IIRC - The original SunOS was a BSD Unix derivative. Solaris was built by combining Sys V and SunOS. Since then, Sun has hacked up (read improved) Solaris substantially. So at this point, I don't know how much of the original AT&T code still resides in Solaris, but I'd wager that there is not too much left. In the mean time, BSD derivatives have also improved over time.
It would take some investment in man-hours, but Sun might be able to comb through Solaris and rip out all the remaining AT&T Sys V stuff and replace it (as a place holder) with either *BSD code or some of their own re-writes. This would be a process similar to what BSDi/UC-Berkley had to do with 4.3/Net2 in order to reach the unencumbered 4.4BSD-Lite. Depending on how much of the old AT&T stuff still exists, this might be either a trivial or Herculean task.
Once that is done, Sun is left with an OS that contains BSD code along with its own Sun-originated Solaris code. At that point, they are free to license it as they see fit.
From looking at the x.org website, I get the impression that x.org is the official decendent of the orginal closed-source X11, while historically XFree86 was just a free implementation. However, it looks like X.org adopted XFree86 4.x as their own code base, which makes me wonder what happened to the old closed-source X11. Did the old X11 just die a silent death as everyone adopted the open source implementation?
I think Brown is somehow trying to imply that Linux is an illicit underground copy of Unix. In Soviet Russia (no this isn't a joke, really) and the East Bloc, banned books were distributed underground by copying (not only photocopying, but sometimes with a typewriter too). Samizdat could also properly be used to refer to the famous Jons Lions Book "A Commentary on the UNIX Operating System", which was driven undergound by AT&T because it contained so much UNIX source code. Salon has a story abou the Lions book and AT&Ts attempts to suppress it. In the old days, this book was secretely copied and distrubited by people who wanted to see how UNIX worked internally.
I wonder if it would be worthwhile to make a "port" which would just run a script to create a bunch of soft links and manpage additons of linuxisms linked to the BSD commands. That might help people transition (although some of the options might be profoundly different).
I agree with you and duff (above) that at the root this is a glibc issue. But, GNU being what it is, and Linux being what it is, I don't expect to see glibc/Linux development practices changing anytime soon. Because there are several Linux vendors with different release cycles, trying to get them all to use compatible libraries is impossible. The pratical fact remains that Suse version 9, Redhat AS version 3, and Debian version 3, can (and do) ship with different and incompatable version of glibc. So saying that a single binary is "for Linux" is a recipie for trouble.
However, I don't think it is unreasonable for Oracle to offer "Oracle for Suse 9" and "Oracle for Redhat AS 3.0" side by side, when they offer "Oracle for Solaris" and "Oracle for AIX" side by side.
I would be interested in knowing how much of Postgres has ended up in Oracle. Since Oracle is only in binary, we will probably never know (unless some Oracle employee gets mad because he didn't get his bonus).
Oracle needs to drop the "one Linux" fits all concept and to recompile against different (and up to date) distributions on a more frequent basis. Right now, Oracle for Linux is compiled against old versions of Suse with ancient glibc libraries. This causes its installation to fail on any modern distributation, unless you apply lots of compatibility patches and some ugly hacks to the configuation.
Because of glibc differences, saying there should be "one binary Oracle for all Linux" is like saying there should be one binary for all of Unix. Granted, the differnces betweeen Suse, Redhat, & Debian are not quite as drastic as the differences between Solaris, HP-UX and AIX, but fact remains that you can't install Oracle compiled against Suse 8 on Fedora without jumping through some major hacks.
Oracle needs to do frequent recompiles and offer different binaries for the various versions of Suse, Redhat AS, Fedora, Debian, and whoever else they decide to support.
Of the people with the 80 highest scores, five were among the Sept. 11 hijackers, Seisint's presentation said. Forty-five were identified as being or possibly being under existing investigations, while 30 others "were unknown to FBI."
This database looks like it could be one valid tool among many, but not conclusive evidence all by itself. So, it alone won't prove anybody guilty, but if some other independent reasons cast attention on someone and they showed up high on this list, then it would be prudent to take a deeper look at them. Doing otherwise would be negligent and could cost the lives of thousands of innocents, versus a fairly non-obtrusive background check or questioning of somebody with a suspicious history.
