I don't know if any of you have noticed it yet but today is the second week in October and we all know what that means. Daylight savings time is here and it's time to "fall back"!! Where does the time go?
You will need to turn your clocks back an hour at 2:00 A.M on Sunday morning or you can do what I do and change them when you go to bed on Saturday night.
Based upon what my best clients are asking about I have a few ideas.
1)We'll see more of a move away from the P2P top tier layer of distributed computing and a slight shift back to the homogeneuos (n+ 1) client/server paradigm. The main thrust behind this is security reasons, or the lack thereof as the standard Client/server architecture is an order of magnitude more secure than P2P.
2) The computing industry as a whole will begin to migrate to more secure computing languages and stop using java and c/c++. This is already taking place as there is a group of coders looking at re-writing the Linux kernal in Fortran 77 for the speed improvements as well as the improved security over straight C.
3) NSA will finally release their Linux distro which will take the world by storm as it will be the most secure server OE and very user friendly as well. Agian, I've seen some of the GUI work they have in place and it will take your breath away. Since I'm still under NDA I can't say any more about that.
To summarize, as computers are now pervasive and hardware is a commodity we will now start to see the focus on security.
I don't know if you remember me but we sat together at the IOAC conference in Cupertino back in 1998. I still impress my friends with that story. Anyways, what I want to know is:
If two trains are exactly 150 miles apart and one of them heads north on the tracks at 30 mph heading into wind gust of 25 mph which only occur at 15 second intervals for exactly 10 seconds and the second one heads south at 27 mph on a 15 degree incline with a crosswind of 12 mph at an angle of 42 degrees, how long will it take for them to meet?
I've recently been commissioned by a client to do some work on a linux server farm for them. I think I'm going to initially recommend that they look at using Linux 8.0 since it's got the new Bluetooth interface integrated into the distro and it's a very mature product. This should be a great help to them as they all carry palm pilots and I think wireless data access is the wave of the future.
However Lindows might be worth looking into for some of the client workstations but I usually won't touch any software until it's in it's fourth or fifth rev.
Really. And despite all the passion RMS may have about free software and why it's important almost nobody else cares.[1] Most people care about stuff like politics or keeping their jobs or the failing health of a loved one or whatever. Software is a tool they use to send emails to Aunt Susie or make a cool greeting card for thier bridge club. The whole concept that using software might have some moral issues involved means nothing to them. I don't mean that to be cruel or harsh but I honestly believe that to be true.
And, if they do care they certainly have lots of opportunities to look at the software they are using and read some docs about it or click help->about or whatever. It's pretty easy to do and it's actually how I learned about GNU and the FSF. I was learning to use emacs in RedHat 5.0 and read a little about it and was lead to RMS and GNU. Nothing too difficult about it all.
But once again, this all leads to the bottom line which is that RMS wants people to use "free" software as opposed to proprietary software and Linux has been the most effective tool for that to happen. Regardless of whether the people understand it or not it serves his ultimate purpose.
[1] - Relatively speaking anyways. What percentage of the worlds population do you think would care even if they did know all about it? Less than one percent I'll bet. I know my mom and sisters and grandparents certainly wouldn't give a crap.
I mostly agree with everything you just said, however that doesn't change anything.
It's certainly true that timing played a big part of Linux becoming popular and Linus certainly wasn't a visionary or a prophet, just a guy who wanted a better terminal emulater to read his news from the Uni, and things fell into place from there.
So no, I'm not claiming that Linus should be held in high esteem due to his brilliant forsight, I'm merely claiming that Linus/Linux was in the in the right place at the right time and it was the reason that all the great Gnu stuff got bundled and blown out to the masses.
Sure, if not Linux than maybe something else, but that's a moot point here. For the life of me I can't understand why Richard isn't thrilled about the millions of people who are using Gnu stuff who weren't using it before instead of being so hung up on the name issue. You don't see the XFree86 people making this an issue do you?
