tomshardware.com has one online... and obviously Athlon systems SMOKE P4-systems at this benchmark. Even 1.7GHz overclocked P4 is unable to outperform 1Ghz Athlon with 100Mhz FSB + SDRAM. P4 looks really bad also in FPU-benchmarks and direct CPU-benchmark. IMO - Save your dollars.
This is absolutely a good thing for Linux, but I hope they can resist the temptation of only supporting their own distribution....And it can be that they turn out as vaporware aswell.
Atleast to my understanding, the Microsoft's implementation of Kerberos is uncompatible in a such way that the Directory service is only available while running W2K Kerberos server. W2K is able to authenticate from UNIX Kerberos server, but I've heard a claim that UNIX clients will be unable to authenticate from W2K Kerberos. None of this I have tried out myself, not willing to touch W2K with even a long, very, very long stick.
Considering that the whole thing is 100% Java, it is more surprising that they actually have to port the thing than that they ported it.
Did you think about that it probably produces native code? The compiler will need some writing, atleast. But what I cannot understand is that why people code Java because they seem to only make Java-programs that run under Windows only. Native win32-code would give a lot better results.
But when will they design a good architechture around the CPU? Practically speaking, everything ooutside the cpu is CRAP on x86-platforms. Yes, we're talking about 1GB/s burst-rate with AGP-bus, but that's only for gfx-board. Back in...was it 1992 SGI had 1.6GB/s continous transfer rate on Indy. And that was for all components. Just compare x86 to any Sun or SGI workstation. I'd choose old Sparc anytime over new PC.
FUD? One is able to for example to write web page that can send every private key in plain text in few seconds, I think this would be possible via Outlook also. I'm uncertain but I think that this has been fixed since NT4sp6, but every older version is vulnerable. M$ CryptoAPI is fundamentally insecure.
Of course, if that's abused, I'm sure they'll turn off some of that
They are already offering free accounts to anyone, so why bother about such abuse possibility. No sane person would ever use that kind of account for anything important. If those cgi-bins are run with user's own account it should not lead into root-compromise either.
I've been using MS ergo, Logitech Ergo and Cirque WaveKeyboard. I didn't like the MS keyboard at all, it was better than normal Keytronic but missed 'the touch'. Then I bought Cirque for home use, which I found excellent from the very beginning. At work I use Logitech keyboard, which looks like the MS model but has better touch and layout.
A piece of addition, many new gpl-softwares use horrible coding practices, void pointer arithmetics allowed by gcc, assumptations on element sizes (like int == long) etc. Which means you cannot compile the source using decent compilers like cc on Solaris and good portability becomes compromised, most likely the software will only work on x86-platform.
I've also experienced controlling mouse with right foot. Mouse sensitivy needs some adjusting but after that atleast focus changing goes easily. For typing I use Cirque Wavekeyboard, which is the best ergo-keyboard I've found for the x86-computers.
If you take closer glance at the list of cards they support so far, you'll notice that only a few of current chipsets are supported. I'm sure the list will increase rapidly, and eagerly await support f.ex. the GeForce 256 chipsets....So I could buy one.:)
LZW and LZ77/78 are only common in that they are both called dictionary algorithms. While LZW calculates a dictionary of 4096 entries on the fly while decoding or encoding, LZ77/78 commonly use a sliding dictionary of 4096 bytes of previous data. Look for some implementations of LZW with adaptive code length and/or huffman encoding combined and LZ77/78 with huffman or mathematical encoding to find the signifigance of those sizes by yourself.
Data compression FAQ will also be a good source of information and several good books have been written on the subject.
Quite an interesting point on the patent quarrel is to note that v42bis-compression used on every modem operating on phone lines uses LZW compression, from the days of the good old v32 modems. (9600bps) Unisys could try to sue the big companies like Rockwell, unless they have paid the fees.
If I recall right the group played around 40 farewell gigs, now they must have changed their minds since they call it a reunion. Nice to know that they'll continue playing since I like their music a lot.
noticed the author mentioned that the Hardware Abstraction Layer (HAL) was used for security purposes.
