On iliad you can annotate, but the method ain't perfect. See the end of this article [arstechnica.com] for a review.
Ah, it stores everything separately, and doesn't seem to have anything but a "pen" mode. Since my handwriting is somewhat poor (and my tablet-writing is even worse), the ability to add typed notes would be nice (via a little on-screen keyboard, perhaps? I'm not asking for OCR to read my scribbles). The biggest thing for me is underlining/highlighting - this can be done neatly and efficiently in any PDF which isn't simple scanned. Okular for KDE4 seems to do a decent job at it (the annotations are also stored separately), but it's still a bit in the early stages of functionality.
The big screen is nice - if/when I ever get an ebook I'd be tempted to want one that's about a 8.5x11" screen to view pages 1:1. Still, the eink and battery life would have to be awfully nice to choose it over a low-end tablet (e.g. HPs start at $900).
This is a neat hack; I'm mildly surprised that you can fit a decent version of Wikipedia in under 4 GB. The text, sure (especially bzip2 compressed), but a decent set of images? Anyone have a breakdown of exactly which version of Wikipedia this is?
The static Wikipedia pages appear to have not been updated since April 2007 (the February 2008 ones stop just before "en"). That version comes in larger than 4GB, but static HTML pages are less efficient, I would think, than what this guy did parsing the XML data.
These days, though, WiFi is available in so many places that even if I owned one of these devices I probably wouldn't use up the flash space with an offline version of Wikipedia.
Side note about the iRex. The ebook version of the reader (which, notably, lacks WiFi compared to the more expensive version) appears to be $599 MSRP. I personally thought the Kindle was expensive at $400, wireless service included. The WiFi iRex is $700, which is getting into the territory of a few low-end (or used, I'm sure) tablet notebooks. I understand that the battery life and screen readability of these things is supposed to be pretty good, though.
Anybody know if the iRex or any other ebook reader has the capability to annotate PDF files? I do a quite a bit of reading of PDF documents, and I find myself printing them all too often so that they're easier to read and I can make notes. These ebook screens are supposed to be easier on the eyes than a standard laptop screen, so all that's left is the ability to make annotations.
On the Media had an interview (transcript and mp3 download available) last week with Internet Archive co-founder Brewster Kahle. about his personal experience with the national security letter. Interesting stuff, but perhaps not much new if you've been keeping up with this.
Anyone looked at KDE 4.0?
I cranked it up in a VM and had to look twice to be sure it wasn't GNOME. Most of KDE's signature customizability is gone, and (like GNOME) it's not just a matter of missing GUIs for tweaking settings; the settings themselves are gone into hard code.
This is temporary, and is a common complaint about KDE 4.0. The idea with KDE 4.0 was to ship what they had to encourage further application development. There are lots of changes to KDE, including using a new version of QT (the underlying toolkit).
The basics are there, but customizeability, as you noted, is lacking. From what I understand, that flexibility (especially in terms of the main panel) will return with KDE 4.1, to be released this July.
KDE 4.0 isn't for everybody. After reading about some of these limitations, I decided to wait until KDE 4.1 before upgrading my Kubuntu laptop's KDE version. As I understand it, KDE 4.1 will bring applications like the PIM framework up to speed, and I should be able to make my desktop look and work like I'm used to with KDE 3.5 (a substantial alteration from the default).
KDE hasn't abandoned the philosophy of a very flexible user interface, it's just taking time to re-implement the features in the serious overhaul that is KDE 4. I can wait.
There are add-on repositories for Fedora, check them out if you need something not included in the "Everything" Fedora repository. I am continually surprised that when I find an interesting application on the Internet that it is usually already in Fedora, or in Rawhide and I can rpmbuild it myself.
It's been my experience that the more repositories you add, the more you run into issues of conflicting package offerings between them. I try to avoid it if at all possible and simply use the "official" community-supported stuff.
This is because apt-cache only searches the local cache, while Yum always ventures out onto the 'net to fetch the latest package and file lists.
I know that used to be true, but when I just tried it yum gave no indication that it was going out and retrieving any file lists (but perhaps it still was in the background). For me, that seems like a dumb default setting. The "apt-get update" system seems to work well: update your repository info when you want to, and work with it from then on. That way if you do a few searches you aren't looking for updated repository info for each and every search. Even if there is a way around this with a command-line switch, it still seems like a dumb default.
Comparing RPM to apt-get is apples to oranges. Either compare RPM to DEB, or yum to apt-get. I never had to bother with dependencies when using yum, just as you've never had to bother with dependencies using apt-get.
I completely agree. Since my distros of choice over the last 5 years have been Fedora and Debian/Ubuntu, I've had a fair bit of experience with both yum and apt-get. Yum, at least as of the Fedora 8 install on my desktop, is simply not as good (IMO) as apt-get in Debian or Ubuntu for two reasons:
1.) yum is slow, horribly horribly slow. I think it may have gotten a little better in Fedora 8, and I've heard that they're putting serious work into it. Hopefully Fedora 9 will be better, but it never ceases to amaze me how long it takes to do a "yum search" to look for a package compared to "apt-cache search".
2.) The package repositories for Ubuntu (which is derived from the huge repository from Debian) are larger and more complete, at least for the random software I tend to look for. Again, Fedora is gaining in this regard, the community-supported package setup is starting to rival Ubuntu's universe, making this a huge step up over the old RedHat 7/8/9 days compared to Debian at that time. When it comes to software outside of either repository, RPMs tend to be more common than debs, which is an advantage for Fedora.
So yum (and the standard underlying repositories) are behind in those respects compared to apt-get, but the difference is shrinking. In yum's defense, I think they implemented package signing as a default requirement before Debian did, but I could be wrong on that.
I've run Fedora on my desktop for a while, but Kubuntu on my laptop. I honestly don't know what I'll install on my desktop next. I usually skip every other release, and since I'm on FC 8, that means waiting until FC10. This might be good anyway; I'm a KDE user, and KDE 4.0 just doesn't look feature complete. Best to wait until KDE 4.1 polishes everything a bit more, perhaps. I'm debating whether to try out the latest Kubuntu on my laptop when it's released this month to try out KDE 4.0.