The popularity of Linux has made the GNU software available for millions of people who never used it before and many of those might not have ever used it otherwise. Thus Linux has proven to be a fantastic avenue for Richard and his dream of people using the useful applications the FSF people have built.
But now his ego is getting in the way and he wants to make it all about something it's not. The name is irrelevant, the fact that people are using free software is not. Well... until RMS decided he wanted to make the name an issue. Honestly, if having the GNU name on it was so impotant he should have made it part of the GPL and forced people to use it.
Richard should be happy about the success of Linux because without Linux nobody but a few geeks would know about GNU and the FSF, let alone give a damn about it.
Did you think that you were somehow above the law or do you view stealing as something that's okay as long as you are the one doing it and you don't get caught?
Also:
Did you ever think about the money that you were taking away from honest people who work hard and rely on the profits they generate from software sales or did you only think about yourself while committing your crimes?
In conjunction with two other large companies who wish to remain anonymous at this time, Wagner Consulting LLC is instituting a Macintosh buyback program. We are offering $3,000 for any and all makes of Macintosh computers, no questions asked.
Simply contact one of my employees at 1-888-WAGNERC and they will provide you with the details on how to send your Mac to me for full payment. No longer will you need to be oppressed and chastized for your lack of computer knowledge. There is no need to feel ashamed and we won't ask for your name. I realize when you bought your Mac you were tricked into it with slick marketing and you just didn't know any better. Well now I'm here to help you.
I realize you will lose a little money on the deal but we have worked out a special with Dell Computers to receive a %20 discount on all Dells purchased when you show them you Wagner Consulting voucher. Bascially this will enable you to purchase two new faster Dells with more memory and hard drive space.
I plan on donating the obsolete Macs to a third world country.
This is really bogus folks. I've been a Q1 member of the IOOC 911.11 committee for programming languages since the early 90's and I can tell you scientifically with no reservation that this method is faulty beyond words.
The goal of a programming language is to provide a machine with a set of instructions, not to sit down and read it a story. Do you expect your car to be made of parts which have little embedded notes explainging how they were engineered? Of course not, that's just silly.
The natural course of a programming language is to become more powerful and less encapsulated and as a direct member of the standards committe I will do everything in my power to defend that philosophy. These types of gimmicks are mainly for those who cannot remember to match a free() with a malloc() and think that every Joe and Charlie should be able to program a computer.
This is 100% at odds with the famous paradox by Pablo Cruise, Pablo's Paradox, which states that every problem that can be solved will move linearaly until it hits O(n) and then it will move in the opposite direction as commoners try to improve on the already perfect solution.
Warmest regards, --Jack
Re:What's up with this?
on
Linuxworld Fun
·
· Score: 2, Insightful
No biggie, really.
Wagner LLC did some backend work for Linuxexpo as a level II sub contractor for IBM (a big shout out to the chief IBM tech Pablo Cruise in the design lab!!!) and it turns out the company who did the webpage design/artwork was a photoshop/Windows only shop and needed the new.asp extensions that are only experimental in Apache so the whole works ended up on and NT server. This stuff is always contracted out and means nothing regarding the whole webserver debate. It's just a matter of what tools they are comfortable using.
Quotes Linux Torvalds: "- the page cache works with index/offset, and that should be your first
priority, since the page cache is all that matters from a performance
standpoint."
That's all well and good for litle Endian OS's, but since you are dealing with a static offset you have one extra instruction lookup for all big endian machines. Thus if you port Linux to Sparc or Alpha you not only see a performance degredation of O(logN) but you loose one register spot on the level II chache for the offest lookup. In other words it will be slower, much much slower.
I actually spent a bit of time on the SCSI board of standards back when they were commited towards making great strides in I/O throughput, and they have some real terrific advances that will be released in the next year when their patent's get approved. It's all hush hush and on the QT at the moment and my NDA runs through the end of 2003 so that's all I can say about it.
