Well I certainly can't speak to the linguistics aspect, but I didn't recognize date and population size numbers to be totally made up; there is some research (peer reviewed at least -- this isn't my area) putting initial expansion ~65K to 100K years ago [1,2] and some supporting a tight population bottleneck down to a few thousand individuals (*effective population size) at that point as well [2, 3].
In addition, the list of alarms has the general "list of stuff" UI on the iphone. Having a consistent UI for certain types of tasks (adding something to a list) aids in learning to use the UI; once someone learns to add an item to one list they'll transfer that knowledge to new situations (won't they?!). Though I'll admit that "Add" may be better than "+" in these cases (though you again run into localization issues, but since it's a general UI item Apple could probably handle that system wide).
It's a tricky trade off: UI concepts should be consistent and generalizable, but also as descriptive as possible. These are at odds at times.
9. Why did you decide to make the Android version free and is that going to change any time soon?
“Free is the way to go with Android. Nobody has been successful selling content on Android. We will offer a way to remove the ads by paying for the app, but we don’t expect that to be a huge revenue stream.”
Note: that article is something like two months old now, things may have changed since then for them.
Agreed. The biggest usability change for me so far, aside from the overgenerous whitespace, are the folded preview-comments. I noticed that Re: subjects are missing the original subject (probably a plus, since it's redundant information), and (Score: X) information seems to be missing from them unless they are top-level posts. That's a shame, since I routinely use that as a filter for whether a post is likely to be interesting enough to fold out and read.
While I agree that the result is almost certainly incorrect, IMO his arguments appear to be analytical enough to be considered a proof (perhaps a "sketch," and again, he's likely made a mistake). Even though the arguments are quite rambling, he insists on using too much of his own terminology, and he cites almost no previous literature (though he attempts to justify this near the end.)
Also, he does offer a runtime analysis: O(mn^4) in Section 6. As for the space/time tradeoff mentioned by others, I'm no complexity theorist but as I understand it no poly-time algorithm can use exponential space (how can you access exponential memory locations in polynomial time?)
Finally, the inclusion of real-machine performance isn't that surprising given that he has software: I myself have been asked for this by reviewers even after giving a big-O analysis of the algorithms (in cases where I had produced real software).
Personally, I prefer the instant feedback when scrolling. I generally feel like I'm moving a physical object when scrolling on an iOS device, whereas with Android, I feel like I'm using a gesture UI. There's nothing wrong with either--it's personal preference.
I've thought about this exactly, that apple does a great job of making their devices seem physical instead of virtual. (Please excuse the handwavy language.)
One can call it personal preference, but in reality I think a lot of people's preferences lean that way because it's ingrained in their heads. When they move an object with their finger, certain basal parts of the brain expect that object to react in a smooth, predictable manner, and this is decoupled from the expectation of the information they want to see scrawled on the object (the webpage). This was important even back in keyboard and mouse days (old linux window managers would give an option to just show an empty rectangle when you dragged a window so that the movement and placement was smooth), but now we're literally pushing things around like they're representations of the physical world.
And it's a hell of a lot nicer to use a device that agrees with built-in models of reality. I dunno if apple has patented that very idea or what, but it's all over iOS and conspicuously absent in Android. Fluid screen rotation, resistance to overscroll and springback, acceleration and deceleration of page flipping and scrolling... and simple framerate (though the Droid 2 is admittedly getting a lot closer in that last regard).
Very much agreed, though I'm not a simulationist per se anymore (these days I just do all kinds of crazy things to large data sets, and need to remember what crazy things I did, and when, and as you said, what the output means!)
Here's another tip, which I also thought about last time someone asked slashdot about scientific data organization: keep a wiki, and write down all the things you do (at least everything that isn't trivial to reproduce) there. Commands, paramaters used, input and output files created, etc. I organize chronologically. Having a digital "lab notebook" can be invaluable, and it makes the problem of organizing things much easier, since everything important is indexed in the wiki and can be looked up based on the timeframe of the project.