Electricity use for lighting in North America is only about 1% of the total.
Got a source for that? Some quick Googling shows that there is serious disagreement about the number, but I've seen estimates from about 3% to 20%. This post illustrates the wildly different numbers.
It's amazing to me that we haven't pinned down this number better, but some people like to include things like the amount of extra air conditioning required to compensate for the heat generated in the summer and subtract the amount of heat that the lights generate that something else didn't need to in the winter. That seems like it's complicating things unnecessarily, but amongst the sources I've seen, I haven't found anything below 3%.
Also, even if heavy industry is such a large user of power, that doesn't mean that home energy conservation won't have a significant effect. The reason is straightforward: electricity is very costly to transport long distances. Heavy industry can locate near large dams and other sources of inexpensive power, reducing both production costs and transportation costs. Most people don't live that close to cheap power, and thus the cost savings that they realize by energy conservation provides a decent measure of the burden they're eliminating from the overall system; that's not even getting into the externalities of power production like pollution and greenhouse gases, which only serve to boost the returns from home energy conservation.
Even in my home, if I turn all the ligths off, I use less electricity made with hydro/nuclear/wind power and more natural gas in the furnace, so overall turning the lights off generates more smoke than leaving them on.
That might be true for you, in the winter. Coal is the predominate source of electricity for most people. Transporting natural gas is costly, but it's not burned until it gets to your home. On the other hand, a non-trivial amount of electricity is simply lost due to resistance in transmission - that's coal burned.
Oh, sure, there are probably PhDs out there blogging. Okay, okay, I'm kidding - I sincerely doubt it - unless they were useless in their fields to begin with.
Actually, there are quite a few PhDs out there blogging. They are hardly "useless in their fields", at least the ones I read; they tend to be some of the more high profile people (and the blogs simply give them an even higher profile). Two cases in point for economics: Greg Mankiw's blog and Marginal Revolution, a blog by two George Mason profs with occasional guest bloggers.
Blogging is actually fairly amenable to the goal of many academics: to share information and debate about it. The biggest downside that I see is that blogging is fairly time consuming. Mankiw turned off comments to his blog because he didn't have time to moderate them, so his blog became more of a one-way street.
Of course, econ is just one field; I honestly don't know how prevalent blogging is in other fields.
Why does a professor have a copyright on his lectures, anyway?
If anybody owns a copyright on those lectures, shouldn't it be the school?
This is an interesting difference between academia and the business (i.e. "real") world. I suspect that it comes from a combination of considerations and cultural aspects:
1.) Professors do research, and submit that research to journals for publication. Those journals often require the professor to sign over the copyright of the paper before publication. It's easier for professors to do that if they have the copyright in the first place.
2.) What about lectures? Some (a few) professors make a ton of money (or little money and a lot of recognition, in some cases) by selling their books. These books start out as lecture notes, typically, especially at the graduate level. Professor's salaries don't vary that much, so this is one way in which the better/harder working/better known professors can earn relatively more pay. That keeps them at a university that can't afford to pay them what they'd get if they quit and just published their textbooks, which is good for the university.
3.) A big consideration is probably the culture that a professor's work is not so much for the university itself, in the sense that professors move between universities all the time and take their research/lab/lecture notes with them. Would you honestly expect professors to have to somehow re-write their lecture notes upon moving to a different university? It just doesn't happen.
4.) Tenured faculty have a fair bit of power over university policies, if they collectively put their minds to something. While works created by staff and non-faculty might be the automatic property of the university, faculty (in the U.S. at least) typically get the copyright for much of what they create.
It's tempting to draw parallels between programming and research/lecture notes. The cultures, though, are quite different. In general, academics share their resources pretty openly, at least up until the point it becomes a textbook. To the extent that academics write code, it often isn't under an explicit license at all (which can be inconvenient if you want to properly include it in something for redistribution).
Patents are another issue altogether, and one where the university stands to make a great deal of money. I'm not at all familiar with the general breakdown of rights about those, but it seems that both the inventors and the university get a cut in many cases.
So what's up with this professor? It sounds like somebody is peeved that his students aren't attending class and would rather pay somebody to come in and take notes for them. Many good professors do the exact opposite and post class notes online, though they may not include quite everything that's worth getting from a lecture.
If notes were a substitute for a good lecture, most of us would learn by buying the best notes from the best professor in the world on a subject (which is only sometimes available as a textbook). On the other hand, a bad lecture is worse than a decent set of written notes. The solution is not to sue them, but to improve your lectures!
How many people in a computer lab, or small business read through the license of "free" software that's as common and trusted as Adobe Reader? (And would expect this strange, "you have to download it each time for each computer" implication?) Not many. I bet if you contacted the sales people they wouldn't even have a way to sell you a site license for reader.
That's the whole point: the license is completely at odds with both de-facto policy by Adobe and common practice by users.
If the point was to keep people from distributing copies of Reader, why not just simply add a clause that you can only re-distribute it for your own, personal use or for the internal use of your organization? Perhaps the overkill in the license is the result of the way contract law seems to work (IANAL): if one part is found unenforceable, other parts can be salvaged. This gives lawyers the incentive to write more restrictive licenses and then selectively enforce the terms. Users end up just clicking through the pages and pages of legalese and ignore the terms, both reasonable and not. It's hardly an appealing outcome.
Basically they want you to download a fresh copy of the Reader installer each time you install it.
Actually, I'm not even sure you can do this. I'd be curious what the law would say (if anything) about whether downloading another copy constitutes another new license, and thus another "Permitted" computer.
It's great to see some outrage about stupid license agreements. There is widespread apathy about these agreements and their restrictions, leaving companies with the option to selectively enforce crazy terms if and when it is in interest.
