So he's saying that we'll have self-aware robots in 23 years. This seems pretty unrealistic to me, being that we have yet to design a computer that has demonstrated anything close to human conciousness.
That does seem ludicrous to me as well. However, I'm inclined to believe that in order to develop human conciousness in a machine, you need to give the machine comparable facilities/methods for interpretting the world. Eg. two eyes, sense of touch-smell-sight-taste-hearing, two arms two legs, etc.
And why humanoid? Seems like the current factory robots (massive robots at the auto factories, for example) are doing pretty well without a humanoid design.
Modularity? Mass produced? Instead of having one machine capable of doing X you'd have 1000 able to do A, B, C, D, X,....
Legal departments aren't challenging shrink-wrap licenses... ... because they feel they're not really enforceable contracts. ... because they realize that most of the time they're dealing with a powerful monopoly.
Call me cynical, but I think the reasons are deeper than this. Most individuals I know would scream bloody murder when shown the EULA and all its implications. However corporate entities are more interested in securing control over their own products. This is especially true in content/service/software oriented companies. If they go into a huge legal battle against MS over EULA agreements, this means they can not issue similar agreements of their own. Not only that, it sets a precident for consumer rights over corporate rights. Not to mention that it will hurt their relations with MS, a company on which much of their business may rely. So while every individual in the IT Dept may be screaming bloody murder, where is the incentive for a corporate entity to take MS to court?
What we need is a completely unrelated company like Dole or Florida Orange to challenge the agreement. But they probably don't care so much and would expect a more natural MS Enemy (AOL) to shoulder the burden of a lawsuit. Companies like AOL, would have no interest in challenging corporate rights until they directly infringe against their business plans.
Of course I don't know jack about these kind of operations, but you would think constantly reorganizing the Nasa budget would result in untold amounts of wasted cash. Many projects take a long time to go from development to realization. When you are constantly cutting back and reorganizing resources, you are wasting the moeny and effort already invested. Nasa needs smarter, better, cheaper, but they also need to have guarantees that projects they start will be funded throughout their proposed duration.
True enough, but forgetting about Linux for a sec, Windows 98's biggest competitor was Windows 95. Windows 2000's biggest competitor was NT (2000 was more about replacing NT than it was about replacing *nix). In order to keep a strong incomming revenue stream, any software vendor has to increase the feature set enough to attract people to a new release. As someone who's finally about to get rid of 98, I can say there's some steep competition from older version software (...of course maybe there wouldn't be with more competition, but who can say for sure).
Re:Creative adaptation
on
Arguing A.I.
·
· Score: 1
I agree with you in spirit, but I don't think the boy mentioned in the parent or the checkers playing program are necessarily exhibitting Creative Adaptation. They both learned the rules by playing or watching games being played. This is roughly as creative as holding a rock letting it go and predicting it will drop to the ground (based on X years living on a planet, under the rule of gravity).
Within 50 years, there will be a computer that will pass the Turing Test
Maybe, the Turing test is kind of vague and rather dependent on the tester. A few months ago, there was reported to be a computer that passed a "baby" version of the Turing test (some of which was marketting hype). At any rate, I bet your right in that within 50 yrs there will be a Turing test a computer can pass.
after that (100 years total), computers will be able to parse speech flawlessly
Humans can't even do that : )... seriously though. I believe that if this kind of thing can be done, it will be done much sooner than that. We are already fairly close, so many of the problems with speech recognition today are due to accents/variations (also sentence fragments/pauses). These are personalization issues that feel like they will always be around.
Within 50 years of that (150 years total) we'll have computers that can respond to voice commands like in Star Trek. The computer will not only understand the syntax of language, but it will be able to determine, on its own, the difference between a question asked in conversation and a question asked to the computer in conversation.
Computer: Warp Speed! 150 years is an awfully long time away. Think what things were like in the 1850's. This could happen sooner. It's important to note that this isn't strictly a linguistic problem. The difference between a question to your friend Bill and one to the computer can be as simple as a look, a change in tone, or a change in subject. In order to do this, a computer would have to watch a person's body language and facial expressions as well as listen to what they are saying. In addition it would probably help the computer immensely if it had a physical representation (android/hologram/bowl of petunas) that the human could address in a human-like fashion. My guess is that this kind of technology will mature once we have halfway decent voice recognition, decent facial recognition, and some kind of entity that can be interacted with (optional). Something like this will probably exist in 25-75 years. I wouldn't expect anything mainstream for another 15-25+ years after that. The trick for commercialization is this would never be a small computer in a box. There would be lots of little components like cameras and possibly biometric devices.
Re:I'm doubtful
on
Arguing A.I.
·
· Score: 2, Informative
Is it really something called "intelligence" that can read a face? Nobody every wrote an academic tome called "Advanced Face Reading" that details an intellectual process to go thru to read an opponent that might be turned into some algorithm.
