That word was unknown before the LCD era. It's a strictly LCD-related problem and still serious though great advances have been made......unless you think opposite and consider it an advantage that nobody looking at your screen from a side would see a thing. CRT has full 180 degrees viewing angle. So does Plasma. LCDs are the ones with problems in this area.
What were the alternatives from Microsoft at the time of creation of Perl? Isn't PHP a "favoured default" for Apache, backed by the Apache Foundation? What was the state of development of ASP when PHP has emerged from Perl? And it's not Java. It SUN/Java.
I can say that sentence at about 200WPM without rushing. Can you type it that fast? I'm a bad typist but I can write 100-200K in a single nightly session. Can you say that much without hurting your vocal strings? And for a skilled typist, 200WPM isn't all that much.
If you put that into command-line-ese, you'd get almost exactly what you typed
No. I'd get "copy from drive c windows system32 file kernel.dll to drive c" And that's the point - command line is still much more efficient than GUI, no matter if you use voice or keyboard (or even mouse for GUI) - so command line won't die even after keyboard dies. I just presented two ways of performing a task using voice commands, one in command line style, the other in GUI style. In this we agree...
Typing for a skilled typist is both much faster and easier than speaking. Especially for prolonged periods of time.
There's a rather long while till we get reliable thought-controlled interfaces, and even then they may need some extra training. How many more thoughts you create than you actually say/type? You cull most of the plans before they take any serious shape and are converted into words. Untrained thought-writing would be just a feast of spurts of senseless text and undoing them.
Reliable voice recognition may replace the keyboard in some cases, but even then it won't necessarily replace the command line. It's easier to form a command sentence than to call options being displayed on GUI.
say "changedir/usr/src/linux; copy vmlinuz/boot" versus "my computer, open. drive C, open. Windows, open. System32, open. kernel32.dll copy. parent. parent. parent. paste."
We're not talking about "generic software" or some "pet desktop app". We're talking about a new programming language that has the intention of replacing one that is strong on the market already and is meant for pretty specific usage - mostly in areas where the last word belongs to suits, where multi-million-dollar transactions are made and so on. If your failure/success cost is difference between earning and losing a million dollars, will you pick a mostly untested in such an environment, written largely by amateurs and almost unknown language, possibly very good but of unknown risks, or go with a solution where all the weaknesses are known, it was tested hundreds of times in such conditions, and methods to avoid the bugs and weaknesses are all known to professionals you hire? We're not talking about nerd's or his grandmother's computer. We're talking of real, big and damn expensive production environment.
Apache wouldn't be near anywhere to what it is now without support of the giants behind the Apache Foundation. Firefox gains recognition because millions of home users are interested in it. PHP rode on the shoulders of Apache and Perl. Linux has enough attention to be trustworthy. In the world where decreasing risk by 0.1% means $10.000 profit, picking untested alternatives is out of question.
ASP 1.0 - end of 1996. JSP - Sun's response to ASP. PHP/FI 1995, PHP3 - 1997.
And of course dubious fame of ISS which was the only supported platform for ASP until recently and similarity of PHP to Perl which was the language of CGI for ages.
'cause OO is way easier for team writing and huge projects. It's way easier to split the project into many "single man" or "single small team" tasks, then bind them all together through an easy to use and strictly defined methods with well defined "responsiblity" areas. The difference isn't all that big, except of some "protectionism" (private, public), simplification of some processes (inheritance instead of notorious evil "copy&paste") and strict defining of "responsiblity areas" (objects), instead of guessing whether convert_hostname_to_lowercase() belongs to hostnameconv.h or tolowercase.h:)
I don't believe any open source solution in any near future could crush the Microsoft alternatives in the software development field. The problem is that HERE marketing matters. Home users are free to pick a web browser or operating system of their choice. But when a big system for some business/industry is being developed, the platform decisione are made by the middle-to-upper management. And these guys really -believe- what Microsoft marketing people tell them. So the programmers, people who actually know a thing about the options don't really get the voice in most of the projects. "So... This guy at EXPO told me Visual Basic would solve all these problems. So we write the application in Visual Basic." There is no way the majority of the "big fishes" in programming could accept a hardly known free software language instead of the "famous, widely used Microsoft product" without the right marketing, and without some large funding behind the marketing...
