>
Instead of looking at this acquisition from a purely rational, coldly analytical perspective, we should and must begin to look at the quality of the lives of the employees.
Why analytical perspective is cold? Isn't the employee quality of life a rational issue? I praise your mindfulness, but you don't have to be wishy-washy. True rationality takes into account the Natural Law and its godly, human-care values.
>
Increasingly, with the influx of H-1B's and "free" trade, American companies are becoming the ruthless of ogres of the early part of the 20th century.
Free movement of persons and merchandise (that is, free immigration & trade) is good for poor people in poor countries, who can then sell their wares, their labour and the produce from it anywwhere instead of being in eternal dependence from foreign aid. It is also good for consumers, with cheaper, more abundant, better quality goods. It is not good for the lazy youngsters of Europe and North America, who think that they are entitled to a nine-to-five attitude. The only alternative would be to Europe and North America stop selling their luxury goods, patents and copy rights elsewhere.
The ruthlesness of the modern culture has nothing to do with free trade, but with decadence and greed. Free trade is conductive to the common good of humankind.
No, this will only disperse Oracle's energies. At this moment, Oracle should be focusing on creating a relational successor to its quasi-SQL engine, while upgrading the current offering to full SQL compliance. And it should have a credible free software story, not only "we run on Linux" -- they should run on the BSDs, on all platforms besides Intel, and be free software itself.
>
maybe, at the end of the day, customers will win because of the integration of two not-too-bad software suites.
Customers would be benefitted by a cross-platform, free software, focus on the product. Oracle won't offer any of these. PeopleSoft at least is focused, and isn't tied to a specific DBMS, much less a nonstandard one like Oracle.
>
"Play fair" means not dumping your inferior word processor
About dumping I had already agreed, but with the caveat the the way for Satellite would have been free software, not high prices. About MS Word being inferior, there are two sides: MS Word for Windows is indeed inferior to not only WordPerfect but also to Lotus Ami Pro and even MS Word for DOS; but at the time WordPerfect got beaten in the marketplace MS Word needed just a more reasonable amount of system resources. Not only WordPerfect cost more, it required expensive upgrades.
>
manipulating vendors and breaking your competitor's code on your OS. It would also mean not buying 10% of your competitor so you can keep it off alternate platforms, like Microsoft did right after Corel made Linux binaries.
And how this affected WordPerfect? All you say happened, but to other competitors.
>
Word Perfect 6.0 took up lots of space but was still easier to use and had more useful features than Word.
It didn't matter to most people that templates were much inferior to the former stylesheets, as few people did use them. I still miss that, but the fact is that people will take lower price and ease of use at any time over features and cross-platform portability.
>
if MS had played fair, none of this would have happened.
Would play fair in this case mean be late with a bloated version for MS Windows, and overpriced?
Because that was the case. The first versions of Satellite Software WordPerfect for MS Windows were confusing and bloated. There was one version requiring 6 MB RAM when the norm was 4 MB; MS WinWord at the time asked for only 2 MB. And MS WinWord was cheaper, especially if acquired in MS Office, which was also better integrated. Obviously MS could offer lower prices because of the monopoly, but Satellite could have gone the free software route; at a time, Borland even tried to sell a version of Emacs.
Do you have small scale potential to make for such a huge dam?
Please do your homework. Alternative sources are still (a) insufficient (b) unpredictable, unstable, unreliable (c) expensive. They are very nice for a rich country to burn unneeded money in, and perhaps to reduce fuel consumption, but they add nothing to the total potential in any given day of unfavorable climatic conditions.
These catastrophic scenarios may be realistic, given levels of corruption in the Far East. But they sound suspect to me, just as transgenicphobia and similar luddite talk. Sorry.
>
How can you steal something that's given away for free?
By not giving credit.
BSD in particular quite explicitly required credit, even in advertisements. But even if it never did so, copying withoug acknowledgement of authorship is a violation of copyright laws, besides being immoral.
I love Gecko, use it everyday all day long in Galeon.
But I don't want GtkHTML to be dumped. I want it to get feature parity with Gecko, so that I can dump Gecko and its portability layer.
I find it preposterous that I am running a program that needs emulation of proprietary COM in standard POSIX...
Valid point, already taken care of by the FSF
on
SCO SCO SCO!
·
· Score: 1
In SCO & Unix: a comedy of errors there is a valid point about Linux being in risk for lack of clear copyrights management.
