th ()
{
_ if [ -n "${DISPLAY}" ]; then
_ _ # Change X11 kbd layout to QWERTY somehow
_ else
_ _ loadkeys us
_ fi
}
and:
kj ()
{
_ if [ -n "${DISPLAY}" ]; then
_ _ # Change X11 kbd layout to Dvorak somehow
_ else
_ _ loadkeys dvorak
_ fi
}
(where the underscores above compensate for Slashdot's problematic lack of <PRE>). This is something i came up with on my own as well, back in 1996 when i switched. (loadkeys is a Linux-ism, and somewhat old; OpenBSD uses kbd and different layout names, while other BSDs do it yet differently. YMMV).
I found that it took me several weeks to learn (two-handed) Dvorak well enough to touch-type as fast as i could before with a QWERTY layout. Then it took me several more weeks to re-learn QWERTY, which i'm a little slower at now. Thenceforth, i've been able to switch between them by pausing a moment, then deliberately starting to type in the appropriate layout.
I rather like their hypothesis that people pay more attention to half a conversation than a full one and it seems it may be dead on.
Unfortunately, that's all Monk's study is at the moment, a hypothesis. Nowhere in the article does a standard error or standard deviation appear, only the assertion that "mobile-phone conversations are judged more negatively than loud conversations." The numbers are with 0.1 of each other! If the presented numbers turn out to be accurate within +/- 0.5, then Monk's statement is flat-out false.
There's not even a link to the raw data, nor to the description of the method used to calculate the numbers presented in the article. Are they arithmetic mean? Median? Mode? Geometric mean?
Until these sorts of actual facts are provided, this article is nothing more than figures massaged to support a hunch, and Monk may as well have dispensed with the numbers and merely written, "After conducting this study, it seems our hypothesis may be dead on".
Have you ever tried to extract a single file from a gzip'ed tar archive? It's not possible without unpacking everything and throwing away the bits that you don't want.
First, use your computer's multitasking capability to uncompress and extract a tarball archive member at the same time:
gzip -dc blah.tar.gz |pax -r blah/haha.txt
Or, for those without the POSIX utility:
gzip -dc blah.tar.gz |tar xf - blah/haha.txt
Second, if you don't want to have to uncompress the entire archive, use gzip and tar in the reverse order (though admittedly without the pipe):
gzip -9 blah/*
pax -w blah.tar blah
You extract and uncompress in the reverse order. You also fail to compress header data (much like with zipfiles), fail to take advantage of redundant data between archive members (same as with zipfiles) and can still extract uncorrupted members from an archive where a compressed member has been corrupted (much like with zipfiles).
There's bzip2 as well. POSIX pax and Jrg Schilling's star follow well-defined standards, and both are able to handle large (>2GB) files, as is bzip2. Last i knew, Info-Zip's zip and unzip showed their age and had troubles with files larger than 2GB on ILP32 platforms....
The GNOMEPro mockup looks even more like the NeXT file selector. Take "Favorite Locations" and put it horizontally on top to become the shelf. Use scroll bars instead of those useless scroll button things at either end. The clickable pathname elements are virtually identical to NeXT's scrollable pane with folder icons, only slightly less functional. And there's only one level of folder listing instead of the multiple levels of context that NeXT provides. But the essential elements are pretty much the same.
As other posters have mentioned, why reinvent what's already been designed well, only poorly?
(Tigert's mockup didn't come out well at all; the antialiased text has been color-quantized to the point where it's illegible. I expect more from someone with Tigert's reputation.)
CPM did not have support for hierarchal directory structure!
That's correct. CP/M did, however, have "user areas" numbered 0..7, which worked in effect like a limited number of directories in a single-level hierarchy. Unfortunately, no user area was visible from any other, except to pip.com (the "peripheral interchange program", sort of a combination of cp and cat). Even then, you could only use pip.com to transfer from another user area to the current one; to get pip.com into a different user area, you had to run pip.com, then use a debugger to save an image of the program.
It wasn't until ZCPR that you could get actual directories under CP/M.
[...] I discovered a Turkish company named Arcelik. Better still was the discovery that their parent company are named Koc Holding (google cache link). Those crazy Turks!
Unfortunately, those Turks are crazier than you think: the letter "c" in the modern Turkish writing system is pronounced like the "j" in the English words "jelly" and "jingoism". Imagine them spelled "Arjelik" and "Koj Holding".
