LSL always had a "not adult" mode too, just so they could sell to kids. One more reason to know just what's going on your computer, especially if you lend it to a kid.
How much of the outcry wasn't due to the security testing, but to the fact that most software/hardware tends to be untested at the levels a proper network security tests exerts it?
(i.e. how much of it is due to the fact that the machine might be unavailable to regular users during the test, because a service might crash, and not the impact of testing itself?)
It's a pretty well known anecdote that there are some network vulnerability tests that will find "vulnerable" machines... Those vulnerable machines blue-screening, as a matter of fact.
On a related note, you have your vulnerability tests down path, but what's your strategy if you DO find vulnerable machines? Will you wait till the next maintenance window? Apply the fix immediately? Do you even know which vulnerabilities you're scanning for? Did you scan security focus for your os/application/hardware mix beforehand, to get some feel for what might go wrong?
clarifying the tinfoil hat's position...
on
Privacy in the Woods?
·
· Score: 2, Interesting
Let's see, I'm not a card-carrying member of the tinfoil hat brigade, but I do think privacy is underrated. Let's see just how your system can do wrong, and maybe you can get it right(ok so I'm an IDEALIST member of the tinfoil hat brigade)
Annoying safety features that conflict with privacy: 1) thinking your privacy is being respected, when it's not: Make sure you label clearly EVERYWHERE anyone's position is being monitored, that includes the most remote areas that can be monitored, put your guardhouses/first responders barracks close to the points that AREN'T monitored, as the teenagers who are looking for a thrill will look for the out of the way spots first
2) Having someone trusted with my privacy that abuses it: Make sure all your employees know that using the data on your hikers for any reason but those that fit on the signs your applied to conform to 1) will result in rightful termination of their employment, and expose them to criminal prosecution
Your hikers will 1) be warned they are being watched 2) not fear stalkers/other criminals hiding among your staff/otherwise abusing the privacy data that belongs to your hikers...
If you can get a 100% rate of respect for those rules, you should be ok.
You have to be the worse sort of dreamer to think you can get even 90% on the second one, but I stress, 100% is the only thing that will not harm your park in the worse way. Any breach of two will mean: "They don't watch to prevent people from dying at park XYZ, they watch to supply children to child prostitution ring ABC, avoid it like the plague."
Yes, it sounds hysterical, in this case, hysteria is your friend. If hystericals can't find fault with you, you're safe, otherwise, plan better.
The tinfoil hats are there because "someone in power" has the power to abuse, and too little reason not to... If you can give him a big enough reason, maybe I'll remove my own tinfoil hat.
No mention of data=writeback, or any other optimisation tweaks, however. Kinda sad. The article is nice, the graphs are.... Err too much of a good thing? And basically the results just reiterate the design imperatives of each filesystem(how unsurprising!)
- ext2 predates them all - ext3 is a low-impact, let's reuse what we know as much as possible, kinda file system - reiser's b-trees reflect it's "once we put the data in, how do we find it again" orientation - XFS was at least at one point, designed for "Media" files(think renderfarms), aka LARGE files, much of the benchmarks it won were on such files, although its design was also influenced by large-scale server needs(a renderfarm is a large-scale server cluster right?) - JFS was influenced by large-scale server needs(databases), but tampered by OS/2's needs, and other systems, resulting in a filesystem that's a bit more nimble than XFS, but less handy with huge files(normal, since databases try to use raw-io if necessary on huge files, unlike render clusters)
I think this demonstrates the implications of early design imperatives on long-term software trends. XFS and JFS were developed for other platforms and ported to linux, yet notice how they can't really change their strengths(good thing too!).
Anyone try the same benchmark on the 2.6 kernel just to contrast it? Wouldn't the new IO-system help to mitigate those weird ext2/ext3 slowdowns the article mentions, but doesn't explain?
That also raises the question of the legal status of an international not-for-profit ad-hoc group meant to develop software
The individual getting extra income isn't always clear(say a project of ten people, who agree only two should receive the product of the donations)... Think back on how perl "hires" Larry Wall and other developers... That's a tricky situation methinks. Because the donations go to the project(who is theoretically taxable) who gives it back, without making a dime, to members.
paraphrased summary of the comic: -"I want to ask why you remove beloved feature XYZ" -"We don't support it anymore" -"Was it buggy?" -"No we just don't support it anymore" -"But it worked fine!" *embarassed silence* -"That's why you removed feature XYZ isn't it?" -"Thank you for calling ubersoft tech support" linkfor the curious.
Not sure about their other hardware, but it seems my Microsoft mouse works just fine, it works too well in fact, I don't notice it's Microsoft, and I have little to no brand-awareness(I picked it because it worked, without any drivers but what came with the oses I use, I also resist downloading any "custom" drivers that would make my Microsoft experience more Microsoft-y). Maybe that's why? Could it be that they tried to use their routers to increase sales/free(for now) use of some other Microsoft product/service and were disappointed by the result? (MSN perhaps?)
