Those only apply if it's reasonable that someone could see/hear the statement and believe that it represented reality.
If I say that you are a "cock gobbler" it's unlikely people will think that you actually perform fellatio. However, if I say "I saw Zang sucking some guy's sick behind that gay bathhouse" it is believable and thus probably is defamation.
It's also (in all civilized countries - ie, not the UK or Australia) a defense if it's true. If I saw you doing that, I could say it, assuming I could prove to a judge that it is true.
Further, stating an opinion rarely is defamation. Saying "I think Valenti looks like a mobster" isn't defamation. It's part of that believability test. If I say "I think he sucks cock for money" it's clearly my opinion, which means people probably wouldn't accept it as truth.
Thus, the original poster stopped short of defamation. (Assuming he lives in a reasonable country.)
Yeah, Tolkien did say that he didn't edit out Bombadil despite him not really fitting, because he was a mystery that wasn't answered anywhere (none of the books or letters say what he was) and Tolkien thought that some questions were best left unanswered.
However, Bombadil still didn't really seem to fit the story well. But then, I think the books would have been better if they started with the leaving of the Shire (maybe two pages in) and flashed back to everything else. It took me a few tries to get past the dull stuff to the story. (I later learned that it made a bit more sense if read post-hobbit...)
The trolls are in the movie, if you mean the stone trolls from the Hobbit? Look up in the scene where they're camped in the forrest trying to heal Frodo - just before Arwen shows up.
And as for Bombadil... I'm glad he was cut. IMHO he was a hold-over from when Tolkien started the book as a sequel to the Hobbit, in the same young-adult style. (You know, "Biffer, Boffer, etc" and other silly things.)
About halfway through FotR it got "older" and darker. Supposedly (in Tolkien's letters) he talked about how he changed his mind partway through. And how Bombadil was based on a doll(?) that his children loved, so he thought he'd toss him into the book.
I learned this later, but it backed up my feel from after reading all three that Bombadil didn't really belong.
The solution to the complex laws that people need to hire lawyers to follow isn't to make ignorance a valid excuse. It's to simplify laws.
Part of high school (the required schooling) should be used to teach the law. Anything that can't then be understood by a graduate of this course with an average reading comprehension should either be revised until it can, or removed from the books. Ditto with the volume of law. If it can't be reduced to where the average person (with this mandatory training) could keep enough in their head at once to determine the legality of any action they contemplated, it should be simplified.
We've got a situation today where nobody, even a trained lawyer, can know all of the law. It's so complex that you pretty well can't avoid breaking some laws or opening yourself to liability in some way. This just means that you get selective enforcement. Everyone is guilty and the system punishes those they wish, everyone else who committed the same crime (even if only by technicality) goes free, through no fault or innocence of their own.
I'd say that this type of course (and tests based on it) should be part of the requirement for being a legal adult and being able to go to war, drink, and vote.
Moreover, to avoid abuse (setting the bar to high) there should be at least a 90% passing grade. If the schools can't manage this either the test is too hard (reduce the legal complexity) or better the schools.
An interesting side benefit of this is that if someone wants to study outside school they can challenge the tests. A precocious 12-year old can get themselves declared legally adult while a dull 35-year old perhaps could not. (It's not perfect, but makes at least as much sense as a hard age limit.)
I can't understand what Bombadil is supposed to add to the story...
The first part of LotR was written as more of a kid's book, a romp, like The Hobbit. Bombadil was based on a favorite doll (?) of Tolkien's children.
He later said in some of his letters that the first half (or so) of the first book didn't suit the tone he later decided on.
That's why so many parts like the party and selling BagEnd, Bombadil, and Brie seem silly and out of place.
IMHO it was a very good choice to decide to stick with a consistent feeling throughout and remove excess bits as an editor should have done long ago.
Really, Tolkien wasn't a great writer. His books are enduring because of the rich world he creates and the attention to detail in its design. But to stick to every last word he wrote doesn't make sense when in many cases he simply didn't get around to fixing it up.
I thought Sauruman's role was handled well. He intends to betray Sauron, as evidenced by the scene with the Uru-kai where it says it serves him, not Sauron which directly contradicts the orders Sauruman received from Sauron.
The things they could have left out imho were some of the effects like the worms and bugs crawling over the hobbits when they were hidden, and the rock walkway in Moria where they rocked back and forth.
This is only true on DVD writable media. If you buy actual blanks and press them in a $100k machine this isn't true.
This is beyond what any home user can do, but the Asian pirates have been doing it for years. They even forge the packaging (or at least the silk-screening on the disk.)
Would I go through a huge hassle to make my web browser and MP3 player work 20% faster, or even Quake3? No. But I don't wait on those.
Would I go through the same hassle for 20% more bandwidth? Hell yes. I often end up waiting for downloads because I download 700MB images. Would I pay much for 20% faster CD Burning? Perhaps, if it still took 70 minutes like on my old single speed. But I wouldn't pay for the extra minute saved now that I can burn a disk in five minutes.
So perhaps you wouldn't care much about big savings in downloading from Usenet. But do you download much from Usenet? If not, does your opinion matter? Why not ask the people who use it? I think they've strongly indicated their views by starting to use yENC.
This isn't the same as a company forcing people to use something. This is the people choosing for themselves to find a tool to fill an important niche. It may not be terribly standard, but it's open. And as long as its consistent, that'll enable it to become a standard.
You want an alternative? Become a hardware engineer. Or a landscape architect. Seriously, there's nothing saying you have to make a living in any certain field.
I just wanted to get that out of the way. I do think that programmers can make a living even in a world full of GPLed software. But I don't think we should take it as a given that we'll be supported in the profession we'd like. I know artists who wish they could make a living but instead do it as a hobby while they work other jobs.
Is there any reason why you couldn't write a closed-source game? Use only LGPL libraries, rewrite code, etc. If there's less competition then shouldn't you make more money when it's released?
As for books (and music), they're just finally entering the realm of "information" after hundreds of years. It's been out of the realm of practicality for individuals to copy books until recently so copyright law really only had to deal with the professionals (other publishing houses). Now all digital information is in the same spot at the same time and we're seeing how absurd laws that treat it as physical property really are.