On a semi-related note, the choice for many people is not to buy some expensive software or use the open source equivalent, but to illegally copy the expensive software or use its open source equivalent. Open source software does fairly well competing on features and stability alone, but the total lack of restrictions on use is something that is often overlooked, at least by home users. There may still be some misunderstanding about open source regarding use, though I hope that's fading. The availability of source is great, but frankly, that's not going to directly affect what home users can do with their software.
Rather than ignore licenses and laws we don't like, it's far more productive if people make an issue about them to get them changed. The outrage over AT&T's policies is one example of this (though that seems to have stalled, aside from an ongoing lawsuit), and this is another. Ignoring these terms and hoping they aren't enforceable is just a recipe for trouble. As a side benefit, if people become disgusted by closed source license agreements, open source software will look more appealing.
Anyone who chooses to upload anything to a public forum/gallery should be aware that some of these websites will claim the right to do whatever they want with that material. Back in 2003, I even stayed at a hotel where the internet access had such a clause; they claimed the right to reproduce whatever you uploaded through the system. How enforceable are such terms? I'm not a lawyer, but I don't think we've seen a sufficient number of court cases come out of license agreements like these.
Adobe's not exactly known for their reasonable EULAs. Just take a look at the EULA for Adobe Reader. This is software that Adobe is trying to get on all the computers it can. The license, however, permits only the installation on one primary computer and one mobile computer (note that "Permitted Number" is 1). I've gone so far as to contact Adobe customer service and ask them what's going on - this goes completely against their marketing policy. Amusingly, they send all their customer service responses via PDF over email. Their official response?
With regard to installing the software on more than two computers and its use at the same time. I need to inform you that although Adobe Reader is a free software, Adobe maintains its distribution rights. Thus, as per Adobe policy there is no provision to use the software on more than two computers simultaneously.
We apologize for the inconvenience this may cause.
Please note that, single-user Adobe branded product that is installed on a computer at home, you can also install and use the software on one secondary computer of the same platform at office or on a portable computer. However, you may not run the software simultaneously on both the primary and secondary computers.
It's clear that Adobe has no intention to actually try to enforce this restriction, but it suggests that organizations with computer labs and such are supposed to negotiate a volume license with Adobe. I think the Reader license is simply boilerplate recycled from other Adobe software, but it's clear that whoever is responsible for Adobe's licenses isn't in touch with what Adobe actually wants to have regarding its licensing (at least from a marketing perspective).
Does this mean if I replace my nice Porterhouse Steak with a Sirloin Steak "It's just as good"? I would be tempted to call that inflation by any name.
I agree. The quality of the sirloin steak is clearly lower, and they're not terribly great substitutes. Typically beef prices move up and down together. Substitution bias is if (just an example) porterhouses became much more expensive relative to ribeyes/t-bones/your pick here. You might be very close to indifferent between them, and you'll pick the one with the lowest price per pound. If the CPI only looked at porterhouses every year (it doesn't, this is a simplification), it would show higher inflation than you really "feel" - you can have very close substitutes where it only takes a small price change for you to prefer one over the other.
As you can imagine, this is hard to measure. To the extent that the CPI tries, there seems to be a consensus that it still biases inflation upwards.
The real CPI has not been reported since 1986. Here's some of the tricks used.
The "real CPI" you claim is a crude measure which deserved to be fixed. A fixed basket doesn't represent what people actually do, and the current methods of collecting price data get at that better.
Even if you disagree with the methods (and your link is not exactly an objective source, show me a paper published somewhere respectable or at least written by a respected academic), you won't arrive at 10-12% inflation that you claimed. That's quite simply a blatant exaggeration.
Gasoline has more than tripled in price in the last decade (1.04 to 3.27) .
1.) Gas does not make up a terribly large portion of people's expenditures 2.) By picking the last decade, you conveniently start with the lowest recent price of gas and end with the highest (by recent I mean since the 1980s, but we really should be talking about a measure of the real price of gas). In the 8 or so years before that, the price of gas fell about 30 cents in nominal terms
Housing? Doubled or tripled. Food? Don't even ask. Sure, you can substitute for some items, but for the stuff you actually NEED, like a roof over your head, food in your stomach, and transportation to and from work?
Like I said in the grandparent post, the full CPI index, with its current ~3-4% rate of inflation, includes housing, energy, transportation, etc. No amount of fiddling with quality or substitution or anything like that is going to change that into the 10-12% you claimed.
Also, the calculators of the CPI have already done the "substitution", to such an extent that they use USED cars instead of new cars, and "owner's equivalent rent" instead of the actual cost of the roof over your head. Its a lie.
Much of your link is simply FUD. Seasonal adjustment biasing the CPI downward? No, it simply makes the CPI less volatile - parts of the year where gas is relatively more expensive are smoothed out with the parts where gas is relatively cheaper. The stuff about the "core" rate? Most news articles I read make a pretty clear distinction if the inflation measure they're using includes energy and food. To the extent that the CPI tries to fix for substitution bias, reputable work using a large dataset shows that the CPI is still likely overstating inflation significantly; their result is that inflation is an average of 0.8% lower than reported, on data covering 40% of consumer expenditures from 1994, 1999-2001, and 2002-2003.
Again, there's simply no evidence for your claim of 10-12% inflation.
That won't even keep pace with inflation. Real inflation (not the CPI bs that the government hands out every year, which excludes stuff like fuel) is running between 10% and 12%.
The CPI is released in several forms. It's usually reported in the news as either the overall CPI index (which includes food and energy), or the CPI less food and energy (sometimes referred to as the "cold and hungry" CPI). Neither is anywhere close to 10-12%. See for yourself. Overall inflation, at an annual rate, based on the last 3 months is 3.4%. Based on the last 12 months, it's 4.4%. Without food and energy, these numbers are 2.4% and 2.3%. Inflation is up from its relatively low values in the last couple of decades, but still far away from the early 80s. Also, many economists believe that the CPI in fact overstates inflation. Why? People will substitute from goods which became relatively more expensive to those which haven't. To the extent that the basket of goods that the Bureau of Labor Statistics uses to calculate CPI doesn't take this into account, it will make inflation seem larger than the average person will really feel.