Actually, this exact style of intelligence has been the subject of many papers. See MIT's Sociable Machines site for more information. This kind of thing is a very hot topic in AI, not ot mention psychology or even some areas of linguistics.
Another potential type of product placement that I don't think would be so bad would be user activated stuff. It would be kind of cool if after whacking the coke machine with a crowbar, you heard the sound of a can opening and your character saying, "ah, crisp refreshing coke." Or if your character hovered in front of a "gap" store for more than a few seconds, he might say, "I wish I had a dress like that." It would have to be comedic and the action would have to be deliberately triggered.... sort of like Easter egg ads.
After only a few hundred years of thorough genetic mixing, I think it's a little early to call the game.
There's always genetic engineering, resistances to potential new diseases and environmental changes. Let's not forget the potential for speciation via space travel. If it takes 7 years, at light speed, to get from Star System A to Star System B, you can bet there won't be a lot of intermixing between those populations.
And even if humans are slowly reaching the point where the weaker don't get killed off. We are still selecting who we mate with using criteria like Looks, Intelligence, Success, Looks, Sense of Humor, and Looks. So who's to say this kind of social selection won't be a major evolutionary force?
how long before people get tired enough of this crap to start watching everything on TiVO/Replay/etc.?
I love commercials. Consider what might happen if everyone skipped all the ads via Tivo. Television Revenue would slip because commercials would become less and less valuable. So now you've got an industry whose entire business model is based on selling ads that everyone is skipping.
How do you adapt this business model so you can continue to sell advertisements in a profitable manner. Simple, you incorporate the advertisements in the actual shows via product placement. Rachel and Ross (friends) will only drink crisp refreshing Coke and Diet Coke from now on. The Enterprise will be powered by Sun systems. Maybe instead of the scifi channel logo in the lower right, we'll have a dancing can of Budweiser. The CNN ticker will read, "... tragic crash killing 247, in other news McDonalds is having a sale on Big Macs buy one get one free, get yours now..."
I would much much rather have additoinal commercials than any of the above.
I went to OSU. Whenever this question was posed to the students, one of the biggest requests always involved mice in the labs. Most computer labs on campus used old mice and had no mouse pads. Every mouse was perpetually in need of cleaning. Before you go out and spend lots of money on anything innovative, make sure all the basic stuff works well... and if you can, get Optical Mice so no one ever has to clean another lab rat again.
Other suggestions: Improve Documentation: One of the biggest questions at CS-OSU was, "How do I get an X session?"
Improve network infrastructure: This can always be improved.
Improve WebCT/remote learning: WebCT/Remote learning tools typically need improvement. Usually, the biggest problem is not the software but the Teachers who are unfamiliar with it but required to teach course through it. Student aids for these teachers are not always adequate.
Wireless: This may be a bit much, but the students would love it if you could get it working.
Subsidized/Discounted Software: At OSU we had the Buckeye Bundle. It included every MS product (any OS, any Office, Studio) for $100. We also had a Software to Go website where we could download some stuff like SSH for free. This was very popular with me and my friends.
Well sure, often if you can get something online, you can get it for free online... but the question is how long is it going to take you to find what you are looking for?
Specialized services like Lexis Nexis (used to work there [in the mail room]) can greatly reduce the time it takes to find certain kinds of information. The information may be freely available elsewhere in 10 billion different places. Just think of it as paying to save your own time.
We could all probably save money riding the bus, but driving to work in a ferrari is a little faster.
I don't know about anyone else, but I've never trusted online bargains. Once I was trying to order some parts off of pricewatch and came across a series of companies that were lowering their prices at night when the store was closed and raising them when the stores opened. The only way to get the pricewatch price was to call and mention the current price. However there was no way to place an order over the phone when the store was closed. When I asked about getting last midnight's price, they refused and would not comment on their pricing scheme.
Another time, I tried ordering a computer kit online (pricewatch), I was about to give him my credit card number when he suggested I upgrade to their special CPU cooler which was "better" than the one that came with their product. I asked him if there was something wrong with the cpu cooler that came with it and he said that while the cpu cooler that came with it was perfectly good, it only consisted of a fan. I spoke with him a few minutes to try and figure out exactly what he meant and determined the actual cpu cooler included in his 1.33GHz Athlon kit was somehow without a heatsink. Click.
Yet another time, in a fit of insanity, I decided to order a new video card online saving about $60. Everything went smoothly, I ordered from a company in California for about $270 (US). A week and a half later, it arrived with a return address somewhere in Brazil. I have no idea what I would of had to do had the product been defective (thank god it wasn't), but I'm sure it would have been hell.