Now, I'm not saying we should abandon copyrights. I'm just saying that they aren't based on reality in any way, they're just a polite fiction that happens to work. But it's not working as well now as ten years ago. So we're inventing draconian crap like the DMCA and the SSSCA (not CBDTPA or something) trying to push the genie back into the bottle.
When do we just say that enough is enough, that the world has changed, and we aren't going to try to protect obsolete business models?
This relates to a lot of things. Why keep passing laws doomed to failure (the DMCA and SSSCA) to protect a failed copyright law. Why expect that software will make more billionares?
Really, as long as your company turns any profit, and pays a decent wage to the staff, isn't it a success? I know many businesses (in various fields) that can't manage that properly. Why see it as a failure just because it can't go for generic software (like MS) and make more? Perhaps that's the tail end of a short-lived business model.
Or maybe not. But I don't see why your desire (or mine) to make more money at it is any more relevant than the RIAA's desire to make more by stopping music sharing.
I think there're ways for programmers to make money in a different economy, but I'm not willing to cripple a bunch of industry and pass insane laws to make it happen. If we could revise copyrights that'd be a huge step forward, but it'll never happen. Not in a world where political bribery is an accepted fact.
If you want a database for a file system they've been out for decades.
However, it's debatable if this is a good thing. Providing transactions is good, but why write an SQL interface for the whole file system? And then have to massage files into specific data types, or a whole bunch of BLOBs which you can't perform any complex operations on...
If MS does come out with this filesystem, they won't make it a default. It'd rock for some uses and suck for others. Much like NTFS.
Why would the government want to put an FBI surveillance team on someone? It had better be a good reason.
Why would they detail one agent to checking into your library borrowing, your shopping, your phone calls? It'd have to be a suspicion that you were worth it.
Why would the government pull your debit card purchase record and correlate it to "suspicious" profiles? Perhaps because you're in the same city as a suspected criminal with an odd profile.
Why would the government force you to identify yourself in all transactions, making a digital log of your every move and purchase? Because at a negligible cost they get information that *may* be useful.
As the cost goes down, the reasonable ammount of surveilance on someone goes up. At some point the cost is close enough to zero that they can put cameras with face recognition on every corner, monitor all purchases, record all phone calls and automatically transcribe them looking for keywords, etc.
And when they need to "think of the children" to stop "terrorists" who "look just like us" they might decide that perpetual surveillance, "for their own good" would benefit we the people.
I'm not paranoid enough now to think that I stand out enough for anyone to care about me. But if this information starts to be collected who knows what bad uses will be found for it. Hell, maybe I'll piss off a scientologist and be declared "fair game" and they'll get these records and use them against me.
I don't do much that is a "secret", but I'm sure someone could find something embarrasing or that if taken out of context looks bad, and use it to hurt my reputation.
So, why collect that information if it's so easy to abuse?
So don't exist by selling software. Exist by selling hardware that uses it, or by supporting software other people write, or writing custom software that is of much more value to the company that commissioned it than it would be to anyone else (and thus, would never be programmed without their sponsorship.)
Nobody whines that there's no market for ice now that everyone can afford a freezer. (There used to be a thriving market in selling large blocks of ice for homeowners to use basically as a refrigerator.) It could be that selling software has only temporarily been a means to make a ton of money. It wouldn't be the end of the world.
There isn't much precedent in the world for intangible goods. Even art used to take a master to forge, and if an identical copy was made people would still value the original more simply because of its status as an original.
Now we have software though, which can be copied essentially for free, and which has no special original that people want. Any copy is the same as any other. Why should we expect a market based around this to work like other markets?
Amazon could download Morpheus, install it, and go to their own site from referrer's pages. Any time the referrer ID doesn't match the page they came from they log another of Morpheus's IDs. Then they lock that account out.
If you're accusing me of not paying full attention to what you're saying, I suggest you look in a mirror.
I'm not trying to deny Microsoft the right to anything. I'm simply calling for them to be honest for once. If they say open source is bad, let them avoid using it. If they use it, maybe they should say that it's not a bad thing.
If they can't be honest about something that obvious it really makes you wonder what else they're lying about.
You conjecture that because some software is free, a manager will object to the idea of paying someone an hourly rate to make other software?
Carpentry is an easy at-home task and almost everyone has done some. However, carpenters seem to make a living.
It's a market economy. If they can find someone as qualified as I am, who will work cheaper coding annoying doo-dads for their database, they're free to hire them. It's happened before. Sometimes I've been called back by the sheepish client to fix the mess they made.
If I have a job though, I want it to be providing a real service to a customer with freedom of choice. I'm sorry you don't have confidence in your job options in a new economy.
I agree. You don't have to be socialist to want big government.
I was trying to point out some hypocrisy on RR's arguments though. He lambastes open source as being communist, and says there (basically) "needs to be a law!" I'm just pointing out that for someone who thinks might makes right and has a generally Randian point of view, he sure seems to want government protection when something comes along that threatens him or his comfortable world.
I know he can be greedy and want a strictly regulated economy (for his benefit). But he shouldn't throw around terms like "socialist" and "communist" as slurs unless he's against a regulated market.
Myself, I'm quite socialist. Both because I think it's "right" that people don't starve, but also because I'd rather pay slightly higher taxes to ensure that the poor aren't so poor or downtrodden they feel the need to overthrow society. However, I get a kick out of tweaking psuedo-libertarians who want a free market (for them to abuse) but a set of very strict rules that force people to put up with it, and not pull similar tricks on them.
I see your point now. I think it's infeasible, but at least it makes sense.
Yeah, companies in general can be asses. I'm in the middle of RMAing my IBM HD right now. They won't ship me a different model until this fails twice, and then it'll only be a 120GXP which isn't great either.
I think there's a way to change the MAC address on most cards, but you have to have a working card. Google could help more than I could.
I've got my HDs mounted in 5.25 slots right behind a fan. I've got two rear fans, a front bottom fan, and a huge CPU fan. Mine runs the same temp with the case on or off. (Which I figure is good. I can close it up and get it quiet without it heating up.)
