No one building the twin towers said "What if a plane crashed into the building?" when they were building the WTC.
actually, the WTC towers' designers *did* consider the possibility of an aircraft hitting the building, and designed the structure to stay upright in the event of such an impact. what they could not design for was the intense heat of a few thousand pounds of burning fuel; it was the fire that brought them down, not the planes.
The article notes that the "annoying thing" about copyright when it was only 56 years is that the author could outlive his monopoly on the work.
But that's intentional. That's the incentive for the author not just to sit back and milk one cow until it's dead, but to keep creating new stuff to further enrich the culture. Oh, and the enriching the culture bit requires things to go public domain.
as an author of a web page or even a log file, you have the right to publish and de-publish it. just because it's on the net does not give google the right to cache it indefinitely.
by the publishing analogy, doesn't this mean that libraries don't have the right to lend books that are no longer in print? in that respect i see google's cache as a library's copy of a book; they let you look at it, and you can see when it was published. they don't claim it's the most up-to-date, and at any time you can go to the source and see for yourself (e.g. go to a bookstore and buy a new copy).
"What people tend to forget is that there are gatekeepers in the open-source community, too," he said. "It's not a free-for-all. On every one of the open-source projects, there are two or three people who are the gatekeepers. And you have to make a pretty good case, accurate and technically astute, to get them to allow changes. That's how it should be."
interesting; they assume that no one would ever want to make a modification just for private use. yes, there are gatekeepers in the open software world, but only if you wish to get your changes committed to the "official" distribution. with shared source, since you aren't allowed to compile, the source access gets you diddly squat, whereas with most OSS, you can actually benefit from being able to modify the code. you don't even have to make your changes public if you never distribute that changed program.
i guess that gives us insight into their model of customers: dumb users with a minority of dumb developers.
[ weird system config things you have to do to install new experimental software for playing media formats not supported directly by open source software ]
yes, that sucks, but you should only have to do it *once*. and if you pick the right distro, they'll have done it for you -- despite all my other gripes about mandrake, i have to admit that it's nice that they have this sort of thing already set up.
It is slower. End of story. No matter what you say, no matter what benchmarks or other stuff you come up with, qt/gtk widgets are STILL slower than win32 widgets, watching dvd with XINE takes 40% of my CPU while under windows it takes 5%(five), process spawning is slower (under windows if i run iexplore.exe repeatedly, it pops up new windows at a rate about 5 windows/second. Under linux, the best i could do is 0.5 new windows/sec. Dirty test, i agree, but...
okay, you've conflated several things here. the spawn() call on windows is MUCH slower than the fork() / exec() pair on linux. run a configure script on linux and on cygwin and you'll see what i mean.
as for the gui stuff --- X has lots more libraries to load than windows, because windows puts the GUI stuff in the kernel where it's really fast and can bring your system down if anything goes wrong. X is separate from the kernel to avoid this.
clicking on iexplore.exe doesn't actually pop up a new window from a new process; the new process sees that a previous instance exists, and sends an "open a new window" message to that instance and then exits. it's the same one that was already running that actually pops up the new window. most X apps don't do that (no instance handles), but mozilla i think can do remote connection by a command line option (the gnome-moz-remote command does this).
not entirely refuting your speed evidence here, just pointing out that it's not caused by what you said.
Lack of Games
the lack of games is a market share problem, more political than technical. not a good answer, just an explanation.
Sure, you might say "but windows isn't perfect either!" but don't you want your linux to be?
they probably think it *is* perfect for them; they're scratching an itch after all. that doesn't necessarily mean they're trying to please everyone. most apps that i've written have been specifically designed to do what i needed. i make the code available because i think somebody else might find it useful, but so far, nobody has. does that mean i should give up my program and work with somebody else on a project that has only passing interest in solving my problem?
can't help you with non-english, never tried. sorry. but if you want to write in Lyx, they have all sorts of users who also don't do english. poke around on their website or mailing lists, i'm sure there's *something*. latex should print fine, that's what it's designed for.
1. start your computer 2. unregister from MSN 3. shut down your computer 4. disconnect all cables from computer 5. open nearest 10th sorty or higher window 6. defenestrate afflicted computer
or, somewhat more practically, dd if=/dev/zero of=/dev/hda bs=512 , let run long enough to erase the partition table, then install to clean disk as though it was new from the factory.
