I don't just want to know how to program with functional languages in the real world. I'd like to be able to link C/C++ code with functional code. I've discovered that functional languages are great for the things imperative languages are terrible for, and vice versa.
If I could link the two together, I might actually succeed in being able to use the Right tools for the Job.
Ahh. . . you must be talking about any one of the host of books falling along the lines of "They Mythical Man-Month" or "Rapid Development"
Really, you're not going to find much outside of journal articles on this subject, since software engineering is still such a fuzzy science. Heck, I don't think anyone has even gotten around to figuring out what it is yet, let alone pinning down how to do it.
Not the gloss-overs that are put in most books, but an in-depth look at how to get set up w/ graphics and sound programming on linux - something that picks a specific toolkit or toolkits and really talks about it in gory detail. Performance issues, compatibility issues, getting the most out of OpenGL/DRI on that platform, whatever.
No, really. While we're building a 22,000 mile long cable strong enough to hold 20 tons, where are we going to put it? It's probably too big to actually/store/ anywhere, and, even if we did, imagine trying to transport the sucker.
We can't start stringing it off into space as we build it, because it'd keep tending to fall back towards earth until it were about... oh. . 22,000 miles long. (assuming they planned it so that 22,000 miles long would put its center of mass in a stationary orbit)
And, once we've figured out all that, how do we get it/up/ there? Build a 22,000 mile high crane? Really, I'd think that the rocket we would use to get the other end of a space elevator up there would be a greater feat of engineering than the space elevator itself, and building it in sections would probably be an even greater feat of engineering.
That said, it'd be a damn cool thing if we had it, and if a team ever succeeds in constructing one, I'll personally buy a beer for every member of the project.
This probably isn't a document saying what qualifications you must meet to get onto the space station.
This is something saying what qualifications you must meet to get on the space shuttle going to and from the space station (which is also NASA property). This means they effectively do control tourists to the entire space station, not just the US modules.
Until, of course, another country starts up in the space tourism industry. . .
"platform agnostic" refers to the kind of agnosticism you get with people who are unsure whether they agree more with the Roman Catholic or Greek Orthodox churches. Every "platform agnostic" system I have seen lately still requires IE, meaning you get a choice between Windows and MacOS.
"Linux-friendly" is something I have seen many businesses put up just to get business, but aren't willing to back up. Take Netgear. I bought a Netgar FA411 PCMCIA ethernet card a while ago. It said it works with Redhat 6.2. Didn't come w/ drivers on the CD (what, did they run out of space!?), so I had to download them. Precompiled module refused to work with any kernel I had, and the source version refused to work with gcc. Called tech support, and they said they didn't provide support for linux installs. Some "linux friendly."
We don't need platform agnosticism and linux friendliness, we need a return to thinking of "industry standard" as an agreed upon standard that everyone can use, rather than a piece of intellectual property that happens to be owned by the company with the most popular product.
Only, that prolly won't happen 'cos we're all so busy playing copycat that we get more excited about figuring out how to decode Quicktime or Windows Media files than we do trying to push the success of the Ogg project or anything else that might save the open source community from its rapidly approaching death at the hands of incompatibility.
I'd add to this that, if I were an admin worth his salt and was told that security was my prime concern, linux would never even cross my mind. If I were to pick any open source unix-like, it would be OpenBSD, hands down. About 10 minutes looking through Bugtraq should be enough explanation for my reasoning.
No offense, Mr. Barr, but the idea of Linux running on sensitive CIA or FBI computers seems patently ludicrous to me.
I have no problem with advertising in and of itself, so long as it doesn't potray itself obtrusively. Magazine ads on their own page are fine. Magazine ads inline in the article are not okay. TV Commerical breaks I can live with. TV ads in the form of bars across the bottom of the screen complete with sounds that make it hard to make out whatever is being said in the program you're watching are not. I feel similarly about the net - banner ads, ok. inline banners, no. popovers, no.