Microsoft doesn't need to explain why it's system is better with the browser integrated into everything, everyone takes it as fact(or debunks it at myth) Why treat mozilla differently?
No seriously, I imagine the goal is that since mozilla is cross-platform and has a bunch of nifty features, a full-blown desktop written in it would be able to compete with java's desktop system for thin clients and similar ideas(probably with great success, as while Mozilla itself is fairly large, it's also quite a capable system, and fairly self-contained).
It has many features modern thin clients would need or at the very least, like to have(software updates downloaded from the web, ssl/tls based security, multiple user profiles), it supports most "thin clients" activities except for document production(by itself: the ibm-related announcement on slashdot today, about a web-available office suite makes that a non-issue) With the proper XUL environment available, you have almost an os-toolkit, themable/skinnable for those so enclined... What more could you want? (Yes you need an OS under it, but at least, you're not limited to the choice of any particular one)
I'd also sugges that the fact that there is several MPMs, while beneficial in the long run, is slowing down apache 2.0 acceptance.
Think about it, all the third party modules had to support one apache(1.3) now they have to be thread-safe, or recommend that you use prefork. Prefork is ok, but most of the hype that was generated when apache 2.0 was launched was for the other mpms (perchild is an especially nice idea) but most third-parties have either never heard of them, or decide not to support them. Now if you run a server for other people and take requests for apache modules(think "good" virtual hosters) that means prefork is the only acceptable MPM, as it's the most compatible. Prefork is approximately as performant as 1.3 for a lot of loads, but heavier in some cases, and is less tested. Until more of the extant code gets thread-safety(off-topic question: is slash thread-safe?) that means unless your hardware DOES run faster with 2.0, why change? And of course, most third party apps developers are in no hurry to invest the time to thread-check their apps, if noone is using 2.0 with worker. It's a chicken and egg problem.
In an interactive situation(like a game), being in a familiar setting/knowing familiar characters/objects is an advantage, as it reinforces the "I can do this" feeling, important as you start a game. This is especially visible in games with online communities, where each player can build something, either his reputation(most combat games), or some sort of character(Diablo II), or a collection of objects(The Sims and similar games).
Most sequels/expansion packs allow you to improve what you've already built. And since expansion packs reuse game engines, your investment in the expansion usually comes either a little cheaper(you pay less for an expansion) or you get more game for your money(you get the game, AND the expansion, which is less than two games, usually) Watching a movie isn't a "challenge" except for the odd movie about investigation(where the plot is the challenge, but your actions aren't as much your responsability as in a game(you connect plot elements you're given, but you can't see them in more detail, etc...), and your investment is always a full movie ticket... You don't get a "frequent movie fan" type ticket that costs less if you went to see the other movie. Games also get played longer(not many people still go to a movie that came out the time diablo ii was released, yet the online us east item trading is still doing brisk business)
So basically, it's easier to make a quality game sequel from the point of view of the consumer(who gets value from owning two games) than from the producer(who gets a little bit less from an expansion than a truly new game, but also can find ways to invest less).
Natural, in this case, is opposed to artificial which generally denotes the product of willful activity of sentient beings. Humans use "natural" to mean "without conscious thought" in many cases and "what these other humans didn't do" in many others.
The fact that nature ignores many of our willful changes(because it has inertia, and the power granted by being a system in instable equilibrium for the best part of 10E5 years does not mean we WANT nature to regulate against us. It doesn't mean nature can't adjust to our changes, it means we might not like the consequences.
Global warming means the overall temp might go up, but with the weather being a dynamic system, it might only mean that the temps at the poles rise up 10 degrees, and the equator and tropics go down 2 degrees, and guess what, it's warmer, globally, but more people are cold, because less people live in the Arctic Circle than in Quito, Ecuador, despite the fact that Quito is a bit smaller. I haven't gotten into altitude either... It's interesting that the article refers to the surface temperature of the "rock" we live on, as it might be a long term effect of the warming of the air on top of it.
It might also be related to the magnetic shifts covered in other slashdot posts(geologic events that take millenia, and could share a cause with the ice ages[it would be surprising, but it's still possible]) and yet the article speaks of the evolution over decades.
The problem with the climate is that we have trillions of life forms who all contribute part of the problem(breathing) and the solution(photosyntheis) to global warming. But approximately 6 billion of the actors are going on strike, saying we want to be a bigger part of the problem. While the other actors just watch, shake their head, and hope they don't get hit by the backlash of whatever "Correction" to use the economics term, the "gas balance" of the planet will apply.
After all, that is a self-regulating equilibrium people "tend" to understand a (admittedly little) better.
Being regulated by your equilibrium hurts... Much better to regulate yourself, when you can.