Since apparently they think $#*! is a word in the language, I guess there are an infinite number of alternatives: $, $%, $$%, $#%,... I just it just depends on how many of them would be offensive;-)
(Filter error: please use fewer 'junk' characters. I guess slashdot does find them offensive.)
Because results published in major journals typically have very obvious statistical errors [eye roll].
Sigh. I know it's fashionable to bash scientific results these days, but....
Applying the Fischer’s exact test to the respective
CAZyme counts gives a P-value of P = 1.68x10^-6, indicative of a
statistically significant association between populations and the occurrence of porphyranase sequences in gut microbiota.
Transfer of carbohydrate-active enzymes from marine bacteria to Japanese gut microbiota, Jan-Hendrik Hehemann, Gaelle Correc, Tristan Barbeyron, William Helbert, Mirjam Czjzek & Gurvan Michel, Nature Letters, April, 2010.
Yeah, my first thought was that perhaps he is trying to test hardware longevity independently of error correction techniques (e.g., how many times on average can a particular bit be flipped before it's reading starts to degrade?) Perhaps even across different manufacturers or drive densities. This seems like it could be useful information to a) people who buy tons of hard drives and want to fine tune their reliability, or b) some enterprising young individual doing a science project, or c) what do I care why he wants to know?
Seriously, why is everyone hating on him so much? Is this not slashdot?
I was going to say something about the relative rates of drift in mitochondrial dna, but I think you've hit it spot on. I take it there aren't any microsatellites in mtDNA?
Anyway, for those interested, here's an interesting paper regarding a relatively variant region in mitochondrial dna, but for butterflies, rather than humans. Notice that even in butterflies (which have a generation per year), there are some variants which are present over a vast portion of the range of the species---definitely not useful for identifying individuals.
I used to work master control for a small TV station. Almost all the adds (especially the network produced/national ads) came in with color bars and 1kz (I think?) tones before the ad, we were supposed to record the spot* to our local systems with the audio normalized such that the tone was at a particular level.
I'd be surprised if there isn't some standard for normalizing the audio of the content to the test tone, but many commercial producers (again, epecially the national ads) break from that and have louder content. I frequently renormalized these to regular levels, even though I don't think I was technically supposed to. Then again, alone in master control at a tiny station, "my buttons, my call."
*Our station was also fairly out of date. During my time there we had only started moving away from beta tapes toward computerized systems for the programs, although we did have an ancient computer for playing commercials. Overall though, as a C.S. person, I'm surprised at how untechnological the television industry is as a whole.
Although, I suspect that virus caused mutations are only a small portion: even random single nucleotide polymorphisms will tent to survive (if they're not deadly) in the geographic area they originated in, because people generally don't move far from where they grew up before reproducing and passing on those mutations (historically, anyway).
Is that really what they say on arrest? I had to read that 3 times to understand it and the implications, I'm sure I'd have to have it said to me at least five times out loud.
... The most immediate, local scale is the intracellular chemical environment of the genes; so, for example, there is recent work on heritable methylation patterns which block the expression of some genes.
Something similar happened to me in high school. Coming home after a party (WITH a d.d.) we got pulled over for "not dimming our lights" behind the cop in front of us (read: being a bunch of kids in a car at 11pm), and the three of us not driving got taken to the police station and our parents called.
The other two were over 17, so they went before a judge and got MIPs. I was only 16 at the time, so I had to visit the juvenile prosecutor guy. He let me off the hook, noting that it's not actually illegal to have your brights on behind another car, and the cop should not have pulled us over. (I bet the fact that we all only blew ~0.04 had something to do with it too.)
I'm not sure they implied it was anything other than a coincidence. About the most substantive thing said (which also mentions time as a factor) was:
"It's surprising that even though cost isn't an issue, the browsers that offer the best performance are also the least popular, at least for now."
As for myself, I'm really glad somebody tagged this article with "correlationisnotcausation." For a second there I heard this result and just assumed that either popularity caused browser slowness, or that people liked waiting./sarcasm
My first guess is that the activity is worse than anyone actually knows about.
