Huh. I wonder what else could have happened that week which could have also been the primary cause, if not the outage. (I no longer even bother to argue correlation != causation, because only idiots and trolls feel empowered by reciting that Stats101 cliche.)
Ok, how about we get rid of the factorial, because that just a linear of a linear, and anyone who knows how to program would never do that. Right?... RIGHT?
I'm sure you know what a hash is, if you snarked me about factorial search. If you don't read up first.
So, you scan the edge of a piece of paper, say.1mm per pixel, a 10cm strip will have 100 bits. I'm assuming courier typed fonts.
Then you hash the 100 bit digits (barely two IEEE double precision values) that point to the strip ID.
Trivial case: if there is a hash collision, the edges match and the strips were adjacent at one point. Hash functions are O lookup speed, so we simply break up strips until the # of bits to represent and edge creates a reasonable table size (based on memory).
Increasing the resolution increases the hash key and makes erroneous collisions less common.
If you end up with a bunch of strips (or strip fragments) that didn't collide, THEN you have to do things the hard way. I haven't figured out what to do here, but I think I'm clearly an absolute fucking genius so far. Bookmark this conversation, because I bet it's what the winners do. (/endsarcasm)
I think I'll go make a vodka milkshake now and surf 4chan/b/. (/truestatement)
Scan all the shreds. Encode them based on the first 0.1 mm of ink on the long edge Compare all similar edge strings Recombine the ID of matching strings Done
Error correction: If multiple edge strings match, do an OCR to see which solution fits best
I moved my plan from streaming+1DVD to streaming+3DVDs.
Long live netflix!
Too bad the studios are gonna squeeze them dry and then turn around and provide a streaming/dvd service for each studio. I figure 15 major studios under 5 corporate entities equals 5 to 15 new subscriptions at $9.99 a month each. Looks like DVD brick-and-mortar might make a comeback! Either that or I'll start reading again.
"A vague general sketch of the user interface should be worked out before any code starts getting written,"
Agreed, especially if the intent is to PROVIDE a UI (as opposed to command-line apps that become UIs). The best people to spec that are the users and the teachers and the professionals use who use the applications. Grandma can provide as much useful input to a UI as Nerdy McNerdstein over in the compiler group.
"So asking just some users does no good, research would imply involvement of a large number of users, which is hard."
Again, I completely disagree. Asking "some" users (whatever that means, let's assume, oh, 20) will greatly help. Asking many (oh, >1000) will also help. Any kind of feedback is useful. Simply dismissing feedback means you aren't listening to your customers at all. I don't think that is a successful business plan.
"A few users' "preference" will be suboptimal for large numbers of other users with slight variations on usage."
Which is exactly why programmers shouldn't design UIs, in general. Thanks for making my point. I think you are trying to say here is that democratization of features makes the whole product worse. That sounds like Apple. Looking at open source GUIs by comparison, I believe Apple does a better job. But that would seemingly run counter to my initial argument, if it weren't for the fact that Apple does respond to overwhelming user feedback (iOS5 notifications).
"General purpose applications such as web browsers or word processors are a different animal, totally different from CAD software"
I don't see how they are totally different. Word Processors and Web Browsers are quite complex, and can have as many tool bars as CAD software (I'm looking at you, MS ribbon). I think browsers should be simpler, but that is my opinion.
"A programmer should be intimately familiar with the generic requirements to have a web browser, having used other products that already exist."
Programmers are a very, very small %age of users of web browsers and word processors. I don't see how limiting feedback on UI design to just programmers will make a better product. This is a good argument that more people know how to use a word processor or browser than CAD software, and I could devil's advocate myself and say that I don't think that automatically qualifies them to make good judgement calls on UI design.
I think I get your drift: I understand that "focus group" attempts to wag the dog often fail horribly, look at Microsoft. But you can't blow off your user base and assume the ivory tower programmers know what is best. Some users will need to be told how to use something, but a good UI design teaches inherently, and that is why there is value in UI design as a discipline orthogonal to programming.
This issue is a perfect example of the gap between Apple and Droid. You people are flaming each other about a fairly small usability feature. This is right up there with complaining about not being able to change icons. This is why Droid is a mess right now (from bugs, to security, to low customer satisfaction) and iOS is dominating. I know giving up software flexibility is the worst sin ever, but sometimes you need to just except small things you can't change (like the size of your caps lock key or the color of your keyboard cord) and focus on the things that matter, like getting work done. Speaking of which, why am I procrastinating right now? Oops!
I 100% disagree with you. Emphatically. Having been in the CAD development world for 20+ years, programmers are THE LAST PEOPLE who should be designing user interfaces. The vast majority of programmers have no idea what usability means to a general audience, and even worse sense of aesthetic. The worst offenders are programmers who think they know better without ever having met a customer.
