Has anyone run strings (or its Windows equivalent) on the driver binary, and seen if any well known executable names come up?
If they do, it will give us a list of programs where Nvidia is doing application specific optimizations, or (as in the 3dmark case) are doing a dirty and substituting faster but wrong operations.
If they don't then it seems clear that they're deliberately disguising the fact, given that there's empirical proof they're doing it for 3dmark, and this would demonstrate that Nvidia are fully aware that what they're doing is underhand.
It's not an optimization if it does not produce the same results! Recall that the shader code that the driver used did not produce the same visual results as the shader code it replaced.
More tellingly, the driver deliberately flaunts the D3D spec by omitting buffer clears, mucking about with clip planes, etc.... based purely on application-specific pattern matching, which by its very nature is fragile. As was demonstrated so aptly by the 'off the rails' mode in Futuremark. This isn't an accidental bug: it is obvious that such mechanisms are highly fragile, and are almost certain to cause bad rendering on these applications when they are modified in small ways.
As others have said, Futuremark's statement is just covering their legal arse. If someone modifies their code to get better scores in some benchmarks while introducing deliberate bugs (i.e. incorrect rendering), it's a cheat in my book.
Secondly, why are the traditional "beige box" makers so reluctant to follow?
I think it's for the same reason that most of the cars produced in the last 15 years look so similar. Distinctive design tends to polarise people - those who like it like it a lot, while those who don't, hate it. Large manufacturers have tended to take the conservative approach with lower risk, even when a higher-risk approach might have brought higher rewards.
That said, there are certainly a number of companies producing functional and aesthetically pleasing designs around x86 hardware. The first that comes to my mind is Soldam (who also have an English site.) Check out their Pandora Dino or Polo R Figaro for example.
The chief advantage of this sort of thing, as far as I can tell (IANAWM), is that it allows the wine maker to practice their craft with more consistent and better quality raw materials.
Like all good tools, it's how it is used that counts. Certainly it could allow greater homogenisation. On the other hand, it can allow the wine maker to create better and more interesting wines, when they do not have to cater to the vagaries of the environment to such a degree. If anything, I think such technology will have more of a positive effect than a negative, because the "consistently good but not great, dull but predictable and affordable" market seems to be sewn up already by the large wine manufacturers.
Almost all the software I get paid to write is written to allow my employer to do their work better, faster, or at all. Only a small portion is for distribution outside the company. Even if the distributed code were to be given away free of cost, the other code I write - which again, is by far the majority - would keep me employed and valuable to the company.
I would go so far as to say that most code written is written to perform some task for the people employing the programmer, rather than for resale.
That said, note that free (as in GPL) software does not mean that a company producing it gets no revenue. For starters, it need not be given to customers for free; while the customers can then create derivatives, redistribute source and so forth, they still need to buy the program in the first place. Given a choice of buying it with support from the vendor, or compiling it themselves from source gained from a 3rd party, many would (and do!) buy it from the vendor.
Further in products which contain a mixture of code and other material (for example, computer games, databases with data, etc.) the code component can be free-in-GPL and free-in-cost, while still generating revenue for the creators as part of a product which is very much not free.
You like being paid to write software. That's good, getting payed for practicing an art that (I presume) you enjoy and are skilled at. If free software became the norm, only one particular avenue of revenue for potential employers is removed, and it is one which probably does not account for more than a fraction of employed coders. There will still be many opportunities for you to practice your art. And with much free software available, you have a much greater opportunity to learn from and build upon the work of others, potentially allowing you to be a better and more efficient programmer.
These machines will run X just fine. My old Cyrix 120Mhz / 32MB machine ran X perfectly well, using fvwm2 as window manager, though netscape was certainly slow. For emacs/xterm/clock/image-viewing it was just dandy. I honestly think that when people say X is bloated and slow, they are confusing X with some of the toolkits people use on top of it. Netscape became tolerable when the machine was given an additional 32MB of RAM.
Gnome ran tolerably on the machine, but slow enough to make me consider upgrading if wanted to use it over fvwm.
For example, as far as I understand it, in popular forms of Buddhism there are malicious spirits (mara) who cause suffering through the use of deceit and illusion. The sad thing though is that these spirits themselves are trapped in the cycle, and can not escape from the greater illusion that is the world.
In the original Matrix movie, Agent Smith, in the interrogation sequence with Morpheus, gives the impression of being caught up in this artificial world. He has become very much like the humans themselves in fact, with a capacity for disgust and anger. And like most of the humans, he can not escape the Matrix - his existence is inextricably tied to it.
A mara might be reborn as a human, and start on the path to enlightment and ultimately nirvana. As Agent Smith becomes more human, is there the chance that he too will start to see the Matrix as the illusion it is?
The analogy breaks down, as in the film there really is a 'real world' to which the protagonists can escape. Still as a metaphor, it can give some grounding to the limitations of the agents as observed in the film.