Re:because it's an ugly, lumbering dinosaur
on
Postfix 2.1 Released
·
· Score: 1
I believe he meant the delivery time, not the transition. Although postfix is meant to be a drop-in replacement for sendmail, and many facilities are provided to make the transition as smooth, easy and painless as possible.
On redhat, from sendmail to postfix, the transition DOES take five minutes... Provided you don't have thousands of special cases...
That's certainly something I'd be worried about...
A universal format has to be used everywhere(in theory) yet most games(for instance) avoid anything that isn't uber-optimized for a specific "target" which in many cases translates to a subset of one or two platforms.
Is U3D going to be a standard for ray tracing/modeling? For 3d games? For 3d web environments? Only for static objects, or also for movement?
For all of the above? How will it perform well on environments which are so very disparate? (3d movement modeling can have much tighter tolerances, for instance, than 3d web environments, in terms of precision.)
Will it be like opengl was/used to be, yet patent-free? Or QuickTime 3d?
Or are we talking about another beast entirely?
It's interesting that the article already wants an "extensible" format, yet doesn't say the extensions themselves have to be covered by the standard. (theoretical conversion, using a lame analogy) -"You can have a graphic 3d format" -"Cool! How big are the files?" -"Very small if you use the compression extension" -"How do I get that?" -"3000$/display node" -"What? but this is a standard for cripes sake!" -"Yes it is, but nothing in the standard says anything about compression, that's covered by an extension, so you either take 1Gb files, or you pay for compression"
Let's keep things straight, SPAM isn't "any unwanted email" it's UCE. Unwanted email is probably already outnumbering wanted email. But viruses are ALREADY illegal, so fudging them in with the spam, reduces the credibility of those who complain about spam, in lawmaker's eyes, who associate people who don't like spam with whiney people with no sense of discernment.
The article is about spam, which is probably reducing its "inbox percentage of total emails received"(for people who don't have gateway-level virus filters) and increasing it's "inbox percentage"(for people who block those at the gateway level, and never see the viruses).
Lumping our enemies together is great, as long as you like them outnumbering us, a faceless myriad of enemies. If you want to fight them, we gotta categorize them, unanonymize them, and take em out, one at a time.
-- I still remember the internet before spam It was idyllic
Most readers will trade off accuracy for immediacy if that means they get the information "in time". By mentioning "chasing the chimaera of objectivity" it's pretty obvious that we aren't talking about "more accurate isn't ok if you're second", but more "we don't want to wait till the politician announces his resignation to discover he's crooked".
There are degrees of innacuracy involved. There's also the fact that for true objectivity to come from big media, they'd have to stop being large corporations beholden to shareholders and selling ad space to other large corporations...
Most people don't worry about objectivity from the (current) media because they trust that the competition is fierce enough that journalists ARE actively seeking scandals in competing news sources(at least, that's the plan) and the government. I'm sure they'd be disappointed by how much of that money is funneled through press agencies or spent chasing press releases.
webwasher, like most filtering software, uses static "blacklists", so they can't tell if someone really uses a single domain for multiple purposes, or if it's someone using the "not-so-raunchy" bits to do like spam, and intersperse "not blacklisted" material to contour the blacklist.
Unfortunately until some form of dynamic(updated to the minute or less: bandwidth costs for the filtering providers...) blacklisting of pages occur, this type of "block everything unless we know its good" will stay prevalent.
(some) execs of companies dream of patents they don't have to sink a few half-million in research for
The patent system is made to encourage them to sink those few millions.
By that logic, the companies in the first situation are abusing the system, if we had a way to know for sure, we (sh/w)ould take away the first companies' patents away.
I'm a consultant, paid to keep current on many topics, so on some topics, yes... It's however a case of "do more with the same money" instead of "do the same with less money" if that's what you mean.
I used to buy the animal books on several topics, mostly perl programming
Then I got the safari subscription imagine this: oreilly comes up with fourth edition of dns and bind I have paper third edition of dns and bind I use safari to get fourth edition, and I don't need the paper one anymore. Since a lot of the animal books I use are very sucessful, and get updated every so often, just because I can replace one edition with the next at no charge, I save a bundle of money, provided I don't need hardcopy of the work in question, the web interface to it might actually save me time(mostly searching, although with practice, the internal binary-page search is pretty damn hard to beat, it's the "read entire TOC" that takes a while.)
Of course, I've been known to read entire online volumes on topics I was less familiar with(I can't say I'd do it with something like the perl cookbook) but so far, Safari is working out for me.
That's why the patents process normally applies to technology, not science. But ever since patents started covering ideas, and not the technological items that represent them, tne line has been blurring... Software patents(and the dreaded algorithm patent sure to come out someday if the trend doesn't stop) further blur the distinction.