Still, VisualC++ has had support for string and wstring classes at least from v5. And even though their STL implementation is all wrong, and they add some extensions to the language, at least, you can write almost proper standard C++.
BZZZT! WRONG, oh so annoyingly wrong. try this code in VC++6:
by the definition of C++'s loop scode (not an ISO thing, but a basic C++ thing that i've seen in books reaching back many years), that should NOT work, and in fact won't compile with g++, but compiles fine in MSVC++.
on the converse, this code:
void foo2()
{ for (int i=0; i<10; i++) { cout << i << " loops to bind them...\n"; }
for (int i=10; i>0; i--) { cout << "...them bind to loops " << i << endl; } }
won't compile with MSVC++6 --- the error it gives: redefinition of i!!! but i wasn't declared in function scope!!!!
note that both of these compile perfectly with g++. i've run into this problem when writing code from my C++ books and when porting code from unix to windows ever since i started using MSVC++ at 4.something. this bug really pisses me off.
Where are more people like you? I've looked, honestly. All the female 'geeks' I've ran into have been maried. Oh well, guess I'll have to stick to dating art/music women for now.
they're taken becuase they who find them know they've found A Good Thing.
not that there's anything wrong with the art/music women, either. for me, the ultimate is a programmer is who is also a musician.
holding shift while clicking with the pencil or brush tool connects with a straight line the new point with the last one you made with that tool.
so, to make a straight line, click somewhere, shift+click somewhere else.
the development series gimp will show you where the connecting line will be if you just press shift. quite cool. in fact, nearly everything about gimp1.1 is better than 1.0.
Okay, I understand this and agree with you, the software value-add of a good driver is A Good Thing for a hardware company.
However, I think that a more important thing, which the feature mentions but to which you didn't reply, is the hardware specs. The hardware specs allow anyone to write the type of driver you're talking about --- to complete with you, but not steal your stuff because they're writing their own algorithms.
Creating hardware without publishing the specs dooms your hardware to be used only by those with the supported platform. By publishing the specs, those on unsupported platforms who want to use it can write drivers, and, very possibly, when the user base for the unsupported platform grows large enough on the renegade driver, there just may be enough of a market to justify the release (for sale by you, the vendor) of a souped-up driver using your super-secret patented software algorithms. If this software is Good, people will buy it.
Everyone wins. Only the "all-software-must-be-completely-open" people are unhappy, but they would be unhappy anyway.
To run the point into the ground, if you're going to make hardware, you should tell people how to use it. Period.
the guy asks if the linux community will accept RedHat's contributions to the source if they're making money for it... is this guy daft? they [the modifications] would be judged and accepted or denied the same way they always have been --- by their merit!
but what worried me more was the subheading immediately above the linux story, telling that the cellular industry had finally admitted that cell phones cause brain cancer. how disturbing... for one, my roommate just signed up for cellular service, and two, what will happen to scully and mulder?
as long as there's a netscape in your path, there should be no problem. the best thing you could do would be to install netscape to some version-dependent directory (as you've done), create a symlink from the current version dir to/usr/local/netscape, then have a wrapper script in/usr/local/bin that calls/usr/local/netscape/netscape. then the upgrading is only a question of fixing the symlink, and you don't have to add a directory to the system-wide PATH.
there are excellent wrapper scripts available which account for many of netscape's insane command-line quirks, and make it much friendlier. look on freshmeat for nss or netscape-wrapper.
i've pulled the same trick with realplayer.
to stay on topic, i haven't tried the new G2, but the most recent stable version works very well on my libc6 system. i had to re-download it, tho', because the previous libc5 version did nothing but dump core all over the place.
see the subject line. the whole "drug him till he's normal" mentality just pisses me off...
different people are wired differently. it's A Good Thing. why people feel compelled change it?
Re:Problems with addressbook
on
Netscape 4.6
·
· Score: 1
i've had this problem as well... the addressbook created several list items for the same entry, and i could delete none of them. eventually, it came up with blank, empty entries, and started crashing at random.
i pretty much had to resort to blowing away my ancient.netscape directory and rebuilding the whole thing from scratch, addressbook, bookmarks and all.
it made me not very happy.
and i couldn't export the addressbook, either. it would crash (silently) when trying to export. i think there were null pointers in the linked list or something...