We used Visual Source Safe at the beginning of our project (about 10 programmers), despite calls for using CVS instead. VSS did not last three weeks. It is the embarassment of the revision control world. It is Just Broken.
Note that Microsoft most certainly do not eat their own dog food. At least, they certainly did not then (1998-1999).
We threw out VSS. Moved to CVS, despite losing Visual Studio integration and past revision history. In fact, we just used CVS from a command window. It worked, and worked well.
There are commercial and free products now which probably fit your requirements. But if you can't find one, you still ought to ditch VSS and go with something that actually works. It doesn't matter how convenient your source control system is if it doesn't actually control it, or like VSS, actively corrupts it.
That's doesn't follow at all. The "service differentiator" is the only reasonable way one can tell professionally run mailservers from the trojaned Winboxes running spam relays (with some margin of error).
One of the reasons so much spam was sent from dynamic addresses was the lack of accountability - it was hard to associate the spam with an individual spammer... or owned Windows box. A static IP makes the link between the two quite plain, and thus removes one of the chief advantages of using a dynamic IP from the spammer's point of view.
Even if every address was static, there still would be a residential service shitlist.
Even if there were still a blacklist based on residential services, a static IP for any individual user makes it much easier and/or more secure to arrange a relay with a 3rd party. Authentication of the mail sender can be done by IP address (and a secure way of verifying that, but luckily there are numerous options here.) A shitlist would be far more easily worked around by those for whom it was important to do so.
And lastly, it's sad but true: 'professional' servers get trojaned too. Security may be poor by default for residential boxes, but it often is for machines on a fixed IP as well. That fixed IP though allows for much more finely grained accountability, firewalling, etc. etc.
Many ISPs, companies and mail services refuse to accept mail sent from dynamically assigned IP blocks. AOL are simply joining a large and growing body of organisations that choose to blindly block as a weak defense against spam.
The irony is that the whole problem stems from the policy of assigning dynamic IPs to dial-up and DSL users. Had ISPs made a serious (or any!) effort to support IPv6, or stopped using static IPs as a service differentiator that allowed them to charge much higher business rates, this would all be moot.
What pisses me off the most is that T-Online (Deutsche Telekom's 'net service) is dynamic IP only if you want to avoid per-MB charging, yet they charge extra for the use of their SMTP relay. In my mind, this is just despicable. The lack of any real competition for affordable high speed services means that this sort of thing isn't going to go away any time soon.
So here's a big hello to all my peers on the residential second class Internet!
It is easy to pick on the US because it is the most prosperous, visible and free of any state in the world.
But it's not! That's the whole point of articles like these... they demonstrate that the US is not an island utopia of freedom in a sea of despots. The US is certainly visible, and on average is certainly propserous. But you'll find more freedom in most other Western democracies than you will in the US, I'll wager. Certainly you'll get more in (for example) Australia, and as far as I know, in Germany and Canada too.
Having a nice bill of rights doesn't mean squat if they aren't respected.
The sooner US citizens as a group realize that their country isn't the best in the world, the sooner they can do something about improving it. This patriotic blindness is bad for the US, and it's bad for the rest of the world too.
If fighting a war requires people be killed, perhaps they ought not be fought save as a last resort.
The view that this attack on Iraq at present is legal (with respect to international law) certainly is not uniformly held, and I would go so far as to say that the view that it is, is held only by a minority of interested parties.
No clear evidence linking Hussein or Iraq generally to a credible threat to the US has been presented. If it is self-defense, it has certainly not been convincingly argued.
I am not a US citizen, but Australian. Still, our government has committed troops to the conflict, and so it's still pertinent.
The question is: why is killing suddenly not wrong, if it's an order?
I am sure there is coercion of sorts for troops to obey orders, even when they are immoral or illegal. For each individual soldier it becomes (or should become) a choice between committing or aiding murder or facing the consequences of failing to obey. When these consequences are severe, then it is understandable that soldiers obey. Certainly though it is far from clear morally. Court-martial and the loss of a military career is one thing, but of course if it is regarded as treason then there is the threat of the death penalty in the US.
Remember that in this war, the US and its allies are the aggressors; it is not a matter of self-defense. Perhaps soldiers should be making those moral decisions right now, and refusing to fight, even to the loss of their careers. Because at the end of the day they're killing people.
Note also that this war is certainly not a legal one in any international sense.
If I understand it correctly, the author is lamenting that neither of the standard ways of parsing XML in a scripting language fit the straightforward model of scanning for something relevant and then acting upon it, where the two models are: 1) read in whole file and make a tree (take sup too much memory, is slow, etc.); or 2) use a callback interface.
The style of perl script he was seeking was a simple loop model:
while () {
next if/ignorable/;
if (/thing-one/) {... }
elsif (/thing-two/) {... } ...