Once again, everyone blames "evil hackers" instead of addressing the real problem or even so much as hinting that sysadmins, or beter yet, PHBs should take part of the blame. So, what else is new?
Sysadmins and PHBs? Where I work, sysadmins and PHBs are probably the only people working to stop the viruses - and believe me, if we could get away with doing the things we want to do to stop them, we would. We already virus scan everything that comes through email, and we've gone so far as to create worms that search peoples' computers for viruses and sends a message to a server to shut off that computer's port when one is found. We remove MSIE and Outlook from all computers, actively discourage their use, and the helpdesk answers all questions related to the two with, "Get Netscape and then we'll talk."
And despite all this, we got hit hard by Code Red and have been trying for months to get all traces of w32.nimda cleared off the network. Problem is, no matter how much we deprecate as much unsafe software as we can, no matter how much we repeat things like, "don't open unexpected attatchments" at workshops, people still do all the krap we tell them not to do, and have more excuses for why they do it than I care to mention.
If we had our way, we would probably have blocked every kind of attatchment from *.mp3 to *.doc by now.
We can sit and whine about script kiddies and place the blame on them, but that won't ever solve the problem. People will always pull shenanigans because they can. We can blame stupid users, but if there's one thing I've learned from mopping up other peoples' shit, it's that people will always do dumb things.
As far as I'm concerned, there's really only one place we can have much of an affect on this problem, and it is the companies that make software. It's a bit like the problem of auto accidents - you can sit and blame drunks and bad drivers, but the only way you're going to save lives is by imposing safety standards on the automobile industry. Similarly, the only way we're going to fix the problem of viruses is by having some sort of accountability process for companies. I can understand not holding the company accountable when an exploit involves using some obscure bug in the program - we can't reasonably expect bugs to never occur in programs. We can hold companies accountable for idiotic things that nobody should do in commercial software, such as 8-bit XOR encrypting passwords - one of the things that nimda exploited to be able to spread across Microsoft networks so easily.
I'm curious if popunder and popunder ads and inline banners really get more hits than the traditional top-o-the-page banner
I've clicked one or two banner ads that attracted my interest. I don't really mind them. On the other hand, I find other forms of ads so irritating that I ignore purely out of spite. And I doubt they really succeed in getting the user's attention any more than other forms of ad - heck, they probably get less of my attention since I usually close ads that open in separate windows before they can load. And I've never actually followed a link on one of them.
So, at least if you limit the research sample group to me, the top-o-the-page banner is infinitely more effective than other ad types, which are generally a complete waste of money.
I agree that control of the code has a lot to do with open source, but I think that Bill is hitting on the reasons why OSS has become popular rather than te reasons why it exists at all.
The GNU project started in 1983 - after the first MS operating system, but long before Microsoft because the industry-dominating force it is right now. Bill surely knows this, and from that alone he can't claim any responsibility for the ideological roots of Open Source.
He's either talking about kids like me, who got into OSS to be different. I'm staying with open source for many other reasons, but if it weren't for the fact that Linux is the only way to try something different in a world where the only choices are Windows or MacOS (contrast with the days of MS-DOS, CP/M, Apples, Amigas, c64s, Spectrums, etc.) And before anyone jumps up and down shouting "BSD! BeOS!," sit down. The fact of life is, most people hear about Linux long before they hear about anything else.
If he's not talking about that, he's talking about the open source movement realizing just how powerful of a force standardization is. No, we don't all use the same OS or hardware platform, but overall, most of us are still using the same stuff - a POSIX operating system, X, KDE or Gnome, StarOffice or KOffice, and so on. POSIX and X aside, I don't think that anyone can tell me that most OSS projects don't at times steal heavily from "industry standard" software. Like a post further up said, we're riding on MS's coattails. And that's what is making the movement grow, too - when showing people what Linux can do, I generally have to show them that it can be just like Windows, hence why even though I usually use WindowMaker for my desktop and console-jockey around, when I'm showing someone else my computer, I generally show them WindowMaker, but also a Gnome desktop I keep around and let them see something that's more akin to what they are used to, with Drag 'n Drop and dekstop integration and registered file types and all that jazz that MS originally brought to the PC.
