Yeah, as always people look at the cost of the technology, but not how much it costs to own the technology.
I laughed out loud when this guy asked if $7000 was enough to network a three story building. I helped build a wired network in a similar building, and we didn't even spend $200 -- which included an expensive crimping tool that's nice to have anyway. Of course it would have added about $500 to do it wireless (assuming you need three WEPs) and maybe another $1,000 if we'd had to buy wireless NICs for everybody (all of the comptuers in the building, most of which were Macs, already had wired NICs) but that's still a long way from $7000.
But a system, any system, needs to be maintained! I got involved in this project because a non-techie friend asked me to help out. My first advice to him was not to try to sign up everybody in the building right off the bat. Instead, they should start with the two households minimum (my friend didn't live in the same apartment as the DSL connection) and then expand it slowly. In the event, I think he decided that it wasn't worth the hastle to have that many more people involved.
Well, microcomputers appeared in the mid 70s, and by then the most common teleprinter was the Modell 33 Teletype. These were pretty much the standard console for non-IBM computers. (IBM, of course, used telecom versions of their own electric typewriters.) In fact, the Model 33 seems to have been the choice for most of the non-mainframes even before the microprocessor turned computing on its economic head.
According to this article, the first practical teleprinter was patented in 1910.
Thanks for a very cool link. But that's a Model 15. Doesn't that mean there were 14 more impact printers preceding it?
For good and ill, the Teletype is still with us. Look at the keyboard in front of you. The Escape, Control, Backspace and Delete keys are all holderovers from the Teletype, where their usage was radicially different from what it is now.
OK, educate us. Why can't you just print the same form multiple times? Googling gives me lots of references to tamper-proofing, but not enough context to understand what this means.
One thing I found really irritating about the MIT article: the author's assumption that "impact" is just a fancy name for "dot matrix". Apparently it's never occured to him that printers can use type elements, just like typewriters.
I live on a planet where "the market" is just a theory. Products exists if and only if people want them? Jeez, half the cost of most consumer items is advertising -- something that's needed to create demand.
Cell phone manufacturers like Motorola spend millions of dollars every year on market research. They ask people exactly what features they want, let them try new features to test usability, test colors, sizes, weight, everything you can think of.
That's the theory. The reality is that market research is not as reliable as its practitioners pretend. It's usually full of statistical fallacies, and easily stacked to give the "right" answers.
I suppose that featuritis is partly caused by market research ("whould you buy this plain phone that doesn't do anything, or this fancy phone which has all these cool features?"). But that has nothing to do with "what people want".
This is comparable to the serial, parallel, and PS/2 ports on a standard PC. All the devices that require them now have USB alternatives -- in fact, the USB version is often easier to find. There's legacy devices, of course, but that's not a big issue for most of us. So why do most PCs still have these ports? It's not economical to take them out.
Still, some manufacturers do so, when they're trying conserve space rather than dollars. By the same token, we might well see cell devices that reduce built-in features to a bare minimum. But what's a bare minimum? It'd be too painful to configure a phone that lacked a built-in keypad and LCD. Mic and speaker are really necessary too, as a fallback when some glitch prevents your Bluetooth headset from working.
For me, the big issue is not all these "legacy" features, but getting them out of the way when you don't need them. If I ever get round to buying a Bluetooth phone, I'll want one that I can just configure, stick in my pocket, and forget. In theory, I don't even need to take it out to dial. But for this to work, I need to be able to disable the buttons on the phone, so they don't get pressed accidentally. The simplest way to do this is with a clamshell design. And yet very few Bluetooth phones are clamshells.
I beg to differ - one of the few things stopping me from purchasing a phone is the fact that I do not want to pay for hundreds of features that I will never use. All I want is an address book and a way to make calls.
That's my attitude as well, and probably the attitude of most cell users. But so what? It's not up to us. It's up to the people who make and sell them, and they need to grow their market and steal customers from their competitors. Which means they need incentives for people to use their phones instead of somebody else or even sticking with landlines. And the only incentive that creates any excitement is features, features, features, and maybe some more features.
