What level of expertise are your manuals? For example, I never looked at manual for Windows or Microsoft Word back in my Windows days because they are such intuitive, easy-to-use programs. However, at the same time, I could easily see a disaster occuring if Cisco decided to do this. Some of their reference libraries are several volumes and total 2-3 feet of printed material each. Killing them would not be a good idea. Remember two things - this printed manuals are far, far easier to search quickly through (even with no word/regex search, it's just a lot easier to browse through one and find things than with a manpage), and two, it is uncomfortable to read on a computer for extended periods of time.
So, if your company makes mice, by all means, distribute a readme.txt file and be done with it. However, if they make Inverted Confustication Delivery Systems with added Defrillication Modules, then please continue to distribute hardcopy documentation.
No can do. Dreamweaver accomplishes this through a bunch of ungly.LCK files which is spatters through the whole directory structure at will.
I do have a little bit of an idea tho. I realize that most web developers prolly don't want to learn Unix or CVS if at all possible, but realistically that combo is the only viable option. With that in mind, you should explore ways to make CVS easy and convenient to use. There are a few good, intuitive GUIs for CVS available (LinCVS, more on freshmeat). Setup a combo CVS/X/file server and allow people to run the graphical CVS interface over X (there is a free X server for both PC and Mac called TNTlite, check yahoo) to check out HTML into their home directory. Then share the home directory via Samba (or Netatalk, if you're a Mac shop). Voila - people can checkout files over the network, edit them on the server w/filesharing, save the changes, and check them back in. It's kind of a roundabout way of doing things, but it works.
Dreamweaver three has a rudimentary checkin/checkout system in place, if that's what you are looking for. I think it includes support for versions, but I'm not sure.
I do believe that quite a few commerical versioning packages may be designed for C++, but the plain vanilla Revision Control System (RCS) will work with basically any text file. Using RCS, to built on top of it, CVS, works very well for HTML and is quite common. I have seen $Id$ tags on very many web pages. I don't understand what you mean by "web based code" - if you mean stuff that resides on another server, then that is the least of your worries. CVS handles distributed projects like a champ - like every major software package available for Linux right now, pretty much. I would go with CVS.
Differences? Well, notice how the pro's image has colors that are incredibly washed out. Notice how the pro's image has no depth of field.
The short depth of field is a "feature, not a bug," so to speak. It draws attention to the foreground subject by flattening the background. People pay lots of money to get this affect - he was probably using a 300mm-400mm f/2.8 lens, cost: ~$3000. The color washout can be fixed in 5 seconds in Photoshop, BTW.
The digital's shot is good, but it actually proves one of the flaws of digital cameras: they can't handle the light range the way a 35mm can. See the underexposure in McNabb's face and under his armpits on your shot? There is no shadow detail there. The highlights are also missing a little detail. Compare with the other shot, where you can see his open mouth, eyes, nose (these elements really make the shot). Digitals are not as versatile as plain old film in this regard. In turn film is not nearly as versatile as your eyes.
He didn't spend $10,000 for nothing, BTW. Your $900 digital would have a helluva time focusing on air airborne, running quarterback (It wouldn't be able to, until after the shot was already missed). His expensive SLR has a predictive autofocus computer which can determine the direction the subject is moving and compensate accordingly - severl times a second. By the time your digital got a shot off, he could tear through half a roll of film.
I'm not knocking your shot or the camera in any way. For low-res stuff, low-action shots like pretty much everything you find on the web, digitals are a godsend. Just keep in mind that they have a long way to go before they can hold a candle to even a mid-range SLR. People are not spending thousands of dollars in vain. For someone whose livelihood depends on consistently producing great shots, you want the best equipment available. That's why the pro went with the SLR and you with the digital.
I fail to see why anything over 1.3 Megapixels would increase the probabities of taking over 35mm.
Filling your screen is a pretty abstract benchmark of performance considering it's totally dependent on your screen resolution, and it says nothing of clarity, color balance, etc. Also, for serious 35mm users, 1.3Mpixels is not even close to sufficient. It just doesn't work for cropping small areas of the picture; I know, I've tried. I work for a small newspaper and we bought an Olympus digital camera of that resolution. It's nice and fast and all, but you just can't do tight crops on it. The detail is noticeably substandard w/r/t 35mm, even after printing on newspaper.
True, really high-res digital cameras are out there, but they cost about $40,000. I have seen reporters from larger papers like the LA times using these and I'm told they work great. However, it's cheaper and easier just to scan a slide at 4000dpi and really pull even the most minute details out of a frame. In their current state, cheap digital cameras simply cannot match that. That's why they have a long way to go to beat 35mm.
Okay, for a good laugh I suggest everyone check out Old Man Murray. They've been ripping Romero a new one for a good year now over this game, and I suspect actually having a playable demo will fuel their fire. They'll trash this baby in style. Check out some of the older Daikatana coverage to see what I mean.
