It's not the most flexible platform, but it is surprising (a) how much you can do with it and (b) how much creativity you can harness when you have a platform that's approachable by neophyte programmers with good ideas.
I've been involved with the App Inventor community for a while; anyone who has been using AAI has known this time was coming. The hard deadline was set by Google a while back, the target from MIT was to get an analogue to the appinventorbeta.com service up and running by 12/31. They're a little short of that goal for a few important reasons.
Most importantly, the original App Inventor engine (at google) ran on top of google-proprietary internals. That is to say, it was not possible to spin out the App Inventor backend (which handles building and packaging an APK file from the blocks program) onto MIT servers without doing what amounted to a complete rewrite to enable it to run on App Engine. Google supported this effort and handed it off to MIT, who have been working hard to get an up-and-running system ASAP. As for right now:
If you absolutely rely on App Inventor, you can now run your own parallel instance of the backend by deploying the system as linked in the OP. This works nicely, because now the system is completely under your control (and you can hack it if you choose). This is useful for people teaching classes that use AAI as a platform (as I will be doing later this semester), but isn't so great for hobby programmers.
If you want something that runs like the old appinventorbeta.com site, allowing you to write apps but not have to worry about putting up your own backend server, wait a few weeks. There will be something up on appinventoredu.mit.edu fairly soon.
The reason why this transition is taking so long is nothing so nefarious as Google or MIT being evil or bad citizens. It's simple, really: there's not that many people actually doing coding for the project, and there's been a lot of coding to do. It'll be out soon. Patience, young apprentices.
App Inventor isn't going away, and as a matter of fact, the list of new features and useful extensions that are targeted for the coming few years is exciting and compelling.
Now imagine ten thousand of them, crammed together. Toss a few more in for good measure, since a 100W bulb produces about 98W of heat (2W of light, which is why we're trying to phase them out).
Now consider what that would do to the plastic bucket. And stop nitpicking on Joules vs Joules/sec. Shorthanding watts into "energy production" makes sense, because having a measure of total energy produced is kind of meaningless (a 1MJ plant that took a thousand years to produce that MJ wouldn't be that interesting).
This isn't really a legal avenue for *gold buying*. Instead it's an avenue for purchasing a vanity pet that you can sell to other players for in-game gold. The difference is two-fold:
1) No gold is added to the economy in this transaction, which means this process doesn't add inflationary pressure. 2) There's no guarantee of value on the pet. These things are going to flood the marketplace, as people buy them for resale. Who knows what the final market value of these things will be?
Blizzard could have just instituted some kind of $1 = 1,000 gold transfer, but they didn't. This will be at best a minimal gold buying avenue (I don't predict the value of these pets to be all that high in the long term.
On the other hand, the bigger deal is what the article missed- with all the previous vanity pets, $10 got you a pet that you could use on all your characters (Bind to Account). This new (Bind on Equip) pet means that you have to pick the character that gets the pet. That is somewhat disappointing, as I have a fair number of alts that I could conceivably give this pet, but I'm not going to spend $50 doing so.
While you're on a good train of reasoning there, Plausible Deniability isn't what we need when dealing with the RIAA. Remember that they send you a letter saying, "Pay us $1,500 now. If you want to try and plausibly deny this, you'll be on the hook for legal fees and we'll also end up getting a lot more in damages from you". The goal is to avoid getting one of those *cough* extortion *cough* letters in the first place.
While your claim is potentially true, it requires there to be Many (as in more than five) networks for each kind of data service (call them just wireless and wired), so you can choose amongst competing providers. When the providers own the physical layer (as comcast, att, etc do), there's not a lot of competition possible unless someone goes and builds another network in parallel to the cable / fiber networks the major players in the US already own.
It's also worth noting that those cable / fiber networks were highly subsidized by federal and local dollars. A ton of that infrastructure was effectively paid for by the US and given as a gift to the cable companies, each of which eventually got glommed up into whatever single cable monopoly is operating in your locality. I feel as though a federal regulation on net neutrality is in order when the feds paid for the network in the first place. Creating an alternative wireless network is even more difficult, given the finite spectrum resources we have to work with in the first place.
I agree that a highly-competitive market is likely to fix some things without regulation (but not necessarily all: there are, for example, plenty of highly-competitive companies in the automobile market right now, yet none actually produce a product that I want, or consider the results of our deregulation of the financial services industry- that was a market that failed to invisible-hand correct itself, everybody involved just looked away). But until the capacity for actual competition exists, and as long as the markets in question are benefiting from a capital investment where the government amortized all the risk on behalf of the cable (and phone for that matter) providers, then some regulations are probably in order.