Unless Sun, IBM or someone else with enough $$$ and not too much love for Microsoft backs up the project and takes care of marketing and promoting it. But the chances are very slim.
I didn't get the article but I think that's not it. Add an entry "mac" and entry "macintosh" and point both to Apple and you have the synonyms problem solved. Many words to describe the same thing, multiple entries describing the same page. Worse about homonyms, where one word has several meanings. Wikipedia solves that by "disambiguation" pages.
I think we should think of a manned mission... But not "bring them there, pick samples, bring them back". But of a real base...
Send another 50 or so robotic missions. With robot "builders", "planters", machinery etc. Remotely build a base, whole with a launch pad, enough supplies and equipment to be self-sufficient for some 5 years, and with enough labs and assembly line that new robots delivered there in parts can be assembled and deployed locally. Then send a crew that will have specific tasks: - control the robots locally to avoid the delay - repair broken robots - communicate and report in wider range than robots can. - gradually assemble a rocket to get back home, from parts delivered in the meantime:)
By the way, a lot of this stuff could be made from Mars orbit. I think we could profit a LOT from a manned martian orbiting base, just like the Alpha station. A reusable shuttle or two for landing to perform the land-based maintenance, and the life area orbiting around Mars. Note getting between Mars orbit and Earth orbit takes vastly less resources than getting from surface to orbit of any of the planets...
Yes, send humans to Mars. But don't rush it. Take enough time to make it worthwhile, and build enough robotic infrastructure to give the astronauts actual job for which they would be necessary, not just the PR thing.
we can do nothing to prevent, only to try to mitigate.
or send a warning early enough. With 15-minute warning (easy for science/engineering, harder for bureaucracy/business) the casualities could have been halved
..amongst many others:) There were some challenges to crack (brute force) some of "better" encryption schemes to estimate -real- computational power to crack them. MD5, SHA, etc. I don't know about others but I remember MD5 took about a year with several thousands computers cracking it. Of course that wasn't a "general" solution, just decrypting a single encrypted message. But it was proven it's possible.
Ah, and there are quite a few maths problems where some new solutions were found.
Barcodes are very handy but way too easy to copy. You don't need special software. You don't need a computer to copy one. You can xero a barcode, and with enough skill, recreate one using pen and paper. Then, you don't need to go even there! Go, buy Half-Life 1 from bargain bin, then remove the sticker with barcode, slap it on HalfLife 2 over the original one...
If I want to contribute to the community, I release Open Source. But if some commercial bastard pisses me off and releases their "free" proprietary program with some essential functionality missing "for strategical reasons", I release the alternative as GNU GPL'd Free Software - screw you, you want my features in your junk, re-release as GNU or write them yourself!
No. They didn't circumvent any mechanisms protecting copyrighted data in order to use that data. (and this is strictly what DMCA is about) You could say they circumvented the protection (doubtful, the protection wasn't anywhere near to "efficient" as DMCA states) to access the copyrighted firmware. Except their aim is not to steal the original firmware but to replace it with their own, so the intent part isn't fulfilled at all. If they downloaded the firmware and started spreading it over BitTorrent, sure, then they are in violation of DMCA. But if they just make a backup for personal use and then write new software, sorry, nope. Sure they could be SUED under DMCA. But they would win the case hands down. Even if they were spreading original but -modified- (not written from scratch) firmware, a good lawyer could argue it's fan art and as such, fair use, but that's more tricky.