This is an risk that the FSF has reckoned with from immemorial times by their copyright assignment policy. Linus shortsightedness disguised as pragmatism unfortunately prevented him from following it. Hopefully SCO claims will prove unfounded, but had the GNU project been more expedient with the Hurd or Linus been more careful, we wouldn't be running into such risks and FUD now.
Judging from the photo at the linked home page, people's tastes are degrading so much as to warrant my suspicion about us slipping into a new Middle Ages.
More. MS Windows NT 4 already could run very limited programs based on the most restricted section of POSIX. Interix then developed a much better system, effectively replacing the POSIX subsystem of MS WNT by a CygWin lookalike, including GNU GCC. MS bought it and renamed the product.
This is one of the funny parts of MS's FUD on the GNU GPL: they ship it themselves, and wouldn't have an useable product if it were not for the FSF. Actually MS could use it along the lines of, we had to pay this huge amount of cash to SCO because we have this GPL product with contaminated code; can you afford to? Not? Then commit to MS W32.
>
if they're CHRP, than doesn't that make the whole comment about not needing an Apple BIOS to run OS X become wrong?
Kinda.
>
Apple computers don't have a BIOS, they have OpenFirmware
Yes, but a BIOS isn't a BIOS anymore; I mean, no one uses it anymore as a Basic Input and Output System, as it was used by the original PC-DOS. Nowadays the word BIOS is used to describe just motherboard firmware, of which OpenFirmware is a standard.
>
the CHRP specification requires computers to boot using OpenFirmware. Sounds like it's probably using pretty much the same boot process to me.
Yes, your point being?
What can cause the original statement to be construed as misleading is that not needing Apple ROM is a feature of recent versions of Mac OS, not of the motherboard firmware.
an extremely stable OS, it's called Windows 2000, as long as you use MS certified drivers the OS should never crash
You mean as long as you only use MS certified drivers: any driver can compromise the stability of the whole system in monolithic designs like MS WNT or GNU/Linux.
This is a problem MS can't work around; it is inherent to monolithic kernels. And they simply are too closed to integrate driver development like Linux and the BSDs do. The long-term solution perhaps is something like GNU/Hurd with its multiple-server microkernel and translators, but meanwhile the Linux kernel does a pretty reasonable job at being open enough to get the best of the limitted development and debugging resources available. While MS may have deeper pockets, the aggregate of resources available to free software developers, and their efficiency at using them, enable a far better result.
>
there is a Flash plugin for Netscape-compatible browsers that is for Linux
Which being proprietary isn't available for my energy-efficient, silent Debian GNU/Linux PowerPC iBook.
Where did I last saw this attitude of it works for me even if it is proprietary and evil? Some free software desktop that catered only to C++ coders and used to require a proprietary library?
>
the NatSemi/Unisys machines have a paper trail, and I recalled an article indicating that a paper trail was part of Brazil's specification for the machines.
Part of the specification yes, but not widely deployed. The actual deployment of the ballot printer and container was something like 2-3%, not enough to warrant confidence in the election had it been contested.
>
some of the machines don't have a paper trail, and I stand corrected.
So it was incorrect to say they had without qualifying it, and it is incorrect to say some don't have. To be more precise, most don't have it.
Sorry about the nitpicking, but in security issues one must nitpick.
>
Neither Windows CE nor the previous OS were imposed by the government. Those were choices made by the bidding companies, based upon the requirements.
Requirements which didn't include the necessary code audit.
>
I also think your take on Unicamp and USP is too harsh. There are world-class experts in any one of those two and in many more places in Brazil.
Yes, there are. Yet the institutions are not trusted enough to serve as warrants of the whole process.
>
These machines was simple, using standard and cheaps technologies
MS Windows CE is anything but standard and cheap. Make it proprietary, non-standards-compliant, not-so-popular and expensive. The previous system was based on a proprietary VM DOS clone.
>
The paper trail was used on some machines on the last election, and as the only moving part of it, was the main defect source.
Well, that is exactly one of the advantages of mechanics: unlike software, it tend to fail badly enough to be noted by humans. Software can introduce or allow all kinds of errors without anyone noticing.
>
The only concern with these machine are that the rom code aren't public available, due to problems with vendor's copyrights, but are audited by well know and trusted Universities like Unicamp.
Far from being the only concern, since computer illiteracy and lack of audit trail were also big concerns. Also it was not the ROM code, but the whole system from firmware to application software, including MS WinCE in between, wasn't properly audited. Universities aren't trusted in Brazil, being underfunded and having generally low standards of achievement if compared to Europe, North America and Far East. And not "Universities", but several individuals representing Universities, companies and political parties were allowed only a few days in a clean room under NDA. IBM's representative even refused to sign the NDA, since it made the whole process innocuous.
