I believe that some compilers have switches along the lines of "-fSPEC95", which produce optimizations just for that SPECmark. The lesson: all benchmarking is useless, unless they're running the same program you want to run, and you only care about the items that they're benchmarking. And even then I wouldn't trust them to have gotten it right.
I'm not really going for this unethical business. All a benchmark shows is how fast that benchmark runs. Extrapolation to other programs is left to marketing departments and naive buyers.
I like the Quake3 benchmarks, because I play Quake3. I don't care about the multimedia creation benchmarks, because they're totally irrelevant. And few people benchmark the stuff I do use: LaTeX, gcc/gmake, Mozilla, and Emacs. You win some, you lose some.
And please don't ask, I have no idea how to benchmark Emacs (zippy quotes per second?), or why you'd bother. And LaTeX was fast on a 286, reducing the need for benchmarking.
Hardware gets used differently by different operating systems, which can lead to different levels of stability and performance.
I know one funny story about hardware acting different with different operating systems. I might have some details wrong, but here goes:
People had run VMS on VAX machines for a fair bit when UNIX was finally available for VAX. When UNIX was installed, there was either data corruption or instability, probably both. Investigation revealed that memory refresh circuitry for the dynamic rams (were these really dynamic rams, or something older that still needed refresh?) wasn't working properly. VMS ran fine because it frequently accessed most all of the memory, which as a side effect refreshed the memory. UNIX, however, made less frequent addresses to some parts of the memory, and these memory cells would not be properly refreshed by the hardware.
"Beg to differ. Most colleges use a) what ever they've been using forever, or b) what the profs who have funding want to use for a given course."
You left out
c) Whatever software was "generously" donated to the University. I've seen MS do this quite a bit, and I scoff at the "generosity" for several reasons:
i) They provide site licenses or many licenses,
but only allow for one set of disks and one
set of manuals. That's not enough for even
a smallish University.
ii) I believe they get to write off those
licenses at full retail. However, full
retail purchases include manuals, support
for each licensee (to some meager extent),
and media for each licensee. By donating
only one set of media and manuals, Microsoft
doesn't even have to pay manufacturing costs
to obtain the tax write off.
To defend myself against expected responses, I'll add that I am not talking about Bill and Melinda Gate's charitable organization. I'm talking about the company Microsoft doing some very saavy marketing, financial manouevers, and public relations, but end users aren't benefitted in the same measure as Microsoft is for its "generosity". This is meant to be a quantitative argument, not qualitative -- however, I don't have numbers available, so my argument is weak. I believe the point still stands.
-Paul Komarek
His point may be valid, though. The secretaries I work with have spent a lot of time dealing with both Windows and Linux applications. They like some of each, but all of neither. It is possible to pay for training on Windows apps, just as with Unix apps -- neither is cheap. If you're the kind of place that pays people to train your employees, you'll do it for both Windows and Unix.
Overall, I think this training issue is a red herring when it comes to end users. If you are hiring inexperienced admins, then I would expect training to be a huge issue.
First time I installed linux, it didn't support the bizarre-o serial ports on my Packard Hell 386 SUX-20, so I gave up on it (that was early 1993). Later, I got a copy on CD with a small manual in the sleeve ("How to Install X Without Calling the Fire Department") and we installed it on a friends 486DX 50 (not many of those were made). xroach was terrifice! I was so let down when I had a 486DX-2 100, because xroach ran too fast to be enjoyable. But at least I was hooked. This was about 1995. That's 2 quality years of linux I wasted because of crappy serial ports!
Choplifter was an awesome game! I played it primarily on Apples if I recall correctly. I wonder if anyone remembers Lemonade Stand? I really liked Lemonade Stand. It required some business sense combined with gambling. And there was this little Star Wars game, where you could jump the landspeeder over barrels, and attack AT-ATs with one of those small fighters ships. It seems like there were four modes, but those are the only two I remember.
Good heavens. Now I'm remembering learning about the DATA statement in basic, and trying to decide if it was really worth the effort. I think the most use I made of it was when playing music.
Oh dear, that reminds me of when I was going to "write a video game" because I'd learned enough TI-BASIC. I decided that I first would scan through my basic manual (blue and white, about 8.5" by 11", softcover -- does anyone else remember this book?) to be sure I hadn't missed any useful commands. Sure enough, I hadn't learned about for-loops yet! "Ready to write a game" without using for-loops. Heh. Oh, and "My First Book of Microcomputers", with some kid named Chip who interviewed computers about what they were made of (yellow, paperback, anyone remember it?). I've still got that book, it's fun to browse.
