I have a mouse like this, from Kensington. They don't make the exact same model I have any more, but the Studio Mouse appears to be similar. It's a three-button corded optical mouse, but the middle button is small, and above it is a little trackpad. The trackpad only scrolls vertically, though.
Someone would need to write an nsswitch "plugin" for bonjour. So far as I know, this has not been done.
It has been done - nss-mdns. It's been packaged for most Linux distributions. Getting it working on my Debian box was a simple matter of apt-get install libnss-mdns and then edit/etc/nsswitch.conf to use the new nss plugin.
Actually, "Nausicaa" comes from classical Greek. I have the first volume of the manga, and it has a foreword that explains that Miyazaki was partially inspired by a Greek myth featuring a princess of that name. Here's the first page Google found for me about the original Nausicaa: http://www.loggia.com/myth/nausicaa.html.
I'd suggest that you, and anyone else interested in the question of why Mars is a better target for a base and eventual colonization, should read Robert Zubrin's book The Case for Mars. He advances in it a number of arguments as to why Mars is better, but the one I found the most interesting was the suggestion that Mars is the best place in the solar system for farming, aside from the Earth (see pages 185-199).
We know that Mars has lots of water relatively accessible in its polar ice caps, as well as likely underground deposits all over the planet. By contrast, the only known water on the Moon is the recently discovered ice lurking in shadowy craters at the south pole. And based on the results of Viking and analysis of Martian meteorites, we know that its soil has all of the necessary nutrients for growing plants in.
The other key to agriculture is a source of energy for growing crops. To use natural sunlight for growing crops on the Moon or a space habitat would require huge greenhouses. To shield the crops from radiation, the glass would have to be 10cm thick! Mars has enough of an atmosphere to filter out most of the radiation, so farms on Mars would simply require a pressurized tent. Artificial lighting is also a possibility, but the energy costs become prohibitive, particularly when free energy from the Sun is available.
Most of the above factors apply, less generally, to human habitation. The Martian atmosphere will be easier for humans to get around, and it'll mean that settlements require less shielding against radiation. The Martian day is close in length to that of Earth. The surface gravity of Mars is higher than that of the Moon, so adapting to it will be easier. And so on.
Mars is the best place for human habitation outside of the Earth. That makes it the most desirable colonization target, but even for temporary outposts, the same factors will make those bases cheaper and easier to sustain.
Probably my favorite thing about Andale Mono is that the zero has a dot in the center, making it trivial to distinguish from the letter O, which does not have the dot. Few other monospaced fonts today have that feature.
To programmers, that's a big win. In fact, making C-syntax characters look different ("1" v. "l", "{}" v. "()", "O" v. "0", "." v. ",", ":" v. ";", "'" v. "`") should be a priority for anybody working on an Andale Mono replacement. (Andale Mono could be improved on a few of these).
You'd probably be interested in ProFont - a font designed for programmers, which has existed for years, but few outside of the Mac programming community know about it. It was specifically designed to be readable at 9 point, with similar characters distinctly different, as this page demonstrates. The full distribution includes TrueType, Type 1, and bitmap versions of the font for Mac and Windows. You can also download a look-alike bitmap version for Windows here.
I've been using ProFont for years as the font in my editor when coding, and found it very helpful.
I agree that the books on the various Java APIs quickly become outdated and that it's easy to pick up the core language from the online references. However, Effective Java by Joshua Bloch doesn't fall into either of those categories and is very useful. It's the Java equivalent of Effective C++ by Meyer - roughly 50 concrete dos and don'ts to improve your code, no matter what version of Java you're using.
I'll also second the nod for Python Essential Reference, which is the only Python book I've found useful so far, even if only because it's faster to look things up in than in the excellent included language documentation.
It is completely wrong that the administration of such an important trans-national medium as the Internet is in effect in the uncontrolled hands of just so few people.
I'm not usually given to conspiracy theories, but I think that this is exactly the point of the whole ICANN debacle. The DNS is one of, if not the only, chokepoint in the Internet, the rest of which is largely run in a decentralized, rational, and transparent manner. An open, uncontrolled and uncontrollable Internet is a threat to many established power groups.
The only theory I can come up with that fits all of the facts surrounding ICANN's behaviour is that some powerful group is behind ICANN and using the DNS to gain control over the Internet, for power and/or money. The way the management has given Auerbach the runaround makes it clear that the elected board members were intended to be powerless rubber stampers, which is why they're now dropping all pretenses of elections, since one of those elected directors is actually trying to do his job.
Kudos to Auerbach and Gilmore for standing up to these bullies. Meanwhile, I think that the technical Internet community should be building alternatives. Going up against the established powers on their own turf of the courts and political system is foolhardy. Even if we win this particular battle, the power nexus that the DNS roots imply will continue to attract the corrupt or corruptable. A technical solution that completely decentralizes the problem that DNS tries to solve is the best long term strategy, eliminating this particular threat forever.
This idea is completely in accord with one of Gilmore's maxims: "The net treats censorship as damage, and routes around it." The censorship and damage due to ICANN is clear, and we need to find a way around it.
You didn't need to buy XTools for NEdit. It also works pretty well with XFree86, although I've noticed an issue where NEdit doesn't set the mouse pointer correctly with it. And with Fink, installing NEdit and XFree86 is a snap.
Right of first sales gives you the right to re-sell the original, not make a photocopy for yourself and sell the original.
Yes, but the prohibited act here is making the photocopy, not selling the original.
No, the prohibited act is keeping the copy after selling the original. Making backups is permissible under any reasonable interpretation of fair use, but if you do create additional copies, you are supposed to destroy the copies when you sell the original.
Hmm... Is there OS X Photoshop available? No. Is there OS X Office
available? No. OS X Quicken? No.