My position is the opposite of yours: let's consider bits just like any other utility - electricity, water, natural gas. Create a public utility that has a monopoly license in the region if necessary, but strongly differentiate between the utility provided and the things that use that utility. My natural gas provider doesn't get to sell me a furnace (nor do they have any say in what kind of furnace I buy - no vendor lock-in there). Why should my bandwidth provider have thing one to say about what I do with the bits I bought?
Recursion jokes aside, I wonder how long it will be before we have fully-functioning ports of (say) java running in javascript. Because that'd be a hilarious way around Apple's language / appstore restriction on idevices.
Hmm, so I can have my "theory" that the internet is a series of tubes driven by hamsters? Or that you're actually a frog from space? Since you can't disprove a theory..... oh wait. You can disprove theories, can't you? You just can't logically prove them. Hmm.
Actually, gravity is *not* a theory. Newton's law of universal gravitation is a law. The difference is that a law is typically a direct relationship supported by empirical evidence. If you drop something, it falls. Newton worked out all the numbers. Same thing for Charles' law about gases and so on. What those *laws* don't do is say *why* or *how* the law works. A theory of gravitation includes a mechanism.
Furthermore, while it's not technically possible to *prove* any theory (by default, theories deal with unobservable entities that we can only understand by looking at their effects on the macroscopic world), it is certainly possible to disprove a theory. I have a theory that all matter is made up of continuous material, I shoot electron beams at a thin gold foil and observe (like Rutherford) a scattering that is inconsistent with that theory, and BLAM. Dead theory. Pardon my violent language, it's just that radical relativism makes me somewhat angry.
What I'm getting at is that you have an incorrect (and provably so) conceptualization of what the word "theory" means. Anyone who uses the phrase "still a theory" or "just a theory" has the same incorrect conceptualization of theory. The scientific community does itself a disservice by not educating people about how they use language better, but then again I've observed time and again that when the broader scientific community attempts to educate people, they tend to do things like cover their ears and say "lalala".
In closing, the "theory" of evolution is actually more of a set of interconnected theories that successfully aligns several hypothesized mechanisms with the empirically observed differentiation of species, geological age of the earth, and direct observations on the microscopic timescale of genetic drift. The "theory" of creation has no empirical support, and when confronted with actual disconfirmatory evidence its supporters either cover their ears and say "lalala" or they wave their hands and say "just a theory". BLAM. Dead theory.
There's nothing wrong with believing in a creation myth. I personally am a pastafarian (my heaven is waaaay better). But don't conflate irrational clinging to a belief with "proof" that that belief fits scientific models. And don't conflate the status of evolution as "theory" with "uncertainty". All the theory tag attaches is the notion that the entity it describes includes mechanistic and/or causal reasoning and appeals to logical reasoning in addition to empirical observations.
tl;dr: You're wrong, but you have a very common misconception about the scientific method. Plenty of people think that because you can never prove something true in all ways that any claim will do (since you can't prove that you're right, you can't prove that I'm wrong). This is known as "radical relativism", and is a dead end in reasoning.
You know, I was just going to post something like this myself. Except you beat me to it. So hey, I'll just post here instead. Curse you computers for taking my brains away or something.
It's also trivially easy with a few python scripts to strip the DRM from any kindle book you've purchased legally (think about it- the kindle has to be able to decrypt the book, and it's running on a pretty small chip). All you need to do is extract your decrypt key from the kindle, which turns out to be a function of the kindle's serial number.
I've done this for all the books I've bought for the kindle, to save a "just in case" version. It's also worth noting that the majority of piratebay books are pretty lousy OCR scans of books, with lots of markup and text errors. All the harry dresden books, for example, decided to be in a bold fond in the version I downloaded. Makes purchasing them a LOT more worthwhile (which I ended up doing for the first few, until I decided to give up on the series, but that's another story).
Still, I recognize that purchasing from amazon/bn/whoever is just supporting the business model of DRM, even if I strip out the DRM later. Would be nice to get somebody who didn't use platform lockin techniques, but that's probably unlikely in the near term.
Well lookit that. I guess I could have figured that out by clicking the homepage link in your comment page. I must say, I wouldn't be keeping current on developments in Mary Worth if it weren't for your blog over there.
Are you also the same Josh Fruhlinger that writes the Comics Curmudgeon blog? Name seems familiar, and although I at first thought it was just a coincidence, the connection is understandable. After all, plenty of people are interested in programming, blogging, and comics. Like me.
There is a great iPhone application programming course on iTunesU, which includes an introduction to ObjC and rapidly moves into powerful programming techniques for iOS. Better yet, it uses a lot of the examples from the SDK as the course material (there's no book), and the slides are provided as part of the program. It's all free to boot. Definitely worth watching for a good introduction to the platform. Plus, it's all free.