I think Slackware will always be my lightweight choice. A few weeks ago, I needed to get linux working on a 486DX/33 w/ 8 megs of RAM and no ethernet card and no CD-ROM drive. . . I was left with only one choice. Slack claims you need 16MB of RAM to get the system working well. . . I can't use X on this machine, but otherwise it's fine.
My first question would be, if that were true, why hasn't anyone fixed up the linux manpages and infofiles? They're notoriously bad, and have been so for as long as I have been using Linux (several years).
In many open source projects, documentation, usability, design, interface, etc. are deliberately made bad. Take SourceForge, whose business model appears to be to focus only on power and not bother making the product something that is downright painful to configure, because installation / consulting is one of the few ways an open source software company can make money on their product.
And our unfortunate victim's head is 1.5m off the ground (thus making the distance the coconut falls 33.5m.
After having falling 33.5m, the coconut will be moving at sqrt(2*9.8*33.5), or 25.6 m/s.
If the person's head stops the coconut instantly, the deceleration will be infinite, meaning infinite force. Let's hope this is the case, because our hapless victim will likely die too quickly to feel the pain.
Sadly, the world does not work this way, so let's assume it takes 1/2 cm for the coconut to fully decelerate after hitting his/her head. (It's a soft head.)
Using the equation used above, v^2 = v0^2 + 2a(x - x0), we know that the acceleration will be 656.6m/s/s.
With a 1 kg coconut, that means 656.6 newtons of force into your head. And that's assuming you have a truly soft head.
No, really. A square wheel will ride over a surface made of a series of half-circles (think: one long string of speedbumps) as if it were a round wheel going across a flat surface.
Although I doubt it could bring a system to its knees, and I don't know how you could make a virus continue to propagate itself that way, since I doubt you could get at the webserver from a user account, any exploits using this would likely be limited to trojan status. A single infection, however, could still be just as damaging from the standpoint of a user. Lost data is still lost data.
It seems to me the problem is that it won't spread the music - they're doing their first huge release on a band that they probably know is full of people who aren't going to avoid buying a CD because they can't burn it, and probably won't look at the case long enough to realize it won't play on their computer before they buy it. That way, the hit they'll take from people rejecting the product will be minimal on the first widespread attempt at using it, and they can more effectively argue that this technology will not hinder sales.
I have a feeling it's because many people don't care about internet access enough to pay $70/month for it, and many more (myself included) would rather pay $40/month for a personal DSL connection, considering that if you get a good provider, a DSL connection can be every bit as fast as T1.
As for all the other perks, I doubt joe schmoe has a use for unlimited webspace or sees a need for backed up filespace on the network.
Hotels and dorm rooms are an entirely different issue in many peoples' minds, especially since the costs are very different. . . My school has an OC3 connection, but when you divide the cost among 1,200 students, the cost is much less per person. As for hotels, if they even jump the price by $5/night, that ends up being a potential of $150/month per room, but to the person leasing the room it's still an okay price because if you want internet access bad enough to have it in your hotel room, you're probably willing to pay $5 for a night of it.
in my school, we did a very similar thing, but all our work is done on a Sun450 that the professor has root access on. All we have to do is use RCS or CVS, inital all our comments, and comment what we did when we check in the code, and the professor has a script set to run at the due time that snapshots all the log files.
When it comes to video games, the wave of the future is force feedback, because when we get input from the input devices it makes them more intuitive to use.
When we're actually trying to get something done with computers, we wave our hands in the air because it removes contact from devices, gets rid of all force feedback, and. . . well. . . er. .
