Soon new versions of Firefox will start installing themselves in user-writable locations in each user account separately by default, just to make sure all network administrators hate it with a fiery passion.
Actually that would make our lives easier. So they won't do that. I think if the Firefox devs really cared what us administrators thought they'd have provided centralised Group Policy configuration back when we asked for it - ten years or so ago - instead of making us edit custom.js files and manually ship them to the workstations. But still, they haven't completely forgotten us; they keep the old Netscape Communicator era "profile" system with randomised path name, which nobody ever uses and which breaks as soon as you try to log into more than one workstation at once. Just to keep the old memories alive. All those arguments... all the flames... the futility, the anger, the betrayal.... so beautiful.
Now we feel nothing anymore, not even a vague sense of schadenfreude. Another gaping launch-day security hole? Yawn. Join the queue behind Adobe, Adobe, Adobe, Oracle and Microsoft.
Following your logic, a mugger is good example of a free market capitalist
Yes, pretty much. In fact, it's not just street crime that obeys the laws of laissez-faire economics. If you look at crime syndicates, they are pretty much pure capitalist concerns: they do things which make money, and make rational choices to maintain their market position, end of story. Because they are considered "illegal" enterprises they can't rely on government handouts for security, so they apply their own self-reliance and gumption and employ fully private security contractors, and when there's a fully free market in violence with no natural monopoly security provider we call that a "gang war".
If it turns out that people at large don't like organisations like the Mafia and the Zetas? Then people logically ought to deduce from the results that they don't actually want unrestrained pure capitalism, because the violence is a result of the rules those organisations are playing by - perfectly rational, intelligent, free-market rules.
You're right in saying that capitalism is not about altruism, but it's also not about exploiting the weaknesses of the system, and certainly isn't about using that advantage to curtail the efforts of others.
No, you don't want capitalism to be about those things. Capital itself, however, doesn't care about your irrelevant human wants unless it can turn them into money. It is what it is: an alien, mechanical force, utterly amoral, without kindness, without empathy, without sympathy for life of any kind. Murders and genocides and extinctions don't concern it, only dollars and basis points. Whatever we put into the game as an abstract "score", it will tend to optimise, right up to to the point of self-destruction.
We don't actually want capital to be our master; but the system we've built has enshrined it in that role. The results aren't going to be enjoyable unless we change our fundamental valuations of things.
What's actually wrong with Unity? Is there something you can point to, instead of just "ZOMG it's new I don't like it?"
Ok, here goes. A (possibly the) major problem with Unity - and the entire "app-centric" GUI ecosystem from the iDevice and tablet world which it and Windows 8 are aping - is that its focus on applications comes at the expense of documents. This reverses the trend from the 1980s onwards where GUIs were becoming increasingly about the user manipulating rich documents, and puts us right back in the old world of "your data is hard-coded into applications". But that simply isn't the case. Documents transcend applications; the application is just a means to an end.
Why? Two reasons. One, because applications churn faster than data does. For example: my music is a collection of.MP3 and.OGG files. It's over a decade old, and it's not going anywhere. My music player application, however, could be any of Rhythmbox, Banshee, Songbird, VLC. My photograph collection is a bunch of.JPG files. It's also not going to change. The default "photo manager" application (which I'm not sure there's even a need for) in Ubuntu, however, has switched from F-Spot to Shotwell, and then there's the GNOME Viewer if I just want to view them.
Second, there are multiple actions you might want to take with documents, and those different actions may require different applications. If I have a JPG, I *might* want to view it, or I might want to edit it. In that case I'm going to want to open it in GIMP, not Shotwell or Viewer. There's no way the OS can know in advance how I want to work with my data, so it shouldn't attempt to presume that it knows best.
The primary way this broken "applications first" mindset manifests in Unity is with the dock, and the way it groups windows by application rather than document. For instance, if I have two PDF files open, they're two completely separate documents; I want them to appear as two different icons. But no. Dock shows them as one instance of the PDF Viewer, and only once I click on them does it ask me which one I want. That's not at all what the user requires; it's an objective regression in usability (from the document-centric perspective) from even Windows 95's interface. But it's not a bug, it's a design decision, and it's come from inhaling uncritically the iDevice approach of "the app is everything".
I hope this app-focus is just a passing fad in the industry, because it reverses more than thirty years of user interface progress. It's been good news to app developers, as it assures them a privileged industry position and a revenue stream. But it's not good news to the user who wants the ability to sculpt their own document-centric workflow.
Also, while there wasn't third party native software initially on the iPhone, why didn't Palm and other devices end up having such huge third party software libraries?
Palm did. Their app market included e-books, too - I was buying and reading Stephen King's last three Dark Towet novels on my Palm Tungsten, in my pocket, the day they came out (2003), while the iPod was still just an overpriced movie player. I could play MP3s on my Tungsten too. About the same time we had a line of Palm-based smartphones as our corporate standard.
