His analysis of the kindle as a vehicle for college textbooks doesn't work.
Most students buy their books used and sell most of them back to the bookstore at the end of the semester. If publishers started offering textbooks for the kindle, they'd presumably be DRM'd, and you wouldn't be able to sell them back. The publishers hate the used textbook market, and they do anything they can to kill it off (e.g., a new edition of a calculus textbook every 2-3 years), so there's no question in my mind that they'd use DRM to eliminate it.
Most lower-division textbooks in most subjects are in a large, color format with a layout so complex that it makes every page look like the cockpit of a 747. This doesn't work on the kindle.
He seems to assume that the cost of a college textbook mainly has to do with paper, printing, and binding (ppb), so that it would be much cheaper in electronic form. Actually, ppb is no more than a small fraction of the cost of most textbooks.
He seems to assume that the only way to read an electronic book is on a special e-book reader, and then he goes on to calculate how long it would take to earn back the high cost of a kindle. But nearly all college students either have a laptop or a desktop machine, so the only logical reason for them to buy a kindle would be the same as for anyone else: convenience.
I get a good chuckle out of the folks who say OO is to slow as if they're using it to build an embedded real-time OS.
"Oh noes! it takes 20 seconds to boot!"
This attitude amazes me. In 1980 I ran a word processor called Electric Pencil on a machine whose clock speed was in the low MHz range. If I recall correctly, it only took a few seconds to start. I can't imagine why anyone would tolerate a word processor that took 20 seconds to start on a machine a thousand times faster than the one I used back then. Twenty seconds is twenty seconds out of my life. It adds up.
I tried it. They have versions for Windows and Linux, apparently not for Mac. It's not open source. They have a trial version that you can download for free and use for 30 days. The trial version is crippled: can't save to any other format besides their own proprietary.tmd format. They also offer a non-crippled 2006 Windows version for free -- but not the 2008 version, and not for Linux. The download page wants your name, country, and email address, and tells you that you'll automatically be subscribed to their email newsletter. It doesn't say that you can opt out of the newsletter. However, down below the form where they ask for this information, it says, in microscopic type, "Leave empty if you do not wish to register." It works if you simply click through to the download without filling anything in. They have the Linux download packaged with installers in rpm, deb, and shell flavors. I downloaded the deb version, but it wouldn't install on my machine, because my machine is x64. I copied the deb to an x86 box, and it installed fine. It made menu entries for itself in the Gnome Applications/Office menu. The first time you run it, it wants to set up a documents folder for you, which defaults to ~/SoftMaker. (I find this kind of thing annoying, and believe that it discourages people from developing good habits for organizing their files.)
I'm a little bit baffled right now as to why anyone would choose it. They claim "compactness" as an advantage, but the download is 80 MB, which doesn't seem very compact to me. (The 2006 Windows version is smaller.) Their web site says, "The Microsoft Word-compatible word processor that is so easy to use that you will wonder why you bothered with Microsoft Word or OpenOffice.org for so long," and then lists some bullet points. One is "Fast, powerful, reliable." Actually it didn't really seem any faster than OOo. On the machine I tried it on, the startup time was basically the same as OOo. "Reads and writes your Word documents seamlessly (Microsoft Word.doc 6.0 to 2007)" AFAICT the only advantage over OOo would be if it can read OOXML. (Although OOo can't write OOXML, I can't see why anyone would care; if you save in an older Word format and give the file to Word 2007 users, they'll still be able to read it.) The price is $80 US. Although that may be a lot less than full retail price for Word, it's a heck of a lot more than OOo. And of course I'd have to live with all the usual hassles of proprietary software. I won't get an x64 version unless they deign to compile one for me at some point in the future. I won't be able to upgrade without paying money. Sorry, I'm just spoiled -- apt-get and OSS work fine for me.
I run Linux on my ARM-based slug, which I use as a music player. I'm happy with the result (a cheap, always-on, low-energy music server), but it was kind of a pain to do the install, and the resulting system is broken in enough ways that I wouldn't want to use it very often as a general-purpose computer. Ubuntu is known as an easy-to-install, easy-to-use, full-featured desktop system that Just Works. If they can extend that to ARM-based subnotebooks, etc., then IMO it really is a big deal. Most people who own an x86 or x64 desktop machine really don't need one. All they need is a computer for word-processing, web browsing, and email. A lot of people really are finding out that they're just as happy with a cheap eeePC or whatever. If I could have Linux on an ARM-based notebook computer, and the battery would last for weeks in sleep mode, that would really be something I'd consider buying. But no way would I consider it if I didn't get the same freedom from hassles that I get with Ubuntu.
The OP gives Ndiswrapper as a specific example, but asks a general question, and so far all the replies have been about the general question.
What about the specific situation of Ndiswrapper? There's a saying that "bas cases make bad laws," and Ndiswrapper is sort of like that -- it isn't a typical example of OSS.
Okay, first off let me say that I have two machines on my home network that have Ndiswrapper on them, and I'm grateful that it exists, because it saved me from having to drill holes through my hardwood floors and pull cables from the downstairs to the upstairs.
However, I'd be surprised if anyone had ever been under the impression that Ndiswrapper was anything more than a horrible, nasty, dirty kludge with no future ahead of it. The basic problem is that the manufacturers of the wifi cards don't disclose the relevant technical information that would allow third parties to write drivers, and they also don't support operating systems other than Windows. Anything the OSS community does to try to work around that is bound to work badly and be unsatisfactory. I've already seen that any time I upgrade from one release of Ubuntu to the next, wifi breaks, and I have to go back through all the steps of installing the drivers again. There's also the problem that binary blobs make it difficult to debug kernel crashes.
All of these problems show that ndiswrapper has always been nothing more than a band-aid, and nobody should have ever expected it to have a future.
The only real solution for the future is to spread good information about what cards work with OSS (no binary blobs). The FSF has some info here:
http://www.fsf.org/resources/hw/net/wireless/cards.html . The trouble I always have with this kind of situation is that these online lists are always out of date and inaccurate, and they also tend to systematically overstate the quality of support, e.g., when you I the OSS driver, I can't get it to work at all, or if I do get it to work it crashes all the time, or the full functionality isn't supported.