And I did not understand if-expressions that didn't have a binary operator; what did they mean by if (value) then whatever? Hoooeeee. I guess that trip down memory lane was probably only fun for me. Well, I'll thank myself for sharing...;-)
I really miss my TI-99 4/A. We gave it to relatives around 1986 or so. It constantly had ROM problems, but for about $50 you could send it in and get a refurb unit. And the peripheral expansion box, with a disk drive and memory expansion card! It had the worst connector known to man, with a cable so thick you could tow vehicles with it. But the speech synthesizer rocked. It sounded great, especially that female voice used in the game Parsec.
Then there was the 5 year dry spell until my family could afford a PC -- especially since I would be pretty much the only one to use it. I guess that five years without a computer is why I'm a math grad who really loves humanities. Maybe that's a good thing, after all.
I don't disagree with most of what you say (though I can't help but wonder how an R5900@294MHz could do fp as slow as a K6/233). However, there is something funny about the line "...unless you want to hand code MIPS ASM.". You make it sound like hand coding MIPS ASM is hard or unpleasant. Hand coding MIPS ASM is a joy. The ugliest thing I can think of is the branch delay slot (does it exist on the R5900?), and even that is beautiful if you know why it is there.
-Paul Komarek
One other warning, don't put them on opposite sides of a finger. While it won't necessarily squeeze your finger off, the problem comes if and when they slip and lock together with some of your skin between them.
Build with -p if you wish to analyze with prof (I've never done this), or -gp if you want to analyze with gprof. Then learn how to use prof or gprof. Learning to use gprof is a good investment for your time. The only difficult part is correctly interpreting the analysis returned by gprof, which is very detailed. It's not that bad, but it can be daunting at first.
Or you can use the old "ctrl-c" profiling method: run the program in a debugger, and stop it at random times with ctrl-c. Each time, make a note of which function you interrupted. If one function shows up a lot, then optimizing it is probably your best bet for improving performance.
Reliability is one issue. Moving parts make noise and fail early. I've seen a quite a few cpu fans go bad in under a year, and most seem to go bad in under two years.
> It just means they're getting more powerful, like real computers.
More speed and computational power doesn't mean more heat and electrical power consumption. Compare ENIAC to modern wristwatches with calculators. More fairly, compare my K6-233 to the 209MHz Strongarm 1110 in my iPAQ.
Your comment is amusing. What do you mean by "real computers"? I'm guessing that you mean "whatever crap Dell told me is a real computer". That's an uncharitable suggestion, of similar naivite to your comment.
You've given me an idea for a deterrent. After winning an auction, ask the seller for a final quote for payment purposes. Specifically mention that you require this so that the seller is subject to Mail Fraud laws. While some sellers might take offense, and some criminals may laugh at you and send it anyway, I expect most small criminals would think twice before doing "business" with you.
I've seen it suggested that it wasn't until the 1950's or so that "computer" referred to a machine. Whether the machines in question were IBM punch machines or the UNIVAC, I don't recall. Since my copy of "History of Modern Computing" hasn't been returned yet, that's where this comment ends.
Backing up the comment on the filesystem: don't forget to check all the "etc" directories when looking for binaries like "ping"!
-Paul Komarek
Re:Not biased, just practical
on
Mozilla 0.9.5
·
· Score: 2
> There are so many happy Konq users, you would be surprised!
Well, actually I wouldn't be surprised. I've heard a lot of support for Konq. As it turns out, though, I'm not really personally interested in KDE except insomuch as I might mention it to others (please, no offense!), for instance my parents (again, no offense!).
I hated CDE, and probably for similar reasons I haven't been happy with KDE's window manager. In fact, I'm not completely convinced that Gnome is worthwhile; I think I could still be very happy with fvwm2-derived window managers like Afterstep. I know how their config file systems work (and edit them manually), and I know I can configure everything I want. I've become accustomed to configuring Sawfish mostly how I want, and have never figured out how to make KDE's window manager do what I want.
