How the hell is it costing $100,000 per course? Most coursera courses are just some guy talking to a webcam or a fixed camera. Most of the stanford courses(many very good) are just a guy working a camera from the back of the class with the instructor miked up. So unless the check boxes on the quizzes cost $200 each something is fishy with this number.
As per the article, "This is probably a somewhat overindulgent price for appearance, rather than substance." IE. the author believes her sources have rounded up for shock value.
Of course, this is VERY HARD WORK. Most people are lazy which is why we get the crap we do. It takes real work to put together something appropriate and it cannot be slap-dashed together in a few hours.
That's not it at all -- most people just don't know how to structure videos, and right now the orthodox line is just "stand in front of a camera and show a couple of slides, film it, and stick it on the net". Nobody's training teachers to do this stuff right -- they're just expected to do it.
The thing that made Ng's machine learning course really good was his focus on "intuitions" -- a common sense description of how the task works and why -- before getting down to the fine detail. Collins's NLP course presented lots of technical information without adequately demonstrating how, when or why it was applicable. Jurafsky's NLP course dealt, like Ng, with the whys before the hows. Jurafsky's videos were better than his book, actually, as he slipped into the same trap of too-much-technical-information in writing the book. (Personally, I disagree about Odersky's Scala course, though -- I thought it was far too dense for me, even despite already being familiar with FP -- Standard ML at uni + Coursera refresher with Dan Grossman)
TL,DR: I agree: it's not the medium, it's the execution. (So stick with the cheaper/quicker/easier/more convenient medium.)
For me, the lecture uses a more approachable language than the more formal format of the book (good for further exploration and lookups).
Yup, and the reason that the lecture survives to this day is because people speak more naturally and use a lot more supportive redundancy. Natural redundancy tends to be edited out of books on the erroneous grounds of being needless repetition. Natural pacing is lost too, and the text becomes too dense.
A video demonstration is just more compact and more effective because it is multi-modal, than the full description in text.
> because the three forms crowd each other out.
A video has the potential to be more effective because it is multimodal, but it also has the potential to do more harm than good, which brings us back to the thrust of the article -- doing video properly takes more time than it's worth
In a well-done presentation
Well done presentations are the exception -- don't try to build a rule on them.
I applaud Ms. Lorena's determination to keep the cost of her video to the absolute minimum
[...]
It is thus advisable for one to adopt the moderate approach --- and try to avoid the two extremes
Read again (my emphasis):
The short answer is budget and time: making good-quality videos is expensive & making simple yet effective educational videos is time consuming, if not necessarily costly.
Making videos takes a lot of time and/or money -- often both. Even if you keep monetary costs to a minimum, the time expense is still huge. And what do you have at the end of it? Video is a write-once, read-many medium and it's very tricky to go back and revise any errors -- look at some of the heavily repeated courses on Coursera (for example Dan Grossman's University of Washington Programming Languages course) and you'll see corrections patched onto previously-live slidecasts, annotations overlaid onto live code windows and on-screen comments about things the presenter said wrong. Video you either get right first time, or its wrong for all time.
Furthermore, the big elephant in the room is how MOOCs claim that the lecture is ineffective and dead... and then simply put the lecture in a different form. It is very difficult to embed interactive learning into a passive medium like video, and you get many courses that tell you to stop the video and go and do something else... not much good when you've downloaded the video for offline viewing in the provider's own app, and you don't actually have access to that something else.
From TFA:
Despite their popularity in MOOCs and flipped classrooms, “lecture videos” have the same pitfalls as regular lectures: they provide a false sense of clarity and are utterly forgettable.
Exactly -- there are much better things we could be doing with our time.
Barba makes a heck of a lot of important points, and I think anyone writing an online course should read the article.
I'd like to add one observation, though. She doesn't talk about personalised learning. When I go into a MOOC and it has lots of information that I already know, it's either integrated into the videos so that I have no choice but to watch, or its in an "optional" video that (despite being optional) sits in my to-do list either until I load and cancel it, or the course website closes. The path doesn't adapt to my knowledge. And with video, it couldn't even if it wanted to -- dynamically editing and mixing custom videos with alternative explanations etc would be a huuuuuge task.