}
To me the thing that distinguishes this the most from the provided XML parsing interfaces is that it has a minimal amount of state.
So isn't what is needed a corresponding structure to the while () above that iterates over the tree-nodes of the XML-encoded data structure, in a depth-first preorder traversal (to avoid having to build the whole tree first)? One could imagine a parser object that scans through the XML file returning nodes (and their parent history) while maintaining an absolute minimum of state. If one wanted to build an in-memory representation of a subtree given a node, then one can always do so when one finds the node one wants.
Such an interface wouldn't be good for integrity verification or the like, but for the sort of application the author was talking about, it would seem ideal. Much less flexible than the normal models, sure, but much easier to work with when the problem fits this sort of description. Perhaps I'm underestimating the difficulty of the task, but it doesn't sound too hard to write, given that it is doing so much less than the fully-featured XML parsing interfaces.
The other problem is the awkwardness of the use of XML in O-O languages such as addressed in the
article linked-to by Tim Bray in his article. Though I haven't used this particular program, this seems to be the problem that FleXML is trying to address. When you don't need all of the flexibility that XML can provide, but instead have a fixed schema that your XML-representation follows, why not have your parser automatically built to read it? People have used lex/flex for scanning text files for decades --- in these days of XML Schema, it should be even easier. If FleXML lives up to its promise, it will be. Has anyone here used FleXML and are willing to comment on how well it addresses these sorts of problems?
Having worked for a game developer for a bit (but not any more), I can vouch for the sad state of software engineering, project management and management generally in at least one company. Talking to peers and browsing the likes of Fat Babies indicates that if not the norm, it's certainly wide spread.
We made a great game, but it was in spite of, not because of, the engineering and management practices. It is worth noting that every employed programmer from that game has since left the company. Good people, but bad practices. I still hope they can turn things around, but I'm not holding my breath.
If only Blizzard could provide server-side scripting support, we could conquer lag!
This is said only half in jest. First point: for a game designed to be played over the Internet, Diablo II is shockingly lag intolerant. If you're on the same continent as a server, then it's not too bad. If you're stuck on a modem in Australia, whole swathes of skills or gameplay styles just don't work well or at all.
Second point: server side scripts represent a way of dealing with a game at a higher level. Instead of making a click-fest of a game where latency and fast mousing skills count -- such as Warcraft 3 for example -- what about a competetive game where all the twitch aspects are handled by programs at the business end of the game, instead of by hand over a slow internet link? The skill and fun then comes into selection, deployment and generally higher level strategy. Or even into script writing. (Self and friends are working on such a game, but even we aren't holding our breaths for it to become a playable thing. Free time coding and all that.)
PS: It was always more fun writing client robots for LPMUDs than it was to play the MUDs themselves.
Just to second this, cdjapan have been realy good for recently published music. For back catalogue though, it's not so good - but what is? Small (now vanished) labels' CDs from the early 90s seem to be impossible to track down, sadly.
No matter how many times you might repeat that, it still isn't true.
On the contrary, no matter how often the claim otherwise is repeated, "intellectual property" is a legal fiction, maintained only by goverment fiat.
The only idea owned, in any normal sense of the word 'own', is the idea that is not shared. You can talk about the source of an idea, the inspiration for an idea, the scope of an idea, and so on. But the owners of an idea are exactly those who have it, and this includes everyone to whom the idea is transmitted.
Copyrights, trademarks and patents are legal mechanisms used to restrict the propogation of the representations of ideas, or to restrict their implementation. The aim is nominally to encourage the creation of more entertaining or useful ideas and knowledge, though it seems to be used more as a tool to enrich the rich and restrict competition. That said, none of these tools, even in a legal sense, claim to restrict ownership of an idea; at most they restrict its representations, expressions and transmissal.
Ideas are not like bricks or cabbages. Ideas can not be owned in any sense akin to that of material objects. At most, they can be kept secret and thus said to be in the possession of a few.
The "high drug prices and patents" vs. "no drugs and misery and death" is a false dilemma.
Medical research is expensive and has to be paid for somehow. The question is, can this be done more efficiently than the current model?
rknop suggests it can -- and does address the issue, if you'd take the time to read the post.
The current system has two obvious and large economic inefficiencies: people who can't afford drugs remain sick and don't contribute as much economically; and the very large marketing budgets of the pharma companies aren't paying for research. Given the size of these advertising budgets, estimated to match or be as much as twice the R&D budgets, the current system is revealed as being at least 50% inefficient, before considering any other aspects.
Then there are less obvious and harder to quantify inefficiencies, such as money wasted on ill-advised business acquisitions, political lobbying, excess costs borne by drug purchasers (who are pretty much over a barrel here) and so on.
