You miss the point of steganography. Encryption assumes that it's acceptable for an attacker to know there's a communications channel, the requirement is to keep the attacker from finding out the contents of the channel. Steganography is intended to conceal the very existence of the communications channel from a potential attacker.
Consider the situation a dissident in China might be in. Merely concealing what he's posting won't help him. The government doesn't care what the content is, the mere fact that he's hiding it from them's enough to convict him as far as they're concerned. For him encryption isn't required, an encrypted message the government can't read is just as damning as the plaintext would be. What he needs is a channel that's so unobtrusive that the government doesn't realize he's posting anything at all. And if the government doesn't realize there's a message there, they aren't even going to try to read it.
Hmm. Infrared and visible-light pulses. No electricity. Needs to have a message encoded on-the-fly in the field. That sounds an awful lot like a small lantern with an oil wick or candle burning in it and a shutter to conceal or reveal the light/heat source at will. We could even use Morse code. All you'd need is an ignition source to light the lamp. Matches would work, or steel-and-flint to light a small piece of tinder scavenged from the surroundings.
This shouldn't be allowed. Public notices in newspapers serve two purposes. The first is the one mentioned, publishing the notice where interested parties can see it. The second isn't mentioned, though, and that's to create a record of the notice outside the control of the party required to post it. The notice can't be changed later, can't be quietly made to never have happened. We've already seen entities change stories posted on their Web sites when what was in those stories became inconvenient later. Yes, it's going to cost a little extra to maintain that independent record of the notices. When we make a big payment or an important one where not making it has big consequences, it definitely costs for them to give us a receipt that we can use later to prove we did pay and what we paid for. We don't accept the cost savings as a valid reason for not being given a receipt, we don't accept "Trust us, we've got a record of your payment.".
Well, I suppose I can be considered a newcomer. I've only been programming professionally for 25 years, I came in after microcomputers came on the scene (although I do pre-date Microsoft itself). I've worked with Xerces, libxml2, Rogue Wave's XML parse, Microsoft's built-in parser, Java's XML packages and a couple of home-brew monstrosities, and I do suggest you go look at their APIs before you go claiming they're radically different.
As for your objections, I'll just take point A as a case study. If ODF 1.1 is so incapable of storing formulas, then how do OpenOffice, KOffice, Google Docs and Symphony all manage to store spreadsheet formulas in it and manage to understand each other's formulas and have theirs be understood in turn? Certainly it may not be possible the way Microsoft would like to do it, but that's like saying it's impossible to drive a car with an automatic transmission because there's no clutch pedal. And yes, Symphony currently does manage to not mess it up. Your confusion becomes apparent in point E, as OOo hasn't used 1.1 in a while (versions since at least 2.4 have used ODF 1.2). You appear to be working with very old versions, which is likely the cause of your failures.
Up-front costs. Yes, it's a better deal to have them pressed yourself, but you have to shell out that $999 up front. At $8.98 each you have to sell 112 copies just to cover manufacturing costs, and then there's packaging and shipping and storage costs. With Amazon, you give up some potential profit but you also don't have to shell out that 1 grand up front. If you aren't sure you'll sell 112 copies, or at least won't sell that many quickly, Amazon's deal may be a good bargain.
One thing: an existing contract. That's how the labels mostly do it. They don't go after the big-name bands. They go after the small bands who're just getting started, and get them into a contract for a certain number of albums. Note that that's for a certain number released, not produced, and the label gets to decide whether an album will be released. That can be tempting because of the advance and access to studio facilities and backup performers and such that new bands can't always get on their own. And if the band is a hit and becomes a big-name band, well, there's still that contract. The label owns all the rights to songs done under the contract, so the band can't even perform their own works without paying the label. If they try releasing anything, the label can come after them because they owe those songs to the label under the contract. The only way out of this once you're in it is to be a big enough hit that the label winds up having to release enough albums for you to complete your contract with them.