Dynamic text, on the other hand, is mostly a solved problem, so adaptive courses are reasonably easy to make. Yet video distracts us from the real goal.
Or just have a senior member of cabin crew step onto the flight deck whenever one of the flight crew steps out so that there's always two people in the cockpit at any given time, and one can open the door if the other is incapacitated.
So to crash the plane, you only need to get the lead flight attendant under your control. They then kill the cp and your good to go. Ground based unlock? Just get leverage on the IT guy to spoof the signal. Video just gives them a chance to do q selfie diatribe. Life is dangerous. accept it.
Are you thick? The flight attendant wouldn't have a key -- it would be a matter of the flight crew having to physically allow him/her in before one of them leaves the flight deck. The only way this would be any less secure than the current arrangements is if the terrorist attack was an inside job, and cabin crew are no more (or less) corruptible than flight crew, as a general rule.
Yes, the US court system in particular is a bugger for everyone. But the principle of not needing to file trademarks is sound, even if the execution is flawed.
Well I learned something today. I guess I should have assumed that lawyers would make the problem as difficult to resolve as possible.
Sorry, but this one is good for the little guy. It means that granny can make Mrs Mullins' Marmalade without needing to file or register anything, and The Big Conserve Co can't just walk up and register, steal her reputation and force her to stop trading when she's just starting to make a name for herself. This is protection for the little guy.
The people who "generate the most wealth" are often responsible for massive amounts of unemployment. If we look at the Walton family empire (Walmart Stores Inc), we see a string of relatively labour-light retail outlets that push smaller companies out of business, resulting in a net loss of jobs. Amazon has taken this trend to ridiculous levels, exports profits to tax havens whenever and wherever possible, and then whenever criticised for sharp practices stands up and demands that we "celebrate" the billions in wealth and the thousands of jobs they "create".
As for "rewarding people who don't work and contribute nothing to society", you have clearly never had to live on benefits. They hardly constitute a "reward" for anything.
I think that was a joke, because any device with a remote would have a standby mode, and would therefore be soft-power-down rather than hard-power-down.
It turns out that Smart Projects had trademarked the Arduino brand in Italy in December 2008, before Arduino LLC got around to filing in April 2009 in the USA.
So... what's to discuss. I don't think there's a law against being a complete asshole, so smart projects wins.
You don't "trademark" things -- you register a trademark. Registration is not strictly necessary in most jurisdictions -- as long as you are actively using the branding, it is automatically considered your trademark. Order of registration only matters if you're talking about two commercial entities independently coming up with the same brand. Here we have two entities with an existing contractual relationship. Smart Projects therefore was not ignorant of the informal body that later became Arduino LLC. Who initiated the contract and who came up with the name? What did the original contract say about the name? Clearly the initial contracts were poorly drafted, or there would be a clear answer, and we wouldn't be having this conversation!!
Arduino turned 11 yesterday, and like many children of that age, the celebrations were kind of interrupted by its divorced parents' continuing battle for custody....
Or just have a senior member of cabin crew step onto the flight deck whenever one of the flight crew steps out so that there's always two people in the cockpit at any given time, and one can open the door if the other is incapacitated.
I think the prudent response to this incident would be to mandate that two members of crew should be on the flight deck at all times, so that there is always someone available to open the door. In practice, that would mean that the senior cabin crew member would have to step in whenever the pilot or co-pilot needs to go for a pee.
I think you're confusing "fuck" with "Frigg". "Frigg" was a Norse god, and thus gets a capital. "Fuck" was a norse profanity, so doesn't. Except in sentences like that one, where it occurs at the beginning of the sentence.
Well, that's basically what flying is. If you're dumb enough to do that, you deserve to crash in a pile of smouldering flames.