The CPI is supposed to measure what typical households buy, but if you can only pick one rate of "inflation", it's usually the most reasonable. Even if you were to argue that NASA spends a great deal of its budget on fuel (which I highly, highly doubt), that fuel is not directly petroleum-based. The solid rockets are based on ammonium percholorate (according to this). The shuttle itself has engines based off of liquid hydrogen and oxygen.
The real reason Blackle is a waste of time is that dark pixels on a flat panel don't save electricity. The backlight is still on, it's just that the LCD pixels are blocking the light emanating from it. (I'm rather sure you don't have your computer hooked up to a plasma display.)
I agree. But even if you take their claims of energy saving as face value, the magnitude is pitiful. Another poster in this story calculated that a CFL saves a large fraction of a megawatt hour in its lifetime. Replacing a single bulb with a CFL will save more electricity in its lifetime than all the users of Blackle have for its entire existence so far (over a year), going by Blackle's own numbers.
So... don't burn coal for power and don't use compact fluorescents. When did two wrongs become a right?
It's not a matter of two wrongs, it's a matter of tradeoffs (as with most things). I'm a big fan of non-coal power (including nuclear), but the existing coal plants aren't going to go away any time soon, and we seem to keep building more around the world.
For most people, artificial light is a necessity. These days, they have a choice between incandescents, CFLs/fluorescents, halogen, and maybe a few LED options. CFLs are much more efficient than incandescents or halogens. LED lights are still somewhat expensive. If you use a CFL, don't recycle it, and its total mercury emissions are less than the emissions from the power plants used to produce the extra electricity required to power an incandescent, it's clearly better to go with the CFL.
Of course, the best thing to do from an environmental perspective is to simply recycle your bulbs. I've mentioned this before on/.; there are a number of household items that we need to dispose of properly. Things like CRT monitors/TVs, large or lithium-ion batteries, etc. The easiest thing to do is to set all these items aside until they build up a bit, and then cart them off to the nearest recycling center. For me, that's just across town, but I'm lucky in that regard.
Being a rational environmentally-conscious person means that you should take actions which require the least expense (in terms of both time and money) which cause energy/pollution reductions in the greatest quantity. That's why Blackle is a total waste of time (500,000 watt hours saved over all of its users? We're talking $50 in electricity at $0.10/kWH...).
I am interested in the Standard Deviation, SEM, and a one- or two-tailed T-test. As a molecular immunologist, that's about all I need for 99% of my data analysis.
[...] But if you can suggest a good data analysis application that runs on Linux, I will listen, and will surely try it.
I'm usually the first to encourage people to move beyond spreadsheets and use better tools for statistical analysis. That said, a spreadsheet is a really quick and easy way of doing simple data analysis, and it's perfectly fine to use it at such.
The problem comes in when people start trying to use spreadsheet applications for more complicated analysis or want to do more complicated graphics than a spreadsheet easily allows. If and when that time comes, it becomes really worthwhile to have at least one other tool in which to work. As the other reply suggested, R is a free (and excellent) implementation of the SPLUS language. The package is explicitly designed with statistical analysis and graphics in mind. In fact, a nice introduction to the language is Data Analysis and Graphics Using R - An Example-Based Approach by John Maindonald and John Braun . You might be able to find the book at a university library before deciding whether to plunk down the money to buy it.
MATLAB is more of a general purpose language, which can be very useful for some fields and not as useful in others. It's definitely overkill to buy MATLAB to do basic statistical analysis, and it's probably not the best tool for the job unless you already know the language well. Most other commercial statistics packages (SAS, SPSS, Stata) have Linux versions, as this community has tended to be more server/unix-oriented historically.
To bring this back on topic, it's nice that OpenOffice.org is expanding its feature set in the statistical/graphing arena - I've personally found it quite lacking compared to Excel. That said, it's also important to know when you've moved beyond what a spreadsheet is relatively good at and find a package which can do the more complicated analysis. Spreadsheets and stats programs are both complements and substitutes in various ways.
Ok, that's fine and well that if the journal allows it, Harvard makes a copy of the article freely available. What about those journals (Nature and Science, maybe?) that do not allow this. Does this mean that Harvard faculty will not publish in Nature and Science? Somehow I doubt that. Does this mean that Harvard will break copyright agreements? Maybe? The article doesn't quite say.
My understanding of this system is that it's opt-out rather than opt-in. Faculty members retain the copyright to their papers if they're included in the archive, and they have to right to remove them from the archive (opting out). Publishing to many (most?) journals entails signing over the copyright of the final form of the paper to the journal.
It seems entirely conceivable that some journals will require Harvard profs to remove the article from the archive as a condition for publication. On the other hand, in some fields it's common for "working paper" versions of a paper to circulate widely before they are officially published. Official publication does not usually entail the removal of these working paper versions. I suspect that this is part logistical (it's hard to revoke something that's been made available free on the web), part non-competing (the final version of the paper tends to be more polished and you'll almost certainly prefer citing it over the working paper version), and part publicity (it's easy to find working papers, and if you really like it you'll seek out the published version, serving as advertising).
So basically, this archive can serve as a working paper repository for Harvard profs. They don't need to put it up on their own web page or have a website in their field dedicated to it, so hopefully this will make it even more convenient to have research available freely on the web.
The Game Boy was pretty tough, I admit. The other items in that "article", though, are significantly less so. A three-year-old phone? A 6-year-old Mac? C'mon, that stuff is still way too young to be deemed tough. These things need to stand the test of time.
Besides, there's all sorts of old tech which stood the test of time. Rather than rummage through your pile of unsold stuff for eBay, why not pick a few items which are infamous for their longevity?
The IBM Model M Keyboard is the first thing that came to my mind when I thought of durable technology. Bonus points in that it's still useful, and even preferred by some to this day (though I'm not among them - the feel is nice, the noise is not).