My friend has a front and rear fan, a top exhaust fan, and a 120mm side fan over another huge CPU fan. His runs cooler with the case on. But he had to go to extremes to get it.
The mess of cabling in most computers means that the fans don't do a lot of good. That's one reason I'm looking forward to serial ATA; no more ribbon cables.
So? One company decided they didn't like the GPL, mostly based on RMS. Oh well.
For a single counter-example, http://www.merilus.com/ is a company making a linux-based encrypted VPN router/packet filter on a card based on the Crusoe chip. They release all their software.
Companies may find it hard to make money when giving software away for free, but they can always leverage their trademark and sell it retail. Nobody can forge it or they can sue. Sure, you can get it for free, but the average consumer isn't going to know that. If you include a nice installer on the CD and don't on the downloaded version you've got the attention of most regular users.
But... Let's play pretend for a minute. Let's imagine that there isn't a market for selling GPLed software.
Oh wow! The twisting worldview. There also isn't a market for selling ice to eskimos. Or really, to anyone with a fridge/freezer. But you don't hear the president of an ice-cube company complaining that putting the power of ice making in the hands of the common man is destroying the economy and rendering millions unemployable.
Things change. There's no reason the software industry has to be this big. I might even find myself out of a job, but if I'm displaced by someone who can do a better job for less money, I'm willing to go. I don't want a job that exists only because of a government mandate.
It's amazing that for a professed capitalist you have all these facist, big-government leanings. I thought the idea was that the market would sort it out. If mega-corporations can't manage because of competition from hobbyists then they aren't providing anything of value. (See Artificial Scarcity.)
This isn't an unbreakable scheme. It's basically a dongle, but built into the CPU.
You let the software call the built-in decrypt function and then once it's decrypted itself you suspend operation and write it out to file. Fix up the loader and you've got a working application.
So never have it decrypt the whole app you say. So you have to save it in smaller chunks.
So it checks file integrity. So you edit that out.
You can make it a pain, but never more than that. Plus this requires assembly programming which few people can do these days. (Seriously, ask any group of programmers. Maybe 20% have done it, and 5% are capable of it. Perhaps.5% are expert enough to code self modifying code that won't thrash the cache, yet doesn't keep much of itself in memory. It requires interleaving the decryption code, a few bytes at a time, in the execution code. And either decrypting the code like this, or storing decryption code for the next segment in all code segments you write.) Stuff much more complex than this has been hacked.
I remember an old game that used diagonal tracks on the Apple 2. You read a sector, after having given a head movement command. If your code isn't cycle for cycle identical you'll start to get dropped bits. But that's part of the trick, their code took a little too long in a storage loop and eventually got a bit out of sync, which was intentional. Ugh. Nasty stuff.
But those protection methods took a lot of programmer time and meant that they had to write the disks themselves in modified drives, instead of paying a duplication company. It's just not feasible.
It's almost never feasible, especially these days, to muck with hardware protection schemes. (Look at how useful it is with CDs. There are point-and-click cracker programs for all the common laser-burn protections.)
The thing to do is generate a number that passes about twenty tests. Check two-three (all the similar tests) at code entry. Check other ones later and trigger delayed problems.
For instance (a very basic instance), generate keys that, are evenly divisble by 7, 13, and 23, and 29.
At the time of key entry check if the key is divisible by 7 and 29. If it is, pass it. Later on, when the user does something like "cut", check if it's divisible by 13. If not, set a flag to do something later, like crash, scramble output, etc.
When the next version comes out, check if the keys (for upgrading) are divisible by 23. If they are, it's a valid key. If not, it was a key for the old version.
With some checks causing later delayed issues you make the key generation a nightmare simple because most crackers won't spend as much time in the program as a user and won't notice subtle errors.
By saving a few checks for later versions (If you allow upgrades withing 2.x for example) you force people to use a new crack for new versions.
This way there are tons of cracks floating around for your program, most for old versions which aren't on your website anymore. And half of them miss something and the program doesn't work well.
Now, picking a number composed of certain primes isn't terribly secure, so use your own method. But really, the two points above (1: checks not just at startup, 2: new checks with each version) are more important than the specific key strength. (Really though, no reason to skimp.)
There are other important issues too. Ease of use is an important one. Instead of Base64, I'd recommend using alpha/num in pairs, storing 9 bits per two characters. Avoid 0/O and 1/I/l issues.
Print the key with a few check digits. Don't use simple addition (or digital root, or any other commutative idea) for it, otherwise you'll miss transposed digits. (One idea is multiple the first character by 3, the next by 5, etc. Then take the lowest five bits, compress a little (get rid of 0/1, etc) and use it as a check character.
This measure will save you a *ton* of support email. I can't count the number of issues that relate to keys being case sensitive, to '1's being 'L's, and the like.
(Horror story: My company ordered Perfect Keyboard licenses. They were to be tied to the workstations, not users, so we didn't give a name. Now, PK requires you enter ALL your info exactly or it won't work. The issue is that the name field must be left blank, though a blank name field is usually a prompt to enter your name. I think everyone who used it ran into this issue a few times.)
Adding this security to the key does make it longer. From using 6 to 4.8 bits, you go from 22 chars to 26. Not a bad tradeoff. Then add a three or four check characters. Still below thirty.
--
As for the later checks, you could introduce some subtle "bugs" into the code. If you were keeping track of the number of characters in an editor (for display) you could use an intentionally awkward piece of code. In some circumstances (deleting a character to from from 10,000 to 9,999 perhaps) it'd drop through a loop and perform a key check before updating the display properly. If the check fails then the key settings are corrupted when quitting, or the edit menu stops working (Make it look accidental, set the high bit on a few characters, remap "copy" to SaveAs, etc...)
You need to make sure that you inline your key checks, otherwise the cracker only need patch the main routine to always return true.
It's funny you mention "bulletproof" in your post, because BulletProofFTP uses a complex scheme like this and there are a ton of non-working cracks for it.
It's funny that you complain about the number of projects on Freshmeat. That's not where you're supposed to go to find a finished project. It hosts development projects. Some of them are finished, but the majority, not suprisingly, are under development. It's like blaming GeoCities for the bad web pages they host.
Yes, and who do you suppose will fund this if all software is free software?