One reason the labels can do this is that they've managed to pretty well disrupt any legitimate distribution channel for music that doesn't involve physical production of media. The labels control the distribution chain to stores, so even if a band can afford to press and package their own CDs they have a hard time getting them on shelves without signing with a label. Electronic channels where the band might be able to make money, the labels have either sued out of existence or saddled with such onerous restrictions that consumers aren't willing to deal with it. The only avenue left is for a band to set up it's own Web site with it's own on-line store. That depends on search services like Google to give consumers a way to easily find the band's store, and you'll notice the recent attacks on Google's ability to link to things. Make no mistake, those attacks aren't aimed at keeping infringing material off Google's link lists. You'll notice, for instance, that in the Copiepresse case and others, when Google offered to simply stop listing the plaintiff's material completely the plaintiffs objected vociferously, demanding that Google not be allowed to do that. Their goal isn't to get their material off Google, it's to make it so difficult and expensive that Google can't stay in business at all. Search services are a critical link in an electronic distribution channel that completely bypasses the labels, and they want to disrupt them before they fully form.
In an RIAA contract, the artist's looking at maybe a 5-8% nominal on sales.
Note that the artist doesn't get paid that much. There's any advance, studio time, promotional fees, packaging fees, distribution fees, breakage allowance, all sorts of charges the artist is responsible for that come out of gross royalties. The artist gets paid what's left after the label takes all of those out. Most artists are lucky to see 1-2% after all that.
You see now why the labels are so dead-set against any legal electronic distribution of music.
Like Microsoft, you're hopelessly confused about the difference between external and internal representations. It's rather trivial to attach arbitrary XML DOM elements to C++ objects. After all, those XML elements are, when the parser gets done with them, just C++ objects themselves. ODF even makes it easy, since formulas are stored in attributes you don't even need to handle the full general case of nested elements to get it right. A simple linked list of unrecognized attributes (which is what the parser hands you on a silver platter) is all you need. Anybody with more than 5 or so years experience had probably already done that or the equivalent several times, and I'm positive Microsoft doesn't put people with less than 5 years experience in lead designer slots. And that's not even counting the fact that they already had done it right. That leaves only one other option: Microsoft's programmers were fully capable of supporting ODF spreadsheet formulas in a compatible way, and they were instructed to break them.
Someone needs to remind Mr. Yee that, at least in all the cases I've heard reported on, the store didn't sell the video game to the kid. They sold it to an adult relative of the kid, who then gave it to the kid without bothering to check on what exactly their "little angel" had been bugging them for. And then when they found out exactly what little Timmy had gotten, they dove headfirst into that river in Africa and started looking around for someone else to take the blame for their failure. No law about selling video games to minors will do a single blessed thing about that, where there's no video game ever sold to the minor.
There's a lot more than 2 implementations. Besides OpenOffice and MS Office there's AbiWord, KOffice, Google Docs, WordPerfect Office X4, IBM's Lotus Symphony, the Sun ODF plug-in for MS Word and the BSD-licensed ODF plug-in for Word that Microsoft funded and hosted on SourceForge. That last is important, BTW. Not only is Office 2007's implementation of ODF incompatible with OpenOffice, it's incompatible with Microsoft's own other implementation of ODF.
Pretty much, yes. Bear in mind that Microsoft already has code that does handle the spreadsheet formulas correctly. The plug-in that Microsoft itself commissioned and that they own the code for not only preserves the formulas, it correctly parses and interprets them so that cells get recalculated properly as data changes and it correctly writes changed formulas back out. All Microsoft had to do was to not do all the work a second time. And even if they had re-done the work, the XML parser automatically populates the DOM with the formula strings and the internal implementation in Excel already can preserve arbitrary metadata from external formats even when it can't interpret it. All they'd've had to do is not touch things the user hadn't edited and the preservation would've happened automatically. I do this all the time when dealing with XML code, to the point where I have to make a deliberate effort not to write data-preserving code.
Already done, spreadsheet formulas are being specifically addressed in ODF 1.2. But in 1.1 there was already a set of conventions for handling formulas, and Microsoft were the only ones out of all the ODF 1.1-using applications that couldn't follow those conventions. In fact their implementation even specifically violated one of the bits that was in the ODF 1.1 spec: the spec calls for cell names to be enclosed in square brackets, while Microsoft's implementation omits the brackets. Then you have just plain malicious stuff like actively removing formula information that's present. Even if you can't parse the formulas, XML makes it easy to preserve what was there. Every other implementation behaves that way: if they can't understand the formulas at least they leave them intact for applications that do understand them. Microsoft's is the only implementation that deliberately removes formulas from the spreadsheet.