It would be surprising if one couldn't beat 50% (and I believe it's considerably lower) efficiency, presuming one's goal here is in fact to produce medicines to cure and treat illnesses that plague our various societies.
One possible mechanism is university or research institute funding. It seems to work pretty well in physics or mathematics for example, and the budgets for large physics projects are getting pretty enormous. Why is this money spent? Because it has a good chance to better quality of life in the long term, and past experience has demonstrated this to be the case.
Medical research spending has much quicker payoffs than fundamental physics research.
Why would researchers bother developing treatments, if there were no big carrot being dangled before them? Two answers: firstly, the big carrots are being dangled in front of investors and board members, not the researchers; secondly, fame and renown work perfectly well in other scientific fields of endeavour, and are likely to work even more effectively in medicine where the benefits are so much more apparent to the public at large.
As rknop said, things aren't likely to change in a hurry, because the status quo benefits the people with the power more than the alternatives would. The alternatives though have the potential not only to be much more morally sound, but also to be more economically sound. Isn't nice to be able to take both an economic and moral highground?
This device looks really nice: switchable piggyback storage, FM transmittery thing, not too heavy, not too bulky, 20Gb, etc. etc. Of course ogg support is a big plus too. It could be the iPod competition we've all been waiting for! But...
USB 1.1?! What were they thinking? How could they get so close and still drop the bundle?! Transfering a CD's worth of music onto the device would take well over a minute at any decent quality. Transferring a collection onto the drive would take hours. If there were no alternatives, then sure, it's certainly not too bad. But with a disk attached to the device, there's no good reason why transfers couldn't be ten times as fast, if only they used USB2 or firewire.
As a portable harddrive, USB1.1 speeds are apalling.
Would putting firewire or USB2 on really have been so hard? As it stands, the player seems to be in the 'so close but' category.
As Mosfet said, the solution is better organization, not wholesale throwing out of features!
Gnome 2 is the biggest disappointment for me. I was hoping for a Gnome 1.4 without the bugs, with a saner and friendlier GTK+ toolkit underpinning, with great internationalization support from Pango. If anything, gnome-terminal is now even buggier (with more features), while the severe cutting of options makes it impossible for me to set up a Gnome 2 desktop the way I would like to, short of hacking the code.
Examples:
instead of 5 clocks, we have one. Don't like it? Well, cope. Blechh!
want to use the Window List applet on a vertical panel? With the 'new and improved' option selection, one can't see the titles (due to the icons), and the buttons are supersized with no way of fixing it.
No [apply] buttons. Great, so now when I make a selection, it applies it immediately, even if I want to do a number of things in one go and the individual steps are time-consuming to apply. Now changing settings takes longer.
... and so on, so on.
There was an old slashdot article reporting an interview with some UI expert [1], who made claims that configurability was a terrible thing. Yet the arguments used there were just invalid; they simply didn't apply. The principle of 'less-is-more' is at its base a good one: clutter is unattractive and unhelpful. But one should achieve this goal through better organisation, not by discarding things people are actually using! On my desktop I'd like the 'less-is-more' principle applied to reducing or eliminating excess toolbars, and maximising work space for example, but I don't see that happening in a hurry.
I may have to give KDE 3.1 another shot.
[1] speaking of lack of options limiting usability, how about a more functional slashdot search?!
The money argument does not hold water, because the very things that typically tie a site to a single platform are those which are the most expensive to produce.
Extensive javascript menus, elaborate flash 'navigators', exotic ActiveX controls -- all these things take time to produce, a lot more time in fact that the simple option which would have worked anywhere.
The problem isn't money or market-share, it's that so-called web designers are pandering to ignorant clients who want something pretty on their personal desktop rather than a useful web presence. Two groups are at fault: web-designers with no pride in their profession, and clients who are much more interested in spending their company's money on attractive interactive wallpaper than on an effective web site.
The 5% market share argument is an old canard parrotted by web quacks who won't learn new tricks.
The provision of so-called 'residential' services by broadband providers really disturbs me.
Typically in any area (thinking Europe, UK, Australia - true for US too?) there are only one or two high-speed providers to choose from. They offer two tiers of service: one is with a fixed IP, costing $lots per month and where one is charged by the incoming MB; the other is a residential service with a temporary IP -- that is often forcibly expired, killing connections etc. once or twice a day -- with an affordable cost and a relatively high cap before per-byte charging comes in.
These residential services though don't offer the Internet per se, but some sort of diluted version. No fixed IP means no reliable servers. No home-served content for you! I haven't yet seen a mainstream provider that offers IPv6 addresses; if lack of IPv4 addresses were the only motivation for this IP cycling game, then surely they'd offer a stable IPv6 address. The access agreements further compound the situation, with restrictions such as this 'no VPN', or no web serving, or only one computer on the connection, or no multiple accounts, or so on.