A variety of old 80s computers: the Mac Classic, IBM XT, NeXTcube, etc. These things tend to sit around for years and still work just fine.
The point is, why make such a list with tech that's newer than 2000?
The Wii deserves some consideration to be a top game console. [...] I pretty much stopped playing consoles for about 8 years until the Wii came out. I know this is true for a lot of people. During the past winter holidays, my parents, grandparents, and other "older" relatives all loved the Wii too. My 80 year old grandmother won the family bowling tournament, and talked smack about it! My mom got addicted to WarioWare. There's no way those events could happen on any other console.
I think it's still a little too early to judge the Wii as "best console". The Wii has sold about 20 million unit worldwide, which is quite a sum. However, the NES sold 60 million and the Playstation over 100 million (if you believe Wikipedia and the companies own stats). The best console is very subjective, of course, but trying to come up with an objective measure one might include aspects like:
1.) Immensely popular in its own time, with a large selection of good games. The NES and Playstation had a number of duds, but the selection of truly great games for each was remarkable. The Wii is no where near this point, not yet.
2.) Had a huge impact on the video game market at the time; in the NES's case, it brought back the video game console from a rough patch where the future of console gaming was uncertain. The Playstation really helped move 3D gaming from the realm of the PC to popular console games.
3.) Had a huge impact on video game consoles which came after. This could take the form of the style of game, the technologies used, etc. The NES brought out games like RPGs and games with a bit of a storyline. The Playstation sealed the game cartridge's fate by proving how superior CD storage could benefit games through pre-rendered video and good sound.
The Wii might have a dramatic effect on the way we control games. I own a Wii, and I think the controller is pretty great for a lot of things. Still, the Wii gets played largely with groups, and the selection of games is still somewhat limited. I only own Nintendo-branded games at this point, since it seems to be taking some time for the other game companies to devote time and money into making games well-built for the Wii.
Wii Sports and a few other games (though I find the new Mario Party game a bit grating) make excellent party games for people of hugely varied age. That's cool for getting a new audience to play video games, but it's not clear that it will be a huge market.
We'll know whether the Wii has had a big impact on gaming in large part by whether the next generation of Playstation and xbox feature a similarly-new controller.
That certainly would be an attractive option, if it weren't illegal in the United States. Write your Senators and complain.
It wouldn't be illegal if Apple got the right licenses and agreements. They're clearly a big enough player to be able to negotiate rentals through iTunes. They might have to do something annoying to get the studios to agree - like have ripped versions expire after X days. The most obvious thing that studios will worry about besides sharing is mass renting through Netflix/library and ripping them all to disk for a large permanent library.
Still, it would allow you to load up your laptop with movies for the road for a trip and not have to worry about the disks, and in the case of this new Macbook Air, having to bring an optical drive.
Slashdot Mirror
Ah, it stores everything separately, and doesn't seem to have anything but a "pen" mode. Since my handwriting is somewhat poor (and my tablet-writing is even worse), the ability to add typed notes would be nice (via a little on-screen keyboard, perhaps? I'm not asking for OCR to read my scribbles). The biggest thing for me is underlining/highlighting - this can be done neatly and efficiently in any PDF which isn't simple scanned. Okular for KDE4 seems to do a decent job at it (the annotations are also stored separately), but it's still a bit in the early stages of functionality.
The big screen is nice - if/when I ever get an ebook I'd be tempted to want one that's about a 8.5x11" screen to view pages 1:1. Still, the eink and battery life would have to be awfully nice to choose it over a low-end tablet (e.g. HPs start at $900).
This is a neat hack; I'm mildly surprised that you can fit a decent version of Wikipedia in under 4 GB. The text, sure (especially bzip2 compressed), but a decent set of images? Anyone have a breakdown of exactly which version of Wikipedia this is?
The static Wikipedia pages appear to have not been updated since April 2007 (the February 2008 ones stop just before "en"). That version comes in larger than 4GB, but static HTML pages are less efficient, I would think, than what this guy did parsing the XML data.
These days, though, WiFi is available in so many places that even if I owned one of these devices I probably wouldn't use up the flash space with an offline version of Wikipedia.
Side note about the iRex. The ebook version of the reader (which, notably, lacks WiFi compared to the more expensive version) appears to be $599 MSRP. I personally thought the Kindle was expensive at $400, wireless service included. The WiFi iRex is $700, which is getting into the territory of a few low-end (or used, I'm sure) tablet notebooks. I understand that the battery life and screen readability of these things is supposed to be pretty good, though.
Anybody know if the iRex or any other ebook reader has the capability to annotate PDF files? I do a quite a bit of reading of PDF documents, and I find myself printing them all too often so that they're easier to read and I can make notes. These ebook screens are supposed to be easier on the eyes than a standard laptop screen, so all that's left is the ability to make annotations.
On the Media had an interview (transcript and mp3 download available) last week with Internet Archive co-founder Brewster Kahle. about his personal experience with the national security letter. Interesting stuff, but perhaps not much new if you've been keeping up with this.
This is temporary, and is a common complaint about KDE 4.0. The idea with KDE 4.0 was to ship what they had to encourage further application development. There are lots of changes to KDE, including using a new version of QT (the underlying toolkit).
The basics are there, but customizeability, as you noted, is lacking. From what I understand, that flexibility (especially in terms of the main panel) will return with KDE 4.1, to be released this July.
KDE 4.0 isn't for everybody. After reading about some of these limitations, I decided to wait until KDE 4.1 before upgrading my Kubuntu laptop's KDE version. As I understand it, KDE 4.1 will bring applications like the PIM framework up to speed, and I should be able to make my desktop look and work like I'm used to with KDE 3.5 (a substantial alteration from the default).
KDE hasn't abandoned the philosophy of a very flexible user interface, it's just taking time to re-implement the features in the serious overhaul that is KDE 4. I can wait.
0.15 cents or 0.15 dollars?