Okay, I realize it's a moot point, but actually attempting to submit the songs over SSL and "Pay Lars" gives a 403.15 error, which is some sort of Microsoft incarnate HTTP error code saying that the server has run out of "licenses". It even comes with a pretty little link to where you can buy more on IE 5.0. Argh.. LICENSES? I now know the answer to a question I recieve quiet frequently from my less-technical-savvy friends and relatives ("How does Microsoft do it?") - by being greedy assholes. Someone oughta drop this guy a Redhat CD in the mail with a link to www.apache-ssl.org while they're at it.
RF jamming is hardly a new technology, and the people who came up with GPS aren't stupid. This possibility has without a doubt already been considered and dealt with - give them a little credit here. Jamming a spread-spectrum carrier means using a broadband jammer, which is expensive, sucks a lot of power, generates tons of heat, and has a paltry range. Even a theater-based GPS jammer would be inconcievable; barring a massive technological investment, at most only a small portion of the theater could be jammed (realistically). Also, neither smart bombs nor cruise missiles are solely reliant on GPS.
Heh you can say that again. CA grade GPS (what civilians get) is accurate to about 100m, and gets even better if differential GPS is used. In terms of positioning a nuclear bomb, that's like arguing over a few microns.. utterly pointless. Let's not forget that nukes could be accurately aimed long before the advent of GPS, BTW.
Let's pause for a moment and consider what LoC stands for: "Library of Congress". People often forget this, but the entire reason this repository was created, and theoretically the reason it exists now, is to meet the research needs of congress. Taken from their webpage:
1. THE FIRST PRIORITY of the Library of Congress is to make knowledge and creativity available to the United States Congress.
The Congress is the lawmaking body of the United States. As the repository of a universal collection of human knowledge and the creative work of the American people, the Library has the primary mission to make this material available and to identify, analyze and synthesize the information it contains to make it useful to the lawmakers who are the elected representatives of the American people.
This means that it's really not his decision what goes digital and what doesn't - it's up to congress. We all know how convenient having digitized works online would be, and so do your congressmen, who a.) do a lot of research using the LoC, and b.) have way, way less time than any of us. Once they awaken to the possibility, I'm quite sure that they well either browbeat this guy into submission or legislate it. One snooty government employee isn't going to stand in the way of progress (well, except for Greenspan, but that's a whole other story).
Also, his decree directly contravenes the "third priority" of the Library, also off their webpage:
3. THE THIRD PRIORITY of the Library of Congress is to make its collections maximally accessible to (in order of priority)
A. the Congress; B. the U. S. government more broadly; C. the public.
I just got back from Cal Day at Berkeley where they had a Q&A between all the people admitted to the EECS program (me) and their fourth year students. Here are some of the things I heard from the graduating class (I am not exaggerating at all):
I met one girl who had nineteen job offers.
Met a girl who had been picked up in a limo, taken to probably the most expensive restaraunt in San Francisco, returned in limo, and sent flowers the next day, by like 8 different companies (Oracle, MS, etc.).
Met a guy whose friend is getting 6 figures when he graduates (so he claims).
Met a guy who didn't have to buy himself dinner for two weeks because recruiters were doing it
This was all two days ago. Imagine my surprise then, to hear of your plight. As far as I know, it's a real bear market for computer jobs right now. Companies really are starving for engineers, especially for University trained OO programmers (this is like the creme-de-la-creme, from what I gather). Either you are not looking far or hard enough. Put your resume up where a lot of people from all over can see it (Monster.com, etc.) and be ready to relocate, and I guarantee that something will come along.
Agreed. I attend one of the pilot schools for the Netschools program, and criticism has been far and wide. Our school sunk about $500,000 into the project and two years later they don't have that much to show for it. There is little proven correlation between test scores and the presence/absence of laptops in the program. Having kids type up papers and surf the web in class is nice, but is it really worth it when there are certain classes at my school that can't afford textbooks? That's a question we've been pondering since 1998, and the answer seems to be no. The laptops are distributed only to freshmen, who routinely trash them (see other posting regarding this), and generally don't use them to the fullest of their potential. The web as a learning source is a novel solution, but one that's in its infancy and really not much better than a well-written textbook, at this point.
Also, the laptops cost over $1500 per unit. If they we're $200, as you mention, I'd be all for it. But the return on investment, as it stands now, for this program, just isn't good enough to take money away from other programs in our cash-strapped district and divert them to things like this.
I attend one of the Netschools pilot high schools where these laptops are actually distributed to every single incoming freshmen. In fact, I'm actually an employee of the Netschools Corporation; I worked in the Netschools office at my school for awhile. Here is what I can tell you:
w/r/t cost, it's about $1500 per "StudyPro". They consist of a 486 and a 32mb compressed flash card holding Windows 95.
