Piff. Base-10 only "makes sense" because you're used to it. I wish to hell everyone would switch to hexadecimal so I wouldn't have to keep converting from useless, inefficient base-10 to base-16.
The only hitch, of course, is deciding on how to pronounce long hexadecimal numbers. Two-digit numbers, of course, are easy enough -- A7 and CA could work out to something like "alphty-seven" and "ceety-alph" -- but past that, you either saddle yourself with the old base-10 forms -- so A77 becomes "alph hundred seventy-seven" even though "A" isn't in the hundreds position because there is no hundreds position -- or else you come up with something new like "alph hexdred seventy-seven" and irritate the heck out of people until someone decides it's all the work of Satan and we're stuck not only with the old English system of weights and measures, but base-10 as well.
As, indeed, we are now.
And God knows, everyday life is made miserable to the point of being intolerable by having to use pounds instead of kilos and 15 instead of F.
I have to admit I was all wrong about SUN -- this is a very nice operating environment for our porpoises.
I'm with you, buddy. I've had it with this Linux crap. I'm moving all of our servers over to FreeDOS just as soon as I can rustle up a few more ISA ethernet cards. Bye bye XTerm, hellooooo Telix!
I was facing this predicament recently and found a few solutions. They're all BASIC dialects, but I learned BASIC on the Apple II decades ago and still turned out to be a decent programmer when I learned "real" languages. (Personally, I think that if you can be "ruined" by using a language, you were too brittle to be a programmer anyway.)
The first was BLASSIC, which is as close to the BASIC dialects of early personal computers as it gets, complete with line numbers, PEEK, and POKE. The cool thing about it is that it supports simple graphics, which is what really got me interested in the early 80's. It's free and purports to be multiplatform, though I've only tried it under Linux.
The other two interesting BASICs were DarkBasic and Blitz. DarkBasic actually consists of several different packages. There's one called the 3D Game Designer which lets you create first-person shooters by dragging and dropping. When your little one has exhausted the rather limited possibilities there, you can upgrade to the full DarkBasic package and he or she can begin implementing simple behaviors in dirt-simple BASIC.
The Blitz product is primarily interesting (to me, anyway) because they offer a dedicated 2D version of the language devoted to 80's-style arcade games, which are orders of magnitude easier to program than 3D games for a novice programmer. Blitz doesn't use line numbers and optionally supports C-style syntax for function calls, and has a GUI toolkit that is pretty easy to use once your beginner gets a toehold in programming.
My project with my daughter was a clone of Pac Man with Blitz. It took several weekends, much of which had more to do with helping her understand cartesian coordinates than actual programming, but in the end she did most of the coding for a single level of Pac Man, and she was quite proud of herself. I doubt she'll pursue it very far -- which wasn't the point anyway -- but now she has a much better understanding of how software works.
There is no way in hell I have the time or resources to keep track of VAT rates for fifteen different countries. It's possible that if I did a lot of business with EU states that I would have no choice, but for the smattering of EU customers that I have, I'm not going to bother with it. As far as I'm concerned, any transaction conducted with me here in the United States is taking place within in the United States and is subject to US laws. If I were buying something from Europe, I would operate under the reverse assumption and pay the local taxes -- which I presume would be collected by the seller and included in the price.
Provided there's no actual enforcement, I plan to ignore this. If I get a notice from an EU tax agency that I need to pay up or face extradition on tax evasion charges, I will cut a final check to the Europeans and not deal with them in the future.
This is not, BTW, some flag-waving anti-European rant on my part -- I like the EU a good deal better than my own country -- but from a business standpoint, it isn't worth the hassle to me. I'm not sure this is such a hot idea anyway. I'm not viscerally opposed to sales taxes on net sales -- it would help curb the obliteration of thousands of local businesses by giants like Amazon -- but it ought to be collected by the seller and the seller's government. For the seller to have to keep track of the buyers' governments and their innumerable laws is an unreasonable burden on trade. Giant corporations have the resources to deal with that sort of thing; small businesses do not.
I'd think anyone concerned enough to protect their archives would want to use a serious encryption format.
For all the talk about how weak PKZip encryption supposedly is, I have a file I encrypted with PKZip back in the DOS era, forgot the passphrase, and have been unable to decrypt for fourteen years. I have tried every putative zip cracker out there without success.
While I'm pretty sure I must have used one or two English words concatenated without spaces and therefore could probably crack it with a dictionary attack, that's a complaint that could be levelled against any encryption program that uses passphrases, including mcrypt. Had I used a genuinely random string, I'd have no hope at all. (The file in question, incidentally, is not a major concern of mine, but that's beside the point.)
The point which encryption geeks often fail to get is that for most opponents, even weak encryption is good enough, and serious opponents (read: the NSA) will have either the supercomputing power to crack it or the physical might to break your kneecaps to get the key.
Frankly, I'm not sure being able to write with a pen is likely to matter too much longer. (I prefer writing over typing for making notes, but I was born before ARPAnet, much less personal computers and PDAs.)
What is disturbing is the steep decline in actual language skills. Most of my coworkers are unable to spell properly or form grammatical sentences. Many of them are unable to think clearly enough to communicate effectively even within the context of the pidgin dialect they speak. Granted, I work for a rather small and idiosyncratic company, but this was no less true when I was an Intel contractor.
Personally, it doesn't matter to me that my coworkers are semiliterate clods; it's actually an advantage for me. On the other hand, the general decline is making it harder for me to ensure that my daughter gets a decent education in the public schools, and I shudder to think that these people are voting, driving, and registering handguns.