Yes, HAL was used on NT3, but after the GUI evolved, it turned out that using HAL was painfully slow, so now on recent NT versions, too many things just bypass HAL and all of them give their little contribution into making NT unstable.
Originally HAL was a good idea, but the implementation just evolved into a horrible kludge, like the whole NT itself. Please correct if I'm wrong on this.
I like how I go to one website, and it automatically tells me what I do or do not have installed.
Have your considered what that webpage is allowed to do on your system (without even you knowing) because it can tell you what do you have patched and what you haven't?
And why it doesn't work correctly with Netscape browsers is that they won't let a web page to run programs that search your system. (Well Netscapes have holes also but..) Anyway, my point is that I would never grant such rights for a remote automated robot, and not even trained human other than mysel.
Anyone could write a program that completely screws your hard drive if you IE and surf to that page. Only barrier between disaster being that notorius window asking user does he trust the program, asking for permission to execute it. After that...
Interesting indeed, why do they feel they are held back. I think they have still as good possibility as all others for new innovations and the even have the resources for huge research.
Still lately M$-attitude has improved a lot. Specs for office formats etc. have been released to public, I just hope it also means that new versions would be compatible. Nowadays they also admit security flaws and are keen to provide users with patches.
Win2k includes many huge improvements ddns, kerberos, mount points, etc. I just don't like the way they are implemented because you can't take the advance of them if you are using w2k workstations and unix servers. Could be they are really pushing w2k to replace unix on server side. Let's hope they will provide us with specs.
I wouldn't trust IE running any java-applications. Personally I use appletviewer etc. every time for my java developement. Just take some time investigating what a java-application can do under IE without the user even knowing. And yes, IE3 crashed my whole system many times during test phases and that was a java-applet.
By trademarking this in country X, he is probably trying to sue companies using the trademark in the country X and get rich... Just some greedy guy behind that. What he hasn't realised though is that it would never hold in court unless he has some product with a name that can be trademarked.
Slashdot Mirror
tomshardware.com has one online ... and obviously Athlon systems SMOKE P4-systems at this benchmark. Even 1.7GHz overclocked P4 is unable to outperform 1Ghz Athlon with 100Mhz FSB + SDRAM. P4 looks really bad also in FPU-benchmarks and direct CPU-benchmark. IMO - Save your dollars.
This is absolutely a good thing for Linux, but I hope they can resist the temptation of only supporting their own distribution. ...And it can be that they turn out as vaporware aswell.
Atleast to my understanding, the Microsoft's implementation of Kerberos is uncompatible in a such way that the Directory service is only available while running W2K Kerberos server. W2K is able to authenticate from UNIX Kerberos server, but I've heard a claim that UNIX clients will be unable to authenticate from W2K Kerberos. None of this I have tried out myself, not willing to touch W2K with even a long, very, very long stick.
Considering that the whole thing is 100% Java, it is more surprising that they actually have to port the thing than that they ported it.
Did you think about that it probably produces native code? The compiler will need some writing, atleast. But what I cannot understand is that why people code Java because they seem to only make Java-programs that run under Windows only. Native win32-code would give a lot better results.
But when will they design a good architechture around the CPU? Practically speaking, everything ooutside the cpu is CRAP on x86-platforms. Yes, we're talking about 1GB/s burst-rate with AGP-bus, but that's only for gfx-board. Back in ...was it 1992 SGI had 1.6GB/s continous transfer rate on Indy. And that was for all components. Just compare x86 to any Sun or SGI workstation. I'd choose old Sparc anytime over new PC.
...it comes bundled with programs.
FUD? One is able to for example to write web page that can send every private key in plain text in few seconds, I think this would be possible via Outlook also. I'm uncertain but I think that this has been fixed since NT4sp6, but every older version is vulnerable. M$ CryptoAPI is fundamentally insecure.
Of course, if that's abused, I'm sure they'll turn off some of that
They are already offering free accounts to anyone, so why bother about such abuse possibility. No sane person would ever use that kind of account for anything important. If those cgi-bins are run with user's own account it should not lead into root-compromise either.