After briefly looking at the spec for Serial ATA I can see several limitations which will most certainly cause a bottelneck for any I/O simply due to the insane RPM's they require to actually hit their maximum VtR/mg (moving heads [sic] to magnetic resonence ratio). They need to have a look at perhaps doing some embedded hardware hacks which will allow them less resistance in the SCSI channel on the motherboard as opposed to quantum magnetic research. Just my opinion of course and not a reflection of the opinion of my company.
Wagner Consulting did some consulting work for the Chinese ministry of telecom infrastructure in the mid 90's when I was an expat and at that time they were attempting to migrate from the old BSDish SYSV3 TCP/IP to the newest, which at that time was still BSD based and wide open to hackers. The real issue was the need to encrypt the packets at the final TC link before they were run through the RF translator and sent into the air. At that time they are open to intercept by anyone with a mobile phone and a simple electric razor. It's quite simple to use the steel filiments which reside in the razor as an conductor for the mobile phone to pick up any satalitte channel you wish.
At that point you simply program your mobile phone to send out the pirated signals and you pretty much own them. Meanwhile a simple RC3 encryption scheme would have prevented this whole mess.
As a side note I left my position with them due to this very issue so at not to disparage the reputation of Wagner Consulting.
This is simply not true. I happen to know, having done some consulting work for NASA that they are using a proprietary planet tracking system that runs on openBSD which is reporting false positives.
I don't want to get too technical here but basically the openBSD kernel doesn't use any real-time scheduling techniques and instead relies on an O(log(N)) algorithm which causes all double precicsion calculations to be off by +/-.0005. A smal number to be sure but when you are talking about large planets that are billions of miles away it make a huge difference.
For the record I resigned my post after they refused to address my concerns with using a substandard operating system to deal with these important matters.
They still have the vestiges of some of the *BSD core components in Solaris (as you should be aware, SunOS was BSD based until they made the switch to SVR4 and renamed it Solaris) which cause some horrible bottlenecks due to some of the legacy single threaded caching code and it's 32 bit code origin. That's one of the main reasons that Sun has had a hard time recruiting some of the top shelf 3d engineers from the SGI side. I did some consulting work for Sun and I tried to port some of the BSD code but it was a real train wreck and not a simple task. It was loaded with lots of bit field shifting and unsafe typecasts, quite frankly it was an abortion.
Heed this warning, if you buy one of these to do rendering make sure to benchmark it before you send Sun a check.
When I was doing some contractor work for a huge *nix shop (think purple)I met a fellow who told me an interesting tale. It seems this huge *nix company (think purple) had actually spent a week with RMS and some of the HURD developers to talk with them about using the base code from the HURD for a project they were kicking about. The company would have been willing to give back some of the code, under a community type BSD license, which would have brought the HURD up to a Version 1.0 level. Now bear in mind I got this story second hand but the guy who told me was a very reputable source who had been part of the compiler team for years there. He let it slip out while we were discussing the flaws in the BSD threading model and once the cat was out of the bag he spilled his guts.
Anyways, the long and the short of it was that RMS threw a giant hissy fit about the license so they never did business together. It seems that RMS can't see the forrest for the trees sometimes. Instead of giving the community a rock-n-roll new kernel, he decided to cut off his nose to spite his face.
The P4 is only optomized for a Microsoft compiler. It's true. I was doing some consulting work for a major fortune 100 company and they were looking into migrating from Windows to openBSD. However after they did some testing they found that their database applications were running 15 - 20% slower on BSD than Windows. I expected them to be a little slower due to the threading problems wiht BSD, but not that slow.
After one full week of testing we found the problem wasn't with BSD at all, it was with the P4 on BSD. It would seem Intel has an enhanced instruction set cache which is only available on Microsoft compilers. This is not a trivila thing to implement so I doubt the OSS camp will be able to migrate it into their compilers anytime soon.
Obviously this "test" is a crock, as FreeBSD would not be that far behind, what with it's surperior network stack and thread handling
Thanks AC for the nice plug. I'm a frontline developer for FreeBSD and I've been tasked with doing a complete overhaul for the tcp/ip stack, as well as doing a POSIX implementaion of threads, something it sorely needed.