What killed the Palm is, well, Palm. The company ate a lot of silly pills and stopped evolving the hardware or the OS, so all those zillions of apps won't run on a modern platform any more than all the Commodore 64 or Apple II apps will run on your desktop PC or Mac.
2003 wasn't that long ago, you know. So young, and already so forgetful. It's kinda cute really.
what sort of lot will get all bent out of proportion because someone made a law with a stiff penalty to combat something that is morally and ethically wrong?
Philosophers?
There's probably an argument to be made that piracy is morally wrong but ethically right.
Beats me what the difference is between those two words, but I guess there must be one or they wouldn't both exist, eh?
Except that the only way for anti-piracy to "win" is to take general purpose computers out of the public's hands and move everyone into walled garden ecosystems
Precisely. I see you've caught up with the industry's move to Tablet and Cloud.
The same equations that led to the idea of a black hole have spinning solutions. The non-spinning solutions are a little easier and were found first, but the spinning case was quickly found afterward
Geek alert: Only if by "quickly" you mean "nearly fifty years later, due to a fluke, and during which time the field of General Relativity was almost abandoned".
Schwarzchild published his solution for non-rotating spherical masses (containing the singularity which implied black holes, which incidentally Einstein considered unphysical) in 1916, in the middle of World War I, right after Einstein released GR 1.0.
Roy Kerr didn't find the solution for rotating black holes until 1963, the year that Kennedy was assassinated and Doctor Who started broadcasting. And at the time, the mathematical mainstream opinion was that this solution was impossible. The Kerr metric led directly to the equating of quasars with black holes and was one of the key discoveries making the 1960s the "Golden Age of General Relativity", compared to the decades previously when GR had been a bit of a backwater. The "Golden Age" is also one of the reasons why Star Trek had warp drives, because space-time warping was suddenly a hot topic again in physics.
The Kerr-Newman solution for electrically charged rotating black holes, on the other hand, did come out "quickly" - only two years after Kerr's original breakthrough.
Disclaimer: I'm a complete physics layman who doesn't really grok GR at all, but I'm intrigued by the history of science and as a New Zealander I've paid some attention to Roy Kerr's story. And I'm also intrigued by the later Einstein and the fact that most of his work post 1915 was, then and now, considered as scientifically useless as his pre-1915 work was considered brilliant, even while he had become a cult pop icon.
Yes, I know the entire commercial world in 2012 has decided to remap the dictionary and call "design" what the world of commerce in 1982 would have called "style", and "architecture" and "engineering" what the world of 1982 would have called "design". And product designers no longer actually design things but just draw sketches of what the colouring of the pictures on the skin of the 3D printer will look like, while the product architects, who don't have architecture degrees, build flowcharts for the engineers, who don't hold engineering degrees, to build.
But darnit, I still remember when "design" meant how a product works at a technical level, and that's what I came to the article expecting to read, and that's the opposite of what I got.
Get off my perfectly manicured ironically Le Corbusier-inspired post-post-postmodernist lawn.
This is why you theoretically parse data before you accept it as input.
But we all know that in the real world, parsing is for sissies, mathematical proofs of correctness are for NASA and ivory tower perfectionists, and languages with automatic array bounds validation is something no self-respecting genius C++ programmer needs. You just keep hammering on the keys until it compiles, test it once on perfectly-formed input until it stops crashing long enough to ship it, and then it's Operations' problem to work out how to patch it, and Marketing's to change the product name to something hipper-sounding once the users have put out the fires.
Any data in a QR code that is invalid should only be marked as invalid
It shouldn't matter
There should be nothing processed but an ack that the data doesn't correspond to correct ranges.
Ah, "should". The system administrator's favourite word.
Followed closely by: "it can't" "did it just" "there's no way it just" "they say it's impossible that it could have" "their lawyers say they could never have foreseen that it would" "marketing give us every assurance that it absolutely probably maybe won't again"
Yes, let's go ahead and presume that the institutions that figuratively and in some cases literally built the first world nations we sit on our asses in have no idea how to sandbox and bound check a code read from a scanner in order to stop an "infection" from taking over.
Yes, that's pretty much it. We're that stupid.
As evidence for the prosecution, I present: Flash, Java, JPEG, PNG, PDF, Word.DOC, SQL, PHP, ASN.1, and TCP/IP.
All of the above are either sandboxed-by-design programming languages that don't expose binary code, or somple data encapsulation formats that aren't even Turing-complete. They're all in common use in industry. We, our peers, our industry, trumpeted their safety and deployed them. You'd expect that it would be pretty simple for implementation of a parser for any of these formats to at least not fail in catostraphically hilarious ways, eg, randomly snarfing a bunch of raw untrusted bytes into its runtime code page and then attempting to execute it. In much the same way that you might expect that the bricks a skyscraper is made from will not unexpectedly one day turn into penguins and fly away.