The dynamic IP stuff also means that one is pretty much forced to use an SMTP relay for outgoing mail, as so many sites blacklist known dynamic IP blocks out of hand. T-online here in Germany is about to start charging for their SMTP relay service!
The whole point of course is to extract the maximum amount of money out of the market. These service restrictions aren't there to cover otherwise present costs or the like, they're there to provide a differential betweeen their services, so that the providers can extract more money out of anyone who might possibly want to use the 'net for anything serious.
In the same way that major Telcos dragged their feet with ISDN and the like in the UK and in Australia, pricing it per minute _and_ per byte, and thereby siginificantly delaying the adoption of the 'net by businesses at large, the current practices are also limiting the adoption of the Internet as a tool for anything other than passive content consumption.
If there were a level marketplace for internet services, then the situation probably wouldn't last. But of course this isn't the case when there are $10^8 barriers to entry against an entrenched monopoly or duopoly.
Slashdot Mirror
Coming from TeX-land, I'm tempted to say that msword can't really do formulae either.
:)
Honestly, if you're typesetting maths, do yourself a favour and use a tool that is designed for the job
Has anyone run strings (or its Windows equivalent) on the driver binary, and seen if any well known executable names come up?
If they do, it will give us a list of programs where Nvidia is doing application specific optimizations, or (as in the 3dmark case) are doing a dirty and substituting faster but wrong operations.
If they don't then it seems clear that they're deliberately disguising the fact, given that there's empirical proof they're doing it for 3dmark, and this would demonstrate that Nvidia are fully aware that what they're doing is underhand.
It's not an optimization if it does not produce the same results! Recall that the shader code that the driver used did not produce the same visual results as the shader code it replaced.
... based purely on application-specific pattern matching, which by its very nature is fragile. As was demonstrated so aptly by the 'off the rails' mode in Futuremark. This isn't an accidental bug: it is obvious that such mechanisms are highly fragile, and are almost certain to cause bad rendering on these applications when they are modified in small ways.
More tellingly, the driver deliberately flaunts the D3D spec by omitting buffer clears, mucking about with clip planes, etc.
As others have said, Futuremark's statement is just covering their legal arse. If someone modifies their code to get better scores in some benchmarks while introducing deliberate bugs (i.e. incorrect rendering), it's a cheat in my book.
That said, there are certainly a number of companies producing functional and aesthetically pleasing designs around x86 hardware. The first that comes to my mind is Soldam (who also have an English site.) Check out their Pandora Dino or Polo R Figaro for example.
The chief advantage of this sort of thing, as far as I can tell (IANAWM), is that it allows the wine maker to practice their craft with more consistent and better quality raw materials.
Like all good tools, it's how it is used that counts. Certainly it could allow greater homogenisation. On the other hand, it can allow the wine maker to create better and more interesting wines, when they do not have to cater to the vagaries of the environment to such a degree. If anything, I think such technology will have more of a positive effect than a negative, because the "consistently good but not great, dull but predictable and affordable" market seems to be sewn up already by the large wine manufacturers.
Almost all the software I get paid to write is written to allow my employer to do their work better, faster, or at all. Only a small portion is for distribution outside the company. Even if the distributed code were to be given away free of cost, the other code I write - which again, is by far the majority - would keep me employed and valuable to the company.
I would go so far as to say that most code written is written to perform some task for the people employing the programmer, rather than for resale.
That said, note that free (as in GPL) software does not mean that a company producing it gets no revenue. For starters, it need not be given to customers for free; while the customers can then create derivatives, redistribute source and so forth, they still need to buy the program in the first place. Given a choice of buying it with support from the vendor, or compiling it themselves from source gained from a 3rd party, many would (and do!) buy it from the vendor.
Further in products which contain a mixture of code and other material (for example, computer games, databases with data, etc.) the code component can be free-in-GPL and free-in-cost, while still generating revenue for the creators as part of a product which is very much not free.
You like being paid to write software. That's good, getting payed for practicing an art that (I presume) you enjoy and are skilled at. If free software became the norm, only one particular avenue of revenue for potential employers is removed, and it is one which probably does not account for more than a fraction of employed coders. There will still be many opportunities for you to practice your art. And with much free software available, you have a much greater opportunity to learn from and build upon the work of others, potentially allowing you to be a better and more efficient programmer.
Adding to what others have said ...
These machines will run X just fine. My old Cyrix 120Mhz / 32MB machine ran X perfectly well, using fvwm2 as window manager, though netscape was certainly slow. For emacs/xterm/clock/image-viewing it was just dandy. I honestly think that when people say X is bloated and slow, they are confusing X with some of the toolkits people use on top of it. Netscape became tolerable when the machine was given an additional 32MB of RAM.
Gnome ran tolerably on the machine, but slow enough to make me consider upgrading if wanted to use it over fvwm.
It makes sense in an allegorical way ...