Theoretically they are "ruggedized," but in reality you should never overestimate a typical idiot freshman. For some reason Netschools chose to paint their plastic case with a magnesium colored paint, which is both an interesting marketing ploy and complete idiocy. The laptop does look a lot cooler, but most of the kids think they are slinging around a hunk of metal when they are not. This leads to quite a few problems; I've seen StudyPros used as stepladders to high lockers, kids piling three or four of them up and jumping them with their skateboard, and assorted other activities that would make whoever was paying for these things shit in their pants. Invariably when we questioned them they would answer with variations on the same response: "It's metal; it's indestructable." Wrong.
This problem could be fixed by actually using a metal case, but those are too expensive for most school districts to justify an expenditure. Other problems that make me wonder if having laptops in a rugged (read: non-collegiate) environment? LCDs, mainly. There's really no way to protect (cheaply) an LCD screen from puncturing/cracking. Invariably we'd have 3-4 laptops a month come in with orangish goo emanating from the LCD along with an assorted & amusing explanation - "shot it with a BB gun," "tried to crush a nut between the screen & CPU," etc. This is a $400 repair and there's not a cheap way to guard against it. Truly tough LCDs, supposedly, cost much more money, more than the laptop itself. Yet this was probably one of the more common repairs.
Concering the pre-censored aspect: all the laptops are forced to run through a proxy server, yes, but calling it censorship does not do justice to the full issue. You cannot expect to use the laptop as an educational aid without some form of content control. When a teacher tells students to research Africa on the Internet, they need to actually do work, and not surf over to www.slashdot.org, as about 28 of my friends tried to do. Also, there is a possible liability issue here; a parent would have a real case if their student learned how to make a bomb off his StudyPro at school and ended up blowing himself up. It is censorship, yes, but the alternative is no program at all. So it's either: limited access to the web, or no access to the web. I think it's clear what most people would choose.
Sorry that you "detest" it, but there's really no way around the problem. I know it may be hard for people here to grok the idea of a problem that cannot be solved by group collaboration and the free and unbridled exchange of ideas, but they do exist, and this is one of them.
Think about what you are saying: "I detest the fact that you have to pay a Trusted Certificate Authority before you can seamlessly secure these sites." The only way to seamlessly secure "these sites" is to have someone who proactively ensures that these sites are who they say they are (the Trusted Certificate Authority). If you've ever purchased an SSL cert before, you know what an arduous process this is - typically three or more separate forms of identifcation are required, articles of incorporation, etc. Verifying that you are actually "you" is a costly and time consuming thing, and barring an unusally pious CA, someone is going to charge you to do it. The money you pay ensures that they are issuing certs with truthful and correct data on them.
The alternative is not a pretty picture. OpenCA will not, and should not, "get into" the Trusted CA list of browsers because it isn't. They do not perform identity checking (at least as far as I can tell based on a cursory glance at their signup page). Telling several million browsers to take anything OpenCA tells them as gospel is just asking for disaster. It would essentially be like authorizing the DMV to sell photo IDs with whatever information you ask for on it - you can be anyone, any server, any thing, and as long as OpenCA is "trusted," no one can tell the difference.
This may not seem like a big deal now, but it will be in the very near future, when one's digital certificate signature carries the same legal force as a handwritten one (this will happen). Scrutiny on digital certs needs to be increased, if anything. They shouldn't be handed out like candy.
Incorrect. The contracts they sign for CSS specs bar them from releasing any of their hardware specs, CSS related or not. E-mail them and ask them - I did.
Huh? Why bother? Yahoo is free, it works now, and their servers are probably up more than yours. They have a client for Palm that does exactly what this guy wants. Why reinvent the wheel?
I guess I'm a fan of the KISS theory when it comes to stuff like this. I see the rabid and often pointless impulse on/. and esp. in the Linux/FreeBSD communities to code whatever you want to prove your hackerish balls - go download one of the 53 different IRC clients off Freshmeat and you'll see what I mean. What a waste of resources. The economist in me screams.
Yes, I neglected to mention to communal aspect to it. The fact that mankind has congregated to eat since basically we came into existence suggests that there are more than just pragmatic reasons for it. There is something deeply psychological and primal about eating, on the same level as sex or breathing. I'm not sure if messing a tradition hundreds of thousands of years old is that great of an idea.
Read my post earlier in this thread. They can't release the drivers because the Sigma card decrypts CSS in software, viz. in the driver itself. Releasing the specs would entail telling the world how to decrypt CSS. We all know how great that went over last time;)
Ok, so they lack tact, but you have to understand that their hands are tied anyways. They can't release the driver specs because CSS decryption for this particular card is accomplished in software e.g. in the driver. Releasing the specs would essentially mean telling people how to crack the CSS algorithm, which, in case you've been living in a cave on Mars for the past year, is verboten. Look at the fervor with which the MPAA is going after obscure people and random Geocities accounts for distributing DeCSS. As a company, with a physical address to recieve subpoenas at and revenue to be sued for, you too would be afraid of MPAA. If Sigma were to release these drivers, MPAA would eat them alive in court.