It's been my experience that the more repositories you add, the more you run into issues of conflicting package offerings between them. I try to avoid it if at all possible and simply use the "official" community-supported stuff.
I know that used to be true, but when I just tried it yum gave no indication that it was going out and retrieving any file lists (but perhaps it still was in the background). For me, that seems like a dumb default setting. The "apt-get update" system seems to work well: update your repository info when you want to, and work with it from then on. That way if you do a few searches you aren't looking for updated repository info for each and every search. Even if there is a way around this with a command-line switch, it still seems like a dumb default.
I completely agree. Since my distros of choice over the last 5 years have been Fedora and Debian/Ubuntu, I've had a fair bit of experience with both yum and apt-get. Yum, at least as of the Fedora 8 install on my desktop, is simply not as good (IMO) as apt-get in Debian or Ubuntu for two reasons:
1.) yum is slow, horribly horribly slow. I think it may have gotten a little better in Fedora 8, and I've heard that they're putting serious work into it. Hopefully Fedora 9 will be better, but it never ceases to amaze me how long it takes to do a "yum search" to look for a package compared to "apt-cache search".
2.) The package repositories for Ubuntu (which is derived from the huge repository from Debian) are larger and more complete, at least for the random software I tend to look for. Again, Fedora is gaining in this regard, the community-supported package setup is starting to rival Ubuntu's universe, making this a huge step up over the old RedHat 7/8/9 days compared to Debian at that time. When it comes to software outside of either repository, RPMs tend to be more common than debs, which is an advantage for Fedora.
So yum (and the standard underlying repositories) are behind in those respects compared to apt-get, but the difference is shrinking. In yum's defense, I think they implemented package signing as a default requirement before Debian did, but I could be wrong on that.
I've run Fedora on my desktop for a while, but Kubuntu on my laptop. I honestly don't know what I'll install on my desktop next. I usually skip every other release, and since I'm on FC 8, that means waiting until FC10. This might be good anyway; I'm a KDE user, and KDE 4.0 just doesn't look feature complete. Best to wait until KDE 4.1 polishes everything a bit more, perhaps. I'm debating whether to try out the latest Kubuntu on my laptop when it's released this month to try out KDE 4.0.
Got a source for that? Some quick Googling shows that there is serious disagreement about the number, but I've seen estimates from about 3% to 20%. This post illustrates the wildly different numbers.
It's amazing to me that we haven't pinned down this number better, but some people like to include things like the amount of extra air conditioning required to compensate for the heat generated in the summer and subtract the amount of heat that the lights generate that something else didn't need to in the winter. That seems like it's complicating things unnecessarily, but amongst the sources I've seen, I haven't found anything below 3%.
Also, even if heavy industry is such a large user of power, that doesn't mean that home energy conservation won't have a significant effect. The reason is straightforward: electricity is very costly to transport long distances. Heavy industry can locate near large dams and other sources of inexpensive power, reducing both production costs and transportation costs. Most people don't live that close to cheap power, and thus the cost savings that they realize by energy conservation provides a decent measure of the burden they're eliminating from the overall system; that's not even getting into the externalities of power production like pollution and greenhouse gases, which only serve to boost the returns from home energy conservation.
That might be true for you, in the winter. Coal is the predominate source of electricity for most people. Transporting natural gas is costly, but it's not burned until it gets to your home. On the other hand, a non-trivial amount of electricity is simply lost due to resistance in transmission - that's coal burned.
Actually, there are quite a few PhDs out there blogging. They are hardly "useless in their fields", at least the ones I read; they tend to be some of the more high profile people (and the blogs simply give them an even higher profile). Two cases in point for economics: Greg Mankiw's blog and Marginal Revolution, a blog by two George Mason profs with occasional guest bloggers.
Blogging is actually fairly amenable to the goal of many academics: to share information and debate about it. The biggest downside that I see is that blogging is fairly time consuming. Mankiw turned off comments to his blog because he didn't have time to moderate them, so his blog became more of a one-way street.
Of course, econ is just one field; I honestly don't know how prevalent blogging is in other fields.
This is an interesting difference between academia and the business (i.e. "real") world. I suspect that it comes from a combination of considerations and cultural aspects:
1.) Professors do research, and submit that research to journals for publication. Those journals often require the professor to sign over the copyright of the paper before publication. It's easier for professors to do that if they have the copyright in the first place.
2.) What about lectures? Some (a few) professors make a ton of money (or little money and a lot of recognition, in some cases) by selling their books. These books start out as lecture notes, typically, especially at the graduate level. Professor's salaries don't vary that much, so this is one way in which the better/harder working/better known professors can earn relatively more pay. That keeps them at a university that can't afford to pay them what they'd get if they quit and just published their textbooks, which is good for the university.
3.) A big consideration is probably the culture that a professor's work is not so much for the university itself, in the sense that professors move between universities all the time and take their research/lab/lecture notes with them. Would you honestly expect professors to have to somehow re-write their lecture notes upon moving to a different university? It just doesn't happen.
4.) Tenured faculty have a fair bit of power over university policies, if they collectively put their minds to something. While works created by staff and non-faculty might be the automatic property of the university, faculty (in the U.S. at least) typically get the copyright for much of what they create.
It's tempting to draw parallels between programming and research/lecture notes. The cultures, though, are quite different. In general, academics share their resources pretty openly, at least up until the point it becomes a textbook. To the extent that academics write code, it often isn't under an explicit license at all (which can be inconvenient if you want to properly include it in something for redistribution).
Patents are another issue altogether, and one where the university stands to make a great deal of money. I'm not at all familiar with the general breakdown of rights about those, but it seems that both the inventors and the university get a cut in many cases.
So what's up with this professor? It sounds like somebody is peeved that his students aren't attending class and would rather pay somebody to come in and take notes for them. Many good professors do the exact opposite and post class notes online, though they may not include quite everything that's worth getting from a lecture.