Now is an artless programmer better than a bad UI designer? That is debatable. But in my experience, the people who should develop the UI are the users and the trainers, together, and then provide a spec to the development team. With that feedback, even a mediocre programmer can make life a lot easier on the users.
I know this sounds counter-intuitive, but the programmers rarely have any idea how to actually use the software. Especially when it is a large modular project, and each programmer may only have a slight idea what the entire application actually does. Sure the lead integrator has a clue, but they are usually way too busy to put any thought into a UI design, let alone collect feedback from the people who use it; they often delegate to another tertiary programmer (intern, co-op) who knows even less.
I've seen this in 3D animation, CAD/CAM, medical software, automotive UI, factory and assembly line flow control, local government utilities control systems, etc.
Whenever I read an article critical of how a corporation chooses to manage a market, I try to pay attention to where they glean their insight. As with this article, many use a few anecdotal interviews. Had the author backed up the assertion with an example of, say, a company with over 50 content creators switching to PC, then perhaps their claims would have merit. The very rare exception is a scientific survey (non-web-based). So without any kind of solid evidence, this is pure blog filler and naval gazing. Fun to talk about, but ultimately meaningless, which is fine. I would just rather read articles about market trends annotated with sources or reasonable proof.
...it is awkward to pronounce. The two phonemes require an aspired stop, and it doesn't roll off the tongue nicely, unless you say eeeeenk (like Ren Hoek) instead of the stuttering Eee-Eeenk.
Slashdot Mirror
FTA:
”Android has been like the Wild West,” says Miller. “And this bug basically reduces the security of iOS to that of Android.”
Lolz.
SparkFun doesn't carry universal programmers, just ones specific to the particular ucontroller.
MicroController Pros are where the good stuff is at:
http://microcontrollershop.com/default.php?cPath=92&osCsid=c0eab1d9c7b6bbc4a4ec8f048d529bc2
Indian engineers have been reported camping known spawn points in Ungoro. News at 11.
Ahhhh crap. Looks like there are enough studies on both sides to make this another -ism war.
Turning my brain off now...
How the heck did they implement a grammar parser and lexer in 40k, let alone a compiler?!?!?
I suck as a programmer.
Huh. I wonder what else could have happened that week which could have also been the primary cause, if not the outage. (I no longer even bother to argue correlation != causation, because only idiots and trolls feel empowered by reciting that Stats101 cliche.)
wow, those are interesting claims. data references please?
because my sources say speed cameras DO work, and on TWO DIFFERENT continents.
http://www.physorg.com/news140443278.html
http://alttransport.com/2010/10/7966/
nice try, Speed Trollster.
Ok, how about we get rid of the factorial, because that just a linear of a linear, and anyone who knows how to program would never do that. Right? ... RIGHT?
I'm sure you know what a hash is, if you snarked me about factorial search. If you don't read up first.
So, you scan the edge of a piece of paper, say .1mm per pixel, a 10cm strip will have 100 bits. I'm assuming courier typed fonts.
Then you hash the 100 bit digits (barely two IEEE double precision values) that point to the strip ID.
Trivial case: if there is a hash collision, the edges match and the strips were adjacent at one point. Hash functions are O lookup speed, so we simply break up strips until the # of bits to represent and edge creates a reasonable table size (based on memory).
Increasing the resolution increases the hash key and makes erroneous collisions less common.
If you end up with a bunch of strips (or strip fragments) that didn't collide, THEN you have to do things the hard way. I haven't figured out what to do here, but I think I'm clearly an absolute fucking genius so far. Bookmark this conversation, because I bet it's what the winners do. (/endsarcasm)
I think I'll go make a vodka milkshake now and surf 4chan /b/. (/truestatement)
Aaaaaand... your point?
My idea sucks? What's yours? Don't have one? I see.
Please refer the first sentence of my reply.
pfff. puh-leez. don't bother me with your trivial details.
just use quantum computers.
and lasers.
i'll be sipping a mai tai by the pool, please deliver my $50k in bicentennial quarters.
Scan all the shreds.
Encode them based on the first 0.1 mm of ink on the long edge
Compare all similar edge strings
Recombine the ID of matching strings
Done
Error correction:
If multiple edge strings match, do an OCR to see which solution fits best
I moved my plan from streaming+1DVD to streaming+3DVDs.
Long live netflix!
Too bad the studios are gonna squeeze them dry and then turn around and provide a streaming/dvd service for each studio. I figure 15 major studios under 5 corporate entities equals 5 to 15 new subscriptions at $9.99 a month each. Looks like DVD brick-and-mortar might make a comeback! Either that or I'll start reading again.
"Yes, mobile apps that are crowd sourced should be built on open standards to achieve maximum disruptionability."
This made me LOL.
I think you just coined a new word.
Oh, crap, I'm not pre-adolescent anymore. Hmpf.