Quicken 2002 Deluxe is a Carbon program that works great under OS X. I know, because I bought just it last week.
The next release of Office, due this fall, will also run on OS X. The major Mac news sites have all been shown demos of it. Adobe's silence on an OS X version of Photoshop is indeed troubling.
I dispute the contention that choosing Be over NeXT as the basis of OS X would have made a difference in software availability. Be's APIs are just as foreign from the Classic MacOS ones as NeXT's are. No major software house with a large body of working code was going to rewrite it for either Be or NeXT-derived APIs. If Apple had have gone with Be, it would still had to have had something like Carbon for established MacOS software.
There might have been a difference in what native software was available. I haven't looked at BeOS in a while, so I don't know if it has more or fewer apps to bring over than the NeXT world did. But I do know that the BeOS certainly had nothing like OmniWeb, which is very impressive.
If you read the DMCA, what you emphasized is strictly prohibited, yes.
I've glanced at the DMCA several times, and it's not clear to me that it criminalizes these acts. Rather, it criminalizes the means of doing so. There appears to be nothing in it that says you can't print an ebook for example, but if the publisher uses an access control measure to prevent that, it's illegal to circumvent that access control. It's also illegal to give the means to do so to other people (the trafficing provisions that Sklyarov is being charged with).
This is a clever sidestep that makes the DMCA such a strange and dangerous law. The U.S. Congress has effectively delegated its law-making authority in this area to corporations. If Adobe or some other corporation had wanted to make it illegal for someone to print the ebook they just bought, it would have formerly been necessary to lobby and get a law passed. Now, they just add some options to their access control and it's effectively illegal. Sure, no law has been passed removing one's right to print ebooks, but if there's no legal way to exercise that right, does it matter?
This also shunts decision making away from the public view (although, there doesn't seem to have much in the way of public debate when the DMCA was passed). Formerly, any attempt to change the rights and freedoms associated with the use of intellectual property would have had to been done through a vigorous public debate involving all affected parties. Now, new restrictions can be added by unaccountable individuals without any public debate. Any attempt by third parties to undo those restrictions and restore the former status quo is conveniently made illegal.
The distinction between copying and access that you noted is again another clever tactic on the part of the DMCA's framers. U.S. caselaw already has numerous precedents defending the right to make personal copies (the Audio Home Recording Act for example, or the Betamax case), so a law totally denying the right to make copies would be hard to defend. But under the Catch-22 of the DMCA, it doesn't matter, because to make copies you either need to get access to the work (which is illegal) to copy it, or make a copy somehow else, in which case it's still illegal to access that copy.
Defenders of the DMCA are quick to allege that the only possible use for circumvention devices is widespread unauthorized distribution of copyrighted materials (ie, piracy), and give the prevention of things like Napster as a reason for enacting it. It's interesting to note that in the DMCA cases brought forward so far, this is clearly not the primary use of the devices in question. DeCSS was written to allow the development of a Linux DVD player. Felten and company's research was ordinary academic study in cryptography. AEBPR allows the purchasers of ebooks the rights they would have enjoyed by buying books in physical form. Even if one argues that it is not necessarily the case that making copies was not the primary purpose of any of these devices, it could certainly be plausibly argued in a court of law. While all three devices could theoretically be used for unauthorized distribution, so too could the normal playing devices being sold. It certainly does not appear to be the case that the DMCA is effective in preventing piracy in this way.
So yes, I do feel that the items I emphasized should not be illegal. They do no harm to others. Copyright holders are not emperiled because someone prints out an ebook or copies it to another device for offline reading. Likewise, a computer audibly reading text is no danger to anyone. The only item on that list which is harmful is the distribution of copies, which is already illegal. I find it hard to support a law that criminalizes harmless actions. The DMCA offers no benefit to the public at large, only to those corporations who hold large numbers of copyrights. Given the choice between padding the bottom line of a large corporation and defending the freedoms associated with reading, I know which I'll choose every time.
If longstanding copyright law no longer applies in today's world, it should not be rewritten from scratch by fiat, but rather hammered out in a compromise protecting the interests of all affected parties. The DMCA is far too one-sided to be described as such.
Well, it seems to be official. Fair use in the U.S. is dead. Look at what the indictment boils down to:
When an ebook purchased for viewing in the Adobe eBook Reader format was sold by the publisher or distributor, the publisher or the distributor of the ebook could authorize or
limit the purchaser's ability to copy, distribute, print, or have the text read audibly by the computer [emphasis mine]. Adobe designed the Ebook Reader to permit the management of such digital rights so that in the ordinary course of its operation, the eBook Reader effectively permitted the publisher or distributor of the ebook to restrict or limit the exercise of certain copyright rights of an owner of the copyright for an ebook distributed in the eBook Reader format.
On a date prior to June 20, 2001, defendant Dmitry Sklyarov and others wrote a program called the Advanced eBook Processor ("AEBPR") the primary purpose of which was to remove any and all limitations on an ebook purchaser's ability to copy, distribute, print, have the text read audibly by the computer, or any other limitations imposed by the publisher or distributor of an ebook in the eBook Reader format, as well as certain other ebook formats.
Note that the indictment clearly indicates that AEBPR is only useful to purchasers of ebooks in Adobe's format, so there can be no allegation of it being used for widespread piracy. Instead, Sklyarov's apparent crime is to allow people to actually use the ebooks they've bought and paid for. Of the items enumerated as being restrictable by the publisher or distributor, only distribution is forbidden by copyright law prior to the DMCA, and then only when the fair use exemptions don't apply. It seems rather overreaching to me that the DMCA criminalizes being able to do such ordinary actions with an ebook such as having the computer read it aloud or print it, let alone making copies for backup or use on another machine.