I presume the same people as now. The developers. Interested users.
Why? All this excess baggage that Linux carries around--consumers don't need this.
Why get rid of it? All the formats are open and documented. All the shells except Bash are deprecated, etc. The rest are just there for people who grew up with old systems.
And proprietary vendors tend to either make their formats work,
Nobody is claiming that proprietary formats don't work. People are claiming proprietary formats don't work with anything else.
If WordPerfect can't import MS Word files, WordPerfect gets blamed, despite the fact that MS made to convoluted and undocumented format.
Each shell is practically a whole new incompatible platform that software must work with.
When a script runs it specifies the shell it wants. Programs that run under a shell (such as an installer script) pick one and support it. They then work fine when run under any shell.
Have you used unix? For more than a day or two perhaps?
And plumbers get the same treatment after they fix a nasty leak.
Perhaps after they install new plumbing perhaps. But what's wrong with that? I don't see why programming should be held to be anything other than a specialized trade. I'm not ashamed at the idea that I provide a valuable product/service and get paid for it.
There were paid programmers, I'm sure. But, I'm willing to bet that they were paid very modest sums. I believe they were viewed more as "secretarial" positions, rather than professions such as Doctors or Lawyers. [snip] Are they responsible, in part, for raising programmer's salary?
Chuckle. An older friend of mine bitches about the low wages these days. He made $150k+ per year, once over $220k, 1970s dollars, for programming back on old IBM mainframes. Today he makes $60k or something. Not a bad wage by any means, but a fraction of what he made before.
Why would anyone want to destroy this perceived value by making it appear as if software is easy to create and doesn't really matter? When mom-and-pop understand fully that compatible quality software can be made with programmer's free time, for free, what is stopping them from demanding this from Microsoft, Intuit, etc.?
I'm sure you also argue against including compilers with an OS, or making them freely available. I mean, if people see that they can write programs they're going to write their own and never use Microsoft's right?
I am very glad that I can program. I automate many tasks that take my friends hours. Even the ones who can use 3rd-party macro programs can't compete with a perl script I can hack together. If I want to see how a fractal changes if I modify the formula, I can. They have to ask me or hope that someone on the net had the same curiosity as them.
This is a gift that I want to share with everyone. They may never use it, but they'll be able to. Linux will never be locked down, but I can easily imagine a day when to "combat viruses" all code run on a Windows computer must be cryptographically signed. When users are crippled because a company wants to potentially squeeze more money from them in the future.
The reason your salary has not been cut is because software, for the most part, still has value.
One of the contracts I've taken was an ordering/tracking system for a company that made circuit boards. Previously an order (of anything complex) could take up to 30-40 minutes, with a few binders full of pricing charts, to price. The calculation screen(s) I made had space for 80+ variables, accessing hundreds of tables for pricing data. But you could give the customer a price as soon as you were done entering the data. It saved further time by passing the order to work stations at the various steps in the process. From ordering to a sealed computer hung over the drill press in the machine room, to accounting and shipping at the end.
The project saved an average of 15 minutes per order. It also meant that they passed notes and all design docs (the cirsuit diagrams) along as files, instead of taking a folder from station to station.
They mentioned a years or so after the project was finished that they hadn't lost an order since it was put in, and that they loved being able to pull up a spreadsheet that told them how many dollars worth of product were due to be done at any time, where work was backed up, etc. They were working to integrate it into a bonus system for the workers, as well as to let them know ahead of time about potential work shortages.
At my estimate (just of time saved initially) it let them do three times as much work per customer service rep. It eliminated one job (a guy who was moved into customer service instead of fired) of lugging paper around, keeping files straight, etc. It removed the requirement for a room of files, plus the printing costs, storage for old files, probably 5% of machinist time... Hell, even one of the accountants said he was happier because he didn't get stinky files with fiberglass shavings and etching fluid stains on them.
And you don't think that has value? Probably $400 / day, or more. That was the best $20k they ever spent.
But there's no way an off-the-shelf package would do what they want. Too much custom stuff. I've seen some systems for designing a pricing layout by drag and drop... fairly nice, but nowhere near the level of complexity something like this takes. And I doubt anyone will bother making it that good, for the.1% of customers who want it.
I've seen places where almost any business could benefit from custom software. Let me talk to the employees and identify their bitches and I can find even more. 10 minutes a day of hassle doesn't seem like much, but if you're paying $20/hour that's $3.33. Multiply by five employees, times 250 (working days per year) and you're at $4k. Figure in saved training time, and it's starting to look pretty sweet. (That was solved by a few simple batch files to open the right applications, perform incremental backups, etc.)
This will always have value.
[...] but it will be the FSF that brings the entire software industry to its knees.
Once again. Only if the work of paid professionals can't match the hobby work of a bunch of geeks. (Which you seem to think is really really crappy...)
But if they can't match the free software, what right to they have to bitch? They just want corporate welfare. "Rise up Joe Sixpack, cast down the shackles of free software made by the people and send half a month's wage to the world's richest man to sustain his lifestyle." It's not really concerning anyone except the rich who've sunk their mutual funds into MS stock.
Upping the voltage by a 10th or 20th of a volt, or overclocking by a bit, isn't going to kill anything. In fact, many motherboards do just that, they run various components with a bit more voltage to help stability. Sure, heat increases, but as long as it's withing spec you're fine. (Yes, some motherboards overclock by a bit. Check how fast a P4 CPU is running with CPUID, chances are it's not exactly at the rated speed.)
Furthermore, CPUs don't need to be matched at all for doing MP with Athlons. At least one reviewer tested with mismatched CPUs when the chipset first came out and it worked just fine. (Contrast to the Intel SMP which sometimes needs chips of the same stepping.)
Knuth seems more like the type who just doesn't enjoy tweaking people like Feynman did, so he excludes topics he knows they'd get into a thither about. Especially since they're really off-topic to the course.
Feynman seemed like my type of guy. Some people just need to be poked a bit when they're too serious about "The One Truth", whatever they feel it is.
Slashdot Mirror
Those only apply if it's reasonable that someone could see/hear the statement and believe that it represented reality.
