On the other hand, autofocus isn't as bad as you imply. I have mine set to "center point only", and simply put the little dot over what I want to focus on, focus, reframe the shot how I want it to be, and shoot. It's pretty painless.
Oh, how do I hate that. Let me count the reasons:
Focusing before composing is back-asswards, if you ask me.
There's the shutter release button half-press tango, if you press it too hard, then you have to start all over again. At least nowadays with digital wasting a frame is almost completely costless.
It doesn't work if you're shooting close-up, because when you recompose after locking focus, you will very likely change the sensor/subject distance by a significant fraction. The lens I most often use on my Nikons focuses to 25cm, achieves a 1:4 reproduction ratio, and I often use it very close.
But I have a couple of (wonderful) old OM manual focus lenses, and those are a pain in the butt to use without a split prism. Note that you CAN get these focus screens installed as an aftermarket thing -- the company that makes them is called Katz Eye, I think, and I've heard good things about them.
Actually, what I'm thinking of isn't the split prisms (which as you should notice, don't help at all with the close focus situation); it's just the plain old ground glass + fast lens, where it was possible to judge focus decently even without the split prism. I don't know what they do today (I think it has something to do with the AF beamsplitter), but judging focus from the screen without an aid on a modern small format camera is too hard. Hard enough that in the end, at the sub-$1,000 price point, I find that the EVF on the Panasonic G1 and GH1 shows the way forward.
Good lenses are becoming a lost art these days. When autofocus SLRs came out, Canon had a f1.0 50mm lens, some insane telephoto lenses, and many other items. Nikon had a 2000mm zoom lens. Lens quality was also better, because with digital cameras, the camera can recognize a lens, then apply corrections for lens imperfections to the data before it gets dropped into the memory card. This is impossible with film, so the image had to be perfect the first time.
Actually, I don't think most of those are an example of a real problem:
The f/1.0 lenses are simply not as important anymore, because quality at higher ISO sensitivities has gone up. If you used that f/1.0 lens with ISO 100 color film, you can probably get better results today by using f/2.0 at ISO 400 on a DSLR (more depth of field!). If you were using that f/1.0 with ISO 400 color film, then you could use an f/1.4 lens at ISO 800 on digital.
I think digital correction of lens imperfections is a perfectly fine idea. All lenses are tradeoffs between lots of factors. If by correcting curvilinear distortion digitally you can improve some other parameter of the lens, then that's a very good idea. (And the parameter improved could be price or size, for all I care). In fact, I recently bought a Panasonic DMC-G1, whose kit lens was designed with rather severe barrel distortion at the widest zoom setting, and digital correction thereof. I did so in full knowledge, and I'm not complaining; the lens is pretty sharp, small and light.
Which lens do you refer to by the "2000mm zoom"? Is it the 1200-1700mm f/5.6-8 zoom, or is it the 2000mm f/11 reflex? That kind of lens have always been built-to-order products that hardly anybody has ever seen, much less taken photos with, so for nearly everybody that has ever lived, the difference is zero. Which mega-exotic lenses are offered is really a function of demand by a very few users, but also the potential for substituting other products; for example, an APS-C sensor camera attached to a 600mm f/4 on a 2x teleconverter will produce the same angle of view as a 1800mm lens on a 35mm camera, and be a lot cheaper than asking Nikon to custom-make a lens for you. Though lens manufacturers still build some pretty exotic stuff, just not the same as before. Think of Sigma's 200-500mm f/2.8 Not nearly as long as those Nikons, but it's f/2.8.
The "awful interface" is not a problem with $500 dSLRs...it's a problem with most dSLRs today, regardless of price.
I don't disagree with you, but really, I'm thinking of something very concrete. A Nikon D40 has one control dial in the grip. A D90 has two, one in the back and one in the front; with the appropriate custom settings, it's actually reasonably usable. (Set the program exposure mode, use the back dial for program shift, set custom mode that enables the front dial to change exposure compensation without having to press a button).