If notes were a substitute for a good lecture, most of us would learn by buying the best notes from the best professor in the world on a subject (which is only sometimes available as a textbook). On the other hand, a bad lecture is worse than a decent set of written notes. The solution is not to sue them, but to improve your lectures!
That's the whole point: the license is completely at odds with both de-facto policy by Adobe and common practice by users.
If the point was to keep people from distributing copies of Reader, why not just simply add a clause that you can only re-distribute it for your own, personal use or for the internal use of your organization? Perhaps the overkill in the license is the result of the way contract law seems to work (IANAL): if one part is found unenforceable, other parts can be salvaged. This gives lawyers the incentive to write more restrictive licenses and then selectively enforce the terms. Users end up just clicking through the pages and pages of legalese and ignore the terms, both reasonable and not. It's hardly an appealing outcome.
Actually, I'm not even sure you can do this. I'd be curious what the law would say (if anything) about whether downloading another copy constitutes another new license, and thus another "Permitted" computer.
It's great to see some outrage about stupid license agreements. There is widespread apathy about these agreements and their restrictions, leaving companies with the option to selectively enforce crazy terms if and when it is in interest.
On a semi-related note, the choice for many people is not to buy some expensive software or use the open source equivalent, but to illegally copy the expensive software or use its open source equivalent. Open source software does fairly well competing on features and stability alone, but the total lack of restrictions on use is something that is often overlooked, at least by home users. There may still be some misunderstanding about open source regarding use, though I hope that's fading. The availability of source is great, but frankly, that's not going to directly affect what home users can do with their software.
Rather than ignore licenses and laws we don't like, it's far more productive if people make an issue about them to get them changed. The outrage over AT&T's policies is one example of this (though that seems to have stalled, aside from an ongoing lawsuit), and this is another. Ignoring these terms and hoping they aren't enforceable is just a recipe for trouble. As a side benefit, if people become disgusted by closed source license agreements, open source software will look more appealing.
Adobe's not exactly known for their reasonable EULAs. Just take a look at the EULA for Adobe Reader. This is software that Adobe is trying to get on all the computers it can. The license, however, permits only the installation on one primary computer and one mobile computer (note that "Permitted Number" is 1). I've gone so far as to contact Adobe customer service and ask them what's going on - this goes completely against their marketing policy. Amusingly, they send all their customer service responses via PDF over email. Their official response?
It's clear that Adobe has no intention to actually try to enforce this restriction, but it suggests that organizations with computer labs and such are supposed to negotiate a volume license with Adobe. I think the Reader license is simply boilerplate recycled from other Adobe software, but it's clear that whoever is responsible for Adobe's licenses isn't in touch with what Adobe actually wants to have regarding its licensing (at least from a marketing perspective).
The "real CPI" you claim is a crude measure which deserved to be fixed. A fixed basket doesn't represent what people actually do, and the current methods of collecting price data get at that better.
Even if you disagree with the methods (and your link is not exactly an objective source, show me a paper published somewhere respectable or at least written by a respected academic), you won't arrive at 10-12% inflation that you claimed. That's quite simply a blatant exaggeration.
1.) Gas does not make up a terribly large portion of people's expenditures 2.) By picking the last decade, you conveniently start with the lowest recent price of gas and end with the highest (by recent I mean since the 1980s, but we really should be talking about a measure of the real price of gas). In the 8 or so years before that, the price of gas fell about 30 cents in nominal terms
Like I said in the grandparent post, the full CPI index, with its current ~3-4% rate of inflation, includes housing, energy, transportation, etc. No amount of fiddling with quality or substitution or anything like that is going to change that into the 10-12% you claimed.
Much of your link is simply FUD. Seasonal adjustment biasing the CPI downward? No, it simply makes the CPI less volatile - parts of the year where gas is relatively more expensive are smoothed out with the parts where gas is relatively cheaper. The stuff about the "core" rate? Most news articles I read make a pretty clear distinction if the inflation measure they're using includes energy and food. To the extent that the CPI tries to fix for substitution bias, reputable work using a large dataset shows that the CPI is still likely overstating inflation significantly; their result is that inflation is an average of 0.8% lower than reported, on data covering 40% of consumer expenditures from 1994, 1999-2001, and 2002-2003.
Again, there's simply no evidence for your claim of 10-12% inflation.
The CPI is released in several forms. It's usually reported in the news as either the overall CPI index (which includes food and energy), or the CPI less food and energy (sometimes referred to as the "cold and hungry" CPI). Neither is anywhere close to 10-12%. See for yourself. Overall inflation, at an annual rate, based on the last 3 months is 3.4%. Based on the last 12 months, it's 4.4%. Without food and energy, these numbers are 2.4% and 2.3%. Inflation is up from its relatively low values in the last couple of decades, but still far away from the early 80s. Also, many economists believe that the CPI in fact overstates inflation. Why? People will substitute from goods which became relatively more expensive to those which haven't. To the extent that the basket of goods that the Bureau of Labor Statistics uses to calculate CPI doesn't take this into account, it will make inflation seem larger than the average person will really feel.
The CPI is supposed to measure what typical households buy, but if you can only pick one rate of "inflation", it's usually the most reasonable. Even if you were to argue that NASA spends a great deal of its budget on fuel (which I highly, highly doubt), that fuel is not directly petroleum-based. The solid rockets are based on ammonium percholorate (according to this). The shuttle itself has engines based off of liquid hydrogen and oxygen.
I agree. But even if you take their claims of energy saving as face value, the magnitude is pitiful. Another poster in this story calculated that a CFL saves a large fraction of a megawatt hour in its lifetime. Replacing a single bulb with a CFL will save more electricity in its lifetime than all the users of Blackle have for its entire existence so far (over a year), going by Blackle's own numbers.
It's not a matter of two wrongs, it's a matter of tradeoffs (as with most things). I'm a big fan of non-coal power (including nuclear), but the existing coal plants aren't going to go away any time soon, and we seem to keep building more around the world.
