And why would you believe me? No idea. Why do you believe slashdot authors that don't even bother reading the ASN introductionary site? In fact, why believe anyone? Go read the ASN site, the standard, and make up your own mind.
Any piece of data with recurring sentences can be compressed. We already have excellent and fast compression algorithms that do that for us. LZW, Huffmann, etc. Why do we need ANOTHER one?
Besides - ANS.1 is definitely an encoding scheme. (Click the link to look at the website about ANS.1)
If you'd ask me, ANS.1 is not meant for everyday internet traffic. Why EEtimes or Slashdot seem to suggest it is, is beyond me.
With the current over-supply of domestic bandwidth and the move to database-driven, customised web sites, is it worth spending CPU cycles compressing small data files on-the-fly?
Absolutely. Positively. 100% YES!
Say you have a fairly large HTML page. (Take, em... slashdot.) You can compress it from, say, 100Kb to about 11Kb in a fraction (a 800Mhz P3 does it in 0.009 seconds) of a second. That saves an enormous amount of bandwidth and speeds up your browsing too. Definitely worth it.
That's why the protocol and compressing the payloud should be seperate. Like the different HTTP encoding schemes.
I'm all for compression especially since it saves bandwidth (money) AND is generally way faster then transferring the uncompressed payload. But that compression should not get in the way of maintanability and extensibility of a protocol. That has proven to be a bad idea(tm) in the past.
Compressed TCP/IP over IP wouldn't be such a bad idea though, except that compression usually belongs in the application layer. Stuff like HTML compresses like crazy. Good thing mod_gzip usage is on the rise.
Microsoft doesn't have to take this to court. They can just make a small "update" to windows, so that suddenly McAfee software becomes highly unstable. Then they can just sit back, and watch the stock drop...
Sueing a company that provides the platform your total income is based upon is a bad idea(tm) if I ever saw one. Probably took a sue-happy American to come up with that one too.
I don't know. Frankly, the editors that post the MS articles are making fools out of themselves. Slashdot these days seems more and more about posting negative MS articles, and less about "stuff that matters."
But 99 out of 100 MS-related articles here on/., (there's your punctuation) *ARE* indeed lame and mostly untrue anti-MS articles.
Slashdot does a great job spreading FUD about MS and MS products. Slashdot cannot be regarded as a credible newssource for MS related stories. In fact, the amount of anti-MS FUD coming from slashdot has outgrown the amount of anti-OSS FUD coming from MS.
Re:If this can't break Microsoft's back nothing wi
on
Code Red Back For More
·
· Score: 1
A patch was released months ago.
It's not Bills fault you live in a cave.
Replace "Windows" with "unpatched Bind on Unix", and you have an equally disturbing comment.
It breaks many of the code optimization tricks that us assembly language programmers have used for years. In turn this shows up as a decrease in performance in things like device drivers which tend to contain more assembly code than most applications, and it will shows up in slower execution speed of compiled applications, for example Windows and Linux applications written in C++.
That guy has some valid points, from his limited point of view that is. However, what he regards as a crime from intel is in fact intel's biggest and probably best step. Ditch the legacy crap.
Benchmarks show that programs compiled with intels compiler using P4 optimisations, beat the crap out of the competition - including T-birds.
Besides, the average codes does not participate in OSS projects at all.
The average coder has an appartment, or a house, a car, perhaps a wife or girl, and maybe even some children. He needs to get the cash rolling, and hence works for a healthy commercial IT business or a not-so-healthy.com startup. The code he creates will be the property of the company he works for, and although some companies do opensource their code, most don't.
Let's not kid ourselves in thinking that software engineer equals opensource advocate.
I would like to see this O'Reilly guy in a room filled with MS employees, participating in a panel with Mundie, Gates, and people from companies like IBM, Adobe, etc.
The "movement" actually consists of a very select few that actually have the time and energy to participate in opensource projects. The rest are just people that like to have free software.
They talk the talk, but don't walk the walk or something like that.
Then, ofcourse, ISPs 'd take the next logical step - Charge more for high priority data.
In other words - the rich elite gets the better connections.
QoS allocations on the net can NOT be trusted to mere users. It is something the providers will have to do. The only thing users will do with it, is abuse it. (always demand the best, fastest, most bandwidth they can get their grubby hands on.)
IP was never meant to be used on a scale like we use it today. The whole net structure reeks of filthy hacks. A total rewrite of low-level net protocols would, from a strictly theoretical point of view, be preferred.
Because anyone telling you that has been slashdot brainwashed.
XP, and about every new windows released on top of the NT codebase, *DOES* have new features.
Granted, W2K -> XP may not be as big a step as NT4 -> W2K, but that doesn't mean there aren't any new features. There are. Good ones like cleartype and the.NET programming API and bad ones like even more big brother stuff, more bloat, etc.
But the problem isn't really outlook. It's the combination Outlook + Ignorant user.
I've been running outlook (professional, not express) for years, without any problems or successful attacks from viri.
And anyone that suggest pine as a real alternative to Outlook XP as a business mail suite needs to wake up, check his calender, and realize this is the 21st century, not the 70's.
Slashdot Mirror
Three words: Dual Monitor Output
I've worked with MIB. It's a f****ng disaster. The 'S' in SNMP is out of place. Whatever beast SNMP is, it's not Simple.