Which often results in products that suck, of course, cause the work that makes for a really good product is usually subtle, or even invisible. Which means you can't sell it. So you concentrate on crap that actually makes your product less useful. You might call it the Copeland Effect.
One word: addiction. People, even addicts, tend to look as addiction as a simple failure of will. Whatever "will" is, it doesn't show up on an MRI. What does show up is your limbic system getting rewired so that gratifying your addiction overrides all other urges. Including eating, sleeping, and, yes, pissing.
I'm not even going to try to refute anything on junkscience.com. The guy just picks whatever studies seem to back up his agenda, and. Like when he claimed that abestos insulation would have prevented the fall of the WTC towers. And when somebody points out the flaws in his claims (abestos is not that superior to other kinds of insulation), he just insists that he never said what you think he said. That makes any link to his site a non-argument. And plenty of reputable scientists do consider DDT a health hazard. Hey, by the time it was banned, it was reaching toxic levels in human milk.
There actually was a JavaOS for a while. Meant to run on a Java-based diskless workstation. One of those "network computer" concepts that was going to demolish Windows.
Sun will never rename Solaris or any of its other well-established brands. Not because it would be dumb. But because they're totally in love with the idea of "selling the brand". Which means that you don't get rid of brands that are working for you. (Good idea!) It also means you try an squeeze everything you can out of your best brands, even if it destroys any meaning for the name. (Not so good.)
It's Vaporware if the product exists only in the minds of its promoters. In this case the product is supposed to be a Linux distro that lacks the cost-of-ownership issues of existing Linux distros. A tenative package list that "sledge-hammers" on top of an existing distro is not even close to that product. It's just a demo of what the product might look like.
I suppose the project is too new to expect much. But what effort they have done seems to have a discouragingly high ratio of High Concept Brainstorming, and a low ratio of actual design.
Unfortunately, Nullsoft set the standard for Shoutcast-based software: it has to look cool, and support a ton of cute (but mostly useless) plugins. But it doesn't have to be reliable or useable.
Still, you've got the right idea. What's needed is software that was never proprietary at any level. And voila: VideoLAN! Which is not only top quality software, it started out as somebody's class project. What could be geekier?
Psst! Most people who work at the Java Division of Sun agree with you!
IBM is going to steal Java away from Sun within 5 years.
Depends on what you mean by "steal". If you mean, "become the dominant player in the Java marketplace", that's already happened. But if Sun continues to make a nuisance of itself, IBM will want to own the rights to Java (trademark, specs, etc.). Sun would never sell these things, but if Sun's market cap continues to decline, IBM might find it worthwhile just to buy the whole shooting match. 5 years sounds about right.
Not true, actually. You can license it, and a lot of people have done a good job of reverse-engineering it. What's undocumented -- and impossible to reproduce -- is all the subtle ways that Word uses that format.
In it it had a special section where you could go and check out a text book for a few hours. And for $0.15/page you could make copies of it. Or if you were really poor, walk a few blocks with it and make those copies for 5 cents.
Dude, that's illegal. And no, I don't care, either. But if any publisher reads your post, you can be sure that UoD is going to be told to exercise more control over textbook copying. And they'll comply too, rather than face expensive legal sanctions.
Jeez, pay attention. We all know about buying used books on the web. (I like Alibris myself.) So what? College bookstores have always carried used textbooks. But there are never going to be a lot of them, in the store or online, as long as publishers "revise" them every couple of years, rending previous editions worthless.
Slashdot Mirror
Oops. Cold War over. Never mind!
I laughed out loud when this guy asked if $7000 was enough to network a three story building. I helped build a wired network in a similar building, and we didn't even spend $200 -- which included an expensive crimping tool that's nice to have anyway. Of course it would have added about $500 to do it wireless (assuming you need three WEPs) and maybe another $1,000 if we'd had to buy wireless NICs for everybody (all of the comptuers in the building, most of which were Macs, already had wired NICs) but that's still a long way from $7000.