It's the Stanford University CS 193P, iPhone Application Programming, Spring 2010 course that I watched.
"An obscure component manufacturer somewhere in the Pacific Rim announces a major order for some bleeding-edge piece of technology that could conceivably become part of an expensive, digital-lifestyle-enhancing nerd toy"
There are two problems with the premise of this article. The first is inherent- while there are clearly better and worse ways to teach (and lectures in particular are usually pretty poor), there is no One Right way to do it. Teaching is like Perl. Different students require different kinds of experiences to learn stuff, and different material lends itself to different forms of teaching. Requiring "tech" is stupid. Teaching is a complex activity requiring specialized knowledge and skills, not just "tech". Requiring that faculty keep abreast of best practices in the field and implement those best practices is probably a good idea. Of course, that's not their job, which brings me to the other problem.
Faculty at prestigious institutions are hired for their understanding of the advanced content in their field, their ability to carry out publishable research in that field, and their ability to secure funding to do more research in that field. In other words: faculty are judged by how well they raise the profile of the institution. Faculty teaching ability is weighted at or near to last place when considering the "quality" of the professor. Sure, a good teacher is a good idea, but anybody spending lots of time on their teaching is taking away time from their research and grantwriting, which are the activities that will lead to tenure, promotion, and other forms of recognition from the broader community.
In short: as long as we keep selecting faculty based on their academic achievements, we will have faculty that's good at research but not good at teaching. No amount of anything will change that- teaching is a difficult job, and simply layering on requirements to use courseware or videoconferencing will not help.
Software patch cannot fix signal attenuation from a hand.
Actually, a software patch *can* help when it changes how the software-tunable capacitors in the antenna system respond. Not that there's anything about doing that in this particular press release, but you're being a bit under-optimistic here.
I had heard of a scam wherein hackers change your outgoing voicemail message to be "I accept the charges", and then call you collect from one of those strange high-priced calling codes. Effectively, you end up responsible for a huge phone bill, some percentage of which goes to the hackers.
This could be one of those urban legends too- it's late and I'm too tired to confirm it right now, but one can at least see how this isn't necessarily a non-issue.
Slashdot Mirror
I'm sort of curious if you've ever actually tried to use app inventor to do anything interesting. Because as an educational platform, or a place for young adults to learn about software, or a way to give youth a voice using digital storytelling, it seems to do pretty well.
It's not the most flexible platform, but it is surprising (a) how much you can do with it and (b) how much creativity you can harness when you have a platform that's approachable by neophyte programmers with good ideas.
Relevant: http://appinventoredu.mit.edu/faq-app-inventor-transition-mit
I've been involved with the App Inventor community for a while; anyone who has been using AAI has known this time was coming. The hard deadline was set by Google a while back, the target from MIT was to get an analogue to the appinventorbeta.com service up and running by 12/31. They're a little short of that goal for a few important reasons.
Most importantly, the original App Inventor engine (at google) ran on top of google-proprietary internals. That is to say, it was not possible to spin out the App Inventor backend (which handles building and packaging an APK file from the blocks program) onto MIT servers without doing what amounted to a complete rewrite to enable it to run on App Engine. Google supported this effort and handed it off to MIT, who have been working hard to get an up-and-running system ASAP. As for right now:
If you absolutely rely on App Inventor, you can now run your own parallel instance of the backend by deploying the system as linked in the OP. This works nicely, because now the system is completely under your control (and you can hack it if you choose). This is useful for people teaching classes that use AAI as a platform (as I will be doing later this semester), but isn't so great for hobby programmers.
If you want something that runs like the old appinventorbeta.com site, allowing you to write apps but not have to worry about putting up your own backend server, wait a few weeks. There will be something up on appinventoredu.mit.edu fairly soon.
The reason why this transition is taking so long is nothing so nefarious as Google or MIT being evil or bad citizens. It's simple, really: there's not that many people actually doing coding for the project, and there's been a lot of coding to do. It'll be out soon. Patience, young apprentices.
App Inventor isn't going away, and as a matter of fact, the list of new features and useful extensions that are targeted for the coming few years is exciting and compelling.
Imagine a hundred-watt light bulb.
Now imagine ten thousand of them, crammed together. Toss a few more in for good measure, since a 100W bulb produces about 98W of heat (2W of light, which is why we're trying to phase them out).
Now consider what that would do to the plastic bucket. And stop nitpicking on Joules vs Joules/sec. Shorthanding watts into "energy production" makes sense, because having a measure of total energy produced is kind of meaningless (a 1MJ plant that took a thousand years to produce that MJ wouldn't be that interesting).
So yeah, seems unlikely.
This isn't really a legal avenue for *gold buying*. Instead it's an avenue for purchasing a vanity pet that you can sell to other players for in-game gold. The difference is two-fold:
1) No gold is added to the economy in this transaction, which means this process doesn't add inflationary pressure.
