There is no "for" macro in Lisp, though there is a "for" keyword in the loop macro. There's nothing stopping you from defining one yourself, of course.
When the game World War II Online first came out, there were no levels, no equipment other than stock equipment, no concept of a grind at all. If you were killing newbs, it was because you were a better player, not because you had a Super Vorpal Sword of Doom. At 25, at the time, I was one of the younger players. Most of these guys were in their 30s and 40s with wives and kids, and weren't about to play for 40 or more hours a week.
There were other problems with the game which kept it from being more successful (mainly it was the buggiest POS in the history of gaming, with 70MB of patches required out of the box at a time when most people were still on dialup, and many many bugs and missing features). But it was a proof of concept of a non-grind MMOG for people with lives.
I agree, I meant that the popularity of OS X with Unix geeks is a threat to Linux. The move to x86 shouldn't make a difference. Actually, for a year or so it may slow the switch rate as the Osborne effect kicks in and people decide to wait for the new chips.
I understand open source, and it's quite likely that someone will pick up xscreensaver. The point is that when developers leave Linux, that reduces the amount of manpower available to improve it. The people I know, and at the forums I read, seem to be switching in droves, so that could be serious. Maybe these people aren't representive, I don't know.
Switchers will likely continue to develop at least non-GUI stuff for multiple *nix platforms, but they aren't likely to develop things that'll make Linux into a better Desktop because that stuff comes with their Macs.
Last time I setup a Linux system, it was just as easy as setting up a Windows. OK, so I got lucky. The time before that, it took about three days of fussing with it to get it to run all of my hardware properly and install needed software. With Windows, it would have taken a day, most of the time being spent running install wizards for the essential things that don't come with Windows. Windows wins up front. But a year and a half later, I've never updated my anti-virus definitions, worried about spyware, or had to hunt down needed software (apt-get beats tucows.com hands down). So who's really wasted their time?
Last time I visited my dad, he could not access the web from his Windows system, apparently due to haywire anti-virus software. This is the OS that puts such value on its users time?
It's a threat to Linux because quite a few Linux users have switched to Macs, and those are people who won't be contributing to Linux anymore. Case in point, Slashdot ran a story a few days ago about jwz's defection - he had written a mp3 jukebox, and more importantly, xscreensaver, which I've been using myself for 3 years or so. So those are now probably orphaned projects, because he will probably just use the software that comes with his Mac. Jwz's defection, by itself, is not that big of a deal, but if a lot of people switch, it could add up to a big deal.
Generally new printers come with half full (or less) cartridges, so you didn't do yourself any favors. Refills or 3rd party cartridges are the way to go, and if a printer is so "smart" that it doesn't allow that, maybe it'll make a nice doorstop.
I run Debian unstable, and I installed Ubuntu for my mom, so I have some experience with both. Ubuntu seemed nice and coherent, but the problem was that it runs Gnome, which is far too bloated for the piece of shit box I installed it on (if she had a machine capable of running Gnome with reasonable speed, I would have just left Windows XP on it). In the end, I had to mess around with config files to get rid of all of the Gnome crap, and it was about as much hassle to install Ubuntu as Debian.
To use Emacs just for a diary would be overkill, but some of us live in Emacs and it makes sense to do stuff in that environment. Once you're already running Emacs, the diary is just a 1500 line extension that comes with it.
If you behave like a human shell script and slavishly fllow the instructions, you won't learn anything. If you make sure you understand every step, you learn a lot.
The problem is that old emacs and Vim users have climbed up a steep learning curve, and aren't going to be eager to do that again for a text editor. I can't speak for Vim, but emacs is in many ways obsolete - it's written in a dialect of Lisp that should have gone out with disco, and it lacks the ability to do a lot of GUI eye-candy stuff that can actually display useful information. But it's got tons of stuff written for it, and for those of us who know it, the keybindings are hard-coded into our motor neurons, so there's not likely to be a stampede away from it anytime soon.
jEdit looks nice, but I'd have to see a very big advantage to want to switch now. And I'd be nervous about the fact that it's essentially one man's open source project (Slava Pestov). If I have to bet on which one is more likely to still be supported in 10 years, I'm betting on emacs.
Yes, you can remap your keybindings, and you can map keys to emulate the "missing" modifier keys (the machines emacs originally ran on had keyboards with Hyper, Alt (not the same as your "alt" key), and Super keys, and you can make something useless like the windows key do their job), and map the overly complex 80's guitar god chords to something simpler using these. I've done so, and probably with a lot less hassle than hooking up an Erector set to my keyboard.
