One of the proposed rules for unknown pronouns is that the writer uses a pronoun that agrees with his/her own sex. Thus, in the phrase "what X needs" where X is an unknown third person singular, a woman would write "what she needs" and a man would write "what he needs".
David Weber uses this rule consistently in his fiction. A female character will even say something like "we don't know what the average woman-on-the-street thinks" rather than "man-on-the-street".
P.S. I think political correctness made a travesty out of the opening narration from Star Trek: The Next Generation. The original "where no man has gone before" clearly was using "man" in the sense of "member of humanity", or "human being". The new text, "where no one has gone before", carries with it the assumption that all those aliens out there living on those planets don't count; they aren't "anyone". Thus a desire to be PC caused a perfectly neutral statement to become offensively humanocentrically chauvinistic. It is only one short step from there to a new "Manifest Destiny", with the Federation becoming an evil expansionist empire. Such is the power of words! [1]
Does anyone seriously think that "mankind" only refers to the male half of humanity? (Sadly, I'm almost certain that some hard-core feminists would take that position, and argue for "humankind" or some such.)
[1] Tongue firmly in cheek, of course, but I have met people who might seriously advance this argument. Of course those same people also jump all over any words they deem sexist. I think we would all be better off if we were less upset over mere words.
GCC provides (portable) vector types, and if you declare your variables as these then it just has to try to use SSE / AltiVec / Whatever instructions for the operations
I very much want to know more about this.
Is there a book or web site that you recommend where I can learn more about this? The GCC manual doesn't have much about it.
As for redundancy... put two GPS receivers on your ship.
Suppose that someone shoots down enough GPS satellites to disable the GPS system. What then? I'm in favor of some sort of backup system.
It's difficult to shoot down satellites, but not impossible.
That said, I have to assume that ships could still navigate the old-fashioned way, with an sextant, a chronometer, and some charts. You can automate the calculations, so you just take the readings with the sextant and put the numbers in to a computer program. It is even possible to automate the whole process, although nobody seems to bother anymore in these days of GPS.
the Galactica didn't have networked computers, so the Cylons needed agents in place.
I salute you, sir! You have managed to come up with an explanation that holds together pretty well. I wish I had thought of that.
Sadly, I'll bet that the actual writers just never thought about it. "Hey, it would be cool if <X> was also a Cylon and didn't even know it! We still have 5 models of Cylon we haven't shown yet!"
the idea of a convenient star system with dozens of planets and hundreds of moons able to be terraformed to some extent, and conveniently having nearly earth-normal gravity, is a workable plot device
I hope it was clear that I agree with you here; in fact that was my point. I have issues with the science, but the issues were not bad enough to jar me out of suspension of disbelief; and that is all you can really ask of any show.
As a counterexample, look at the new Battlestar. They had FTL in the form of "jump drive", but it was really out of place, because all the rest of their technology was really not that much advanced from our own: nuclear-propelled Viper ships, machine guns just like our own, nuclear missiles for shooting at planets or enemy ships, seriously low-tech computers, etc. However, because of the nature of Battlestar's story, FTL was an absolute requirement.
Well, to nitpick, they had low-tech computers because they didn't want to be hacked by the Cylons. The Cylons were shown to have magical hacking abilities; all you had to do was connect multiple computers in a "network" and then a dramatic Hollywood computer hacking scene would follow instantly.
My big problem with the science in Battlestar Galactica is that they didn't have any kind of "shields" technology, yet they were shown taking hits and not losing the whole ship. If you have the ability to make those kinds of ships, you would have the ability to make missiles that could impact with a serious amount of kinetic energy. And you would lose lots of fighters to tiny bits of debris hitting at high relative speeds.
I also found it rather silly that, while there were only 12 models of human-appearing Cylon, there were at least three "sleepers" on board the Galactica. That might have made sense in the original, where the Galactica was the best ship in the fleet (and Adama was a revered figure). But one of the things I liked best about the new show was that the Galactica was an old ship, the last of the old Battlestars; and Adama was just some stubborn old guy who would be retiring soon. The reason the Galactica escaped the Cylon Trojan Horse was that Adama had successfully resisted all new computer upgrades; and he got away with that because it was an old ship that nobody really cared about. (Hmmm... how did the Pegasus escape the Trojan Horse? I don't remember.) Anyway, if the Galactica somehow had three, the whole fleet must have been shot through with sleepers. It makes less sense the more I think about it.
The other thing which both shows seem to have which is out-of-place is artificial gravity. But it's nearly impossible to make a TV series that doesn't have artificial gravity (Avatar had a brief scene at the beginning with zero-g, but that was a half-billion dollar movie).
I'm always willing to believe in artificial gravity and FTL for an SF show, and as you said, it's nearly impossible to not have it. Avatar not only had a lot of money going into it, it was computer-generated. To make actual actors appear to float, you need wires, and digital wire removal, and it's going to be a pain and cost a lot.
tidy up the 'I don't know the difference between a galaxy and a star system' bug in the original
It was explicitly said that the Firefly ships (like the Serenity) and the other ships we saw did not have FTL ability, and spent all their time in one star system. It was also said that that one system had dozens of planets and hundreds of moons. It was never said how people managed to travel to this one star system (I'm guessing sub-light travel with the passengers in suspended animation), and the only reason ever given for why people traveled here was "Earth got all used up".
I don't think the solar system in Firefly is very plausible. And I have issues with some other science (a box of concentrated food was shown as very valuable, valuable enough to be worth flying the Serenity out to a remote moon for a trade; surgery to re-attach a severed ear required highly special medical equipment; getting victim ships to fly through a giant hoop that killed everyone on board was extremely silly; it's hard to believe that Reavers could even fly ships, let alone keep their ships in good working order; etc.). But I felt it was not so bad that it jarred me out of suspension of disbelief.