This is all qualitatively different from the situation where you just have an OSS project that doesn't have ongoing support. A more typical example of that kind of thing would be sox, which is a command-line utility for converting sound files between different formats, adding effects, and playing sounds. Its author hasn't been supporting it properly for a long time, so less and less of its functionality is working on, e.g., a fresh install of ubuntu. It's gotten to the point where, for me, it's basically useless. But that's no big problem, because other people have picked up the slack by writing similar software to replace it. The difference with Ndiswrapper is that the problem is more fundamental. The things that make Ndiswrapper a kludge are inherent to its purpose, which is to be a kludge.
Anyway, that doesn't stop them from using an old copy of Word '97 to create their own documents: it just stops them from reading other people's files in it. You'd have the same problem if you used Emacs or LaTeX or (to some extent) even OpenOffic.org.
I agree with you about Ooo; that was the point I was making in the g'g'g'g'grandparent post. Of course, upgrading Ooo may be less of a problem than upgrading Word, since it doesn't cost money.
I disagree with you about emacs+latex; the point I was making in the g'g'g'g'grandparent post was that if you really want stability and the ability to keep on using old software, you need to pick software that is high quality and extremely mature. Emacs+latex is a good example of that. The basic implementation of tex/latex is extremely stable and mature now, and all that's been changing in the last couple of decades is some of the macro packages that people use layered on top of that. Likewise, I don't see how I could get in a situation where an old version of emacs was unable to read files written by newer versions of emacs -- should be good until ascii and unicode become obsolete, which I don't think will happen very soon.
I don't want to say the s-word, but after I bought something from ThinkGeek, they started sending me marketing emails. I don't recall being presented with a choice about whether to opt in or out of marketing emails when I made the purchase. It was UCE (unsolicited commercial email), but you could argue that I had already established a commercial relationship with them. All I can say is that personally, if I buy from an online retailer and then they send me ads via email, my personal decision is not to do business with that retailer again. One very practical reason is that once they send me ads, I'm going to blacklist them in my email filter, and that would make it difficult to do business again. I'm not accusing ThinkGeek of being evil criminals with handlebar moustaches or anything, but it's just like any other business -- if I don't find it pleasant to buy from them, then they've lost my business.
MS makes sure that compatibility is great in the (b) direction, and lousy in the (a) direction. I guarantee you that word '97 will not be able to read all files written by word 2007.
That doesn't matter, because nothing is stopping you from using your old copy of Word '97 today. It's impossible to use an old version of Google Apps.
Something is stopping people from using an old copy of Word '97, and what's stopping them is what I stated in the g'g'grandparent post: they won't be able to read files that other people send them, written by newer versions of word.
Besides, Microsoft did make some backwards-compatibility converters available.
Interesting link, thanks. But note that the converters won't work with word '97, which was the case that the g'grandparent poster proposed. Also note that it only works on windows, so if you run word on macos you're out of luck.
(me):
you'll soon find yourself unable to open other people's files, written by newer versions
(you):
Word 2007 still opens (and can save) file formats in Word '97 format. That's 10+ years old -- not a length of time I typically associate with "soon". And since the format (as near as I can tell, anyway) doesn't seem to have dramatically changed between the '97 through '03 versions, I don't think it's unreasonable to assume that in 2027 you'll still be able to open files from Word '97. Maybe Word '27 won't be able to save in the Word '97 format, however.
I was talking about case (a), using your old version of the wordprocessor to read a new-format file written by someone else. You're talking about case (b), using your new version of the wordprocessor to read an old-format file. MS makes sure that compatibility is great in the (b) direction, and lousy in the (a) direction. I guarantee you that word '97 will not be able to read all files written by word 2007.
Ummm... no. One is free of charge only, the other is both free of charge and free as in freedom. One stores your data on computers you have no control over and leaves you at the whim of unexpected feature changes by a publically-traded company whose customers are their advertisers and whose product is your eyeballs; the other leaves you firmly in control over your own data and your own software.
You beat me to a bunch of the points that I tend to trot out every time ajaxy stuff comes up on slashdot. But to be fair, I think there are some counter-points to be made in google's favor.
As far as customer and product, what you say is true, but it's also true of newspapers, for example. The subscription price I pay for the New York Times is only a small fraction of my share of what it cost to make the newspaper. The majority of their revenue comes from advertising. Why, then, isn't the NY Times as bad a paper as USA Today? Well, market forces are at work. The NY Times has a market niche (showing ads to pointy-headed intellectuals), and wants to protect that market niche, so they have to keep on producing the kind of newspaper that attracts the eyeballs that make up its niche. Same with google. If gmail, for example, stopped being better than the competing free email services, google would be hurt by marketplace mechanisms, because they wouldn't have as many eyeballs to show ads to.
As far as free-as-in-speech versus free-as-in-beer, that's a relevant comparison for google docs versus ooo, but for most people the actual choice is google docs versus word, which isn't free-as-in-anything.
Arbitrary changes: actually, nobody is protected against changes to software. Ooo and word could both change overnight. Sure, you can continue to run the old version that you liked better, but that's a pretty poor option with word (you'll soon find yourself unable to open other people's files, written by newer versions), and even with ooo it's not very practical, because the old version won't get bug fixes or new features, and will eventually be incompatible with your OS's libraries and APIs. If you really want stability, the only real solution is to pick old, mature software that's extremely stable (e.g., emacs and latex), not recent software like word, ooo, or google docs.
I'm a community college physics teacher in California, which has the best public college and university system (community college, cal state, UC) in the world. What makes me want to stab myself in the eyeballs with number 2 pencils is the way my colleagues on the faculty all want to send their kids to expensive private schools. It's nuts. If their kids are hot shots, they can go to a UC. If they're really, really hot shots, they'll get into the US flagships, Berkeley and UCLA. If their kids are not quite that hot academically, they can go to a lower-ranked UC, or to a cal state. If they blew high school but are now a little more motivated, they can go to that community college, and then transfer. I have a ton of students who transfer to Berkeley and UCLA after 2 years of community college. Many of them are not particularly brilliant, just ordinary students who worked hard. By comparison, it's just insanely competitive if you're trying to get into those two schools straight out of high school -- about as hard as getting into an Ivy.