Transportation is transportation, why make it overly complicated by learning to drive when you can just walk?
Because it'll save you a lot of time in the (not so) long run, after a minor loss of time learning to do something new. If your editing needs are no more demanding than occasionally editing a.foorc file, use nano, but if you're coding, learn to use a editor. Nano doesn't touch emacs or vi.
I've got commonly used 3+ key chords rebound to Super+(a single key). Super is the left windows key on my setup. This'll allow me to keep the use of my hands for a few extra years, I hope.
Setting up Debian unstable was a pain in the ass, but having suffered through that a year and a half ago, it's been mostly trouble free. Every week or so I use apt-get to upgrade packages I really care about, and every several months I run apt-get dist-upgrade to upgrade everything. That's all the maintenance I do, and it probably amounts to 15 minutes/week. My uptime is now 80 days, and that's because of a power outage.
I set up Ubuntu for my mom, and it seems to be easier. The printer just worked, which was a first for any Linux distro in my experience. It's such a resource hog (it uses Gnome) that I still had to fuss with it to make it usable on her 7 year old, low-end when it was new, eMachines piece of crap (got rid of all the gnome crap and installed icewm, xdm, etc), but I think it would "just work" on a decent computer made in the last 5 years. And maintenance has been trouble-free like debian.
There'll be new bugs with IE7, no doubt, but it'll be a long damn time before the market share of IE6 drops to the point where you can ignore it for anything more important than posting pictures of your cat.
A real Lisper doesn't want to declare his variables.. this would start to look at least superficially like lisp if I had a clue about how to post code that indents properly on/.
#define STMT(a) a; #define ARG(b) void* b #define DEFUN(b,c,d) void* b c { d return 0;} #define LETARG(a, b) void* a = b; #define LET(a,b) (a = (int) b) #define ADD(a,b) ((int) a + (int)b) #define MUL(a,b) ((int) a * (int) b)
I don't think the problem is lack of libraries, because new languages like Python have come along and outstripped Lisp's library support. I think Lisp's lack of popularity stems mainly from 2 things:
1) It's alien. It's a bigger jump to go from C to Lisp than it is to go from C to Perl or Python. Technology succeeds by disturbing peoples' habits as little as possible.
2) Its roots are in academe and AI. The Lisp culture was never oriented towards what most grunt programmers would consider practical work. That's why it's a pain in the ass to certain things in Lisp that are easy in Perl - the people who designed Lisp didn't care about those things. There's nothing inherent in Lisp that makes what would be a simple one-liner in Perl more difficult, it's just not the focus of the language. Name one successful language that was created by academics.
There's also no inherent reason why there couldn't be a Visual Lisp, other than that Lispers like Emacs and aren't likely to develop something like that. (actually, if you count Scheme as a dialect of Lisp, DrScheme comes close).
I have no personal experience with it, but the Prolog embedded in the latest Allegro Common Lisp is supposed to be pretty fast (like not-a-toy fast). This is new within the last year or so.
It's not C-influenced-language programmer friendly. If you learn Lisp before you're indoctrinated into Algol-descended languages, the opposite is true - the syntax of those languages is too complicated and annoying, with too many things to remember (like 10 levels of operator precedence).
It seems difficult to you for the same reason Japanese seems difficult to native English speakers, and English seems difficult to native Japanese speakers.
It seems to be a widely held misconception that Lisp is oriented towards functional programming and recursion; actually, tail recursion is not required by the standard, it's only provided in an obscure, implementation-dependent, often declaration-dependent manner. Real Common Lisp code generally makes liberal use of iteration and assignment.
Maybe people take a class in Scheme and get confused.
Slashdot Mirror
There is no "for" macro in Lisp, though there is a "for" keyword in the loop macro. There's nothing stopping you from defining one yourself, of course.
When the game World War II Online first came out, there were no levels, no equipment other than stock equipment, no concept of a grind at all. If you were killing newbs, it was because you were a better player, not because you had a Super Vorpal Sword of Doom. At 25, at the time, I was one of the younger players. Most of these guys were in their 30s and 40s with wives and kids, and weren't about to play for 40 or more hours a week. There were other problems with the game which kept it from being more successful (mainly it was the buggiest POS in the history of gaming, with 70MB of patches required out of the box at a time when most people were still on dialup, and many many bugs and missing features). But it was a proof of concept of a non-grind MMOG for people with lives.