D'oh. Kevin J Anderson. I deeply apologise to all the hardworking Kevin J Adamsons in fanfic whose life I have inadvertently ruined by comparing to the perpetrator of Jedi Academy and Prelude to Dune.
"If you want to understand how English works and how it came to be, linguistics is for you. If you just want to be a writer, it's probably not."
That's exactly why Languages and Literature departments in universities are opposed warring camps.
You Philistines over in Languages only care about the mechanics! You have no soul! Literature has as much to do with brute words as... as computer science has to do with telescopes! Besides, it's all about the deconstruction of the articulation of the biopolitics of the transgressive post-post-ironic feminist inter(de)mediation now. You'd never understand.
Huh. Buncha latte-slurping coffee-shop hippies. Lets see you guys trek into Kazakhstan to catalogue the Indo-Ayran migration drift of the Mongolian antelope herders' nasal pluperfect tenses, or construct a crude field-expedient LL(R) parser for their iconographic system out of sticks and dead beetles. I did that last week. Had to kill a buncha snow leopards that got in my way.
You think that's hard? Over in Library Sciences we had to convert our whole stack from Dewey to Library of Congress classification. At midnight. By hand.
Hey guys! I'm the visiting speaker from the fiction writing workshop over in Adult Education! I've been writing professionally all my life and I just made a million bucks from my latest novel! Can I join in?
Eliminating the battery would be great as long as you can manage to make it through your day never getting more than a few inches from a charging station. I envision thousands of miles of charging trays built into every sidewalk, wainscott, chair-rail, railing, escalator, and countertop. You could have hanging charge-lines you could take with you when you cross streets like the overhead power used for trolley-cars.
Thank you for that awesome 1930s-radiopunk-dystopia mental image.
Seriously, I want a wallpaper of that.
(Once a day, like clockwork, New York City shuts down as Pirate Nikola Tesla broadcasts tendrils of free charge-lightning all across the Eastern Seaboard from his secret Magnifying Transmitter base. Desperate power-hungry citizens raise dirty vacuum tubes to the heavens to harvest illegal St Elmo's Fire. Meanwhile mammoth Edison Company marketing dirigibles, fresh from destroying the Martian invading force, drop Tripods full of patent lawyers in the Los Angeles desert to storm Fortress Hollywood...)
In short, what we need is not a new system, but a standard message format for social networks and a standard protocol for connecting to a social networking system
Yes. This.
I had high hopes, back in the day, of RSS/Atom, OpenID and Jabber/XMPP becoming just this - in fact I'm still constantly surprised that Twitter managed to get anywhere with Jabber being out there for free. Facebook at least does active matchmaking of friends and maintains a mapping of real name to handle - but Twitter doesn't even do that. It's just a web interface to an instant messaging system with persistent logs - however did that become something that couldn't be repeated by all the IRC servers in the world, let alone XMPP?
What I would like though, if we're starting afresh, is something like a data publishing protocol rather than oldschool blogging or messaging. Let's say JSON (it's not the best possible data standard, but it is widely supported). Can we get a protocol for giving me my own persistent namepace for publishing arbitrary JSON objects? Define a way of publishing standard data formats themselves described by JSON objects, such as data/time/timezone stamps, text blocks, HTML links, user names, photos, books, movies, GPS coordinates. Then let me link them together how I want: let's say I want to say "at universal/local time X I was in place Y, here's the OpenStreetMap reference. Here are my friends and hobbies. Here's my photos. Here's a review about a movie I just saw. Here's the Wikipedia link. Here's the IMDB link." And let me build views which can easily do lookup to find universal references like ISBNs rather than raw text strings for any of these which have been categorised.
The trick is to standardise as little as possible in the core - just the basics of publish, subscribe, and identity - and leave the system open to be evolved by the users. Turn "social media" back into a Web, but a web of personally-published data, where we can choose how much to publish and to whom.
Hey, I just invented Semantic Web! And schema.org. And a lot of other attempts. But how about we try to make it simple and doable, and just get it done? I still don't understand what the holdup is, and why we're still letting the Facebooks of the world set the tone.
if you want sub forums and privilidged users then you want forum sw..
No. Nobody really wants web forums. They're all universally awful both to admin and to use. People only put up with them because they don't realise how much better it could be.
What I want, and have wanted for some time, is a really basic publishing platform for small snippets of text. Something that I can use to assemble a blog, a wiki, a forum conversation, anything like that. Something that will automatically subscribe and publish to RSS, email, SMS, Twitter, Facebook gadget, whatever the cool viewing tech of the day is, but still keep the underlying content online, filed neatly and accessible. Something I can host on a public service or on my own server. Something that has a shared interlinked namespace so I can reply to friends' posts wherever they are.