As far as cursive is concerned, if you do plan to write with pen and paper, it's worth learning. Provided you practice enough to be good at it, you can write much faster in cursive than in regular script, which is why cursive was invented in the first place. I'm inclined to note that manual writing has a number of other advantages, but I doubt that they would appeal to anyone for whom those advantages are not self-evident.
That's definitely not cool by me, especially since, back in 1993, I submitted several terms I made up with fabricated histories, and ESR failed to include them in the dictionary.
OTOH, some other putative references have been happily absorbing all the bullshit I can make up for some time now, so I guess it all evens out in the end.
Buy out SCO. Hostile style. Buy up enough of the stock to have them vote to merge under IBM.
They can't. 80% of SCO is owned by a single shareholder. Grabbing all of the SCO stock out on the open market wouldn't do diddly for IBM.
From the other side of the equation, if you're that shareholder, what do you have to lose -- considering you've already lost your ethics and basic human decency -- by sticking to your guns? SCO is doomed in any event, but suing IBM gives them a distant chance at having a judge hand them a big chunk of change.
SCO's move is very daring and audacious -- and stupid, morally bankrupt, and damaging to society as a whole, but I digress -- and the only real alternative to closing their doors sometime within the next 18 months.
Unless something emerges to radically change the equation, I doubt either IBM or SCO will back down, and the question the rest of us must face is how to mitigate the damage from SCO's IP terrorism.
I find that I'm drawn to a lot of technical hobbies, and almost all of them are archaic or obsolete in some way -- wood engraving, traditional lead typesetting, handmade cameras and 19th century photographic processes, plant breeding and grafting -- and its so consistent that I suspect it's a sort of subconscious Luddism reacting to the modern tech I stare at for 10 hours a day or more. That they are all complex and technical is, I think, a sign that both my career and my hobbies stem from the same basic personality trait, not that the hobbies are in some way an outgrowth or extension of my career.
I've found this is a pretty common thread among my fellow techies, generally growing in strength, complexity, and expense with age. I think it stems from the love-hate relationship a lot of us have with our machines.
Everyone who has ever bought a copy of Caldera/Linux should return it. Since it was marketed as GPL code and it's not. If everyone does this right now they won't have $$$ to pay the lawyers and the whole thing might go away.
Puh-leeeaaase. It's not like they don't have the cash on hand to buy back all fifty copies.
Maybe it's time to buy a copy of The 3D Game Creator and whip out a quick Quake-style game which eschews violence against law enforcement officials but gives the player plenty of opportunity to shoot up the Washington state legislature.
Of course, this is just a cynical move on my part, as an Oregonian, to tie up the legislature of our chief economic competitor with pointless bills while our legislature concentrates on stimulating the local economy. BWAHAHAHAHA!
Oh wait, our legislature is hog-tied by anti-tax fanatics while the roads and schools crumble. But that's off-topic and not even very funny, so I'll let the matter drop.
A little marketing wouldn't hurt
on
OSI vs SCO
·
· Score: 1
While having little legal bearing, it would be nice to have an executive-summary-style comparison of the featuresets of Linux vs. SCO's offerings. Not having worked with SCO's products, I'm not in a position to do this alone (I've administered various Linux distros, Solaris, and AIX), but if some sysadmins with SCO experience would like to put some thought into this, I'll help do the writeup and the suit-friendly presentation. Contact me at eodell@sfront.com if you're interested.
All the prior art examples I've seen posted have been about autocompletion or searching a users previously entered text. They are taking this and expanding it to search the entered text of a group of users, giving the benefit of possible autocompetion of text you may have never typed.
Okay, try this -- this feature has been implemented in the Index tab of the Microsoft Help applet since the early 90's. What Amazon is talking about is autocompletion based on comparison with a list of preprogrammed list of keys.
Moreover, application of a search trie to a user interface element does not score "non-obvious" in my book.
Of course, you shouldn't be able to patent this shit in the first place, but that's an issue I keep taking up with my congressthing.
At the moment, there's not much that can be done with respect to the SCO vs. IBM case if you don't happen to be a litigant. This raises the question, at least for those who want to do more than bitch about SCO on Slashdot (no matter how gratifying it may be), of what we can do to harm SCO legally.
Eric Raymond's official rebuttal is a nice start insofar as it shows how ridiculous SCO's claims are by demonstrating that Linux is so technically superior to SCO's *nix that the idea of Linux stealing from SCO is silly. How about a nice executive-level (read: pretty and simplified) feature comparison chart showing the hundreds of features Linux has that SCO doesn't. This shouldn't be a very hard task for someone who is familiar with both systems. Distributed widely among decision-makers in the corporate world, it might have some impact.
Another thing that would be worth assembling is a migration guide from SCO to Linux. Again, someone well familiar with both systems would be needed to assemble this. Give a step-by-step HOWTO on ditching your existing SCO setup and replacing it with Linux, preferably demonstrating several improvements or new features offered by the Linux alternative.
On a related note, why are the big Linux players so quiet? I hear propaganda from SCO almost daily through various channels, but barely a peep from IBM, Red Hat, Suse, etc. You'd think they'd realize that this is a marketing action being conducted partly through the courts and would respond with marketers as well as lawyers.
You know, I hate to be the one to point this out, but "Firebird" is a damn stupid name for either a browser or a database, especially considering the implication that both products rose from the ashes of other products that were abandoned by their original developers when they ceased to be competitive.