For example, as far as I understand it, in popular forms of Buddhism there are malicious spirits (mara) who cause suffering through the use of deceit and illusion. The sad thing though is that these spirits themselves are trapped in the cycle, and can not escape from the greater illusion that is the world.
In the original Matrix movie, Agent Smith, in the interrogation sequence with Morpheus, gives the impression of being caught up in this artificial world. He has become very much like the humans themselves in fact, with a capacity for disgust and anger. And like most of the humans, he can not escape the Matrix - his existence is inextricably tied to it.
A mara might be reborn as a human, and start on the path to enlightment and ultimately nirvana. As Agent Smith becomes more human, is there the chance that he too will start to see the Matrix as the illusion it is?
The analogy breaks down, as in the film there really is a 'real world' to which the protagonists can escape. Still as a metaphor, it can give some grounding to the limitations of the agents as observed in the film.
We used Visual Source Safe at the beginning of our project (about 10 programmers), despite calls for using CVS instead. VSS did not last three weeks. It is the embarassment of the revision control world. It is Just Broken.
Note that Microsoft most certainly do not eat their own dog food. At least, they certainly did not then (1998-1999).
We threw out VSS. Moved to CVS, despite losing Visual Studio integration and past revision history. In fact, we just used CVS from a command window. It worked, and worked well.
There are commercial and free products now which probably fit your requirements. But if you can't find one, you still ought to ditch VSS and go with something that actually works. It doesn't matter how convenient your source control system is if it doesn't actually control it, or like VSS, actively corrupts it.
And lastly, it's sad but true: 'professional' servers get trojaned too. Security may be poor by default for residential boxes, but it often is for machines on a fixed IP as well. That fixed IP though allows for much more finely grained accountability, firewalling, etc. etc.
Many ISPs, companies and mail services refuse to accept mail sent from dynamically assigned IP blocks. AOL are simply joining a large and growing body of organisations that choose to blindly block as a weak defense against spam.
The irony is that the whole problem stems from the policy of assigning dynamic IPs to dial-up and DSL users. Had ISPs made a serious (or any!) effort to support IPv6, or stopped using static IPs as a service differentiator that allowed them to charge much higher business rates, this would all be moot.
What pisses me off the most is that T-Online (Deutsche Telekom's 'net service) is dynamic IP only if you want to avoid per-MB charging, yet they charge extra for the use of their SMTP relay. In my mind, this is just despicable. The lack of any real competition for affordable high speed services means that this sort of thing isn't going to go away any time soon.
So here's a big hello to all my peers on the residential second class Internet!
Having a nice bill of rights doesn't mean squat if they aren't respected.
The sooner US citizens as a group realize that their country isn't the best in the world, the sooner they can do something about improving it. This patriotic blindness is bad for the US, and it's bad for the rest of the world too.
Three quick points:
This concerns me.
I am not a US citizen, but Australian. Still, our government has committed troops to the conflict, and so it's still pertinent.
The question is: why is killing suddenly not wrong, if it's an order?
I am sure there is coercion of sorts for troops to obey orders, even when they are immoral or illegal. For each individual soldier it becomes (or should become) a choice between committing or aiding murder or facing the consequences of failing to obey. When these consequences are severe, then it is understandable that soldiers obey. Certainly though it is far from clear morally. Court-martial and the loss of a military career is one thing, but of course if it is regarded as treason then there is the threat of the death penalty in the US.
Remember that in this war, the US and its allies are the aggressors; it is not a matter of self-defense. Perhaps soldiers should be making those moral decisions right now, and refusing to fight, even to the loss of their careers. Because at the end of the day they're killing people.
Note also that this war is certainly not a legal one in any international sense.
If I understand it correctly, the author is lamenting that neither of the standard ways of parsing XML in a scripting language fit the straightforward model of scanning for something relevant and then acting upon it, where the two models are: 1) read in whole file and make a tree (take sup too much memory, is slow, etc.); or 2) use a callback interface.
The style of perl script he was seeking was a simple loop model: /ignorable/; ... } ... }
...
while () {
next if
if (/thing-one/) {
elsif (/thing-two/) {
}
To me the thing that distinguishes this the most from the provided XML parsing interfaces is that it has a minimal amount of state.
So isn't what is needed a corresponding structure to the while () above that iterates over the tree-nodes of the XML-encoded data structure, in a depth-first preorder traversal (to avoid having to build the whole tree first)? One could imagine a parser object that scans through the XML file returning nodes (and their parent history) while maintaining an absolute minimum of state. If one wanted to build an in-memory representation of a subtree given a node, then one can always do so when one finds the node one wants.
Such an interface wouldn't be good for integrity verification or the like, but for the sort of application the author was talking about, it would seem ideal. Much less flexible than the normal models, sure, but much easier to work with when the problem fits this sort of description. Perhaps I'm underestimating the difficulty of the task, but it doesn't sound too hard to write, given that it is doing so much less than the fully-featured XML parsing interfaces.