Slashdot Mirror
I'd like to add to that:
I don't just want to know how to program with functional languages in the real world. I'd like to be able to link C/C++ code with functional code. I've discovered that functional languages are great for the things imperative languages are terrible for, and vice versa.
If I could link the two together, I might actually succeed in being able to use the Right tools for the Job.
Ahh. . . you must be talking about any one of the host of books falling along the lines of "They Mythical Man-Month" or "Rapid Development"
Really, you're not going to find much outside of journal articles on this subject, since software engineering is still such a fuzzy science. Heck, I don't think anyone has even gotten around to figuring out what it is yet, let alone pinning down how to do it.
So, let's say I have an X program to write and just want to get the damn program written and don't really feel like becoming an X god in the process?
Not the gloss-overs that are put in most books, but an in-depth look at how to get set up w/ graphics and sound programming on linux - something that picks a specific toolkit or toolkits and really talks about it in gory detail. Performance issues, compatibility issues, getting the most out of OpenGL/DRI on that platform, whatever.
No, really. While we're building a 22,000 mile long cable strong enough to hold 20 tons, where are we going to put it? It's probably too big to actually
We can't start stringing it off into space as we build it, because it'd keep tending to fall back towards earth until it were about.
And, once we've figured out all that, how do we get it
That said, it'd be a damn cool thing if we had it, and if a team ever succeeds in constructing one, I'll personally buy a beer for every member of the project.
This probably isn't a document saying what qualifications you must meet to get onto the space station.
This is something saying what qualifications you must meet to get on the space shuttle going to and from the space station (which is also NASA property). This means they effectively do control tourists to the entire space station, not just the US modules.
Until, of course, another country starts up in the space tourism industry. . .
Problems:
"platform agnostic" refers to the kind of agnosticism you get with people who are unsure whether they agree more with the Roman Catholic or Greek Orthodox churches. Every "platform agnostic" system I have seen lately still requires IE, meaning you get a choice between Windows and MacOS.
"Linux-friendly" is something I have seen many businesses put up just to get business, but aren't willing to back up. Take Netgear. I bought a Netgar FA411 PCMCIA ethernet card a while ago. It said it works with Redhat 6.2. Didn't come w/ drivers on the CD (what, did they run out of space!?), so I had to download them. Precompiled module refused to work with any kernel I had, and the source version refused to work with gcc. Called tech support, and they said they didn't provide support for linux installs. Some "linux friendly."
We don't need platform agnosticism and linux friendliness, we need a return to thinking of "industry standard" as an agreed upon standard that everyone can use, rather than a piece of intellectual property that happens to be owned by the company with the most popular product.
Only, that prolly won't happen 'cos we're all so busy playing copycat that we get more excited about figuring out how to decode Quicktime or Windows Media files than we do trying to push the success of the Ogg project or anything else that might save the open source community from its rapidly approaching death at the hands of incompatibility.
I'd add to this that, if I were an admin worth his salt and was told that security was my prime concern, linux would never even cross my mind. If I were to pick any open source unix-like, it would be OpenBSD, hands down. About 10 minutes looking through Bugtraq should be enough explanation for my reasoning.
No offense, Mr. Barr, but the idea of Linux running on sensitive CIA or FBI computers seems patently ludicrous to me.
5) Steve Ballmer will be chased out of Redmond and forced to become a male stripper after he is accused of "penguin lust."
(Sorry, Berke, couldn't resist)
'cos I can't get blackbox to play nice when I'm running it remotely. Same w/ IceWM, Gnome, and KDE.
I have no problem with advertising in and of itself, so long as it doesn't potray itself obtrusively. Magazine ads on their own page are fine. Magazine ads inline in the article are not okay. TV Commerical breaks I can live with. TV ads in the form of bars across the bottom of the screen complete with sounds that make it hard to make out whatever is being said in the program you're watching are not. I feel similarly about the net - banner ads, ok. inline banners, no. popovers, no.