Note also that the indictment makes no mention of the AEBPR being used to violate copyright law. No evidence is offered that any of the handful of its purchasers used the program for any illegal purpose. The mere fact that it allows the purchaser full use of a bought ebook and the theoretical possibility of commiting an act (unpermitted distribution) which is already illegal under century old copyright law, is reason enough to send a man to jail for 25 years. Scary.
And publishers wonder in vain why ebooks aren't selling very well? Gee, if you don't let the purchaser do anything with them, making ebooks far more restricted and less useful than print books, and totally upset the balance between public and private interests enshrined in copyright law, you should expect this. Indeed, I'm frightened that ebooks have sold as well as they have. The freedoms and rights associated with reading seem to no longer apply in the digital world if the interests that bought the DMCA have their way.
My favorite Poul Anderson work is the Time Patrol story "The Sorrow of Odin the Goth", which brings a tear to my eye every time I read it. It would have to be closely followed by The Boat of a Million Years, which makes being immortal sound like a burden. These show Anderson at his best, either taking the long view of history or making some ancient piece of the past seem alive and exciting. No other author I've read has had that delicate touch for history, and for that he will be missed.
Little known-fact: I read somewhere that Anderson's novel Three Hearts and Three Swords was one of the strong influences on the creation of Dungeons and Dragons.
The trouble with Python is that it isn't a mainstream language. A CS program should really only teach a few languages, and as such the ones that are taught must be choosen carefully, and be fairly mainstream.
I disagree strongly. A CS program should teach several languages, and the issue of whether those languages are popular or not should be irrelevant; rather, the languages should be chosen to be the ones that best facilitate the teaching of the particular course. A non-mainstream language actually has a small advantage for the first year intro to programming course, since it means that few students coming in will already know it and be bored out of their skulls.
You then also still need to teach them a "heavy" mainstream language, and may not have time (even in 4 years) to teach them another mainstream language.
I agree that one of the languages learned during a four year CS should be strongly mainstream (probably C, the current lingua franca of computer languages), but if a four year CS student can't pick up several languages during that period, something is wrong. In my alma mater's CS department, computer science students learn four new languages during the second year alone (Eiffel, C, Prolog, and MIPS assembly) for the required classes. In third and fourth year, which are all electives, most classes require learning new languages as well. With a strong foundation on a good teaching language in first year, upper year CS students should be able to learn new languages fairly easily.
Python makes an excellent language for teaching beginners. In no way should its lack of popularity be held against it in that role, save possibly the lack of introductory CS textbooks using it.
I've heard about rootless X on OS X, but I have no idea where to find it! Any leads on where I can find it?
Try the XFree86 on Darwin and Mac OS X page.
Most of the work is done through the XonX project at Sourceforge.
I believe that in order to get the rootless X in Aqua working, you need to build the latest XFree86 and XonX from CVS.
Check the archives of the Darwin Development mailing list at darwinfo.org for postings by Torrey Lyons on how to compile it.
My question is, is there any kind of plan to do the same kind of thing
for Objective C and the Cocoa (NextStep) and/or GNUStep APIs? If not,
would it be feasable, or maybe even productive, to do so as a third
party hack?
Check out the Python - Objective-C bridge project at Sourceforge. It appears to be coming along slowly. In the mailing list archives you'll also find discussion of using Jython with Apple's Java bridge as well.
I'm using NFS between my Linux server and my OS X workstation, but it's a bit flaky. OS X can import NFS filesystems if you set some options (look in the forums on macnn.com for details), but I've found that I need to stop and restart the NFS server processes every time the OS X box tries to import a new filesystem. OS X can export via NFS so long as it's from a partition formatted as UFS (the default is HFS+ for Classic compatibility, which does not work with NFS).
There is no fancy GUI to enable NFS, so you have to be familiar with the Unix command line and the NeXT-derived NetInfo system in order to get NFS working.
In the Public Beta, I found some problems where non-Unix applications could not see NFS-imported filesystems, but that appears to have been fixed in the Final Release.
First off, I'm assuming that this is the case where aliens come to make contact with humans on Earth, and not vice versa. The balance of power is very heavily weighted in favor of the party that's going out and making contact on the other party's territory. Our own history of the European settlement of the Americas and Australia shows this quite clearly, and in that case both parties were of the same species. With the even greater potential for misunderstanding between different species, in the long term such contact is going to be even more messy.
Anyhow, the short answer, any human government with good sense will treat alien visitors like visiting royalty, or at a minimum, like foreign diplomats with immunity to local laws. Any being with the technology to cross the interstellar gap is powerful enough to threaten every living being on Earth, so it only makes sense to treat them very carefully, avoiding giving offense or letting them come to harm.
Scientist and science fiction author Charles Pellegrino has pointed out that the energy needed to accelerate a spaceship up to nearly the speed of light is astronomical, and could do unbelievable damage. Life-bearing planets are large and move slowly and predictably, and it's impossible to defend against something coming at you almost before you detect it. If a missile of large mass travelling at 0.99c were to collide with a planet, it would do damage that makes the dinosaur killer look like peanuts. This scenario is graphically portrayed in his novel The Killing Star, which posits that this is the reason why we haven't detected any alien intelligences. It's impossible to defend your homeworld against this sort of attack, so the only rational thing to do is be as invisible as possible, and then go around killing potential threats before they materialize.
Even without planet-killers used by a Berserker race, and hypothesizing some sort of FTL travel with much lower energy, there are lots of ways star-faring aliens could do serious harm to Earth people. The short story "The Flies of Memory" had a particularly amusing example. In it, aliens looking like giant mosquitos came to Earth to tour our famous landmarks. After a human lunatic killed one, it was discovered that the things the aliens view become tied to them in their memory, such that if the alien dies, all of the landmarks it remembers ceases to exist.