I don't think they have a bad attitude; they just can't legally do anything about this. On the contrary, I think they have a good attitude, considering that they told you they would support Linux with their next product. That's a lot cooler than most of the companies out there today.
Okay, here's my totally unscientific, probably flat wrong, view on this...
1. Ever seen those people on 20/20 who are taking like 20000 times the RDA for various vitamins to fend off, among other things, old age, cancer, liver disease, or the plague? Those are always, invariably, the least health, most fucked up people I have ever seen. Their skin is a burnt-piss-yellow color, they have buggy little eyes. Something is wrong here. Some sort of imbalance exists.
2. I had some bad experiences with beef and became a vegetarian for about a year of my life. I basically stopped growing (I was 16 at the time, right in the middle of puberty - bad idea) because I didn't eat enough protein, even though I drank protein shakes and had plenty of protien supplements, hated beans but ate a little bit anyways, and generally had a pretty all-around balanced diet. I took up meat in a big way after that and gained about 20 lbs.
I posit that, for whatever reason, taking man-made nutritional supplements just doesn't cut it after awhile. There is something about natural vitamins and carbos and proteins that our body needs. I don't know what it is, but I know it exists. I don't think it would be possible to live on Vitapatches for an extended period even in the best of situations, not to mention humping it through the jungle/desert as a footsoldier all day long.
So many people here are pining for more features - IrDA, Ethernet, etc, when in reality this is the exact wrong approach to take.
Let's think about why the IO's became so successful. It was the price! For me, and I'm sure for many other people, we saw the original story on/., thought "Cool; another machine hacked to run Linux" and kind of glossed over it. Later I came back and realized: this thing costs like 100 bucks! And I went out and bought one, and so did everyone else. I/we don't really care about features, or processor speed, or RAM, or hard drive space. For most of the things this thing has been hacked to do, those are worthless. I can still type a paper or browse the web whether it's 100 or 1000 mhz. I've got one on my coffee table as a web brower, and it's slow, but it's the coolness factor that I care about: this little box is browsing the net and leaves a teeny little footprint.
Adding a bunch of specialized features can only serve to increase the price and thus make it less popular. For every potential user that needs Ethernet, there are ten people who don't and for whom this feature would be a dead weight waste of money. This holds true for most of the proposed changes on the board. Integrate GPS would be great, for the four people in the world who would use it. The changes may be good for me, but they won't for a lot of people.
I submit that very little should change from the original IO, and anything that does should at least follow these two guidelines:
1. Whatever peripheral is added, make it based on well documented standards. 2. Don't alter the form factor of the case. The fact that the IO was cute was a big selling point for a lot of people, no matter how lame that sounds.
Thus, within the guidelines, I would suggest adding a standard PC Card slot - and that's it- which is of course well supported and an open standard, small (so it doesn't mess with the size/weight), and, needless to say, extremely versatile. Possibly even two (or more) PC cards could be added; one for a hard drive and one for miscellaneous. The IO is certainly thick enough to accomodate a few.
PC Card slots are cheap ($60 for a PC version) and ubiquitous, and could be made to suit everyone's fancy. I think that any proposed addition to the IO should be looked at in this light, and nixed if it doesn't work.
This is definitely not the second coming of the 300A; I'm suprised to poster, who wrote the article, would say this. What made the 300A (and the 366, to some extent) beautiful was that it matched, sometimes even outperformed a P2 at equal clockrate. For about a quarter the price. It even matched P3s on non-SSE apps.
They managed to pull some incredible clockrates out of the FCPGA Celeries, but in no way are they comparable to an equal Pentium 3:
While the original Celeron 300A@450MHz offered the same performance as a similarly clocked Pentium II, we can see from the benchmarks that the new Celeron will be significantly slower than a Coppermine P3 of the same speed. At 901MHz, the Celeron only outperforms the P3 by a minimal amount.
It's still a pretty good deal; spend about $180 for a Celeron 566 vs. $230 for a P3-600 133mhz FSB. Just keep in mind that a P3-900, when it comes out, will mop the floor with your Celeron;)
WTF are you talking about? I'm going to assume you intended your response to be germane to the question that was originally posed, namely "What Discman-like devices can play CDs full of MP3s." In which case you are patently incorrect. MP3s are perfectly legitimate here because it is a lot, lot more convenient to have one CD with hundreds of your favorite tracks loaded up and encoded as MP3s than to have to swap CDs, or even burn a compilation with a few of your favorite tracks. Encoding MP3s of CDs that you own is covered under "Personal Use" of the copyright laws and is perfectly, 100% legal.