As a database name, it's especially stupid. What's the phoenix symbolism here? "Hey, try our database! All your data will crash and burn periodically, but you'll be able to recover it." C'mon, even freaking IBM can come up with better marketing themes. At least when you hear the name of their product, DB2, you know it's a database.
The same argument could be applied to the browser, though. Or indeed to most browsers. If you're not already aware of the product, what do names like Firebird, Mozilla, and Opera mean to you? Nothing. Look at MS product names: Internet Explorer, Word, Publisher, Flight Simulator. Bigod, the product name tells you what the product is. (Of course, on the other hand, you have Excel, Access, and Bob.)
Instead of squabbling over an uninformative name that has an uninspiring thudding cadence and making the Open Source world look like a bunch of petty jerks, how about the Firebird people and the other Firebird people go back to their corners and choose new names as if their choice of names actually matters.
The rest of society has already figured this out. Ex-criminals can be useful for information, but it's not very often that they get put into positions of *trust*. I sure wouldn't want someone who's already proven their disregard for security controls designing them.
It's not quite that simple. The reason you don't hire a hacker to be a security consultant or a bank robber to be a security guard is that the connection between the two, in terms of the skills required, is tenuous at best. Yes, a cracker will tend to have decent insight into the weaknesses of a system, but it doesn't mean he has a clue about how to strengthen system security.
Most of the self-proclaimed "crackers" I've dealt with on the job have the naive notion that you probe for vulnerabilities and then plug them. While that's certainly part of securing a system, it's a bit like saying that the best way to design a sturdy boat is to take any old rotten rowboat and patch the leaks. Security -- at least if it's done right -- is akin to architecture. Mere system cracking is like throwing bricks through windows. How does vandalism qualify someone to be an architect?
The other big strike against firms composed wholly or partly of crackers is that they tend to be little more than teenage cults of personality for people whose security credentials are on a par with Ozzy Osbourne's Satanist credentials. The same may often be true of suit-driven organizations within their peer group, but I'd rather take advice from someone with no criminal record and an advanced degree than from some bush league rawkstar who doesn't understand why Back Orifice didn't really say anything at all about Windows security.
Actually, they don't know if it was a software bug. At this point that is pure (though somewhat educated) speculation.
Yes, but what can you expect from closed-source Russian software? If only they'd been running an early version of the 2.4 kernel, they'd --
Oh, right. It would have still been swapping madly as it gouged a fifty-meter crater into the steppes of Kazakhstan.
Really, folks, one would expect an audience with a higher-than-usual proportion of programmers to be at least a little sympathetic to the fact that all of us, no matter how good we are, no matter how well we design, implement, and test, will inevitably let some bugs slip through. And as noted in the parent post, the backup system did kick in and work like a charm. I wish that would happen when X started to crash!
If necessary, us geeks will revert to uucp-style networking - for those too young to remember, that's dial up connections from place to place, covering the whole planet.
That's great, except that back then, we were politically powerless and numbered in the tens of thousands at best. Not that I don't miss dialing into machines running Waffle -- after waiting a few hours for the other users to hang up -- but as rewarding as the BBS era was (and in ways the always-had-the-internet crowd can't imagine), we were way below ham radio operators in terms of influence and maybe -- maybe -- slightly above the radio-controlled modelling crowd.
All of this is a moot point, of course, since the law being discussed would apply to dialup communications, too, unless you're proposing running your own telco. And I hate to hit you with this, but private citizens can't lay their own fiber or copper through their neighborhoods because it would involve crossing utility right-of-ways and public streets. Even if they could, at some point you're going to have to tap into the "public" internet and play by ISP rules.
Laser communications -- well, they're good over short distances and in good weather. Unless you live in a vacuum, their applicability is pretty damn limited.
The idea that technology can magically overcome the power of the state is one that, by now, ought to have been thoroughly discredited. The last technology that suddenly overpowered a national entity was carried in the belly of the Enola Gay and, gross security lapses at national labs notwithstanding, wasn't open source.
the 'experience' that GW wants customers to have (of coming into their own stores and getting a hard sell)
Huh? Every time I walk into a GW store, a teenage employee saunters up and asks me if he or she can help me find something, and I say, "No thanks, I'm just browsing," and that's it. Which is pretty much my experience everywhere that the salespeople aren't paid on commission. If you're being pushed around by the pasty RPG geeks at GW, consider exercising.
The authors spend four lines discussing the best computer language for the job (C/C++)
Am I the only C and C++ programmer who finds the "C/C++" label annoying as hell? Having it come from HR people who don't know any better is one thing, but hearing it from programmers drives me up the wall. I sometimes suspect it comes from C++-only programmers operating under the mistaken assumption that because C++ is a superset of C, they know C, too.
Despite similar syntax, C and C++ are completely different languages. C++-only programmers write C code that's on a par with the code produced by C-only programmers dabbling in C++. Perl, PHP, Objective C, and several dozen other Algol-descended languages have really similar syntax, but no one says "Algol/BCPL/C/C++/PHP/Perl/Pascal" with a straight face.
My guess is that if the authors are extolling the virtues of objects, they are primarily extolling the virtues of C++.
Strange things happen. Whenever the technicalities of music compression are discussed you will find a lot of people claiming that mp3 is indistinguishable from the original CD. When it comes to downloading everyone will say that the compressed music is vastly inferior...
It could be in part because people tend to listen to MP3s with crappy PC speakers or equally crappy earbud headphones.