Of course, if the companies are establish in countries where English is widely spoken, then those crazy Turks certainly managed to come up with names that aren't, ehm, culturally appropriate....
This coming from a company who 95% of computer users never heard, and who never even added the functionality of replying to emails even though it was really good until just recently, makes me believe his just looking for his two minutes of fame.
If you haven't heard of F-Prot, Fridrik Skulason, or Frisk Software International in connection with virus-scanning software, you've had your head some pretty nifty-smelling sand for over ten years. Back in 1992, F-Prot was one of the best virus-scanning software packages available for MS-DOS, and it was even freely downloadable over what was in those days called "the Internet".
Fridrik Skulason has actually had a bit more than two minutes of recognition (viz. this History of Computer Viruses); F-Prot has been continually available and updated since 1990---by my reckoning, that's more like thirteen years.
Beware of those who ascribe their own ignorance to 95% of everyone else.
It's about time - CDs have been overpriced for years.
What it's about time for is for CD albums to be between $5 and $10 US. I'm still aghast when i walk in and see one-disc albums on shelves for $18 or $21, when twelve-dollar CDs were normal back in 1990.
$12.98 isn't a price reduction, it's a miniscule token gesture, like giving customers a lollipop with their purchase.
"Tired of high CD prices? Buy your music from Universal Music Group, and you get to apply the lube yourself before you bend over!"
"t's like saying Honda's Cars are safer because American made cars had 300 accidents and Hondas 100. While it is true there are hundreds of thousands more Amierican made cars on the road the Hondas."
Actually, quite a number of Hondas are manufactured in North America... this is like saying: "Computers made by Dell are safer, because computers sold by Dell had fewer security problems reported than computers sold in the United States".
[WinCE] uses unicode for all internal character representation, so all apps that use standard functions should be
able to use all language features supported by unicode.
Using Unicode/ISO-10646 for internal character representation isn't nearly enough for reasonable i18n, let alone good multilingual support.
You still have to deal with rendering glyphs made from combining characters. You still have to deal with equivalences between such combined characters and precomposed characters. You still have to deal with right-to-left writing directions. You still have to deal with required ligatures in certain writing systems (e.g., Arabic or Devanagari).
In CJKV writing systems, you still have to deal with "language" tagging so that you know whether to represent a given "unified Han character" with a Japanese, Korean, Traditional, Simplified, or Vietnamese glyph (or know whether a given Hanzi/Kanji/Hanja/chuHan has code points for both simplified and traditional versions). You still have to deal with knowing which full-width punctuation characters should be shifted left or right and printed as half-width under which circumstances. You still need to know how to perform proper line breaking (CJKV text doesn't generally use spaces between words, but you can't just break a line between arbitrary characters).
You still need to deal with proper glyph rotation in vertical writing orientations---roman characters are usually rotated 90 degrees clockwise when printed vertically; likewise certain CJKV characters (for example, the hiragana/katakana lengthening mark U+30FC). Certain punctuation is also rotated in vertical orientation.
And that's just dealing with text printing or display! What about interoperability with other text encodings---like ISO-2022 or EUC, or funny locale-specific encodings like Shift-JIS---some of which are often used in things like, oh, email? How should you interpret character code 0x5c in EUC-JP or Shift-JIS encoded text? Is that an ASCII backslash, or a JIS-Roman narrow Yen sign? Do your Windows pathnames work when the directories are delimited by U+00A5 instead of U+005C? You prefer EUC-KR encoding instead? So is that a backslash or a Won sign?
I'm not even going to talk about fonts.
Good multilingual software is not only not easy to produce, it's damn hard, and Unicode is no panacaea.
the correct gender-neutral description in english is "he". by saying "she", the author is specifically writing about females (to the exclusion of males),
What century are you living in again? Were you asleep during all those Cultural Studies courses you were supposed to take in college?
Some speakers or writers use "she" instead of "he" as the singular gender-indeterminate pronoun, not to somehow make up for centuries (millennia?) of oppression and by the male-dominated power base, but rather to make the language slightly less female-exclusive.
It's becoming increasingly common in written American English (especially in academic usage) for writers either to use "she" exclusively or to alternate use of "she" and "he" for the singular gender-indeterminate pronoun.