I've been using MS ergo, Logitech Ergo and Cirque WaveKeyboard. I didn't like the MS keyboard at all, it was better than normal Keytronic but missed 'the touch'. Then I bought Cirque for home use, which I found excellent from the very beginning. At work I use Logitech keyboard, which looks like the MS model but has better touch and layout.
www.cirque.com
A piece of addition, many new gpl-softwares use horrible coding practices, void pointer arithmetics allowed by gcc, assumptations on element sizes (like int == long) etc. Which means you cannot compile the source using decent compilers like cc on Solaris and good portability becomes compromised, most likely the software will only work on x86-platform.
I've also experienced controlling mouse with right foot. Mouse sensitivy needs some adjusting but after that atleast focus changing goes easily. For typing I use Cirque Wavekeyboard, which is the best ergo-keyboard I've found for the x86-computers.
If you take closer glance at the list of cards they support so far, you'll notice that only a few of current chipsets are supported. I'm sure the list will increase rapidly, and eagerly await support f.ex. the GeForce 256 chipsets. ...So I could buy one. :)
LZW and LZ77/78 are only common in that they are both called dictionary algorithms. While LZW calculates a dictionary of 4096 entries on the fly while decoding or encoding, LZ77/78 commonly use a sliding dictionary of 4096 bytes of previous data. Look for some implementations of LZW with adaptive code length and/or huffman encoding combined and LZ77/78 with huffman or mathematical encoding to find the signifigance of those sizes by yourself.
Data compression FAQ will also be a good source of information and several good books have been written on the subject.
Quite an interesting point on the patent quarrel is to note that v42bis-compression used on every modem operating on phone lines uses LZW compression, from the days of the good old v32 modems. (9600bps) Unisys could try to sue the big companies like Rockwell, unless they have paid the fees.
btw, LZW stands for Lempel, Ziv and Welch.
If I recall right the group played around 40 farewell gigs, now they must have changed their minds since they call it a reunion. Nice to know that they'll continue playing since I like their music a lot.
noticed the author mentioned that the Hardware Abstraction Layer (HAL) was used for security purposes.
Yes, HAL was used on NT3, but after the GUI evolved, it turned out that using HAL was painfully slow, so now on recent NT versions, too many things just bypass HAL and all of them give their little contribution into making NT unstable.
Originally HAL was a good idea, but the implementation just evolved into a horrible kludge, like the whole NT itself. Please correct if I'm wrong on this.
I like how I go to one website, and it automatically tells me what I do or do not have installed.
Have your considered what that webpage is allowed to do on your system (without even you knowing) because it can tell you what do you have patched and what you haven't?
And why it doesn't work correctly with Netscape browsers is that they won't let a web page to run programs that search your system. (Well Netscapes have holes also but..) Anyway, my point is that I would never grant such rights for a remote automated robot, and not even trained human other than mysel.
Anyone could write a program that completely screws your hard drive if you IE and surf to that page. Only barrier between disaster being that notorius window asking user does he trust the program, asking for permission to execute it. After that...
Interesting indeed, why do they feel they are held back. I think they have still as good possibility as all others for new innovations and the even have the resources for huge research.
Still lately M$-attitude has improved a lot. Specs for office formats etc. have been released to public, I just hope it also means that new versions would be compatible. Nowadays they also admit security flaws and are keen to provide users with patches.
Win2k includes many huge improvements ddns, kerberos, mount points, etc. I just don't like the way they are implemented because you can't take the advance of them if you are using w2k workstations and unix servers. Could be they are really pushing w2k to replace unix on server side. Let's hope they will provide us with specs.
I wouldn't trust IE running any java-applications. Personally I use appletviewer etc. every time for my java developement. Just take some time investigating what a java-application can do under IE without the user even knowing. And yes, IE3 crashed my whole system many times during test phases and that was a java-applet.
By trademarking this in country X, he is probably trying to sue companies using the trademark in the country X and get rich... Just some greedy guy behind that. What he hasn't realised though is that it would never hold in court unless he has some product with a name that can be trademarked.