Though don't get me started on autofocus, because I just hate that shit. I mean, I understand that there are some people who actually do need it (e.g., sports photographers), but for most of us, well, all that stuff with focus points and focus modes and tracking and focus lock is just too much more complicated than the good old manual focus interface: "turn this ring until the thing you want to be sharp looks sharp." But of course, to give autofocus SLRs to the few people who actually need them, they killed off manual focus SLRs, replaced the viewfinder focus screens with these new ones that are hard to focus manually with, and replaced all the manual focus lenses with AF lenses that have too short focusing throws and crappy feel (and increasingly often, are throwing out the focus distance and depth of field scales). Grrrr.
Anyway, a D40 kit is under $500, while a D90 kit is about $1,000. The D90 is a lot more usable than the D40 thanks to having two control dials and a bigger viewfinder. That's the most important difference between them.
The results you get from the average $300 Canon Powershot are far better than what you had with your $500 SLR back in the day, especially considering the lenses you could afford then.
No, not at all. A really basic 35mm SLR kit back in the day of the Pentax K1000 would have had a 50mm f/2 lens or better, which completely trounces the lens on a Powershot, period.
You have to understand that over the history of photography, most of the advances haven't been about image quality; they've been about size, speed and convenience. Even changes that at first glance seem to be about image quality (e.g., aspherical lens elements, special types of glass) were actually about improving the optical performance of small format cameras to get acceptable pictures out of them. So that 50mm f/2 film SLR kit lens actually gave way to crappy slow kit zoom lenses, because, well, zooms are convenient. Same applies to the Powershot vs. old 35mm SLR. The Powershot doesn't have the optical quality, but the users don't care, period, because (a) they can see their photos right away, (b) they can carry less stuff, (c) they can email the photos easily. Same sorts of advantages apply to DSLRs, plus (d) you get comparatively less noise at higher ISO speeds, and (e) you don't have to muck around with color correction filters to get the correct white balance.
Both Canon and Nikon still make 35mm film SLRs. Canon only makes an autofocus film body for about $900, but Nikon sells a student manual focus camera, the FM10. Amazon's selling it (through a third party vendor) for $290 with a kit zoom lens (35-70mm, f/3.5-4.8, not very good for a film student). They even still sell a decent selection of manual focus lenses for it (and you can still use the Nikon non-G autofocus lenses on those cameras, so the selection is even wider).
Still, you'd be silly to buy this kind of camera new. There's a big glut of comparable used manual focus bodies out there that will make perfectly good cameras. I can wholly recommend Minolta SRT (which I've used), and the other systems (Olympus, Canon FD, Pentax K, etc.) are also good.
But you can still get great photos with a $500 SLR. You won't get a full frame sensor in that price range, but either of the Canon or Nikon entry level dSLRs with interchangable lenses will give great photos. And, just like an entry level SLR film camera, you can still build a nice lens collection that's ready to use with a more expensive full frame sensor camera later.
The problem at $500 isn't the image quality, indeed. The problems are (a) awful tiny viewfinders, (b) awful interface (one control dial does everything, so you must hold down all sorts of tiny indistinct buttons while turning it), (c) often you don't have the option but to buy without the kit zoom lens.
Indeed. Kodachrome for many years has been a fringe old-style color slide film that's been mostly replaced with E-6 process film for many years now (the E-6 process dates from 1977). Kodak's certainly not discontinuing their Ektachrome E-6 process films.
Yeah, that's more or less what I figured after reading a bit more through stuff. The article Slashdot is sourcing this from is just clueless about what the real differentiating point is; it's not the fact that it's OLAP, it's the UI and integration with other Google or web data.
Looking over the actual Google blog announcement, this looks more like a case of the F article getting it all wrong. The "dimensionality" stuff is clearly not intended to be the innovation or selling point of Google's service; much less a differentiator relative to database vendors, who've had OLAP for ages.
The real selling points seem to be an easy UI, a lot of predefined public data sets available to combine and correlate with your own data, and the collaboration features.
Now, the news media's choice of tone and language in reporting on H1N1 is another matter entirely.
Now, I don't want to say that the media did a perfect job by any means, but in the early days of H1N1's spread in Mexico, the number of deaths reasonably attributed to it and the age distribution of the victims was truly very worrisome. It was only clear after a couple of weeks that those early numbers weren't truly representative.