What annoys me most about Microsoft's pseudo-support is that it had to be deliberate. They had to actually expend additional effort to be this incompatible. If they'd simply been lazy and taken the easiest way out, they would've been far more compatible with everybody else than they ended up being.
If I were IBM, I'd be looking back at my own portfolio. IBM's been doing pay-to-unlock on their mainframe OSes since the 60s. They've even been doing it with their hardware. And I'd imagine they've got at least a couple of patents related to this stuffed away in their files.
If you hit Google, for instance, you'll find at least 3 Todd Knarrs out there. What's an ISP to do when they receive a complaint from one of them claiming that another's using his name? And I'd add that Google isn't complete, I know of at least a 4th Todd Knarr who doesn't show up. Names aren't unique identifiers.
I sum it up as "One in a million? That means there's 250 like me in the US alone.".
And that's exactly it: most newspapers and news sites nowadays aren't doing any of that. They whine about Google News aggregating links to them, but they themselves do little more than aggregate and reprint verbatim the same old AP wire stories (or other stories gotten from other outlets). That's a losing proposition. There's too many people doing exactly the same thing and at least some of them'll have a reason to make the articles freely visible, so as soon as a site puts up any kind of barrier to freely reading the articles readers simply find the same commodity cheaper elsewhere. Commodities are always a low-margin business and branding's pretty much irrelevant. Readers who're interested in more than a reprint of a wire-service story, OTOH, aren't going to other news sites or to news aggregators, they're using Google search to find the original sources and reading those instead. Even as little as ten or fifteen years ago news sites could get away with that, but not now when it's trivial for me to get access to news outlets world-wide and not very hard to track down the original sources by Web search.
As far as aggregating themselves, the problem there is they've got to compete with Google News. And Google's doing it for free, while they're trying to make people pay for it. Google's simply got too great an advantage with their search technology. Google's simply better at aggregating news than the newspapers and news sites are. All they can do is try and block access to the source of the stories, and there's just no way they can get every news site in the entire world to put all AP wire stories behind a pay-wall. If even a handful of news sites make the stories freely viewable, the whole thing collapses as readers go for the cheapest source of the commodity. And AP wire stories are a commodity: it's the exact same story no matter where you read it.
My feeling is that "journalism" on the net is headed more in the direction of Groklaw than of CNN or newspaper web sites. Groklaw does for the various SCO legal cases what reporters ought to be doing in general: pulling together all the information, researching the background, providing the analysis of why things are the way they are, showing in detail where and how the claims of the various parties match up with or contradict reality. That's what newspapers and mainstream news sites simply aren't doing anymore. And if they don't start doing it, amateurs like PJ will simply steamroll them.
I'm afraid it's not Google killing the news sites. It's the Internet itself. The Internet made it possible for anybody who wants to to publish cheaply and get read by a world-wide audience, and in the process killed the mere reporting of news as a paying job.
Why should I go to a news site to read a reprint of a press release from a company when I can go to that company's own Web site and read the original press release? Why should I read a news report of the latest scientific breakthrough when I can go to the scientist's own site and read his own paper on it? Why should I read the news reports of a disaster when I can go to the Twitter feeds and Livejournals of people who're actually there and read their first-hand reports, or go to the web sites of the emergency-services agencies in the area and read their updates on the situation? And in all of those cases, those first-hand sources aren't in the business of reporting news. They don't particularly care whether they get paid for generating their content, they've got other reasons of their own for wanting that content visible. And, as in so many things, the Internet's making it harder and harder for those middlemen whose business model is to get between the source of something and the eventual consumer and charge for transferring that something from the source to the destination.
Now, news sites aren't doomed. But to survive they're going to have to do something more than just report the news. They're going to have to start pulling together many sources of different information, analyzing all of it and putting together the pieces that it isn't immediately obvious fit together. Of course, that's going to be kind of hard seeing as they've spent the last decade or so wiping all traces of that out of their organizations because investigative journalism of any quality doesn't produce the Holy ROI.
Repeat after me: mirroring is not a backup. Backups are physically removed from the machine and stored where they can't be altered until they're needed for a restore. If they aren't removed from the machine, well, as we've just seen that only ends in tears. Observe their pain and learn from it!