One common justification I hear my colleagues give for sending their kids to expensive private schools is that they don't want them stuck in large lecture classes where they won't thrive. Sorry, but that's not a reasonable reason to double the cost of college (or quadruple it, compared to community college followed by UC). There are only a few of those huge freshman lecture classes, and students who try hard enough can avoid them completely, especially if they've taken AP exams.
Another common justification I hear is that it should be the kid's choice. Well, that might make sense if it was only the kid paying for it, but usually that's not the case.
Work while doing it? Even better.
That could be good advice if the student is both very smart and very dedicated. What I see much more often is that students have unrealistic expectations about the lifestyle they're going to live at the age of 18. They want to have a nice car, and they don't want to live at home with Mom and Dad. So they work 30 hours a week while taking 18 units. Oops, then they wake up one day and find out that (a) their GPA is 2.0, (b) they took W's in so many classes that college is going to take them 6 years instead of 4, and (c) they missed the part of college where you're supposed to get an education, as opposed to just cramming for tests and doing the minimum to get by.
Yeah, in general I'm wondering what's the right thing to do with stocks I own that are very low in share price. Normally I don't try to "pick stocks" at all, and I don't buy or sell based on movement in prices -- the only reason I own individual stocks is that I wanted to have a socially screened index fund that met my own idiosyncratic social screening criteria. But my understanding (corrections would be appreciated!) is that when a stock goes down below a certain share price or a certain market capitalization, it can get delisted. After delisting it can only trade as a pink sheet, and there may be problems with liquidity. Here are the NYSE's listing criteria, which include (with some oversimplification on my part) market capitalization of 100 M$ and a share price of $4. They don't lower the boom immediately if a company falls below their standards, but when you're getting down to share prices of $1 and $2 it's pretty clear that it could be delisted. So what's the smart thing to do? If I sell, it costs me $7 for the trade, and I could just be locking in losses. If I hang onto it, I could lose my ability to liquidate it, and even if it retains liquidity I'm also not clear on what happens as a practical matter if I own a stock via an online brokerage, and then it goes to the pink sheets market -- hassles for me? expenses?
The question you ask is wrong...since people are no longer answering questions on usenet.
How about (a) toning down the arrogance, and (b) not making statements that are demonstrably untrue. I ask and answer questions on usenet pretty much every day of my life.
That sounds like how I voted. In North Carolina, we use paper ballots, a pen, and an optical scanner (for speed; the paper trail is what's used for a recount).
Where I live, in California, they use electronic voting (rotary wheel and buttons, no touch screen) with a backup in the form of a paper strip that the voter can see being printed behind a plastic window. So it's pretty similar to yours in NC, except that there's no optical scanning required. And let me tell you as a teacher who's run scantrons, optical scanners are extremely unreliable. It's very common for them to get read incorrectly, especially if the user had to erase and change something. Paper jams also happen pretty frequently, and you can pretty easily get a scantron damaged by the machine. The company that sells the scanners we use claims that they're very reliable if they're calibrated properly. That might imply that the ones we have where I teach aren't getting serviced often enough, but if so, that doesn't make me feel very safe about the idea of trusting votes to something similar. How do we know that the ones used in NC are getting serviced often enough? On the voting machines we use here in CA, which print on a paper strip, I'm sure it's possible for there to be problems with the printer, such as running out of ink, running out of paper, or a paper jam -- but all of those would be immediately obvious to the voter. Having your vote miscounted by an optical scanner is something you'll never know happened.
I think a lot of the problems with these systems in the U.S. flow from the fact that they don't want the voter to be able to prove to a third party that he voted a certain way, i.e., you have deniability about how you voted. This seems like a good thing in theory, so that you can prevent coercion or vote-selling. But in reality I've never seen any evidence that either of those has ever been an issue within living memory, and in any case when people vote by mail it is possible to sell votes or be coerced. If we don't demand this deniability feature for voting by mail, why should we be so hung up on it when it comes to voting at a polling place on election day? It would be much easier to satisfy voters that they were getting their votes counted correctly if we got rid of deniability. Give the voter a code number when he votes, and when he goes home, he get on the web, enter his code number, and verifies that his vote was counted correctly.
Sorry about the email regarding our Head First book. Would you mind sending me a copy of it so I can get to the bottom of this? You can email betty@oreilly.com. We don't want you or anyone else to feel like we are spamming you. I can also add you to our no email list. Thanks.
Thanks for your response. I deleted the email, but if you want to find it in your logs, the string to grep for would be either lightandmatter.com or fullcoll.edu (I think it was the former). If you don't want anyone to feel like you're spamming them, then the solution would be pretty simple: don't send unsolicited commercial email to people who have no business relationship with you. Adding me to a no-email list would be irrelevant, because I'm now blocking oreilly.com in my spam filter, and in any case asking people to opt out of spam isn't a reasonable way to deal with the spam problem.
advice on marketing ... from spammers?
on
Website Optimization
·
· Score: 3, Informative
I'm afraid I'd want to steer clear of any book on internet marketing that's published by a company that spams. I'm a college physics professor, and O'Reilly has spammed me to advertise their book Heads First Physics. I was really surprised and disappointed by the spam, because I have a whole bunch of O'Reilly books on the shelf next to my desk at home, and I'd been under the impression that O'Reilly really "got it" when it came to open source and the internet. But I really am pretty firm about not being willing to do business with spammers.
what do you mean by a proof being valid in another set of axioms? that there is a simple way to translate the proof into the 2nd system and it'll be true then?
yes
Which central theorems
aren't already proven? Principia Mathematica did the basics and I always assumed more advanced theorems became proven as required.
Maybe someone just wants to do Principia Mathematica volumes II through L but doesn't that already in effect exist?
Well, one issue is that the Principia Mathematica undoubtedly has errors in it. No human being could ever write a book of that length without making a single mistake. So there could be some value in producing the same results on a computer, which doesn't make the same kind of mistakes a human does. Another issue is that the Principia Mathematica works within a certain axiomatic framework, and I believe that framework is different from the most popular axiomatic framework used today, which is Zermelo-Fraenkel set theory, with the axiom of choice (ZFC). (They were both produced around 1910, but the WP article says that PM used a complicated system of types, which is probably different from anything in ZFC.)