I had never seen the term "smartbook" before. This article defines a "smartbook" as a netbook with a non-x86 processor (likely ARM).
I guess it's a portmanteau of "smart phone" and "netbook". Or maybe it means "smart enough finally to use something other than x86 for an ultra-portable device".
I suggest you get some books that make math more interesting.
For any software developer, the math in the "Aha!" books is appropriate. I was able to understand this math in high school, so I'm sure hard-core math nerds will sneer that this is too easy, but I stand by the claim that these books are worth your time.
From my own personal experience, I have never needed calculus or differential equations or any of that advanced sort of math in my whole career. I'm now doing somewhat advanced DSP work, and even there I haven't needed advanced math. (I don't entirely understand how the FFT works; I just know how to use its properties to get the result I need. Other people wrote C versions of the FFT for me; I haven't needed to write it.)
The math that has been useful to me is basic logic stuff, to know how to write conditions for if statements and the like; O(n) estimations, to help you choose the best algorithm to solve your problem; and basic probability stuff, to help me understand how caches work and such. So the first class sounds better to me than the second.
I'll recommend one more book to you. It's a sort of encyclopedia of algorithms... it will expand your mind with possibilities (if you don't even know something exists, you won't be able to use it to solve your problems). It's engagingly written, with "war stories" that make it more lively.
Disclaimer: I'm not any sort of expert on this stuff. Ignore me or laugh at me as you like.
I'm dubious about this. The scale of the thing is staggering, and it's hard to believe it will produce a better electrical output than if you spent a similar amount of money building a molten-salt solar thermal plant instead. Unlike molten salt solar thermal, this won't make electricity at night.
The one thing that makes this interesting is that it combines a giant greenhouse with the energy generation. If you can somehow make the greenhouse part very profitable (growing exotic fruit that is expensive to transport, or some such) then maybe you might have a payoff to match the expense. Maybe. But I'm dubious.
they actually had a video phone product in the 1980s, and then, nobody wanted it
I remember that product. It used a modem to put video over ordinary phone lines. That's right, less than 32 kbps for full motion video! It had a screen the size of a postage stamp and the video updated as smoothly as a slideshow where the slides were all coated in honey. (I never had one, but (a) I read about them, and (b) I know just how little bandwidth 32 kbps is for motion video.)
A better product for you to mention would have been H.261 video over ISDN. That actually had a nonzero number of users, but still didn't light the world on fire. DSL and the cable TV infrastructure were better ways to connect people to the Internet and ISDN became irrelevant and unloved.
We actually have the technology now to put practical video phones in every home. We just don't bother because it is not worth the cost. People do want Internet on their cell phones (to Google for restaurants, or movie show times, or whatever) and people do want WebEx and similar tools (sharing a PC desktop is way more valuable than showing video of someone talking, and with a USB webcam you can also do the talking head thing if you like).
Someday every home will have a video phone, if only because it will be so cheap to do it that it will be built in to every computer, and some computers will be wall-sized flat panel home theatre convergence devices. The main early adopters probably will be families with babies whose grandparents live far away.
The article submitter must be trolling. Decades ago there existed a one-off prototype, which was never widely deployed, that was hugely expensive. Now there exists an inexpensive learning gadget that might actually be in the hands of actual kids, and this is "moving backwards"?
Next up: is the phone industry moving backwards? At a world's fair, AT&T demonstrated a working two-way color video phone, yet I don't have a video phone in my house yet. Of course, millions of people have full-color Internet on their phones, and can do things like view a photo of their home taken from orbit. And millions of people have practical teleconferencing via WebEx et al. But never mind that. The phone company doesn't have video phones in every house; we're moving backwards!
Thank you for taking the time to write a nice long answer to my questions. I appreciate it.
I hope you are able to work out your Ardour issues. If not, you might want to look into running Cubase under VirtualBox on your Linux computer; I have been generally pleased by how seamless and bulletproof VirtualBox has been in my testing. (But I haven't tried anything like Cubase; if it does realtime USB, that would be far more serious than anything I have tried.)
Hard to reconcile your claim that fights would be over quickly with the reality that fights between trained martial artists (e.g., UFC fights, Olympic judo matches, boxing matches, Olympic TKD matches, etc...) the majority of the time go to decision.
These fights do not allow crippling blows. No eye gouging, no cheap shots below the belt (or else, if they are allowed, the men are wearing cups), definitely no neck snapping and I doubt full-on hits to the brain case are allowed. These guys aren't even allowed surprise attacks; they can't attack until the referee says it is time.
Unless you know of a martial arts fighting contest where the losers are typically hauled off to the hospital or the morgue, I will not accept that they perfectly model real full-on fights.
On the other hand, I'll grant you that martial artists doing full-contact kick each other pretty hard in various parts of the body without the fight being decided.
Also hard to reconcile your claim that a kick to the knee would end a fight with the reality that kicks to knees are legal in lots of martial arts competitions (MT matches, UFC fights, K1 fights) but almost never end fights.
Hmmm. And these people have unpadded, unprotected knees? I've constantly heard this meme that the knee is rather fragile. If you are correct, and martial artists are allowed to attack each others' knees full-force, then I must be mistaken on this point. (I am not any kind of expert on lethal hand-to-hand fighting and I don't claim to be one.) How often do the losers have to go to hospital for a messed-up knee?
Next up for Linux, media production software. What the fuck is up with Hydrogen and Ardour? Can't they get at least one real musician on their design staff?
I want to learn how to use this sort of software. I have read some positive reviews of Ardour, so I'm interested to hear what you don't like about it, and what you do like. What do you usually use, if not Ardour? And whatever you use, does it run okay on WINE?
The fight scenes are epic, but in real life they'd be... more epic.
I think that in real life, fights tend to be over a lot faster than in movies/TV. The first guy to land a really good hit gains an advantage, then presses the advantage, then wins. If I let you get in a good shot to my chest and I'm sucking air, you can then land several more attacks and lay me out. If you even land a good hit against my knee, I've probably lost the fight right there. This is all assuming a real fight and not just posturing in front of an audience.