In informal spoken American English, it's more common to use "they" as the singular gender-indeterminate pronoun.
On the other hand, if people would just suck it in and not gripe so much, I wonder how many of them
get just how easy it *is* to carry around a fair-sized laptop provided the carrying case is constructed
well. [...] Except that they probably already *have* a PDA, and only want the laptop in
addition for its gadget value.
You must not do any significant commuting via bicycle. I used to have a 7-pound Toshiba laptop. When i sold it and got a 3-pound Sony VAIO, i found that my bicycle commutes were much less of a drag (uh... pun not intended). Sure, i could have "sucked it in" and "not griped so much", but instead i've had a top-notch, lightweight, incredibly portable notebook machine for the past three-and-a-half years.
And now i lift weights at the gym, not the office.
Park said that he will seek to commercialize the next generation of his
protocol that he has been fine-tuning over the past year. Those
refinements, he said, make the protocol less theoretical and more
appropriate for real-world use.
Translation:
Park said that he has already filed an application for patents on this "technology" and any similar "technology" that looks like it might be useful to anyone. Current protocols already in widespread use may infringe on his pending patents, and he plans to form an intellectual property corporation in order to begin infringement lawsuits once the patents have been granted.
I think
certain combos of VMS and hardware allow true hotswapping, where the CPU can just be plucked out
with no loss to the OS.
Certain models of VAX hardware were fault-tolerant and had more than one CPU (usually three); the CPUs would "vote" by running the same code and comparing results. If two CPUs agreed and one didn't, the one that didn't was marked faulty, and a message was sent to the operator.
The future was ten years ago; now it's just in syndication.
[...] the writer(s) advised that one should "go limp, and roll into a fatal position". [emphasis mine]
Heh. When attacked by a bear, most all positions are fatal positions.
(Bonus: cf. the words of Ruby Rhod: ---"Mr. Rhod, you're going to have to assume your individual position." ---"I don't one one position! I want all positions!.")
However, if the event is at the Drafthouse Downtown, you may find a slightly greater likelihood at Polly Esther's Culture Club across the street. Remember to bathe, groom, and put on clean clothes.
Is that spitting hairs? I think so. I think that frankly, it's bullshit. Welcome to the USPTO.
If you're splitting hairs, it's not novel or innovative. IRC is non-patent prior art, it's extremely similar to AIM, and the move from automated client-based polling to server-based "push" notification is a rather obvious one to an expert in the field. ("The client polling requests are overloading the server. How can i reduce the load on the server? Oh---i'll just move the notification functionality from the client to the server. Bingo!")
IM is actually two independant services coupled together--messaging and notification. AOL's patent is specific to a system that couples these two services together. talk, this TERM-talk thing, IRC, zephyr all all allow for messaging, but none of them allow for presence notification.
So you mean that when, my ~/.ircrc file contains:
notify OoklaTheMok
and the IRC server notifies me when OoklaTheMok signs onto the server, that the server is lying, and OoklaTheMok really hasn't signed on?
And you mean that when gaim signs me onto the AIM server, it doesn't say anything to the server about ThundarrTheBarbarian being in my Buddy List, but the AIM server somehow magically knows to notify me when ThundarrTheBarbarian signs onto AIM?
Wow. So i guess ircII's notification facility that i was using back in 1995 wasn't really notification at all. You must know a lot more than i do about instant messaging....
Slashdot Mirror
I found that it took me several weeks to learn (two-handed) Dvorak well enough to touch-type as fast as i could before with a QWERTY layout. Then it took me several more weeks to re-learn QWERTY, which i'm a little slower at now. Thenceforth, i've been able to switch between them by pausing a moment, then deliberately starting to type in the appropriate layout.
For what it's worth....
[Score -1: Dyslexia]
172.16.0.0/12 is 172.16.0.0..172.31.255.255.
WhiteBox is obsoleted by CentOS. See this FAQ and answer.
Err ... something new to wine about, perhaps?
{Groans}
Translation: Jesuits are American. They tango grandly with eggs and practice golf.
Unfortunately, that's all Monk's study is at the moment, a hypothesis. Nowhere in the article does a standard error or standard deviation appear, only the assertion that "mobile-phone conversations are judged more negatively than loud conversations." The numbers are with 0.1 of each other! If the presented numbers turn out to be accurate within +/- 0.5, then Monk's statement is flat-out false.