More broadly, for people who are interested in the foundations of mathematics, there's no clear way to decide that one set of axioms is superior to another. Some people feel that ZFC is too strong, because it asserts the existence of things that can't be explicitly constructed. Those people may be interested in seeing how much of mathematics can be proved using purely constructive methods. With a weaker set of axioms, some results may be impossible to prove, while others may be possible to prove, but the proofs may be extremely long compared to the proofs in ZFC -- so long that using a computer may be a real help.
Still other people are interested in seeing what can be done in an axiomatic system that's stronger than ZFC. For instance, such a system may have sizes of infinities that are larger than the sizes of the infinities in ZFC, and they may even have creatures like the set of all sets. One thing that tends to happen when you try to make stronger axiomatic systems is that it becomes much more difficult to avoid internal inconsistencies. I can imagine that a computer could be helpful in finding things like that. If you're fiddling around with the computer proof system and it comes up with a result that 1+1=3, then you've learned something interesting: your set of axioms is bogus.
One thing you really have to be careful about if you're working in this kind of field is that you don't want to inadvertently assume some result that someone else has proved in some other axiomatic system, and that seems obvious to you, but that actually isn't provable (or hasn't yet been formally proved) in the system you're working in. That's the kind of thing where I'd imagine a computer system would be really helpful. It wouldn't allow you to make those assumptions. In general, if you look at almost all published mathematical work, it never states which axiomatic system it's assuming, and that's because in most fields of mathematics we expect that the results are unlikely to depend on which particular foundation you're working in. E.g., I doubt that Wiles' proof Fermat's last theorem even bothers to state that it starts from ZFC, and most mathematicians probably have a strong expectation that it doesn't matter whether you pick some other foundation, Wiles' proof will still be valid. Nevetheless, it's possible that that's not true. Abraham Robinson, for example, claimed that ZFC had been carefully engineered to make the real number system work correctly, and therefore claimed if you wanted to use a different system of numbers (such as the hyperreals, a system that includes Newton- and Leibniz-style infinities), you might be better off using diffrent axioms.
Godel's completeness theorem states that every statement that follows from the axioms is in fact deducible from the axioms in finitely many logical steps
How would you even define a notion of "follows from the axioms" that didn't presuppose proof in a finite number of steps? If there were a distinction between "provable" and "provable in finite steps," it seems like you'd have to define what it meant to prove something in an infinite number of steps, and then show that for every infinite-length proof there was also a simpler, finite-length one. But I'm really having a hard time imagining what it would mean to have an infinite-length proof of something.
That might work on some or most files, but there still is no replacement for Acrobat.
Huh? I create PDFs all the time, and don't own a copy of Acrobat. I use pdftex and inkscape, but there's scads of other software that can do it, e.g., Scribus if you want GUI desktop publishing. This is all on linux, but there's tons of PDF-creating software on Windows as well.
Does Adobe Reader come with a "safe mode" with just plain old PDF enabled?
To disable js, go to Edit, Preferences, JavaScript, and uncheck "Enable Acrobat JavaScript".
Even if the js-related security bugs are fixed, it's still a privacy issue, because js in a pdf file can be used to track who's reading a particular document.
Personally, when I see that a piece of software has a long history of security problems, I take that as my cue to remove it from my system. I don't really care that they keep fixing the bugs. The fact that it has this history demonstrates that the software wasn't written with the correct attention to security, and it's likely to have more such problems in the future.
If you're running Linux, xpdf starts up extremely fast, and that's why I use it as my pdf plugin in Firefox. If you want something a little more modern, try evince.
People have posted saying that on Windows, you should switch to Foxit, but the article says that the security flaw was found first in Foxit, and only later in Adobe Reader. I actually tried to get the science division at the community college where I teach to switch to putting Foxit on machines in the student labs as the default pdf plugin. However, when the faculty were testing it, they found that it was not correctly displaying some of the pdfs they were using.
This is an approach that's been worked on for years and years now, and there hasn't been any rapid progress. Electromagnetic shielding may ultimately work, but it has a lot of problems to overcome. Without some kind of significant technological progress, the radiation dose for astronauts going to Mars is a real showstopper: http://en.wikipedia.org/wiki/Health_threat_from_cosmic_rays
Engineers have studied a variety of electromagnetic field configurations for this. Electric fields have a problem because any field that repels positively charged particles will attract negatively charged ones, and vice versa. Also, large DC electric fields tend to discharge violently.
With magnetic fields, one problem is that you need magnet coils that carry huge currents. If they're not superconducting, then you're talking about huge amounts of power, way more than is really practical. If they're superconducting, then you're trusting your life to a type of technology that's notoriously prone to failure; normal superconductors need liquid helium temperatures, which are very hard to maintain reliably, and high-temperature superconductors still need liquid nitrogen temperatures (and also may be decades away from being ready for this type of application).
In this article they talk about using an AC field, although they don't provide many details. One thing I wonder about with an AC field is radiative losses. They say, "The approach will probably also work with a field that is not on constantly, but cycles on and off - conserving the power that is precious on long-term missions." I totally don't get the idea here. They make it sound like they're just going to have it be pulsed, with a low duty cycle. But then won't it only provide some small fraction of the desired protection?
Another approach that looks promising is to make the spacecraft from low-Z materials like plastics; most of the hazard from cosmic rays is actually from secondary radiation (which is why thicker shielding actually *hurts* you, for any practical thickness).
I think the real question to ask is whether there is any valid, objective reason for sending a human to Mars. Crewed spaceflight has never been a good value for the money in terms of scientific research, compared to probes. Or if the motivation is some kind of romantic vision of heroic exploration of a new frontier, then I think we need to be more realistic about the prospects for permanent, economically viable settlements, which are probably at least a century away.
There's another project called Y Windows, which also aims to replace X with something that has less historical cruft.