Resident sniffer/logger. Simple Firewall. Monitor, blinking LEDs on certain kinds of packets arriving. "Wake on ring" if not present by default. "extra secret storage" in unused flash. Changing MAC address... *less* bandwidth (throttling your uplink, etc)
Imagine a beowulf cluster of these! With nodes changing location within the cluster according to some obscure formula... and killing unsuspecting operators:)
Debian Woody (chosen for stability, not size, though it ran well on half of the 2GB hdd before I upgraded to 20GB (and still does)), most of "generic" applications, just postfix for mail (I'm not a sendmail masochist), boa for webserver (apache is way too big for my needs), 128M of swap space (only 16M of RAM), NOTHING X. Webbrowsers: lynx, links. Mailers: mutt, pine. All common shells. It runs quite fine, though heavier tasks like compiling the kernel upgrade are performed on my other box because this one takes some 3 days... Also standard more computationally intense tasks take some time, like I wait some 5-8s for ssh login (then ssh runs smoothly). I recently got a new monitor for my primary box so I moved the old 17" CRT to that 486 (replacing a crappy 14" mono) and I can run even svgalib programs i.e. zgv to view pics (although it takes quite some time to decompress and display bigger JPEGs)
Uptimes reach a year and are usually ended by power failure or kernel upgrade. I had one major failure: the power supply died and I had a hard time finding an AT power supply.
I'd say the key to running it fast is giving up ALL of X-windows. I know that if I increased the swap size, I could run KDE on it. But it wouldn't be usable. In my case it runs great with console - and not only as firewall. I play some 'doze game and check hints from the FAQ on the other screen, or do some programming and keep some chat open on the other screen... The only problem is I keep confusing the keyboards:)
Take such Saiuz from Russians. No doubt it starts vertically. All of them do. Then the capsule lands on parachutes, mostly vertically too. Only shuttles don't land vertically. So essentially most of our spaceships are VTOL.
Slashdot Mirror
I will choose CRT over LCD anytime until I earn $10K/month or so ;) Then I will discard old LCD and buy a new one way before it degrades :)
Viewing angle??????
...unless you think opposite and consider it an advantage that nobody looking at your screen from a side would see a thing.
That word was unknown before the LCD era. It's a strictly LCD-related problem and still serious though great advances have been made...
CRT has full 180 degrees viewing angle. So does Plasma. LCDs are the ones with problems in this area.
What were the alternatives from Microsoft at the time of creation of Perl?
Isn't PHP a "favoured default" for Apache, backed by the Apache Foundation? What was the state of development of ASP when PHP has emerged from Perl?
And it's not Java. It SUN/Java.
I can say that sentence at about 200WPM without rushing. Can you type it that fast?
I'm a bad typist but I can write 100-200K in a single nightly session. Can you say that much without hurting your vocal strings? And for a skilled typist, 200WPM isn't all that much.
If you put that into command-line-ese, you'd get almost exactly what you typed
No. I'd get "copy from drive c windows system32 file kernel.dll to drive c" And that's the point - command line is still much more efficient than GUI, no matter if you use voice or keyboard (or even mouse for GUI) - so command line won't die even after keyboard dies. I just presented two ways of performing a task using voice commands, one in command line style, the other in GUI style. In this we agree...
Typing for a skilled typist is both much faster and easier than speaking. Especially for prolonged periods of time.
/usr/src/linux; copy vmlinuz /boot"
There's a rather long while till we get reliable thought-controlled interfaces, and even then they may need some extra training. How many more thoughts you create than you actually say/type? You cull most of the plans before they take any serious shape and are converted into words. Untrained thought-writing would be just a feast of spurts of senseless text and undoing them.
Reliable voice recognition may replace the keyboard in some cases, but even then it won't necessarily replace the command line. It's easier to form a command sentence than to call options being displayed on GUI.
say "changedir
versus
"my computer, open. drive C, open. Windows, open. System32, open. kernel32.dll copy. parent. parent. parent. paste."
We're not talking about "generic software" or some "pet desktop app". We're talking about a new programming language that has the intention of replacing one that is strong on the market already and is meant for pretty specific usage - mostly in areas where the last word belongs to suits, where multi-million-dollar transactions are made and so on. If your failure/success cost is difference between earning and losing a million dollars, will you pick a mostly untested in such an environment, written largely by amateurs and almost unknown language, possibly very good but of unknown risks, or go with a solution where all the weaknesses are known, it was tested hundreds of times in such conditions, and methods to avoid the bugs and weaknesses are all known to professionals you hire?