It's not rocket science, or it shouldn't be, but we've somehow allowed Web publishing to stagnate into a hard division between pages, forums, wikis, and "social networks"; there really needn't be any such walls.
I just don't want to try to write it because the #1 web development tool for availability is PHP, which means it will get rooted instantly. I'd like to see, say, Wordpress, which has shaken most of its security bugs out now, evolve into something more general like this; but the signs aren't promising.
Start by replacing the "page" concept with "posts" and "views", and we'll be getting somewhere.
Twenty years ago if you wanted to put your pictures online you had to run your own server.
Fifteen years ago, though, you had GeoCities.
Which, if you weren't there (and I'm guessing most of you weren't - kids today, harumph) was basically a Facebook wall without messaging. We had "guest books" for that. But we did have image memes. The dancing baby. The Java "water droplets under the cursor" thing. The tiled background JPGs And the animated "under construction" GIFs, aieee. My eyes are still burning.
Do you have any reason to suspect that our understanding of quantum mechanics is generally unreliable or flawed in a relevant way?
Sure we do. One, the existence of extra infinities in QED which renormalisation works around shows that quantum field theories can't be the last mathematical word; they're at best an approximation which works as long as we don't extend it to all cases.
Two, the Standard Model containing a huge number of plugged-in variables shows that we don't yet have a fundamental theory which generates those numbers.
Three, the fact that we've yet to achieve a workable theory of quantum gravity despite >50 years of extremely bright effort shows that there are major fundamental mathematical incompatibilities between QED/QCD and General Relativity. The continuing failure of string theory to find its way out of the landscape problem suggests that such a mathematical resolution might not even exist.
Four, the laser, one of the most spectacular achievements of post-WW2 physics, actually was considered impossible by some of the leading QM theorists of the time. It works, but why did the theory suggest it wouldn't? Food for thought.
Five (although this generally worries philosophers and laypeople a lot more than it worries trained physicists, for some reason), at a deeper philosophical level the number of arbitrary postulates and principles which get waived and fudged on an ad-hoc fashion in the entire field of QM really ought to give us much pause. Relativity assumes a smooth continuum but QM assumes integer values of very small quantities, except where it doesn't (1/2 spin and 1/3 charge; why are fractions allowed in a supposedly final theory?) Parity is always conserved, except where it isn't. Spacetime ought to be quantized, but maybe it isn't. The wavefunction itself probably isn't quantized, but why? Spin isn't really spin, it's just a number. Fundamental particles are modelled as infinitely small points, but that would assume infinite energy density, which not only isn't quantizable, it doesn't really make sense. Antimatter is indistinguishable from ordinary matter travelling backwards in time (Feynman), but there's apparently no predictions that this makes, despite it being a huge reversal to our ordinary notions of causality. The proton is made of three quarks, except that it isn't; it's actually a uncountable sea of virtual quarks. What are virtual particles made of - it's obviously not "nothing" if it interacts - and why do we need to assume all this?
There's no wider motivation to any of these assumptions and many of them are mutually contradicting; except that each was introduced to solve a specific problem and then generated its own. To the outside observer, all this looks disturbingly like a series of Ptolemaic epicycles. Classical physics from Newton to Maxwell was clean and simple, required very few postulates to explain a wide range of phenomena, and was easy to explain. The fundamentals were small and logical and worked together without edge cases. Post-Einstein physics seems increasingly to be moving in the opposite direction: a huge bloated kernel where the "fundamental" layer is actually more complicated than the simpler emergent phenomena it claims to explain. This intuitively feels like it's putting the cart before the horse? But since Einstein and Bohr, "intuition" has been increasingly discredited in physics, and "counterintutitive" is seen as high praise, so there's not much leverage for philosophical criticism. And the sheer brute success of, eg, atomic explosions has been seen as empirical justification of the whole subatomic physics enterprise. But - to an outsider - it feels that, despite there being a huge volume of paper produced, basic physics progress (as opposed to engineering) is actually slowing exponentially since WW2, not accelerating.
Slashdot Mirror
Soon new versions of Firefox will start installing themselves in user-writable locations in each user account separately by default, just to make sure all network administrators hate it with a fiery passion.
Actually that would make our lives easier. So they won't do that. I think if the Firefox devs really cared what us administrators thought they'd have provided centralised Group Policy configuration back when we asked for it - ten years or so ago - instead of making us edit custom .js files and manually ship them to the workstations. But still, they haven't completely forgotten us; they keep the old Netscape Communicator era "profile" system with randomised path name, which nobody ever uses and which breaks as soon as you try to log into more than one workstation at once. Just to keep the old memories alive. All those arguments... all the flames... the futility, the anger, the betrayal.... so beautiful.