Also, in movies/TV, people seem to be able to run flat out and fight with 100% intensity for minutes at a time. You especially see this in superhero movies, even where the hero has no actual powers (Daredevil, Batman, etc.)
P.S. My favorite fight scene in fiction is probably from the story "Gulf" by Robert A. Heinlein. The protagonist is a super-intelligent spy, and he is expecting some sort of trouble from the bad guys; since he is in a public place he figures their first move will be some kind of diversion. The bad guys light off some fireworks, and almost everyone around the spy turns to look in the direction of the bang; two people near the spy do not look and head straight for him. Without hesitating he attacks them, and lays them out instantly, and then walks away as if nothing had happened. I thought that was a lot cooler than a three-page fight scene would have been.
IMHO everything is wrong with C as first language. It gives you dreadful programming style and is not a right tool for application programming. You can mod me as a troll if you want but you've got to chose the best tool for the job and C is a tool for writing operating systems for fuck's sake. It isn't even a high level language.
Not so. C is the king of the third-generation languages. With respect to your points, in order:
C doesn't enforce any particular programming style, and all major applications are written in C or C++ (so I don't think you can claim it is "not a right tool" for that). C is a tool for writing operating systems or anything else, and I am going to claim that C is a slightly higher level language than Pascal; it can do anything you can do in Pascal, often with cleaner code. (For example, it is much cleaner to simply return a value, than to assign the return value to the function name and then use a series of if statements to make sure that nothing else happens until the end of the function body is reached.)
All those buffer overflow security holes happen because of both typical "clever hack" C programming style and choice of using a language for writing operating systems to write business applications.
Buffer overflows are caused by using a language that doesn't do run-time bounds checking for you, combined with a standard library that includes functions that don't do explicit bounds checking either. strcpy() is considered harmful, you need to use functions that know how long the buffer is and refuse to write past the end. I note that there is nothing in Pascal to make it better than C here, except for the fact that it is really impossible to write a useful standard library in Pascal. (Different length strings have different types so there is no way to write a general-purpose library function that can accept strings of different lengths! That's just crazy! It's so crazy that every major Pascal implementation offers some way to work around it. Whereas C got this, like so many other things, right; and there are plenty of standard C string functions.)
Pascal wasn't a very good teaching language for nothing - it forced you to write software in a very clean and readable way.
It did nothing of the sort. You can write wretched and awful code in Pascal just like anything else.
The closest I have seen to a language that forces you to write clean software is Python; it forces you to put colons at the end of if statements and function definitions, just for readability; it forces you to indent your code consistently, or it won't work. And yet you can write wretched and awful code in Python just like anything else.
A language cannot force good code. At best, it can support good code. C does this decently; Python does it better.
Pascal is pretty much irrelevant now, and C is still very relevant now, which should tell you something.
P.S. Getting back to 12-year-olds, I think Python would be the ideal first language. C would not be horrible; it would be better than C++ because there is much less to learn. Pascal isn't bad, but it is a dead-end irrelevant language, so I wouldn't bother with it.
When the OLPC was first announced, I was surprised that they were making things so hard on themselves. A clamshell, with keyboard, with color screen? Trying to hit a $100 price point? It seemed like that would be hard, and it was... hard enough that they didn't hit their target. The OLPC XO costs double the target price, it is glacially slow, and at least the one I bought has a totally unusable touch pad.
What I thought they should have done was to make something rather like a Handspring Visor, but bigger. A tablet with a touchscreen and a stylus. One piece, no hinges. It could have a flip cover, and the cover could have little pins that go into molded dimples on the tablet; the cover would protect the touchscreen when closed, but would not be actually part of the tablet and would be easy to replace. But a plastic cover isn't a strict necessity; a slip case of cloth or vinyl or whatever would serve just fine.
Color is great, and kids love it, but it is not necessary and hitting the price point is more important.
I remember when the Palm PDA first came out, I read about attachments you could get to hook up science probes (thermometers and pH probes) and how teachers were taking kids out to ponds or wetlands to measure things. No color screen there, not needed.
I have carried a Palm or Handspring PDA since the Palm first came out. I have used them for reading books (mostly fiction but some nonfiction). The core mission of the OLPC really is to serve as the textbooks for the children. No color screen needed.
So, let's imagine a tablet, with a touchscreen and a stylus. The stylus stores in a "silo" as with a Palm PDA. The screen is maybe 6 inches (15 cm). Like a Palm PDA, it has a few "hard keys" along with the screen, and in fact let's give it a game pad (direction pad) like a Nintendo DS, and standard 1/8" mini jacks for headphones and microphone. It has whatever CPU makes technical sense, likely an ARM (certainly not an x86). It even has a USB connector or two, making it possible to prop this thing up and plug in a keyboard (whatever you can buy cheapest from some Chinese manufacturer). It would be great if it had an SD card slot, but with the USB connector you don't actually need an SD card slot and it's more important that it be simple and cheap to produce. Design it brain dead simple: take it apart, and you have the mainboard, the screen, the battery pack, and the case. Could this be built for the sub-$100 price point? Yes. Now upgrade the screen to color, if you can hit the price point. Then maybe add WiFi, if you can hit the price point. Cameras are fun too, if you can hit the price point.
If you can save money by making the screen the exact size and shape as would be used in a hugely mass-produced portable DVD player, do that. If you have to, make a dual-screen design like a Nintendo DS that uses standard cell phone displays. As long as you can make a device that can be field-repaired and cheaply mass-produced, you can be flexible with the design. But it doesn't have to look like a laptop and it shouldn't have any hinges or other moving parts. Maybe it doesn't even have a DC power jack; maybe you charge it by plugging in to the USB port. Or maybe not; I'm not an expert on this stuff, have an expert advise you on what is most cost-effective.