And this is just addressing potential physical harm. I imagine that no matter what they do, any visiting aliens would irreparably transform the human psyche. First contact is rather a cultural singularity - after it, everything will be different. Assuming aliens trying to minimize the damage to our worldview, just the demonstration of non-human intelligence and the technology they possess will be a huge hurdle.
So, I guess I'm with the cynics on this one. If aliens come to Earth, the question is not so much what rights they have (in all likelihood, they have enough power to do whatever the hell they please), and the real disturbing issue is how the poor indigenes (ie, us) will fare. Our own history is a pretty disturbing precedent.
First contact and its effects have been discussed and thought about in written science fiction for decades. There are many more stories than the ones I mentioned above considering angles of this issue, so you might want to take a trip to the library.
A lot of people complain about Python's required use of whitespace,
but I've never found it to be a problem. I always use consistent
indentation in my programs no matter what language they're in. So
I find this feature to be a plus, because I never have to run into
the common mistake in C of confusing indentation with for and if
statements, like the following examples:
C
if (a==b)
pr intf("a equals b\n");
else
pr intf("a not equal to b\n");
pr intf("Always runs, regardless of a and b\n");
Python
if a == b:
pr int "a equals b"
else:
pr int "a not equal to b"
pr int "This is part of else clause"
(Blame Slashcode for the extra spaces in the print statements.)
Note that in the Python version, the behaviour is exactly what one
would naively expect from a single glance at the arrangement of the
code. Even non-programmers can read Python code and understand it
readily. I find that Python implementations of algorithms look a
lot like the pseudo-code I had to generate for assignments when I
was in school.
Indeed, I consider Python's use of whitespace to be an advantage
beyond merely preventing this common mistake.
After starting my internship at a big company with a multi-million line
codebase, I'm strongly of the opinion that the source code should
clearly communicate its intent to human readers. Python makes it
difficult not to write clear, readable programs, without relying
on the strong programmer discipline needed to write maintainable
programs in other languages.
Along these lines, the one feature of Python syntax that does get in my way is the colon. I sometimes forget to add colons (like the one on the end of the else statement) precisely because I've gotten used
to the increased indentation being sufficient and necessary to
communicate my intent.
Another annoying thing is that the line ending
characters must be the ones defined on the local platform, so I can't
move a script from my Linux box to my Mac without running it through
a translator. This will become a big issue as the Mac and Unix versions
of Python merge together for Mac OS X.
I haven't looked in any great depth at either Perl or Ruby, but
my initial impression is that their syntax is a little foreign. I was
able to learn Python quickly precisely because the syntax was familiar
and predictable. This is consistent for me - I originally came to
Python from the Dylan world, a language whose claim to fame is being
Lisp-inspired but with an infix syntax to be palatable to C
programmers.
Since taste in syntax is a personal thing, I'm not going to flame
Perl or Ruby users (indeed, both languages are cool and have their
strengths). But I do encourage people not to obsess on the whitespace
issue when considering whether to try Python or not.
Dealers of Lightning: Xerox Parc and the Dawn of the Computer Age by Michael A. Hiltzik
This covers the history of the hugely influential Xerox PARC research center from its founding through the glory days of the 70s when they invented modern desktop computing. It's an important part of computer history that was typically only recorded as "Steve Jobs steals the GUI from a visit to Xerox PARC" until this book was published.
No one has yet pointed out that turn-based games work really well in multiplayer if the players are't all free at the same time. The PBEM (Play By EMail) mode of games like Alpha Centauri lets you carry on a game even when all of the players are on completely different schedules.
I have a good friend who's as into turn-based strategy games as I am. In the past, we've gotten together for a full day playing Master of Orion II, FreeCiv, or Alpha Centauri on a LAN, but that takes a lot of time and requires us to both be free at the same time. He and I now live in different cities in different time zones with different schedules, so it'd be very difficult to set up a simultaneous game over the Internet.
Instead, since August, we've been playing near continuous games of Alpha Centauri via email. When I finish a turn, I send it to him in email and vice versa. This lets the game go whenever we have time in our respective schedules, and gives time to think about strategy and tactics between turns without holding up the other players.
This is easily extensible up to the maximum of seven players allowed in Alpha Centauri. The Apolyton Alpha Centauri Multiplayer forum regularly has games going with players from all over the world, playing their turns whenever is convenient for them in their own timezone.
Alpha Centauri is a long game and takes a lot of time, so you can't complete a full game in a single sitting as with RTS games. Coming from a long line of turn-based games it works very well in that mode, and turn-based gaming is very compatible with PBEM. I'm also aware of other games like VGA Planets which also are turn-based with PBEM modes, so the idea that turn-based games aren't multiplayer is a myth.
Turn-based games aren't as popular as RTS games, but there's a sizable contingent of people who prefer them, and will continue to buy them as long as there are companies that produce good ones. The next year looks very promising with Civilization III and Master of Orion III coming out, and I hope they both have PBEM capability.
The difference (as far as I could tell) is that apt downloads binaries whereas the ports system downloads source and compiles, which is OK when you've got an OK connection, but not when you're on a stinking modem;) Seeing as how source usually is bigger (at least the size of the tarball).
I'm running Debian on my main Linux machine, and have NetBSD installed on another. I like both methods better than the RPM-based systems I've used in the past, with Debian's being my current favorite. I have a couple of points to add to the comparison:
Because the BSD ports defaults to compiling everything from source, it can be very slow on an older machine. My NetBSD box is a Decstation from circa 1992, and compiling a single ported package on it typically takes about half an hour. Before I ever install anything on it, I prefer to look at the NetBSD ftp site to see if anyone has built a binary pmax package first.