Now we feel nothing anymore, not even a vague sense of schadenfreude. Another gaping launch-day security hole? Yawn. Join the queue behind Adobe, Adobe, Adobe, Oracle and Microsoft.
Following your logic, a mugger is good example of a free market capitalist
Yes, pretty much. In fact, it's not just street crime that obeys the laws of laissez-faire economics. If you look at crime syndicates, they are pretty much pure capitalist concerns: they do things which make money, and make rational choices to maintain their market position, end of story. Because they are considered "illegal" enterprises they can't rely on government handouts for security, so they apply their own self-reliance and gumption and employ fully private security contractors, and when there's a fully free market in violence with no natural monopoly security provider we call that a "gang war".
If it turns out that people at large don't like organisations like the Mafia and the Zetas? Then people logically ought to deduce from the results that they don't actually want unrestrained pure capitalism, because the violence is a result of the rules those organisations are playing by - perfectly rational, intelligent, free-market rules.
You're right in saying that capitalism is not about altruism, but it's also not about exploiting the weaknesses of the system, and certainly isn't about using that advantage to curtail the efforts of others.
No, you don't want capitalism to be about those things. Capital itself, however, doesn't care about your irrelevant human wants unless it can turn them into money. It is what it is: an alien, mechanical force, utterly amoral, without kindness, without empathy, without sympathy for life of any kind. Murders and genocides and extinctions don't concern it, only dollars and basis points. Whatever we put into the game as an abstract "score", it will tend to optimise, right up to to the point of self-destruction.
We don't actually want capital to be our master; but the system we've built has enshrined it in that role. The results aren't going to be enjoyable unless we change our fundamental valuations of things.
What's actually wrong with Unity? Is there something you can point to, instead of just "ZOMG it's new I don't like it?"
Ok, here goes. A (possibly the) major problem with Unity - and the entire "app-centric" GUI ecosystem from the iDevice and tablet world which it and Windows 8 are aping - is that its focus on applications comes at the expense of documents. This reverses the trend from the 1980s onwards where GUIs were becoming increasingly about the user manipulating rich documents, and puts us right back in the old world of "your data is hard-coded into applications". But that simply isn't the case. Documents transcend applications; the application is just a means to an end.
Why? Two reasons. One, because applications churn faster than data does. For example: my music is a collection of .MP3 and .OGG files. It's over a decade old, and it's not going anywhere. My music player application, however, could be any of Rhythmbox, Banshee, Songbird, VLC. My photograph collection is a bunch of .JPG files. It's also not going to change. The default "photo manager" application (which I'm not sure there's even a need for) in Ubuntu, however, has switched from F-Spot to Shotwell, and then there's the GNOME Viewer if I just want to view them.
Second, there are multiple actions you might want to take with documents, and those different actions may require different applications. If I have a JPG, I *might* want to view it, or I might want to edit it. In that case I'm going to want to open it in GIMP, not Shotwell or Viewer. There's no way the OS can know in advance how I want to work with my data, so it shouldn't attempt to presume that it knows best.
The primary way this broken "applications first" mindset manifests in Unity is with the dock, and the way it groups windows by application rather than document. For instance, if I have two PDF files open, they're two completely separate documents; I want them to appear as two different icons. But no. Dock shows them as one instance of the PDF Viewer, and only once I click on them does it ask me which one I want. That's not at all what the user requires; it's an objective regression in usability (from the document-centric perspective) from even Windows 95's interface. But it's not a bug, it's a design decision, and it's come from inhaling uncritically the iDevice approach of "the app is everything".
I hope this app-focus is just a passing fad in the industry, because it reverses more than thirty years of user interface progress. It's been good news to app developers, as it assures them a privileged industry position and a revenue stream. But it's not good news to the user who wants the ability to sculpt their own document-centric workflow.
Er. Overpriced music player, I meant.
Also, while there wasn't third party native software initially on the iPhone, why didn't Palm and other devices end up having such huge third party software libraries?
Palm did. Their app market included e-books, too - I was buying and reading Stephen King's last three Dark Towet novels on my Palm Tungsten, in my pocket, the day they came out (2003), while the iPod was still just an overpriced movie player. I could play MP3s on my Tungsten too. About the same time we had a line of Palm-based smartphones as our corporate standard.
What killed the Palm is, well, Palm. The company ate a lot of silly pills and stopped evolving the hardware or the OS, so all those zillions of apps won't run on a modern platform any more than all the Commodore 64 or Apple II apps will run on your desktop PC or Mac.
2003 wasn't that long ago, you know. So young, and already so forgetful. It's kinda cute really.
what sort of lot will get all bent out of proportion because someone made a law with a stiff penalty to combat something that is morally and ethically wrong?