Once again, everyone blames "evil hackers" instead of addressing the real problem or even so much as hinting that sysadmins, or beter yet, PHBs should take part of the blame. So, what else is new?
Sysadmins and PHBs? Where I work, sysadmins and PHBs are probably the only people working to stop the viruses - and believe me, if we could get away with doing the things we want to do to stop them, we would. We already virus scan everything that comes through email, and we've gone so far as to create worms that search peoples' computers for viruses and sends a message to a server to shut off that computer's port when one is found. We remove MSIE and Outlook from all computers, actively discourage their use, and the helpdesk answers all questions related to the two with, "Get Netscape and then we'll talk."
And despite all this, we got hit hard by Code Red and have been trying for months to get all traces of w32.nimda cleared off the network. Problem is, no matter how much we deprecate as much unsafe software as we can, no matter how much we repeat things like, "don't open unexpected attatchments" at workshops, people still do all the krap we tell them not to do, and have more excuses for why they do it than I care to mention.
If we had our way, we would probably have blocked every kind of attatchment from *.mp3 to *.doc by now.
We can sit and whine about script kiddies and place the blame on them, but that won't ever solve the problem. People will always pull shenanigans because they can. We can blame stupid users, but if there's one thing I've learned from mopping up other peoples' shit, it's that people will always do dumb things.
As far as I'm concerned, there's really only one place we can have much of an affect on this problem, and it is the companies that make software. It's a bit like the problem of auto accidents - you can sit and blame drunks and bad drivers, but the only way you're going to save lives is by imposing safety standards on the automobile industry. Similarly, the only way we're going to fix the problem of viruses is by having some sort of accountability process for companies. I can understand not holding the company accountable when an exploit involves using some obscure bug in the program - we can't reasonably expect bugs to never occur in programs. We can hold companies accountable for idiotic things that nobody should do in commercial software, such as 8-bit XOR encrypting passwords - one of the things that nimda exploited to be able to spread across Microsoft networks so easily.
I'm curious if popunder and popunder ads and inline banners really get more hits than the traditional top-o-the-page banner
I've clicked one or two banner ads that attracted my interest. I don't really mind them. On the other hand, I find other forms of ads so irritating that I ignore purely out of spite. And I doubt they really succeed in getting the user's attention any more than other forms of ad - heck, they probably get less of my attention since I usually close ads that open in separate windows before they can load. And I've never actually followed a link on one of them.
So, at least if you limit the research sample group to me, the top-o-the-page banner is infinitely more effective than other ad types, which are generally a complete waste of money.
I agree that control of the code has a lot to do with open source, but I think that Bill is hitting on the reasons why OSS has become popular rather than te reasons why it exists at all.
The GNU project started in 1983 - after the first MS operating system, but long before Microsoft because the industry-dominating force it is right now. Bill surely knows this, and from that alone he can't claim any responsibility for the ideological roots of Open Source.
He's either talking about kids like me, who got into OSS to be different. I'm staying with open source for many other reasons, but if it weren't for the fact that Linux is the only way to try something different in a world where the only choices are Windows or MacOS (contrast with the days of MS-DOS, CP/M, Apples, Amigas, c64s, Spectrums, etc.) And before anyone jumps up and down shouting "BSD! BeOS!," sit down. The fact of life is, most people hear about Linux long before they hear about anything else.
If he's not talking about that, he's talking about the open source movement realizing just how powerful of a force standardization is. No, we don't all use the same OS or hardware platform, but overall, most of us are still using the same stuff - a POSIX operating system, X, KDE or Gnome, StarOffice or KOffice, and so on. POSIX and X aside, I don't think that anyone can tell me that most OSS projects don't at times steal heavily from "industry standard" software. Like a post further up said, we're riding on MS's coattails. And that's what is making the movement grow, too - when showing people what Linux can do, I generally have to show them that it can be just like Windows, hence why even though I usually use WindowMaker for my desktop and console-jockey around, when I'm showing someone else my computer, I generally show them WindowMaker, but also a Gnome desktop I keep around and let them see something that's more akin to what they are used to, with Drag 'n Drop and dekstop integration and registered file types and all that jazz that MS originally brought to the PC.