There's still no test, but I'll definitely fight my way to the front lines for this vaccination.
bah, another C programmer attempts lisp. bah! /kidding :-)
"A vague general sketch of the user interface should be worked out before any code starts getting written,"
Agreed, especially if the intent is to PROVIDE a UI (as opposed to command-line apps that become UIs). The best people to spec that are the users and the teachers and the professionals use who use the applications. Grandma can provide as much useful input to a UI as Nerdy McNerdstein over in the compiler group.
"So asking just some users does no good, research would imply involvement of a large number of users, which is hard."
Again, I completely disagree. Asking "some" users (whatever that means, let's assume, oh, 20) will greatly help. Asking many (oh, >1000) will also help. Any kind of feedback is useful. Simply dismissing feedback means you aren't listening to your customers at all. I don't think that is a successful business plan.
"A few users' "preference" will be suboptimal for large numbers of other users with slight variations on usage."
Which is exactly why programmers shouldn't design UIs, in general. Thanks for making my point. I think you are trying to say here is that democratization of features makes the whole product worse. That sounds like Apple. Looking at open source GUIs by comparison, I believe Apple does a better job. But that would seemingly run counter to my initial argument, if it weren't for the fact that Apple does respond to overwhelming user feedback (iOS5 notifications).
"General purpose applications such as web browsers or word processors are a different animal, totally different from CAD software"
I don't see how they are totally different. Word Processors and Web Browsers are quite complex, and can have as many tool bars as CAD software (I'm looking at you, MS ribbon). I think browsers should be simpler, but that is my opinion.
"A programmer should be intimately familiar with the generic requirements to have a web browser, having used other products that already exist."
Programmers are a very, very small %age of users of web browsers and word processors. I don't see how limiting feedback on UI design to just programmers will make a better product. This is a good argument that more people know how to use a word processor or browser than CAD software, and I could devil's advocate myself and say that I don't think that automatically qualifies them to make good judgement calls on UI design.
I think I get your drift: I understand that "focus group" attempts to wag the dog often fail horribly, look at Microsoft. But you can't blow off your user base and assume the ivory tower programmers know what is best. Some users will need to be told how to use something, but a good UI design teaches inherently, and that is why there is value in UI design as a discipline orthogonal to programming.
Oops, meant "accept", not "except". Totally changes the meaning. Ha.
This issue is a perfect example of the gap between Apple and Droid. You people are flaming each other about a fairly small usability feature. This is right up there with complaining about not being able to change icons. This is why Droid is a mess right now (from bugs, to security, to low customer satisfaction) and iOS is dominating. I know giving up software flexibility is the worst sin ever, but sometimes you need to just except small things you can't change (like the size of your caps lock key or the color of your keyboard cord) and focus on the things that matter, like getting work done. Speaking of which, why am I procrastinating right now? Oops!
I 100% disagree with you. Emphatically. Having been in the CAD development world for 20+ years, programmers are THE LAST PEOPLE who should be designing user interfaces. The vast majority of programmers have no idea what usability means to a general audience, and even worse sense of aesthetic. The worst offenders are programmers who think they know better without ever having met a customer.
Now is an artless programmer better than a bad UI designer? That is debatable. But in my experience, the people who should develop the UI are the users and the trainers, together, and then provide a spec to the development team. With that feedback, even a mediocre programmer can make life a lot easier on the users.
I know this sounds counter-intuitive, but the programmers rarely have any idea how to actually use the software. Especially when it is a large modular project, and each programmer may only have a slight idea what the entire application actually does. Sure the lead integrator has a clue, but they are usually way too busy to put any thought into a UI design, let alone collect feedback from the people who use it; they often delegate to another tertiary programmer (intern, co-op) who knows even less.
I've seen this in 3D animation, CAD/CAM, medical software, automotive UI, factory and assembly line flow control, local government utilities control systems, etc.
Can someone explain the pros and cons of this? Seems like troll food.
A mind is a terrible thing to not use.
Educate yourself, son:
http://en.wikipedia.org/wiki/Retreat_of_glaciers_since_1850
Whenever I read an article critical of how a corporation chooses to manage a market, I try to pay attention to where they glean their insight. As with this article, many use a few anecdotal interviews. Had the author backed up the assertion with an example of, say, a company with over 50 content creators switching to PC, then perhaps their claims would have merit. The very rare exception is a scientific survey (non-web-based). So without any kind of solid evidence, this is pure blog filler and naval gazing. Fun to talk about, but ultimately meaningless, which is fine. I would just rather read articles about market trends annotated with sources or reasonable proof.
what IPA phoneme is "yn"?
...it is awkward to pronounce. The two phonemes require an aspired stop, and it doesn't roll off the tongue nicely, unless you say eeeeenk (like Ren Hoek) instead of the stuttering Eee-Eeenk.