It's quite possible this test was done without the BSD_booey_stack extensions compiled in, that would account for the less than stellar results. It's almost impossible for the tcp/ip stack to be the bottleneck for any problems anymore since I introduced double pass ID caching into the CVS. Basically it uses user space memory to reduce the overhead on the kernel, while performing table lookup translations on a stack hashtable. It's getting still better times if you enable my optimized dynamic MTU settings, even though that's still in the experimental stages.
I don't let this stuff get me down though, when we release the next point version at comdex next year the industry will watch with awe, it's going to be a smoking little OS by then.
It seems we have now entered the age of paranoia where we are hell bent on throwing the baby out with the bathwater. Say what you will about Microsoft but the facts are these, with their massive resources they have a collective intelligence that is well above and beyond anything that the average IT person has. It's time to make the choice, do I bury my head in the sand and stand alone, or embrace it and let myself grow and expand my horizons.
It's not so much about MS owning content, that's a silly premise, it's all about becoming one with the collective groupthink, it's about becoming one with society. Technology changes at the speed of light, you can either stay abreast with those changes, or go home and line your walls with tinfoil. I can promise you this, information is power, and you will soon see a change in the social classes based upon information. I refuse to be left behind based upon nameless fears and unfounded paranoia.
Netscape 6.0 is an abomination. It's a crime against humanity. Why should they feel obligated to bundle that monstrosity just because it's says Netscape on it? For my money,.90 is not there yet either which leaves linux with either Konquerer or Opra I guess. Konquerer is not all there so that leaves Opra. I don't know but it seems like a pretty sad state of affairs for the Linux camp. With the major importance of a browser you would think that the Mozilla folks would have tried to build a world class one first, then worry about the 200 applications they want to build into it. But instead they let the engineers run the show which will ultimatly be their downfall.
There is a lesson to be learned there for the OS community. It's not always about scratching an itch when you are trying to compete with commercial applications. You need some strategic planners who are more big picture oriented and can overcome the engineering nonsense. But if you try to suggest anything like that to the *community* you are labeled a troll. Oh well, I guess they can fail on their own and of their own accord.
What are you using your Ultra for? What tests did you do? Remeber the Ultra 5 is, as someone else already mentioned, pretty much a peeeeceeee with a sparc processor.
Linux does somethings faster WRT normal daily tasks, disk read/writes for example, but solaris on a sparc machine under load will completely fry linux. Solaris has the distinct advantage of being desinged specifically for the sparcs, this is very different from Linux which is pretty vanilla so as to work on many different platforms. Solaris is more like a big 18 wheeler compared to Linux which is like a pickup truck. Sure the pickup might get off the line faster but the 18 wheeler will go a lot farther with a much heavier load than the pickup
For the record, I have solaris 8 on an old Sparcstation 20 with a Ross Hyperspac 150 and it is still a very capable machine. I have thought about putting one of the BSD's on it for fun, put I have no trouble with Solaris 8's preformance.
At the risk of repeating an old cliche.. we need to work smarter, not harder. For instance I work about 45 hours a week and I make a very nice six figure salary. It's "the myth of incompetence" that the tech industry loves to propagate with regards to working 80 hour weeks. Maybe your working an 80 hour week becasue you are using the wrong tools or trying to implement the wrong solutions, etc...
I see it all the time though, some people will only use Windows because they are a "windows only shop" when it's obvious that Solaris or HP/UX would do the job with half the effort. I also see people jumping on the bandwagon and using languages like java or python when they should be using a RAD language. But hey, I make money because most people attack problems from the wrong direction so I guess I shouldn't bitch.