Um, "markup language", "text formatting language", what's the diff?
well, the name kinda says it all, doesn't it?
the difference is the intent, and semantics. with a text-formatting language, you indend to format the text in a very particular way. with a markup language, you simply intend to denote which parts are of which type, letting something else later parse that logical markup for whatever purposes.
and this is the difference in practice. LaTeX defines things exactly, but html defines things loosely.
You should be able to explain your reasoning behind Good Web Design to your boss. If your boss won't listen to a cogent argument about usability, thinks flash is all that's necessary, and would fire you for trying to do your job well, do you really want to work there?
Slashdot Mirror
No one building the twin towers said "What if a plane crashed into the building?" when they were building the WTC.
actually, the WTC towers' designers *did* consider the possibility of an aircraft hitting the building, and designed the structure to stay upright in the event of such an impact. what they could not design for was the intense heat of a few thousand pounds of burning fuel; it was the fire that brought them down, not the planes.
i recall hearing this in the incessant news coverage at the time, but google corroborates my story.
The article notes that the "annoying thing" about copyright when it was only 56 years is that the author could outlive his monopoly on the work.
But that's intentional. That's the incentive for the author not just to sit back and milk one cow until it's dead, but to keep creating new stuff to further enrich the culture. Oh, and the enriching the culture bit requires things to go public domain.
"What people tend to forget is that there are gatekeepers in the open-source community, too," he said. "It's not a free-for-all. On every one of the open-source projects, there are two or three people who are the gatekeepers. And you have to make a pretty good case, accurate and technically astute, to get them to allow changes. That's how it should be."
interesting; they assume that no one would ever want to make a modification just for private use. yes, there are gatekeepers in the open software world, but only if you wish to get your changes committed to the "official" distribution. with shared source, since you aren't allowed to compile, the source access gets you diddly squat, whereas with most OSS, you can actually benefit from being able to modify the code. you don't even have to make your changes public if you never distribute that changed program.
i guess that gives us insight into their model of customers: dumb users with a minority of dumb developers.
[ weird system config things you have to do to install new experimental software for playing media formats not supported directly by open source software ]
yes, that sucks, but you should only have to do it *once*. and if you pick the right distro, they'll have done it for you -- despite all my other gripes about mandrake, i have to admit that it's nice that they have this sort of thing already set up.
It is slower. End of story. No matter what you say, no matter what benchmarks or other stuff you come up with, qt/gtk widgets are STILL slower than win32 widgets, watching dvd with XINE takes 40% of my CPU while under windows it takes 5%(five), process spawning is slower (under windows if i run iexplore.exe repeatedly, it pops up new windows at a rate about 5 windows/second. Under linux, the best i could do is 0.5 new windows/sec. Dirty test, i agree, but...
okay, you've conflated several things here. the spawn() call on windows is MUCH slower than the fork() / exec() pair on linux. run a configure script on linux and on cygwin and you'll see what i mean.
as for the gui stuff --- X has lots more libraries to load than windows, because windows puts the GUI stuff in the kernel where it's really fast and can bring your system down if anything goes wrong. X is separate from the kernel to avoid this.
clicking on iexplore.exe doesn't actually pop up a new window from a new process; the new process sees that a previous instance exists, and sends an "open a new window" message to that instance and then exits. it's the same one that was already running that actually pops up the new window. most X apps don't do that (no instance handles), but mozilla i think can do remote connection by a command line option (the gnome-moz-remote command does this).
not entirely refuting your speed evidence here, just pointing out that it's not caused by what you said.
Lack of Games
the lack of games is a market share problem, more political than technical. not a good answer, just an explanation.
Sure, you might say "but windows isn't perfect either!" but don't you want your linux to be?
they probably think it *is* perfect for them; they're scratching an itch after all. that doesn't necessarily mean they're trying to please everyone. most apps that i've written have been specifically designed to do what i needed. i make the code available because i think somebody else might find it useful, but so far, nobody has. does that mean i should give up my program and work with somebody else on a project that has only passing interest in solving my problem?
can't help you with non-english, never tried. sorry. but if you want to write in Lyx, they have all sorts of users who also don't do english. poke around on their website or mailing lists, i'm sure there's *something*. latex should print fine, that's what it's designed for.