An MP3 encoded at 128kbps -- when played on quality audio equipment -- sounds noticeably different than the CD original if you are paying any attention at all. Some kinds of music sound worse than others, depending on the predominant waveforms and the particular encoder implementation. Music with high levels of distortion seems particularly prone to really noticeable encoding artifacts.
To me, the difference disappears somewhere between 256 and 320kbps, depending on whether I'm actually listening to the music or just using it as audio wallpaper. Consequently, whether the MP3 is an incentive to buy the CD depends on how I listen to it. If I really like the artist and actually sit down and listen to the music without doing anything else, I'll buy the CD. If I don't care that much and use the music as wallpaper, I won't buy it.
I should note, however, that for "wallpaper" music, I wouldn't buy the CD even if I couldn't get the MP3. OTOH, there's a lot of music I like enough to buy that I never would have heard of if I hadn't listened to the MP3 first.
But I digress. Pump a CD track through a good stereo system, and then play the 128 or 192kbps MP3 version. MP3s are definitely not CD quality, at least not at the bitrates at which they are commonly encoded.
Well, it either is or isn't a flaw, depending on what you want to get out of Google. The author of the article plainly wants to get search results in which websites are ranked in a manner that reflects their offline significance, which is, as you note, probably not possible.
OTOH, Google's apparent over-reliance on link density as a measure of "authoritativeness" pretty obviously breaks down when blog culture is thrown into the mix. Exactly what practical use can be derived from searching blogs is beyond me. They are useless as a metric of current trends -- there is, after all, a reason why surveys are conducted in accordance with fairly strict statistical standards. As simple sources of information, a blog is roughly equivalent to listening to a random conversation on a bus or a university cafeteria.
Consider the search keys "meme" and "paradigm". Of the people who use one or both terms, there's probably not one person in a hundred who actually understands them and uses them correctly. And particularly in the case of "paradigm," there are several correct usages particular to different fields -- the CS usage has little to do with the philosophy-of-science usage, and neither have much to do with the New Age usage. How far down the list of Google search results do you have to go to get to useful information about memetics or paradigm shifts? Quite a ways.
Another example, and one that personally vexes me, is "visualization". As a Google search key, the term retrieves link after link after link about data visualization. Good luck finding anything about mental visualization unless you figure out some additional keys to select psychology-related pages. (Hint: "cognitive" is a good secondary key.)
The root problem, I think, is that Google doesn't really take into account topic-specific usages. If you were to make a graph of links between "visualization"-related sites, you would see a big island composed of computer science sites, a smaller island of cognitive psychology sites, and a still smaller island of Shakti Gawain fans. Odds are that any given user will be interested in only one of those islands. Northern Light, before it closed its public front to become a CIA contractor, actually did a pretty good job of grouping related subjects in a convenient UI; one can only hope Google eventually will, too.
The point of the article isn't about competing "memes", it's about flaws in Google's PageRank system. In this case, the weighting of site importance on the basis of the number and authoritativeness is being thrown by the incestuous linking behavior of weblogs.
If there's a flaw in the article, it's that it implies -- without ever quite coming out and saying it -- that there was some sort of conspiracy or malice aforethought. Obviously, there was not.
Also implied but not stated directly is that Google has some kind of responsibility to make sure that its results aren't skewed by anomalous inputs. Whether they do or not is a matter of opinion, but the article would have been much stronger if the author had addressed the point directly. OTOH, considering how little traction the idea of social responsibility has among the center-right libertarian crowd that reads the Register, I can see why the author declined to wander into that quagmire.
While those of us who have ideological sympathies with RMS may not be happy to hear the Indian decision, it is nonetheless worth pointing out that a significant economic goal of free software -- choice -- is very definitely being fostered by the widespread availability of free software.
If Microsoft is forced to lower its prices, relax its licensing conditions, or make "donations" to state governments (in lieu of the taxes they don't pay), then this is all to the good. Microsoft is finally being forced to compete, which was a major and laudable goal of the ESR/Open Source half of the movement.
It's not a total loss for the Free Software side, either. That Microsoft is being forced to compete is a concrete sign that we are making credible inroads -- that the software equivalent of the Republican Guard, as it were, is withdrawing into the streets of Redmond for a last stand. The outcome, however, is not in doubt.
By the way, does anyone wonder what would happen if the government taxed the bible (which is in public domain)?
I hate to break it to you, but no it's not. The King James Version -- arguably one of the worst translations in terms of faithfulness to the original text -- is the only significant translation in the public domain. All the rest are copyrighted and earning royalties for their owners. Oh sure, there's the Vulgate, but that won't do you much good unless you're comfortable with a Latin dialect that isn't taught anymore. Photographic reproductions of the surviving source documents -- which aren't much use unless you know ancient Hebrew, Aramaic, or Koine Greek -- are very much under copyright.
This isn't my specific concern, since it's not my religion, but I've met a number of Christian copyright activists who are quite pissed about the way their holy books and hymns are controlled by avaricious lawyers from the religious publishing industry who would rather have people go without access to the Bible if they can't pay for access to it. Similar problems afflict members of other religions, since the same problem applies -- the chief texts are written in dead languages, and most or all of the translations into modern languages are someone's intellectual "property". Those of us who belong to younger religions are in even worse shape, since the source texts themselves suffer from modern perpetual copyrights.
I think it could get a lot of people angry.
So get angry. Fight the law or, better yet, devote yourself to being a first-rate scholar of the languages involved, and produce an open source translation of such quality that it forces the intellectual property parasites out of business. At least until our government abolishes the public domain, too.