I can't wait for some of the Linux distros to start to fail. Yes, competition is a good thing, but the mainstream public doesn't view this as competition, they view it as fragmentation. Several of my clients have told me point blank that they refuse to even look at Linux because they feel the environment is too unstable for a long term commitment. Mind you, they didn't say the product was unstable, but rather the environment. That's the part of the equation that tends to get glossed over by the open source techy types, they tend to focus on the technical merits and they can't understand the need to align towards common goals and reduce any collaborative filtering which quite frankly ends up confusing John. Q. Consumer.
I will continue to recommend Linux as a low TOC and high ROI solution but it's an uphill battle. It's a shame that the "powers that be" won't listen when anyone mentions this line of thinking but I can tell you first hand that out in the real world it's the truth.
Slashdot Mirror
Yes, Hello,
I don't know if any of you have noticed it yet but today is the second week in October and we all know what that means. Daylight savings time is here and it's time to "fall back"!! Where does the time go?
You will need to turn your clocks back an hour at 2:00 A.M on Sunday morning or you can do what I do and change them when you go to bed on Saturday night.
Warmest regards,
--Jack
Based upon what my best clients are asking about I have a few ideas.
1)We'll see more of a move away from the P2P top tier layer of distributed computing and a slight shift back to the homogeneuos (n+ 1) client/server paradigm. The main thrust behind this is security reasons, or the lack thereof as the standard Client/server architecture is an order of magnitude more secure than P2P.
2) The computing industry as a whole will begin to migrate to more secure computing languages and stop using java and c/c++. This is already taking place as there is a group of coders looking at re-writing the Linux kernal in Fortran 77 for the speed improvements as well as the improved security over straight C.
3) NSA will finally release their Linux distro which will take the world by storm as it will be the most secure server OE and very user friendly as well. Agian, I've seen some of the GUI work they have in place and it will take your breath away. Since I'm still under NDA I can't say any more about that.
To summarize, as computers are now pervasive and hardware is a commodity we will now start to see the focus on security.
Warmest regards,
--Jack
I don't know if you remember me but we sat together at the IOAC conference in Cupertino back in 1998. I still impress my friends with that story. Anyways, what I want to know is:
If two trains are exactly 150 miles apart and one of them heads north on the tracks at 30 mph heading into wind gust of 25 mph which only occur at 15 second intervals for exactly 10 seconds and the second one heads south at 27 mph on a 15 degree incline with a crosswind of 12 mph at an angle of 42 degrees, how long will it take for them to meet?
Warmest regards,
--Jack
I've recently been commissioned by a client to do some work on a linux server farm for them. I think I'm going to initially recommend that they look at using Linux 8.0 since it's got the new Bluetooth interface integrated into the distro and it's a very mature product. This should be a great help to them as they all carry palm pilots and I think wireless data access is the wave of the future.
However Lindows might be worth looking into for some of the client workstations but I usually won't touch any software until it's in it's fourth or fifth rev.
--Jack
They don't care.
Really. And despite all the passion RMS may have about free software and why it's important almost nobody else cares.[1] Most people care about stuff like politics or keeping their jobs or the failing health of a loved one or whatever. Software is a tool they use to send emails to Aunt Susie or make a cool greeting card for thier bridge club. The whole concept that using software might have some moral issues involved means nothing to them. I don't mean that to be cruel or harsh but I honestly believe that to be true.
And, if they do care they certainly have lots of opportunities to look at the software they are using and read some docs about it or click help->about or whatever. It's pretty easy to do and it's actually how I learned about GNU and the FSF. I was learning to use emacs in RedHat 5.0 and read a little about it and was lead to RMS and GNU. Nothing too difficult about it all.
But once again, this all leads to the bottom line which is that RMS wants people to use "free" software as opposed to proprietary software and Linux has been the most effective tool for that to happen. Regardless of whether the people understand it or not it serves his ultimate purpose.
[1] - Relatively speaking anyways. What percentage of the worlds population do you think would care even if they did know all about it? Less than one percent I'll bet. I know my mom and sisters and grandparents certainly wouldn't give a crap.
I mostly agree with everything you just said, however that doesn't change anything.