1. start your computer
2. unregister from MSN
3. shut down your computer
4. disconnect all cables from computer
5. open nearest 10th sorty or higher window
6. defenestrate afflicted computer
or, somewhat more practically, dd if=/dev/zero of=/dev/hda bs=512 , let run long enough to erase the partition table, then install to clean disk as though it was new from the factory.
Still, VisualC++ has had support for string and wstring classes at least from v5. And even though their STL implementation is all wrong, and they add some extensions to the language, at least, you can write almost proper standard C++.
BZZZT! WRONG, oh so annoyingly wrong. try this code in VC++6:
by the definition of C++'s loop scode (not an ISO thing, but a basic C++ thing that i've seen in books reaching back many years), that should NOT work, and in fact won't compile with g++, but compiles fine in MSVC++.
on the converse, this code:
won't compile with MSVC++6 --- the error it gives: redefinition of i!!! but i wasn't declared in function scope!!!!
note that both of these compile perfectly with g++. i've run into this problem when writing code from my C++ books and when porting code from unix to windows ever since i started using MSVC++ at 4.something. this bug really pisses me off.
muppet
they're taken becuase they who find them know they've found A Good Thing.
not that there's anything wrong with the art/music women, either. for me, the ultimate is a programmer is who is also a musician.
holding shift while clicking with the pencil or brush tool connects with a straight line the new point with the last one you made with that tool.
so, to make a straight line, click somewhere, shift+click somewhere else.
the development series gimp will show you where the connecting line will be if you just press shift. quite cool. in fact, nearly everything about gimp1.1 is better than 1.0.
unpack the zip archive somewhere, maybe c:\temp...
this will create a subdirectory, bin, if i recall correctly...
in that subdirectory you will find tons and tons of subdirectories and dlls and exes.
i presume you've already done this, and the mess of files is the source of your complaint ---- NEVER FEAR!
simply execute apprunner.exe --- everything else just happens.
Okay, I understand this and agree with you, the software value-add of a good driver is A Good Thing for a hardware company.
However, I think that a more important thing, which the feature mentions but to which you didn't reply, is the hardware specs. The hardware specs allow anyone to write the type of driver you're talking about --- to complete with you, but not steal your stuff because they're writing their own algorithms.
Creating hardware without publishing the specs dooms your hardware to be used only by those with the supported platform. By publishing the specs, those on unsupported platforms who want to use it can write drivers, and, very possibly, when the user base for the unsupported platform grows large enough on the renegade driver, there just may be enough of a market to justify the release (for sale by you, the vendor) of a souped-up driver using your super-secret patented software algorithms. If this software is Good, people will buy it.
Everyone wins. Only the "all-software-must-be-completely-open" people are unhappy, but they would be unhappy anyway.
To run the point into the ground, if you're going to make hardware, you should tell people how to use it. Period.
but what worried me more was the subheading immediately above the linux story, telling that the cellular industry had finally admitted that cell phones cause brain cancer. how disturbing... for one, my roommate just signed up for cellular service, and two, what will happen to scully and mulder?
there are excellent wrapper scripts available which account for many of netscape's insane command-line quirks, and make it much friendlier. look on freshmeat for nss or netscape-wrapper.
i've pulled the same trick with realplayer.
to stay on topic, i haven't tried the new G2, but the most recent stable version works very well on my libc6 system. i had to re-download it, tho', because the previous libc5 version did nothing but dump core all over the place.
different people are wired differently. it's A Good Thing. why people feel compelled change it?
i pretty much had to resort to blowing away my ancient .netscape directory and rebuilding the whole thing from scratch, addressbook, bookmarks and all.
it made me not very happy.
and i couldn't export the addressbook, either. it would crash (silently) when trying to export. i think there were null pointers in the linked list or something...
but how can i tell? it's closed-source...
well, the name kinda says it all, doesn't it?
the difference is the intent, and semantics. with a text-formatting language, you indend to format the text in a very particular way. with a markup language, you simply intend to denote which parts are of which type, letting something else later parse that logical markup for whatever purposes.
and this is the difference in practice. LaTeX defines things exactly, but html defines things loosely.
You should be able to explain your reasoning behind Good Web Design to your boss. If your boss won't listen to a cogent argument about usability, thinks flash is all that's necessary, and would fire you for trying to do your job well, do you really want to work there?