If I say that you are a "cock gobbler" it's unlikely people will think that you actually perform fellatio. However, if I say "I saw Zang sucking some guy's sick behind that gay bathhouse" it is believable and thus probably is defamation.
It's also (in all civilized countries - ie, not the UK or Australia) a defense if it's true. If I saw you doing that, I could say it, assuming I could prove to a judge that it is true.
Further, stating an opinion rarely is defamation. Saying "I think Valenti looks like a mobster" isn't defamation. It's part of that believability test. If I say "I think he sucks cock for money" it's clearly my opinion, which means people probably wouldn't accept it as truth.
Thus, the original poster stopped short of defamation. (Assuming he lives in a reasonable country.)
Yeah, Tolkien did say that he didn't edit out Bombadil despite him not really fitting, because he was a mystery that wasn't answered anywhere (none of the books or letters say what he was) and Tolkien thought that some questions were best left unanswered.
However, Bombadil still didn't really seem to fit the story well. But then, I think the books would have been better if they started with the leaving of the Shire (maybe two pages in) and flashed back to everything else. It took me a few tries to get past the dull stuff to the story. (I later learned that it made a bit more sense if read post-hobbit...)
Here's a post which mentions the details of the New Footage.
... I'm glad he was cut. IMHO he was a hold-over from when Tolkien started the book as a sequel to the Hobbit, in the same young-adult style. (You know, "Biffer, Boffer, etc" and other silly things.)
Looks pretty good. There's some Gimli/Galadriel action missing still. (That sounds sick)
The trolls are in the movie, if you mean the stone trolls from the Hobbit? Look up in the scene where they're camped in the forrest trying to heal Frodo - just before Arwen shows up.
And as for Bombadil
About halfway through FotR it got "older" and darker. Supposedly (in Tolkien's letters) he talked about how he changed his mind partway through. And how Bombadil was based on a doll(?) that his children loved, so he thought he'd toss him into the book.
I learned this later, but it backed up my feel from after reading all three that Bombadil didn't really belong.
The solution to the complex laws that people need to hire lawyers to follow isn't to make ignorance a valid excuse. It's to simplify laws.
Part of high school (the required schooling) should be used to teach the law. Anything that can't then be understood by a graduate of this course with an average reading comprehension should either be revised until it can, or removed from the books. Ditto with the volume of law. If it can't be reduced to where the average person (with this mandatory training) could keep enough in their head at once to determine the legality of any action they contemplated, it should be simplified.
We've got a situation today where nobody, even a trained lawyer, can know all of the law. It's so complex that you pretty well can't avoid breaking some laws or opening yourself to liability in some way. This just means that you get selective enforcement. Everyone is guilty and the system punishes those they wish, everyone else who committed the same crime (even if only by technicality) goes free, through no fault or innocence of their own.
I'd say that this type of course (and tests based on it) should be part of the requirement for being a legal adult and being able to go to war, drink, and vote.
Moreover, to avoid abuse (setting the bar to high) there should be at least a 90% passing grade. If the schools can't manage this either the test is too hard (reduce the legal complexity) or better the schools.
An interesting side benefit of this is that if someone wants to study outside school they can challenge the tests. A precocious 12-year old can get themselves declared legally adult while a dull 35-year old perhaps could not. (It's not perfect, but makes at least as much sense as a hard age limit.)
I can't understand what Bombadil is supposed to add to the story...
The first part of LotR was written as more of a kid's book, a romp, like The Hobbit. Bombadil was based on a favorite doll (?) of Tolkien's children.
He later said in some of his letters that the first half (or so) of the first book didn't suit the tone he later decided on.
That's why so many parts like the party and selling BagEnd, Bombadil, and Brie seem silly and out of place.
IMHO it was a very good choice to decide to stick with a consistent feeling throughout and remove excess bits as an editor should have done long ago.
Really, Tolkien wasn't a great writer. His books are enduring because of the rich world he creates and the attention to detail in its design. But to stick to every last word he wrote doesn't make sense when in many cases he simply didn't get around to fixing it up.
I thought Sauruman's role was handled well. He intends to betray Sauron, as evidenced by the scene with the Uru-kai where it says it serves him, not Sauron which directly contradicts the orders Sauruman received from Sauron.
The things they could have left out imho were some of the effects like the worms and bugs crawling over the hobbits when they were hidden, and the rock walkway in Moria where they rocked back and forth.
This is only true on DVD writable media. If you buy actual blanks and press them in a $100k machine this isn't true.
This is beyond what any home user can do, but the Asian pirates have been doing it for years. They even forge the packaging (or at least the silk-screening on the disk.)
Would I go through a huge hassle to make my web browser and MP3 player work 20% faster, or even Quake3? No. But I don't wait on those.
Would I go through the same hassle for 20% more bandwidth? Hell yes. I often end up waiting for downloads because I download 700MB images. Would I pay much for 20% faster CD Burning? Perhaps, if it still took 70 minutes like on my old single speed. But I wouldn't pay for the extra minute saved now that I can burn a disk in five minutes.
So perhaps you wouldn't care much about big savings in downloading from Usenet. But do you download much from Usenet? If not, does your opinion matter? Why not ask the people who use it? I think they've strongly indicated their views by starting to use yENC.
This isn't the same as a company forcing people to use something. This is the people choosing for themselves to find a tool to fill an important niche. It may not be terribly standard, but it's open. And as long as its consistent, that'll enable it to become a standard.
You want an alternative? Become a hardware engineer. Or a landscape architect. Seriously, there's nothing saying you have to make a living in any certain field.
I just wanted to get that out of the way. I do think that programmers can make a living even in a world full of GPLed software. But I don't think we should take it as a given that we'll be supported in the profession we'd like. I know artists who wish they could make a living but instead do it as a hobby while they work other jobs.
Is there any reason why you couldn't write a closed-source game? Use only LGPL libraries, rewrite code, etc. If there's less competition then shouldn't you make more money when it's released?
As for books (and music), they're just finally entering the realm of "information" after hundreds of years. It's been out of the realm of practicality for individuals to copy books until recently so copyright law really only had to deal with the professionals (other publishing houses). Now all digital information is in the same spot at the same time and we're seeing how absurd laws that treat it as physical property really are.