Perhaps, bad experiences online have driven consumers away from these vendors and forced the online retailers to raise prices. Maybe most of the bargains out there never really existed anyway. As a kid scrounging for money to buy stuff, it sometimes makes sense. As a professional with a decent salary, buying online often isn't worth the risk.
AOL/Time-Warner knows its business quite well, and its business has nothing to do with software and everything to do with charging people for access to content they desire.
This is only partially true. The main distinguishing feature AOL has over other ISPs is in fact their "easy to use" software. To take this another step, AOL has many software/content companies that don't charge for use at all. Instead they make their money through outside advertising or through the promotion of other AOLT products. ICQ, Digital City, iPlanet, and Mapquest just to name a few. (for more http://www.aoltimewarner.com/about/index.html). Ultimately, I think they are interested in controlling both content and distribution (software).
This means AOL has "network appliance" in their heads.
I'm not so sure about this. A year or two ago, AOL/Gateway came out with a network appliance (http://www.idg.net/idgns/2000/04/05/AOLGatewayToO fferInstantAOL.shtml). It failed for a variety of reasons... but mostly because people/world weren't ready for devices like it.
Perhaps this is just me talking out of my ass, but maybe AOL really wants to release their own consumer branded OS to compete with Microsoft. Normal people like my mom and dad are fed up with buggy Microsoft products, but they don't see anything else to switch to. They would never try an OS made by some little known Linux startup (aka RedHat), but they would consider trying something made by AOL. More to the point, there are about 30 million AOL subscribers out there that also might consider a switch (about 100 million AIM users). Add to that the close ties AOL has had with certain manufacturers (like Gateway) and the poor relationships MS has developed with the same people, who knows maybe it could be the beginning of true competition.
The problem is more one of diversity. If you place 500 million machines out in the wild all running the same software. Then any exploits found in that software will leave all those machines vulnerable. It doesn't matter if its Windows or Linux.
Since when is choice bad? Well then I also think drving a car is a complicated thing there are so many models to choose from when buying one.
Choice isn't bad, but having many choices makes making decisions more difficult. My point was, for many things, linux bombards you with choices while windows provides few or none... You have to spend more time figuring out what is best for you. In this sense, Windows is simpler than Linux. Bad/Worse/Better are ambiguous loaded terms. Simpler != Better.
I've lastly installet some dual boot machines for friends. (Win98, SuSE 7.3) SuSE was always less work, you just pressed enter all the time, while Win98 had bath problems with the graphic card ane th CDROM.
I've never used SUSE. My experience was with Debian and RedHat, but my point about complexity in many Linux installs was "if the user wants to do any customizing." In your example where you "Press Enter" all the time you are not doing any customizing. Furthermore, you are comparing the latest version of SUSE to a four year old/3 generation Old Windows OS. That's not exactly very sportsmanlike.
Until some exceptions most (hardware) is (supported), or not more than having on windows. Especially try to get windows drivers fop newer windors for older hardware. Or the other way around try to get drivers for older windows. I still not have a windows 3.11 PCMCIA driver for my little 386 notebook. Nor do I get windows NT drivers for my old SB CDROM.
Most of my hardware/software was manufactured in the past 5 years so I can't really speak to your concerns about not being able to get PCMCIA drivers for windows 3.1 or old SB CDROM drivers for NT. When I go out to the local electronics store to buy a consumer grade modem, webcam, scanner, hard drive, motherboard, or anything else, it always works with Windows, but not always with Linux. Even if it is supported by Linux, it often won't say so on the box. Drivers for new hardware in Linux will always lag behind drivers for Windows until Manufacturers see greater profit potential in Linux. Furthermore, you often won't get the full power of your hardware. I've had to not use or not fully use a great deal of hardware in Linux because it was not fully supported (ATA66/100 channels, video cards, sound cards, tv cards, etc).
You can also write
chmod a=rwx,u=rx,o=rx public_html
hiding how flags are stored inside, of course it's more to write thats why most use the octal form.
The point was that Linux does a poor job abstracting normal operations. In Linux the easiest or "best" way of doing something often involves lots of uneccessary knowledge. It's like programming in C. You have access to all sorts of things that are denied in a higher language like Java, but at the same time you have to learn how to do everything at a very nitty gritty level.
I'm not saying the Linux is worse than Windows nor am I saying it is better. I use and enjoy Linux daily.
Give me a bigger case! I want room for my 8 IDE devices. Give me more than 2 mm between my each hard drive. I want more space between my actively cooled GeForce2 and the rest of the PCI cards, let's get some circulation in there.
But most of all, can't we do something about all the damn wires. Run power cables inside or along the walls of the case and provide a small connector where each drive goes. For the love of god, get rid of ribbon cables. Do Not Place the 2-4 IDE channels and the Floppy plug next to each other on the motherboard. How am I supposed to plug a cable in there with all the other cables in my way? Do NOT place the memory underneath the mess of IDE cables. Put it between the processor and AGP slot or anywhere that makes it so you don't have to unplug all your drives to add 128 MB of ram.