There's not even a link to the raw data, nor to the description of the method used to calculate the numbers presented in the article. Are they arithmetic mean? Median? Mode? Geometric mean?
Until these sorts of actual facts are provided, this article is nothing more than figures massaged to support a hunch, and Monk may as well have dispensed with the numbers and merely written, "After conducting this study, it seems our hypothesis may be dead on".
First, use your computer's multitasking capability to uncompress and extract a tarball archive member at the same time:
Or, for those without the POSIX utility:
Second, if you don't want to have to uncompress the entire archive, use gzip and tar in the reverse order (though admittedly without the pipe):
You extract and uncompress in the reverse order. You also fail to compress header data (much like with zipfiles), fail to take advantage of redundant data between archive members (same as with zipfiles) and can still extract uncorrupted members from an archive where a compressed member has been corrupted (much like with zipfiles).
There's bzip2 as well. POSIX pax and Jrg Schilling's star follow well-defined standards, and both are able to handle large (>2GB) files, as is bzip2. Last i knew, Info-Zip's zip and unzip showed their age and had troubles with files larger than 2GB on ILP32 platforms....
The GNOMEPro mockup looks even more like the NeXT file selector. Take "Favorite Locations" and put it horizontally on top to become the shelf. Use scroll bars instead of those useless scroll button things at either end. The clickable pathname elements are virtually identical to NeXT's scrollable pane with folder icons, only slightly less functional. And there's only one level of folder listing instead of the multiple levels of context that NeXT provides. But the essential elements are pretty much the same.
As other posters have mentioned, why reinvent what's already been designed well, only poorly?
(Tigert's mockup didn't come out well at all; the antialiased text has been color-quantized to the point where it's illegible. I expect more from someone with Tigert's reputation.)
That's correct. CP/M did, however, have "user areas" numbered 0..7, which worked in effect like a limited number of directories in a single-level hierarchy. Unfortunately, no user area was visible from any other, except to pip.com (the "peripheral interchange program", sort of a combination of cp and cat). Even then, you could only use pip.com to transfer from another user area to the current one; to get pip.com into a different user area, you had to run pip.com, then use a debugger to save an image of the program.
It wasn't until ZCPR that you could get actual directories under CP/M.
Yee-haw.
No, never Z:. CP/M (at least, CP/M-80---did anyone ever actually use CP/M-86?) limited the available drives from A: to P:.
Unfortunately, those Turks are crazier than you think: the letter "c" in the modern Turkish writing system is pronounced like the "j" in the English words "jelly" and "jingoism". Imagine them spelled "Arjelik" and "Koj Holding".
Of course, if the companies are establish in countries where English is widely spoken, then those crazy Turks certainly managed to come up with names that aren't, ehm, culturally appropriate....
If you haven't heard of F-Prot, Fridrik Skulason, or Frisk Software International in connection with virus-scanning software, you've had your head some pretty nifty-smelling sand for over ten years. Back in 1992, F-Prot was one of the best virus-scanning software packages available for MS-DOS, and it was even freely downloadable over what was in those days called "the Internet".
Fridrik Skulason has actually had a bit more than two minutes of recognition (viz. this History of Computer Viruses); F-Prot has been continually available and updated since 1990---by my reckoning, that's more like thirteen years.
Beware of those who ascribe their own ignorance to 95% of everyone else.
What it's about time for is for CD albums to be between $5 and $10 US. I'm still aghast when i walk in and see one-disc albums on shelves for $18 or $21, when twelve-dollar CDs were normal back in 1990.
$12.98 isn't a price reduction, it's a miniscule token gesture, like giving customers a lollipop with their purchase.
"Tired of high CD prices? Buy your music from Universal Music Group, and you get to apply the lube yourself before you bend over!"
Actually, quite a number of Hondas are manufactured in North America ... this is like saying: "Computers made by Dell are safer, because computers sold by Dell had fewer security problems reported than computers sold in the United States".
Lutefisk is mostly Norwegian.
It is, however, possible to order lutefisk online. Also, there is apparently an annual lutefisk eating contest near Seattle, WA.
Using Unicode/ISO-10646 for internal character representation isn't nearly enough for reasonable i18n, let alone good multilingual support.