Slashdot Mirror
What level of expertise are your manuals? For example, I never looked at manual for Windows or Microsoft Word back in my Windows days because they are such intuitive, easy-to-use programs. However, at the same time, I could easily see a disaster occuring if Cisco decided to do this. Some of their reference libraries are several volumes and total 2-3 feet of printed material each. Killing them would not be a good idea. Remember two things - this printed manuals are far, far easier to search quickly through (even with no word/regex search, it's just a lot easier to browse through one and find things than with a manpage), and two, it is uncomfortable to read on a computer for extended periods of time.
So, if your company makes mice, by all means, distribute a readme.txt file and be done with it. However, if they make Inverted Confustication Delivery Systems with added Defrillication Modules, then please continue to distribute hardcopy documentation.
--
No can do. Dreamweaver accomplishes this through a bunch of ungly .LCK files which is spatters through the whole directory structure at will.
I do have a little bit of an idea tho. I realize that most web developers prolly don't want to learn Unix or CVS if at all possible, but realistically that combo is the only viable option. With that in mind, you should explore ways to make CVS easy and convenient to use. There are a few good, intuitive GUIs for CVS available (LinCVS, more on freshmeat). Setup a combo CVS/X/file server and allow people to run the graphical CVS interface over X (there is a free X server for both PC and Mac called TNTlite, check yahoo) to check out HTML into their home directory. Then share the home directory via Samba (or Netatalk, if you're a Mac shop). Voila - people can checkout files over the network, edit them on the server w/filesharing, save the changes, and check them back in. It's kind of a roundabout way of doing things, but it works.
--
Dreamweaver three has a rudimentary checkin/checkout system in place, if that's what you are looking for. I think it includes support for versions, but I'm not sure.
I do believe that quite a few commerical versioning packages may be designed for C++, but the plain vanilla Revision Control System (RCS) will work with basically any text file. Using RCS, to built on top of it, CVS, works very well for HTML and is quite common. I have seen $Id$ tags on very many web pages. I don't understand what you mean by "web based code" - if you mean stuff that resides on another server, then that is the least of your worries. CVS handles distributed projects like a champ - like every major software package available for Linux right now, pretty much. I would go with CVS.
--
Differences? Well, notice how the pro's image has colors that are incredibly washed out. Notice how the pro's image has no depth of field.
The short depth of field is a "feature, not a bug," so to speak. It draws attention to the foreground subject by flattening the background. People pay lots of money to get this affect - he was probably using a 300mm-400mm f/2.8 lens, cost: ~$3000. The color washout can be fixed in 5 seconds in Photoshop, BTW.
The digital's shot is good, but it actually proves one of the flaws of digital cameras: they can't handle the light range the way a 35mm can. See the underexposure in McNabb's face and under his armpits on your shot? There is no shadow detail there. The highlights are also missing a little detail. Compare with the other shot, where you can see his open mouth, eyes, nose (these elements really make the shot). Digitals are not as versatile as plain old film in this regard. In turn film is not nearly as versatile as your eyes.
He didn't spend $10,000 for nothing, BTW. Your $900 digital would have a helluva time focusing on air airborne, running quarterback (It wouldn't be able to, until after the shot was already missed). His expensive SLR has a predictive autofocus computer which can determine the direction the subject is moving and compensate accordingly - severl times a second. By the time your digital got a shot off, he could tear through half a roll of film.
I'm not knocking your shot or the camera in any way. For low-res stuff, low-action shots like pretty much everything you find on the web, digitals are a godsend. Just keep in mind that they have a long way to go before they can hold a candle to even a mid-range SLR. People are not spending thousands of dollars in vain. For someone whose livelihood depends on consistently producing great shots, you want the best equipment available. That's why the pro went with the SLR and you with the digital.
--
I fail to see why anything over 1.3 Megapixels would increase the probabities of taking over 35mm.
Filling your screen is a pretty abstract benchmark of performance considering it's totally dependent on your screen resolution, and it says nothing of clarity, color balance, etc. Also, for serious 35mm users, 1.3Mpixels is not even close to sufficient. It just doesn't work for cropping small areas of the picture; I know, I've tried. I work for a small newspaper and we bought an Olympus digital camera of that resolution. It's nice and fast and all, but you just can't do tight crops on it. The detail is noticeably substandard w/r/t 35mm, even after printing on newspaper.
True, really high-res digital cameras are out there, but they cost about $40,000. I have seen reporters from larger papers like the LA times using these and I'm told they work great. However, it's cheaper and easier just to scan a slide at 4000dpi and really pull even the most minute details out of a frame. In their current state, cheap digital cameras simply cannot match that. That's why they have a long way to go to beat 35mm.