We're not talking about nerd's or his grandmother's computer. We're talking of real, big and damn expensive production environment.
Apache wouldn't be near anywhere to what it is now without support of the giants behind the Apache Foundation. Firefox gains recognition because millions of home users are interested in it. PHP rode on the shoulders of Apache and Perl. Linux has enough attention to be trustworthy.
In the world where decreasing risk by 0.1% means $10.000 profit, picking untested alternatives is out of question.
ASP 1.0 - end of 1996.
JSP - Sun's response to ASP.
PHP/FI 1995, PHP3 - 1997.
And of course dubious fame of ISS which was the only supported platform for ASP until recently and similarity of PHP to Perl which was the language of CGI for ages.
'cause OO is way easier for team writing and huge projects. It's way easier to split the project into many "single man" or "single small team" tasks, then bind them all together through an easy to use and strictly defined methods with well defined "responsiblity" areas. The difference isn't all that big, except of some "protectionism" (private, public), simplification of some processes (inheritance instead of notorious evil "copy&paste") and strict defining of "responsiblity areas" (objects), instead of guessing whether convert_hostname_to_lowercase() belongs to hostnameconv.h or tolowercase.h :)
I don't believe any open source solution in any near future could crush the Microsoft alternatives in the software development field.
The problem is that HERE marketing matters. Home users are free to pick a web browser or operating system of their choice. But when a big system for some business/industry is being developed, the platform decisione are made by the middle-to-upper management. And these guys really -believe- what Microsoft marketing people tell them. So the programmers, people who actually know a thing about the options don't really get the voice in most of the projects. "So... This guy at EXPO told me Visual Basic would solve all these problems. So we write the application in Visual Basic." There is no way the majority of the "big fishes" in programming could accept a hardly known free software language instead of the "famous, widely used Microsoft product" without the right marketing, and without some large funding behind the marketing...
Unless Sun, IBM or someone else with enough $$$ and not too much love for Microsoft backs up the project and takes care of marketing and promoting it. But the chances are very slim.
Unfortunately version 2.0 is known to crash quite badly when deployed. ESA had some problems with that...
I wonder what's the case about -writing- blogs and how many blogs out there aren't read even once.
Anyway, blogs definitely -should- have some kind of mark to help filter them off from Google. Sometimes they badly ruin search results.
I didn't get the article but I think that's not it.
Add an entry "mac" and entry "macintosh" and point both to Apple and you have the synonyms problem solved. Many words to describe the same thing, multiple entries describing the same page.
Worse about homonyms, where one word has several meanings. Wikipedia solves that by "disambiguation" pages.
Okay, I'm another person who didn't understand a thing.
Would someone who actually understood the article please explain wtf, or just translate it to human-readable form?
I think we should think of a manned mission... But not "bring them there, pick samples, bring them back". But of a real base...
:)
Send another 50 or so robotic missions. With robot "builders", "planters", machinery etc. Remotely build a base, whole with a launch pad, enough supplies and equipment to be self-sufficient for some 5 years, and with enough labs and assembly line that new robots delivered there in parts can be assembled and deployed locally. Then send a crew that will have specific tasks:
- control the robots locally to avoid the delay
- repair broken robots
- communicate and report in wider range than robots can.
- gradually assemble a rocket to get back home, from parts delivered in the meantime
By the way, a lot of this stuff could be made from Mars orbit. I think we could profit a LOT from a manned martian orbiting base, just like the Alpha station. A reusable shuttle or two for landing to perform the land-based maintenance, and the life area orbiting around Mars.
Note getting between Mars orbit and Earth orbit takes vastly less resources than getting from surface to orbit of any of the planets...