Now, I'm not saying we should abandon copyrights. I'm just saying that they aren't based on reality in any way, they're just a polite fiction that happens to work. But it's not working as well now as ten years ago. So we're inventing draconian crap like the DMCA and the SSSCA (not CBDTPA or something) trying to push the genie back into the bottle.
When do we just say that enough is enough, that the world has changed, and we aren't going to try to protect obsolete business models?
This relates to a lot of things. Why keep passing laws doomed to failure (the DMCA and SSSCA) to protect a failed copyright law. Why expect that software will make more billionares?
Really, as long as your company turns any profit, and pays a decent wage to the staff, isn't it a success? I know many businesses (in various fields) that can't manage that properly. Why see it as a failure just because it can't go for generic software (like MS) and make more? Perhaps that's the tail end of a short-lived business model.
Or maybe not. But I don't see why your desire (or mine) to make more money at it is any more relevant than the RIAA's desire to make more by stopping music sharing.
I think there're ways for programmers to make money in a different economy, but I'm not willing to cripple a bunch of industry and pass insane laws to make it happen. If we could revise copyrights that'd be a huge step forward, but it'll never happen. Not in a world where political bribery is an accepted fact.
If you want a database for a file system they've been out for decades.
However, it's debatable if this is a good thing. Providing transactions is good, but why write an SQL interface for the whole file system? And then have to massage files into specific data types, or a whole bunch of BLOBs which you can't perform any complex operations on...
If MS does come out with this filesystem, they won't make it a default. It'd rock for some uses and suck for others. Much like NTFS.
It's all about ease.
Why would the government want to put an FBI surveillance team on someone? It had better be a good reason.
Why would they detail one agent to checking into your library borrowing, your shopping, your phone calls? It'd have to be a suspicion that you were worth it.
Why would the government pull your debit card purchase record and correlate it to "suspicious" profiles? Perhaps because you're in the same city as a suspected criminal with an odd profile.
Why would the government force you to identify yourself in all transactions, making a digital log of your every move and purchase? Because at a negligible cost they get information that *may* be useful.
As the cost goes down, the reasonable ammount of surveilance on someone goes up. At some point the cost is close enough to zero that they can put cameras with face recognition on every corner, monitor all purchases, record all phone calls and automatically transcribe them looking for keywords, etc.
And when they need to "think of the children" to stop "terrorists" who "look just like us" they might decide that perpetual surveillance, "for their own good" would benefit we the people.
I'm not paranoid enough now to think that I stand out enough for anyone to care about me. But if this information starts to be collected who knows what bad uses will be found for it. Hell, maybe I'll piss off a scientologist and be declared "fair game" and they'll get these records and use them against me.
I don't do much that is a "secret", but I'm sure someone could find something embarrasing or that if taken out of context looks bad, and use it to hurt my reputation.
So, why collect that information if it's so easy to abuse?
So don't exist by selling software. Exist by selling hardware that uses it, or by supporting software other people write, or writing custom software that is of much more value to the company that commissioned it than it would be to anyone else (and thus, would never be programmed without their sponsorship.)
Nobody whines that there's no market for ice now that everyone can afford a freezer. (There used to be a thriving market in selling large blocks of ice for homeowners to use basically as a refrigerator.) It could be that selling software has only temporarily been a means to make a ton of money. It wouldn't be the end of the world.
There isn't much precedent in the world for intangible goods. Even art used to take a master to forge, and if an identical copy was made people would still value the original more simply because of its status as an original.
Now we have software though, which can be copied essentially for free, and which has no special original that people want. Any copy is the same as any other. Why should we expect a market based around this to work like other markets?
Amazon could download Morpheus, install it, and go to their own site from referrer's pages. Any time the referrer ID doesn't match the page they came from they log another of Morpheus's IDs. Then they lock that account out.
If you're accusing me of not paying full attention to what you're saying, I suggest you look in a mirror.
I'm not trying to deny Microsoft the right to anything. I'm simply calling for them to be honest for once. If they say open source is bad, let them avoid using it. If they use it, maybe they should say that it's not a bad thing.
If they can't be honest about something that obvious it really makes you wonder what else they're lying about.
You conjecture that because some software is free, a manager will object to the idea of paying someone an hourly rate to make other software?
Carpentry is an easy at-home task and almost everyone has done some. However, carpenters seem to make a living.
It's a market economy. If they can find someone as qualified as I am, who will work cheaper coding annoying doo-dads for their database, they're free to hire them. It's happened before. Sometimes I've been called back by the sheepish client to fix the mess they made.
If I have a job though, I want it to be providing a real service to a customer with freedom of choice. I'm sorry you don't have confidence in your job options in a new economy.
I agree. You don't have to be socialist to want big government.
I was trying to point out some hypocrisy on RR's arguments though. He lambastes open source as being communist, and says there (basically) "needs to be a law!" I'm just pointing out that for someone who thinks might makes right and has a generally Randian point of view, he sure seems to want government protection when something comes along that threatens him or his comfortable world.
I know he can be greedy and want a strictly regulated economy (for his benefit). But he shouldn't throw around terms like "socialist" and "communist" as slurs unless he's against a regulated market.
Myself, I'm quite socialist. Both because I think it's "right" that people don't starve, but also because I'd rather pay slightly higher taxes to ensure that the poor aren't so poor or downtrodden they feel the need to overthrow society. However, I get a kick out of tweaking psuedo-libertarians who want a free market (for them to abuse) but a set of very strict rules that force people to put up with it, and not pull similar tricks on them.
I see your point now. I think it's infeasible, but at least it makes sense.
Yeah, companies in general can be asses. I'm in the middle of RMAing my IBM HD right now. They won't ship me a different model until this fails twice, and then it'll only be a 120GXP which isn't great either.
I think there's a way to change the MAC address on most cards, but you have to have a working card. Google could help more than I could.
Depends on the layout.
I've got my HDs mounted in 5.25 slots right behind a fan. I've got two rear fans, a front bottom fan, and a huge CPU fan. Mine runs the same temp with the case on or off. (Which I figure is good. I can close it up and get it quiet without it heating up.)