Here's an idea... Is it possible to use both the front and back side of a motherboard? If half was on the front and half on the back everything would be much easier to reach. Of course the cost of this would likely be prohibitive.
I'm getting basic cable $30, broadband $35, and local phone $21 all from one company and after all the regulatory fees. Where's the other $140 come from?
My guess would be cellular phone, long distance package, caller id, call waiting, voicemail, line protection, giant smurfs, and a really expensive cable package... either way, it's been inflated.
I explained to a newcomer in one hour how to surf, browse, send emails, access a newsnet server, and play little games on it, don't talk about nonsense crap for having to learn weeks.
Linux has a steep learning curve. Period.
1) When you decide to try out Linux, the first thing you realize is you have a lot of choices. Distribution, Window Manager, Sound Support, etc. Choices are great but thoroughly researching them requires a great deal of time. With Windows, you only have to chose between the latest DOS or NT based system.
2) With Linux, you need to figure out if your hardware is supported. Even if it is supported, it is certain that the software (that you paid for) packaged with it isn't. Since Windows is the consumer leader, hardware support for consumer equipment is not a big issue. After building 5 or so completely different windows systems, I can say it has never been a problem for me or anyone I personally know.
3) The actual installation process is much more complicated for most Linux distributions than it is for Windows. My experience is with RedHat62 and Potato. In both cases, if the user wants to do any customizing, he is confronted with complicated stuff like partitioning hard drives and selecting from long lists of obsficated modules.
4) Portability of knowledge. KDE, Gnome, may be great, but what happens when you have to use an alternative system. As far as I can tell, the only way you can truly master Linux is by mastering the Unix command line and everything that it entails. Let's just say RedHat's Gnome package comes with a special user manager program or a task scheduler. Those programs may be completely different or absent on a Debian KDE install. The only way you can learn to manage users or schedule tasks on both systems no matter what is to learn the syntax of adduser and crontab. Learning the command line for everything is a pain in the ass. Since you have far fewer choices with Windows, everything is more or less the same at a fairly intuitive graphical level.
5) Learning the Command line is hard. Command line interfaces are typically designed for brevity of keystrokes. This makes them very user friendly to people who already understand them and very difficult for people who are learning them. Case and point: VI (I use it). Short switches (cp -R, crontab -e) make everything more complicated... you have to remember -R means recursive, -e is for edit... and they don't stay the same for all commands. Or as my mom says, what the hell is recursion? Also, many commands have funny abbreviated names that you have to look up. This is complicated.
6) Linux does a poor job of abstracting the internal workings. For example, when I go to change permissions on a file, I type chmod -R 755 public_html. Changing permissions on a file should NEVER require binary arithmetic or any knowledge about how the permission is implemented. Similarly, when you schedule a task using crontab, you edit or create a special cron file using crontab -e. The user shouldn't have to know a special month day year hour min syntax for scheduling a program, nor should he even have to be aware a special file containing his scheduled programs even exists. This level of knowledge is uneccessary and makes learning the system more difficult. In contrast, changing permissions is done with a right click in windows (and with ACLS) and scheduling is handled through task scheduler in Windows. Please take a moment to compare those names: crontab, task scheduler. Which is easier to remember? To use?
For me it boils down to three key points. Linux has lots of choices that give you power but make figuring out the best way to do things very complex giving it a steep learning curve. Because Linux is a minority holder in the consumer market, it has relatively poor support for consumer hardware (winmodems, sblive's, the latest ATA100 controllers, etc). Finally, Linux does not hide information well and command line interfaces while powerful, are not intuitive. Oh yeah, and most package managers are pure Hell (read DIE RPM, DIE)!
When you explained to a newcomer how to browse the web, use email/news, and play a few games, you weren't telling them how to use Linux. You were telling them how to surf, use email/news, and play a few games. In other words you taught them Opera and Pine and SameGnome NOT LINUX. If they ever had to set that stuff up themselves or god forbid install/setup a different browser/mail program, how easy would that be for them? You gave them specific programs they had to use in order to get around. In essence you took away the single greatest advantage of Linux, choice.
Brief disclaimer: I like Linux. I use it as a server and do most personal development projects there. Linux is many things, but easier than Windows it is not. PS. I have used KDE.
Slashdot Mirror
So he's saying that we'll have self-aware robots in 23 years. This seems pretty unrealistic to me, being that we have yet to design a computer that has demonstrated anything close to human conciousness.
....
That does seem ludicrous to me as well. However, I'm inclined to believe that in order to develop human conciousness in a machine, you need to give the machine comparable facilities/methods for interpretting the world. Eg. two eyes, sense of touch-smell-sight-taste-hearing, two arms two legs, etc.