"Ok, yes, we still have a database of every non-spam/non-forward email (without attachments) that passed through AT&T's servers between 2002 and 2007. As well as meta-information on every call made. No, we don't want to get rid of it." (Hmm, perhaps that would have been infeasible back in 2002...?)
I'm not sure what else 'the secrets' could be, without donning my extra thick tinfoil hat.
Eh, I have a few OCD tendencies. For instance, my productivity decreases as a function of my desk entropy;) Not so much because I can't find stuff, but because I'm easily distracted/annoyed at the incongruences of a messy desk (or non-matching widgets...)
Would it also be bad to suggest that they at least use the same gui toolkit?
I'm always trying to get my gtk apps to look like my qt apps (gtkqt etc.), and it never seems to work quite right. There are a lot of good gtk apps which don't have equivalent qt versions and vice-versa. It keeps my poor linux desktop looking sad and unprofessional:(
Speaking as someone who's played around with my friend's cheap guitars and toyed with the idea of getting one myself... what is it about a cheap guitar that makes it "difficult and painful" to learn on?
Personally, I haven't bought myself one because 1) I don't have a lot of time for yet another hobby, and 2) I do get frustrated just trying to do very basic stuff on guitar as it is.
Speaking of OSX Office.. After my in-law's office install choked this winter, I attempted to fix it by reinstalling Office. Turns out, you can't just "reinstall" office (or a recent update) on a mac. You have to fully remove the program and all its files, install fresh, and go through the 2 or 3 hour process of running AutoUpdate 15 times to get the point upgrades.
Sigh. I've installed neooffice alongside, and encouraged them to migrate to that when possible. So much for Microsoft's last decent product.
Slashdot Mirror
Well I certainly can't speak to the linguistics aspect, but I didn't recognize date and population size numbers to be totally made up; there is some research (peer reviewed at least -- this isn't my area) putting initial expansion ~65K to 100K years ago [1,2] and some supporting a tight population bottleneck down to a few thousand individuals (*effective population size) at that point as well [2, 3].
1. http://www.pnas.org/content/103/25/9381.full
2. http://www.sciencedirect.com/science/article/pii/S0047248498902196
3. http://www.nature.com/nature/journal/v475/n7357/full/nature10231.html?WT.ec_id=NATURE-20110728
In addition, the list of alarms has the general "list of stuff" UI on the iphone. Having a consistent UI for certain types of tasks (adding something to a list) aids in learning to use the UI; once someone learns to add an item to one list they'll transfer that knowledge to new situations (won't they?!). Though I'll admit that "Add" may be better than "+" in these cases (though you again run into localization issues, but since it's a general UI item Apple could probably handle that system wide).
It's a tricky trade off: UI concepts should be consistent and generalizable, but also as descriptive as possible. These are at odds at times.
Clearly angry birds is making money on both platforms.
I think this interview may be interesting, given the reference to Angry Birds in particular:
Peter Vesterbacka, Maker of Angry Birds Talks about the Birds, Apple, Android, Nokia, and Palm/HP
9. Why did you decide to make the Android version free and is that going to change any time soon?
“Free is the way to go with Android. Nobody has been successful selling content on Android. We will offer a way to remove the ads by paying for the app, but we don’t expect that to be a huge revenue stream.”
Note: that article is something like two months old now, things may have changed since then for them.
Agreed. The biggest usability change for me so far, aside from the overgenerous whitespace, are the folded preview-comments. I noticed that Re: subjects are missing the original subject (probably a plus, since it's redundant information), and (Score: X) information seems to be missing from them unless they are top-level posts. That's a shame, since I routinely use that as a filter for whether a post is likely to be interesting enough to fold out and read.
(Devil's advocate post, ymmv)
While I agree that the result is almost certainly incorrect, IMO his arguments appear to be analytical enough to be considered a proof (perhaps a "sketch," and again, he's likely made a mistake). Even though the arguments are quite rambling, he insists on using too much of his own terminology, and he cites almost no previous literature (though he attempts to justify this near the end.)
Also, he does offer a runtime analysis: O(mn^4) in Section 6. As for the space/time tradeoff mentioned by others, I'm no complexity theorist but as I understand it no poly-time algorithm can use exponential space (how can you access exponential memory locations in polynomial time?)
