I seem to remember something from high school biology called the "10% law of energy transfer" - a living organism can only obtain 10% of the energy of its food supply through ingestion.
Although something about fusion power was also mentioned, somebody behind The Matrix wasn't paying too much attention in science class.
What if you took Red Hat Linux 7.2, removed the Linux kernel, and substituted the Solaris kernel?
On x86, you would immediately have 8-cpu scalability with no problem, with much greater efficiency than RH Advanced Server.
However, if you needed greater than 8-cpu scalability, porting Red Hat/Solaris to an e15k would give you a max of 106 cpus in a system architecture originally designed by Cray.
In other words, Sun could get control of the low end, provide a market-wide migration path to SPARC, and cut the throat of any high-end UNIX on Intel. They would instantly own the "Linux momentum."
The moment that Sun releases the Solaris kernel, or a portion of it that scales to 8 cpus (i.e., the "Free Solaris" standard version) in an open license, either BSD or GPL, they will kill Linux.
Apple is becomming a server threat. If Sun distributes a free Solaris kernel that is difficult to scale beyond 8 cpus that Apple takes up, Sun might be able to relegate Apple to the low end of the market, and give Apple vendors a high-end migration path. Apple is already rumored to be maintaining a SPARC port of Mac OS X and is dissatisfied with POWER.
AMD is adopting a NUMA architecture. If Sun works with AMD on a NUMA Solaris, design decisions there may provide Sun with new ideas for the SPARC line (for which they now will not report TPC-C scores, presumably because of shame).
Sun could also kill the Itanium UNIXen (HP-UX, AIX 5L, a future Tru64, and even OpenVMS) with a free Solaris kernel for ia64.
Sun must investigate the question of embedding Solaris technology in the products of other vendors (Apple and Red Hat immediately come to mind), preferably through an open license. The growth of Sun's server market will stagnate without such an envigorating act.
If we could wrap a scalable, sound, SMP-capable GPL kernel around Debian or Red Hat, would we think twice?
Or what if Sun were to release and maintain free Solaris for Itanium as well as x86? Would that be the kiss of death for HP-UX and AIX 5L? Why do they hesitate?
Granted, the Solaris kernel has weaknesses. UFS has to go. I hate/etc/system, I'd much rather tune on the fly with 2.4. patchchk is what up2date was several years ago. Sun's continued reliance on CDE/ksh/zip to get everything done really makes me ill. Solaris needs to be the UNIX of the 21st century.
What is the possibility of Sun convincing Apple to integrate large portions of Solaris into Mac OS X? Would they be willing to give it away to Apple? Why haven't they done so to build up market share?
I am a Sun stockholder. I would like to see Sun publicly considering these actions. I want to see some bombast from Steve and Bill. If Sun, Apple, and possibly AOL collaberate on an x86-os, they will kill Microsoft.
Sun needs to wake up to the potential of its own power. As it stands, they are difficult to distinguish from roadkill.
I want a JavaScript window on every page served to an IE browser saying something like "You are running a browser that is not standards-compliant with dangerous and insecure extensions. Please upgrade your browser to Mozilla, Netscape, Konqueror, Opera, or any other more secure solution."
Seriously, Oracle seems to get a lot of mileage with a single codebase for their "Universal Installer" and "Enterprise Manager" which are cross-platform Java Applets/Applications.
Apache should take a page from this book and silence this UI debate forever.
The only problem is that Apache is then condemned to including a JRE with their distribution forever, just like Oracle.
Or maybe we should just convince Oracle to do it for us?
I heard in a lecture given by Carl Sagan some years ago that the flatulence of cows was the #1 CO2 emission source - fossil fuel was a distance 2nd. Was this/is this true?
MS SQL Server and Sybase were one and the same until release 4.8. They share the same syntax and procedural-SQL extensions (Transact SQL).
Sybase is multi-platform, 64-bit capable, and very cheap (cheaper than MS, especially with the AS bundling). MS's only 64-bit port is to the Itanium, which is still a stranger to the datacenter. Tom Kyte of Oracle fame has criticisms for the Sybase/MS locking model, but the software is capable.