Search CIPSGA for more information, if you can read Portuguese (or Spanish, it's similar enough.
>
National Semiconductor and Unisys (two American companies) made a really good electronic voting system for Brazil, they've been using them since 1996.
No, the original machines were made by Procomp. Please check your facts.
>
It has a tamper resistant paper trail, so it is completely auditable
No, the original machine had. In the last election only a small percentage of the machines had the paper trail. Please double-check all your facts.
>
people are happy with them
I am not. Source code was never properly audited, and next to no paper trail. I don't doubt the results, but had the election being more contested we'd have been in a full-scale fiasco. Had the situation won, the Left would still be complaining.
One of the more derided FSF's policies is copyright assignment, loathed even by some GNU developers themselves. For example, the XEmacs fork was born out of other controversies, but RMS himself explained it endures because of XEmacs not having been assigned the copyright to all code it included; putting unassigned code into GNU Emacs would open it to the same kind of attacks SCO is performing against the kernel Linux.
GNU Hurd for one is totally defensible from such attacks; while probably the *BSDs are vulnerable. It also speaks about the difference in attitude between Linus and the Open Source folks from RMS and the FSF guys that the copyright assignment requirement was imposed exactly to forestall such a thing as this attack from SCO, while the at the time seemingly more pragmatic attitude of accepting every contribution is now a source of a big headache, even if SCO's claims are found to be totally misguided.
I find it very interesting how everyone discusses things we can't quite evaluate and ignore what we can know and do something about.
For example, do the school has a program to eliminate paper? It is not necessarily a buzzword, especially if there isn't a printer nearby or if its use is costly.
And which notebooks? A PowerPC notebook (like an Apple) would use much less energy, and cost much less to manufacture, than an equivalent Intel, since die and battery sizes are bigger for the same given process and performance.
>
we're talking 8086 - 80286 CPUs with less than a megabyte of memory, correct?
Not quite. i80286 with up to 16MB DRAM is more like it. I knew some two or three years ago a guy who maintainted such systems in the port of Santos, SP, Brasil: he was considering upgrading it to GNU/Linux on a 80486 or something the like.
>
I personally would not buy an inferior performing Apple just to make a point. That's why I compared it to a rebel attitude in the first post. I think we have different end goals: you want a continuum moral equality in the corporate world, I just want to get stuff done.
Actually I would say that I have a long-range, global perspective while you look only at one aspect a the short term.
When looking raw performance you look only the mass-market; if you really wanted Performance and had the big bucks, you'd realise Intel doesn't scale; and if you go cluster, than RISC allows better density.
Which brings me to economical and environmental issues. CISC uses up more resources from manufacturing to disposal, and uses more energy, not only to run but to refrigerate.
And in the long range, RISC would enable us to get even higher performance, if Intel's monopoly is checked and therefore RISC gets the economies of scale it should.
Slashdot Mirror
Why analytical perspective is cold? Isn't the employee quality of life a rational issue? I praise your mindfulness, but you don't have to be wishy-washy. True rationality takes into account the Natural Law and its godly, human-care values.
Free movement of persons and merchandise (that is, free immigration & trade) is good for poor people in poor countries, who can then sell their wares, their labour and the produce from it anywwhere instead of being in eternal dependence from foreign aid. It is also good for consumers, with cheaper, more abundant, better quality goods. It is not good for the lazy youngsters of Europe and North America, who think that they are entitled to a nine-to-five attitude. The only alternative would be to Europe and North America stop selling their luxury goods, patents and copy rights elsewhere.
The ruthlesness of the modern culture has nothing to do with free trade, but with decadence and greed. Free trade is conductive to the common good of humankind.
No, this will only disperse Oracle's energies. At this moment, Oracle should be focusing on creating a relational successor to its quasi-SQL engine, while upgrading the current offering to full SQL compliance. And it should have a credible free software story, not only "we run on Linux" -- they should run on the BSDs, on all platforms besides Intel, and be free software itself.
Customers would be benefitted by a cross-platform, free software, focus on the product. Oracle won't offer any of these. PeopleSoft at least is focused, and isn't tied to a specific DBMS, much less a nonstandard one like Oracle.
About dumping I had already agreed, but with the caveat the the way for Satellite would have been free software, not high prices. About MS Word being inferior, there are two sides: MS Word for Windows is indeed inferior to not only WordPerfect but also to Lotus Ami Pro and even MS Word for DOS; but at the time WordPerfect got beaten in the marketplace MS Word needed just a more reasonable amount of system resources. Not only WordPerfect cost more, it required expensive upgrades.