Finally, the inclusion of real-machine performance isn't that surprising given that he has software: I myself have been asked for this by reviewers even after giving a big-O analysis of the algorithms (in cases where I had produced real software).
This'll never work. How are people going to annoyingly talk on their phones while ordering?
Personally, I prefer the instant feedback when scrolling. I generally feel like I'm moving a physical object when scrolling on an iOS device, whereas with Android, I feel like I'm using a gesture UI. There's nothing wrong with either--it's personal preference.
I've thought about this exactly, that apple does a great job of making their devices seem physical instead of virtual. (Please excuse the handwavy language.)
One can call it personal preference, but in reality I think a lot of people's preferences lean that way because it's ingrained in their heads. When they move an object with their finger, certain basal parts of the brain expect that object to react in a smooth, predictable manner, and this is decoupled from the expectation of the information they want to see scrawled on the object (the webpage). This was important even back in keyboard and mouse days (old linux window managers would give an option to just show an empty rectangle when you dragged a window so that the movement and placement was smooth), but now we're literally pushing things around like they're representations of the physical world.
And it's a hell of a lot nicer to use a device that agrees with built-in models of reality. I dunno if apple has patented that very idea or what, but it's all over iOS and conspicuously absent in Android. Fluid screen rotation, resistance to overscroll and springback, acceleration and deceleration of page flipping and scrolling... and simple framerate (though the Droid 2 is admittedly getting a lot closer in that last regard).
How the hell am I supposed to wrap a fish in that?
I've found that my iPad makes an excellent sushi plate.
Very much agreed, though I'm not a simulationist per se anymore (these days I just do all kinds of crazy things to large data sets, and need to remember what crazy things I did, and when, and as you said, what the output means!)
Here's another tip, which I also thought about last time someone asked slashdot about scientific data organization: keep a wiki, and write down all the things you do (at least everything that isn't trivial to reproduce) there. Commands, paramaters used, input and output files created, etc. I organize chronologically. Having a digital "lab notebook" can be invaluable, and it makes the problem of organizing things much easier, since everything important is indexed in the wiki and can be looked up based on the timeframe of the project.
Since apparently they think $#*! is a word in the language, I guess there are an infinite number of alternatives: $, $%, $$%, $#%, ... I just it just depends on how many of them would be offensive ;-)
(Filter error: please use fewer 'junk' characters. I guess slashdot does find them offensive.)
Sigh. I know it's fashionable to bash scientific results these days, but....
Applying the Fischer’s exact test to the respective CAZyme counts gives a P-value of P = 1.68x10^-6, indicative of a statistically significant association between populations and the occurrence of porphyranase sequences in gut microbiota.
Transfer of carbohydrate-active enzymes from marine bacteria to Japanese gut microbiota, Jan-Hendrik Hehemann, Gaelle Correc, Tristan Barbeyron, William Helbert, Mirjam Czjzek & Gurvan Michel, Nature Letters, April, 2010.
Yeah, my first thought was that perhaps he is trying to test hardware longevity independently of error correction techniques (e.g., how many times on average can a particular bit be flipped before it's reading starts to degrade?) Perhaps even across different manufacturers or drive densities. This seems like it could be useful information to a) people who buy tons of hard drives and want to fine tune their reliability, or b) some enterprising young individual doing a science project, or c) what do I care why he wants to know?
Seriously, why is everyone hating on him so much? Is this not slashdot?
I was going to say something about the relative rates of drift in mitochondrial dna, but I think you've hit it spot on. I take it there aren't any microsatellites in mtDNA?
Anyway, for those interested, here's an interesting paper regarding a relatively variant region in mitochondrial dna, but for butterflies, rather than humans. Notice that even in butterflies (which have a generation per year), there are some variants which are present over a vast portion of the range of the species---definitely not useful for identifying individuals.
Somebody mod this guy up if only for his nickname.
I used to work master control for a small TV station. Almost all the adds (especially the network produced/national ads) came in with color bars and 1kz (I think?) tones before the ad, we were supposed to record the spot* to our local systems with the audio normalized such that the tone was at a particular level.
