Well, frankly said, I do not really understand the fanaticism surrounding Linux. GNU is a cool
operating system, and a viable UNIX solution, but it has many weak points. One of them is
the Linux kernel.
What we WOULD need is modern, scalable software technology. What we HAVE is something
huge and monstrous thawed from a prehistorical slab of ice, and then decorated with bells
and whistles beyond recognition. Has anybody wondered why hardware comes with factory
drivers for Windows and MacOS, sometimes even for various *NIX flavors, but almost NEVER
for Linux? Do they know that Linux DOESN'T HAVE a unified interface for binary, user-space
drivers? It's not like the hardware manufacturers stroke some infernal deal with BSA to rid
the earth of Open Source... They simply don't want to give their technology out, and as
things stand now, they don't have the RIGHT by LAW to supply closed source drivers for
Linux - if we step beyond the all too obvious technical problems with this.
Also, have a look at WHAT exactly runs in kernel space? Well, more than half of it could as
well run in user space, providing more stability and scalability. Linux was already an outdated
example of proto-unices when it was a mere idea in the head of Linus. Now it's a dinosaur,
trampling great projects like HURD under its feet. "Everybody uses Linux, why work on
Sure, it has nuclear breath and can wreck Tokyo in less than an hour... But what if I want it
to fly? :D
I believe what the UNIX community and the platform itself needs is a new breed of unices,
based on new ideas but maintaining compatibility to old technologies. Things like TRUE i18n
with universal use of Unicode with conversion routines for backward compatibility,
microkernel architectures, a modern device namespace with compatibility fixes, a unified
hardware device driver interface put down in a core standard for POSIX systems, and so on.
MacOS/X is a great new step in this direction. I will be cheering for it, even from the back of
my Linux system. Probably this new wave will reach the Open Source community too, and
hopefully it won't turn its back...
Re: Japanese is proprietary
> You're just saying that because you
> speak japanese.
> The characters are way too complex, It's
> too dependant on
> being able to draw the characters by
> hand, so you can't
> use japanese with a static charset. And
> you said that there
> were differences between japanese and
> chinese. You only
> know that because you speak japanese.
> You know that
> language is very weird and not very
> ordered. You know
> japanese is an obsolete language. You
> just don't want to
> switch to a roman based language, and
> it's understandable
> that you would want to hang on to your
> native language,
> even though it's not effective for
Well, for one thing, I am not a native Japanese speaker, I am Hungarian. Our language does resemble Japanese a bit in terms of grammar, but uses the roman alphabet (the latin-2 charset), and has LOT more exceptions, 'stupid' and 'obsolete' rules than Japanese. I also speak English, and know that it ALSO has more of the formerly mentioned problems than Japanese. :)
Knowing that there is difference between Japanese and Chinese is not a secret, actually it is common knowledge on MY planet, totally obvious to any grade schooler... (And we're not even Asia...) But I admit, whether this is different in the USA, I know not... :D
Japanese CAN be used with a static charset, I have one here on my Linux system... The problem is NOT getting the computer to display Japanese. It already can. The problem is bringing this whole charset-madhouse to an end, and forget all the nightmare incidents of Kanji being displayed as rubbish, 'o doubleacute' as 'o tilde', and such... :)
The language is not weird, and IS ordered. I think, speaking both Japanese and English (and German and Hungarian as well, the latter as my native tongue), I can say this with certainty - while you, speaking apparently English only, have no basis to fight my arguments.
It's like starting to argue with an Australian on whether a Kangaroo has six legs or four... :D
Still, the roman alphabet does seem to be an optimal choice for inputting text into a computer. Programming languages, scripts, control codes should use that and no other. BUT being able to use different, more complicated charsets for personal text editing, desktop publishing, and such is a basic, reasonable and undenyable demand.
It is not the user who should be subservient to the developer, but the developer who should work to create functions the user wants. :)
Many users want to use different charsets, different languages on ONE computer. Without rebooting, restarting programs and fiddling with environmental variables.
Reasonable enough. It's time to realize it. You don't HAVE TO, but if you write self-contained, unflexible codes, please include a warning text: 'WARNING! Due to obsolete coding methodology, interoperability with third party software is uncertain.'
Talk about 'proprietary'... heh...