This was one of the driving reasons behind the federal requirement that TV stations broadcast in "digital" and "HDTV" by a certain date. The digital signals take up less bandwidth and the FCC knew all along that they needed more bandwidth for wireless phones. So once all the TV stations switch over, they will be required to surrender their old frequencies back to the government, who will re-allocate them to wireless/cell phones.
It looks like the was the missing piece that kept Evolution from connecting to OpenGroupWare. Now that looks to be solved. OGW provides Contacts and Calandar services (an can be integrated with a mail server like Cyrus to provide mail)
IIRC - This was the one of the "missing pieces" that OpenGroupware needs to make it an Exchange replacement. I think this is all Evolution needs to connect to the OpenGroupWare server.
Apparently though, Outlook itself will still require the "ZideLook" plug in to connect to the OGW server. But that is still a small price to pay in lieu of Exchange & MS-Win Server CALs.
Right - UnixWare Libraries possibly at issue
on
AutoZone Responds To SCO
·
· Score: 2, Insightful
IIRC - SCO thinks that AutoZone (in addition to using Linux generally) is using actual UnixWare OS libraries/ABIs to enable binary compatibility between its old UnixWare applications and its new Linux platform. They think that IBM (the vendor who migrated AutoZone to Linux) did the engineering to get AutoZone's UnixWare apps integrated to run on Linux by taking the short cut of actually stripping the libraries off the old UnixWare installation and putting them into the new Linux installation. If that is true, then it *really would* be a case of copyright infringement. But like most SCO claims, it is probably suspect.
It's my understanding that these "new" rules are modifications to the existing FLSA. Under to existing FLSA found at 29 USC Chapter 8 Section 213, people whose duties are "executive, administrative, and professional" are exempted from overtime pay. You can look this up yourself at www.law.cornell.edu
Here is a quote from Title 29 USC Chapter 8 Section 213 paragraph 17 on people exempt from overtime:
(17)
any employee who is a computer systems analyst, computer programmer, software engineer, or other similarly skilled worker, whose primary duty is -
(A) the application of systems analysis techniques and procedures, including consulting with users, to determine hardware, software, or system functional specifications;
(B) the design, development, documentation, analysis, creation, testing, or modification of computer systems or programs, including prototypes, based on and related to user or system design specifications;
(C) the design, documentation, testing, creation, or modification of computer programs related to machine operating systems; or
(D) a combination of duties described in subparagraphs (A), (B), and (C) the performance of which requires the same level of skills, and who, in the case of an employee who is compensated on an hourly basis, is compensated at a rate of not less than $27.63 an hour.
Not only did they treat their customer's like crap, but it sounds like the treated their employees with total contempt. That "buddy program" where the Indian followed an employee around so he could take his job has to be the most classic case of PHB leadership *EVER*. Message to employee - "We want you to do a great job for us, but we consider you expendable and beneath contempt and we don't even try to hide it. By the way.. meet Raji, he'll be taking your job in a few months. Have a enjoyable and productive day!".
No, it is not sick. It is progress. Humanity has been "genetically engineering" crops for many thousands of years. Previously we did it through selective breeding that took many tens of generations to get to the final product, but now we can do it in just a few generations in the lab. Corn is a grass. The orginal wild corn is not much different from the grass growing in your yard. Very small seed-kernels and a small yield. But the ancient Maya, Toltecs, Hopi, Azetcs and others "genetically" engineered it over many years into the Maize Corn that we eat today. It took them hundreds (perhaps thousands) of years to get a final product. We can do this over the course of a few years in a lab by isolating and encouraging (turning on) beneficial genes. Did you know that that thousands of human lives have been saved by the development of high-yield rice that simply involved encouraging it to grow with a short stalk? (short stalks do not sag and rot in the water).
Now I will agree that there are two kinds of genetic engineering and that one suits me fine and the other I find disturbing. I have no problem with isolating and encouraging naturally occuring recessive traits. I do have concerns about "transgenic" engineering that splices genes from another species (especially when we are talking about animals instead of plants).
the need for specific add-ons that are available as part of packages but are available when rolling-my-own.