The Linux version of the Sybase 11.0.3.3 server is also available free of charge from linux.sybase.com - this is free for development and deployment. The Sybase 12 family is very inexpensive compared to Oracle and DB2.
The free 11.0.3.3 server was developed for Red Hat. It's SQL implementation is a bit dated (I don't even know if it is SQL92 entry-level-compliant, but a lot of SQL syntax from Postgres and MySQL doesn't work).
Why anybody uses MS SQL Server when a branch of the same code base is available for free boils down to one word: marketing.
If they were to get a hold of Openmail, they really might be able to slash MS right out of the server space - as long as they could keep MS from messing with the protocols.
AFAIK, Exchange is the number one reason to have MS anywhere near the datacenter.
When one of the DNS root servers switches to NT, please let me know - not that DNS is that stable or secure.
When IIS has a 60% market share (as Apache does now), I might also get a bit concerned.
When the Microsoft Sybase rip-off has a 46% market share (as Oracle currently has), we might start worrying about the datacenter.
When they have a stable, scalable 64-bit version of Windows, we might start worrying.
In order for Microsoft to get any of these markets, they will have to have a good product, good customer service, and good interoperability with other vendors products. I don't see that happening anytime soon.
After all, we gave them SMTP, and look what they did with that.
I mean, how hard do you have to work to convince a developer not to use gets() to parse an.ini file?
I wouldn't call this brainwashing. I remember reading an article about Oracle that they put the top 10 insecure things that you can do in C on a worksheet and they have every package maintainer sign off that these techniques have not been used. These are only touchstones, though, and security problems could easily be introduced while still using valid code.
Think of it more as a "security epiphany" or "security enlightenment" - they were probably just presented with a minimal list of what not to do. Hard to disagree about such things.
I read an article some time ago that Dolly the sheep had developed arthritis and was suffering from obesity, both of these conditions being extremely rare for her age.
This person that has been created may suffer from intense health problems. I consider this action to be extremely unwise, as it will play into the hands of extremists seeking a ban.
I personally would like to see cloning technology developed, but used on humans only when it is both safe and effective.
Here I am 3 hours from Chicago, and there are 3 inches of snow on the ground. I can't get anything resembling the truth from my favorite technical newsite, and our old SCO system in the accounting department finally died this morning.
A Beautiful Mind is a whitewash of a draft-dodging deadbeat dad who is occasionally of the alternate persuasion and gifted with only marginal scholastic ability when compared to his peers of the same time period. Yes, I found its exposition of schizophrenia to be extremely informative, but the events of his life were so thoroughly edited (i.e. his divorce and remarriage) that the movie has very little to do with the truth.
The Fellowship of the Ring was a tremendous gamble for the studio, and they won big. Their efforts at remaining truthful to the novels deserve preference to the pack of lies that was A Beautiful Mind.
The Academy Awards have very little to do with the quality of the motion pictures this year, or the esteem in which they are held by the movie-going public. In the depths of their political pandering they have become entirely irrelevant.
Of course, as a youth I really enjoyed Switched on Bach (I even have the double CD set of Switched on Brandenburgs in my car right now). It's amazing that she was a personal friend of the Moog family.
However, when I consider her later work (Digital Moonscapes, Beauty in the Beast, and even the Tron soundtrack), I see unrealized potential. I think that the recent "Switched on Bach 2000" is really sad from this perspective. My favorite piece that she actually composed herself was "Country Lane" from her Clockwork Orange soundtrack (I love the Dies Irea theme; I wish that she'd had more time to work on this score).
I don't know much about her decision to get a sex change, but sometimes I wonder what sort of an effect that had upon her musical output.
Still, AFAIK, Switched on Bach was the hottest-selling classical music album of all time...
Why does RH62 improve on all preceeding and following versions? Let me count the ways...
It has inetd
It does not have some bizarre mix of 2.2/2.4 kernel components
It does not require two separate compilers (kgcc)
It does not compile the filesystem driver as a kernel module
It runs Oracle 8.1.7 just fine
It has an up2date agent that is every bit as functional as the later versions
It does not implement stupid firewall rules out of the box
It is stable
Lots of people have said that 2.2 was Linux's "sweet spot" and this revision is great for servers - the only thing it lacks is JFS and large files. I use SGI's XFS shim for RH71 for that (should try 72 one of these days).