My friend has a front and rear fan, a top exhaust fan, and a 120mm side fan over another huge CPU fan. His runs cooler with the case on. But he had to go to extremes to get it.
The mess of cabling in most computers means that the fans don't do a lot of good. That's one reason I'm looking forward to serial ATA; no more ribbon cables.
So? One company decided they didn't like the GPL, mostly based on RMS. Oh well.
For a single counter-example, http://www.merilus.com/ is a company making a linux-based encrypted VPN router/packet filter on a card based on the Crusoe chip. They release all their software.
Companies may find it hard to make money when giving software away for free, but they can always leverage their trademark and sell it retail. Nobody can forge it or they can sue. Sure, you can get it for free, but the average consumer isn't going to know that. If you include a nice installer on the CD and don't on the downloaded version you've got the attention of most regular users.
But... Let's play pretend for a minute. Let's imagine that there isn't a market for selling GPLed software.
Oh wow! The twisting worldview. There also isn't a market for selling ice to eskimos. Or really, to anyone with a fridge/freezer. But you don't hear the president of an ice-cube company complaining that putting the power of ice making in the hands of the common man is destroying the economy and rendering millions unemployable.
Things change. There's no reason the software industry has to be this big. I might even find myself out of a job, but if I'm displaced by someone who can do a better job for less money, I'm willing to go. I don't want a job that exists only because of a government mandate.
It's amazing that for a professed capitalist you have all these facist, big-government leanings. I thought the idea was that the market would sort it out. If mega-corporations can't manage because of competition from hobbyists then they aren't providing anything of value. (See Artificial Scarcity.)
This isn't an unbreakable scheme. It's basically a dongle, but built into the CPU.
.5% are expert enough to code self modifying code that won't thrash the cache, yet doesn't keep much of itself in memory. It requires interleaving the decryption code, a few bytes at a time, in the execution code. And either decrypting the code like this, or storing decryption code for the next segment in all code segments you write.) Stuff much more complex than this has been hacked.
You let the software call the built-in decrypt function and then once it's decrypted itself you suspend operation and write it out to file. Fix up the loader and you've got a working application.
So never have it decrypt the whole app you say. So you have to save it in smaller chunks.
So it checks file integrity. So you edit that out.
You can make it a pain, but never more than that. Plus this requires assembly programming which few people can do these days. (Seriously, ask any group of programmers. Maybe 20% have done it, and 5% are capable of it. Perhaps
I remember an old game that used diagonal tracks on the Apple 2. You read a sector, after having given a head movement command. If your code isn't cycle for cycle identical you'll start to get dropped bits. But that's part of the trick, their code took a little too long in a storage loop and eventually got a bit out of sync, which was intentional. Ugh. Nasty stuff.
But those protection methods took a lot of programmer time and meant that they had to write the disks themselves in modified drives, instead of paying a duplication company. It's just not feasible.
It's almost never feasible, especially these days, to muck with hardware protection schemes. (Look at how useful it is with CDs. There are point-and-click cracker programs for all the common laser-burn protections.)
The thing to do is generate a number that passes about twenty tests. Check two-three (all the similar tests) at code entry. Check other ones later and trigger delayed problems.
For instance (a very basic instance), generate keys that, are evenly divisble by 7, 13, and 23, and 29.
At the time of key entry check if the key is divisible by 7 and 29. If it is, pass it. Later on, when the user does something like "cut", check if it's divisible by 13. If not, set a flag to do something later, like crash, scramble output, etc.
When the next version comes out, check if the keys (for upgrading) are divisible by 23. If they are, it's a valid key. If not, it was a key for the old version.
With some checks causing later delayed issues you make the key generation a nightmare simple because most crackers won't spend as much time in the program as a user and won't notice subtle errors.
By saving a few checks for later versions (If you allow upgrades withing 2.x for example) you force people to use a new crack for new versions.
This way there are tons of cracks floating around for your program, most for old versions which aren't on your website anymore. And half of them miss something and the program doesn't work well.
Now, picking a number composed of certain primes isn't terribly secure, so use your own method. But really, the two points above (1: checks not just at startup, 2: new checks with each version) are more important than the specific key strength. (Really though, no reason to skimp.)
There are other important issues too. Ease of use is an important one. Instead of Base64, I'd recommend using alpha/num in pairs, storing 9 bits per two characters. Avoid 0/O and 1/I/l issues.
Print the key with a few check digits. Don't use simple addition (or digital root, or any other commutative idea) for it, otherwise you'll miss transposed digits. (One idea is multiple the first character by 3, the next by 5, etc. Then take the lowest five bits, compress a little (get rid of 0/1, etc) and use it as a check character.
This measure will save you a *ton* of support email. I can't count the number of issues that relate to keys being case sensitive, to '1's being 'L's, and the like.
(Horror story: My company ordered Perfect Keyboard licenses. They were to be tied to the workstations, not users, so we didn't give a name. Now, PK requires you enter ALL your info exactly or it won't work. The issue is that the name field must be left blank, though a blank name field is usually a prompt to enter your name. I think everyone who used it ran into this issue a few times.)
Adding this security to the key does make it longer. From using 6 to 4.8 bits, you go from 22 chars to 26. Not a bad tradeoff. Then add a three or four check characters. Still below thirty.
--
As for the later checks, you could introduce some subtle "bugs" into the code. If you were keeping track of the number of characters in an editor (for display) you could use an intentionally awkward piece of code. In some circumstances (deleting a character to from from 10,000 to 9,999 perhaps) it'd drop through a loop and perform a key check before updating the display properly. If the check fails then the key settings are corrupted when quitting, or the edit menu stops working (Make it look accidental, set the high bit on a few characters, remap "copy" to SaveAs, etc...)
You need to make sure that you inline your key checks, otherwise the cracker only need patch the main routine to always return true.
It's funny you mention "bulletproof" in your post, because BulletProofFTP uses a complex scheme like this and there are a ton of non-working cracks for it.
It's funny that you complain about the number of projects on Freshmeat. That's not where you're supposed to go to find a finished project. It hosts development projects. Some of them are finished, but the majority, not suprisingly, are under development. It's like blaming GeoCities for the bad web pages they host.
.1% of customers who want it.