The other problem is the awkwardness of the use of XML in O-O languages such as addressed in the article linked-to by Tim Bray in his article. Though I haven't used this particular program, this seems to be the problem that FleXML is trying to address. When you don't need all of the flexibility that XML can provide, but instead have a fixed schema that your XML-representation follows, why not have your parser automatically built to read it? People have used lex/flex for scanning text files for decades --- in these days of XML Schema, it should be even easier. If FleXML lives up to its promise, it will be. Has anyone here used FleXML and are willing to comment on how well it addresses these sorts of problems?
Having worked for a game developer for a bit (but not any more), I can vouch for the sad state of software engineering, project management and management generally in at least one company. Talking to peers and browsing the likes of Fat Babies indicates that if not the norm, it's certainly wide spread.
We made a great game, but it was in spite of, not because of, the engineering and management practices. It is worth noting that every employed programmer from that game has since left the company. Good people, but bad practices. I still hope they can turn things around, but I'm not holding my breath.
If only Blizzard could provide server-side scripting support, we could conquer lag!
This is said only half in jest. First point: for a game designed to be played over the Internet, Diablo II is shockingly lag intolerant. If you're on the same continent as a server, then it's not too bad. If you're stuck on a modem in Australia, whole swathes of skills or gameplay styles just don't work well or at all.
Second point: server side scripts represent a way of dealing with a game at a higher level. Instead of making a click-fest of a game where latency and fast mousing skills count -- such as Warcraft 3 for example -- what about a competetive game where all the twitch aspects are handled by programs at the business end of the game, instead of by hand over a slow internet link? The skill and fun then comes into selection, deployment and generally higher level strategy. Or even into script writing. (Self and friends are working on such a game, but even we aren't holding our breaths for it to become a playable thing. Free time coding and all that.)
PS: It was always more fun writing client robots for LPMUDs than it was to play the MUDs themselves.
Just to second this, cdjapan have been realy good for recently published music. For back catalogue though, it's not so good - but what is? Small (now vanished) labels' CDs from the early 90s seem to be impossible to track down, sadly.
The only idea owned, in any normal sense of the word 'own', is the idea that is not shared. You can talk about the source of an idea, the inspiration for an idea, the scope of an idea, and so on. But the owners of an idea are exactly those who have it, and this includes everyone to whom the idea is transmitted.
Copyrights, trademarks and patents are legal mechanisms used to restrict the propogation of the representations of ideas, or to restrict their implementation. The aim is nominally to encourage the creation of more entertaining or useful ideas and knowledge, though it seems to be used more as a tool to enrich the rich and restrict competition. That said, none of these tools, even in a legal sense, claim to restrict ownership of an idea; at most they restrict its representations, expressions and transmissal.
Ideas are not like bricks or cabbages. Ideas can not be owned in any sense akin to that of material objects. At most, they can be kept secret and thus said to be in the possession of a few.
The "high drug prices and patents" vs. "no drugs and misery and death" is a false dilemma.
Medical research is expensive and has to be paid for somehow. The question is, can this be done more efficiently than the current model?
rknop suggests it can -- and does address the issue, if you'd take the time to read the post.
The current system has two obvious and large economic inefficiencies: people who can't afford drugs remain sick and don't contribute as much economically; and the very large marketing budgets of the pharma companies aren't paying for research. Given the size of these advertising budgets, estimated to match or be as much as twice the R&D budgets, the current system is revealed as being at least 50% inefficient, before considering any other aspects.
Then there are less obvious and harder to quantify inefficiencies, such as money wasted on ill-advised business acquisitions, political lobbying, excess costs borne by drug purchasers (who are pretty much over a barrel here) and so on.
It would be surprising if one couldn't beat 50% (and I believe it's considerably lower) efficiency, presuming one's goal here is in fact to produce medicines to cure and treat illnesses that plague our various societies.
One possible mechanism is university or research institute funding. It seems to work pretty well in physics or mathematics for example, and the budgets for large physics projects are getting pretty enormous. Why is this money spent? Because it has a good chance to better quality of life in the long term, and past experience has demonstrated this to be the case.
Medical research spending has much quicker payoffs than fundamental physics research.
Why would researchers bother developing treatments, if there were no big carrot being dangled before them? Two answers: firstly, the big carrots are being dangled in front of investors and board members, not the researchers; secondly, fame and renown work perfectly well in other scientific fields of endeavour, and are likely to work even more effectively in medicine where the benefits are so much more apparent to the public at large.
As rknop said, things aren't likely to change in a hurry, because the status quo benefits the people with the power more than the alternatives would. The alternatives though have the potential not only to be much more morally sound, but also to be more economically sound. Isn't nice to be able to take both an economic and moral highground?