Now mass produce the things and sell them to anyone who wants one. Get the production quantity up. But hit the frakking price point.
For bonus points, make thermometers, pH probes, volt/ohm probes, and such that plug in to the USB or the microphone jack, and mass produce those as well.
Actual devices like this, in actual school children's hands, would be far better than a more ambitious device that doesn't hit the price point and only gets deployed in a very few locations.
P.S. I'm glad the XO-2 seems to be dead. Dual full-color touch screens with a hinge? That is not a design for inexpensive mass production.
Alas, Mono is still a part of the default Gnome distribution, just so they can have a note taking applet
Oh, "just" so they can have a single applet? It couldn't possibly be because they think it is a generally useful way to develop applications, such as F-Stop and Banshee?
Mono may or may not be a good idea, but you are framing your argument in an intellectually dishonest way here. That note-taking applet ("Tomboy") may be the only thing in standard GNOME that needs Mono right now, but I'm pretty sure that there will be others.
Even worse, there are folks pushing Banshee as the default music player so there's another dependency on Mono.
See? Then it won't just be Tomboy, there will be other things using Mono.
I haven't tried C#, but a lot of people seem to like it. If having C# means I get more free software to play with, I'm in favor of that.
The major argument I have seen against Mono is "Microsoft is just waiting and they will assert patent claims!!" In that case, the only thing that they can do is force people to stop using C# and Mono. In which case, all the Mono apps will be pulled or re-written. And at that point, you would have what you seem to want: no more Mono in GNOME.
That is the worst-case scenario. And I don't see it as being bad enough to try to keep people from using Mono. If people want to use Mono to write free software, that's fine with me.
I'm curious: now that Java is becoming fully free, would you support re-writing Tomboy and F-Stop and the others in Java? That way, instead of being bloated and slow C# applications, they could be bloated and slow Java applications. Would you be happier?
In my day job, I write wicked fast C code (small memory footprint, too). When I write software on my own for fun, it tends to be Python, which is even slower than C#. Do you have a problem with Python too?
Please provide a link to the FSF claiming that the LGPL is "less free" than the GPL.
Are you trolling? They renamed LGPL from the "Library" GPL to the "Lesser" GPL, because they feel it is less free. It's baked right into the name that they feel it is less free.
But you asked for a link. Here you go:
Using the Lesser GPL for any particular library constitutes a retreat for free software. It means we partially abandon the attempt to defend the users' freedom, and some of the requirements to share what is built on top of GPL-covered software. In themselves, those are changes for the worse.
Just in case that wasn't clear enough for you, let me rephrase it: according to this gnu.org link, the LGPL does not protect users' freedom as well as the GPL. It does not maximize freedom as well as GPL. In short, it is less free, according to gnu.org.
Remember that GNU and FSF are all about the users' freedom. Freedom of any developer to make proprietary software is not viewed as a good thing. A license like GPL that restricts the ability of developers to make proprietary software is viewed as more free.
On the other hand, fans of the BSD license argue that it is "more free" because anyone may do anything with the software. GNU and FSF reject this idea.
I format my external USB drives to ext3. Most of my machines are Linux anyway, and I can always plug the USB drive into my storage server and backup over Samba to any kind of drive supported by the storage server.
ext3 is pretty much stable and well understood. It just works. That's what I want for backup drives.
And my netbook has Ubuntu Linux on it, and ext3 performs well on the external USB drive there. I haven't tested NTFS over FUSE on the netbook, but I wonder about CPU overhead on the little Atom chip: it might be a little bit slow.
If you want a drive you can take over to your friend's house, and your friend just runs Windows or a Mac, then by all means NTFS.
Do you really purchase every VHS movie on DVD and then Blu-Ray? Do you purchase every LP on cassette tape, CD, SACD, DVD-Audio?
We really did re-purchase every VHS on DVD. We haven't been in a big hurry to re-buy as Blu-Ray... but I'll get my favorites, once I have a giant flat panel TV (which will be soon). For movies that are mostly about the dialog, or for old TV content that is innately standard-def, there is no reason to upgrade from the DVD. But I see the difference with high-def and I like it. (If you really want to drive home how much better high-def is at resolving details, look at the end credits for movies sometime. On DVD they are often unreadable, the text is just too small to resolve!)
I really did re-purchase every LP on DVD. I have not yet bought an SACD or DVD-Audio. I'm willing to buy multi-channel releases for artists I like, if I am convinced that the quality is there. (I'm leery of new releases; read the Wikipedia entry on "loudness war" for details.)
There might be a few I'm willing to upgrade, assuming the new version is remastered.
Yes, I agree. And it needs to be a useful remaster. I've noticed that some movies get released, and re-released, and re-released again, to milk the upgrade market. I don't buy re-releases very often, and if I do, you won't get me with a quick refresh.
Format shifting is nearly automatic at this point, you can digitize tapes and records
Eww, digitizing cassette tapes? Or worse, 8-tracks? Better to bin them and buy a CD, or even buy them as a digital download from an online music retailer. If you have an unscratched LP, that's more reasonable, especially if you have good audio gear. Or if you have an album that just isn't available on CD.
Slashdot Mirror
Pizza Hut didn't do much research on site location, but simply put stores near McDonalds
I read in the newspaper that Tully's has a policy of siting their stores near Starbucks stores. Same deal.
steveha
I wonder if the person who wrote that is female.
One of the proposed rules for unknown pronouns is that the writer uses a pronoun that agrees with his/her own sex. Thus, in the phrase "what X needs" where X is an unknown third person singular, a woman would write "what she needs" and a man would write "what he needs".
David Weber uses this rule consistently in his fiction. A female character will even say something like "we don't know what the average woman-on-the-street thinks" rather than "man-on-the-street".