And why humanoid? Seems like the current factory robots (massive robots at the auto factories, for example) are doing pretty well without a humanoid design.
Modularity? Mass produced? Instead of having one machine capable of doing X you'd have 1000 able to do A, B, C, D, X,
Legal departments aren't challenging shrink-wrap licenses...
... because they feel they're not really enforceable contracts.
... because they realize that most of the time they're dealing with a powerful monopoly.
Call me cynical, but I think the reasons are deeper than this. Most individuals I know would scream bloody murder when shown the EULA and all its implications. However corporate entities are more interested in securing control over their own products. This is especially true in content/service/software oriented companies. If they go into a huge legal battle against MS over EULA agreements, this means they can not issue similar agreements of their own. Not only that, it sets a precident for consumer rights over corporate rights. Not to mention that it will hurt their relations with MS, a company on which much of their business may rely. So while every individual in the IT Dept may be screaming bloody murder, where is the incentive for a corporate entity to take MS to court?
What we need is a completely unrelated company like Dole or Florida Orange to challenge the agreement. But they probably don't care so much and would expect a more natural MS Enemy (AOL) to shoulder the burden of a lawsuit. Companies like AOL, would have no interest in challenging corporate rights until they directly infringe against their business plans.
Work in a small company 50 ppl, you will likely work for a smart boss.
Where is the Heart of the Internet? Who else to ask other than Al Gore?
My guess would be he put it in a "lockbox" probably inside his glove box. Only Al, the joint Chiefs of Staff, and Tipper know for sure.
Of course I don't know jack about these kind of operations, but you would think constantly reorganizing the Nasa budget would result in untold amounts of wasted cash. Many projects take a long time to go from development to realization. When you are constantly cutting back and reorganizing resources, you are wasting the moeny and effort already invested. Nasa needs smarter, better, cheaper, but they also need to have guarantees that projects they start will be funded throughout their proposed duration.
True enough, but forgetting about Linux for a sec, Windows 98's biggest competitor was Windows 95. Windows 2000's biggest competitor was NT (2000 was more about replacing NT than it was about replacing *nix). In order to keep a strong incomming revenue stream, any software vendor has to increase the feature set enough to attract people to a new release. As someone who's finally about to get rid of 98, I can say there's some steep competition from older version software (...of course maybe there wouldn't be with more competition, but who can say for sure).
I agree with you in spirit, but I don't think the boy mentioned in the parent or the checkers playing program are necessarily exhibitting Creative Adaptation. They both learned the rules by playing or watching games being played. This is roughly as creative as holding a rock letting it go and predicting it will drop to the ground (based on X years living on a planet, under the rule of gravity).
Within 50 years, there will be a computer that will pass the Turing Test
... seriously though. I believe that if this kind of thing can be done, it will be done much sooner than that. We are already fairly close, so many of the problems with speech recognition today are due to accents/variations (also sentence fragments/pauses). These are personalization issues that feel like they will always be around.
Maybe, the Turing test is kind of vague and rather dependent on the tester. A few months ago, there was reported to be a computer that passed a "baby" version of the Turing test (some of which was marketting hype). At any rate, I bet your right in that within 50 yrs there will be a Turing test a computer can pass.
after that (100 years total), computers will be able to parse speech flawlessly
Humans can't even do that : )
Within 50 years of that (150 years total) we'll have computers that can respond to voice commands like in Star Trek. The computer will not only understand the syntax of language, but it will be able to determine, on its own, the difference between a question asked in conversation and a question asked to the computer in conversation.
Computer: Warp Speed! 150 years is an awfully long time away. Think what things were like in the 1850's. This could happen sooner. It's important to note that this isn't strictly a linguistic problem. The difference between a question to your friend Bill and one to the computer can be as simple as a look, a change in tone, or a change in subject. In order to do this, a computer would have to watch a person's body language and facial expressions as well as listen to what they are saying. In addition it would probably help the computer immensely if it had a physical representation (android/hologram/bowl of petunas) that the human could address in a human-like fashion. My guess is that this kind of technology will mature once we have halfway decent voice recognition, decent facial recognition, and some kind of entity that can be interacted with (optional). Something like this will probably exist in 25-75 years. I wouldn't expect anything mainstream for another 15-25+ years after that. The trick for commercialization is this would never be a small computer in a box. There would be lots of little components like cameras and possibly biometric devices.
Is it really something called "intelligence" that can read a face? Nobody every wrote an academic tome called "Advanced Face Reading" that details an intellectual process to go thru to read an opponent that might be turned into some algorithm.
Actually, this exact style of intelligence has been the subject of many papers. See MIT's Sociable Machines site for more information. This kind of thing is a very hot topic in AI, not ot mention psychology or even some areas of linguistics.