But a system, any system, needs to be maintained! I got involved in this project because a non-techie friend asked me to help out. My first advice to him was not to try to sign up everybody in the building right off the bat. Instead, they should start with the two households minimum (my friend didn't live in the same apartment as the DSL connection) and then expand it slowly. In the event, I think he decided that it wasn't worth the hastle to have that many more people involved.
According to this article, the first practical teleprinter was patented in 1910.
For good and ill, the Teletype is still with us. Look at the keyboard in front of you. The Escape, Control, Backspace and Delete keys are all holderovers from the Teletype, where their usage was radicially different from what it is now.
One thing I found really irritating about the MIT article: the author's assumption that "impact" is just a fancy name for "dot matrix". Apparently it's never occured to him that printers can use type elements, just like typewriters.
I live on a planet where "the market" is just a theory. Products exists if and only if people want them? Jeez, half the cost of most consumer items is advertising -- something that's needed to create demand.
I suppose that featuritis is partly caused by market research ("whould you buy this plain phone that doesn't do anything, or this fancy phone which has all these cool features?"). But that has nothing to do with "what people want".
Still, some manufacturers do so, when they're trying conserve space rather than dollars. By the same token, we might well see cell devices that reduce built-in features to a bare minimum. But what's a bare minimum? It'd be too painful to configure a phone that lacked a built-in keypad and LCD. Mic and speaker are really necessary too, as a fallback when some glitch prevents your Bluetooth headset from working.
For me, the big issue is not all these "legacy" features, but getting them out of the way when you don't need them. If I ever get round to buying a Bluetooth phone, I'll want one that I can just configure, stick in my pocket, and forget. In theory, I don't even need to take it out to dial. But for this to work, I need to be able to disable the buttons on the phone, so they don't get pressed accidentally. The simplest way to do this is with a clamshell design. And yet very few Bluetooth phones are clamshells.
Which often results in products that suck, of course, cause the work that makes for a really good product is usually subtle, or even invisible. Which means you can't sell it. So you concentrate on crap that actually makes your product less useful. You might call it the Copeland Effect.
Because it's not how much heat you've got, it's how intelligent you are at dispersing it.
I'm not even going to try to refute anything on junkscience.com. The guy just picks whatever studies seem to back up his agenda, and. Like when he claimed that abestos insulation would have prevented the fall of the WTC towers. And when somebody points out the flaws in his claims (abestos is not that superior to other kinds of insulation), he just insists that he never said what you think he said. That makes any link to his site a non-argument. And plenty of reputable scientists do consider DDT a health hazard. Hey, by the time it was banned, it was reaching toxic levels in human milk.
Sun will never rename Solaris or any of its other well-established brands. Not because it would be dumb. But because they're totally in love with the idea of "selling the brand". Which means that you don't get rid of brands that are working for you. (Good idea!) It also means you try an squeeze everything you can out of your best brands, even if it destroys any meaning for the name. (Not so good.)
I suppose the project is too new to expect much. But what effort they have done seems to have a discouragingly high ratio of High Concept Brainstorming, and a low ratio of actual design.
...if UserLinux is ever more than VaporWare.
Still, you've got the right idea. What's needed is software that was never proprietary at any level. And voila: VideoLAN! Which is not only top quality software, it started out as somebody's class project. What could be geekier?
The "Java" in "Java Desktop" is just a brand, not a technology descriptor.
I'm an informative troll! Cool!
100 people rush to the library. Wackiness ensues.
Not true, actually. You can license it, and a lot of people have done a good job of reverse-engineering it. What's undocumented -- and impossible to reproduce -- is all the subtle ways that Word uses that format.
Jeez, pay attention. We all know about buying used books on the web. (I like Alibris myself.) So what? College bookstores have always carried used textbooks. But there are never going to be a lot of them, in the store or online, as long as publishers "revise" them every couple of years, rending previous editions worthless.