It's certainly true that timing played a big part of Linux becoming popular and Linus certainly wasn't a visionary or a prophet, just a guy who wanted a better terminal emulater to read his news from the Uni, and things fell into place from there.
So no, I'm not claiming that Linus should be held in high esteem due to his brilliant forsight, I'm merely claiming that Linus/Linux was in the in the right place at the right time and it was the reason that all the great Gnu stuff got bundled and blown out to the masses.
Sure, if not Linux than maybe something else, but that's a moot point here. For the life of me I can't understand why Richard isn't thrilled about the millions of people who are using Gnu stuff who weren't using it before instead of being so hung up on the name issue. You don't see the XFree86 people making this an issue do you?
Bah. He's a hypocrite at the very least.
The popularity of Linux has made the GNU software available for millions of people who never used it before and many of those might not have ever used it otherwise. Thus Linux has proven to be a fantastic avenue for Richard and his dream of people using the useful applications the FSF people have built.
But now his ego is getting in the way and he wants to make it all about something it's not. The name is irrelevant, the fact that people are using free software is not. Well... until RMS decided he wanted to make the name an issue. Honestly, if having the GNU name on it was so impotant he should have made it part of the GPL and forced people to use it.
Richard should be happy about the success of Linux because without Linux nobody but a few geeks would know about GNU and the FSF, let alone give a damn about it.
Warmest Regards,
--Jack
Also:
Did you ever think about the money that you were taking away from honest people who work hard and rely on the profits they generate from software sales or did you only think about yourself while committing your crimes?
Warmest regards,
--Jack
In conjunction with two other large companies who wish to remain anonymous at this time, Wagner Consulting LLC is instituting a Macintosh buyback program. We are offering $3,000 for any and all makes of Macintosh computers, no questions asked.
Simply contact one of my employees at 1-888-WAGNERC and they will provide you with the details on how to send your Mac to me for full payment. No longer will you need to be oppressed and chastized for your lack of computer knowledge. There is no need to feel ashamed and we won't ask for your name. I realize when you bought your Mac you were tricked into it with slick marketing and you just didn't know any better. Well now I'm here to help you.
I realize you will lose a little money on the deal but we have worked out a special with Dell Computers to receive a %20 discount on all Dells purchased when you show them you Wagner Consulting voucher. Bascially this will enable you to purchase two new faster Dells with more memory and hard drive space.
I plan on donating the obsolete Macs to a third world country.
Warmest regards,
--Jack
This is really bogus folks. I've been a Q1 member of the IOOC 911.11 committee for programming languages since the early 90's and I can tell you scientifically with no reservation that this method is faulty beyond words.
The goal of a programming language is to provide a machine with a set of instructions, not to sit down and read it a story. Do you expect your car to be made of parts which have little embedded notes explainging how they were engineered? Of course not, that's just silly.
The natural course of a programming language is to become more powerful and less encapsulated and as a direct member of the standards committe I will do everything in my power to defend that philosophy. These types of gimmicks are mainly for those who cannot remember to match a free() with a malloc() and think that every Joe and Charlie should be able to program a computer.
This is 100% at odds with the famous paradox by Pablo Cruise, Pablo's Paradox, which states that every problem that can be solved will move linearaly until it hits O(n) and then it will move in the opposite direction as commoners try to improve on the already perfect solution.
Warmest regards,
--Jack
No biggie, really.
.asp extensions that are only experimental in Apache so the whole works ended up on and NT server. This stuff is always contracted out and means nothing regarding the whole webserver debate. It's just a matter of what tools they are comfortable using.
Wagner LLC did some backend work for Linuxexpo as a level II sub contractor for IBM (a big shout out to the chief IBM tech Pablo Cruise in the design lab!!!) and it turns out the company who did the webpage design/artwork was a photoshop/Windows only shop and needed the new
Warmest regards,
-Jack
That's all well and good for litle Endian OS's, but since you are dealing with a static offset you have one extra instruction lookup for all big endian machines. Thus if you port Linux to Sparc or Alpha you not only see a performance degredation of O(logN) but you loose one register spot on the level II chache for the offest lookup. In other words it will be slower, much much slower.