The real question in my mind is whether this kind of thing does anything to address the big problems users are really encountering. Of course, not every open source project has to make large numbers of users happier. But the author of Y Windows says, for example, "I've got tired with the state of desktop GNU/Linux. Most of the problems that I see with it can be traced back to the underlying window system, X. So I decided to write its successor... "
For me as an end-user, the big issues are simply hassles with xorg not correctly recognizing LCD screens, so that it sets them to an inappropriate resolution, or the image appears offset. I have close to zero interest in gaming, so personally I just use the onboard video of my mobo, with only 2-d driver features, but the impression I get from people who do care about gaming (or fancy WMs) is that the big issue is drivers, not the internal structure of X.
As far as programming, the structure of X also seems like kind of a non-issue. Sure, X's APIs are heinously ugly, but almost nobody uses them directly.
The advantages listed by the article are things like a more manageable code base, a smaller memory footprint, and elimination of rendering artifacts. To me, none of those seem like major issues that I was all that worried about.
For anyone who doesn't want to sit through the entire (long) Flash presentation by Bruce Nash, here's a micro-summary. Basically, he predicts that it will be around 8 pm Pacific time that they call Virginia and Pennsylvania for Obama, at which point it will be clear that Obama has won.
Slashdot Mirror
His analysis of the kindle as a vehicle for college textbooks doesn't work.
Most students buy their books used and sell most of them back to the bookstore at the end of the semester. If publishers started offering textbooks for the kindle, they'd presumably be DRM'd, and you wouldn't be able to sell them back. The publishers hate the used textbook market, and they do anything they can to kill it off (e.g., a new edition of a calculus textbook every 2-3 years), so there's no question in my mind that they'd use DRM to eliminate it.
Most lower-division textbooks in most subjects are in a large, color format with a layout so complex that it makes every page look like the cockpit of a 747. This doesn't work on the kindle.
He seems to assume that the cost of a college textbook mainly has to do with paper, printing, and binding (ppb), so that it would be much cheaper in electronic form. Actually, ppb is no more than a small fraction of the cost of most textbooks.
He seems to assume that the only way to read an electronic book is on a special e-book reader, and then he goes on to calculate how long it would take to earn back the high cost of a kindle. But nearly all college students either have a laptop or a desktop machine, so the only logical reason for them to buy a kindle would be the same as for anyone else: convenience.
This attitude amazes me. In 1980 I ran a word processor called Electric Pencil on a machine whose clock speed was in the low MHz range. If I recall correctly, it only took a few seconds to start. I can't imagine why anyone would tolerate a word processor that took 20 seconds to start on a machine a thousand times faster than the one I used back then. Twenty seconds is twenty seconds out of my life. It adds up.
I tried it. They have versions for Windows and Linux, apparently not for Mac. It's not open source. They have a trial version that you can download for free and use for 30 days. The trial version is crippled: can't save to any other format besides their own proprietary .tmd format. They also offer a non-crippled 2006 Windows version for free -- but not the 2008 version, and not for Linux. The download page wants your name, country, and email address, and tells you that you'll automatically be subscribed to their email newsletter. It doesn't say that you can opt out of the newsletter. However, down below the form where they ask for this information, it says, in microscopic type, "Leave empty if you do not wish to register." It works if you simply click through to the download without filling anything in. They have the Linux download packaged with installers in rpm, deb, and shell flavors. I downloaded the deb version, but it wouldn't install on my machine, because my machine is x64. I copied the deb to an x86 box, and it installed fine. It made menu entries for itself in the Gnome Applications/Office menu. The first time you run it, it wants to set up a documents folder for you, which defaults to ~/SoftMaker. (I find this kind of thing annoying, and believe that it discourages people from developing good habits for organizing their files.)
I'm a little bit baffled right now as to why anyone would choose it. They claim "compactness" as an advantage, but the download is 80 MB, which doesn't seem very compact to me. (The 2006 Windows version is smaller.) Their web site says, "The Microsoft Word-compatible word processor that is so easy to use that you will wonder why you bothered with Microsoft Word or OpenOffice.org for so long," and then lists some bullet points. One is "Fast, powerful, reliable." Actually it didn't really seem any faster than OOo. On the machine I tried it on, the startup time was basically the same as OOo. "Reads and writes your Word documents seamlessly (Microsoft Word .doc 6.0 to 2007)" AFAICT the only advantage over OOo would be if it can read OOXML. (Although OOo can't write OOXML, I can't see why anyone would care; if you save in an older Word format and give the file to Word 2007 users, they'll still be able to read it.) The price is $80 US. Although that may be a lot less than full retail price for Word, it's a heck of a lot more than OOo. And of course I'd have to live with all the usual hassles of proprietary software. I won't get an x64 version unless they deign to compile one for me at some point in the future. I won't be able to upgrade without paying money. Sorry, I'm just spoiled -- apt-get and OSS work fine for me.
I run Linux on my ARM-based slug, which I use as a music player. I'm happy with the result (a cheap, always-on, low-energy music server), but it was kind of a pain to do the install, and the resulting system is broken in enough ways that I wouldn't want to use it very often as a general-purpose computer. Ubuntu is known as an easy-to-install, easy-to-use, full-featured desktop system that Just Works. If they can extend that to ARM-based subnotebooks, etc., then IMO it really is a big deal. Most people who own an x86 or x64 desktop machine really don't need one. All they need is a computer for word-processing, web browsing, and email. A lot of people really are finding out that they're just as happy with a cheap eeePC or whatever. If I could have Linux on an ARM-based notebook computer, and the battery would last for weeks in sleep mode, that would really be something I'd consider buying. But no way would I consider it if I didn't get the same freedom from hassles that I get with Ubuntu.
The OP gives Ndiswrapper as a specific example, but asks a general question, and so far all the replies have been about the general question.
What about the specific situation of Ndiswrapper? There's a saying that "bas cases make bad laws," and Ndiswrapper is sort of like that -- it isn't a typical example of OSS.
Okay, first off let me say that I have two machines on my home network that have Ndiswrapper on them, and I'm grateful that it exists, because it saved me from having to drill holes through my hardwood floors and pull cables from the downstairs to the upstairs.