"Mexican Flu". With the politically correct liberal media, we can't have that name or it will possibly hurt the tourism in Mexico.
Let us grant, very strictly for the sake of argument, that the fact that the virus isn't being called "Mexican Flu" (a name that has never been in widespread use for it, but I digress) is because the media, which is, again for the sake of argument is in fact liberal and "politically correct," doesn't want to hurt tourism in Mexico.
Now, what exactly is supposed to be wrong with that goal? Isn't it a virtue to help other people? Are you actually telling us that we should instead change the name of the current H1N1 flu outbreak to "Mexican Flu" in order to hurt tourism to Mexico? Why would anybody want to do this?
Isn't this the sort of thing that a google search appliance would be helpful for?
Yes, but don't make the mistake of thinking that just because Google are the leading web search engine, they must also be the leading document search solution. Google's web search relies heavily on links between HTML documents to assess their relative importance. In an office with a lot of plain old documents, there will be no links.
4) The batteries now have way more battery life, which isn't "worsening" the battery situation in my book. Perhaps you're referring to the fact that the battery is not removable? I don't see that as a major issue. How often does a MacBook Pro user replace their battery?
I'll have to take issue with that one. I had a Powerbook from 2004 to about 2006, and a Macbook Pro from 2006 and ongoing, and in both cases the battery went to hell in less than two years. Maybe the new battery technology they're using improves this, I don't know, but I really am not inclined to trust them on this.
What the statistics actually show is that if you replaced all of the hours you spend in a car on the road with hours in a plane in the sky, you're chances of being injured or killed are still *lower*.
I'll buy that your chances are dying are lower, but brother, if you truly spent that much time in passenger planes, you'd lose your legs to injury!
This is the central argument of an excellent book called The Nurture Assumption by Judith Rich Harris.
No, not quite. Harris is arguing that parents have little influence over their children. I was just challenging what I see as an implicit assumption in the post I was responding to, that mothers have near-exclusive influence over their children's values.
How is it, then, that women find themselves the victim of "social gender roles?" Men, I think, in a very real sense, do not make society. Women do. Women raise kids and instill values in them men's behavior is almost entirely based on doing things that will score and keep women.
Have you considered the possibility that children actually don't acquire their values exclusively from their mothers? But rather, acquire them from their interaction with the culture at large? Have you considered the possibility that, just for example, schools are sites of sustainable transmission of values between the children themselves? So that kids end up learning a very large chunk of their values from peers and kids slightly above their grade.
And what about the constant portrayal of gender roles in the media? Are you also absolutely convinced that that has no part to play? Or, also, what about the fact that until relatively recently in our culture, licit sexual access to women was negotiated between the suitor and the woman's father? Are you absolutely sure that our culture contains no residues of that? Like, for example, are you sure that men's behavior toward women is always truly aimed at gaining the women's favor as an end in itself, and never as, say, a means towards winning an imagined competition between men?
You've considered all of this and more, and correctly discarded all of it, right?
People seem to assume that what is happening is that previously, cultural norms dictated gender inequality when there was no biological basis, and now that those norms have changed, biological equality is restored. Couldn't it be the other way around? I.e. that there is a biological inequality, that is being altered by cultural factors to produce equality?
But the simple problem here is that everybody grows up in a culture, and that this throws the whole concept of "biological ability" out the window. There is no way to establish a baseline level of purely biological ability; all you can do is measure relative differences in the end-results of different cultures.
If you have Culture A with equal outcome, and Culture B with unequal outcome, there is simply no way of deciding that one of these outcomes is "more natural" than the other without making some value judgement about the cultures.
Women tend to gravitate towards fields which there is a degree of socializing, such as education, medicine (Regular and veterinary), and communications. Men tend to gravitate towards either exciting fields, or fields which they feel will be financial rewarding.
You're missing the point by a mile. TFA isn't about whether this is so, but rather about why this is so. There is a relatively prominent set of people who insist in attributing this kind of thing to innate differences between the genders; TFA is mentioning studies that rebut that claim, and rather support the counterclaim that the differences are due to culture.