Philosophers?
There's probably an argument to be made that piracy is morally wrong but ethically right.
Beats me what the difference is between those two words, but I guess there must be one or they wouldn't both exist, eh?
Except that the only way for anti-piracy to "win" is to take general purpose computers out of the public's hands and move everyone into walled garden ecosystems
Precisely. I see you've caught up with the industry's move to Tablet and Cloud.
We're busily swapping the Cluetrain for the Zittrain.
The world will not be a better place when everyone and their dog can download and print their own guns.
Certainly not once the cats get their uranium centrifuges online.
Until someone responds, I'll be at Google.
Larry Page, what are you doing here?
The same equations that led to the idea of a black hole have spinning solutions. The non-spinning solutions are a little easier and were found first, but the spinning case was quickly found afterward
Geek alert: Only if by "quickly" you mean "nearly fifty years later, due to a fluke, and during which time the field of General Relativity was almost abandoned".
Schwarzchild published his solution for non-rotating spherical masses (containing the singularity which implied black holes, which incidentally Einstein considered unphysical) in 1916, in the middle of World War I, right after Einstein released GR 1.0.
Roy Kerr didn't find the solution for rotating black holes until 1963, the year that Kennedy was assassinated and Doctor Who started broadcasting. And at the time, the mathematical mainstream opinion was that this solution was impossible. The Kerr metric led directly to the equating of quasars with black holes and was one of the key discoveries making the 1960s the "Golden Age of General Relativity", compared to the decades previously when GR had been a bit of a backwater. The "Golden Age" is also one of the reasons why Star Trek had warp drives, because space-time warping was suddenly a hot topic again in physics.
The Kerr-Newman solution for electrically charged rotating black holes, on the other hand, did come out "quickly" - only two years after Kerr's original breakthrough.
Disclaimer: I'm a complete physics layman who doesn't really grok GR at all, but I'm intrigued by the history of science and as a New Zealander I've paid some attention to Roy Kerr's story. And I'm also intrigued by the later Einstein and the fact that most of his work post 1915 was, then and now, considered as scientifically useless as his pre-1915 work was considered brilliant, even while he had become a cult pop icon.
These are styling principles.
Yes, I know the entire commercial world in 2012 has decided to remap the dictionary and call "design" what the world of commerce in 1982 would have called "style", and "architecture" and "engineering" what the world of 1982 would have called "design". And product designers no longer actually design things but just draw sketches of what the colouring of the pictures on the skin of the 3D printer will look like, while the product architects, who don't have architecture degrees, build flowcharts for the engineers, who don't hold engineering degrees, to build.
But darnit, I still remember when "design" meant how a product works at a technical level, and that's what I came to the article expecting to read, and that's the opposite of what I got.
Get off my perfectly manicured ironically Le Corbusier-inspired post-post-postmodernist lawn.
This is why you theoretically parse data before you accept it as input.
But we all know that in the real world, parsing is for sissies, mathematical proofs of correctness are for NASA and ivory tower perfectionists, and languages with automatic array bounds validation is something no self-respecting genius C++ programmer needs. You just keep hammering on the keys until it compiles, test it once on perfectly-formed input until it stops crashing long enough to ship it, and then it's Operations' problem to work out how to patch it, and Marketing's to change the product name to something hipper-sounding once the users have put out the fires.
Um, if a programmer writes a single piece of code that _presumes_ anything about input, you fire his sorry ass
Oh, if only that were true. But on the other hand, it would make my life as a sysadmin so much simpler I'd be out of a job.
There's absolutely no reason
Any data in a QR code that is invalid should only be marked as invalid
It shouldn't matter
There should be nothing processed but an ack that the data doesn't correspond to correct ranges.
Ah, "should". The system administrator's favourite word.
Followed closely by:
"it can't"
"did it just"
"there's no way it just"
"they say it's impossible that it could have"
"their lawyers say they could never have foreseen that it would"
"marketing give us every assurance that it absolutely probably maybe won't again"
Yes, let's go ahead and presume that the institutions that figuratively and in some cases literally built the first world nations we sit on our asses in have no idea how to sandbox and bound check a code read from a scanner in order to stop an "infection" from taking over.
Yes, that's pretty much it. We're that stupid.
As evidence for the prosecution, I present: Flash, Java, JPEG, PNG, PDF, Word .DOC, SQL, PHP, ASN.1, and TCP/IP.
All of the above are either sandboxed-by-design programming languages that don't expose binary code, or somple data encapsulation formats that aren't even Turing-complete. They're all in common use in industry. We, our peers, our industry, trumpeted their safety and deployed them. You'd expect that it would be pretty simple for implementation of a parser for any of these formats to at least not fail in catostraphically hilarious ways, eg, randomly snarfing a bunch of raw untrusted bytes into its runtime code page and then attempting to execute it. In much the same way that you might expect that the bricks a skyscraper is made from will not unexpectedly one day turn into penguins and fly away.