I meant ASN indeed.
:)
It's getting late here
And why would you believe me? No idea. Why do you believe slashdot authors that don't even bother reading the ASN introductionary site? In fact, why believe anyone? Go read the ASN site, the standard, and make up your own mind.
Besides - ANS.1 is definitely an encoding scheme. (Click the link to look at the website about ANS.1)
If you'd ask me, ANS.1 is not meant for everyday internet traffic. Why EEtimes or Slashdot seem to suggest it is, is beyond me.
Absolutely. Positively. 100% YES!
Say you have a fairly large HTML page. (Take, em... slashdot.) You can compress it from, say, 100Kb to about 11Kb in a fraction (a 800Mhz P3 does it in 0.009 seconds) of a second. That saves an enormous amount of bandwidth and speeds up your browsing too. Definitely worth it.
That's why the protocol and compressing the payloud should be seperate. Like the different HTTP encoding schemes.
I'm all for compression especially since it saves bandwidth (money) AND is generally way faster then transferring the uncompressed payload. But that compression should not get in the way of maintanability and extensibility of a protocol. That has proven to be a bad idea(tm) in the past.
Not.
Compressed TCP/IP over IP wouldn't be such a bad idea though, except that compression usually belongs in the application layer. Stuff like HTML compresses like crazy. Good thing mod_gzip usage is on the rise.
First....
200 BYTE (!) XML documents are pretty rare. They probably standarized a few headers and instead of sending they just send some code.
Don't believe for a second we're talking about a compression scheme here. The usual slashdot lack of information applies.
Microsoft doesn't have to take this to court. They can just make a small "update" to windows, so that suddenly McAfee software becomes highly unstable. Then they can just sit back, and watch the stock drop...
Sueing a company that provides the platform your total income is based upon is a bad idea(tm) if I ever saw one. Probably took a sue-happy American to come up with that one too.
What is a 'browser' anyway?
Does a program that connects to a socket and gets data using HTTP qualify? (If that is true then telnet is a violating application.)
If so, then does a component that does the same qualify? How about that component extended with some back/forward buttons?
American software patents like this one totally fail to grasp the technical implications of the total bullshit uttered in them.
America. The land of freedom. Freedom for corporations to do whatever they want.
Amen... B5 was, and still is, the best story told on television - ever.
I don't know. Frankly, the editors that post the MS articles are making fools out of themselves. Slashdot these days seems more and more about posting negative MS articles, and less about "stuff that matters."
No. But the dude who wrote the article I replied to seems to think it is.
But 99 out of 100 MS-related articles here on /., (there's your punctuation) *ARE* indeed lame and mostly untrue anti-MS articles.
Slashdot does a great job spreading FUD about MS and MS products. Slashdot cannot be regarded as a credible newssource for MS related stories. In fact, the amount of anti-MS FUD coming from slashdot has outgrown the amount of anti-OSS FUD coming from MS.
A patch was released months ago.
It's not Bills fault you live in a cave.
Replace "Windows" with "unpatched Bind on Unix", and you have an equally disturbing comment.
It breaks many of the code optimization tricks that us assembly language programmers have used for years. In turn this shows up as a decrease in performance in things like device drivers which tend to contain more assembly code than most applications, and it will shows up in slower execution speed of compiled applications, for example Windows and Linux applications written in C++.
That guy has some valid points, from his limited point of view that is. However, what he regards as a crime from intel is in fact intel's biggest and probably best step. Ditch the legacy crap.
Benchmarks show that programs compiled with intels compiler using P4 optimisations, beat the crap out of the competition - including T-birds.
+++ATH0
The average coder has an appartment, or a house, a car, perhaps a wife or girl, and maybe even some children. He needs to get the cash rolling, and hence works for a healthy commercial IT business or a not-so-healthy
Let's not kid ourselves in thinking that software engineer equals opensource advocate.
+++ATH0
Now THAT would be fun.
+++ATH0
They talk the talk, but don't walk the walk or something like that.
+++ATH0
In other words - the rich elite gets the better connections.
QoS allocations on the net can NOT be trusted to mere users. It is something the providers will have to do. The only thing users will do with it, is abuse it. (always demand the best, fastest, most bandwidth they can get their grubby hands on.)
IP was never meant to be used on a scale like we use it today. The whole net structure reeks of filthy hacks. A total rewrite of low-level net protocols would, from a strictly theoretical point of view, be preferred.
+++ATH0
--
XP, and about every new windows released on top of the NT codebase, *DOES* have new features.
Granted, W2K -> XP may not be as big a step as NT4 -> W2K, but that doesn't mean there aren't any new features. There are. Good ones like cleartype and the
--
The first one that pops into mind is cleartype technology. I actually like that.
The second one is the
Admitted, I don't see why they need 256Mb extra RAM and a few Gb of HD space for that... but okay.
--
But the problem isn't really outlook. It's the combination Outlook + Ignorant user.
I've been running outlook (professional, not express) for years, without any problems or successful attacks from viri.
And anyone that suggest pine as a real alternative to Outlook XP as a business mail suite needs to wake up, check his calender, and realize this is the 21st century, not the 70's.
--