P.S. I think political correctness made a travesty out of the opening narration from Star Trek: The Next Generation. The original "where no man has gone before" clearly was using "man" in the sense of "member of humanity", or "human being". The new text, "where no one has gone before", carries with it the assumption that all those aliens out there living on those planets don't count; they aren't "anyone". Thus a desire to be PC caused a perfectly neutral statement to become offensively humanocentrically chauvinistic. It is only one short step from there to a new "Manifest Destiny", with the Federation becoming an evil expansionist empire. Such is the power of words! [1]
Does anyone seriously think that "mankind" only refers to the male half of humanity? (Sadly, I'm almost certain that some hard-core feminists would take that position, and argue for "humankind" or some such.)
[1] Tongue firmly in cheek, of course, but I have met people who might seriously advance this argument. Of course those same people also jump all over any words they deem sexist. I think we would all be better off if we were less upset over mere words.
steveha
GCC provides (portable) vector types, and if you declare your variables as these then it just has to try to use SSE / AltiVec / Whatever instructions for the operations
I very much want to know more about this.
Is there a book or web site that you recommend where I can learn more about this? The GCC manual doesn't have much about it.
steveha
As for redundancy... put two GPS receivers on your ship.
Suppose that someone shoots down enough GPS satellites to disable the GPS system. What then? I'm in favor of some sort of backup system.
It's difficult to shoot down satellites, but not impossible.
That said, I have to assume that ships could still navigate the old-fashioned way, with an sextant, a chronometer, and some charts. You can automate the calculations, so you just take the readings with the sextant and put the numbers in to a computer program. It is even possible to automate the whole process, although nobody seems to bother anymore in these days of GPS.
steveha
The Coast Guard is going to "kill" LORAN? This choice of words worries me. What if LORAN decides to strike first, out of self-defense?
"LORAN", "SKYNET", both are short words with an 'N' in them. COINCIDENCE? I think not!
steveha
the Galactica didn't have networked computers, so the Cylons needed agents in place.
I salute you, sir! You have managed to come up with an explanation that holds together pretty well. I wish I had thought of that.
Sadly, I'll bet that the actual writers just never thought about it. "Hey, it would be cool if <X> was also a Cylon and didn't even know it! We still have 5 models of Cylon we haven't shown yet!"
steveha
the idea of a convenient star system with dozens of planets and hundreds of moons able to be terraformed to some extent, and conveniently having nearly earth-normal gravity, is a workable plot device
I hope it was clear that I agree with you here; in fact that was my point. I have issues with the science, but the issues were not bad enough to jar me out of suspension of disbelief; and that is all you can really ask of any show.
As a counterexample, look at the new Battlestar. They had FTL in the form of "jump drive", but it was really out of place, because all the rest of their technology was really not that much advanced from our own: nuclear-propelled Viper ships, machine guns just like our own, nuclear missiles for shooting at planets or enemy ships, seriously low-tech computers, etc. However, because of the nature of Battlestar's story, FTL was an absolute requirement.
Well, to nitpick, they had low-tech computers because they didn't want to be hacked by the Cylons. The Cylons were shown to have magical hacking abilities; all you had to do was connect multiple computers in a "network" and then a dramatic Hollywood computer hacking scene would follow instantly.
My big problem with the science in Battlestar Galactica is that they didn't have any kind of "shields" technology, yet they were shown taking hits and not losing the whole ship. If you have the ability to make those kinds of ships, you would have the ability to make missiles that could impact with a serious amount of kinetic energy. And you would lose lots of fighters to tiny bits of debris hitting at high relative speeds.
I also found it rather silly that, while there were only 12 models of human-appearing Cylon, there were at least three "sleepers" on board the Galactica. That might have made sense in the original, where the Galactica was the best ship in the fleet (and Adama was a revered figure). But one of the things I liked best about the new show was that the Galactica was an old ship, the last of the old Battlestars; and Adama was just some stubborn old guy who would be retiring soon. The reason the Galactica escaped the Cylon Trojan Horse was that Adama had successfully resisted all new computer upgrades; and he got away with that because it was an old ship that nobody really cared about. (Hmmm... how did the Pegasus escape the Trojan Horse? I don't remember.) Anyway, if the Galactica somehow had three, the whole fleet must have been shot through with sleepers. It makes less sense the more I think about it.
The other thing which both shows seem to have which is out-of-place is artificial gravity. But it's nearly impossible to make a TV series that doesn't have artificial gravity (Avatar had a brief scene at the beginning with zero-g, but that was a half-billion dollar movie).
I'm always willing to believe in artificial gravity and FTL for an SF show, and as you said, it's nearly impossible to not have it. Avatar not only had a lot of money going into it, it was computer-generated. To make actual actors appear to float, you need wires, and digital wire removal, and it's going to be a pain and cost a lot.
steveha
tidy up the 'I don't know the difference between a galaxy and a star system' bug in the original
It was explicitly said that the Firefly ships (like the Serenity) and the other ships we saw did not have FTL ability, and spent all their time in one star system. It was also said that that one system had dozens of planets and hundreds of moons. It was never said how people managed to travel to this one star system (I'm guessing sub-light travel with the passengers in suspended animation), and the only reason ever given for why people traveled here was "Earth got all used up".
I don't think the solar system in Firefly is very plausible. And I have issues with some other science (a box of concentrated food was shown as very valuable, valuable enough to be worth flying the Serenity out to a remote moon for a trade; surgery to re-attach a severed ear required highly special medical equipment; getting victim ships to fly through a giant hoop that killed everyone on board was extremely silly; it's hard to believe that Reavers could even fly ships, let alone keep their ships in good working order; etc.). But I felt it was not so bad that it jarred me out of suspension of disbelief.
steveha
I had never seen the term "smartbook" before. This article defines a "smartbook" as a netbook with a non-x86 processor (likely ARM).