--
Okay, for a good laugh I suggest everyone check out Old Man Murray. They've been ripping Romero a new one for a good year now over this game, and I suspect actually having a playable demo will fuel their fire. They'll trash this baby in style. Check out some of the older Daikatana coverage to see what I mean.
Totally hilarious.
--
Okay, I realize it's a moot point, but actually attempting to submit the songs over SSL and "Pay Lars" gives a 403.15 error, which is some sort of Microsoft incarnate HTTP error code saying that the server has run out of "licenses". It even comes with a pretty little link to where you can buy more on IE 5.0. Argh.. LICENSES? I now know the answer to a question I recieve quiet frequently from my less-technical-savvy friends and relatives ("How does Microsoft do it?") - by being greedy assholes. Someone oughta drop this guy a Redhat CD in the mail with a link to www.apache-ssl.org while they're at it.
--
RF jamming is hardly a new technology, and the people who came up with GPS aren't stupid. This possibility has without a doubt already been considered and dealt with - give them a little credit here. Jamming a spread-spectrum carrier means using a broadband jammer, which is expensive, sucks a lot of power, generates tons of heat, and has a paltry range. Even a theater-based GPS jammer would be inconcievable; barring a massive technological investment, at most only a small portion of the theater could be jammed (realistically). Also, neither smart bombs nor cruise missiles are solely reliant on GPS.
--
Heh you can say that again. CA grade GPS (what civilians get) is accurate to about 100m, and gets even better if differential GPS is used. In terms of positioning a nuclear bomb, that's like arguing over a few microns.. utterly pointless. Let's not forget that nukes could be accurately aimed long before the advent of GPS, BTW.
--
Let's pause for a moment and consider what LoC stands for: "Library of Congress". People often forget this, but the entire reason this repository was created, and theoretically the reason it exists now, is to meet the research needs of congress. Taken from their webpage:
1. THE FIRST PRIORITY of the Library of Congress is to make knowledge and creativity available to the United States Congress.
The Congress is the lawmaking body of the United States. As the repository of a universal collection of human knowledge and the creative work of the American people, the Library has the primary mission to make this material available and to identify, analyze and synthesize the information it contains to make it useful to the lawmakers who are the elected representatives of the American people.
This means that it's really not his decision what goes digital and what doesn't - it's up to congress. We all know how convenient having digitized works online would be, and so do your congressmen, who a.) do a lot of research using the LoC, and b.) have way, way less time than any of us. Once they awaken to the possibility, I'm quite sure that they well either browbeat this guy into submission or legislate it. One snooty government employee isn't going to stand in the way of progress (well, except for Greenspan, but that's a whole other story).
Also, his decree directly contravenes the "third priority" of the Library, also off their webpage:
3. THE THIRD PRIORITY of the Library of Congress is to make its collections maximally accessible to (in order of priority)
A. the Congress;
B. the U. S. government more broadly;
C. the public.
Maximally accessible! Hullo?
Although I fully agree with what he has to say.
--
This was all two days ago. Imagine my surprise then, to hear of your plight. As far as I know, it's a real bear market for computer jobs right now. Companies really are starving for engineers, especially for University trained OO programmers (this is like the creme-de-la-creme, from what I gather). Either you are not looking far or hard enough. Put your resume up where a lot of people from all over can see it (Monster.com, etc.) and be ready to relocate, and I guarantee that something will come along.
--
Agreed. I attend one of the pilot schools for the Netschools program, and criticism has been far and wide. Our school sunk about $500,000 into the project and two years later they don't have that much to show for it. There is little proven correlation between test scores and the presence/absence of laptops in the program. Having kids type up papers and surf the web in class is nice, but is it really worth it when there are certain classes at my school that can't afford textbooks? That's a question we've been pondering since 1998, and the answer seems to be no. The laptops are distributed only to freshmen, who routinely trash them (see other posting regarding this), and generally don't use them to the fullest of their potential. The web as a learning source is a novel solution, but one that's in its infancy and really not much better than a well-written textbook, at this point.
Also, the laptops cost over $1500 per unit. If they we're $200, as you mention, I'd be all for it. But the return on investment, as it stands now, for this program, just isn't good enough to take money away from other programs in our cash-strapped district and divert them to things like this.
--
I attend one of the Netschools pilot high schools where these laptops are actually distributed to every single incoming freshmen. In fact, I'm actually an employee of the Netschools Corporation; I worked in the Netschools office at my school for awhile. Here is what I can tell you:
w/r/t cost, it's about $1500 per "StudyPro". They consist of a 486 and a 32mb compressed flash card holding Windows 95.