Well, there are two that raise questions with me immediately. First, where the specifications called for 4 readings to be averaged, the code uses an exponential moving average instead. That's bad because the moving average weights the last sample at 50% of the final result, while the purpose of averaging is to give all samples equal weight so the inherent errors in each sample will (usually) cancel each other somewhat out making the error of the average less than the error of each individual sample. At the very least it indicates the programmers didn't understand the problem they were being asked to solve. Second, the reduction of the sample resolution from 12 bits (4096 values) to 3 bits (8 values) is strange. That alone introduces a fairly large error into the results just by squashing 256 distinct sample values into one. That means that a reading of 1 (negligible) on the sensor would yield the same result on the display as a reading of 511 (not negligible anymore), and that the device can only report 8 BAC readings to cover the entire range it purports to measure.
Combined with the large number of warnings from lint, that kind of thing is exactly what'd make me very cautions about a codebase. Lots of small bad habits in the code combined with obvious indications the programmers either didn't quite understand what they were doing or weren't at all careful about making sure what they did was what they intended spells waist-deep morass of bugs to me.
No, no more than your browser already monitors how often you visit each site. All ABP does is query your browser's history list for how many times the site appears in the last N days. The result doesn't leave the plugin, it's only used to control whether to bring up the pop-up or not.
And no, ABP isn't negotiating with advertisers. They're considering parsing for a particular mega tag, which sites are free to put on their pages at will. Note that the advertiser doesn't get to control whether the flag's present, it occurs on the site's page and not in the advertiser's content.
So no, the combination doesn't alarm me at all. The whole thing may end up being pointless, but I don't see anything dangerous here.
We've all seen the cases where the big media companies have been caught using people's copyrighted content without permission. There's the case just recently where Sarkozy's own party got caught at it. So, if they want three-strikes, give it to them. If you see one of their political parties using your content without permission, report them. If you find one of the big media companies there using your content without permission, report them. And demand, loudly and publicly, that the law they were so bound and determined to get, that they so loudly demanded, be followed to the letter. If it's "three strikes and you're out", then it's three strikes and they're out too.
Hardly. If you read the proposal, you'll notice that even when the flag's present ABP will not present ads by default. And it won't even immediately prompt you, let alone prompt you every time. It first checks whether you visit the site often. If it sees repeated visits recently, then it brings up a bar at the bottom giving you three options: "Let me see how the site looks with ads.", "Keep blocking the ads and don't ask me about this site ever again." and "Keep blocking the ads, but ask me about it next time it qualifies.". If you choose to see how it looks, then you get the site with ads and two options: "Add an exception for this site." and "Keep blocking ads for this site.". So ABP's never, even with the tag, going to allow ads through by default. And with the repeat-visitor logic, it shouldn't even be popping up the question bar too often (unless you keep using the "Ask me later." option).
I'd prefer it to unblock by service (eg. let me tell it "Allow Google AdSense text-only ads through regardless of site."), but as it stands the proposal is hardly a neutering of ABP in any way.
I'm fine with that, as long as there's a setting to control whether or not to honor the flag. I want the option of saying "No, if I want ads to not be blocked I'll add an exception for that site myself so don't bother bringing up the dialog.". I note that there's already an option to disable ad blocking for the page or the whole site in the right-click menu of ABP's icon, so an easy way to add an exception's already in place.
Re:No, no, no... Did I mention "No"?
on
The "Dangers" of Free
·
· Score: 3, Insightful
No, he does grasp one essential point: the bills have to be paid. Whatever you're producing, there's costs that you've got to have the money to cover. Utility bills, payroll, taxes, cost of materials, it all takes money and you need to come up with that money from somewhere. Either you're funding the whole thing out of your savings, or you need to find a way to earn revenue from the project. And if you intend to give away your product for free, then you'd better know what other source you're going to get revenue from or you'll be finding your bank account emptied at an alarming rate and when it hits zero the bank won't let you write any more checks no matter how many you've still got in your checkbook.
Yes, we as consumers of the free product don't care about any of that. But the guy producing the product had better care, because the bills still need to be paid.
Yes, but the OS you're referring to isn't one where you have to pay for the released version. Whether you like it or not, it's Microsoft's prerogative to charge for their OS. Since they're not charging for the RC, it being testing code and all, it's not reasonable to expect them to just have it automagically upgrade to the released version without you having to pay for it.