Slashdot Mirror
a well thought-out system that makes sense
Piff. Base-10 only "makes sense" because you're used to it. I wish to hell everyone would switch to hexadecimal so I wouldn't have to keep converting from useless, inefficient base-10 to base-16.
The only hitch, of course, is deciding on how to pronounce long hexadecimal numbers. Two-digit numbers, of course, are easy enough -- A7 and CA could work out to something like "alphty-seven" and "ceety-alph" -- but past that, you either saddle yourself with the old base-10 forms -- so A77 becomes "alph hundred seventy-seven" even though "A" isn't in the hundreds position because there is no hundreds position -- or else you come up with something new like "alph hexdred seventy-seven" and irritate the heck out of people until someone decides it's all the work of Satan and we're stuck not only with the old English system of weights and measures, but base-10 as well.
As, indeed, we are now.
And God knows, everyday life is made miserable to the point of being intolerable by having to use pounds instead of kilos and 15 instead of F.
I have to admit I was all wrong about SUN -- this is a very nice operating environment for our porpoises.
I'm with you, buddy. I've had it with this Linux crap. I'm moving all of our servers over to FreeDOS just as soon as I can rustle up a few more ISA ethernet cards. Bye bye XTerm, hellooooo Telix!
I was facing this predicament recently and found a few solutions. They're all BASIC dialects, but I learned BASIC on the Apple II decades ago and still turned out to be a decent programmer when I learned "real" languages. (Personally, I think that if you can be "ruined" by using a language, you were too brittle to be a programmer anyway.)
The first was BLASSIC, which is as close to the BASIC dialects of early personal computers as it gets, complete with line numbers, PEEK, and POKE. The cool thing about it is that it supports simple graphics, which is what really got me interested in the early 80's. It's free and purports to be multiplatform, though I've only tried it under Linux.
The other two interesting BASICs were DarkBasic and Blitz. DarkBasic actually consists of several different packages. There's one called the 3D Game Designer which lets you create first-person shooters by dragging and dropping. When your little one has exhausted the rather limited possibilities there, you can upgrade to the full DarkBasic package and he or she can begin implementing simple behaviors in dirt-simple BASIC.
The Blitz product is primarily interesting (to me, anyway) because they offer a dedicated 2D version of the language devoted to 80's-style arcade games, which are orders of magnitude easier to program than 3D games for a novice programmer. Blitz doesn't use line numbers and optionally supports C-style syntax for function calls, and has a GUI toolkit that is pretty easy to use once your beginner gets a toehold in programming.
My project with my daughter was a clone of Pac Man with Blitz. It took several weekends, much of which had more to do with helping her understand cartesian coordinates than actual programming, but in the end she did most of the coding for a single level of Pac Man, and she was quite proud of herself. I doubt she'll pursue it very far -- which wasn't the point anyway -- but now she has a much better understanding of how software works.
There is no way in hell I have the time or resources to keep track of VAT rates for fifteen different countries. It's possible that if I did a lot of business with EU states that I would have no choice, but for the smattering of EU customers that I have, I'm not going to bother with it. As far as I'm concerned, any transaction conducted with me here in the United States is taking place within in the United States and is subject to US laws. If I were buying something from Europe, I would operate under the reverse assumption and pay the local taxes -- which I presume would be collected by the seller and included in the price.
Provided there's no actual enforcement, I plan to ignore this. If I get a notice from an EU tax agency that I need to pay up or face extradition on tax evasion charges, I will cut a final check to the Europeans and not deal with them in the future.
This is not, BTW, some flag-waving anti-European rant on my part -- I like the EU a good deal better than my own country -- but from a business standpoint, it isn't worth the hassle to me. I'm not sure this is such a hot idea anyway. I'm not viscerally opposed to sales taxes on net sales -- it would help curb the obliteration of thousands of local businesses by giants like Amazon -- but it ought to be collected by the seller and the seller's government. For the seller to have to keep track of the buyers' governments and their innumerable laws is an unreasonable burden on trade. Giant corporations have the resources to deal with that sort of thing; small businesses do not.
I'd think anyone concerned enough to protect their archives would want to use a serious encryption format.
For all the talk about how weak PKZip encryption supposedly is, I have a file I encrypted with PKZip back in the DOS era, forgot the passphrase, and have been unable to decrypt for fourteen years. I have tried every putative zip cracker out there without success.
While I'm pretty sure I must have used one or two English words concatenated without spaces and therefore could probably crack it with a dictionary attack, that's a complaint that could be levelled against any encryption program that uses passphrases, including mcrypt. Had I used a genuinely random string, I'd have no hope at all. (The file in question, incidentally, is not a major concern of mine, but that's beside the point.)
The point which encryption geeks often fail to get is that for most opponents, even weak encryption is good enough, and serious opponents (read: the NSA) will have either the supercomputing power to crack it or the physical might to break your kneecaps to get the key.
Frankly, I'm not sure being able to write with a pen is likely to matter too much longer. (I prefer writing over typing for making notes, but I was born before ARPAnet, much less personal computers and PDAs.)
What is disturbing is the steep decline in actual language skills. Most of my coworkers are unable to spell properly or form grammatical sentences. Many of them are unable to think clearly enough to communicate effectively even within the context of the pidgin dialect they speak. Granted, I work for a rather small and idiosyncratic company, but this was no less true when I was an Intel contractor.
Personally, it doesn't matter to me that my coworkers are semiliterate clods; it's actually an advantage for me. On the other hand, the general decline is making it harder for me to ensure that my daughter gets a decent education in the public schools, and I shudder to think that these people are voting, driving, and registering handguns.