I guess it's a portmanteau of "smart phone" and "netbook". Or maybe it means "smart enough finally to use something other than x86 for an ultra-portable device".
http://www.pcmag.com/article2/0,2817,2357758,00.asp
steveha
I suggest you get some books that make math more interesting.
For any software developer, the math in the "Aha!" books is appropriate. I was able to understand this math in high school, so I'm sure hard-core math nerds will sneer that this is too easy, but I stand by the claim that these books are worth your time.
Aha!: A Two Volume Collection on Google Books
Aha! Insight on Amazon
Aha! Gotcha! on Amazon
From my own personal experience, I have never needed calculus or differential equations or any of that advanced sort of math in my whole career. I'm now doing somewhat advanced DSP work, and even there I haven't needed advanced math. (I don't entirely understand how the FFT works; I just know how to use its properties to get the result I need. Other people wrote C versions of the FFT for me; I haven't needed to write it.)
The math that has been useful to me is basic logic stuff, to know how to write conditions for if statements and the like; O(n) estimations, to help you choose the best algorithm to solve your problem; and basic probability stuff, to help me understand how caches work and such. So the first class sounds better to me than the second.
I'll recommend one more book to you. It's a sort of encyclopedia of algorithms... it will expand your mind with possibilities (if you don't even know something exists, you won't be able to use it to solve your problems). It's engagingly written, with "war stories" that make it more lively.
The Algorithm Design Manual on Google Books
The Algorithm Design Manual on Amazon
The author's web page for the book
steveha
Disclaimer: I'm not any sort of expert on this stuff. Ignore me or laugh at me as you like.
I'm dubious about this. The scale of the thing is staggering, and it's hard to believe it will produce a better electrical output than if you spent a similar amount of money building a molten-salt solar thermal plant instead. Unlike molten salt solar thermal, this won't make electricity at night.
The one thing that makes this interesting is that it combines a giant greenhouse with the energy generation. If you can somehow make the greenhouse part very profitable (growing exotic fruit that is expensive to transport, or some such) then maybe you might have a payoff to match the expense. Maybe. But I'm dubious.
steveha
they actually had a video phone product in the 1980s, and then, nobody wanted it
I remember that product. It used a modem to put video over ordinary phone lines. That's right, less than 32 kbps for full motion video! It had a screen the size of a postage stamp and the video updated as smoothly as a slideshow where the slides were all coated in honey. (I never had one, but (a) I read about them, and (b) I know just how little bandwidth 32 kbps is for motion video.)
A better product for you to mention would have been H.261 video over ISDN. That actually had a nonzero number of users, but still didn't light the world on fire. DSL and the cable TV infrastructure were better ways to connect people to the Internet and ISDN became irrelevant and unloved.
We actually have the technology now to put practical video phones in every home. We just don't bother because it is not worth the cost. People do want Internet on their cell phones (to Google for restaurants, or movie show times, or whatever) and people do want WebEx and similar tools (sharing a PC desktop is way more valuable than showing video of someone talking, and with a USB webcam you can also do the talking head thing if you like).
Someday every home will have a video phone, if only because it will be so cheap to do it that it will be built in to every computer, and some computers will be wall-sized flat panel home theatre convergence devices. The main early adopters probably will be families with babies whose grandparents live far away.
steveha
The article submitter must be trolling. Decades ago there existed a one-off prototype, which was never widely deployed, that was hugely expensive. Now there exists an inexpensive learning gadget that might actually be in the hands of actual kids, and this is "moving backwards"?
Next up: is the phone industry moving backwards? At a world's fair, AT&T demonstrated a working two-way color video phone, yet I don't have a video phone in my house yet. Of course, millions of people have full-color Internet on their phones, and can do things like view a photo of their home taken from orbit. And millions of people have practical teleconferencing via WebEx et al. But never mind that. The phone company doesn't have video phones in every house; we're moving backwards!
steveha
Thank you for taking the time to write a nice long answer to my questions. I appreciate it.
I hope you are able to work out your Ardour issues. If not, you might want to look into running Cubase under VirtualBox on your Linux computer; I have been generally pleased by how seamless and bulletproof VirtualBox has been in my testing. (But I haven't tried anything like Cubase; if it does realtime USB, that would be far more serious than anything I have tried.)
steveha
Hard to reconcile your claim that fights would be over quickly with the reality that fights between trained martial artists (e.g., UFC fights, Olympic judo matches, boxing matches, Olympic TKD matches, etc...) the majority of the time go to decision.
These fights do not allow crippling blows. No eye gouging, no cheap shots below the belt (or else, if they are allowed, the men are wearing cups), definitely no neck snapping and I doubt full-on hits to the brain case are allowed. These guys aren't even allowed surprise attacks; they can't attack until the referee says it is time.
Unless you know of a martial arts fighting contest where the losers are typically hauled off to the hospital or the morgue, I will not accept that they perfectly model real full-on fights.
On the other hand, I'll grant you that martial artists doing full-contact kick each other pretty hard in various parts of the body without the fight being decided.
Also hard to reconcile your claim that a kick to the knee would end a fight with the reality that kicks to knees are legal in lots of martial arts competitions (MT matches, UFC fights, K1 fights) but almost never end fights.
Hmmm. And these people have unpadded, unprotected knees? I've constantly heard this meme that the knee is rather fragile. If you are correct, and martial artists are allowed to attack each others' knees full-force, then I must be mistaken on this point. (I am not any kind of expert on lethal hand-to-hand fighting and I don't claim to be one.) How often do the losers have to go to hospital for a messed-up knee?
steveha
Next up for Linux, media production software. What the fuck is up with Hydrogen and Ardour? Can't they get at least one real musician on their design staff?
I want to learn how to use this sort of software. I have read some positive reviews of Ardour, so I'm interested to hear what you don't like about it, and what you do like. What do you usually use, if not Ardour? And whatever you use, does it run okay on WINE?