There is a separate question here that TFA doesn't discuss, but which your quote does bring up: pay differences. I'm not going to argue this one way or the other, but there's a question to be asked as to what extent men gravitate towards those jobs because they're financially rewarding, versus to what extent the jobs are financially rewarding because they're done by men. I know it's hard to think of the latter alternative, but basically, it comes down to the power to set the relative prices for different kinds of labor being overwhelmingly in mens' hands.
The problem with this reasoning is that executive orders are not laws.
You're mixing up the concepts of law and statute. Executive orders aren't statutes, but they sure are law. Same thing with case law.
Re:Why only one database language?
on
SQL in a Nutshell
·
· Score: 1
It depends on what one means by "eliminating nulls"; there's a reason I mentioned NULL and 3-valued logic in the same breath. I certainly would welcome a 2-valued system where NULL was an optionally allowed value for every column (treated as a true value, and not the funny "no value" semantics).
Or even better, we could just have a simplified version of something like Haskell's labeled union types. This would take care of NULL and enumerated types in one feature.
Re:Why only one database language?
on
SQL in a Nutshell
·
· Score: 2, Insightful
I disagree. Existing query optimization and execution engines are built with the capabilities and constructs of SQL in mind. A new query language, on the other hand, presumably exists to enable things which existing languages (like SQL) don't... otherwise, why would you bother? So, given that, it's very likely that the language would require new, novel technologies for optimization and implementation... and those things ain't easy to get right.
I think DragonWriter is going from the core assumption (which I share) that we're talking about a relational language alternative to SQL, where the differences would be along the following lines:
Closer adherence to the relational model (no NULLs or three-valued logic, no row duplication allowed, etc.).
Friendlier query language with more capabilities for factoring complex repetitive queries.
Note that existing SQL systems also have had many kinds of features added over the years that often require new optimization and implementation techniques (e.g., recursive query extensions, materialized views with automatic query rewrite, MERGE statement). In this sense, really, a relational system with a novel language doesn't have it intrinsically harder than existing SQL databases.
Re:Why only one database language?
on
SQL in a Nutshell
·
· Score: 3, Insightful
I think you very much overstate the "clarity," "simplicity" and "straighforwardness" of SQL; there are many well-studied ways in which a relational language could be better than SQL. But I do think you've hit the nail on the head otherwise: designing and implementing a credible RDBMS is extremely hard compared to designing and implementing a programming language, and no relational query language is going to go anywhere without being paired to a good RDBMS.
Interesting...but what's to stop you from using it to write a smallish library, and then linking the library to other software?
The fact that (a) the compiler doesn't output linkable object files at all, (b) the fact that to get the maximum performance boost, the compiler needs to optimize across the application/library boundary. The latter becomes particularly important in functional languages where you have library routines that accept functions as arguments--a whole program compiler is capable of inlining the code for a lambda into the library routine that it's passed to, and then optimizing the result aggressively.
Slashdot Mirror
Oh, how do I hate that. Let me count the reasons:
Actually, what I'm thinking of isn't the split prisms (which as you should notice, don't help at all with the close focus situation); it's just the plain old ground glass + fast lens, where it was possible to judge focus decently even without the split prism. I don't know what they do today (I think it has something to do with the AF beamsplitter), but judging focus from the screen without an aid on a modern small format camera is too hard. Hard enough that in the end, at the sub-$1,000 price point, I find that the EVF on the Panasonic G1 and GH1 shows the way forward.
Actually, I don't think most of those are an example of a real problem:
I don't disagree with you, but really, I'm thinking of something very concrete. A Nikon D40 has one control dial in the grip. A D90 has two, one in the back and one in the front; with the appropriate custom settings, it's actually reasonably usable. (Set the program exposure mode, use the back dial for program shift, set custom mode that enables the front dial to change exposure compensation without having to press a button).
Though don't get me started on autofocus, because I just hate that shit. I mean, I understand that there are some people who actually do need it (e.g., sports photographers), but for most of us, well, all that stuff with focus points and focus modes and tracking and focus lock is just too much more complicated than the good old manual focus interface: "turn this ring until the thing you want to be sharp looks sharp." But of course, to give autofocus SLRs to the few people who actually need them, they killed off manual focus SLRs, replaced the viewfinder focus screens with these new ones that are hard to focus manually with, and replaced all the manual focus lenses with AF lenses that have too short focusing throws and crappy feel (and increasingly often, are throwing out the focus distance and depth of field scales). Grrrr.