Theoretically they are "ruggedized," but in reality you should never overestimate a typical idiot freshman. For some reason Netschools chose to paint their plastic case with a magnesium colored paint, which is both an interesting marketing ploy and complete idiocy. The laptop does look a lot cooler, but most of the kids think they are slinging around a hunk of metal when they are not. This leads to quite a few problems; I've seen StudyPros used as stepladders to high lockers, kids piling three or four of them up and jumping them with their skateboard, and assorted other activities that would make whoever was paying for these things shit in their pants. Invariably when we questioned them they would answer with variations on the same response: "It's metal; it's indestructable." Wrong.
This problem could be fixed by actually using a metal case, but those are too expensive for most school districts to justify an expenditure. Other problems that make me wonder if having laptops in a rugged (read: non-collegiate) environment? LCDs, mainly. There's really no way to protect (cheaply) an LCD screen from puncturing/cracking. Invariably we'd have 3-4 laptops a month come in with orangish goo emanating from the LCD along with an assorted & amusing explanation - "shot it with a BB gun," "tried to crush a nut between the screen & CPU," etc. This is a $400 repair and there's not a cheap way to guard against it. Truly tough LCDs, supposedly, cost much more money, more than the laptop itself. Yet this was probably one of the more common repairs.
Concering the pre-censored aspect: all the laptops are forced to run through a proxy server, yes, but calling it censorship does not do justice to the full issue. You cannot expect to use the laptop as an educational aid without some form of content control. When a teacher tells students to research Africa on the Internet, they need to actually do work, and not surf over to www.slashdot.org, as about 28 of my friends tried to do. Also, there is a possible liability issue here; a parent would have a real case if their student learned how to make a bomb off his StudyPro at school and ended up blowing himself up. It is censorship, yes, but the alternative is no program at all. So it's either: limited access to the web, or no access to the web. I think it's clear what most people would choose.
--
Sorry that you "detest" it, but there's really no way around the problem. I know it may be hard for people here to grok the idea of a problem that cannot be solved by group collaboration and the free and unbridled exchange of ideas, but they do exist, and this is one of them.
Think about what you are saying: "I detest the fact that you have to pay a Trusted Certificate Authority before you can seamlessly secure these sites." The only way to seamlessly secure "these sites" is to have someone who proactively ensures that these sites are who they say they are (the Trusted Certificate Authority). If you've ever purchased an SSL cert before, you know what an arduous process this is - typically three or more separate forms of identifcation are required, articles of incorporation, etc. Verifying that you are actually "you" is a costly and time consuming thing, and barring an unusally pious CA, someone is going to charge you to do it. The money you pay ensures that they are issuing certs with truthful and correct data on them.
The alternative is not a pretty picture. OpenCA will not, and should not, "get into" the Trusted CA list of browsers because it isn't. They do not perform identity checking (at least as far as I can tell based on a cursory glance at their signup page). Telling several million browsers to take anything OpenCA tells them as gospel is just asking for disaster. It would essentially be like authorizing the DMV to sell photo IDs with whatever information you ask for on it - you can be anyone, any server, any thing, and as long as OpenCA is "trusted," no one can tell the difference.
This may not seem like a big deal now, but it will be in the very near future, when one's digital certificate signature carries the same legal force as a handwritten one (this will happen). Scrutiny on digital certs needs to be increased, if anything. They shouldn't be handed out like candy.
--
Incorrect. The contracts they sign for CSS specs bar them from releasing any of their hardware specs, CSS related or not. E-mail them and ask them - I did.
Could you really expect any more from the MPAA?
--
Huh? Why bother? Yahoo is free, it works now, and their servers are probably up more than yours. They have a client for Palm that does exactly what this guy wants. Why reinvent the wheel?
/. and esp. in the Linux/FreeBSD communities to code whatever you want to prove your hackerish balls - go download one of the 53 different IRC clients off Freshmeat and you'll see what I mean. What a waste of resources. The economist in me screams.
I guess I'm a fan of the KISS theory when it comes to stuff like this. I see the rabid and often pointless impulse on
--
Yes, I neglected to mention to communal aspect to it. The fact that mankind has congregated to eat since basically we came into existence suggests that there are more than just pragmatic reasons for it. There is something deeply psychological and primal about eating, on the same level as sex or breathing. I'm not sure if messing a tradition hundreds of thousands of years old is that great of an idea.
--
Read my post earlier in this thread. They can't release the drivers because the Sigma card decrypts CSS in software, viz. in the driver itself. Releasing the specs would entail telling the world how to decrypt CSS. We all know how great that went over last time ;)
--
Ok, so they lack tact, but you have to understand that their hands are tied anyways. They can't release the driver specs because CSS decryption for this particular card is accomplished in software e.g. in the driver. Releasing the specs would essentially mean telling people how to crack the CSS algorithm, which, in case you've been living in a cave on Mars for the past year, is verboten. Look at the fervor with which the MPAA is going after obscure people and random Geocities accounts for distributing DeCSS. As a company, with a physical address to recieve subpoenas at and revenue to be sued for, you too would be afraid of MPAA. If Sigma were to release these drivers, MPAA would eat them alive in court.