However, I'd be surprised if anyone had ever been under the impression that Ndiswrapper was anything more than a horrible, nasty, dirty kludge with no future ahead of it. The basic problem is that the manufacturers of the wifi cards don't disclose the relevant technical information that would allow third parties to write drivers, and they also don't support operating systems other than Windows. Anything the OSS community does to try to work around that is bound to work badly and be unsatisfactory. I've already seen that any time I upgrade from one release of Ubuntu to the next, wifi breaks, and I have to go back through all the steps of installing the drivers again. There's also the problem that binary blobs make it difficult to debug kernel crashes.
All of these problems show that ndiswrapper has always been nothing more than a band-aid, and nobody should have ever expected it to have a future.
The only real solution for the future is to spread good information about what cards work with OSS (no binary blobs). The FSF has some info here: http://www.fsf.org/resources/hw/net/wireless/cards.html . The trouble I always have with this kind of situation is that these online lists are always out of date and inaccurate, and they also tend to systematically overstate the quality of support, e.g., when you I the OSS driver, I can't get it to work at all, or if I do get it to work it crashes all the time, or the full functionality isn't supported.
This is all qualitatively different from the situation where you just have an OSS project that doesn't have ongoing support. A more typical example of that kind of thing would be sox, which is a command-line utility for converting sound files between different formats, adding effects, and playing sounds. Its author hasn't been supporting it properly for a long time, so less and less of its functionality is working on, e.g., a fresh install of ubuntu. It's gotten to the point where, for me, it's basically useless. But that's no big problem, because other people have picked up the slack by writing similar software to replace it. The difference with Ndiswrapper is that the problem is more fundamental. The things that make Ndiswrapper a kludge are inherent to its purpose, which is to be a kludge.
I agree with you about Ooo; that was the point I was making in the g'g'g'g'grandparent post. Of course, upgrading Ooo may be less of a problem than upgrading Word, since it doesn't cost money.
I disagree with you about emacs+latex; the point I was making in the g'g'g'g'grandparent post was that if you really want stability and the ability to keep on using old software, you need to pick software that is high quality and extremely mature. Emacs+latex is a good example of that. The basic implementation of tex/latex is extremely stable and mature now, and all that's been changing in the last couple of decades is some of the macro packages that people use layered on top of that. Likewise, I don't see how I could get in a situation where an old version of emacs was unable to read files written by newer versions of emacs -- should be good until ascii and unicode become obsolete, which I don't think will happen very soon.
I don't want to say the s-word, but after I bought something from ThinkGeek, they started sending me marketing emails. I don't recall being presented with a choice about whether to opt in or out of marketing emails when I made the purchase. It was UCE (unsolicited commercial email), but you could argue that I had already established a commercial relationship with them. All I can say is that personally, if I buy from an online retailer and then they send me ads via email, my personal decision is not to do business with that retailer again. One very practical reason is that once they send me ads, I'm going to blacklist them in my email filter, and that would make it difficult to do business again. I'm not accusing ThinkGeek of being evil criminals with handlebar moustaches or anything, but it's just like any other business -- if I don't find it pleasant to buy from them, then they've lost my business.
Something is stopping people from using an old copy of Word '97, and what's stopping them is what I stated in the g'g'grandparent post: they won't be able to read files that other people send them, written by newer versions of word.
Interesting link, thanks. But note that the converters won't work with word '97, which was the case that the g'grandparent poster proposed. Also note that it only works on windows, so if you run word on macos you're out of luck.
I was talking about case (a), using your old version of the wordprocessor to read a new-format file written by someone else. You're talking about case (b), using your new version of the wordprocessor to read an old-format file. MS makes sure that compatibility is great in the (b) direction, and lousy in the (a) direction. I guarantee you that word '97 will not be able to read all files written by word 2007.
You beat me to a bunch of the points that I tend to trot out every time ajaxy stuff comes up on slashdot. But to be fair, I think there are some counter-points to be made in google's favor.
As far as customer and product, what you say is true, but it's also true of newspapers, for example. The subscription price I pay for the New York Times is only a small fraction of my share of what it cost to make the newspaper. The majority of their revenue comes from advertising. Why, then, isn't the NY Times as bad a paper as USA Today? Well, market forces are at work. The NY Times has a market niche (showing ads to pointy-headed intellectuals), and wants to protect that market niche, so they have to keep on producing the kind of newspaper that attracts the eyeballs that make up its niche. Same with google. If gmail, for example, stopped being better than the competing free email services, google would be hurt by marketplace mechanisms, because they wouldn't have as many eyeballs to show ads to.
As far as free-as-in-speech versus free-as-in-beer, that's a relevant comparison for google docs versus ooo, but for most people the actual choice is google docs versus word, which isn't free-as-in-anything.
Arbitrary changes: actually, nobody is protected against changes to software. Ooo and word could both change overnight. Sure, you can continue to run the old version that you liked better, but that's a pretty poor option with word (you'll soon find yourself unable to open other people's files, written by newer versions), and even with ooo it's not very practical, because the old version won't get bug fixes or new features, and will eventually be incompatible with your OS's libraries and APIs. If you really want stability, the only real solution is to pick old, mature software that's extremely stable (e.g., emacs and latex), not recent software like word, ooo, or google docs.
I'm a community college physics teacher in California, which has the best public college and university system (community college, cal state, UC) in the world. What makes me want to stab myself in the eyeballs with number 2 pencils is the way my colleagues on the faculty all want to send their kids to expensive private schools. It's nuts. If their kids are hot shots, they can go to a UC. If they're really, really hot shots, they'll get into the US flagships, Berkeley and UCLA. If their kids are not quite that hot academically, they can go to a lower-ranked UC, or to a cal state. If they blew high school but are now a little more motivated, they can go to that community college, and then transfer. I have a ton of students who transfer to Berkeley and UCLA after 2 years of community college. Many of them are not particularly brilliant, just ordinary students who worked hard. By comparison, it's just insanely competitive if you're trying to get into those two schools straight out of high school -- about as hard as getting into an Ivy.
One common justification I hear my colleagues give for sending their kids to expensive private schools is that they don't want them stuck in large lecture classes where they won't thrive. Sorry, but that's not a reasonable reason to double the cost of college (or quadruple it, compared to community college followed by UC). There are only a few of those huge freshman lecture classes, and students who try hard enough can avoid them completely, especially if they've taken AP exams.