Another potential type of product placement that I don't think would be so bad would be user activated stuff. It would be kind of cool if after whacking the coke machine with a crowbar, you heard the sound of a can opening and your character saying, "ah, crisp refreshing coke." Or if your character hovered in front of a "gap" store for more than a few seconds, he might say, "I wish I had a dress like that." It would have to be comedic and the action would have to be deliberately triggered. ... sort of like Easter egg ads.
After only a few hundred years of thorough genetic mixing, I think it's a little early to call the game.
There's always genetic engineering, resistances to potential new diseases and environmental changes. Let's not forget the potential for speciation via space travel. If it takes 7 years, at light speed, to get from Star System A to Star System B, you can bet there won't be a lot of intermixing between those populations.
And even if humans are slowly reaching the point where the weaker don't get killed off. We are still selecting who we mate with using criteria like Looks, Intelligence, Success, Looks, Sense of Humor, and Looks. So who's to say this kind of social selection won't be a major evolutionary force?
Better yet, buy a laptop.
http://www.iceboxllc.com/flipscreen/index.html
Manufacturers Suggested Retail Price $2,995.00
Aaaaaaaaaaaaaaaaaaaaaaaaaaaahhhhhhhhhhhhhh!!!!
Man, just buy a $1000 dell, stick it out of site in a cabinet, and hookup an lcd/washable keyboard.
It looks cool and all, but $3K? No way.
how long before people get tired enough of this crap to start watching everything on TiVO/Replay/etc.?
I love commercials. Consider what might happen if everyone skipped all the ads via Tivo. Television Revenue would slip because commercials would become less and less valuable. So now you've got an industry whose entire business model is based on selling ads that everyone is skipping.
How do you adapt this business model so you can continue to sell advertisements in a profitable manner. Simple, you incorporate the advertisements in the actual shows via product placement. Rachel and Ross (friends) will only drink crisp refreshing Coke and Diet Coke from now on. The Enterprise will be powered by Sun systems. Maybe instead of the scifi channel logo in the lower right, we'll have a dancing can of Budweiser. The CNN ticker will read, "... tragic crash killing 247, in other news McDonalds is having a sale on Big Macs buy one get one free, get yours now..."
I would much much rather have additoinal commercials than any of the above.
I went to OSU. Whenever this question was posed to the students, one of the biggest requests always involved mice in the labs. Most computer labs on campus used old mice and had no mouse pads. Every mouse was perpetually in need of cleaning. Before you go out and spend lots of money on anything innovative, make sure all the basic stuff works well... and if you can, get Optical Mice so no one ever has to clean another lab rat again.
Other suggestions:
Improve Documentation: One of the biggest questions at CS-OSU was, "How do I get an X session?"
Improve network infrastructure: This can always be improved.
Improve WebCT/remote learning: WebCT/Remote learning tools typically need improvement. Usually, the biggest problem is not the software but the Teachers who are unfamiliar with it but required to teach course through it. Student aids for these teachers are not always adequate.
Wireless: This may be a bit much, but the students would love it if you could get it working.
Subsidized/Discounted Software: At OSU we had the Buckeye Bundle. It included every MS product (any OS, any Office, Studio) for $100. We also had a Software to Go website where we could download some stuff like SSH for free. This was very popular with me and my friends.
If you can get it, you can get it for free
Well sure, often if you can get something online, you can get it for free online... but the question is how long is it going to take you to find what you are looking for?
Specialized services like Lexis Nexis (used to work there [in the mail room]) can greatly reduce the time it takes to find certain kinds of information. The information may be freely available elsewhere in 10 billion different places. Just think of it as paying to save your own time.
We could all probably save money riding the bus, but driving to work in a ferrari is a little faster.
At the Grand Canyon:
Buy Grand Canyon keychains!
At the Bank:
Make free money! Refinance now!
In residential areas:
(picture of sexy girl) Enlarge your penis now!
At Disney World:
Visit the gift shop now!
I don't know about anyone else, but I've never trusted online bargains. Once I was trying to order some parts off of pricewatch and came across a series of companies that were lowering their prices at night when the store was closed and raising them when the stores opened. The only way to get the pricewatch price was to call and mention the current price. However there was no way to place an order over the phone when the store was closed. When I asked about getting last midnight's price, they refused and would not comment on their pricing scheme.
Another time, I tried ordering a computer kit online (pricewatch), I was about to give him my credit card number when he suggested I upgrade to their special CPU cooler which was "better" than the one that came with their product. I asked him if there was something wrong with the cpu cooler that came with it and he said that while the cpu cooler that came with it was perfectly good, it only consisted of a fan. I spoke with him a few minutes to try and figure out exactly what he meant and determined the actual cpu cooler included in his 1.33GHz Athlon kit was somehow without a heatsink. Click.