Slashdot Mirror
You miss the point of steganography. Encryption assumes that it's acceptable for an attacker to know there's a communications channel, the requirement is to keep the attacker from finding out the contents of the channel. Steganography is intended to conceal the very existence of the communications channel from a potential attacker.
Consider the situation a dissident in China might be in. Merely concealing what he's posting won't help him. The government doesn't care what the content is, the mere fact that he's hiding it from them's enough to convict him as far as they're concerned. For him encryption isn't required, an encrypted message the government can't read is just as damning as the plaintext would be. What he needs is a channel that's so unobtrusive that the government doesn't realize he's posting anything at all. And if the government doesn't realize there's a message there, they aren't even going to try to read it.
Hmm. Infrared and visible-light pulses. No electricity. Needs to have a message encoded on-the-fly in the field. That sounds an awful lot like a small lantern with an oil wick or candle burning in it and a shutter to conceal or reveal the light/heat source at will. We could even use Morse code. All you'd need is an ignition source to light the lamp. Matches would work, or steel-and-flint to light a small piece of tinder scavenged from the surroundings.
This shouldn't be allowed. Public notices in newspapers serve two purposes. The first is the one mentioned, publishing the notice where interested parties can see it. The second isn't mentioned, though, and that's to create a record of the notice outside the control of the party required to post it. The notice can't be changed later, can't be quietly made to never have happened. We've already seen entities change stories posted on their Web sites when what was in those stories became inconvenient later. Yes, it's going to cost a little extra to maintain that independent record of the notices. When we make a big payment or an important one where not making it has big consequences, it definitely costs for them to give us a receipt that we can use later to prove we did pay and what we paid for. We don't accept the cost savings as a valid reason for not being given a receipt, we don't accept "Trust us, we've got a record of your payment.".
Well, I suppose I can be considered a newcomer. I've only been programming professionally for 25 years, I came in after microcomputers came on the scene (although I do pre-date Microsoft itself). I've worked with Xerces, libxml2, Rogue Wave's XML parse, Microsoft's built-in parser, Java's XML packages and a couple of home-brew monstrosities, and I do suggest you go look at their APIs before you go claiming they're radically different.
As for your objections, I'll just take point A as a case study. If ODF 1.1 is so incapable of storing formulas, then how do OpenOffice, KOffice, Google Docs and Symphony all manage to store spreadsheet formulas in it and manage to understand each other's formulas and have theirs be understood in turn? Certainly it may not be possible the way Microsoft would like to do it, but that's like saying it's impossible to drive a car with an automatic transmission because there's no clutch pedal. And yes, Symphony currently does manage to not mess it up. Your confusion becomes apparent in point E, as OOo hasn't used 1.1 in a while (versions since at least 2.4 have used ODF 1.2). You appear to be working with very old versions, which is likely the cause of your failures.
Up-front costs. Yes, it's a better deal to have them pressed yourself, but you have to shell out that $999 up front. At $8.98 each you have to sell 112 copies just to cover manufacturing costs, and then there's packaging and shipping and storage costs. With Amazon, you give up some potential profit but you also don't have to shell out that 1 grand up front. If you aren't sure you'll sell 112 copies, or at least won't sell that many quickly, Amazon's deal may be a good bargain.
One thing: an existing contract. That's how the labels mostly do it. They don't go after the big-name bands. They go after the small bands who're just getting started, and get them into a contract for a certain number of albums. Note that that's for a certain number released, not produced, and the label gets to decide whether an album will be released. That can be tempting because of the advance and access to studio facilities and backup performers and such that new bands can't always get on their own. And if the band is a hit and becomes a big-name band, well, there's still that contract. The label owns all the rights to songs done under the contract, so the band can't even perform their own works without paying the label. If they try releasing anything, the label can come after them because they owe those songs to the label under the contract. The only way out of this once you're in it is to be a big enough hit that the label winds up having to release enough albums for you to complete your contract with them.