At this point, it is difficult for me to justify spending time using KDE. In the end, I just need emacs and terminal windows, and something that doesn't fall over when I browse the web. I've become very happy with Mozilla. Back when I was unhappy with Mozilla, I was also unhappy with Netscape, Opera, and Konq. In fact, only lynx has never let me down. Well, that and "telnet slashdot.org 80".;-)
Of course, it's difficult for me to justify spending time writing slashdot comments, but I do it anyway. Hmm... =-)
The Tom's Hardware website has noise ratings on a lot of hard drives. Furthermore, they appear to be fairly careful in their noise testing methods. I chose my IBM Ultrastar over alternatives partially because of the low dB ratings I found at Tom's hardware.
Also, I've seen people complain about reliability problems with the IBM Ultrastar drives (while others claim that the IBM drives never quit). FWIW, I've been told that if an IBM drive works at the beginning (i.e. for several weeks), it will work forever.
Having used SCSI at home for the last year, I really appreciate the "SCSI difference". Not only have I seen better benchmarks (kernel compile times, etc), but it *feels* snappier in general. Here are some interseting things I've seen:
1) For random access to a cdrom, SCSI kicks IDE's but (on my system). When ripping a CD, espcially with overlapping sector reads, my 24x scsi cdrw kicks my 52x IDE cdrom's butt.
2) I get better performance from my IDE cdrom when using linux's SCSI emulation. That was quite a surprise.
3) SCSI drives typically have 5 year warranties, whereas IDE drives typically have 3 year warranties.
4) My IBM Ultrastar (SCSI hard drive) is much quieter and cooler than my IDE Maxtors and IDE IBM Deskstar. However, new IDE drives may have caught up.
5) You have to be really careful with IDE drives in order to get good performance. For instance, I've seen an IDE drive unable to sustain more than 2MB/sec when attached to the middle of an IDE cable, but sustain 6MB/sec when attached at the end (these speeds are for writes, not reads). With SCSI, once it works (which can be a pain if you skimp on cabling and termination), it goes *fast* and is *robust*.
6) Processor overhead: transfering data between my SCSI devices requires far less cpu help than transfering data between my IDE devices (I believe I have all the right DMA stuff configured for my IDE devices -- it helps, but doesn't make things as nice as with SCSI). The implication is that writing CDs on a SCSI system is more robust than on an IDE system. I've never had a buffer underflow, even when writing CDs while the system had a sustained load over 2.0.
And once you use SCSI, you can be a SCSI snob! You almost have to in order to justify the price for a home machine (unless you work at home like I do). SCSI really is the right way to do things. However, SCSI is doing its best to kill itself off. In that way, SCSI verus IDE is a lot like OS/2 versus Windows.
-Paul Komarek
Re:You arent a casual user
on
Mozilla 0.9.5
·
· Score: 2
If anyone wants "just a browser", use lynx. If you want integrated mail, news, chat, and html editing, use emacs. =-)
-Paul Komarek
Re:Not biased, just practical
on
Mozilla 0.9.5
·
· Score: 2
KDE is the defacto standard in linux? Your argument is based on what vendors are selling, not what people are using. Overall, I wouldn't be surprised if fvwm has a majority position in the window manager numbers. Your argument is on exactly the same lines as those made by people who want to underestimate the number of linux users; these people only look at how many boxed copies are sold.
I've had more stability problems with Konq than I have with Mozilla, but maybe I was using an old version. Opera also gave me a lot of problems, and strangely so has Galeon. One thing that I believe Konq is missing is Windows support. It seems that many people who have posted on this article have been Windows users. Does anyone know for sure whether or not Konq runs on windows?
Also, comparing the speed of html rendering while working over a modem is probably a silly thing to do. To make a comparison, you probably want to compare the rendering of local files. I can see where the incremental tables bit would be nice for a modem user, but this isn't the same as rendering speed. Another point: you argue that Konq will be the first browser a new linux user uses. What I want to know is which browser will be the last the use!;-)
Anyway, we can all be thankful that whether or not any browser becomes a standard (or defact standard) for linux, we'll still have many good choices. And even with Sun and IBM getting behind Gnome for their AIX and Solaris desktops, you'll still be able to use KDE if you prefer (I'm assuming someone has managed to run KDE on AIX and Solaris).
Slashdot Mirror
Okay, in what way is Q3A not "real world"? ;-)
-Paul Komarek
I believe that some compilers have switches along the lines of "-fSPEC95", which produce optimizations just for that SPECmark. The lesson: all benchmarking is useless, unless they're running the same program you want to run, and you only care about the items that they're benchmarking. And even then I wouldn't trust them to have gotten it right.