Another common justification I hear is that it should be the kid's choice. Well, that might make sense if it was only the kid paying for it, but usually that's not the case.
That could be good advice if the student is both very smart and very dedicated. What I see much more often is that students have unrealistic expectations about the lifestyle they're going to live at the age of 18. They want to have a nice car, and they don't want to live at home with Mom and Dad. So they work 30 hours a week while taking 18 units. Oops, then they wake up one day and find out that (a) their GPA is 2.0, (b) they took W's in so many classes that college is going to take them 6 years instead of 4, and (c) they missed the part of college where you're supposed to get an education, as opposed to just cramming for tests and doing the minimum to get by.
Yeah, in general I'm wondering what's the right thing to do with stocks I own that are very low in share price. Normally I don't try to "pick stocks" at all, and I don't buy or sell based on movement in prices -- the only reason I own individual stocks is that I wanted to have a socially screened index fund that met my own idiosyncratic social screening criteria. But my understanding (corrections would be appreciated!) is that when a stock goes down below a certain share price or a certain market capitalization, it can get delisted. After delisting it can only trade as a pink sheet, and there may be problems with liquidity. Here are the NYSE's listing criteria, which include (with some oversimplification on my part) market capitalization of 100 M$ and a share price of $4. They don't lower the boom immediately if a company falls below their standards, but when you're getting down to share prices of $1 and $2 it's pretty clear that it could be delisted. So what's the smart thing to do? If I sell, it costs me $7 for the trade, and I could just be locking in losses. If I hang onto it, I could lose my ability to liquidate it, and even if it retains liquidity I'm also not clear on what happens as a practical matter if I own a stock via an online brokerage, and then it goes to the pink sheets market -- hassles for me? expenses?
How about (a) toning down the arrogance, and (b) not making statements that are demonstrably untrue. I ask and answer questions on usenet pretty much every day of my life.
Where I live, in California, they use electronic voting (rotary wheel and buttons, no touch screen) with a backup in the form of a paper strip that the voter can see being printed behind a plastic window. So it's pretty similar to yours in NC, except that there's no optical scanning required. And let me tell you as a teacher who's run scantrons, optical scanners are extremely unreliable. It's very common for them to get read incorrectly, especially if the user had to erase and change something. Paper jams also happen pretty frequently, and you can pretty easily get a scantron damaged by the machine. The company that sells the scanners we use claims that they're very reliable if they're calibrated properly. That might imply that the ones we have where I teach aren't getting serviced often enough, but if so, that doesn't make me feel very safe about the idea of trusting votes to something similar. How do we know that the ones used in NC are getting serviced often enough? On the voting machines we use here in CA, which print on a paper strip, I'm sure it's possible for there to be problems with the printer, such as running out of ink, running out of paper, or a paper jam -- but all of those would be immediately obvious to the voter. Having your vote miscounted by an optical scanner is something you'll never know happened.
I think a lot of the problems with these systems in the U.S. flow from the fact that they don't want the voter to be able to prove to a third party that he voted a certain way, i.e., you have deniability about how you voted. This seems like a good thing in theory, so that you can prevent coercion or vote-selling. But in reality I've never seen any evidence that either of those has ever been an issue within living memory, and in any case when people vote by mail it is possible to sell votes or be coerced. If we don't demand this deniability feature for voting by mail, why should we be so hung up on it when it comes to voting at a polling place on election day? It would be much easier to satisfy voters that they were getting their votes counted correctly if we got rid of deniability. Give the voter a code number when he votes, and when he goes home, he get on the web, enter his code number, and verifies that his vote was counted correctly.
Thanks for your response. I deleted the email, but if you want to find it in your logs, the string to grep for would be either lightandmatter.com or fullcoll.edu (I think it was the former). If you don't want anyone to feel like you're spamming them, then the solution would be pretty simple: don't send unsolicited commercial email to people who have no business relationship with you. Adding me to a no-email list would be irrelevant, because I'm now blocking oreilly.com in my spam filter, and in any case asking people to opt out of spam isn't a reasonable way to deal with the spam problem.
I'm afraid I'd want to steer clear of any book on internet marketing that's published by a company that spams. I'm a college physics professor, and O'Reilly has spammed me to advertise their book Heads First Physics. I was really surprised and disappointed by the spam, because I have a whole bunch of O'Reilly books on the shelf next to my desk at home, and I'd been under the impression that O'Reilly really "got it" when it came to open source and the internet. But I really am pretty firm about not being willing to do business with spammers.
what do you mean by a proof being valid in another set of axioms? that there is a simple way to translate the proof into the 2nd system and it'll be true then?
yes
Which central theorems aren't already proven? Principia Mathematica did the basics and I always assumed more advanced theorems became proven as required. Maybe someone just wants to do Principia Mathematica volumes II through L but doesn't that already in effect exist?
Well, one issue is that the Principia Mathematica undoubtedly has errors in it. No human being could ever write a book of that length without making a single mistake. So there could be some value in producing the same results on a computer, which doesn't make the same kind of mistakes a human does. Another issue is that the Principia Mathematica works within a certain axiomatic framework, and I believe that framework is different from the most popular axiomatic framework used today, which is Zermelo-Fraenkel set theory, with the axiom of choice (ZFC). (They were both produced around 1910, but the WP article says that PM used a complicated system of types, which is probably different from anything in ZFC.)
More broadly, for people who are interested in the foundations of mathematics, there's no clear way to decide that one set of axioms is superior to another. Some people feel that ZFC is too strong, because it asserts the existence of things that can't be explicitly constructed. Those people may be interested in seeing how much of mathematics can be proved using purely constructive methods. With a weaker set of axioms, some results may be impossible to prove, while others may be possible to prove, but the proofs may be extremely long compared to the proofs in ZFC -- so long that using a computer may be a real help.
Still other people are interested in seeing what can be done in an axiomatic system that's stronger than ZFC. For instance, such a system may have sizes of infinities that are larger than the sizes of the infinities in ZFC, and they may even have creatures like the set of all sets. One thing that tends to happen when you try to make stronger axiomatic systems is that it becomes much more difficult to avoid internal inconsistencies. I can imagine that a computer could be helpful in finding things like that. If you're fiddling around with the computer proof system and it comes up with a result that 1+1=3, then you've learned something interesting: your set of axioms is bogus.