Anyway, a D40 kit is under $500, while a D90 kit is about $1,000. The D90 is a lot more usable than the D40 thanks to having two control dials and a bigger viewfinder. That's the most important difference between them.
No, not at all. A really basic 35mm SLR kit back in the day of the Pentax K1000 would have had a 50mm f/2 lens or better, which completely trounces the lens on a Powershot, period.
You have to understand that over the history of photography, most of the advances haven't been about image quality; they've been about size, speed and convenience. Even changes that at first glance seem to be about image quality (e.g., aspherical lens elements, special types of glass) were actually about improving the optical performance of small format cameras to get acceptable pictures out of them. So that 50mm f/2 film SLR kit lens actually gave way to crappy slow kit zoom lenses, because, well, zooms are convenient. Same applies to the Powershot vs. old 35mm SLR. The Powershot doesn't have the optical quality, but the users don't care, period, because (a) they can see their photos right away, (b) they can carry less stuff, (c) they can email the photos easily. Same sorts of advantages apply to DSLRs, plus (d) you get comparatively less noise at higher ISO speeds, and (e) you don't have to muck around with color correction filters to get the correct white balance.
Both Canon and Nikon still make 35mm film SLRs. Canon only makes an autofocus film body for about $900, but Nikon sells a student manual focus camera, the FM10. Amazon's selling it (through a third party vendor) for $290 with a kit zoom lens (35-70mm, f/3.5-4.8, not very good for a film student). They even still sell a decent selection of manual focus lenses for it (and you can still use the Nikon non-G autofocus lenses on those cameras, so the selection is even wider).
Still, you'd be silly to buy this kind of camera new. There's a big glut of comparable used manual focus bodies out there that will make perfectly good cameras. I can wholly recommend Minolta SRT (which I've used), and the other systems (Olympus, Canon FD, Pentax K, etc.) are also good.
The problem at $500 isn't the image quality, indeed. The problems are (a) awful tiny viewfinders, (b) awful interface (one control dial does everything, so you must hold down all sorts of tiny indistinct buttons while turning it), (c) often you don't have the option but to buy without the kit zoom lens.
Indeed. Kodachrome for many years has been a fringe old-style color slide film that's been mostly replaced with E-6 process film for many years now (the E-6 process dates from 1977). Kodak's certainly not discontinuing their Ektachrome E-6 process films.
Yeah, that's more or less what I figured after reading a bit more through stuff. The article Slashdot is sourcing this from is just clueless about what the real differentiating point is; it's not the fact that it's OLAP, it's the UI and integration with other Google or web data.
Looking over the actual Google blog announcement, this looks more like a case of the F article getting it all wrong. The "dimensionality" stuff is clearly not intended to be the innovation or selling point of Google's service; much less a differentiator relative to database vendors, who've had OLAP for ages.
The real selling points seem to be an easy UI, a lot of predefined public data sets available to combine and correlate with your own data, and the collaboration features.
How's this three dimensional stuff not just plain old OLAP?
Now, I don't want to say that the media did a perfect job by any means, but in the early days of H1N1's spread in Mexico, the number of deaths reasonably attributed to it and the age distribution of the victims was truly very worrisome. It was only clear after a couple of weeks that those early numbers weren't truly representative.
Let us grant, very strictly for the sake of argument, that the fact that the virus isn't being called "Mexican Flu" (a name that has never been in widespread use for it, but I digress) is because the media, which is, again for the sake of argument is in fact liberal and "politically correct," doesn't want to hurt tourism in Mexico.
Now, what exactly is supposed to be wrong with that goal? Isn't it a virtue to help other people? Are you actually telling us that we should instead change the name of the current H1N1 flu outbreak to "Mexican Flu" in order to hurt tourism to Mexico? Why would anybody want to do this?
Yes, but don't make the mistake of thinking that just because Google are the leading web search engine, they must also be the leading document search solution. Google's web search relies heavily on links between HTML documents to assess their relative importance. In an office with a lot of plain old documents, there will be no links.