-Paul Komarek
I'm not really going for this unethical business. All a benchmark shows is how fast that benchmark runs. Extrapolation to other programs is left to marketing departments and naive buyers.
I like the Quake3 benchmarks, because I play Quake3. I don't care about the multimedia creation benchmarks, because they're totally irrelevant. And few people benchmark the stuff I do use: LaTeX, gcc/gmake, Mozilla, and Emacs. You win some, you lose some.
And please don't ask, I have no idea how to benchmark Emacs (zippy quotes per second?), or why you'd bother. And LaTeX was fast on a 286, reducing the need for benchmarking.
-Paul Komarek
Hardware gets used differently by different operating systems, which can lead to different levels of stability and performance.
I know one funny story about hardware acting different with different operating systems. I might have some details wrong, but here goes:
People had run VMS on VAX machines for a fair bit when UNIX was finally available for VAX. When UNIX was installed, there was either data corruption or instability, probably both. Investigation revealed that memory refresh circuitry for the dynamic rams (were these really dynamic rams, or something older that still needed refresh?) wasn't working properly. VMS ran fine because it frequently accessed most all of the memory, which as a side effect refreshed the memory. UNIX, however, made less frequent addresses to some parts of the memory, and these memory cells would not be properly refreshed by the hardware.
-Paul Komarek
"Beg to differ. Most colleges use a) what ever they've been using forever, or b) what the profs who have funding want to use for a given course."
You left out
c) Whatever software was "generously" donated to the University. I've seen MS do this quite a bit, and I scoff at the "generosity" for several reasons:
i) They provide site licenses or many licenses,
but only allow for one set of disks and one
set of manuals. That's not enough for even
a smallish University.
ii) I believe they get to write off those
licenses at full retail. However, full
retail purchases include manuals, support
for each licensee (to some meager extent),
and media for each licensee. By donating
only one set of media and manuals, Microsoft
doesn't even have to pay manufacturing costs
to obtain the tax write off.
To defend myself against expected responses, I'll add that I am not talking about Bill and Melinda Gate's charitable organization. I'm talking about the company Microsoft doing some very saavy marketing, financial manouevers, and public relations, but end users aren't benefitted in the same measure as Microsoft is for its "generosity". This is meant to be a quantitative argument, not qualitative -- however, I don't have numbers available, so my argument is weak. I believe the point still stands.
-Paul Komarek
His point may be valid, though. The secretaries I work with have spent a lot of time dealing with both Windows and Linux applications. They like some of each, but all of neither. It is possible to pay for training on Windows apps, just as with Unix apps -- neither is cheap. If you're the kind of place that pays people to train your employees, you'll do it for both Windows and Unix.
Overall, I think this training issue is a red herring when it comes to end users. If you are hiring inexperienced admins, then I would expect training to be a huge issue.
-Paul Komarek
I think you missed 'HMS "two sheds" Jackson', a name from a Monty Python sketch. A fairly funny sketch, at that.
-Paul Komarek
First time I installed linux, it didn't support the bizarre-o serial ports on my Packard Hell 386 SUX-20, so I gave up on it (that was early 1993). Later, I got a copy on CD with a small manual in the sleeve ("How to Install X Without Calling the Fire Department") and we installed it on a friends 486DX 50 (not many of those were made). xroach was terrifice! I was so let down when I had a 486DX-2 100, because xroach ran too fast to be enjoyable. But at least I was hooked. This was about 1995. That's 2 quality years of linux I wasted because of crappy serial ports!
-Paul Komarek
Choplifter was an awesome game! I played it primarily on Apples if I recall correctly. I wonder if anyone remembers Lemonade Stand? I really liked Lemonade Stand. It required some business sense combined with gambling. And there was this little Star Wars game, where you could jump the landspeeder over barrels, and attack AT-ATs with one of those small fighters ships. It seems like there were four modes, but those are the only two I remember.
Good heavens. Now I'm remembering learning about the DATA statement in basic, and trying to decide if it was really worth the effort. I think the most use I made of it was when playing music.
Oh dear, that reminds me of when I was going to "write a video game" because I'd learned enough TI-BASIC. I decided that I first would scan through my basic manual (blue and white, about 8.5" by 11", softcover -- does anyone else remember this book?) to be sure I hadn't missed any useful commands. Sure enough, I hadn't learned about for-loops yet! "Ready to write a game" without using for-loops. Heh. Oh, and "My First Book of Microcomputers", with some kid named Chip who interviewed computers about what they were made of (yellow, paperback, anyone remember it?). I've still got that book, it's fun to browse.