That's some major league victim blaming there. Let's see if we can blame the victims of the Titanic for "letting someone stick them in a big tub of riveted metal and setting them afloat on the world's second largest ocean". Or maybe we can blame the dinosaurs for being stupid enough to live on a big rock with only one inadeuately small moon to help protect it from once-in-a-gazillion-years meteor strikes. Stupid dinosaurs -- they really had it coming.
I notice that the article does not give any actual specifications on how dark the test night was. It sounds like this gives someone a slight edge (100% vs 33% for the control group), so it won't replace technological night vision for a while at least.
Or possibly even pirate night-vision, for that matter.
Why did pirates wear eye-patches? Because an eye that receives no light adapts to the dark. After nightfall, they would take the patch off, and the patched eye would perform better than the eye that had been exposed to the glare of the sea all day. The "test subject" in this case wore darkened contact lenses and then stuck on a pair of sunglasses which he wore for a couple of hours before the test. By the time they went out to carry out the distance vision experiment, his eyes would have had longer to adjust to the dark than his companions, so his night vision would have been better than theirs anyway...
Not a patch, but having one eye as "control" wouldn't be a bad idea. If the subject had one "normal" eye and one "doped" one, you could make some kind of meaningful-ish comparison. The eye-patch would be a bad idea as the patched eye would be constantly adjusting to the darkness and would end up being "super-powered" itself by the time of the test.
Actually, if the buy was wearing darkened contact lenses and sunglasses after treatment and before the test, and the "control" subjects weren't, the whole thing's basically just a waste of time, as we don't know how much of his improved accuracy over the others was down to the increased light-sensitivity as a result of blocking out light for several hours.
Soulskill decides to use semi-trendy jargon in a headline, and you think it's some kind of admission from the establishment that they control the horizontal and the vertical? There's a hole in your hat. Take some of this -- it's Bacofoil. Quality stuff.
Slashdot Mirror
How the hell is it costing $100,000 per course? Most coursera courses are just some guy talking to a webcam or a fixed camera. Most of the stanford courses(many very good) are just a guy working a camera from the back of the class with the instructor miked up. So unless the check boxes on the quizzes cost $200 each something is fishy with this number.
As per the article, "This is probably a somewhat overindulgent price for appearance, rather than substance." IE. the author believes her sources have rounded up for shock value.
Yeah... bout as bloody ignorant as them bloody Romans who invented the bloody thing thousands of years ago.
(You do realise that & is an older letter than K and Z, right...?)
Of course, this is VERY HARD WORK. Most people are lazy which is why we get the crap we do. It takes real work to put together something appropriate and it cannot be slap-dashed together in a few hours.
That's not it at all -- most people just don't know how to structure videos, and right now the orthodox line is just "stand in front of a camera and show a couple of slides, film it, and stick it on the net". Nobody's training teachers to do this stuff right -- they're just expected to do it.
The thing that made Ng's machine learning course really good was his focus on "intuitions" -- a common sense description of how the task works and why -- before getting down to the fine detail. Collins's NLP course presented lots of technical information without adequately demonstrating how, when or why it was applicable. Jurafsky's NLP course dealt, like Ng, with the whys before the hows. Jurafsky's videos were better than his book, actually, as he slipped into the same trap of too-much-technical-information in writing the book. (Personally, I disagree about Odersky's Scala course, though -- I thought it was far too dense for me, even despite already being familiar with FP -- Standard ML at uni + Coursera refresher with Dan Grossman)
TL,DR: I agree: it's not the medium, it's the execution. (So stick with the cheaper/quicker/easier/more convenient medium.)
For me, the lecture uses a more approachable language than the more formal format of the book (good for further exploration and lookups).
Yup, and the reason that the lecture survives to this day is because people speak more naturally and use a lot more supportive redundancy. Natural redundancy tends to be edited out of books on the erroneous grounds of being needless repetition. Natural pacing is lost too, and the text becomes too dense.
A video demonstration is just more compact and more effective because it is multi-modal, than the full description in text.
> because the three forms crowd each other out.
A video has the potential to be more effective because it is multimodal, but it also has the potential to do more harm than good, which brings us back to the thrust of the article -- doing video properly takes more time than it's worth
In a well-done presentation
Well done presentations are the exception -- don't try to build a rule on them.