Conversely, Debian's apt-get can be instructed to download source packages and build them locally. It's not as elegant and integrated as the BSD make world approach, though.
The NetBSD ports system tries to ftp source packages from the original distribution sites and a couple of mirrors. In the past, I've had problems with it being unable to find the files it was looking for (because the FTP site has moved things around or is down). By contrast, I always can rely on an official Debian package being on the close fast official mirror site I've put in my/etc/apt/sources.list file.
I think they're both good packaging systems, and not different enough to be a reason to switch between Debian and a BSD or vice versa.
What's more interesting, though, is to consider human languages from the same perspective: does the language you speak constrain the concepts you can think about? If you're fluent in several languages, is your mental universe larger? Can you describe a concept that's easy in one language to users of another?
This is a famous question in natural language linguistics, known as the Sapir-Wharf hypothesis. It posits that what humans can think about is limited by the set of terms in their language. So far, all of the experimental evidence disproves it, even though the idea sounds very intuitive. It appears that all human languages meet a certain level of expressivity that allows us to think and talk about anything, no matter the language used.
In computer languages, which are all artificial and of much smaller scope than any natural language, the question is still open. Anecdotally, programmers know that some programming languages make solving certain categories of problems easier than others, sometimes by orders of magnitude. It's been suggested on Slashdot before that a variant of the Sapir-Wharf hypothesis may hold true for computer languages, even if it doesn't for human ones.
Who'd'a thunk Philosophy and Computer Science would have anything in common?
I've always felt that computer science is in many ways a field of applied philosophy. There are many crossovers between the two fields. Consider, for example, that the Boolean logic used as the basis of all digital computing was invented by a theologian as a way of answering questions in metaphysics and theology.
Slashdot Mirror
I have a mouse like this, from Kensington. They don't make the exact same model I have any more, but the Studio Mouse appears to be similar. It's a three-button corded optical mouse, but the middle button is small, and above it is a little trackpad. The trackpad only scrolls vertically, though.
It has been done - nss-mdns. It's been packaged for most Linux distributions. Getting it working on my Debian box was a simple matter of apt-get install libnss-mdns and then edit /etc/nsswitch.conf to use the new nss plugin.
Actually, "Nausicaa" comes from classical Greek. I have the first volume of the manga, and it has a foreword that explains that Miyazaki was partially inspired by a Greek myth featuring a princess of that name. Here's the first page Google found for me about the original Nausicaa: http://www.loggia.com/myth/nausicaa.html.
The different Canadian result is because the author included our $1 (loonie) and $2 (toonie) coins, over the range of change 0 cents to $4.99.
I'd suggest that you, and anyone else interested in the question of why Mars is a better target for a base and eventual colonization, should read Robert Zubrin's book The Case for Mars. He advances in it a number of arguments as to why Mars is better, but the one I found the most interesting was the suggestion that Mars is the best place in the solar system for farming, aside from the Earth (see pages 185-199).
We know that Mars has lots of water relatively accessible in its polar ice caps, as well as likely underground deposits all over the planet. By contrast, the only known water on the Moon is the recently discovered ice lurking in shadowy craters at the south pole. And based on the results of Viking and analysis of Martian meteorites, we know that its soil has all of the necessary nutrients for growing plants in.
The other key to agriculture is a source of energy for growing crops. To use natural sunlight for growing crops on the Moon or a space habitat would require huge greenhouses. To shield the crops from radiation, the glass would have to be 10cm thick! Mars has enough of an atmosphere to filter out most of the radiation, so farms on Mars would simply require a pressurized tent. Artificial lighting is also a possibility, but the energy costs become prohibitive, particularly when free energy from the Sun is available.
Most of the above factors apply, less generally, to human habitation. The Martian atmosphere will be easier for humans to get around, and it'll mean that settlements require less shielding against radiation. The Martian day is close in length to that of Earth. The surface gravity of Mars is higher than that of the Moon, so adapting to it will be easier. And so on.
Mars is the best place for human habitation outside of the Earth. That makes it the most desirable colonization target, but even for temporary outposts, the same factors will make those bases cheaper and easier to sustain.
You'd probably be interested in ProFont - a font designed for programmers, which has existed for years, but few outside of the Mac programming community know about it. It was specifically designed to be readable at 9 point, with similar characters distinctly different, as this page demonstrates. The full distribution includes TrueType, Type 1, and bitmap versions of the font for Mac and Windows. You can also download a look-alike bitmap version for Windows here.
I've been using ProFont for years as the font in my editor when coding, and found it very helpful.
I agree that the books on the various Java APIs quickly become outdated and that it's easy to pick up the core language from the online references. However, Effective Java by Joshua Bloch doesn't fall into either of those categories and is very useful. It's the Java equivalent of Effective C++ by Meyer - roughly 50 concrete dos and don'ts to improve your code, no matter what version of Java you're using.
I'll also second the nod for Python Essential Reference, which is the only Python book I've found useful so far, even if only because it's faster to look things up in than in the excellent included language documentation.
These books should be on the shelf of every programmer:
And here's a list of the best computer history books I've encountered:
I'm not usually given to conspiracy theories, but I think that this is exactly the point of the whole ICANN debacle. The DNS is one of, if not the only, chokepoint in the Internet, the rest of which is largely run in a decentralized, rational, and transparent manner. An open, uncontrolled and uncontrollable Internet is a threat to many established power groups.
The only theory I can come up with that fits all of the facts surrounding ICANN's behaviour is that some powerful group is behind ICANN and using the DNS to gain control over the Internet, for power and/or money. The way the management has given Auerbach the runaround makes it clear that the elected board members were intended to be powerless rubber stampers, which is why they're now dropping all pretenses of elections, since one of those elected directors is actually trying to do his job.