And I did not understand if-expressions that didn't have a binary operator; what did they mean by if (value) then whatever? Hoooeeee. I guess that trip down memory lane was probably only fun for me. Well, I'll thank myself for sharing...;-)
I really miss my TI-99 4/A. We gave it to relatives around 1986 or so. It constantly had ROM problems, but for about $50 you could send it in and get a refurb unit. And the peripheral expansion box, with a disk drive and memory expansion card! It had the worst connector known to man, with a cable so thick you could tow vehicles with it. But the speech synthesizer rocked. It sounded great, especially that female voice used in the game Parsec.
Then there was the 5 year dry spell until my family could afford a PC -- especially since I would be pretty much the only one to use it. I guess that five years without a computer is why I'm a math grad who really loves humanities. Maybe that's a good thing, after all.
-Paul Komarek
I don't disagree with most of what you say (though I can't help but wonder how an R5900@294MHz could do fp as slow as a K6/233). However, there is something funny about the line "...unless you want to hand code MIPS ASM.". You make it sound like hand coding MIPS ASM is hard or unpleasant. Hand coding MIPS ASM is a joy. The ugliest thing I can think of is the branch delay slot (does it exist on the R5900?), and even that is beautiful if you know why it is there.
-Paul Komarek
One other warning, don't put them on opposite sides of a finger. While it won't necessarily squeeze your finger off, the problem comes if and when they slip and lock together with some of your skin between them.
-Paul Komarek
Build with -p if you wish to analyze with prof (I've never done this), or -gp if you want to analyze with gprof. Then learn how to use prof or gprof. Learning to use gprof is a good investment for your time. The only difficult part is correctly interpreting the analysis returned by gprof, which is very detailed. It's not that bad, but it can be daunting at first.
Or you can use the old "ctrl-c" profiling method: run the program in a debugger, and stop it at random times with ctrl-c. Each time, make a note of which function you interrupted. If one function shows up a lot, then optimizing it is probably your best bet for improving performance.
I prefer gprof. =-)
-Paul Komarek
Reliability is one issue. Moving parts make noise and fail early. I've seen a quite a few cpu fans go bad in under a year, and most seem to go bad in under two years.
> It just means they're getting more powerful, like real computers.
More speed and computational power doesn't mean more heat and electrical power consumption. Compare ENIAC to modern wristwatches with calculators. More fairly, compare my K6-233 to the 209MHz Strongarm 1110 in my iPAQ.
Your comment is amusing. What do you mean by "real computers"? I'm guessing that you mean "whatever crap Dell told me is a real computer". That's an uncharitable suggestion, of similar naivite to your comment.
-Paul Komarek
You've given me an idea for a deterrent. After winning an auction, ask the seller for a final quote for payment purposes. Specifically mention that you require this so that the seller is subject to Mail Fraud laws. While some sellers might take offense, and some criminals may laugh at you and send it anyway, I expect most small criminals would think twice before doing "business" with you.
-Paul Komarek
I've seen it suggested that it wasn't until the 1950's or so that "computer" referred to a machine. Whether the machines in question were IBM punch machines or the UNIVAC, I don't recall. Since my copy of "History of Modern Computing" hasn't been returned yet, that's where this comment ends.
-Paul Komarek
I must be thinking of SunOS instead of Solaris. Most often, I don't bother differentiating between the two.
-Paul
Backing up the comment on the filesystem: don't forget to check all the "etc" directories when looking for binaries like "ping"!
-Paul Komarek
> There are so many happy Konq users, you would be surprised!
;-)
Well, actually I wouldn't be surprised. I've heard a lot of support for Konq. As it turns out, though, I'm not really personally interested in KDE except insomuch as I might mention it to others (please, no offense!), for instance my parents (again, no offense!).
I hated CDE, and probably for similar reasons I haven't been happy with KDE's window manager. In fact, I'm not completely convinced that Gnome is worthwhile; I think I could still be very happy with fvwm2-derived window managers like Afterstep. I know how their config file systems work (and edit them manually), and I know I can configure everything I want. I've become accustomed to configuring Sawfish mostly how I want, and have never figured out how to make KDE's window manager do what I want.