And how this affected WordPerfect? All you say happened, but to other competitors.
Touché. All the previous versions you mention weren't true MS Windows versions, but more or less straightforward DOS ports. Until the respective versions 6, WordPerfect had the upper hand over MS Word. With MS Word 6 for Windows, MS had a cheaper product, which was easier to use, more consistent with the platform, and required less system resources. Not only that, but it could be even cheaper and better integrated if acquired in MS Office.
It didn't matter to most people that templates were much inferior to the former stylesheets, as few people did use them. I still miss that, but the fact is that people will take lower price and ease of use at any time over features and cross-platform portability.
Would play fair in this case mean be late with a bloated version for MS Windows, and overpriced?
Because that was the case. The first versions of Satellite Software WordPerfect for MS Windows were confusing and bloated. There was one version requiring 6 MB RAM when the norm was 4 MB; MS WinWord at the time asked for only 2 MB. And MS WinWord was cheaper, especially if acquired in MS Office, which was also better integrated. Obviously MS could offer lower prices because of the monopoly, but Satellite could have gone the free software route; at a time, Borland even tried to sell a version of Emacs.
What about nights and rainy days?
Unproven.
What about too windy or too quiet days?
Do you have small scale potential to make for such a huge dam?
Please do your homework. Alternative sources are still (a) insufficient (b) unpredictable, unstable, unreliable (c) expensive. They are very nice for a rich country to burn unneeded money in, and perhaps to reduce fuel consumption, but they add nothing to the total potential in any given day of unfavorable climatic conditions.
These catastrophic scenarios may be realistic, given levels of corruption in the Far East. But they sound suspect to me, just as transgenicphobia and similar luddite talk. Sorry.
By not giving credit.
BSD in particular quite explicitly required credit, even in advertisements. But even if it never did so, copying withoug acknowledgement of authorship is a violation of copyright laws, besides being immoral.
Nat says he'd dump GtkHTML for Gecko.
I love Gecko, use it everyday all day long in Galeon.
But I don't want GtkHTML to be dumped. I want it to get feature parity with Gecko, so that I can dump Gecko and its portability layer.
I find it preposterous that I am running a program that needs emulation of proprietary COM in standard POSIX...
In SCO & Unix: a comedy of errors there is a valid point about Linux being in risk for lack of clear copyrights management.
This is an risk that the FSF has reckoned with from immemorial times by their copyright assignment policy. Linus shortsightedness disguised as pragmatism unfortunately prevented him from following it. Hopefully SCO claims will prove unfounded, but had the GNU project been more expedient with the Hurd or Linus been more careful, we wouldn't be running into such risks and FUD now.
Does anyone have more pointers to this Grant, and knows if he has information on IBM comparable to what Bill Parish has on Microsoft?
Jim Gray is a SQL advocate. Obviously he'd promote data storage over processing. The sad thing is that he is among those that want to go with the database status quo instead of progressing towards relational nirvana; and without truly relational systems we will never be able of getting full use of all that data.
Only one is good, the rest is crap.
Judging from the photo at the linked home page, people's tastes are degrading so much as to warrant my suspicion about us slipping into a new Middle Ages.
More. MS Windows NT 4 already could run very limited programs based on the most restricted section of POSIX. Interix then developed a much better system, effectively replacing the POSIX subsystem of MS WNT by a CygWin lookalike, including GNU GCC. MS bought it and renamed the product.
This is one of the funny parts of MS's FUD on the GNU GPL: they ship it themselves, and wouldn't have an useable product if it were not for the FSF. Actually MS could use it along the lines of, we had to pay this huge amount of cash to SCO because we have this GPL product with contaminated code; can you afford to? Not? Then commit to MS W32.
Kinda.
Yes, but a BIOS isn't a BIOS anymore; I mean, no one uses it anymore as a Basic Input and Output System, as it was used by the original PC-DOS. Nowadays the word BIOS is used to describe just motherboard firmware, of which OpenFirmware is a standard.
Yes, your point being?
What can cause the original statement to be construed as misleading is that not needing Apple ROM is a feature of recent versions of Mac OS, not of the motherboard firmware.
Obviously no one can, motivation is intrinsic. What you could do is to encourage, help, give incentives etc.
Moderators, please -1 flamebait me. I was just nitpicking.