You'd expect that, and you'd be dead wrong.
D'oh. Kevin J Anderson. I deeply apologise to all the hardworking Kevin J Adamsons in fanfic whose life I have inadvertently ruined by comparing to the perpetrator of Jedi Academy and Prelude to Dune.
Should all fan fiction be NYT Bestseller List worthy?
It should aspire to be a lot better. Kevin J Adamson's.... material.... regularly features on that list. The bar really isn't that high.
"If you want to understand how English works and how it came to be, linguistics is for you. If you just want to be a writer, it's probably not."
That's exactly why Languages and Literature departments in universities are opposed warring camps.
You Philistines over in Languages only care about the mechanics! You have no soul! Literature has as much to do with brute words as... as computer science has to do with telescopes! Besides, it's all about the deconstruction of the articulation of the biopolitics of the transgressive post-post-ironic feminist inter(de)mediation now. You'd never understand.
Huh. Buncha latte-slurping coffee-shop hippies. Lets see you guys trek into Kazakhstan to catalogue the Indo-Ayran migration drift of the Mongolian antelope herders' nasal pluperfect tenses, or construct a crude field-expedient LL(R) parser for their iconographic system out of sticks and dead beetles. I did that last week. Had to kill a buncha snow leopards that got in my way.
You think that's hard? Over in Library Sciences we had to convert our whole stack from Dewey to Library of Congress classification. At midnight. By hand.
Hey guys! I'm the visiting speaker from the fiction writing workshop over in Adult Education! I've been writing professionally all my life and I just made a million bucks from my latest novel! Can I join in?
(all turn and glare)
Eliminating the battery would be great as long as you can manage to make it through your day never getting more than a few inches from a charging station. I envision thousands of miles of charging trays built into every sidewalk, wainscott, chair-rail, railing, escalator, and countertop. You could have hanging charge-lines you could take with you when you cross streets like the overhead power used for trolley-cars.
Thank you for that awesome 1930s-radiopunk-dystopia mental image.
Seriously, I want a wallpaper of that.
(Once a day, like clockwork, New York City shuts down as Pirate Nikola Tesla broadcasts tendrils of free charge-lightning all across the Eastern Seaboard from his secret Magnifying Transmitter base. Desperate power-hungry citizens raise dirty vacuum tubes to the heavens to harvest illegal St Elmo's Fire. Meanwhile mammoth Edison Company marketing dirigibles, fresh from destroying the Martian invading force, drop Tripods full of patent lawyers in the Los Angeles desert to storm Fortress Hollywood...)
Do you even know who Borges is? Hint: he died long before Wikipedia ever existed...
He's so unhip, when you say Dylan
He thinks you're talking about Dylan Thomas,
whoever he was.
The man ain't got no culture.
In short, what we need is not a new system, but a standard message format for social networks and a standard protocol for connecting to a social networking system
Yes. This.
I had high hopes, back in the day, of RSS/Atom, OpenID and Jabber/XMPP becoming just this - in fact I'm still constantly surprised that Twitter managed to get anywhere with Jabber being out there for free. Facebook at least does active matchmaking of friends and maintains a mapping of real name to handle - but Twitter doesn't even do that. It's just a web interface to an instant messaging system with persistent logs - however did that become something that couldn't be repeated by all the IRC servers in the world, let alone XMPP?
What I would like though, if we're starting afresh, is something like a data publishing protocol rather than oldschool blogging or messaging. Let's say JSON (it's not the best possible data standard, but it is widely supported). Can we get a protocol for giving me my own persistent namepace for publishing arbitrary JSON objects? Define a way of publishing standard data formats themselves described by JSON objects, such as data/time/timezone stamps, text blocks, HTML links, user names, photos, books, movies, GPS coordinates. Then let me link them together how I want: let's say I want to say "at universal/local time X I was in place Y, here's the OpenStreetMap reference. Here are my friends and hobbies. Here's my photos. Here's a review about a movie I just saw. Here's the Wikipedia link. Here's the IMDB link." And let me build views which can easily do lookup to find universal references like ISBNs rather than raw text strings for any of these which have been categorised.
The trick is to standardise as little as possible in the core - just the basics of publish, subscribe, and identity - and leave the system open to be evolved by the users. Turn "social media" back into a Web, but a web of personally-published data, where we can choose how much to publish and to whom.
Hey, I just invented Semantic Web! And schema.org. And a lot of other attempts. But how about we try to make it simple and doable, and just get it done? I still don't understand what the holdup is, and why we're still letting the Facebooks of the world set the tone.
if you want sub forums and privilidged users then you want forum sw..