At this point, it is difficult for me to justify spending time using KDE. In the end, I just need emacs and terminal windows, and something that doesn't fall over when I browse the web. I've become very happy with Mozilla. Back when I was unhappy with Mozilla, I was also unhappy with Netscape, Opera, and Konq. In fact, only lynx has never let me down. Well, that and "telnet slashdot.org 80".
Of course, it's difficult for me to justify spending time writing slashdot comments, but I do it anyway. Hmm... =-)
-Paul Komarek
All these promising suggestions, it almost makes me weep! ;-)
-Paul Komarek
The Tom's Hardware website has noise ratings on a lot of hard drives. Furthermore, they appear to be fairly careful in their noise testing methods. I chose my IBM Ultrastar over alternatives partially because of the low dB ratings I found at Tom's hardware.
Also, I've seen people complain about reliability problems with the IBM Ultrastar drives (while others claim that the IBM drives never quit). FWIW, I've been told that if an IBM drive works at the beginning (i.e. for several weeks), it will work forever.
-Paul Komarek
Having used SCSI at home for the last year, I really appreciate the "SCSI difference". Not only have I seen better benchmarks (kernel compile times, etc), but it *feels* snappier in general. Here are some interseting things I've seen:
1) For random access to a cdrom, SCSI kicks IDE's but (on my system). When ripping a CD, espcially with overlapping sector reads, my 24x scsi cdrw kicks my 52x IDE cdrom's butt.
2) I get better performance from my IDE cdrom when using linux's SCSI emulation. That was quite a surprise.
3) SCSI drives typically have 5 year warranties, whereas IDE drives typically have 3 year warranties.
4) My IBM Ultrastar (SCSI hard drive) is much quieter and cooler than my IDE Maxtors and IDE IBM Deskstar. However, new IDE drives may have caught up.
5) You have to be really careful with IDE drives in order to get good performance. For instance, I've seen an IDE drive unable to sustain more than 2MB/sec when attached to the middle of an IDE cable, but sustain 6MB/sec when attached at the end (these speeds are for writes, not reads). With SCSI, once it works (which can be a pain if you skimp on cabling and termination), it goes *fast* and is *robust*.
6) Processor overhead: transfering data between my SCSI devices requires far less cpu help than transfering data between my IDE devices (I believe I have all the right DMA stuff configured for my IDE devices -- it helps, but doesn't make things as nice as with SCSI). The implication is that writing CDs on a SCSI system is more robust than on an IDE system. I've never had a buffer underflow, even when writing CDs while the system had a sustained load over 2.0.
And once you use SCSI, you can be a SCSI snob! You almost have to in order to justify the price for a home machine (unless you work at home like I do). SCSI really is the right way to do things. However, SCSI is doing its best to kill itself off. In that way, SCSI verus IDE is a lot like OS/2 versus Windows.
-Paul Komarek
If anyone wants "just a browser", use lynx. If you want integrated mail, news, chat, and html editing, use emacs. =-)
-Paul Komarek
KDE is the defacto standard in linux? Your argument is based on what vendors are selling, not what people are using. Overall, I wouldn't be surprised if fvwm has a majority position in the window manager numbers. Your argument is on exactly the same lines as those made by people who want to underestimate the number of linux users; these people only look at how many boxed copies are sold.
;-)
I've had more stability problems with Konq than I have with Mozilla, but maybe I was using an old version. Opera also gave me a lot of problems, and strangely so has Galeon. One thing that I believe Konq is missing is Windows support. It seems that many people who have posted on this article have been Windows users. Does anyone know for sure whether or not Konq runs on windows?
Also, comparing the speed of html rendering while working over a modem is probably a silly thing to do. To make a comparison, you probably want to compare the rendering of local files. I can see where the incremental tables bit would be nice for a modem user, but this isn't the same as rendering speed. Another point: you argue that Konq will be the first browser a new linux user uses. What I want to know is which browser will be the last the use!
Anyway, we can all be thankful that whether or not any browser becomes a standard (or defact standard) for linux, we'll still have many good choices. And even with Sun and IBM getting behind Gnome for their AIX and Solaris desktops, you'll still be able to use KDE if you prefer (I'm assuming someone has managed to run KDE on AIX and Solaris).
-Paul Komarek
Wow, that's too bad. Any bug that disabled the blink tag would be a feature, in my opinion.
-Paul Komarek
Clearly this is not a watch for the "vast majority".
-Paul Komarek