As far as cursive is concerned, if you do plan to write with pen and paper, it's worth learning. Provided you practice enough to be good at it, you can write much faster in cursive than in regular script, which is why cursive was invented in the first place. I'm inclined to note that manual writing has a number of other advantages, but I doubt that they would appeal to anyone for whom those advantages are not self-evident.
That's definitely not cool by me, especially since, back in 1993, I submitted several terms I made up with fabricated histories, and ESR failed to include them in the dictionary.
OTOH, some other putative references have been happily absorbing all the bullshit I can make up for some time now, so I guess it all evens out in the end.
Buy out SCO. Hostile style. Buy up enough of the stock to have them vote to merge under IBM.
They can't. 80% of SCO is owned by a single shareholder. Grabbing all of the SCO stock out on the open market wouldn't do diddly for IBM.
From the other side of the equation, if you're that shareholder, what do you have to lose -- considering you've already lost your ethics and basic human decency -- by sticking to your guns? SCO is doomed in any event, but suing IBM gives them a distant chance at having a judge hand them a big chunk of change.
SCO's move is very daring and audacious -- and stupid, morally bankrupt, and damaging to society as a whole, but I digress -- and the only real alternative to closing their doors sometime within the next 18 months.
Unless something emerges to radically change the equation, I doubt either IBM or SCO will back down, and the question the rest of us must face is how to mitigate the damage from SCO's IP terrorism.
I find that I'm drawn to a lot of technical hobbies, and almost all of them are archaic or obsolete in some way -- wood engraving, traditional lead typesetting, handmade cameras and 19th century photographic processes, plant breeding and grafting -- and its so consistent that I suspect it's a sort of subconscious Luddism reacting to the modern tech I stare at for 10 hours a day or more. That they are all complex and technical is, I think, a sign that both my career and my hobbies stem from the same basic personality trait, not that the hobbies are in some way an outgrowth or extension of my career.
I've found this is a pretty common thread among my fellow techies, generally growing in strength, complexity, and expense with age. I think it stems from the love-hate relationship a lot of us have with our machines.
Everyone who has ever bought a copy of Caldera/Linux should return it. Since it was marketed as GPL code and it's not. If everyone does this right now they won't have $$$ to pay the lawyers and the whole thing might go away.
Puh-leeeaaase. It's not like they don't have the cash on hand to buy back all fifty copies.
Maybe it's time to buy a copy of The 3D Game Creator and whip out a quick Quake-style game which eschews violence against law enforcement officials but gives the player plenty of opportunity to shoot up the Washington state legislature.
Of course, this is just a cynical move on my part, as an Oregonian, to tie up the legislature of our chief economic competitor with pointless bills while our legislature concentrates on stimulating the local economy. BWAHAHAHAHA!
Oh wait, our legislature is hog-tied by anti-tax fanatics while the roads and schools crumble. But that's off-topic and not even very funny, so I'll let the matter drop.
While having little legal bearing, it would be nice to have an executive-summary-style comparison of the featuresets of Linux vs. SCO's offerings. Not having worked with SCO's products, I'm not in a position to do this alone (I've administered various Linux distros, Solaris, and AIX), but if some sysadmins with SCO experience would like to put some thought into this, I'll help do the writeup and the suit-friendly presentation. Contact me at eodell@sfront.com if you're interested.
All the prior art examples I've seen posted have been about autocompletion or searching a users previously entered text. They are taking this and expanding it to search the entered text of a group of users, giving the benefit of possible autocompetion of text you may have never typed.
Okay, try this -- this feature has been implemented in the Index tab of the Microsoft Help applet since the early 90's. What Amazon is talking about is autocompletion based on comparison with a list of preprogrammed list of keys.
Moreover, application of a search trie to a user interface element does not score "non-obvious" in my book.
Of course, you shouldn't be able to patent this shit in the first place, but that's an issue I keep taking up with my congressthing.
At the moment, there's not much that can be done with respect to the SCO vs. IBM case if you don't happen to be a litigant. This raises the question, at least for those who want to do more than bitch about SCO on Slashdot (no matter how gratifying it may be), of what we can do to harm SCO legally.
Eric Raymond's official rebuttal is a nice start insofar as it shows how ridiculous SCO's claims are by demonstrating that Linux is so technically superior to SCO's *nix that the idea of Linux stealing from SCO is silly. How about a nice executive-level (read: pretty and simplified) feature comparison chart showing the hundreds of features Linux has that SCO doesn't. This shouldn't be a very hard task for someone who is familiar with both systems. Distributed widely among decision-makers in the corporate world, it might have some impact.
Another thing that would be worth assembling is a migration guide from SCO to Linux. Again, someone well familiar with both systems would be needed to assemble this. Give a step-by-step HOWTO on ditching your existing SCO setup and replacing it with Linux, preferably demonstrating several improvements or new features offered by the Linux alternative.
On a related note, why are the big Linux players so quiet? I hear propaganda from SCO almost daily through various channels, but barely a peep from IBM, Red Hat, Suse, etc. You'd think they'd realize that this is a marketing action being conducted partly through the courts and would respond with marketers as well as lawyers.
You know, I hate to be the one to point this out, but "Firebird" is a damn stupid name for either a browser or a database, especially considering the implication that both products rose from the ashes of other products that were abandoned by their original developers when they ceased to be competitive.