No. Nobody really wants web forums. They're all universally awful both to admin and to use. People only put up with them because they don't realise how much better it could be.
What I want, and have wanted for some time, is a really basic publishing platform for small snippets of text. Something that I can use to assemble a blog, a wiki, a forum conversation, anything like that. Something that will automatically subscribe and publish to RSS, email, SMS, Twitter, Facebook gadget, whatever the cool viewing tech of the day is, but still keep the underlying content online, filed neatly and accessible. Something I can host on a public service or on my own server. Something that has a shared interlinked namespace so I can reply to friends' posts wherever they are.
It's not rocket science, or it shouldn't be, but we've somehow allowed Web publishing to stagnate into a hard division between pages, forums, wikis, and "social networks"; there really needn't be any such walls.
I just don't want to try to write it because the #1 web development tool for availability is PHP, which means it will get rooted instantly. I'd like to see, say, Wordpress, which has shaken most of its security bugs out now, evolve into something more general like this; but the signs aren't promising.
Start by replacing the "page" concept with "posts" and "views", and we'll be getting somewhere.
Twenty years ago if you wanted to put your pictures online you had to run your own server.
Fifteen years ago, though, you had GeoCities.
Which, if you weren't there (and I'm guessing most of you weren't - kids today, harumph) was basically a Facebook wall without messaging. We had "guest books" for that. But we did have image memes. The dancing baby. The Java "water droplets under the cursor" thing. The tiled background JPGs And the animated "under construction" GIFs, aieee. My eyes are still burning.
Remember when every ISP offered email? I guess they still do - but nobody cares. There are 3ish winners in the west: Yahoo, Hotmail, Gmail.
None of which I use, and the majority of people in my email to-list are not on one of those.
Do you have any reason to suspect that our understanding of quantum mechanics is generally unreliable or flawed in a relevant way?
Sure we do. One, the existence of extra infinities in QED which renormalisation works around shows that quantum field theories can't be the last mathematical word; they're at best an approximation which works as long as we don't extend it to all cases.
Two, the Standard Model containing a huge number of plugged-in variables shows that we don't yet have a fundamental theory which generates those numbers.
Three, the fact that we've yet to achieve a workable theory of quantum gravity despite >50 years of extremely bright effort shows that there are major fundamental mathematical incompatibilities between QED/QCD and General Relativity. The continuing failure of string theory to find its way out of the landscape problem suggests that such a mathematical resolution might not even exist.
Four, the laser, one of the most spectacular achievements of post-WW2 physics, actually was considered impossible by some of the leading QM theorists of the time. It works, but why did the theory suggest it wouldn't? Food for thought.
Five (although this generally worries philosophers and laypeople a lot more than it worries trained physicists, for some reason), at a deeper philosophical level the number of arbitrary postulates and principles which get waived and fudged on an ad-hoc fashion in the entire field of QM really ought to give us much pause. Relativity assumes a smooth continuum but QM assumes integer values of very small quantities, except where it doesn't (1/2 spin and 1/3 charge; why are fractions allowed in a supposedly final theory?) Parity is always conserved, except where it isn't. Spacetime ought to be quantized, but maybe it isn't. The wavefunction itself probably isn't quantized, but why? Spin isn't really spin, it's just a number. Fundamental particles are modelled as infinitely small points, but that would assume infinite energy density, which not only isn't quantizable, it doesn't really make sense. Antimatter is indistinguishable from ordinary matter travelling backwards in time (Feynman), but there's apparently no predictions that this makes, despite it being a huge reversal to our ordinary notions of causality. The proton is made of three quarks, except that it isn't; it's actually a uncountable sea of virtual quarks. What are virtual particles made of - it's obviously not "nothing" if it interacts - and why do we need to assume all this?
There's no wider motivation to any of these assumptions and many of them are mutually contradicting; except that each was introduced to solve a specific problem and then generated its own. To the outside observer, all this looks disturbingly like a series of Ptolemaic epicycles. Classical physics from Newton to Maxwell was clean and simple, required very few postulates to explain a wide range of phenomena, and was easy to explain. The fundamentals were small and logical and worked together without edge cases. Post-Einstein physics seems increasingly to be moving in the opposite direction: a huge bloated kernel where the "fundamental" layer is actually more complicated than the simpler emergent phenomena it claims to explain. This intuitively feels like it's putting the cart before the horse? But since Einstein and Bohr, "intuition" has been increasingly discredited in physics, and "counterintutitive" is seen as high praise, so there's not much leverage for philosophical criticism. And the sheer brute success of, eg, atomic explosions has been seen as empirical justification of the whole subatomic physics enterprise. But - to an outsider - it feels that, despite there being a huge volume of paper produced, basic physics progress (as opposed to engineering) is actually slowing exponentially since WW2, not accelerating.