Kudos to Auerbach and Gilmore for standing up to these bullies. Meanwhile, I think that the technical Internet community should be building alternatives. Going up against the established powers on their own turf of the courts and political system is foolhardy. Even if we win this particular battle, the power nexus that the DNS roots imply will continue to attract the corrupt or corruptable. A technical solution that completely decentralizes the problem that DNS tries to solve is the best long term strategy, eliminating this particular threat forever.
This idea is completely in accord with one of Gilmore's maxims: "The net treats censorship as damage, and routes around it." The censorship and damage due to ICANN is clear, and we need to find a way around it.
You didn't need to buy XTools for NEdit. It also works pretty well with XFree86, although I've noticed an issue where NEdit doesn't set the mouse pointer correctly with it. And with Fink, installing NEdit and XFree86 is a snap.
No, the prohibited act is keeping the copy after selling the original. Making backups is permissible under any reasonable interpretation of fair use, but if you do create additional copies, you are supposed to destroy the copies when you sell the original.
Quicken 2002 Deluxe is a Carbon program that works great under OS X. I know, because I bought just it last week.
The next release of Office, due this fall, will also run on OS X. The major Mac news sites have all been shown demos of it. Adobe's silence on an OS X version of Photoshop is indeed troubling.
I dispute the contention that choosing Be over NeXT as the basis of OS X would have made a difference in software availability. Be's APIs are just as foreign from the Classic MacOS ones as NeXT's are. No major software house with a large body of working code was going to rewrite it for either Be or NeXT-derived APIs. If Apple had have gone with Be, it would still had to have had something like Carbon for established MacOS software.
There might have been a difference in what native software was available. I haven't looked at BeOS in a while, so I don't know if it has more or fewer apps to bring over than the NeXT world did. But I do know that the BeOS certainly had nothing like OmniWeb, which is very impressive.
I've glanced at the DMCA several times, and it's not clear to me that it criminalizes these acts. Rather, it criminalizes the means of doing so. There appears to be nothing in it that says you can't print an ebook for example, but if the publisher uses an access control measure to prevent that, it's illegal to circumvent that access control. It's also illegal to give the means to do so to other people (the trafficing provisions that Sklyarov is being charged with).
This is a clever sidestep that makes the DMCA such a strange and dangerous law. The U.S. Congress has effectively delegated its law-making authority in this area to corporations. If Adobe or some other corporation had wanted to make it illegal for someone to print the ebook they just bought, it would have formerly been necessary to lobby and get a law passed. Now, they just add some options to their access control and it's effectively illegal. Sure, no law has been passed removing one's right to print ebooks, but if there's no legal way to exercise that right, does it matter?
This also shunts decision making away from the public view (although, there doesn't seem to have much in the way of public debate when the DMCA was passed). Formerly, any attempt to change the rights and freedoms associated with the use of intellectual property would have had to been done through a vigorous public debate involving all affected parties. Now, new restrictions can be added by unaccountable individuals without any public debate. Any attempt by third parties to undo those restrictions and restore the former status quo is conveniently made illegal.
The distinction between copying and access that you noted is again another clever tactic on the part of the DMCA's framers. U.S. caselaw already has numerous precedents defending the right to make personal copies (the Audio Home Recording Act for example, or the Betamax case), so a law totally denying the right to make copies would be hard to defend. But under the Catch-22 of the DMCA, it doesn't matter, because to make copies you either need to get access to the work (which is illegal) to copy it, or make a copy somehow else, in which case it's still illegal to access that copy.
Defenders of the DMCA are quick to allege that the only possible use for circumvention devices is widespread unauthorized distribution of copyrighted materials (ie, piracy), and give the prevention of things like Napster as a reason for enacting it. It's interesting to note that in the DMCA cases brought forward so far, this is clearly not the primary use of the devices in question. DeCSS was written to allow the development of a Linux DVD player. Felten and company's research was ordinary academic study in cryptography. AEBPR allows the purchasers of ebooks the rights they would have enjoyed by buying books in physical form. Even if one argues that it is not necessarily the case that making copies was not the primary purpose of any of these devices, it could certainly be plausibly argued in a court of law. While all three devices could theoretically be used for unauthorized distribution, so too could the normal playing devices being sold. It certainly does not appear to be the case that the DMCA is effective in preventing piracy in this way.
So yes, I do feel that the items I emphasized should not be illegal. They do no harm to others. Copyright holders are not emperiled because someone prints out an ebook or copies it to another device for offline reading. Likewise, a computer audibly reading text is no danger to anyone. The only item on that list which is harmful is the distribution of copies, which is already illegal. I find it hard to support a law that criminalizes harmless actions. The DMCA offers no benefit to the public at large, only to those corporations who hold large numbers of copyrights. Given the choice between padding the bottom line of a large corporation and defending the freedoms associated with reading, I know which I'll choose every time.
If longstanding copyright law no longer applies in today's world, it should not be rewritten from scratch by fiat, but rather hammered out in a compromise protecting the interests of all affected parties. The DMCA is far too one-sided to be described as such.
Well, it seems to be official. Fair use in the U.S. is dead. Look at what the indictment boils down to:
Note that the indictment clearly indicates that AEBPR is only useful to purchasers of ebooks in Adobe's format, so there can be no allegation of it being used for widespread piracy. Instead, Sklyarov's apparent crime is to allow people to actually use the ebooks they've bought and paid for. Of the items enumerated as being restrictable by the publisher or distributor, only distribution is forbidden by copyright law prior to the DMCA, and then only when the fair use exemptions don't apply. It seems rather overreaching to me that the DMCA criminalizes being able to do such ordinary actions with an ebook such as having the computer read it aloud or print it, let alone making copies for backup or use on another machine.