*steps on soapbox* now it appears to me that part of the problem with patents and copyright is that the public instinctively associate them with the inventor/discoverer/original author, and we're fine with them, until such rights are sold then we become not so fine with it. For one, it's a lot simpler/more instinctive to sort out who disovered something, if it's an individual making the discovery, not a corporation. In that same line of thought, what happens when those rights are transferred, and for instance, how to get the original author to unlearn what he sold, gets a lo more confusing. Trademarks, as a rule are purely corporate/marketing concepts and have far less contestation going on... They also have to be defended constantly... Copyrights are especially problematic, as copyright ownership gets regularly extended for corporations (technically, you could even view the legatees of an author as a corporation, THEY DIDN'T WRITE THE BOOK) to protect little mouses, and such... Yet the author in many cases is long dead... before the first lawyer gets paid... (Not that I object to heirs of authors and such to benefit from the work of their forebears... I do however insist on the fact that by that point, the original intent of copyright: to encourage the author to produce more work, because his rights are protected, will take a lot of doing to respect: he's dead, he won't write that many books anymore... You might say part of the problem is that the public recognizes a right to a discoverer/author as an individual, for a limited duration, but it becomes a lot less instinctive when those are transferred. Perhaps making them non-transferrable would keep the legislation on this topic, at least understandable by someone without a major in minor in intellectual property...
I'd like to contrast the American experience with the Canadian one, and maybe some contrasts will become apparent...
1) My cable provider has a-la-carte pricing, but only if you take the expensive digital rig (you pay more up-front, you pay less if you stick around for five years) you can also pick a bewildering array of packages(of course, confusion is their friend) 2) The pricing scheme for channels is heavily regulated, for several reasons, some of which follow: a) Part of the per-channel price(indeed any cable bill) goes to promote production of local content b) the CRTC regulates the final price, for consumer-protection, and to prevent many kinds of abuses c) not all per-channel fees are considered equal d) some channels are deemed to be worth more, for various reasons(some are actually one-fee for four channels worth of bandwidth under one theme) for example e) the linguistics aspect also is involved(somehow, I'm neither a lawyer, nor really inclined to learn so much of this, as it oftens to be more politics than law)
It certainly isn't a perfect system, but it works on the following premises: 1) the experts on the CRTC are supposed to know what the costs are 2) The satellite operators and cable operators are "passing the token" about who's unwilling not to charge consumers for unviewed channels, having regulated prices goes around the issue of "whose fault is it" Consider also that: channel operators fight so hard to get a license to have their own channels, they can't actually argue "well gee, we fought like lions to get this and now you take away our revenue stream!" when they actually have to justify that they will have audience once the channel operates, just to get the license to start their channel. 3) Many of the bundles include common mixes, for cheaper, but I did the math, and I found it almost impossible not to get some channels I didn't want along with channels I did... Choosing per-channel for example, was the best way to guarantee I could take the Space Channel along with the Z channel(the local, francophone techno-channel)
Now obviously, so far the system works, and the large, monolithic, much-more-profitable-in-densely-populated-urban-ar eas have their warts... But I think it's important to make the point that the "pay per channel" model is not the only important part of a workable scheme... The idea being is that customers in the ideal system don't pay for what they don't want, not that they pay for those they want...(Yes they are ALSO paying for those they want, but it's very easy to just tax for the unwatched channels by doubling the rate of heavily watched channels if there is no regulation/supervision) It's also noteworthy that they don't offer this choice without a substantial customer investment(the equipment is around 600$US total, perhaps, but I'm not sure how to deduct the programming credits, since they seem to apply only on popular bundles, which go against the grain of this article).
How is this relevant you say?
Well let's see: a good, customer-defending system has safeguards in place to protect customers, both from channel and cable-operator-spun fud(they are there to make a profit, and most of them DO own news-producing companies, so the media can hardly be considered neutral when they are concerned.) So I'd certainly suggest to any south of the border friends to consider enforcement of any safeguards before any "pay per use" schemes are applied "because they are cheaper" it's not really that simple. Without any bundles, consider this pricing: over 1$US(using approximate exchange rate) per channel, if you take no bundle at all, with a 500$+ setup fee Some of the bundles start at 15$US... But do they include what you want? That is the question, so unless you really need only some channels, it might not really be that much cheaper, unless you really aren't interested in the same thing as the suburbian 2.3 kids family next door. Now that just may be the case of many slashdotters, your milage may of course vary.
What makes it like apple is that their software doesn't work on other people's hardware. The Windows version you get from Dell also works(at least, the serial number would) on an HP.
If I may, read that certain carefully. I said their ownership of the platform CAUSED uncertainty, not that it was uncertain. If you read up the thread, you'll see that Sun's ownership was spoken of as a bad thing...
Hmm, Because most languages people think they "know" they know not as much as they think they do, and what they do know is often inefficient/misadapted to writing a large, platform-independant, network-distributed set of API's?
Slashdot Mirror
LSL always had a "not adult" mode too, just so they could sell to kids.
One more reason to know just what's going on your computer, especially if you lend it to a kid.