Yet another time, in a fit of insanity, I decided to order a new video card online saving about $60. Everything went smoothly, I ordered from a company in California for about $270 (US). A week and a half later, it arrived with a return address somewhere in Brazil. I have no idea what I would of had to do had the product been defective (thank god it wasn't), but I'm sure it would have been hell.
Perhaps, bad experiences online have driven consumers away from these vendors and forced the online retailers to raise prices. Maybe most of the bargains out there never really existed anyway. As a kid scrounging for money to buy stuff, it sometimes makes sense. As a professional with a decent salary, buying online often isn't worth the risk.
AOL/Time-Warner knows its business quite well, and its business has nothing to do with software and everything to do with charging people for access to content they desire.
O fferInstantAOL.shtml). It failed for a variety of reasons... but mostly because people/world weren't ready for devices like it.
This is only partially true. The main distinguishing feature AOL has over other ISPs is in fact their "easy to use" software. To take this another step, AOL has many software/content companies that don't charge for use at all. Instead they make their money through outside advertising or through the promotion of other AOLT products. ICQ, Digital City, iPlanet, and Mapquest just to name a few. (for more http://www.aoltimewarner.com/about/index.html). Ultimately, I think they are interested in controlling both content and distribution (software).
This means AOL has "network appliance" in their heads.
I'm not so sure about this. A year or two ago, AOL/Gateway came out with a network appliance (http://www.idg.net/idgns/2000/04/05/AOLGatewayTo
Perhaps this is just me talking out of my ass, but maybe AOL really wants to release their own consumer branded OS to compete with Microsoft. Normal people like my mom and dad are fed up with buggy Microsoft products, but they don't see anything else to switch to. They would never try an OS made by some little known Linux startup (aka RedHat), but they would consider trying something made by AOL. More to the point, there are about 30 million AOL subscribers out there that also might consider a switch (about 100 million AIM users). Add to that the close ties AOL has had with certain manufacturers (like Gateway) and the poor relationships MS has developed with the same people, who knows maybe it could be the beginning of true competition.
Good points. I can't argue with that.
The problem is more one of diversity. If you place 500 million machines out in the wild all running the same software. Then any exploits found in that software will leave all those machines vulnerable. It doesn't matter if its Windows or Linux.
Since when is choice bad? Well then I also think drving a car is a complicated thing there are so many models to choose from when buying one.
Choice isn't bad, but having many choices makes making decisions more difficult. My point was, for many things, linux bombards you with choices while windows provides few or none... You have to spend more time figuring out what is best for you. In this sense, Windows is simpler than Linux. Bad/Worse/Better are ambiguous loaded terms. Simpler != Better.
I've lastly installet some dual boot machines for friends. (Win98, SuSE 7.3) SuSE was always less work, you just pressed enter all the time, while Win98 had bath problems with the graphic card ane th CDROM.
I've never used SUSE. My experience was with Debian and RedHat, but my point about complexity in many Linux installs was "if the user wants to do any customizing." In your example where you "Press Enter" all the time you are not doing any customizing. Furthermore, you are comparing the latest version of SUSE to a four year old/3 generation Old Windows OS. That's not exactly very sportsmanlike.
Until some exceptions most (hardware) is (supported), or not more than having on windows. Especially try to get windows drivers fop newer windors for older hardware. Or the other way around try to get drivers for older windows. I still not have a windows 3.11 PCMCIA driver for my little 386 notebook. Nor do I get windows NT drivers for my old SB CDROM.
Most of my hardware/software was manufactured in the past 5 years so I can't really speak to your concerns about not being able to get PCMCIA drivers for windows 3.1 or old SB CDROM drivers for NT. When I go out to the local electronics store to buy a consumer grade modem, webcam, scanner, hard drive, motherboard, or anything else, it always works with Windows, but not always with Linux. Even if it is supported by Linux, it often won't say so on the box. Drivers for new hardware in Linux will always lag behind drivers for Windows until Manufacturers see greater profit potential in Linux. Furthermore, you often won't get the full power of your hardware. I've had to not use or not fully use a great deal of hardware in Linux because it was not fully supported (ATA66/100 channels, video cards, sound cards, tv cards, etc).
You can also write chmod a=rwx,u=rx,o=rx public_html hiding how flags are stored inside, of course it's more to write thats why most use the octal form.
The point was that Linux does a poor job abstracting normal operations. In Linux the easiest or "best" way of doing something often involves lots of uneccessary knowledge. It's like programming in C. You have access to all sorts of things that are denied in a higher language like Java, but at the same time you have to learn how to do everything at a very nitty gritty level.
I'm not saying the Linux is worse than Windows nor am I saying it is better. I use and enjoy Linux daily.
Smaller footprint? Huh?
Give me a bigger case! I want room for my 8 IDE devices. Give me more than 2 mm between my each hard drive. I want more space between my actively cooled GeForce2 and the rest of the PCI cards, let's get some circulation in there.