My criteria for package vs source is what the program is supposed to do - i.e. a "system package" vs application package. Pre-built packages are a *best guess* of what options the package maintainer thinks you might want included. But that guess is not always right for me. So I use the binary packages from the distrubution vendor for updates to the base system, but for the production apps (such as apache, php, MySQL/PostgreSQL, mail servers, etc), I use source compiles because it gives me finer grain control of the included options. The gives me the best of both worlds. I get the convienence of quick system and security updates with little hassle, while I also get to have maximum control over the applications that really matter productivity.
We can just locate all the credit-card customer call centers in Nigeria. Then they won't to send us those "special" deals via e-mail in order to get our credit card numbers!
If I remember correctly, Mozilla.org was having some heartburn over unofficial builds using the Mozilla name because of Trademark issues (yes, even open source projects can be A-holes when it comes to "stealing" thier stuff). From the mozilla website:
Our code is free - but we do strictly enforce our trademark rights. Our trademarks include, among others, the names Mozilla, Bugzilla and XUL, the Firefox name and logo, the Mozilla text logo and the red lizard logo. This means that, while you have considerable freedom to modify and redistribute our software, there are tight restrictions on your ability to use the Mozilla name and logos, even when built into binaries that we provide.
They go on to say that un-modified versions can be redistributed with the mozilla branding, but "all other uses of the Mozilla marks require prior written permission." So, if you are doing any kind of browswer hacking, changing the name/logos becomes more or less mandatory in order to stay on Mozilla.org's good side.
-- Caldera started by disgruntled Novell employees --MS finances Caldera/SCO to sue IBM --IBM induces Novell to register UNIX copyrights (after waiting 10 years to get around to it). --Novell Buys SuSe --SCO sues Novell --Now IBM pours money into Novell
I know Novell is a company with a glorious past, but, just as SCO is a MS puppet/proxy, I think Novell is getting pulled into being a minor sattelite orbiting IBM.
So, why can't they come to an agreement to pay King a few pennies out of every sale? Everyone wins, Mandrake-Linux keeps its good name and King makes a buck or two.
In business, its all about money. (More on this later).
Can MS-Office be ported to Linux technically? I would say yes, because they were able to make a Mac OS X port, which has BSD-Unix underpinnings. Pretty much anything than can be done on BSD can be done on Linux. So no great feat of technology would be involved on getting MS-Office ported to Linux.
Now lets talk about why MS would or would not want to do this. If enough of a market existed (read: Corporate customers clamoring for a native Linux port), MS might have an opportunity to retain those customers (and maybe get a few new customers) and make some money doing it. So there is an opportunity for them there in the office suite market. The danger is this: MS-Office & MS-Windows are mutually supporting monopolies in the corporate world. . As long as Office effectively requires Windows, every corporate desktop sold with Office almost guarantees an accompanying windows license. So double the revenue for M$. A native Linux version of MS-Office would undermine Windows. Once Windows is undermined, then Office itself might be jeopardized because they are mutually supporting.
A native Linux port of MS-Office is just too much of a threat to the MS monopoly structure. MS knows this, so such a port will never see the light of day.
1 : a branch of metaphysics concerned with the nature and relations of being 2 : a particular theory about the nature of being or the kinds of existents.
So, WTF does that mean and what does it have to do with information?
Slashdot Mirror
IIRC - The original SunOS was a BSD Unix derivative. Solaris was built by combining Sys V and SunOS. Since then, Sun has hacked up (read improved) Solaris substantially. So at this point, I don't know how much of the original AT&T code still resides in Solaris, but I'd wager that there is not too much left. In the mean time, BSD derivatives have also improved over time.
It would take some investment in man-hours, but Sun might be able to comb through Solaris and rip out all the remaining AT&T Sys V stuff and replace it (as a place holder) with either *BSD code or some of their own re-writes. This would be a process similar to what BSDi/UC-Berkley had to do with 4.3/Net2 in order to reach the unencumbered 4.4BSD-Lite. Depending on how much of the old AT&T stuff still exists, this might be either a trivial or Herculean task.
Once that is done, Sun is left with an OS that contains BSD code along with its own Sun-originated Solaris code. At that point, they are free to license it as they see fit.