How much of the outcry wasn't due to the security testing, but to the fact that most software/hardware tends to be untested at the levels a proper network security tests exerts it?
(i.e. how much of it is due to the fact that the machine might be unavailable to regular users during the test, because a service might crash, and not the impact of testing itself?)
It's a pretty well known anecdote that there are some network vulnerability tests that will find "vulnerable" machines... Those vulnerable machines blue-screening, as a matter of fact.
On a related note, you have your vulnerability tests down path, but what's your strategy if you DO find vulnerable machines? Will you wait till the next maintenance window? Apply the fix immediately? Do you even know which vulnerabilities you're scanning for? Did you scan security focus for your os/application/hardware mix beforehand, to get some feel for what might go wrong?
Let's see, I'm not a card-carrying member of the tinfoil hat brigade, but I do think privacy is underrated. Let's see just how your system can do wrong, and maybe you can get it right(ok so I'm an IDEALIST member of the tinfoil hat brigade)
Annoying safety features that conflict with privacy:
1) thinking your privacy is being respected, when it's not:
Make sure you label clearly EVERYWHERE anyone's position is being monitored, that includes the most remote areas that can be monitored, put your guardhouses/first responders barracks close to the points that AREN'T monitored, as the teenagers who are looking for a thrill will look for the out of the way spots first
2) Having someone trusted with my privacy that abuses it:
Make sure all your employees know that using the data on your hikers for any reason but those that fit on the signs your applied to conform to 1) will result in rightful termination of their employment, and expose them to criminal prosecution
Your hikers will 1) be warned they are being watched
2) not fear stalkers/other criminals hiding among your staff/otherwise abusing the privacy data that belongs to your hikers...
If you can get a 100% rate of respect for those rules, you should be ok.
You have to be the worse sort of dreamer to think you can get even 90% on the second one, but I stress, 100% is the only thing that will not harm your park in the worse way. Any breach of two will mean: "They don't watch to prevent people from dying at park XYZ, they watch to supply children to child prostitution ring ABC, avoid it like the plague."
Yes, it sounds hysterical, in this case, hysteria is your friend. If hystericals can't find fault with you, you're safe, otherwise, plan better.
The tinfoil hats are there because "someone in power" has the power to abuse, and too little reason not to... If you can give him a big enough reason, maybe I'll remove my own tinfoil hat.
No mention of data=writeback, or any other optimisation tweaks, however. Kinda sad. The article is nice, the graphs are.... Err too much of a good thing?
And basically the results just reiterate the design imperatives of each filesystem(how unsurprising!)
- ext2 predates them all
- ext3 is a low-impact, let's reuse what we know as much as possible, kinda file system
- reiser's b-trees reflect it's "once we put the data in, how do we find it again" orientation
- XFS was at least at one point, designed for "Media" files(think renderfarms), aka LARGE files, much of the benchmarks it won were on such files, although its design was also influenced by large-scale server needs(a renderfarm is a large-scale server cluster right?)
- JFS was influenced by large-scale server needs(databases), but tampered by OS/2's needs, and other systems, resulting in a filesystem that's a bit more nimble than XFS, but less handy with huge files(normal, since databases try to use raw-io if necessary on huge files, unlike render clusters)
I think this demonstrates the implications of early design imperatives on long-term software trends. XFS and JFS were developed for other platforms and ported to linux, yet notice how they can't really change their strengths(good thing too!).
Anyone try the same benchmark on the 2.6 kernel just to contrast it? Wouldn't the new IO-system help to mitigate those weird ext2/ext3 slowdowns the article mentions, but doesn't explain?
That also raises the question of the legal status of an international not-for-profit ad-hoc group meant to develop software
The individual getting extra income isn't always clear(say a project of ten people, who agree only two should receive the product of the donations)... Think back on how perl "hires" Larry Wall and other developers... That's a tricky situation methinks. Because the donations go to the project(who is theoretically taxable) who gives it back, without making a dime, to members.
To quote the ubersoft.net webcomic:
paraphrased summary of the comic:
-"I want to ask why you remove beloved feature XYZ"
-"We don't support it anymore"
-"Was it buggy?"
-"No we just don't support it anymore"
-"But it worked fine!"
*embarassed silence*
-"That's why you removed feature XYZ isn't it?"
-"Thank you for calling ubersoft tech support"
linkfor the curious.
Not sure about their other hardware, but it seems my Microsoft mouse works just fine, it works too well in fact, I don't notice it's Microsoft, and I have little to no brand-awareness(I picked it because it worked, without any drivers but what came with the oses I use, I also resist downloading any "custom" drivers that would make my Microsoft experience more Microsoft-y). Maybe that's why? Could it be that they tried to use their routers to increase sales/free(for now) use of some other Microsoft product/service and were disappointed by the result? (MSN perhaps?)
Microsoft doesn't need to explain why it's system is better with the browser integrated into everything, everyone takes it as fact(or debunks it at myth)
Why treat mozilla differently?