This was a very strong distribution; I dislike the current requirement that I build the RPMs myself, especially for a major problem like this.
Don't they prefer email these days?
on
SSSCA Hearing
·
· Score: 2
Anyway, I was plastering my bathroom last night; I don't want the FBI at my house over the residue.
Should I send this to my congressmen?
on
SSSCA Hearing
·
· Score: 5, Insightful
Gentlemen, I am writing you today to voice grave objections to Senator Fritz Hollings and his stance on digital rights management via his SSSCA legislation.
The U.S. economy is in the midst of an amazing ascent from recession, having endured more than any of us might have imagined that it could survive.
The digital information infrastructure is in no small part responsible for our recovery. The ability of businesses to easily and quickly exchange large amounts of information is key to our ever-increasing productivity.
This infrastructure is made possible by a number of software applications that are made available for free, and these applications are maintained by organizations that derive their profits by charging for support.
Let us take, for example, the "sendmail" application supported by:
http://www.sendmail.org
It is both unfair and unreasonable to require this company, who gives their product away for free, and plays no direct role in piracy on the Internet, to shoulder the overhead of implementing digital rights management.
Furthermore, "sendmail" is such a widespread product that E-Mail on the Internet would effectively end if all copies of "sendmail" were simultaneously disabled.
The SSSCA will take broad sectors of the IT market into violation of the law with the stroke of a pen. These sectors will include the entire free software movement, including one of my favorite companies, Red Hat Software (one of the most successful IPOs of 1999).
The damage to our information information infrastructure will be incalcuable should this legislation be enacted.
This legislation is a result of the lobbying efforts of the MPAA and the RIIA, who rightly desire some control over the perfusion of their digital content. Such content controls could easily be made voluntary by including a small message in an MP3 asking the user to purchase a legitimate copy of the work if a license key was not found in a local encyption key cache.
Instead, these media organizations wish to trample our constitutional protections on freedom of speech and fair use, and take the economy along with it.
Let there be no mistake; this legislation is a disaster. I urge you to vote against it.
Slashdot Mirror
I seem to remember something from high school biology called the "10% law of energy transfer" - a living organism can only obtain 10% of the energy of its food supply through ingestion.
Although something about fusion power was also mentioned, somebody behind The Matrix wasn't paying too much attention in science class.
What if you took Red Hat Linux 7.2, removed the Linux kernel, and substituted the Solaris kernel?
On x86, you would immediately have 8-cpu scalability with no problem, with much greater efficiency than RH Advanced Server.
However, if you needed greater than 8-cpu scalability, porting Red Hat/Solaris to an e15k would give you a max of 106 cpus in a system architecture originally designed by Cray.
In other words, Sun could get control of the low end, provide a market-wide migration path to SPARC, and cut the throat of any high-end UNIX on Intel. They would instantly own the "Linux momentum."
They need to do it now. Right now.
The moment that Sun releases the Solaris kernel, or a portion of it that scales to 8 cpus (i.e., the "Free Solaris" standard version) in an open license, either BSD or GPL, they will kill Linux.
Apple is becomming a server threat. If Sun distributes a free Solaris kernel that is difficult to scale beyond 8 cpus that Apple takes up, Sun might be able to relegate Apple to the low end of the market, and give Apple vendors a high-end migration path. Apple is already rumored to be maintaining a SPARC port of Mac OS X and is dissatisfied with POWER.
AMD is adopting a NUMA architecture. If Sun works with AMD on a NUMA Solaris, design decisions there may provide Sun with new ideas for the SPARC line (for which they now will not report TPC-C scores, presumably because of shame).
Sun could also kill the Itanium UNIXen (HP-UX, AIX 5L, a future Tru64, and even OpenVMS) with a free Solaris kernel for ia64.
Sun must investigate the question of embedding Solaris technology in the products of other vendors (Apple and Red Hat immediately come to mind), preferably through an open license. The growth of Sun's server market will stagnate without such an envigorating act.