From looking at the x.org website, I get the impression that x.org is the official decendent of the orginal closed-source X11, while historically XFree86 was just a free implementation. However, it looks like X.org adopted XFree86 4.x as their own code base, which makes me wonder what happened to the old closed-source X11. Did the old X11 just die a silent death as everyone adopted the open source implementation?
I think Brown is somehow trying to imply that Linux is an illicit underground copy of Unix. In Soviet Russia (no this isn't a joke, really) and the East Bloc, banned books were distributed underground by copying (not only photocopying, but sometimes with a typewriter too). Samizdat could also properly be used to refer to the famous Jons Lions Book "A Commentary on the UNIX Operating System", which was driven undergound by AT&T because it contained so much UNIX source code. Salon has a story abou the Lions book and AT&Ts attempts to suppress it. In the old days, this book was secretely copied and distrubited by people who wanted to see how UNIX worked internally.
I wonder if it would be worthwhile to make a "port" which would just run a script to create a bunch of soft links and manpage additons of linuxisms linked to the BSD commands. That might help people transition (although some of the options might be profoundly different).
I agree with you and duff (above) that at the root this is a glibc issue. But, GNU being what it is, and Linux being what it is, I don't expect to see glibc/Linux development practices changing anytime soon. Because there are several Linux vendors with different release cycles, trying to get them all to use compatible libraries is impossible. The pratical fact remains that Suse version 9, Redhat AS version 3, and Debian version 3, can (and do) ship with different and incompatable version of glibc. So saying that a single binary is "for Linux" is a recipie for trouble.
However, I don't think it is unreasonable for Oracle to offer "Oracle for Suse 9" and "Oracle for Redhat AS 3.0" side by side, when they offer "Oracle for Solaris" and "Oracle for AIX" side by side.
I would be interested in knowing how much of Postgres has ended up in Oracle. Since Oracle is only in binary, we will probably never know (unless some Oracle employee gets mad because he didn't get his bonus).
Oracle needs to drop the "one Linux" fits all concept and to recompile against different (and up to date) distributions on a more frequent basis. Right now, Oracle for Linux is compiled against old versions of Suse with ancient glibc libraries. This causes its installation to fail on any modern distributation, unless you apply lots of compatibility patches and some ugly hacks to the configuation.
Because of glibc differences, saying there should be "one binary Oracle for all Linux" is like saying there should be one binary for all of Unix. Granted, the differnces betweeen Suse, Redhat, & Debian are not quite as drastic as the differences between Solaris, HP-UX and AIX, but fact remains that you can't install Oracle compiled against Suse 8 on Fedora without jumping through some major hacks.
Oracle needs to do frequent recompiles and offer different binaries for the various versions of Suse, Redhat AS, Fedora, Debian, and whoever else they decide to support.
This database looks like it could be one valid tool among many, but not conclusive evidence all by itself. So, it alone won't prove anybody guilty, but if some other independent reasons cast attention on someone and they showed up high on this list, then it would be prudent to take a deeper look at them. Doing otherwise would be negligent and could cost the lives of thousands of innocents, versus a fairly non-obtrusive background check or questioning of somebody with a suspicious history.
Absolutely Right. It can't be very "green" when it is ultimately powered by a smog-belching coal-fired powerplant.
This was one of the driving reasons behind the federal requirement that TV stations broadcast in "digital" and "HDTV" by a certain date. The digital signals take up less bandwidth and the FCC knew all along that they needed more bandwidth for wireless phones. So once all the TV stations switch over, they will be required to surrender their old frequencies back to the government, who will re-allocate them to wireless/cell phones.
It looks like the was the missing piece that kept Evolution from connecting to OpenGroupWare. Now that looks to be solved. OGW provides Contacts and Calandar services (an can be integrated with a mail server like Cyrus to provide mail)
IIRC - This was the one of the "missing pieces" that OpenGroupware needs to make it an Exchange replacement. I think this is all Evolution needs to connect to the OpenGroupWare server.
Apparently though, Outlook itself will still require the "ZideLook" plug in to connect to the OGW server. But that is still a small price to pay in lieu of Exchange & MS-Win Server CALs.