No seriously, I imagine the goal is that since mozilla is cross-platform and has a bunch of nifty features, a full-blown desktop written in it would be able to compete with java's desktop system for thin clients and similar ideas(probably with great success, as while Mozilla itself is fairly large, it's also quite a capable system, and fairly self-contained).
It has many features modern thin clients would need or at the very least, like to have(software updates downloaded from the web, ssl/tls based security, multiple user profiles), it supports most "thin clients" activities except for document production(by itself: the ibm-related announcement on slashdot today, about a web-available office suite makes that a non-issue) With the proper XUL environment available, you have almost an os-toolkit, themable/skinnable for those so enclined... What more could you want? (Yes you need an OS under it, but at least, you're not limited to the choice of any particular one)
I'd also sugges that the fact that there is several MPMs, while beneficial in the long run, is slowing down apache 2.0 acceptance.
Think about it, all the third party modules had to support one apache(1.3) now they have to be thread-safe, or recommend that you use prefork. Prefork is ok, but most of the hype that was generated when apache 2.0 was launched was for the other mpms (perchild is an especially nice idea) but most third-parties have either never heard of them, or decide not to support them.
Now if you run a server for other people and take requests for apache modules(think "good" virtual hosters) that means prefork is the only acceptable MPM, as it's the most compatible. Prefork is approximately as performant as 1.3 for a lot of loads, but heavier in some cases, and is less tested. Until more of the extant code gets thread-safety(off-topic question: is slash thread-safe?) that means unless your hardware DOES run faster with 2.0, why change?
And of course, most third party apps developers are in no hurry to invest the time to thread-check their apps, if noone is using 2.0 with worker. It's a chicken and egg problem.
Maybe it's odd, maybe not. You'll notice Singles, isn't released in North America yet.
In an interactive situation(like a game), being in a familiar setting/knowing familiar characters/objects is an advantage, as it reinforces the "I can do this" feeling, important as you start a game. This is especially visible in games with online communities, where each player can build something, either his reputation(most combat games), or some sort of character(Diablo II), or a collection of objects(The Sims and similar games).
Most sequels/expansion packs allow you to improve what you've already built. And since expansion packs reuse game engines, your investment in the expansion usually comes either a little cheaper(you pay less for an expansion) or you get more game for your money(you get the game, AND the expansion, which is less than two games, usually)
Watching a movie isn't a "challenge" except for the odd movie about investigation(where the plot is the challenge, but your actions aren't as much your responsability as in a game(you connect plot elements you're given, but you can't see them in more detail, etc...), and your investment is always a full movie ticket... You don't get a "frequent movie fan" type ticket that costs less if you went to see the other movie. Games also get played longer(not many people still go to a movie that came out the time diablo ii was released, yet the online us east item trading is still doing brisk business)
So basically, it's easier to make a quality game sequel from the point of view of the consumer(who gets value from owning two games) than from the producer(who gets a little bit less from an expansion than a truly new game, but also can find ways to invest less).
Natural, in this case, is opposed to artificial which generally denotes the product of willful activity of sentient beings. Humans use "natural" to mean "without conscious thought" in many cases and "what these other humans didn't do" in many others.
The fact that nature ignores many of our willful changes(because it has inertia, and the power granted by being a system in instable equilibrium for the best part of 10E5 years does not mean we WANT nature to regulate against us. It doesn't mean nature can't adjust to our changes, it means we might not like the consequences.
Global warming means the overall temp might go up, but with the weather being a dynamic system, it might only mean that the temps at the poles rise up 10 degrees, and the equator and tropics go down 2 degrees, and guess what, it's warmer, globally, but more people are cold, because less people live in the Arctic Circle than in Quito, Ecuador, despite the fact that Quito is a bit smaller. I haven't gotten into altitude either... It's interesting that the article refers to the surface temperature of the "rock" we live on, as it might be a long term effect of the warming of the air on top of it.
It might also be related to the magnetic shifts covered in other slashdot posts(geologic events that take millenia, and could share a cause with the ice ages[it would be surprising, but it's still possible]) and yet the article speaks of the evolution over decades.
The problem with the climate is that we have trillions of life forms who all contribute part of the problem(breathing) and the solution(photosyntheis) to global warming. But approximately 6 billion of the actors are going on strike, saying we want to be a bigger part of the problem. While the other actors just watch, shake their head, and hope they don't get hit by the backlash of whatever "Correction" to use the economics term, the "gas balance" of the planet will apply.
After all, that is a self-regulating equilibrium people "tend" to understand a (admittedly little) better.
Being regulated by your equilibrium hurts... Much better to regulate yourself, when you can.
I believe he meant the delivery time, not the transition. Although postfix is meant to be a drop-in replacement for sendmail, and many facilities are provided to make the transition as smooth, easy and painless as possible.