But most of all, can't we do something about all the damn wires. Run power cables inside or along the walls of the case and provide a small connector where each drive goes. For the love of god, get rid of ribbon cables. Do Not Place the 2-4 IDE channels and the Floppy plug next to each other on the motherboard. How am I supposed to plug a cable in there with all the other cables in my way? Do NOT place the memory underneath the mess of IDE cables. Put it between the processor and AGP slot or anywhere that makes it so you don't have to unplug all your drives to add 128 MB of ram. Here's an idea... Is it possible to use both the front and back side of a motherboard? If half was on the front and half on the back everything would be much easier to reach. Of course the cost of this would likely be prohibitive.
I'm getting basic cable $30, broadband $35, and local phone $21 all from one company and after all the regulatory fees. Where's the other $140 come from?
My guess would be cellular phone, long distance package, caller id, call waiting, voicemail, line protection, giant smurfs, and a really expensive cable package... either way, it's been inflated.
I explained to a newcomer in one hour how to surf, browse, send emails, access a newsnet server, and play little games on it, don't talk about nonsense crap for having to learn weeks.
Linux has a steep learning curve. Period.
1) When you decide to try out Linux, the first thing you realize is you have a lot of choices. Distribution, Window Manager, Sound Support, etc. Choices are great but thoroughly researching them requires a great deal of time. With Windows, you only have to chose between the latest DOS or NT based system.
2) With Linux, you need to figure out if your hardware is supported. Even if it is supported, it is certain that the software (that you paid for) packaged with it isn't. Since Windows is the consumer leader, hardware support for consumer equipment is not a big issue. After building 5 or so completely different windows systems, I can say it has never been a problem for me or anyone I personally know.
3) The actual installation process is much more complicated for most Linux distributions than it is for Windows. My experience is with RedHat62 and Potato. In both cases, if the user wants to do any customizing, he is confronted with complicated stuff like partitioning hard drives and selecting from long lists of obsficated modules.
4) Portability of knowledge. KDE, Gnome, may be great, but what happens when you have to use an alternative system. As far as I can tell, the only way you can truly master Linux is by mastering the Unix command line and everything that it entails. Let's just say RedHat's Gnome package comes with a special user manager program or a task scheduler. Those programs may be completely different or absent on a Debian KDE install. The only way you can learn to manage users or schedule tasks on both systems no matter what is to learn the syntax of adduser and crontab. Learning the command line for everything is a pain in the ass. Since you have far fewer choices with Windows, everything is more or less the same at a fairly intuitive graphical level.
5) Learning the Command line is hard. Command line interfaces are typically designed for brevity of keystrokes. This makes them very user friendly to people who already understand them and very difficult for people who are learning them. Case and point: VI (I use it). Short switches (cp -R, crontab -e) make everything more complicated... you have to remember -R means recursive, -e is for edit... and they don't stay the same for all commands. Or as my mom says, what the hell is recursion? Also, many commands have funny abbreviated names that you have to look up. This is complicated.
6) Linux does a poor job of abstracting the internal workings. For example, when I go to change permissions on a file, I type chmod -R 755 public_html. Changing permissions on a file should NEVER require binary arithmetic or any knowledge about how the permission is implemented. Similarly, when you schedule a task using crontab, you edit or create a special cron file using crontab -e. The user shouldn't have to know a special month day year hour min syntax for scheduling a program, nor should he even have to be aware a special file containing his scheduled programs even exists. This level of knowledge is uneccessary and makes learning the system more difficult. In contrast, changing permissions is done with a right click in windows (and with ACLS) and scheduling is handled through task scheduler in Windows. Please take a moment to compare those names: crontab, task scheduler. Which is easier to remember? To use?
For me it boils down to three key points. Linux has lots of choices that give you power but make figuring out the best way to do things very complex giving it a steep learning curve. Because Linux is a minority holder in the consumer market, it has relatively poor support for consumer hardware (winmodems, sblive's, the latest ATA100 controllers, etc). Finally, Linux does not hide information well and command line interfaces while powerful, are not intuitive. Oh yeah, and most package managers are pure Hell (read DIE RPM, DIE)!
When you explained to a newcomer how to browse the web, use email/news, and play a few games, you weren't telling them how to use Linux. You were telling them how to surf, use email/news, and play a few games. In other words you taught them Opera and Pine and SameGnome NOT LINUX. If they ever had to set that stuff up themselves or god forbid install/setup a different browser/mail program, how easy would that be for them? You gave them specific programs they had to use in order to get around. In essence you took away the single greatest advantage of Linux, choice.
Brief disclaimer: I like Linux. I use it as a server and do most personal development projects there. Linux is many things, but easier than Windows it is not. PS. I have used KDE.