IIRC - SCO thinks that AutoZone (in addition to using Linux generally) is using actual UnixWare OS libraries/ABIs to enable binary compatibility between its old UnixWare applications and its new Linux platform. They think that IBM (the vendor who migrated AutoZone to Linux) did the engineering to get AutoZone's UnixWare apps integrated to run on Linux by taking the short cut of actually stripping the libraries off the old UnixWare installation and putting them into the new Linux installation. If that is true, then it *really would* be a case of copyright infringement. But like most SCO claims, it is probably suspect.
Here is a quote from Title 29 USC Chapter 8 Section 213 paragraph 17 on people exempt from overtime:
Not only did they treat their customer's like crap, but it sounds like the treated their employees with total contempt. That "buddy program" where the Indian followed an employee around so he could take his job has to be the most classic case of PHB leadership *EVER*. Message to employee - "We want you to do a great job for us, but we consider you expendable and beneath contempt and we don't even try to hide it. By the way.. meet Raji, he'll be taking your job in a few months. Have a enjoyable and productive day!".
No, it is not sick. It is progress. Humanity has been "genetically engineering" crops for many thousands of years. Previously we did it through selective breeding that took many tens of generations to get to the final product, but now we can do it in just a few generations in the lab. Corn is a grass. The orginal wild corn is not much different from the grass growing in your yard. Very small seed-kernels and a small yield. But the ancient Maya, Toltecs, Hopi, Azetcs and others "genetically" engineered it over many years into the Maize Corn that we eat today. It took them hundreds (perhaps thousands) of years to get a final product. We can do this over the course of a few years in a lab by isolating and encouraging (turning on) beneficial genes. Did you know that that thousands of human lives have been saved by the development of high-yield rice that simply involved encouraging it to grow with a short stalk? (short stalks do not sag and rot in the water).
Now I will agree that there are two kinds of genetic engineering and that one suits me fine and the other I find disturbing. I have no problem with isolating and encouraging naturally occuring recessive traits. I do have concerns about "transgenic" engineering that splices genes from another species (especially when we are talking about animals instead of plants).
We can just locate all the credit-card customer call centers in Nigeria. Then they won't to send us those "special" deals via e-mail in order to get our credit card numbers!
They go on to say that un-modified versions can be redistributed with the mozilla branding, but "all other uses of the Mozilla marks require prior written permission." So, if you are doing any kind of browswer hacking, changing the name/logos becomes more or less mandatory in order to stay on Mozilla.org's good side.
-- Caldera started by disgruntled Novell employees
--MS finances Caldera/SCO to sue IBM
--IBM induces Novell to register UNIX copyrights (after waiting 10 years to get around to it).
--Novell Buys SuSe
--SCO sues Novell
--Now IBM pours money into Novell
I know Novell is a company with a glorious past, but, just as SCO is a MS puppet/proxy, I think Novell is getting pulled into being a minor sattelite orbiting IBM.
Man, this is better than a soap opera !
because you have to when it will invaribly lock up
So, why can't they come to an agreement to pay King a few pennies out of every sale? Everyone wins, Mandrake-Linux keeps its good name and King makes a buck or two.
In business, its all about money. (More on this later).
Can MS-Office be ported to Linux technically? I would say yes, because they were able to make a Mac OS X port, which has BSD-Unix underpinnings. Pretty much anything than can be done on BSD can be done on Linux. So no great feat of technology would be involved on getting MS-Office ported to Linux.
Now lets talk about why MS would or would not want to do this. If enough of a market existed (read: Corporate customers clamoring for a native Linux port), MS might have an opportunity to retain those customers (and maybe get a few new customers) and make some money doing it. So there is an opportunity for them there in the office suite market. The danger is this: MS-Office & MS-Windows are mutually supporting monopolies in the corporate world. . As long as Office effectively requires Windows, every corporate desktop sold with Office almost guarantees an accompanying windows license. So double the revenue for M$. A native Linux version of MS-Office would undermine Windows. Once Windows is undermined, then Office itself might be jeopardized because they are mutually supporting.
A native Linux port of MS-Office is just too much of a threat to the MS monopoly structure. MS knows this, so such a port will never see the light of day.
My handy dictionary defines Ontology as:
1 : a branch of metaphysics concerned with the nature and relations of being
2 : a particular theory about the nature of being or the kinds of existents.
So, WTF does that mean and what does it have to do with information?
lynx.browser.org. I recommend it's nice JS implementation by the way.