For most people, artificial light is a necessity. These days, they have a choice between incandescents, CFLs/fluorescents, halogen, and maybe a few LED options. CFLs are much more efficient than incandescents or halogens. LED lights are still somewhat expensive. If you use a CFL, don't recycle it, and its total mercury emissions are less than the emissions from the power plants used to produce the extra electricity required to power an incandescent, it's clearly better to go with the CFL.
Of course, the best thing to do from an environmental perspective is to simply recycle your bulbs. I've mentioned this before on
Being a rational environmentally-conscious person means that you should take actions which require the least expense (in terms of both time and money) which cause energy/pollution reductions in the greatest quantity. That's why Blackle is a total waste of time (500,000 watt hours saved over all of its users? We're talking $50 in electricity at $0.10/kWH...).
I'm usually the first to encourage people to move beyond spreadsheets and use better tools for statistical analysis. That said, a spreadsheet is a really quick and easy way of doing simple data analysis, and it's perfectly fine to use it at such.
The problem comes in when people start trying to use spreadsheet applications for more complicated analysis or want to do more complicated graphics than a spreadsheet easily allows. If and when that time comes, it becomes really worthwhile to have at least one other tool in which to work. As the other reply suggested, R is a free (and excellent) implementation of the SPLUS language. The package is explicitly designed with statistical analysis and graphics in mind. In fact, a nice introduction to the language is Data Analysis and Graphics Using R - An Example-Based Approach by
John Maindonald and John Braun . You might be able to find the book at a university library before deciding whether to plunk down the money to buy it.
MATLAB is more of a general purpose language, which can be very useful for some fields and not as useful in others. It's definitely overkill to buy MATLAB to do basic statistical analysis, and it's probably not the best tool for the job unless you already know the language well. Most other commercial statistics packages (SAS, SPSS, Stata) have Linux versions, as this community has tended to be more server/unix-oriented historically.
To bring this back on topic, it's nice that OpenOffice.org is expanding its feature set in the statistical/graphing arena - I've personally found it quite lacking compared to Excel. That said, it's also important to know when you've moved beyond what a spreadsheet is relatively good at and find a package which can do the more complicated analysis. Spreadsheets and stats programs are both complements and substitutes in various ways.
My understanding of this system is that it's opt-out rather than opt-in. Faculty members retain the copyright to their papers if they're included in the archive, and they have to right to remove them from the archive (opting out). Publishing to many (most?) journals entails signing over the copyright of the final form of the paper to the journal.
It seems entirely conceivable that some journals will require Harvard profs to remove the article from the archive as a condition for publication. On the other hand, in some fields it's common for "working paper" versions of a paper to circulate widely before they are officially published. Official publication does not usually entail the removal of these working paper versions. I suspect that this is part logistical (it's hard to revoke something that's been made available free on the web), part non-competing (the final version of the paper tends to be more polished and you'll almost certainly prefer citing it over the working paper version), and part publicity (it's easy to find working papers, and if you really like it you'll seek out the published version, serving as advertising).
So basically, this archive can serve as a working paper repository for Harvard profs. They don't need to put it up on their own web page or have a website in their field dedicated to it, so hopefully this will make it even more convenient to have research available freely on the web.
The Game Boy was pretty tough, I admit. The other items in that "article", though, are significantly less so. A three-year-old phone? A 6-year-old Mac? C'mon, that stuff is still way too young to be deemed tough. These things need to stand the test of time.
Besides, there's all sorts of old tech which stood the test of time. Rather than rummage through your pile of unsold stuff for eBay, why not pick a few items which are infamous for their longevity?
The IBM Model M Keyboard is the first thing that came to my mind when I thought of durable technology. Bonus points in that it's still useful, and even preferred by some to this day (though I'm not among them - the feel is nice, the noise is not).
A variety of old 80s computers: the Mac Classic, IBM XT, NeXTcube, etc. These things tend to sit around for years and still work just fine.
The point is, why make such a list with tech that's newer than 2000?
I think it's still a little too early to judge the Wii as "best console". The Wii has sold about 20 million unit worldwide, which is quite a sum. However, the NES sold 60 million and the Playstation over 100 million (if you believe Wikipedia and the companies own stats). The best console is very subjective, of course, but trying to come up with an objective measure one might include aspects like:
1.) Immensely popular in its own time, with a large selection of good games. The NES and Playstation had a number of duds, but the selection of truly great games for each was remarkable. The Wii is no where near this point, not yet.
2.) Had a huge impact on the video game market at the time; in the NES's case, it brought back the video game console from a rough patch where the future of console gaming was uncertain. The Playstation really helped move 3D gaming from the realm of the PC to popular console games.
3.) Had a huge impact on video game consoles which came after. This could take the form of the style of game, the technologies used, etc. The NES brought out games like RPGs and games with a bit of a storyline. The Playstation sealed the game cartridge's fate by proving how superior CD storage could benefit games through pre-rendered video and good sound.
The Wii might have a dramatic effect on the way we control games. I own a Wii, and I think the controller is pretty great for a lot of things. Still, the Wii gets played largely with groups, and the selection of games is still somewhat limited. I only own Nintendo-branded games at this point, since it seems to be taking some time for the other game companies to devote time and money into making games well-built for the Wii.
Wii Sports and a few other games (though I find the new Mario Party game a bit grating) make excellent party games for people of hugely varied age. That's cool for getting a new audience to play video games, but it's not clear that it will be a huge market.
We'll know whether the Wii has had a big impact on gaming in large part by whether the next generation of Playstation and xbox feature a similarly-new controller.
It wouldn't be illegal if Apple got the right licenses and agreements. They're clearly a big enough player to be able to negotiate rentals through iTunes. They might have to do something annoying to get the studios to agree - like have ripped versions expire after X days. The most obvious thing that studios will worry about besides sharing is mass renting through Netflix/library and ripping them all to disk for a large permanent library.
Still, it would allow you to load up your laptop with movies for the road for a trip and not have to worry about the disks, and in the case of this new Macbook Air, having to bring an optical drive.