I applaud Ms. Lorena's determination to keep the cost of her video to the absolute minimum
[...]
It is thus advisable for one to adopt the moderate approach --- and try to avoid the two extremes
Read again (my emphasis):
The short answer is budget and time: making good-quality videos is expensive & making simple yet effective educational videos is time consuming, if not necessarily costly.
Making videos takes a lot of time and/or money -- often both. Even if you keep monetary costs to a minimum, the time expense is still huge. And what do you have at the end of it? Video is a write-once, read-many medium and it's very tricky to go back and revise any errors -- look at some of the heavily repeated courses on Coursera (for example Dan Grossman's University of Washington Programming Languages course) and you'll see corrections patched onto previously-live slidecasts, annotations overlaid onto live code windows and on-screen comments about things the presenter said wrong. Video you either get right first time, or its wrong for all time.
Furthermore, the big elephant in the room is how MOOCs claim that the lecture is ineffective and dead... and then simply put the lecture in a different form. It is very difficult to embed interactive learning into a passive medium like video, and you get many courses that tell you to stop the video and go and do something else... not much good when you've downloaded the video for offline viewing in the provider's own app, and you don't actually have access to that something else.
From TFA:
Despite their popularity in MOOCs and flipped classrooms, “lecture videos” have the same pitfalls as regular lectures: they provide a false sense of clarity and are utterly forgettable.
Exactly -- there are much better things we could be doing with our time.
Barba makes a heck of a lot of important points, and I think anyone writing an online course should read the article.
I'd like to add one observation, though. She doesn't talk about personalised learning. When I go into a MOOC and it has lots of information that I already know, it's either integrated into the videos so that I have no choice but to watch, or its in an "optional" video that (despite being optional) sits in my to-do list either until I load and cancel it, or the course website closes. The path doesn't adapt to my knowledge. And with video, it couldn't even if it wanted to -- dynamically editing and mixing custom videos with alternative explanations etc would be a huuuuuge task.
Dynamic text, on the other hand, is mostly a solved problem, so adaptive courses are reasonably easy to make. Yet video distracts us from the real goal.
Or just have a senior member of cabin crew step onto the flight deck whenever one of the flight crew steps out so that there's always two people in the cockpit at any given time, and one can open the door if the other is incapacitated.
So to crash the plane, you only need to get the lead flight attendant under your control. They then kill the cp and your good to go. Ground based unlock? Just get leverage on the IT guy to spoof the signal. Video just gives them a chance to do q selfie diatribe. Life is dangerous. accept it.
Are you thick? The flight attendant wouldn't have a key -- it would be a matter of the flight crew having to physically allow him/her in before one of them leaves the flight deck. The only way this would be any less secure than the current arrangements is if the terrorist attack was an inside job, and cabin crew are no more (or less) corruptible than flight crew, as a general rule.
Yes, the US court system in particular is a bugger for everyone. But the principle of not needing to file trademarks is sound, even if the execution is flawed.
Well I learned something today. I guess I should have assumed that lawyers would make the problem as difficult to resolve as possible.
Sorry, but this one is good for the little guy. It means that granny can make Mrs Mullins' Marmalade without needing to file or register anything, and The Big Conserve Co can't just walk up and register, steal her reputation and force her to stop trading when she's just starting to make a name for herself. This is protection for the little guy.
Who would leave their console on and just turn the TV off?
Quite a lot of people. And they even think they're switching off.
The people who "generate the most wealth" are often responsible for massive amounts of unemployment. If we look at the Walton family empire (Walmart Stores Inc), we see a string of relatively labour-light retail outlets that push smaller companies out of business, resulting in a net loss of jobs. Amazon has taken this trend to ridiculous levels, exports profits to tax havens whenever and wherever possible, and then whenever criticised for sharp practices stands up and demands that we "celebrate" the billions in wealth and the thousands of jobs they "create".