Warmest regards,
-Jack
I actually spent a bit of time on the SCSI board of standards back when they were commited towards making great strides in I/O throughput, and they have some real terrific advances that will be released in the next year when their patent's get approved. It's all hush hush and on the QT at the moment and my NDA runs through the end of 2003 so that's all I can say about it.
After briefly looking at the spec for Serial ATA I can see several limitations which will most certainly cause a bottelneck for any I/O simply due to the insane RPM's they require to actually hit their maximum VtR/mg (moving heads [sic] to magnetic resonence ratio). They need to have a look at perhaps doing some embedded hardware hacks which will allow them less resistance in the SCSI channel on the motherboard as opposed to quantum magnetic research. Just my opinion of course and not a reflection of the opinion of my company.
Best Regards,
-Jack
Wagner Consulting did some consulting work for the Chinese ministry of telecom infrastructure in the mid 90's when I was an expat and at that time they were attempting to migrate from the old BSDish SYSV3 TCP/IP to the newest, which at that time was still BSD based and wide open to hackers. The real issue was the need to encrypt the packets at the final TC link before they were run through the RF translator and sent into the air. At that time they are open to intercept by anyone with a mobile phone and a simple electric razor. It's quite simple to use the steel filiments which reside in the razor as an conductor for the mobile phone to pick up any satalitte channel you wish.
At that point you simply program your mobile phone to send out the pirated signals and you pretty much own them. Meanwhile a simple RC3 encryption scheme would have prevented this whole mess.
As a side note I left my position with them due to this very issue so at not to disparage the reputation of Wagner Consulting.
Yours,
Jack Wagner.
I don't want to get too technical here but basically the openBSD kernel doesn't use any real-time scheduling techniques and instead relies on an O(log(N)) algorithm which causes all double precicsion calculations to be off by +/-.0005. A smal number to be sure but when you are talking about large planets that are billions of miles away it make a huge difference.
For the record I resigned my post after they refused to address my concerns with using a substandard operating system to deal with these important matters.
Yours,
-Jack
Heed this warning, if you buy one of these to do rendering make sure to benchmark it before you send Sun a check.
Anyways, the long and the short of it was that RMS threw a giant hissy fit about the license so they never did business together. It seems that RMS can't see the forrest for the trees sometimes. Instead of giving the community a rock-n-roll new kernel, he decided to cut off his nose to spite his face.
Yours,
-Jack
The P4 is only optomized for a Microsoft compiler. It's true. I was doing some consulting work for a major fortune 100 company and they were looking into migrating from Windows to openBSD. However after they did some testing they found that their database applications were running 15 - 20% slower on BSD than Windows. I expected them to be a little slower due to the threading problems wiht BSD, but not that slow.
After one full week of testing we found the problem wasn't with BSD at all, it was with the P4 on BSD. It would seem Intel has an enhanced instruction set cache which is only available on Microsoft compilers. This is not a trivila thing to implement so I doubt the OSS camp will be able to migrate it into their compilers anytime soon.
Obviously this "test" is a crock, as FreeBSD would not be that far behind, what with it's surperior network stack and thread handling
Thanks AC for the nice plug. I'm a frontline developer for FreeBSD and I've been tasked with doing a complete overhaul for the tcp/ip stack, as well as doing a POSIX implementaion of threads, something it sorely needed.
It's quite possible this test was done without the BSD_booey_stack extensions compiled in, that would account for the less than stellar results. It's almost impossible for the tcp/ip stack to be the bottleneck for any problems anymore since I introduced double pass ID caching into the CVS. Basically it uses user space memory to reduce the overhead on the kernel, while performing table lookup translations on a stack hashtable. It's getting still better times if you enable my optimized dynamic MTU settings, even though that's still in the experimental stages.