Yes, send humans to Mars. But don't rush it. Take enough time to make it worthwhile, and build enough robotic infrastructure to give the astronauts actual job for which they would be necessary, not just the PR thing.
we can do nothing to prevent, only to try to mitigate.
or send a warning early enough. With 15-minute warning (easy for science/engineering, harder for bureaucracy/business) the casualities could have been halved
..amongst many others :)
There were some challenges to crack (brute force) some of "better" encryption schemes to estimate -real- computational power to crack them. MD5, SHA, etc. I don't know about others but I remember MD5 took about a year with several thousands computers cracking it.
Of course that wasn't a "general" solution, just decrypting a single encrypted message. But it was proven it's possible.
Ah, and there are quite a few maths problems where some new solutions were found.
Barcodes are very handy but way too easy to copy.
You don't need special software. You don't need a computer to copy one. You can xero a barcode, and with enough skill, recreate one using pen and paper.
Then, you don't need to go even there! Go, buy Half-Life 1 from bargain bin, then remove the sticker with barcode, slap it on HalfLife 2 over the original one...
- Cracking MD5
- Human Genome
If I want to contribute to the community, I release Open Source. But if some commercial bastard pisses me off and releases their "free" proprietary program with some essential functionality missing "for strategical reasons", I release the alternative as GNU GPL'd Free Software - screw you, you want my features in your junk, re-release as GNU or write them yourself!
No. They didn't circumvent any mechanisms protecting copyrighted data in order to use that data. (and this is strictly what DMCA is about)
You could say they circumvented the protection (doubtful, the protection wasn't anywhere near to "efficient" as DMCA states) to access the copyrighted firmware. Except their aim is not to steal the original firmware but to replace it with their own, so the intent part isn't fulfilled at all. If they downloaded the firmware and started spreading it over BitTorrent, sure, then they are in violation of DMCA. But if they just make a backup for personal use and then write new software, sorry, nope. Sure they could be SUED under DMCA. But they would win the case hands down.
Even if they were spreading original but -modified- (not written from scratch) firmware, a good lawyer could argue it's fan art and as such, fair use, but that's more tricky.
Resident sniffer/logger.
Simple Firewall.
Monitor, blinking LEDs on certain kinds of packets arriving.
"Wake on ring" if not present by default.
"extra secret storage" in unused flash.
Changing MAC address...
*less* bandwidth (throttling your uplink, etc)
Yep. in non-hackable hardware.
If it's made illegal, it doesn't vanish. It only moves deeper under ground.
Imagine a beowulf cluster of these! :)
With nodes changing location within the cluster according to some obscure formula... and killing unsuspecting operators
Debian Woody (chosen for stability, not size, though it ran well on half of the 2GB hdd before I upgraded to 20GB (and still does)), most of "generic" applications, just postfix for mail (I'm not a sendmail masochist), boa for webserver (apache is way too big for my needs), 128M of swap space (only 16M of RAM), NOTHING X. Webbrowsers: lynx, links. Mailers: mutt, pine. All common shells. It runs quite fine, though heavier tasks like compiling the kernel upgrade are performed on my other box because this one takes some 3 days... Also standard more computationally intense tasks take some time, like I wait some 5-8s for ssh login (then ssh runs smoothly). I recently got a new monitor for my primary box so I moved the old 17" CRT to that 486 (replacing a crappy 14" mono) and I can run even svgalib programs i.e. zgv to view pics (although it takes quite some time to decompress and display bigger JPEGs)
:)
Uptimes reach a year and are usually ended by power failure or kernel upgrade. I had one major failure: the power supply died and I had a hard time finding an AT power supply.
I'd say the key to running it fast is giving up ALL of X-windows. I know that if I increased the swap size, I could run KDE on it. But it wouldn't be usable. In my case it runs great with console - and not only as firewall. I play some 'doze game and check hints from the FAQ on the other screen, or do some programming and keep some chat open on the other screen... The only problem is I keep confusing the keyboards
Ah. The CPU is "Cyrix Instead"
Take such Saiuz from Russians. No doubt it starts vertically. All of them do. Then the capsule lands on parachutes, mostly vertically too. Only shuttles don't land vertically. So essentially most of our spaceships are VTOL.