I don't think they have a bad attitude; they just can't legally do anything about this. On the contrary, I think they have a good attitude, considering that they told you they would support Linux with their next product. That's a lot cooler than most of the companies out there today.
--
Okay, here's my totally unscientific, probably flat wrong, view on this...
1. Ever seen those people on 20/20 who are taking like 20000 times the RDA for various vitamins to fend off, among other things, old age, cancer, liver disease, or the plague? Those are always, invariably, the least health, most fucked up people I have ever seen. Their skin is a burnt-piss-yellow color, they have buggy little eyes. Something is wrong here. Some sort of imbalance exists.
2. I had some bad experiences with beef and became a vegetarian for about a year of my life. I basically stopped growing (I was 16 at the time, right in the middle of puberty - bad idea) because I didn't eat enough protein, even though I drank protein shakes and had plenty of protien supplements, hated beans but ate a little bit anyways, and generally had a pretty all-around balanced diet. I took up meat in a big way after that and gained about 20 lbs.
I posit that, for whatever reason, taking man-made nutritional supplements just doesn't cut it after awhile. There is something about natural vitamins and carbos and proteins that our body needs. I don't know what it is, but I know it exists. I don't think it would be possible to live on Vitapatches for an extended period even in the best of situations, not to mention humping it through the jungle/desert as a footsoldier all day long.
--
So many people here are pining for more features - IrDA, Ethernet, etc, when in reality this is the exact wrong approach to take.
/., thought "Cool; another machine hacked to run Linux" and kind of glossed over it. Later I came back and realized: this thing costs like 100 bucks! And I went out and bought one, and so did everyone else. I/we don't really care about features, or processor speed, or RAM, or hard drive space. For most of the things this thing has been hacked to do, those are worthless. I can still type a paper or browse the web whether it's 100 or 1000 mhz. I've got one on my coffee table as a web brower, and it's slow, but it's the coolness factor that I care about: this little box is browsing the net and leaves a teeny little footprint.
Let's think about why the IO's became so successful. It was the price! For me, and I'm sure for many other people, we saw the original story on
Adding a bunch of specialized features can only serve to increase the price and thus make it less popular. For every potential user that needs Ethernet, there are ten people who don't and for whom this feature would be a dead weight waste of money. This holds true for most of the proposed changes on the board. Integrate GPS would be great, for the four people in the world who would use it. The changes may be good for me, but they won't for a lot of people.
I submit that very little should change from the original IO, and anything that does should at least follow these two guidelines:
1. Whatever peripheral is added, make it based on well documented standards.
2. Don't alter the form factor of the case. The fact that the IO was cute was a big selling point for a lot of people, no matter how lame that sounds.
Thus, within the guidelines, I would suggest adding a standard PC Card slot - and that's it- which is of course well supported and an open standard, small (so it doesn't mess with the size/weight), and, needless to say, extremely versatile. Possibly even two (or more) PC cards could be added; one for a hard drive and one for miscellaneous. The IO is certainly thick enough to accomodate a few.
PC Card slots are cheap ($60 for a PC version) and ubiquitous, and could be made to suit everyone's fancy. I think that any proposed addition to the IO should be looked at in this light, and nixed if it doesn't work.
--
This is definitely not the second coming of the 300A; I'm suprised to poster, who wrote the article, would say this. What made the 300A (and the 366, to some extent) beautiful was that it matched, sometimes even outperformed a P2 at equal clockrate. For about a quarter the price. It even matched P3s on non-SSE apps.
;)
They managed to pull some incredible clockrates out of the FCPGA Celeries, but in no way are they comparable to an equal Pentium 3:
While the original Celeron 300A@450MHz offered the same performance as a similarly clocked Pentium II, we can see from the benchmarks that the new Celeron will be significantly slower than a Coppermine P3 of the same speed. At 901MHz, the Celeron only outperforms the P3 by a minimal amount.
It's still a pretty good deal; spend about $180 for a Celeron 566 vs. $230 for a P3-600 133mhz FSB. Just keep in mind that a P3-900, when it comes out, will mop the floor with your Celeron
--
The cryptically titled "Linux Device Drivers" might be a start.
--
You'd think it'd never happened before!
--
WTF are you talking about? I'm going to assume you intended your response to be germane to the question that was originally posed, namely "What Discman-like devices can play CDs full of MP3s." In which case you are patently incorrect. MP3s are perfectly legitimate here because it is a lot, lot more convenient to have one CD with hundreds of your favorite tracks loaded up and encoded as MP3s than to have to swap CDs, or even burn a compilation with a few of your favorite tracks. Encoding MP3s of CDs that you own is covered under "Personal Use" of the copyright laws and is perfectly, 100% legal.
--