As a database name, it's especially stupid. What's the phoenix symbolism here? "Hey, try our database! All your data will crash and burn periodically, but you'll be able to recover it." C'mon, even freaking IBM can come up with better marketing themes. At least when you hear the name of their product, DB2, you know it's a database.
The same argument could be applied to the browser, though. Or indeed to most browsers. If you're not already aware of the product, what do names like Firebird, Mozilla, and Opera mean to you? Nothing. Look at MS product names: Internet Explorer, Word, Publisher, Flight Simulator. Bigod, the product name tells you what the product is. (Of course, on the other hand, you have Excel, Access, and Bob.)
Instead of squabbling over an uninformative name that has an uninspiring thudding cadence and making the Open Source world look like a bunch of petty jerks, how about the Firebird people and the other Firebird people go back to their corners and choose new names as if their choice of names actually matters.
The rest of society has already figured this out. Ex-criminals can be useful for information, but it's not very often that they get put into positions of *trust*. I sure wouldn't want someone who's already proven their disregard for security controls designing them.
It's not quite that simple. The reason you don't hire a hacker to be a security consultant or a bank robber to be a security guard is that the connection between the two, in terms of the skills required, is tenuous at best. Yes, a cracker will tend to have decent insight into the weaknesses of a system, but it doesn't mean he has a clue about how to strengthen system security.
Most of the self-proclaimed "crackers" I've dealt with on the job have the naive notion that you probe for vulnerabilities and then plug them. While that's certainly part of securing a system, it's a bit like saying that the best way to design a sturdy boat is to take any old rotten rowboat and patch the leaks. Security -- at least if it's done right -- is akin to architecture. Mere system cracking is like throwing bricks through windows. How does vandalism qualify someone to be an architect?
The other big strike against firms composed wholly or partly of crackers is that they tend to be little more than teenage cults of personality for people whose security credentials are on a par with Ozzy Osbourne's Satanist credentials. The same may often be true of suit-driven organizations within their peer group, but I'd rather take advice from someone with no criminal record and an advanced degree than from some bush league rawkstar who doesn't understand why Back Orifice didn't really say anything at all about Windows security.
Actually, they don't know if it was a software bug. At this point that is pure (though somewhat educated) speculation.
Yes, but what can you expect from closed-source Russian software? If only they'd been running an early version of the 2.4 kernel, they'd --
Oh, right. It would have still been swapping madly as it gouged a fifty-meter crater into the steppes of Kazakhstan.
Really, folks, one would expect an audience with a higher-than-usual proportion of programmers to be at least a little sympathetic to the fact that all of us, no matter how good we are, no matter how well we design, implement, and test, will inevitably let some bugs slip through. And as noted in the parent post, the backup system did kick in and work like a charm. I wish that would happen when X started to crash!
If necessary, us geeks will revert to uucp-style networking - for those too young to remember, that's dial up connections from place to place, covering the whole planet.
That's great, except that back then, we were politically powerless and numbered in the tens of thousands at best. Not that I don't miss dialing into machines running Waffle -- after waiting a few hours for the other users to hang up -- but as rewarding as the BBS era was (and in ways the always-had-the-internet crowd can't imagine), we were way below ham radio operators in terms of influence and maybe -- maybe -- slightly above the radio-controlled modelling crowd.
All of this is a moot point, of course, since the law being discussed would apply to dialup communications, too, unless you're proposing running your own telco. And I hate to hit you with this, but private citizens can't lay their own fiber or copper through their neighborhoods because it would involve crossing utility right-of-ways and public streets. Even if they could, at some point you're going to have to tap into the "public" internet and play by ISP rules.
Laser communications -- well, they're good over short distances and in good weather. Unless you live in a vacuum, their applicability is pretty damn limited.
The idea that technology can magically overcome the power of the state is one that, by now, ought to have been thoroughly discredited. The last technology that suddenly overpowered a national entity was carried in the belly of the Enola Gay and, gross security lapses at national labs notwithstanding, wasn't open source.
the 'experience' that GW wants customers to have (of coming into their own stores and getting a hard sell)
Huh? Every time I walk into a GW store, a teenage employee saunters up and asks me if he or she can help me find something, and I say, "No thanks, I'm just browsing," and that's it. Which is pretty much my experience everywhere that the salespeople aren't paid on commission. If you're being pushed around by the pasty RPG geeks at GW, consider exercising.
The authors spend four lines discussing the best computer language for the job (C/C++)
Am I the only C and C++ programmer who finds the "C/C++" label annoying as hell? Having it come from HR people who don't know any better is one thing, but hearing it from programmers drives me up the wall. I sometimes suspect it comes from C++-only programmers operating under the mistaken assumption that because C++ is a superset of C, they know C, too.
Despite similar syntax, C and C++ are completely different languages. C++-only programmers write C code that's on a par with the code produced by C-only programmers dabbling in C++. Perl, PHP, Objective C, and several dozen other Algol-descended languages have really similar syntax, but no one says "Algol/BCPL/C/C++/PHP/Perl/Pascal" with a straight face.
My guess is that if the authors are extolling the virtues of objects, they are primarily extolling the virtues of C++.
Strange things happen. Whenever the technicalities of music compression are discussed you will find a lot of people claiming that mp3 is indistinguishable from the original CD. When it comes to downloading everyone will say that the compressed music is vastly inferior...
It could be in part because people tend to listen to MP3s with crappy PC speakers or equally crappy earbud headphones.