They didn't really remove it. They just rebranded it as the Macbook Pro 13".
I'll have to take issue with that one. I had a Powerbook from 2004 to about 2006, and a Macbook Pro from 2006 and ongoing, and in both cases the battery went to hell in less than two years. Maybe the new battery technology they're using improves this, I don't know, but I really am not inclined to trust them on this.
I'll buy that your chances are dying are lower, but brother, if you truly spent that much time in passenger planes, you'd lose your legs to injury!
No, not quite. Harris is arguing that parents have little influence over their children. I was just challenging what I see as an implicit assumption in the post I was responding to, that mothers have near-exclusive influence over their children's values.
Have you considered the possibility that children actually don't acquire their values exclusively from their mothers? But rather, acquire them from their interaction with the culture at large? Have you considered the possibility that, just for example, schools are sites of sustainable transmission of values between the children themselves? So that kids end up learning a very large chunk of their values from peers and kids slightly above their grade.
And what about the constant portrayal of gender roles in the media? Are you also absolutely convinced that that has no part to play? Or, also, what about the fact that until relatively recently in our culture, licit sexual access to women was negotiated between the suitor and the woman's father? Are you absolutely sure that our culture contains no residues of that? Like, for example, are you sure that men's behavior toward women is always truly aimed at gaining the women's favor as an end in itself, and never as, say, a means towards winning an imagined competition between men?
You've considered all of this and more, and correctly discarded all of it, right?
But the simple problem here is that everybody grows up in a culture, and that this throws the whole concept of "biological ability" out the window. There is no way to establish a baseline level of purely biological ability; all you can do is measure relative differences in the end-results of different cultures.
If you have Culture A with equal outcome, and Culture B with unequal outcome, there is simply no way of deciding that one of these outcomes is "more natural" than the other without making some value judgement about the cultures.
You're missing the point by a mile. TFA isn't about whether this is so, but rather about why this is so. There is a relatively prominent set of people who insist in attributing this kind of thing to innate differences between the genders; TFA is mentioning studies that rebut that claim, and rather support the counterclaim that the differences are due to culture.
There is a separate question here that TFA doesn't discuss, but which your quote does bring up: pay differences. I'm not going to argue this one way or the other, but there's a question to be asked as to what extent men gravitate towards those jobs because they're financially rewarding, versus to what extent the jobs are financially rewarding because they're done by men. I know it's hard to think of the latter alternative, but basically, it comes down to the power to set the relative prices for different kinds of labor being overwhelmingly in mens' hands.
You're mixing up the concepts of law and statute. Executive orders aren't statutes, but they sure are law. Same thing with case law.
It depends on what one means by "eliminating nulls"; there's a reason I mentioned NULL and 3-valued logic in the same breath. I certainly would welcome a 2-valued system where NULL was an optionally allowed value for every column (treated as a true value, and not the funny "no value" semantics).
Or even better, we could just have a simplified version of something like Haskell's labeled union types. This would take care of NULL and enumerated types in one feature.
I think DragonWriter is going from the core assumption (which I share) that we're talking about a relational language alternative to SQL, where the differences would be along the following lines:
Note that existing SQL systems also have had many kinds of features added over the years that often require new optimization and implementation techniques (e.g., recursive query extensions, materialized views with automatic query rewrite, MERGE statement). In this sense, really, a relational system with a novel language doesn't have it intrinsically harder than existing SQL databases.
I think you very much overstate the "clarity," "simplicity" and "straighforwardness" of SQL; there are many well-studied ways in which a relational language could be better than SQL. But I do think you've hit the nail on the head otherwise: designing and implementing a credible RDBMS is extremely hard compared to designing and implementing a programming language, and no relational query language is going to go anywhere without being paired to a good RDBMS.
The fact that (a) the compiler doesn't output linkable object files at all, (b) the fact that to get the maximum performance boost, the compiler needs to optimize across the application/library boundary. The latter becomes particularly important in functional languages where you have library routines that accept functions as arguments--a whole program compiler is capable of inlining the code for a lambda into the library routine that it's passed to, and then optimizing the result aggressively.