I don't let this stuff get me down though, when we release the next point version at comdex next year the industry will watch with awe, it's going to be a smoking little OS by then.
It seems we have now entered the age of paranoia where we are hell bent on throwing the baby out with the bathwater. Say what you will about Microsoft but the facts are these, with their massive resources they have a collective intelligence that is well above and beyond anything that the average IT person has. It's time to make the choice, do I bury my head in the sand and stand alone, or embrace it and let myself grow and expand my horizons.
It's not so much about MS owning content, that's a silly premise, it's all about becoming one with the collective groupthink, it's about becoming one with society. Technology changes at the speed of light, you can either stay abreast with those changes, or go home and line your walls with tinfoil. I can promise you this, information is power, and you will soon see a change in the social classes based upon information. I refuse to be left behind based upon nameless fears and unfounded paranoia.
Netscape 6.0 is an abomination. It's a crime against humanity. Why should they feel obligated to bundle that monstrosity just because it's says Netscape on it? For my money, .90 is not there yet either which leaves linux with either Konquerer or Opra I guess. Konquerer is not all there so that leaves Opra. I don't know but it seems like a pretty sad state of affairs for the Linux camp. With the major importance of a browser you would think that the Mozilla folks would have tried to build a world class one first, then worry about the 200 applications they want to build into it. But instead they let the engineers run the show which will ultimatly be their downfall.
There is a lesson to be learned there for the OS community. It's not always about scratching an itch when you are trying to compete with commercial applications. You need some strategic planners who are more big picture oriented and can overcome the engineering nonsense. But if you try to suggest anything like that to the *community* you are labeled a troll. Oh well, I guess they can fail on their own and of their own accord.
What are you using your Ultra for? What tests did you do? Remeber the Ultra 5 is, as someone else already mentioned, pretty much a peeeeceeee with a sparc processor.
Linux does somethings faster WRT normal daily tasks, disk read/writes for example, but solaris on a sparc machine under load will completely fry linux. Solaris has the distinct advantage of being desinged specifically for the sparcs, this is very different from Linux which is pretty vanilla so as to work on many different platforms. Solaris is more like a big 18 wheeler compared to Linux which is like a pickup truck. Sure the pickup might get off the line faster but the 18 wheeler will go a lot farther with a much heavier load than the pickup
For the record, I have solaris 8 on an old Sparcstation 20 with a Ross Hyperspac 150 and it is still a very capable machine. I have thought about putting one of the BSD's on it for fun, put I have no trouble with Solaris 8's preformance.
At the risk of repeating an old cliche.. we need to work smarter, not harder. For instance I work about 45 hours a week and I make a very nice six figure salary. It's "the myth of incompetence" that the tech industry loves to propagate with regards to working 80 hour weeks. Maybe your working an 80 hour week becasue you are using the wrong tools or trying to implement the wrong solutions, etc...
I see it all the time though, some people will only use Windows because they are a "windows only shop" when it's obvious that Solaris or HP/UX would do the job with half the effort. I also see people jumping on the bandwagon and using languages like java or python when they should be using a RAD language. But hey, I make money because most people attack problems from the wrong direction so I guess I shouldn't bitch.
I can't wait for some of the Linux distros to start to fail. Yes, competition is a good thing, but the mainstream public doesn't view this as competition, they view it as fragmentation. Several of my clients have told me point blank that they refuse to even look at Linux because they feel the environment is too unstable for a long term commitment. Mind you, they didn't say the product was unstable, but rather the environment. That's the part of the equation that tends to get glossed over by the open source techy types, they tend to focus on the technical merits and they can't understand the need to align towards common goals and reduce any collaborative filtering which quite frankly ends up confusing John. Q. Consumer.
I will continue to recommend Linux as a low TOC and high ROI solution but it's an uphill battle. It's a shame that the "powers that be" won't listen when anyone mentions this line of thinking but I can tell you first hand that out in the real world it's the truth.