Hilary, you want all the benefits of digital technologies and none of what you perceive to be the drawbacks.
A simple suggestion then. Go back to LPs and cassettes.
Please, put yourself out of our misery, you flat-earther.
A nice radio ad as outlined in the article would be great.
Would this make Linux irrelevant instantaneously?
If we could wrap a scalable, sound, SMP-capable GPL kernel around Debian or Red Hat, would we think twice?
Or what if Sun were to release and maintain free Solaris for Itanium as well as x86? Would that be the kiss of death for HP-UX and AIX 5L? Why do they hesitate?
Granted, the Solaris kernel has weaknesses. UFS has to go. I hate /etc/system, I'd much rather tune on the fly with 2.4. patchchk is what up2date was several years ago. Sun's continued reliance on CDE/ksh/zip to get everything done really makes me ill. Solaris needs to be the UNIX of the 21st century.
What is the possibility of Sun convincing Apple to integrate large portions of Solaris into Mac OS X? Would they be willing to give it away to Apple? Why haven't they done so to build up market share?
I am a Sun stockholder. I would like to see Sun publicly considering these actions. I want to see some bombast from Steve and Bill. If Sun, Apple, and possibly AOL collaberate on an x86-os, they will kill Microsoft.
Sun needs to wake up to the potential of its own power. As it stands, they are difficult to distinguish from roadkill.
There's too many free db engines out there to ignore, plus Oracle may hinder license transfer.
I want a JavaScript window on every page served to an IE browser saying something like "You are running a browser that is not standards-compliant with dangerous and insecure extensions. Please upgrade your browser to Mozilla, Netscape, Konqueror, Opera, or any other more secure solution."
I want this setting in httpd.conf.
Seriously, Oracle seems to get a lot of mileage with a single codebase for their "Universal Installer" and "Enterprise Manager" which are cross-platform Java Applets/Applications.
Apache should take a page from this book and silence this UI debate forever.
The only problem is that Apache is then condemned to including a JRE with their distribution forever, just like Oracle.
Or maybe we should just convince Oracle to do it for us?
I had no idea that methane was such an active greenhouse gas.
The KDE project is attempting to develop a version of KDE/Qt that does not require X.
They've been at this for awhile, and I don't know their status, but have you any thoughts on similar work?
I heard in a lecture given by Carl Sagan some years ago that the flatulence of cows was the #1 CO2 emission source - fossil fuel was a distance 2nd. Was this/is this true?
...and start agressively migrating MS SQL Server to RedHat Advanced Server/Sybase. It's a perfect fit.
MS SQL Server and Sybase were one and the same until release 4.8. They share the same syntax and procedural-SQL extensions (Transact SQL).
Sybase is multi-platform, 64-bit capable, and very cheap (cheaper than MS, especially with the AS bundling). MS's only 64-bit port is to the Itanium, which is still a stranger to the datacenter. Tom Kyte of Oracle fame has criticisms for the Sybase/MS locking model, but the software is capable.
The Linux version of the Sybase 11.0.3.3 server is also available free of charge from linux.sybase.com - this is free for development and deployment. The Sybase 12 family is very inexpensive compared to Oracle and DB2.
The free 11.0.3.3 server was developed for Red Hat. It's SQL implementation is a bit dated (I don't even know if it is SQL92 entry-level-compliant, but a lot of SQL syntax from Postgres and MySQL doesn't work).
Why anybody uses MS SQL Server when a branch of the same code base is available for free boils down to one word: marketing.
If they were to get a hold of Openmail, they really might be able to slash MS right out of the server space - as long as they could keep MS from messing with the protocols.
AFAIK, Exchange is the number one reason to have MS anywhere near the datacenter.
When one of the DNS root servers switches to NT, please let me know - not that DNS is that stable or secure.
When IIS has a 60% market share (as Apache does now), I might also get a bit concerned.
When the Microsoft Sybase rip-off has a 46% market share (as Oracle currently has), we might start worrying about the datacenter.
When they have a stable, scalable 64-bit version of Windows, we might start worrying.
In order for Microsoft to get any of these markets, they will have to have a good product, good customer service, and good interoperability with other vendors products. I don't see that happening anytime soon.