On redhat, from sendmail to postfix, the transition DOES take five minutes... Provided you don't have thousands of special cases...
Hmm, none of the cheering men were interviewed to find out why...
I wonder if any thought even an all expenses date with an unknown woman was potential "work" instead of pleasure.
--
MHO is that it's food for thought, your thought might be starving, you know.
That's certainly something I'd be worried about...
A universal format has to be used everywhere(in theory) yet most games(for instance) avoid anything that isn't uber-optimized for a specific "target" which in many cases translates to a subset of one or two platforms.
Is U3D going to be a standard for ray tracing/modeling?
For 3d games?
For 3d web environments?
Only for static objects, or also for movement?
For all of the above? How will it perform well on environments which are so very disparate? (3d movement modeling can have much tighter tolerances, for instance, than 3d web environments, in terms of precision.)
Will it be like opengl was/used to be, yet patent-free? Or QuickTime 3d?
Or are we talking about another beast entirely?
It's interesting that the article already wants an "extensible" format, yet doesn't say the extensions themselves have to be covered by the standard.
(theoretical conversion, using a lame analogy)
-"You can have a graphic 3d format"
-"Cool! How big are the files?"
-"Very small if you use the compression extension"
-"How do I get that?"
-"3000$/display node"
-"What? but this is a standard for cripes sake!"
-"Yes it is, but nothing in the standard says anything about compression, that's covered by an extension, so you either take 1Gb files, or you pay for compression"
Let's keep things straight,
SPAM isn't "any unwanted email"
it's UCE.
Unwanted email is probably already outnumbering wanted email. But viruses are ALREADY illegal, so fudging them in with the spam, reduces the credibility of those who complain about spam, in lawmaker's eyes, who associate people who don't like spam with whiney people with no sense of discernment.
The article is about spam, which is probably reducing its "inbox percentage of total emails received"(for people who don't have gateway-level virus filters) and increasing it's "inbox percentage"(for people who block those at the gateway level, and never see the viruses).
Lumping our enemies together is great, as long as you like them outnumbering us, a faceless myriad of enemies. If you want to fight them, we gotta categorize them, unanonymize them, and take em out, one at a time.
--
I still remember the internet before spam
It was idyllic
Most readers will trade off accuracy for immediacy if that means they get the information "in time".
By mentioning "chasing the chimaera of objectivity" it's pretty obvious that we aren't talking about "more accurate isn't ok if you're second", but more "we don't want to wait till the politician announces his resignation to discover he's crooked".
There are degrees of innacuracy involved. There's also the fact that for true objectivity to come from big media, they'd have to stop being large corporations beholden to shareholders and selling ad space to other large corporations...
Most people don't worry about objectivity from the (current) media because they trust that the competition is fierce enough that journalists ARE actively seeking scandals in competing news sources(at least, that's the plan) and the government. I'm sure they'd be disappointed by how much of that money is funneled through press agencies or spent chasing press releases.
webwasher, like most filtering software, uses static "blacklists", so they can't tell if someone really uses a single domain for multiple purposes, or if it's someone using the "not-so-raunchy" bits to do like spam, and intersperse "not blacklisted" material to contour the blacklist.
Unfortunately until some form of dynamic(updated to the minute or less: bandwidth costs for the filtering providers...) blacklisting of pages occur, this type of "block everything unless we know its good" will stay prevalent.
a funny thought:
(some) execs of companies dream of patents they don't have to sink a few half-million in research for
The patent system is made to encourage them to sink those few millions.
By that logic, the companies in the first situation are abusing the system, if we had a way to know for sure, we (sh/w)ould take away the first companies' patents away.
I'm a consultant, paid to keep current on many topics, so on some topics, yes...
It's however a case of "do more with the same money" instead of "do the same with less money" if that's what you mean.
*disclaimer safari subscriber*
I used to buy the animal books on several topics, mostly perl programming
Then I got the safari subscription
imagine this:
oreilly comes up with fourth edition of dns and bind
I have paper third edition of dns and bind
I use safari to get fourth edition, and I don't need the paper one anymore.
Since a lot of the animal books I use are very sucessful, and get updated every so often, just because I can replace one edition with the next at no charge, I save a bundle of money, provided I don't need hardcopy of the work in question, the web interface to it might actually save me time(mostly searching, although with practice, the internal binary-page search is pretty damn hard to beat, it's the "read entire TOC" that takes a while.)
Of course, I've been known to read entire online volumes on topics I was less familiar with(I can't say I'd do it with something like the perl cookbook) but so far, Safari is working out for me.
That's why the patents process normally applies to technology, not science. But ever since patents started covering ideas, and not the technological items that represent them, tne line has been blurring...
Software patents(and the dreaded algorithm patent sure to come out someday if the trend doesn't stop) further blur the distinction.