One reason the labels can do this is that they've managed to pretty well disrupt any legitimate distribution channel for music that doesn't involve physical production of media. The labels control the distribution chain to stores, so even if a band can afford to press and package their own CDs they have a hard time getting them on shelves without signing with a label. Electronic channels where the band might be able to make money, the labels have either sued out of existence or saddled with such onerous restrictions that consumers aren't willing to deal with it. The only avenue left is for a band to set up it's own Web site with it's own on-line store. That depends on search services like Google to give consumers a way to easily find the band's store, and you'll notice the recent attacks on Google's ability to link to things. Make no mistake, those attacks aren't aimed at keeping infringing material off Google's link lists. You'll notice, for instance, that in the Copiepresse case and others, when Google offered to simply stop listing the plaintiff's material completely the plaintiffs objected vociferously, demanding that Google not be allowed to do that. Their goal isn't to get their material off Google, it's to make it so difficult and expensive that Google can't stay in business at all. Search services are a critical link in an electronic distribution channel that completely bypasses the labels, and they want to disrupt them before they fully form.
In an RIAA contract, the artist's looking at maybe a 5-8% nominal on sales.
Note that the artist doesn't get paid that much. There's any advance, studio time, promotional fees, packaging fees, distribution fees, breakage allowance, all sorts of charges the artist is responsible for that come out of gross royalties. The artist gets paid what's left after the label takes all of those out. Most artists are lucky to see 1-2% after all that.
You see now why the labels are so dead-set against any legal electronic distribution of music.
If you didn't find the links, you weren't trying very hard.
Like Microsoft, you're hopelessly confused about the difference between external and internal representations. It's rather trivial to attach arbitrary XML DOM elements to C++ objects. After all, those XML elements are, when the parser gets done with them, just C++ objects themselves. ODF even makes it easy, since formulas are stored in attributes you don't even need to handle the full general case of nested elements to get it right. A simple linked list of unrecognized attributes (which is what the parser hands you on a silver platter) is all you need. Anybody with more than 5 or so years experience had probably already done that or the equivalent several times, and I'm positive Microsoft doesn't put people with less than 5 years experience in lead designer slots. And that's not even counting the fact that they already had done it right. That leaves only one other option: Microsoft's programmers were fully capable of supporting ODF spreadsheet formulas in a compatible way, and they were instructed to break them.
Someone needs to remind Mr. Yee that, at least in all the cases I've heard reported on, the store didn't sell the video game to the kid. They sold it to an adult relative of the kid, who then gave it to the kid without bothering to check on what exactly their "little angel" had been bugging them for. And then when they found out exactly what little Timmy had gotten, they dove headfirst into that river in Africa and started looking around for someone else to take the blame for their failure. No law about selling video games to minors will do a single blessed thing about that, where there's no video game ever sold to the minor.
There's a lot more than 2 implementations. Besides OpenOffice and MS Office there's AbiWord, KOffice, Google Docs, WordPerfect Office X4, IBM's Lotus Symphony, the Sun ODF plug-in for MS Word and the BSD-licensed ODF plug-in for Word that Microsoft funded and hosted on SourceForge. That last is important, BTW. Not only is Office 2007's implementation of ODF incompatible with OpenOffice, it's incompatible with Microsoft's own other implementation of ODF.
Pretty much, yes. Bear in mind that Microsoft already has code that does handle the spreadsheet formulas correctly. The plug-in that Microsoft itself commissioned and that they own the code for not only preserves the formulas, it correctly parses and interprets them so that cells get recalculated properly as data changes and it correctly writes changed formulas back out. All Microsoft had to do was to not do all the work a second time. And even if they had re-done the work, the XML parser automatically populates the DOM with the formula strings and the internal implementation in Excel already can preserve arbitrary metadata from external formats even when it can't interpret it. All they'd've had to do is not touch things the user hadn't edited and the preservation would've happened automatically. I do this all the time when dealing with XML code, to the point where I have to make a deliberate effort not to write data-preserving code.
Already done, spreadsheet formulas are being specifically addressed in ODF 1.2. But in 1.1 there was already a set of conventions for handling formulas, and Microsoft were the only ones out of all the ODF 1.1-using applications that couldn't follow those conventions. In fact their implementation even specifically violated one of the bits that was in the ODF 1.1 spec: the spec calls for cell names to be enclosed in square brackets, while Microsoft's implementation omits the brackets. Then you have just plain malicious stuff like actively removing formula information that's present. Even if you can't parse the formulas, XML makes it easy to preserve what was there. Every other implementation behaves that way: if they can't understand the formulas at least they leave them intact for applications that do understand them. Microsoft's is the only implementation that deliberately removes formulas from the spreadsheet.