After all, we gave them SMTP, and look what they did with that.
I mean, how hard do you have to work to convince a developer not to use gets() to parse an .ini file?
I wouldn't call this brainwashing. I remember reading an article about Oracle that they put the top 10 insecure things that you can do in C on a worksheet and they have every package maintainer sign off that these techniques have not been used. These are only touchstones, though, and security problems could easily be introduced while still using valid code.
Think of it more as a "security epiphany" or "security enlightenment" - they were probably just presented with a minimal list of what not to do. Hard to disagree about such things.
I read an article some time ago that Dolly the sheep had developed arthritis and was suffering from obesity, both of these conditions being extremely rare for her age.
This person that has been created may suffer from intense health problems. I consider this action to be extremely unwise, as it will play into the hands of extremists seeking a ban.
I personally would like to see cloning technology developed, but used on humans only when it is both safe and effective.
Here I am 3 hours from Chicago, and there are 3 inches of snow on the ground. I can't get anything resembling the truth from my favorite technical newsite, and our old SCO system in the accounting department finally died this morning.
What a lousy day.
The Academy Awards have very little to do with the quality of the motion pictures this year, or the esteem in which they are held by the movie-going public. In the depths of their political pandering they have become entirely irrelevant.
Of course, as a youth I really enjoyed Switched on Bach (I even have the double CD set of Switched on Brandenburgs in my car right now). It's amazing that she was a personal friend of the Moog family.
However, when I consider her later work (Digital Moonscapes, Beauty in the Beast, and even the Tron soundtrack), I see unrealized potential. I think that the recent "Switched on Bach 2000" is really sad from this perspective. My favorite piece that she actually composed herself was "Country Lane" from her Clockwork Orange soundtrack (I love the Dies Irea theme; I wish that she'd had more time to work on this score).
I don't know much about her decision to get a sex change, but sometimes I wonder what sort of an effect that had upon her musical output.
Still, AFAIK, Switched on Bach was the hottest-selling classical music album of all time...
Why does RH62 improve on all preceeding and following versions? Let me count the ways...
Lots of people have said that 2.2 was Linux's "sweet spot" and this revision is great for servers - the only thing it lacks is JFS and large files. I use SGI's XFS shim for RH71 for that (should try 72 one of these days).
This was a very strong distribution; I dislike the current requirement that I build the RPMs myself, especially for a major problem like this.
Anyway, I was plastering my bathroom last night; I don't want the FBI at my house over the residue.
Gentlemen, I am writing you today to voice grave objections to Senator Fritz Hollings and his stance on digital rights management via his SSSCA legislation.
The U.S. economy is in the midst of an amazing ascent from recession, having endured more than any of us might have imagined that it could survive.
The digital information infrastructure is in no small part responsible for our recovery. The ability of businesses to easily and quickly exchange large amounts of information is key to our ever-increasing productivity.
This infrastructure is made possible by a number of software applications that are made available for free, and these applications are maintained by organizations that derive their profits by charging for support.
Let us take, for example, the "sendmail" application supported by:
http://www.sendmail.org
It is both unfair and unreasonable to require this company, who gives their product away for free, and plays no direct role in piracy on the Internet, to shoulder the overhead of implementing digital rights management.
Furthermore, "sendmail" is such a widespread product that E-Mail on the Internet would effectively end if all copies of "sendmail" were
simultaneously disabled.
The SSSCA will take broad sectors of the IT market into violation of the law with the stroke of a pen. These sectors will include the entire free software movement, including one of my favorite companies, Red Hat Software (one of the most successful IPOs of 1999).
The damage to our information information infrastructure will be incalcuable should this legislation be enacted.
This legislation is a result of the lobbying efforts of the MPAA and the RIIA, who rightly desire some control over the perfusion of their digital content. Such content controls could easily be made voluntary by including a small message in an MP3 asking the user to purchase a legitimate copy of the work if a license key was not found in a local encyption key cache.
Instead, these media organizations wish to trample our constitutional protections on freedom of speech and fair use, and take the economy along with it.
Let there be no mistake; this legislation is a disaster. I urge you to vote against it.