One thing you really have to be careful about if you're working in this kind of field is that you don't want to inadvertently assume some result that someone else has proved in some other axiomatic system, and that seems obvious to you, but that actually isn't provable (or hasn't yet been formally proved) in the system you're working in. That's the kind of thing where I'd imagine a computer system would be really helpful. It wouldn't allow you to make those assumptions. In general, if you look at almost all published mathematical work, it never states which axiomatic system it's assuming, and that's because in most fields of mathematics we expect that the results are unlikely to depend on which particular foundation you're working in. E.g., I doubt that Wiles' proof Fermat's last theorem even bothers to state that it starts from ZFC, and most mathematicians probably have a strong expectation that it doesn't matter whether you pick some other foundation, Wiles' proof will still be valid. Nevetheless, it's possible that that's not true. Abraham Robinson, for example, claimed that ZFC had been carefully engineered to make the real number system work correctly, and therefore claimed if you wanted to use a different system of numbers (such as the hyperreals, a system that includes Newton- and Leibniz-style infinities), you might be better off using diffrent axioms.
Godel's completeness theorem states that every statement that follows from the axioms is in fact deducible from the axioms in finitely many logical steps
How would you even define a notion of "follows from the axioms" that didn't presuppose proof in a finite number of steps? If there were a distinction between "provable" and "provable in finite steps," it seems like you'd have to define what it meant to prove something in an infinite number of steps, and then show that for every infinite-length proof there was also a simpler, finite-length one. But I'm really having a hard time imagining what it would mean to have an infinite-length proof of something.
it's Acrobat Reader 5.0, x86 linux 5.0.10 Nov 8 2004 13:14:17.
You're running an extremely old version. The current version is 9.
Huh? I create PDFs all the time, and don't own a copy of Acrobat. I use pdftex and inkscape, but there's scads of other software that can do it, e.g., Scribus if you want GUI desktop publishing. This is all on linux, but there's tons of PDF-creating software on Windows as well.
To disable js, go to Edit, Preferences, JavaScript, and uncheck "Enable Acrobat JavaScript".
Even if the js-related security bugs are fixed, it's still a privacy issue, because js in a pdf file can be used to track who's reading a particular document.
Personally, when I see that a piece of software has a long history of security problems, I take that as my cue to remove it from my system. I don't really care that they keep fixing the bugs. The fact that it has this history demonstrates that the software wasn't written with the correct attention to security, and it's likely to have more such problems in the future.
If you're running Linux, xpdf starts up extremely fast, and that's why I use it as my pdf plugin in Firefox. If you want something a little more modern, try evince.
People have posted saying that on Windows, you should switch to Foxit, but the article says that the security flaw was found first in Foxit, and only later in Adobe Reader. I actually tried to get the science division at the community college where I teach to switch to putting Foxit on machines in the student labs as the default pdf plugin. However, when the faculty were testing it, they found that it was not correctly displaying some of the pdfs they were using.
This is an approach that's been worked on for years and years now, and there hasn't been any rapid progress. Electromagnetic shielding may ultimately work, but it has a lot of problems to overcome. Without some kind of significant technological progress, the radiation dose for astronauts going to Mars is a real showstopper: http://en.wikipedia.org/wiki/Health_threat_from_cosmic_rays
Engineers have studied a variety of electromagnetic field configurations for this. Electric fields have a problem because any field that repels positively charged particles will attract negatively charged ones, and vice versa. Also, large DC electric fields tend to discharge violently.
With magnetic fields, one problem is that you need magnet coils that carry huge currents. If they're not superconducting, then you're talking about huge amounts of power, way more than is really practical. If they're superconducting, then you're trusting your life to a type of technology that's notoriously prone to failure; normal superconductors need liquid helium temperatures, which are very hard to maintain reliably, and high-temperature superconductors still need liquid nitrogen temperatures (and also may be decades away from being ready for this type of application).
In this article they talk about using an AC field, although they don't provide many details. One thing I wonder about with an AC field is radiative losses. They say, "The approach will probably also work with a field that is not on constantly, but cycles on and off - conserving the power that is precious on long-term missions." I totally don't get the idea here. They make it sound like they're just going to have it be pulsed, with a low duty cycle. But then won't it only provide some small fraction of the desired protection?
Another approach that looks promising is to make the spacecraft from low-Z materials like plastics; most of the hazard from cosmic rays is actually from secondary radiation (which is why thicker shielding actually *hurts* you, for any practical thickness).
I think the real question to ask is whether there is any valid, objective reason for sending a human to Mars. Crewed spaceflight has never been a good value for the money in terms of scientific research, compared to probes. Or if the motivation is some kind of romantic vision of heroic exploration of a new frontier, then I think we need to be more realistic about the prospects for permanent, economically viable settlements, which are probably at least a century away.
There's another project called Y Windows, which also aims to replace X with something that has less historical cruft.
The real question in my mind is whether this kind of thing does anything to address the big problems users are really encountering. Of course, not every open source project has to make large numbers of users happier. But the author of Y Windows says, for example, "I've got tired with the state of desktop GNU/Linux. Most of the problems that I see with it can be traced back to the underlying window system, X. So I decided to write its successor... "
For me as an end-user, the big issues are simply hassles with xorg not correctly recognizing LCD screens, so that it sets them to an inappropriate resolution, or the image appears offset. I have close to zero interest in gaming, so personally I just use the onboard video of my mobo, with only 2-d driver features, but the impression I get from people who do care about gaming (or fancy WMs) is that the big issue is drivers, not the internal structure of X.
As far as programming, the structure of X also seems like kind of a non-issue. Sure, X's APIs are heinously ugly, but almost nobody uses them directly.
The advantages listed by the article are things like a more manageable code base, a smaller memory footprint, and elimination of rendering artifacts. To me, none of those seem like major issues that I was all that worried about.
For anyone who doesn't want to sit through the entire (long) Flash presentation by Bruce Nash, here's a micro-summary. Basically, he predicts that it will be around 8 pm Pacific time that they call Virginia and Pennsylvania for Obama, at which point it will be clear that Obama has won.