Yes, and who do you suppose will fund this if all software is free software?
I presume the same people as now. The developers. Interested users.
Why? All this excess baggage that Linux carries around--consumers don't need this.
Why get rid of it? All the formats are open and documented. All the shells except Bash are deprecated, etc. The rest are just there for people who grew up with old systems.
And proprietary vendors tend to either make their formats work,
Nobody is claiming that proprietary formats don't work. People are claiming proprietary formats don't work with anything else.
If WordPerfect can't import MS Word files, WordPerfect gets blamed, despite the fact that MS made to convoluted and undocumented format.
Each shell is practically a whole new incompatible platform that software must work with.
When a script runs it specifies the shell it wants. Programs that run under a shell (such as an installer script) pick one and support it. They then work fine when run under any shell.
Have you used unix? For more than a day or two perhaps?
And plumbers get the same treatment after they fix a nasty leak.
Perhaps after they install new plumbing perhaps. But what's wrong with that? I don't see why programming should be held to be anything other than a specialized trade. I'm not ashamed at the idea that I provide a valuable product/service and get paid for it.
There were paid programmers, I'm sure. But, I'm willing to bet that they were paid very modest sums. I believe they were viewed more as "secretarial" positions, rather than professions such as Doctors or Lawyers. [snip] Are they responsible, in part, for raising programmer's salary?
Chuckle. An older friend of mine bitches about the low wages these days. He made $150k+ per year, once over $220k, 1970s dollars, for programming back on old IBM mainframes. Today he makes $60k or something. Not a bad wage by any means, but a fraction of what he made before.
Why would anyone want to destroy this perceived value by making it appear as if software is easy to create and doesn't really matter? When mom-and-pop understand fully that compatible quality software can be made with programmer's free time, for free, what is stopping them from demanding this from Microsoft, Intuit, etc.?
I'm sure you also argue against including compilers with an OS, or making them freely available. I mean, if people see that they can write programs they're going to write their own and never use Microsoft's right?
I am very glad that I can program. I automate many tasks that take my friends hours. Even the ones who can use 3rd-party macro programs can't compete with a perl script I can hack together. If I want to see how a fractal changes if I modify the formula, I can. They have to ask me or hope that someone on the net had the same curiosity as them.
This is a gift that I want to share with everyone. They may never use it, but they'll be able to. Linux will never be locked down, but I can easily imagine a day when to "combat viruses" all code run on a Windows computer must be cryptographically signed. When users are crippled because a company wants to potentially squeeze more money from them in the future.
The reason your salary has not been cut is because software, for the most part, still has value.
One of the contracts I've taken was an ordering/tracking system for a company that made circuit boards. Previously an order (of anything complex) could take up to 30-40 minutes, with a few binders full of pricing charts, to price. The calculation screen(s) I made had space for 80+ variables, accessing hundreds of tables for pricing data. But you could give the customer a price as soon as you were done entering the data. It saved further time by passing the order to work stations at the various steps in the process. From ordering to a sealed computer hung over the drill press in the machine room, to accounting and shipping at the end.
The project saved an average of 15 minutes per order. It also meant that they passed notes and all design docs (the cirsuit diagrams) along as files, instead of taking a folder from station to station.
They mentioned a years or so after the project was finished that they hadn't lost an order since it was put in, and that they loved being able to pull up a spreadsheet that told them how many dollars worth of product were due to be done at any time, where work was backed up, etc. They were working to integrate it into a bonus system for the workers, as well as to let them know ahead of time about potential work shortages.
At my estimate (just of time saved initially) it let them do three times as much work per customer service rep. It eliminated one job (a guy who was moved into customer service instead of fired) of lugging paper around, keeping files straight, etc. It removed the requirement for a room of files, plus the printing costs, storage for old files, probably 5% of machinist time... Hell, even one of the accountants said he was happier because he didn't get stinky files with fiberglass shavings and etching fluid stains on them.
And you don't think that has value? Probably $400 / day, or more. That was the best $20k they ever spent.
But there's no way an off-the-shelf package would do what they want. Too much custom stuff. I've seen some systems for designing a pricing layout by drag and drop... fairly nice, but nowhere near the level of complexity something like this takes. And I doubt anyone will bother making it that good, for the
I've seen places where almost any business could benefit from custom software. Let me talk to the employees and identify their bitches and I can find even more. 10 minutes a day of hassle doesn't seem like much, but if you're paying $20/hour that's $3.33. Multiply by five employees, times 250 (working days per year) and you're at $4k. Figure in saved training time, and it's starting to look pretty sweet. (That was solved by a few simple batch files to open the right applications, perform incremental backups, etc.)
This will always have value.
[...] but it will be the FSF that brings the entire software industry to its knees.
Once again. Only if the work of paid professionals can't match the hobby work of a bunch of geeks. (Which you seem to think is really really crappy...)
But if they can't match the free software, what right to they have to bitch? They just want corporate welfare. "Rise up Joe Sixpack, cast down the shackles of free software made by the people and send half a month's wage to the world's richest man to sustain his lifestyle." It's not really concerning anyone except the rich who've sunk their mutual funds into MS stock.
Quit fear mongering.
Upping the voltage by a 10th or 20th of a volt, or overclocking by a bit, isn't going to kill anything. In fact, many motherboards do just that, they run various components with a bit more voltage to help stability. Sure, heat increases, but as long as it's withing spec you're fine. (Yes, some motherboards overclock by a bit. Check how fast a P4 CPU is running with CPUID, chances are it's not exactly at the rated speed.)
Furthermore, CPUs don't need to be matched at all for doing MP with Athlons. At least one reviewer tested with mismatched CPUs when the chipset first came out and it worked just fine. (Contrast to the Intel SMP which sometimes needs chips of the same stepping.)
Knuth seems more like the type who just doesn't enjoy tweaking people like Feynman did, so he excludes topics he knows they'd get into a thither about. Especially since they're really off-topic to the course.
Feynman seemed like my type of guy. Some people just need to be poked a bit when they're too serious about "The One Truth", whatever they feel it is.
The reason the "Woodpecker" quote bugs me is that it lacks this corrollary...
"... But once the first house was built the whole world would have free housing."