What annoys me most about Microsoft's pseudo-support is that it had to be deliberate. They had to actually expend additional effort to be this incompatible. If they'd simply been lazy and taken the easiest way out, they would've been far more compatible with everybody else than they ended up being.
If I were IBM, I'd be looking back at my own portfolio. IBM's been doing pay-to-unlock on their mainframe OSes since the 60s. They've even been doing it with their hardware. And I'd imagine they've got at least a couple of patents related to this stuffed away in their files.
If you hit Google, for instance, you'll find at least 3 Todd Knarrs out there. What's an ISP to do when they receive a complaint from one of them claiming that another's using his name? And I'd add that Google isn't complete, I know of at least a 4th Todd Knarr who doesn't show up. Names aren't unique identifiers.
I sum it up as "One in a million? That means there's 250 like me in the US alone.".
And that's exactly it: most newspapers and news sites nowadays aren't doing any of that. They whine about Google News aggregating links to them, but they themselves do little more than aggregate and reprint verbatim the same old AP wire stories (or other stories gotten from other outlets). That's a losing proposition. There's too many people doing exactly the same thing and at least some of them'll have a reason to make the articles freely visible, so as soon as a site puts up any kind of barrier to freely reading the articles readers simply find the same commodity cheaper elsewhere. Commodities are always a low-margin business and branding's pretty much irrelevant. Readers who're interested in more than a reprint of a wire-service story, OTOH, aren't going to other news sites or to news aggregators, they're using Google search to find the original sources and reading those instead. Even as little as ten or fifteen years ago news sites could get away with that, but not now when it's trivial for me to get access to news outlets world-wide and not very hard to track down the original sources by Web search.
As far as aggregating themselves, the problem there is they've got to compete with Google News. And Google's doing it for free, while they're trying to make people pay for it. Google's simply got too great an advantage with their search technology. Google's simply better at aggregating news than the newspapers and news sites are. All they can do is try and block access to the source of the stories, and there's just no way they can get every news site in the entire world to put all AP wire stories behind a pay-wall. If even a handful of news sites make the stories freely viewable, the whole thing collapses as readers go for the cheapest source of the commodity. And AP wire stories are a commodity: it's the exact same story no matter where you read it.
My feeling is that "journalism" on the net is headed more in the direction of Groklaw than of CNN or newspaper web sites. Groklaw does for the various SCO legal cases what reporters ought to be doing in general: pulling together all the information, researching the background, providing the analysis of why things are the way they are, showing in detail where and how the claims of the various parties match up with or contradict reality. That's what newspapers and mainstream news sites simply aren't doing anymore. And if they don't start doing it, amateurs like PJ will simply steamroll them.
I'm afraid it's not Google killing the news sites. It's the Internet itself. The Internet made it possible for anybody who wants to to publish cheaply and get read by a world-wide audience, and in the process killed the mere reporting of news as a paying job.
Why should I go to a news site to read a reprint of a press release from a company when I can go to that company's own Web site and read the original press release? Why should I read a news report of the latest scientific breakthrough when I can go to the scientist's own site and read his own paper on it? Why should I read the news reports of a disaster when I can go to the Twitter feeds and Livejournals of people who're actually there and read their first-hand reports, or go to the web sites of the emergency-services agencies in the area and read their updates on the situation? And in all of those cases, those first-hand sources aren't in the business of reporting news. They don't particularly care whether they get paid for generating their content, they've got other reasons of their own for wanting that content visible. And, as in so many things, the Internet's making it harder and harder for those middlemen whose business model is to get between the source of something and the eventual consumer and charge for transferring that something from the source to the destination.
Now, news sites aren't doomed. But to survive they're going to have to do something more than just report the news. They're going to have to start pulling together many sources of different information, analyzing all of it and putting together the pieces that it isn't immediately obvious fit together. Of course, that's going to be kind of hard seeing as they've spent the last decade or so wiping all traces of that out of their organizations because investigative journalism of any quality doesn't produce the Holy ROI.
Repeat after me: mirroring is not a backup. Backups are physically removed from the machine and stored where they can't be altered until they're needed for a restore. If they aren't removed from the machine, well, as we've just seen that only ends in tears. Observe their pain and learn from it!