Do I have to use a sarcasm tag? I keep forgetting.
This device looks really nice: switchable piggyback storage, FM transmittery thing, not too heavy, not too bulky, 20Gb, etc. etc. Of course ogg support is a big plus too. It could be the iPod competition we've all been waiting for! But ...
USB 1.1?! What were they thinking? How could they get so close and still drop the bundle?! Transfering a CD's worth of music onto the device would take well over a minute at any decent quality. Transferring a collection onto the drive would take hours. If there were no alternatives, then sure, it's certainly not too bad. But with a disk attached to the device, there's no good reason why transfers couldn't be ten times as fast, if only they used USB2 or firewire.
As a portable harddrive, USB1.1 speeds are apalling.
Would putting firewire or USB2 on really have been so hard? As it stands, the player seems to be in the 'so close but' category.
As Mosfet said, the solution is better organization, not wholesale throwing out of features!
Gnome 2 is the biggest disappointment for me. I was hoping for a Gnome 1.4 without the bugs, with a saner and friendlier GTK+ toolkit underpinning, with great internationalization support from Pango. If anything, gnome-terminal is now even buggier (with more features), while the severe cutting of options makes it impossible for me to set up a Gnome 2 desktop the way I would like to, short of hacking the code.
Examples:
- instead of 5 clocks, we have one. Don't like it? Well, cope. Blechh!
- want to use the Window List applet on a vertical panel? With the 'new and improved' option selection, one can't see the titles (due to the icons), and the buttons are supersized with no way of fixing it.
- No [apply] buttons. Great, so now when I make a selection, it applies it immediately, even if I want to do a number of things in one go and the individual steps are time-consuming to apply. Now changing settings takes longer
.
... and so on, so on.There was an old slashdot article reporting an interview with some UI expert [1], who made claims that configurability was a terrible thing. Yet the arguments used there were just invalid; they simply didn't apply. The principle of 'less-is-more' is at its base a good one: clutter is unattractive and unhelpful. But one should achieve this goal through better organisation, not by discarding things people are actually using! On my desktop I'd like the 'less-is-more' principle applied to reducing or eliminating excess toolbars, and maximising work space for example, but I don't see that happening in a hurry.
I may have to give KDE 3.1 another shot.
[1] speaking of lack of options limiting usability, how about a more functional slashdot search?!
The money argument does not hold water, because the very things that typically tie a site to a single platform are those which are the most expensive to produce.
Extensive javascript menus, elaborate flash 'navigators', exotic ActiveX controls -- all these things take time to produce, a lot more time in fact that the simple option which would have worked anywhere.
The problem isn't money or market-share, it's that so-called web designers are pandering to ignorant clients who want something pretty on their personal desktop rather than a useful web presence. Two groups are at fault: web-designers with no pride in their profession, and clients who are much more interested in spending their company's money on attractive interactive wallpaper than on an effective web site.
The 5% market share argument is an old canard parrotted by web quacks who won't learn new tricks.
The provision of so-called 'residential' services by broadband providers really disturbs me.
Typically in any area (thinking Europe, UK, Australia - true for US too?) there are only one or two high-speed providers to choose from. They offer two tiers of service: one is with a fixed IP, costing $lots per month and where one is charged by the incoming MB; the other is a residential service with a temporary IP -- that is often forcibly expired, killing connections etc. once or twice a day -- with an affordable cost and a relatively high cap before per-byte charging comes in.
These residential services though don't offer the Internet per se, but some sort of diluted version. No fixed IP means no reliable servers. No home-served content for you! I haven't yet seen a mainstream provider that offers IPv6 addresses; if lack of IPv4 addresses were the only motivation for this IP cycling game, then surely they'd offer a stable IPv6 address. The access agreements further compound the situation, with restrictions such as this 'no VPN', or no web serving, or only one computer on the connection, or no multiple accounts, or so on.
The dynamic IP stuff also means that one is pretty much forced to use an SMTP relay for outgoing mail, as so many sites blacklist known dynamic IP blocks out of hand. T-online here in Germany is about to start charging for their SMTP relay service!
The whole point of course is to extract the maximum amount of money out of the market. These service restrictions aren't there to cover otherwise present costs or the like, they're there to provide a differential betweeen their services, so that the providers can extract more money out of anyone who might possibly want to use the 'net for anything serious.
In the same way that major Telcos dragged their feet with ISDN and the like in the UK and in Australia, pricing it per minute _and_ per byte, and thereby siginificantly delaying the adoption of the 'net by businesses at large, the current practices are also limiting the adoption of the Internet as a tool for anything other than passive content consumption.
If there were a level marketplace for internet services, then the situation probably wouldn't last. But of course this isn't the case when there are $10^8 barriers to entry against an entrenched monopoly or duopoly.