As for "rewarding people who don't work and contribute nothing to society", you have clearly never had to live on benefits. They hardly constitute a "reward" for anything.
But the continual heating of the transformer coil also shortens the lifespan of the device.
I think that was a joke, because any device with a remote would have a standby mode, and would therefore be soft-power-down rather than hard-power-down.
From TFA:
It turns out that Smart Projects had trademarked the Arduino brand in Italy in December 2008, before Arduino LLC got around to filing in April 2009 in the USA.
So... what's to discuss. I don't think there's a law against being a complete asshole, so smart projects wins.
You don't "trademark" things -- you register a trademark. Registration is not strictly necessary in most jurisdictions -- as long as you are actively using the branding, it is automatically considered your trademark. Order of registration only matters if you're talking about two commercial entities independently coming up with the same brand. Here we have two entities with an existing contractual relationship. Smart Projects therefore was not ignorant of the informal body that later became Arduino LLC. Who initiated the contract and who came up with the name? What did the original contract say about the name? Clearly the initial contracts were poorly drafted, or there would be a clear answer, and we wouldn't be having this conversation!!
Arduino turned 11 yesterday, and like many children of that age, the celebrations were kind of interrupted by its divorced parents' continuing battle for custody....
Or just have a senior member of cabin crew step onto the flight deck whenever one of the flight crew steps out so that there's always two people in the cockpit at any given time, and one can open the door if the other is incapacitated.
I think the prudent response to this incident would be to mandate that two members of crew should be on the flight deck at all times, so that there is always someone available to open the door. In practice, that would mean that the senior cabin crew member would have to step in whenever the pilot or co-pilot needs to go for a pee.
Oh, with these cheap, low cost, discount fliers today . . . yes, it sure feels like retail!
Cattle markets are considered wholesale.
Although a certain large dating website manages to catch cocktail in its obscenity filter.
I think you're confusing "fuck" with "Frigg". "Frigg" was a Norse god, and thus gets a capital. "Fuck" was a norse profanity, so doesn't. Except in sentences like that one, where it occurs at the beginning of the sentence.
Well, that's basically what flying is. If you're dumb enough to do that, you deserve to crash in a pile of smouldering flames.
That's some major league victim blaming there. Let's see if we can blame the victims of the Titanic for "letting someone stick them in a big tub of riveted metal and setting them afloat on the world's second largest ocean". Or maybe we can blame the dinosaurs for being stupid enough to live on a big rock with only one inadeuately small moon to help protect it from once-in-a-gazillion-years meteor strikes. Stupid dinosaurs -- they really had it coming.
I notice that the article does not give any actual specifications on how dark the test night was. It sounds like this gives someone a slight edge (100% vs 33% for the control group), so it won't replace technological night vision for a while at least.
Or possibly even pirate night-vision, for that matter.
Why did pirates wear eye-patches? Because an eye that receives no light adapts to the dark. After nightfall, they would take the patch off, and the patched eye would perform better than the eye that had been exposed to the glare of the sea all day. The "test subject" in this case wore darkened contact lenses and then stuck on a pair of sunglasses which he wore for a couple of hours before the test. By the time they went out to carry out the distance vision experiment, his eyes would have had longer to adjust to the dark than his companions, so his night vision would have been better than theirs anyway...
Not a patch, but having one eye as "control" wouldn't be a bad idea. If the subject had one "normal" eye and one "doped" one, you could make some kind of meaningful-ish comparison. The eye-patch would be a bad idea as the patched eye would be constantly adjusting to the darkness and would end up being "super-powered" itself by the time of the test.
Actually, if the buy was wearing darkened contact lenses and sunglasses after treatment and before the test, and the "control" subjects weren't, the whole thing's basically just a waste of time, as we don't know how much of his improved accuracy over the others was down to the increased light-sensitivity as a result of blocking out light for several hours.
Soulskill decides to use semi-trendy jargon in a headline, and you think it's some kind of admission from the establishment that they control the horizontal and the vertical? There's a hole in your hat. Take some of this -- it's Bacofoil. Quality stuff.
Fags? I don't smoke.