Note also that the indictment makes no mention of the AEBPR being used to violate copyright law. No evidence is offered that any of the handful of its purchasers used the program for any illegal purpose. The mere fact that it allows the purchaser full use of a bought ebook and the theoretical possibility of commiting an act (unpermitted distribution) which is already illegal under century old copyright law, is reason enough to send a man to jail for 25 years. Scary.
And publishers wonder in vain why ebooks aren't selling very well? Gee, if you don't let the purchaser do anything with them, making ebooks far more restricted and less useful than print books, and totally upset the balance between public and private interests enshrined in copyright law, you should expect this. Indeed, I'm frightened that ebooks have sold as well as they have. The freedoms and rights associated with reading seem to no longer apply in the digital world if the interests that bought the DMCA have their way.
My favorite Poul Anderson work is the Time Patrol story "The Sorrow of Odin the Goth", which brings a tear to my eye every time I read it. It would have to be closely followed by The Boat of a Million Years, which makes being immortal sound like a burden. These show Anderson at his best, either taking the long view of history or making some ancient piece of the past seem alive and exciting. No other author I've read has had that delicate touch for history, and for that he will be missed.
Little known-fact: I read somewhere that Anderson's novel Three Hearts and Three Swords was one of the strong influences on the creation of Dungeons and Dragons.
I disagree strongly. A CS program should teach several languages, and the issue of whether those languages are popular or not should be irrelevant; rather, the languages should be chosen to be the ones that best facilitate the teaching of the particular course. A non-mainstream language actually has a small advantage for the first year intro to programming course, since it means that few students coming in will already know it and be bored out of their skulls.
I agree that one of the languages learned during a four year CS should be strongly mainstream (probably C, the current lingua franca of computer languages), but if a four year CS student can't pick up several languages during that period, something is wrong. In my alma mater's CS department, computer science students learn four new languages during the second year alone (Eiffel, C, Prolog, and MIPS assembly) for the required classes. In third and fourth year, which are all electives, most classes require learning new languages as well. With a strong foundation on a good teaching language in first year, upper year CS students should be able to learn new languages fairly easily.
Python makes an excellent language for teaching beginners. In no way should its lack of popularity be held against it in that role, save possibly the lack of introductory CS textbooks using it.
Try the XFree86 on Darwin and Mac OS X page. Most of the work is done through the XonX project at Sourceforge. I believe that in order to get the rootless X in Aqua working, you need to build the latest XFree86 and XonX from CVS. Check the archives of the Darwin Development mailing list at darwinfo.org for postings by Torrey Lyons on how to compile it.
Check out the Python - Objective-C bridge project at Sourceforge. It appears to be coming along slowly. In the mailing list archives you'll also find discussion of using Jython with Apple's Java bridge as well.
I'm using NFS between my Linux server and my OS X workstation, but it's a bit flaky. OS X can import NFS filesystems if you set some options (look in the forums on macnn.com for details), but I've found that I need to stop and restart the NFS server processes every time the OS X box tries to import a new filesystem. OS X can export via NFS so long as it's from a partition formatted as UFS (the default is HFS+ for Classic compatibility, which does not work with NFS).
There is no fancy GUI to enable NFS, so you have to be familiar with the Unix command line and the NeXT-derived NetInfo system in order to get NFS working.
In the Public Beta, I found some problems where non-Unix applications could not see NFS-imported filesystems, but that appears to have been fixed in the Final Release.
First off, I'm assuming that this is the case where aliens come to make contact with humans on Earth, and not vice versa. The balance of power is very heavily weighted in favor of the party that's going out and making contact on the other party's territory. Our own history of the European settlement of the Americas and Australia shows this quite clearly, and in that case both parties were of the same species. With the even greater potential for misunderstanding between different species, in the long term such contact is going to be even more messy.
Anyhow, the short answer, any human government with good sense will treat alien visitors like visiting royalty, or at a minimum, like foreign diplomats with immunity to local laws. Any being with the technology to cross the interstellar gap is powerful enough to threaten every living being on Earth, so it only makes sense to treat them very carefully, avoiding giving offense or letting them come to harm.
Scientist and science fiction author Charles Pellegrino has pointed out that the energy needed to accelerate a spaceship up to nearly the speed of light is astronomical, and could do unbelievable damage. Life-bearing planets are large and move slowly and predictably, and it's impossible to defend against something coming at you almost before you detect it. If a missile of large mass travelling at 0.99c were to collide with a planet, it would do damage that makes the dinosaur killer look like peanuts. This scenario is graphically portrayed in his novel The Killing Star, which posits that this is the reason why we haven't detected any alien intelligences. It's impossible to defend your homeworld against this sort of attack, so the only rational thing to do is be as invisible as possible, and then go around killing potential threats before they materialize.
Even without planet-killers used by a Berserker race, and hypothesizing some sort of FTL travel with much lower energy, there are lots of ways star-faring aliens could do serious harm to Earth people. The short story "The Flies of Memory" had a particularly amusing example. In it, aliens looking like giant mosquitos came to Earth to tour our famous landmarks. After a human lunatic killed one, it was discovered that the things the aliens view become tied to them in their memory, such that if the alien dies, all of the landmarks it remembers ceases to exist.
And this is just addressing potential physical harm. I imagine that no matter what they do, any visiting aliens would irreparably transform the human psyche. First contact is rather a cultural singularity - after it, everything will be different. Assuming aliens trying to minimize the damage to our worldview, just the demonstration of non-human intelligence and the technology they possess will be a huge hurdle.