I'd be surprised if there isn't some standard for normalizing the audio of the content to the test tone, but many commercial producers (again, epecially the national ads) break from that and have louder content. I frequently renormalized these to regular levels, even though I don't think I was technically supposed to. Then again, alone in master control at a tiny station, "my buttons, my call."
*Our station was also fairly out of date. During my time there we had only started moving away from beta tapes toward computerized systems for the programs, although we did have an ancient computer for playing commercials. Overall though, as a C.S. person, I'm surprised at how untechnological the television industry is as a whole.
Behold the awesome pictures: Genes mirror geography within Europe
Although, I suspect that virus caused mutations are only a small portion: even random single nucleotide polymorphisms will tent to survive (if they're not deadly) in the geographic area they originated in, because people generally don't move far from where they grew up before reproducing and passing on those mutations (historically, anyway).
Is that really what they say on arrest? I had to read that 3 times to understand it and the implications, I'm sure I'd have to have it said to me at least five times out loud.
... The most immediate, local scale is the intracellular chemical environment of the genes; so, for example, there is recent work on heritable methylation patterns which block the expression of some genes.
Kim Sterelny, Rethinking Inheritance, Journal of Biological Theory, 2007.
Something similar happened to me in high school. Coming home after a party (WITH a d.d.) we got pulled over for "not dimming our lights" behind the cop in front of us (read: being a bunch of kids in a car at 11pm), and the three of us not driving got taken to the police station and our parents called.
The other two were over 17, so they went before a judge and got MIPs. I was only 16 at the time, so I had to visit the juvenile prosecutor guy. He let me off the hook, noting that it's not actually illegal to have your brights on behind another car, and the cop should not have pulled us over. (I bet the fact that we all only blew ~0.04 had something to do with it too.)
I'm not sure they implied it was anything other than a coincidence. About the most substantive thing said (which also mentions time as a factor) was:
/sarcasm
"It's surprising that even though cost isn't an issue, the browsers that offer the best performance are also the least popular, at least for now."
As for myself, I'm really glad somebody tagged this article with "correlationisnotcausation." For a second there I heard this result and just assumed that either popularity caused browser slowness, or that people liked waiting.
My first guess is that the activity is worse than anyone actually knows about.
"Ok, yes, we still have a database of every non-spam/non-forward email (without attachments) that passed through AT&T's servers between 2002 and 2007. As well as meta-information on every call made. No, we don't want to get rid of it." (Hmm, perhaps that would have been infeasible back in 2002...?)
I'm not sure what else 'the secrets' could be, without donning my extra thick tinfoil hat.
Eh, I have a few OCD tendencies. For instance, my productivity decreases as a function of my desk entropy ;) Not so much because I can't find stuff, but because I'm easily distracted/annoyed at the incongruences of a messy desk (or non-matching widgets...)
Would it also be bad to suggest that they at least use the same gui toolkit?
:(
I'm always trying to get my gtk apps to look like my qt apps (gtkqt etc.), and it never seems to work quite right. There are a lot of good gtk apps which don't have equivalent qt versions and vice-versa. It keeps my poor linux desktop looking sad and unprofessional
Speaking as someone who's played around with my friend's cheap guitars and toyed with the idea of getting one myself... what is it about a cheap guitar that makes it "difficult and painful" to learn on?
Personally, I haven't bought myself one because 1) I don't have a lot of time for yet another hobby, and 2) I do get frustrated just trying to do very basic stuff on guitar as it is.
Finally, whatever happened to these guys? http://www.guitarrising.com/
Speaking of OSX Office.. After my in-law's office install choked this winter, I attempted to fix it by reinstalling Office. Turns out, you can't just "reinstall" office (or a recent update) on a mac. You have to fully remove the program and all its files, install fresh, and go through the 2 or 3 hour process of running AutoUpdate 15 times to get the point upgrades.
Sigh. I've installed neooffice alongside, and encouraged them to migrate to that when possible. So much for Microsoft's last decent product.