You still have to deal with rendering glyphs made from combining characters. You still have to deal with equivalences between such combined characters and precomposed characters. You still have to deal with right-to-left writing directions. You still have to deal with required ligatures in certain writing systems (e.g., Arabic or Devanagari).
In CJKV writing systems, you still have to deal with "language" tagging so that you know whether to represent a given "unified Han character" with a Japanese, Korean, Traditional, Simplified, or Vietnamese glyph (or know whether a given Hanzi/Kanji/Hanja/chuHan has code points for both simplified and traditional versions). You still have to deal with knowing which full-width punctuation characters should be shifted left or right and printed as half-width under which circumstances. You still need to know how to perform proper line breaking (CJKV text doesn't generally use spaces between words, but you can't just break a line between arbitrary characters). You still need to deal with proper glyph rotation in vertical writing orientations---roman characters are usually rotated 90 degrees clockwise when printed vertically; likewise certain CJKV characters (for example, the hiragana/katakana lengthening mark U+30FC). Certain punctuation is also rotated in vertical orientation.
And that's just dealing with text printing or display! What about interoperability with other text encodings---like ISO-2022 or EUC, or funny locale-specific encodings like Shift-JIS---some of which are often used in things like, oh, email? How should you interpret character code 0x5c in EUC-JP or Shift-JIS encoded text? Is that an ASCII backslash, or a JIS-Roman narrow Yen sign? Do your Windows pathnames work when the directories are delimited by U+00A5 instead of U+005C? You prefer EUC-KR encoding instead? So is that a backslash or a Won sign?
I'm not even going to talk about fonts.
Good multilingual software is not only not easy to produce, it's damn hard, and Unicode is no panacaea.
What century are you living in again? Were you asleep during all those Cultural Studies courses you were supposed to take in college?
Some speakers or writers use "she" instead of "he" as the singular gender-indeterminate pronoun, not to somehow make up for centuries (millennia?) of oppression and by the male-dominated power base, but rather to make the language slightly less female-exclusive.
It's becoming increasingly common in written American English (especially in academic usage) for writers either to use "she" exclusively or to alternate use of "she" and "he" for the singular gender-indeterminate pronoun.
In informal spoken American English, it's more common to use "they" as the singular gender-indeterminate pronoun.
Where have you been living?
You must not do any significant commuting via bicycle. I used to have a 7-pound Toshiba laptop. When i sold it and got a 3-pound Sony VAIO, i found that my bicycle commutes were much less of a drag (uh ... pun not intended). Sure, i could have "sucked it in" and "not griped so much", but instead i've had a top-notch, lightweight, incredibly portable notebook machine for the past three-and-a-half years.
And now i lift weights at the gym, not the office.
Translation:
Or something....
Certain models of VAX hardware were fault-tolerant and had more than one CPU (usually three); the CPUs would "vote" by running the same code and comparing results. If two CPUs agreed and one didn't, the one that didn't was marked faulty, and a message was sent to the operator.
The future was ten years ago; now it's just in syndication.
Urp. That would actually be:
Eye seam two knead uh spilling chucker.
Heh. When attacked by a bear, most all positions are fatal positions.
(Bonus: cf. the words of Ruby Rhod: ---"Mr. Rhod, you're going to have to assume your individual position." ---"I don't one one position! I want all positions!.")
Not at the event.
However, if the event is at the Drafthouse Downtown, you may find a slightly greater likelihood at Polly Esther's Culture Club across the street. Remember to bathe, groom, and put on clean clothes.
If you're splitting hairs, it's not novel or innovative. IRC is non-patent prior art, it's extremely similar to AIM, and the move from automated client-based polling to server-based "push" notification is a rather obvious one to an expert in the field. ("The client polling requests are overloading the server. How can i reduce the load on the server? Oh---i'll just move the notification functionality from the client to the server. Bingo!")
So you mean that when, my ~/.ircrc file contains:
and the IRC server notifies me when OoklaTheMok signs onto the server, that the server is lying, and OoklaTheMok really hasn't signed on?
And you mean that when gaim signs me onto the AIM server, it doesn't say anything to the server about ThundarrTheBarbarian being in my Buddy List, but the AIM server somehow magically knows to notify me when ThundarrTheBarbarian signs onto AIM?
Wow. So i guess ircII's notification facility that i was using back in 1995 wasn't really notification at all. You must know a lot more than i do about instant messaging....