing code is unsafe, for any large amount of code you will never be able to review it enough. This leads to the risk of signing code with flaws.
What we have seen is a flawed design which was accidentially broken by an insider. But it might not be a DMCA violation, because it is not a security model introduced to prevent unauthorized copying. (Even if microsoft claims so, it is not.) It is a security model designed to control who is allowed to develop code for the architecture.