I agree, I meant that the popularity of OS X with Unix geeks is a threat to Linux. The move to x86 shouldn't make a difference. Actually, for a year or so it may slow the switch rate as the Osborne effect kicks in and people decide to wait for the new chips.
Switchers will likely continue to develop at least non-GUI stuff for multiple *nix platforms, but they aren't likely to develop things that'll make Linux into a better Desktop because that stuff comes with their Macs.
How are we ever going to have a decent looking design without the gays????
Last time I visited my dad, he could not access the web from his Windows system, apparently due to haywire anti-virus software. This is the OS that puts such value on its users time?
It's a threat to Linux because quite a few Linux users have switched to Macs, and those are people who won't be contributing to Linux anymore. Case in point, Slashdot ran a story a few days ago about jwz's defection - he had written a mp3 jukebox, and more importantly, xscreensaver, which I've been using myself for 3 years or so. So those are now probably orphaned projects, because he will probably just use the software that comes with his Mac. Jwz's defection, by itself, is not that big of a deal, but if a lot of people switch, it could add up to a big deal.
Generally new printers come with half full (or less) cartridges, so you didn't do yourself any favors. Refills or 3rd party cartridges are the way to go, and if a printer is so "smart" that it doesn't allow that, maybe it'll make a nice doorstop.
So why not use Debian?
Take your paternalistic, we-know-what-you-need-better-than-you-do ideas and shove them up your ass.
To use Emacs just for a diary would be overkill, but some of us live in Emacs and it makes sense to do stuff in that environment. Once you're already running Emacs, the diary is just a 1500 line extension that comes with it.
If you behave like a human shell script and slavishly fllow the instructions, you won't learn anything. If you make sure you understand every step, you learn a lot.
jEdit looks nice, but I'd have to see a very big advantage to want to switch now. And I'd be nervous about the fact that it's essentially one man's open source project (Slava Pestov). If I have to bet on which one is more likely to still be supported in 10 years, I'm betting on emacs.
Yes, you can remap your keybindings, and you can map keys to emulate the "missing" modifier keys (the machines emacs originally ran on had keyboards with Hyper, Alt (not the same as your "alt" key), and Super keys, and you can make something useless like the windows key do their job), and map the overly complex 80's guitar god chords to something simpler using these. I've done so, and probably with a lot less hassle than hooking up an Erector set to my keyboard.
Because it'll save you a lot of time in the (not so) long run, after a minor loss of time learning to do something new. If your editing needs are no more demanding than occasionally editing a .foorc file, use nano, but if you're coding, learn to use a editor. Nano doesn't touch emacs or vi.
I've got commonly used 3+ key chords rebound to Super+(a single key). Super is the left windows key on my setup. This'll allow me to keep the use of my hands for a few extra years, I hope.
I set up Ubuntu for my mom, and it seems to be easier. The printer just worked, which was a first for any Linux distro in my experience. It's such a resource hog (it uses Gnome) that I still had to fuss with it to make it usable on her 7 year old, low-end when it was new, eMachines piece of crap (got rid of all the gnome crap and installed icewm, xdm, etc), but I think it would "just work" on a decent computer made in the last 5 years. And maintenance has been trouble-free like debian.
Try Brad Delong. Similar point of view, similar credentials, lower price point.
There'll be new bugs with IE7, no doubt, but it'll be a long damn time before the market share of IE6 drops to the point where you can ignore it for anything more important than posting pictures of your cat.
True, but for a lot of people, web and email is all they really use. For those people, any money spent on Microsoft products is a total waste.
1) It's alien. It's a bigger jump to go from C to Lisp than it is to go from C to Perl or Python. Technology succeeds by disturbing peoples' habits as little as possible.
2) Its roots are in academe and AI. The Lisp culture was never oriented towards what most grunt programmers would consider practical work. That's why it's a pain in the ass to certain things in Lisp that are easy in Perl - the people who designed Lisp didn't care about those things. There's nothing inherent in Lisp that makes what would be a simple one-liner in Perl more difficult, it's just not the focus of the language. Name one successful language that was created by academics.
There's also no inherent reason why there couldn't be a Visual Lisp, other than that Lispers like Emacs and aren't likely to develop something like that. (actually, if you count Scheme as a dialect of Lisp, DrScheme comes close).
I have no personal experience with it, but the Prolog embedded in the latest Allegro Common Lisp is supposed to be pretty fast (like not-a-toy fast). This is new within the last year or so.
It seems difficult to you for the same reason Japanese seems difficult to native English speakers, and English seems difficult to native Japanese speakers.
Maybe people take a class in Scheme and get confused.