An MP3 encoded at 128kbps -- when played on quality audio equipment -- sounds noticeably different than the CD original if you are paying any attention at all. Some kinds of music sound worse than others, depending on the predominant waveforms and the particular encoder implementation. Music with high levels of distortion seems particularly prone to really noticeable encoding artifacts.
To me, the difference disappears somewhere between 256 and 320kbps, depending on whether I'm actually listening to the music or just using it as audio wallpaper. Consequently, whether the MP3 is an incentive to buy the CD depends on how I listen to it. If I really like the artist and actually sit down and listen to the music without doing anything else, I'll buy the CD. If I don't care that much and use the music as wallpaper, I won't buy it.
I should note, however, that for "wallpaper" music, I wouldn't buy the CD even if I couldn't get the MP3. OTOH, there's a lot of music I like enough to buy that I never would have heard of if I hadn't listened to the MP3 first.
But I digress. Pump a CD track through a good stereo system, and then play the 128 or 192kbps MP3 version. MP3s are definitely not CD quality, at least not at the bitrates at which they are commonly encoded.
First, I don't see it as a flaw.
Well, it either is or isn't a flaw, depending on what you want to get out of Google. The author of the article plainly wants to get search results in which websites are ranked in a manner that reflects their offline significance, which is, as you note, probably not possible.
OTOH, Google's apparent over-reliance on link density as a measure of "authoritativeness" pretty obviously breaks down when blog culture is thrown into the mix. Exactly what practical use can be derived from searching blogs is beyond me. They are useless as a metric of current trends -- there is, after all, a reason why surveys are conducted in accordance with fairly strict statistical standards. As simple sources of information, a blog is roughly equivalent to listening to a random conversation on a bus or a university cafeteria.
Consider the search keys "meme" and "paradigm". Of the people who use one or both terms, there's probably not one person in a hundred who actually understands them and uses them correctly. And particularly in the case of "paradigm," there are several correct usages particular to different fields -- the CS usage has little to do with the philosophy-of-science usage, and neither have much to do with the New Age usage. How far down the list of Google search results do you have to go to get to useful information about memetics or paradigm shifts? Quite a ways.
Another example, and one that personally vexes me, is "visualization". As a Google search key, the term retrieves link after link after link about data visualization. Good luck finding anything about mental visualization unless you figure out some additional keys to select psychology-related pages. (Hint: "cognitive" is a good secondary key.)
The root problem, I think, is that Google doesn't really take into account topic-specific usages. If you were to make a graph of links between "visualization"-related sites, you would see a big island composed of computer science sites, a smaller island of cognitive psychology sites, and a still smaller island of Shakti Gawain fans. Odds are that any given user will be interested in only one of those islands. Northern Light, before it closed its public front to become a CIA contractor, actually did a pretty good job of grouping related subjects in a convenient UI; one can only hope Google eventually will, too.
The point of the article isn't about competing "memes", it's about flaws in Google's PageRank system. In this case, the weighting of site importance on the basis of the number and authoritativeness is being thrown by the incestuous linking behavior of weblogs.
If there's a flaw in the article, it's that it implies -- without ever quite coming out and saying it -- that there was some sort of conspiracy or malice aforethought. Obviously, there was not.
Also implied but not stated directly is that Google has some kind of responsibility to make sure that its results aren't skewed by anomalous inputs. Whether they do or not is a matter of opinion, but the article would have been much stronger if the author had addressed the point directly. OTOH, considering how little traction the idea of social responsibility has among the center-right libertarian crowd that reads the Register, I can see why the author declined to wander into that quagmire.
While those of us who have ideological sympathies with RMS may not be happy to hear the Indian decision, it is nonetheless worth pointing out that a significant economic goal of free software -- choice -- is very definitely being fostered by the widespread availability of free software.
If Microsoft is forced to lower its prices, relax its licensing conditions, or make "donations" to state governments (in lieu of the taxes they don't pay), then this is all to the good. Microsoft is finally being forced to compete, which was a major and laudable goal of the ESR/Open Source half of the movement.
It's not a total loss for the Free Software side, either. That Microsoft is being forced to compete is a concrete sign that we are making credible inroads -- that the software equivalent of the Republican Guard, as it were, is withdrawing into the streets of Redmond for a last stand. The outcome, however, is not in doubt.
By the way, does anyone wonder what would happen if the government taxed the bible (which is in public domain)?
I hate to break it to you, but no it's not. The King James Version -- arguably one of the worst translations in terms of faithfulness to the original text -- is the only significant translation in the public domain. All the rest are copyrighted and earning royalties for their owners. Oh sure, there's the Vulgate, but that won't do you much good unless you're comfortable with a Latin dialect that isn't taught anymore. Photographic reproductions of the surviving source documents -- which aren't much use unless you know ancient Hebrew, Aramaic, or Koine Greek -- are very much under copyright.
This isn't my specific concern, since it's not my religion, but I've met a number of Christian copyright activists who are quite pissed about the way their holy books and hymns are controlled by avaricious lawyers from the religious publishing industry who would rather have people go without access to the Bible if they can't pay for access to it. Similar problems afflict members of other religions, since the same problem applies -- the chief texts are written in dead languages, and most or all of the translations into modern languages are someone's intellectual "property". Those of us who belong to younger religions are in even worse shape, since the source texts themselves suffer from modern perpetual copyrights.
I think it could get a lot of people angry.
So get angry. Fight the law or, better yet, devote yourself to being a first-rate scholar of the languages involved, and produce an open source translation of such quality that it forces the intellectual property parasites out of business. At least until our government abolishes the public domain, too.