I think Slackware will always be my lightweight choice. A few weeks ago, I needed to get linux working on a 486DX/33 w/ 8 megs of RAM and no ethernet card and no CD-ROM drive. . . I was left with only one choice. Slack claims you need 16MB of RAM to get the system working well. . . I can't use X on this machine, but otherwise it's fine.
My first question would be, if that were true, why hasn't anyone fixed up the linux manpages and infofiles? They're notoriously bad, and have been so for as long as I have been using Linux (several years).
In many open source projects, documentation, usability, design, interface, etc. are deliberately made bad. Take SourceForge, whose business model appears to be to focus only on power and not bother making the product something that is downright painful to configure, because installation / consulting is one of the few ways an open source software company can make money on their product.
Let's assume the coconut weighs 1kg.
And our unfortunate victim's head is 1.5m off the ground (thus making the distance the coconut falls 33.5m.
After having falling 33.5m, the coconut will be moving at sqrt(2*9.8*33.5), or 25.6 m/s.
If the person's head stops the coconut instantly, the deceleration will be infinite, meaning infinite force. Let's hope this is the case, because our hapless victim will likely die too quickly to feel the pain.
Sadly, the world does not work this way, so let's assume it takes 1/2 cm for the coconut to fully decelerate after hitting his/her head. (It's a soft head.)
Using the equation used above, v^2 = v0^2 + 2a(x - x0), we know that the acceleration will be 656.6m/s/s.
With a 1 kg coconut, that means 656.6 newtons of force into your head. And that's assuming you have a truly soft head.
No, really. A square wheel will ride over a surface made of a series of half-circles (think: one long string of speedbumps) as if it were a round wheel going across a flat surface.
Although I doubt it could bring a system to its knees, and I don't know how you could make a virus continue to propagate itself that way, since I doubt you could get at the webserver from a user account, any exploits using this would likely be limited to trojan status.
A single infection, however, could still be just as damaging from the standpoint of a user. Lost data is still lost data.
It seems to me the problem is that it won't spread the music - they're doing their first huge release on a band that they probably know is full of people who aren't going to avoid buying a CD because they can't burn it, and probably won't look at the case long enough to realize it won't play on their computer before they buy it. That way, the hit they'll take from people rejecting the product will be minimal on the first widespread attempt at using it, and they can more effectively argue that this technology will not hinder sales.
I have a feeling it's because many people don't care about internet access enough to pay $70/month for it, and many more (myself included) would rather pay $40/month for a personal DSL connection, considering that if you get a good provider, a DSL connection can be every bit as fast as T1.
As for all the other perks, I doubt joe schmoe has a use for unlimited webspace or sees a need for backed up filespace on the network.
Hotels and dorm rooms are an entirely different issue in many peoples' minds, especially since the costs are very different. . . My school has an OC3 connection, but when you divide the cost among 1,200 students, the cost is much less per person. As for hotels, if they even jump the price by $5/night, that ends up being a potential of $150/month per room, but to the person leasing the room it's still an okay price because if you want internet access bad enough to have it in your hotel room, you're probably willing to pay $5 for a night of it.
in my school, we did a very similar thing, but all our work is done on a Sun450 that the professor has root access on. All we have to do is use RCS or CVS, inital all our comments, and comment what we did when we check in the code, and the professor has a script set to run at the due time that snapshots all the log files.
Just think about how much more healthy the online pr0n industry is, thanks to Ray Tomlinson.
When it comes to video games, the wave of the future is force feedback, because when we get input from the input devices it makes them more intuitive to use.
When we're actually trying to get something done with computers, we wave our hands in the air because it removes contact from devices, gets rid of all force feedback, and. . . well. . . er. .