*steps on soapbox*
now it appears to me that part of the problem with patents and copyright is that the public instinctively associate them with the inventor/discoverer/original author, and we're fine with them, until such rights are sold then we become not so fine with it. For one, it's a lot simpler/more instinctive to sort out who disovered something, if it's an individual making the discovery, not a corporation. In that same line of thought, what happens when those rights are transferred, and for instance, how to get the original author to unlearn what he sold, gets a lo more confusing.
Trademarks, as a rule are purely corporate/marketing concepts and have far less contestation going on... They also have to be defended constantly...
Copyrights are especially problematic, as copyright ownership gets regularly extended for corporations (technically, you could even view the legatees of an author as a corporation, THEY DIDN'T WRITE THE BOOK) to protect little mouses, and such... Yet the author in many cases is long dead... before the first lawyer gets paid... (Not that I object to heirs of authors and such to benefit from the work of their forebears... I do however insist on the fact that by that point, the original intent of copyright: to encourage the author to produce more work, because his rights are protected, will take a lot of doing to respect: he's dead, he won't write that many books anymore...
You might say part of the problem is that the public recognizes a right to a discoverer/author as an individual, for a limited duration, but it becomes a lot less instinctive when those are transferred. Perhaps making them non-transferrable would keep the legislation on this topic, at least understandable by someone without a major in minor in intellectual property...
I'd like to contrast the American experience with the Canadian one, and maybe some contrasts will become apparent...
r eas have their warts... But I think it's important to make the point that the "pay per channel" model is not the only important part of a workable scheme...
1) My cable provider has a-la-carte pricing, but only if you take the expensive digital rig (you pay more up-front, you pay less if you stick around for five years) you can also pick a bewildering array of packages(of course, confusion is their friend)
2) The pricing scheme for channels is heavily regulated, for several reasons, some of which follow:
a) Part of the per-channel price(indeed any cable bill) goes to promote production of local content
b) the CRTC regulates the final price, for consumer-protection, and to prevent many kinds of abuses
c) not all per-channel fees are considered equal
d) some channels are deemed to be worth more, for various reasons(some are actually one-fee for four channels worth of bandwidth under one theme) for example
e) the linguistics aspect also is involved(somehow, I'm neither a lawyer, nor really inclined to learn so much of this, as it oftens to be more politics than law)
It certainly isn't a perfect system, but it works on the following premises:
1) the experts on the CRTC are supposed to know what the costs are
2) The satellite operators and cable operators are "passing the token" about who's unwilling not to charge consumers for unviewed channels, having regulated prices goes around the issue of "whose fault is it" Consider also that: channel operators fight so hard to get a license to have their own channels, they can't actually argue "well gee, we fought like lions to get this and now you take away our revenue stream!" when they actually have to justify that they will have audience once the channel operates, just to get the license to start their channel.
3) Many of the bundles include common mixes, for cheaper, but I did the math, and I found it almost impossible not to get some channels I didn't want along with channels I did... Choosing per-channel for example, was the best way to guarantee I could take the Space Channel along with the Z channel(the local, francophone techno-channel)
Now obviously, so far the system works, and the large, monolithic, much-more-profitable-in-densely-populated-urban-a
The idea being is that customers in the ideal system don't pay for what they don't want, not that they pay for those they want...(Yes they are ALSO paying for those they want, but it's very easy to just tax for the unwatched channels by doubling the rate of heavily watched channels if there is no regulation/supervision)
It's also noteworthy that they don't offer this choice without a substantial customer investment(the equipment is around 600$US total, perhaps, but I'm not sure how to deduct the programming credits, since they seem to apply only on popular bundles, which go against the grain of this article).
How is this relevant you say?
Well let's see:
a good, customer-defending system has safeguards in place to protect customers, both from channel and cable-operator-spun fud(they are there to make a profit, and most of them DO own news-producing companies, so the media can hardly be considered neutral when they are concerned.) So I'd certainly suggest to any south of the border friends to consider enforcement of any safeguards before any "pay per use" schemes are applied "because they are cheaper" it's not really that simple. Without any bundles, consider this pricing: over 1$US(using approximate exchange rate) per channel, if you take no bundle at all, with a 500$+ setup fee
Some of the bundles start at 15$US... But do they include what you want? That is the question, so unless you really need only some channels, it might not really be that much cheaper, unless you really aren't interested in the same thing as the suburbian 2.3 kids family next door. Now that just may be the case of many slashdotters, your milage may of course vary.
What makes it like apple is that their software doesn't work on other people's hardware.
The Windows version you get from Dell also works(at least, the serial number would) on an HP.
If I may, read that certain carefully.
I said their ownership of the platform CAUSED uncertainty, not that it was uncertain.
If you read up the thread, you'll see that Sun's ownership was spoken of as a bad thing...
Hmm, Because most languages people think they "know" they know not as much as they think they do, and what they do know is often inefficient/misadapted to writing a large, platform-independant, network-distributed set of API's?