You mean as long as you only use MS certified drivers: any driver can compromise the stability of the whole system in monolithic designs like MS WNT or GNU/Linux.
This is a problem MS can't work around; it is inherent to monolithic kernels. And they simply are too closed to integrate driver development like Linux and the BSDs do. The long-term solution perhaps is something like GNU/Hurd with its multiple-server microkernel and translators, but meanwhile the Linux kernel does a pretty reasonable job at being open enough to get the best of the limitted development and debugging resources available. While MS may have deeper pockets, the aggregate of resources available to free software developers, and their efficiency at using them, enable a far better result.
Which being proprietary isn't available for my energy-efficient, silent Debian GNU/Linux PowerPC iBook.
Where did I last saw this attitude of it works for me even if it is proprietary and evil? Some free software desktop that catered only to C++ coders and used to require a proprietary library?
Part of the specification yes, but not widely deployed. The actual deployment of the ballot printer and container was something like 2-3%, not enough to warrant confidence in the election had it been contested.
So it was incorrect to say they had without qualifying it, and it is incorrect to say some don't have. To be more precise, most don't have it.
Sorry about the nitpicking, but in security issues one must nitpick.
And yes, now I can agree with you.
Requirements which didn't include the necessary code audit.
Yes, there are. Yet the institutions are not trusted enough to serve as warrants of the whole process.
MS Windows CE is anything but standard and cheap. Make it proprietary, non-standards-compliant, not-so-popular and expensive. The previous system was based on a proprietary VM DOS clone.
Well, that is exactly one of the advantages of mechanics: unlike software, it tend to fail badly enough to be noted by humans. Software can introduce or allow all kinds of errors without anyone noticing.
Far from being the only concern, since computer illiteracy and lack of audit trail were also big concerns. Also it was not the ROM code, but the whole system from firmware to application software, including MS WinCE in between, wasn't properly audited. Universities aren't trusted in Brazil, being underfunded and having generally low standards of achievement if compared to Europe, North America and Far East. And not "Universities", but several individuals representing Universities, companies and political parties were allowed only a few days in a clean room under NDA. IBM's representative even refused to sign the NDA, since it made the whole process innocuous.
Search CIPSGA for more information, if you can read Portuguese (or Spanish, it's similar enough.
No, the original machines were made by Procomp. Please check your facts.
No, the original machine had. In the last election only a small percentage of the machines had the paper trail. Please double-check all your facts.
I am not. Source code was never properly audited, and next to no paper trail. I don't doubt the results, but had the election being more contested we'd have been in a full-scale fiasco. Had the situation won, the Left would still be complaining.
Once more the FSF's prescience is vindicated.
One of the more derided FSF's policies is copyright assignment, loathed even by some GNU developers themselves. For example, the XEmacs fork was born out of other controversies, but RMS himself explained it endures because of XEmacs not having been assigned the copyright to all code it included; putting unassigned code into GNU Emacs would open it to the same kind of attacks SCO is performing against the kernel Linux.
GNU Hurd for one is totally defensible from such attacks; while probably the *BSDs are vulnerable. It also speaks about the difference in attitude between Linus and the Open Source folks from RMS and the FSF guys that the copyright assignment requirement was imposed exactly to forestall such a thing as this attack from SCO, while the at the time seemingly more pragmatic attitude of accepting every contribution is now a source of a big headache, even if SCO's claims are found to be totally misguided.
I find it very interesting how everyone discusses things we can't quite evaluate and ignore what we can know and do something about.
For example, do the school has a program to eliminate paper? It is not necessarily a buzzword, especially if there isn't a printer nearby or if its use is costly.
And which notebooks? A PowerPC notebook (like an Apple) would use much less energy, and cost much less to manufacture, than an equivalent Intel, since die and battery sizes are bigger for the same given process and performance.
Not quite. i80286 with up to 16MB DRAM is more like it. I knew some two or three years ago a guy who maintainted such systems in the port of Santos, SP, Brasil: he was considering upgrading it to GNU/Linux on a 80486 or something the like.
Actually I would say that I have a long-range, global perspective while you look only at one aspect a the short term.
When looking raw performance you look only the mass-market; if you really wanted Performance and had the big bucks, you'd realise Intel doesn't scale; and if you go cluster, than RISC allows better density.
Which brings me to economical and environmental issues. CISC uses up more resources from manufacturing to disposal, and uses more energy, not only to run but to refrigerate.
And in the long range, RISC would enable us to get even higher performance, if Intel's monopoly is checked and therefore RISC gets the economies of scale it should.