So, I guess I'm with the cynics on this one. If aliens come to Earth, the question is not so much what rights they have (in all likelihood, they have enough power to do whatever the hell they please), and the real disturbing issue is how the poor indigenes (ie, us) will fare. Our own history is a pretty disturbing precedent.
First contact and its effects have been discussed and thought about in written science fiction for decades. There are many more stories than the ones I mentioned above considering angles of this issue, so you might want to take a trip to the library.
A lot of people complain about Python's required use of whitespace, but I've never found it to be a problem. I always use consistent indentation in my programs no matter what language they're in. So I find this feature to be a plus, because I never have to run into the common mistake in C of confusing indentation with for and if statements, like the following examples:
C
if (a==b)
pr intf("a equals b\n");
else
pr intf("a not equal to b\n");
pr intf("Always runs, regardless of a and b\n");
Python
if a == b:
pr int "a equals b"
else:
pr int "a not equal to b"
pr int "This is part of else clause"
(Blame Slashcode for the extra spaces in the print statements.)
Note that in the Python version, the behaviour is exactly what one would naively expect from a single glance at the arrangement of the code. Even non-programmers can read Python code and understand it readily. I find that Python implementations of algorithms look a lot like the pseudo-code I had to generate for assignments when I was in school.
Indeed, I consider Python's use of whitespace to be an advantage beyond merely preventing this common mistake. After starting my internship at a big company with a multi-million line codebase, I'm strongly of the opinion that the source code should clearly communicate its intent to human readers. Python makes it difficult not to write clear, readable programs, without relying on the strong programmer discipline needed to write maintainable programs in other languages.
Along these lines, the one feature of Python syntax that does get in my way is the colon. I sometimes forget to add colons (like the one on the end of the else statement) precisely because I've gotten used to the increased indentation being sufficient and necessary to communicate my intent.
Another annoying thing is that the line ending characters must be the ones defined on the local platform, so I can't move a script from my Linux box to my Mac without running it through a translator. This will become a big issue as the Mac and Unix versions of Python merge together for Mac OS X.
I haven't looked in any great depth at either Perl or Ruby, but my initial impression is that their syntax is a little foreign. I was able to learn Python quickly precisely because the syntax was familiar and predictable. This is consistent for me - I originally came to Python from the Dylan world, a language whose claim to fame is being Lisp-inspired but with an infix syntax to be palatable to C programmers.
Since taste in syntax is a personal thing, I'm not going to flame Perl or Ruby users (indeed, both languages are cool and have their strengths). But I do encourage people not to obsess on the whitespace issue when considering whether to try Python or not.
Dealers of Lightning: Xerox Parc and the Dawn of the Computer Age by Michael A. Hiltzik
This covers the history of the hugely influential Xerox PARC research center from its founding through the glory days of the 70s when they invented modern desktop computing. It's an important part of computer history that was typically only recorded as "Steve Jobs steals the GUI from a visit to Xerox PARC" until this book was published.
No one has yet pointed out that turn-based games work really well in multiplayer if the players are't all free at the same time. The PBEM (Play By EMail) mode of games like Alpha Centauri lets you carry on a game even when all of the players are on completely different schedules.
I have a good friend who's as into turn-based strategy games as I am. In the past, we've gotten together for a full day playing Master of Orion II, FreeCiv, or Alpha Centauri on a LAN, but that takes a lot of time and requires us to both be free at the same time. He and I now live in different cities in different time zones with different schedules, so it'd be very difficult to set up a simultaneous game over the Internet.
Instead, since August, we've been playing near continuous games of Alpha Centauri via email. When I finish a turn, I send it to him in email and vice versa. This lets the game go whenever we have time in our respective schedules, and gives time to think about strategy and tactics between turns without holding up the other players.
This is easily extensible up to the maximum of seven players allowed in Alpha Centauri. The Apolyton Alpha Centauri Multiplayer forum regularly has games going with players from all over the world, playing their turns whenever is convenient for them in their own timezone.
Alpha Centauri is a long game and takes a lot of time, so you can't complete a full game in a single sitting as with RTS games. Coming from a long line of turn-based games it works very well in that mode, and turn-based gaming is very compatible with PBEM. I'm also aware of other games like VGA Planets which also are turn-based with PBEM modes, so the idea that turn-based games aren't multiplayer is a myth.
Turn-based games aren't as popular as RTS games, but there's a sizable contingent of people who prefer them, and will continue to buy them as long as there are companies that produce good ones. The next year looks very promising with Civilization III and Master of Orion III coming out, and I hope they both have PBEM capability.
I'm running Debian on my main Linux machine, and have NetBSD installed on another. I like both methods better than the RPM-based systems I've used in the past, with Debian's being my current favorite. I have a couple of points to add to the comparison:
I think they're both good packaging systems, and not different enough to be a reason to switch between Debian and a BSD or vice versa.
This is a famous question in natural language linguistics, known as the Sapir-Wharf hypothesis. It posits that what humans can think about is limited by the set of terms in their language. So far, all of the experimental evidence disproves it, even though the idea sounds very intuitive. It appears that all human languages meet a certain level of expressivity that allows us to think and talk about anything, no matter the language used.
In computer languages, which are all artificial and of much smaller scope than any natural language, the question is still open. Anecdotally, programmers know that some programming languages make solving certain categories of problems easier than others, sometimes by orders of magnitude. It's been suggested on Slashdot before that a variant of the Sapir-Wharf hypothesis may hold true for computer languages, even if it doesn't for human ones.
Who'd'a thunk Philosophy and Computer Science would have anything in common?I've always felt that computer science is in many ways a field of applied philosophy. There are many crossovers between the two fields. Consider, for example, that the Boolean logic used as the basis of all digital computing was invented by a theologian as a way of answering questions in metaphysics and theology.