Well, there are two that raise questions with me immediately. First, where the specifications called for 4 readings to be averaged, the code uses an exponential moving average instead. That's bad because the moving average weights the last sample at 50% of the final result, while the purpose of averaging is to give all samples equal weight so the inherent errors in each sample will (usually) cancel each other somewhat out making the error of the average less than the error of each individual sample. At the very least it indicates the programmers didn't understand the problem they were being asked to solve. Second, the reduction of the sample resolution from 12 bits (4096 values) to 3 bits (8 values) is strange. That alone introduces a fairly large error into the results just by squashing 256 distinct sample values into one. That means that a reading of 1 (negligible) on the sensor would yield the same result on the display as a reading of 511 (not negligible anymore), and that the device can only report 8 BAC readings to cover the entire range it purports to measure.
Combined with the large number of warnings from lint, that kind of thing is exactly what'd make me very cautions about a codebase. Lots of small bad habits in the code combined with obvious indications the programmers either didn't quite understand what they were doing or weren't at all careful about making sure what they did was what they intended spells waist-deep morass of bugs to me.
No, no more than your browser already monitors how often you visit each site. All ABP does is query your browser's history list for how many times the site appears in the last N days. The result doesn't leave the plugin, it's only used to control whether to bring up the pop-up or not.
And no, ABP isn't negotiating with advertisers. They're considering parsing for a particular mega tag, which sites are free to put on their pages at will. Note that the advertiser doesn't get to control whether the flag's present, it occurs on the site's page and not in the advertiser's content.
So no, the combination doesn't alarm me at all. The whole thing may end up being pointless, but I don't see anything dangerous here.
We've all seen the cases where the big media companies have been caught using people's copyrighted content without permission. There's the case just recently where Sarkozy's own party got caught at it. So, if they want three-strikes, give it to them. If you see one of their political parties using your content without permission, report them. If you find one of the big media companies there using your content without permission, report them. And demand, loudly and publicly, that the law they were so bound and determined to get, that they so loudly demanded, be followed to the letter. If it's "three strikes and you're out", then it's three strikes and they're out too.
Hardly. If you read the proposal, you'll notice that even when the flag's present ABP will not present ads by default. And it won't even immediately prompt you, let alone prompt you every time. It first checks whether you visit the site often. If it sees repeated visits recently, then it brings up a bar at the bottom giving you three options: "Let me see how the site looks with ads.", "Keep blocking the ads and don't ask me about this site ever again." and "Keep blocking the ads, but ask me about it next time it qualifies.". If you choose to see how it looks, then you get the site with ads and two options: "Add an exception for this site." and "Keep blocking ads for this site.". So ABP's never, even with the tag, going to allow ads through by default. And with the repeat-visitor logic, it shouldn't even be popping up the question bar too often (unless you keep using the "Ask me later." option).
I'd prefer it to unblock by service (eg. let me tell it "Allow Google AdSense text-only ads through regardless of site."), but as it stands the proposal is hardly a neutering of ABP in any way.
I'm fine with that, as long as there's a setting to control whether or not to honor the flag. I want the option of saying "No, if I want ads to not be blocked I'll add an exception for that site myself so don't bother bringing up the dialog.". I note that there's already an option to disable ad blocking for the page or the whole site in the right-click menu of ABP's icon, so an easy way to add an exception's already in place.
No, he does grasp one essential point: the bills have to be paid. Whatever you're producing, there's costs that you've got to have the money to cover. Utility bills, payroll, taxes, cost of materials, it all takes money and you need to come up with that money from somewhere. Either you're funding the whole thing out of your savings, or you need to find a way to earn revenue from the project. And if you intend to give away your product for free, then you'd better know what other source you're going to get revenue from or you'll be finding your bank account emptied at an alarming rate and when it hits zero the bank won't let you write any more checks no matter how many you've still got in your checkbook.
Yes, we as consumers of the free product don't care about any of that. But the guy producing the product had better care, because the bills still need to be paid.
Yes, but the OS you're referring to isn't one where you have to pay for the released version. Whether you like it or not, it's Microsoft's prerogative to charge for their OS. Since they're not charging for the RC, it being testing code and all, it's not reasonable to expect them to just have it automagically upgrade to the released version without you having to pay for it.