And, have you looked at Ubuntu Studio?
http://ubuntustudio.org/
steveha
The fight scenes are epic, but in real life they'd be ... more epic.
I think that in real life, fights tend to be over a lot faster than in movies/TV. The first guy to land a really good hit gains an advantage, then presses the advantage, then wins. If I let you get in a good shot to my chest and I'm sucking air, you can then land several more attacks and lay me out. If you even land a good hit against my knee, I've probably lost the fight right there. This is all assuming a real fight and not just posturing in front of an audience.
Also, in movies/TV, people seem to be able to run flat out and fight with 100% intensity for minutes at a time. You especially see this in superhero movies, even where the hero has no actual powers (Daredevil, Batman, etc.)
P.S. My favorite fight scene in fiction is probably from the story "Gulf" by Robert A. Heinlein. The protagonist is a super-intelligent spy, and he is expecting some sort of trouble from the bad guys; since he is in a public place he figures their first move will be some kind of diversion. The bad guys light off some fireworks, and almost everyone around the spy turns to look in the direction of the bang; two people near the spy do not look and head straight for him. Without hesitating he attacks them, and lays them out instantly, and then walks away as if nothing had happened. I thought that was a lot cooler than a three-page fight scene would have been.
IMHO everything is wrong with C as first language. It gives you dreadful programming style and is not a right tool for application programming. You can mod me as a troll if you want but you've got to chose the best tool for the job and C is a tool for writing operating systems for fuck's sake. It isn't even a high level language.
Not so. C is the king of the third-generation languages. With respect to your points, in order:
C doesn't enforce any particular programming style, and all major applications are written in C or C++ (so I don't think you can claim it is "not a right tool" for that). C is a tool for writing operating systems or anything else, and I am going to claim that C is a slightly higher level language than Pascal; it can do anything you can do in Pascal, often with cleaner code. (For example, it is much cleaner to simply return a value, than to assign the return value to the function name and then use a series of if statements to make sure that nothing else happens until the end of the function body is reached.)
All those buffer overflow security holes happen because of both typical "clever hack" C programming style and choice of using a language for writing operating systems to write business applications.
Buffer overflows are caused by using a language that doesn't do run-time bounds checking for you, combined with a standard library that includes functions that don't do explicit bounds checking either. strcpy() is considered harmful, you need to use functions that know how long the buffer is and refuse to write past the end. I note that there is nothing in Pascal to make it better than C here, except for the fact that it is really impossible to write a useful standard library in Pascal. (Different length strings have different types so there is no way to write a general-purpose library function that can accept strings of different lengths! That's just crazy! It's so crazy that every major Pascal implementation offers some way to work around it. Whereas C got this, like so many other things, right; and there are plenty of standard C string functions.)
Pascal wasn't a very good teaching language for nothing - it forced you to write software in a very clean and readable way.
It did nothing of the sort. You can write wretched and awful code in Pascal just like anything else.
The closest I have seen to a language that forces you to write clean software is Python; it forces you to put colons at the end of if statements and function definitions, just for readability; it forces you to indent your code consistently, or it won't work. And yet you can write wretched and awful code in Python just like anything else.
A language cannot force good code. At best, it can support good code. C does this decently; Python does it better.
Pascal is pretty much irrelevant now, and C is still very relevant now, which should tell you something.
P.S. Getting back to 12-year-olds, I think Python would be the ideal first language. C would not be horrible; it would be better than C++ because there is much less to learn. Pascal isn't bad, but it is a dead-end irrelevant language, so I wouldn't bother with it.
steveha
When the OLPC was first announced, I was surprised that they were making things so hard on themselves. A clamshell, with keyboard, with color screen? Trying to hit a $100 price point? It seemed like that would be hard, and it was... hard enough that they didn't hit their target. The OLPC XO costs double the target price, it is glacially slow, and at least the one I bought has a totally unusable touch pad.
What I thought they should have done was to make something rather like a Handspring Visor, but bigger. A tablet with a touchscreen and a stylus. One piece, no hinges. It could have a flip cover, and the cover could have little pins that go into molded dimples on the tablet; the cover would protect the touchscreen when closed, but would not be actually part of the tablet and would be easy to replace. But a plastic cover isn't a strict necessity; a slip case of cloth or vinyl or whatever would serve just fine.
Color is great, and kids love it, but it is not necessary and hitting the price point is more important.
I remember when the Palm PDA first came out, I read about attachments you could get to hook up science probes (thermometers and pH probes) and how teachers were taking kids out to ponds or wetlands to measure things. No color screen there, not needed.
I have carried a Palm or Handspring PDA since the Palm first came out. I have used them for reading books (mostly fiction but some nonfiction). The core mission of the OLPC really is to serve as the textbooks for the children. No color screen needed.
So, let's imagine a tablet, with a touchscreen and a stylus. The stylus stores in a "silo" as with a Palm PDA. The screen is maybe 6 inches (15 cm). Like a Palm PDA, it has a few "hard keys" along with the screen, and in fact let's give it a game pad (direction pad) like a Nintendo DS, and standard 1/8" mini jacks for headphones and microphone. It has whatever CPU makes technical sense, likely an ARM (certainly not an x86). It even has a USB connector or two, making it possible to prop this thing up and plug in a keyboard (whatever you can buy cheapest from some Chinese manufacturer). It would be great if it had an SD card slot, but with the USB connector you don't actually need an SD card slot and it's more important that it be simple and cheap to produce. Design it brain dead simple: take it apart, and you have the mainboard, the screen, the battery pack, and the case. Could this be built for the sub-$100 price point? Yes. Now upgrade the screen to color, if you can hit the price point. Then maybe add WiFi, if you can hit the price point. Cameras are fun too, if you can hit the price point.
If you can save money by making the screen the exact size and shape as would be used in a hugely mass-produced portable DVD player, do that. If you have to, make a dual-screen design like a Nintendo DS that uses standard cell phone displays. As long as you can make a device that can be field-repaired and cheaply mass-produced, you can be flexible with the design. But it doesn't have to look like a laptop and it shouldn't have any hinges or other moving parts. Maybe it doesn't even have a DC power jack; maybe you charge it by plugging in to the USB port. Or maybe not; I'm not an expert on this stuff, have an expert advise you on what is most cost-effective.
Now mass produce the things and sell them to anyone who wants one. Get the production quantity up. But hit the frakking price point.
For bonus points, make thermometers, pH probes, volt/ohm probes, and such that plug in to the USB or the microphone jack, and mass produce those as well.
Actual devices like this, in actual school children's hands, would be far better than a more ambitious device that doesn't hit the price point and only gets deployed in a very few locations.
P.S. I'm glad the XO-2 seems to be dead. Dual full-color touch screens with a hinge? That is not a design for inexpensive mass production.
steveha
Alas, Mono is still a part of the default Gnome distribution, just so they can have a note taking applet
Oh, "just" so they can have a single applet? It couldn't possibly be because they think it is a generally useful way to develop applications, such as F-Stop and Banshee?
Mono may or may not be a good idea, but you are framing your argument in an intellectually dishonest way here. That note-taking applet ("Tomboy") may be the only thing in standard GNOME that needs Mono right now, but I'm pretty sure that there will be others.
Even worse, there are folks pushing Banshee as the default music player so there's another dependency on Mono.
See? Then it won't just be Tomboy, there will be other things using Mono.
I haven't tried C#, but a lot of people seem to like it. If having C# means I get more free software to play with, I'm in favor of that.
The major argument I have seen against Mono is "Microsoft is just waiting and they will assert patent claims!!" In that case, the only thing that they can do is force people to stop using C# and Mono. In which case, all the Mono apps will be pulled or re-written. And at that point, you would have what you seem to want: no more Mono in GNOME.
That is the worst-case scenario. And I don't see it as being bad enough to try to keep people from using Mono. If people want to use Mono to write free software, that's fine with me.
I'm curious: now that Java is becoming fully free, would you support re-writing Tomboy and F-Stop and the others in Java? That way, instead of being bloated and slow C# applications, they could be bloated and slow Java applications. Would you be happier?
In my day job, I write wicked fast C code (small memory footprint, too). When I write software on my own for fun, it tends to be Python, which is even slower than C#. Do you have a problem with Python too?
steveha
Please provide a link to the FSF claiming that the LGPL is "less free" than the GPL.
Are you trolling? They renamed LGPL from the "Library" GPL to the "Lesser" GPL, because they feel it is less free. It's baked right into the name that they feel it is less free.
But you asked for a link. Here you go:
http://www.gnu.org/licenses/gpl-faq.html#WhySomeGPLAndNotLGPL
Just in case that wasn't clear enough for you, let me rephrase it: according to this gnu.org link, the LGPL does not protect users' freedom as well as the GPL. It does not maximize freedom as well as GPL. In short, it is less free, according to gnu.org.
Remember that GNU and FSF are all about the users' freedom. Freedom of any developer to make proprietary software is not viewed as a good thing. A license like GPL that restricts the ability of developers to make proprietary software is viewed as more free.
On the other hand, fans of the BSD license argue that it is "more free" because anyone may do anything with the software. GNU and FSF reject this idea.
steveha
I format my external USB drives to ext3. Most of my machines are Linux anyway, and I can always plug the USB drive into my storage server and backup over Samba to any kind of drive supported by the storage server.
ext3 is pretty much stable and well understood. It just works. That's what I want for backup drives.
And my netbook has Ubuntu Linux on it, and ext3 performs well on the external USB drive there. I haven't tested NTFS over FUSE on the netbook, but I wonder about CPU overhead on the little Atom chip: it might be a little bit slow.
If you want a drive you can take over to your friend's house, and your friend just runs Windows or a Mac, then by all means NTFS.
steveha
Eh, making up new words seems like a granfalloonian thing to do.
But one good thing about your word, it sounds kind of fugnutish. I like that in a word.
steveha
I'll bet SSH tunneling will be illegal, and will be considered evidence on its face that you are Up To Something. I wonder what the penalty will be.
On the other hand, the Chinese government does like money, and lack of SSL would make it rather hard to move money around via the Internet.
Either way, this is going to be about as successful as Prohibition was in the USA.
steveha
Do you really purchase every VHS movie on DVD and then Blu-Ray? Do you purchase every LP on cassette tape, CD, SACD, DVD-Audio?
We really did re-purchase every VHS on DVD. We haven't been in a big hurry to re-buy as Blu-Ray... but I'll get my favorites, once I have a giant flat panel TV (which will be soon). For movies that are mostly about the dialog, or for old TV content that is innately standard-def, there is no reason to upgrade from the DVD. But I see the difference with high-def and I like it. (If you really want to drive home how much better high-def is at resolving details, look at the end credits for movies sometime. On DVD they are often unreadable, the text is just too small to resolve!)
I really did re-purchase every LP on DVD. I have not yet bought an SACD or DVD-Audio. I'm willing to buy multi-channel releases for artists I like, if I am convinced that the quality is there. (I'm leery of new releases; read the Wikipedia entry on "loudness war" for details.)
There might be a few I'm willing to upgrade, assuming the new version is remastered.
Yes, I agree. And it needs to be a useful remaster. I've noticed that some movies get released, and re-released, and re-released again, to milk the upgrade market. I don't buy re-releases very often, and if I do, you won't get me with a quick refresh.
Format shifting is nearly automatic at this point, you can digitize tapes and records
Eww, digitizing cassette tapes? Or worse, 8-tracks? Better to bin them and buy a CD, or even buy them as a digital download from an online music retailer. If you have an unscratched LP, that's more reasonable, especially if you have good audio gear. Or if you have an album that just isn't available on CD.
steveha