2) There's no guarantee of value on the pet. These things are going to flood the marketplace, as people buy them for resale. Who knows what the final market value of these things will be?
Blizzard could have just instituted some kind of $1 = 1,000 gold transfer, but they didn't. This will be at best a minimal gold buying avenue (I don't predict the value of these pets to be all that high in the long term.
On the other hand, the bigger deal is what the article missed- with all the previous vanity pets, $10 got you a pet that you could use on all your characters (Bind to Account). This new (Bind on Equip) pet means that you have to pick the character that gets the pet. That is somewhat disappointing, as I have a fair number of alts that I could conceivably give this pet, but I'm not going to spend $50 doing so.
I am pretty sure you should have used "clearly" in that post title. Which might be just a teensy bit ironic. Not sure.
Thanks for everything. Been a reader for ages. Love this place.
Don't they already have 4chan for their social networking? Isn't that sufficient?
While you're on a good train of reasoning there, Plausible Deniability isn't what we need when dealing with the RIAA. Remember that they send you a letter saying, "Pay us $1,500 now. If you want to try and plausibly deny this, you'll be on the hook for legal fees and we'll also end up getting a lot more in damages from you". The goal is to avoid getting one of those *cough* extortion *cough* letters in the first place.
While your claim is potentially true, it requires there to be Many (as in more than five) networks for each kind of data service (call them just wireless and wired), so you can choose amongst competing providers. When the providers own the physical layer (as comcast, att, etc do), there's not a lot of competition possible unless someone goes and builds another network in parallel to the cable / fiber networks the major players in the US already own.
It's also worth noting that those cable / fiber networks were highly subsidized by federal and local dollars. A ton of that infrastructure was effectively paid for by the US and given as a gift to the cable companies, each of which eventually got glommed up into whatever single cable monopoly is operating in your locality. I feel as though a federal regulation on net neutrality is in order when the feds paid for the network in the first place. Creating an alternative wireless network is even more difficult, given the finite spectrum resources we have to work with in the first place.
I agree that a highly-competitive market is likely to fix some things without regulation (but not necessarily all: there are, for example, plenty of highly-competitive companies in the automobile market right now, yet none actually produce a product that I want, or consider the results of our deregulation of the financial services industry- that was a market that failed to invisible-hand correct itself, everybody involved just looked away). But until the capacity for actual competition exists, and as long as the markets in question are benefiting from a capital investment where the government amortized all the risk on behalf of the cable (and phone for that matter) providers, then some regulations are probably in order.
My position is the opposite of yours: let's consider bits just like any other utility - electricity, water, natural gas. Create a public utility that has a monopoly license in the region if necessary, but strongly differentiate between the utility provided and the things that use that utility. My natural gas provider doesn't get to sell me a furnace (nor do they have any say in what kind of furnace I buy - no vendor lock-in there). Why should my bandwidth provider have thing one to say about what I do with the bits I bought?
Recursion jokes aside, I wonder how long it will be before we have fully-functioning ports of (say) java running in javascript. Because that'd be a hilarious way around Apple's language / appstore restriction on idevices.
Buy two and keep a spare.
Isn't Nathan Myhrvold the guy who wrote the latest food pornstravaganza, Modernist Cuisine?
Hmm, so I can have my "theory" that the internet is a series of tubes driven by hamsters? Or that you're actually a frog from space? Since you can't disprove a theory..... oh wait. You can disprove theories, can't you? You just can't logically prove them. Hmm.
Actually, gravity is *not* a theory. Newton's law of universal gravitation is a law. The difference is that a law is typically a direct relationship supported by empirical evidence. If you drop something, it falls. Newton worked out all the numbers. Same thing for Charles' law about gases and so on. What those *laws* don't do is say *why* or *how* the law works. A theory of gravitation includes a mechanism.
Furthermore, while it's not technically possible to *prove* any theory (by default, theories deal with unobservable entities that we can only understand by looking at their effects on the macroscopic world), it is certainly possible to disprove a theory. I have a theory that all matter is made up of continuous material, I shoot electron beams at a thin gold foil and observe (like Rutherford) a scattering that is inconsistent with that theory, and BLAM. Dead theory. Pardon my violent language, it's just that radical relativism makes me somewhat angry.
What I'm getting at is that you have an incorrect (and provably so) conceptualization of what the word "theory" means. Anyone who uses the phrase "still a theory" or "just a theory" has the same incorrect conceptualization of theory. The scientific community does itself a disservice by not educating people about how they use language better, but then again I've observed time and again that when the broader scientific community attempts to educate people, they tend to do things like cover their ears and say "lalala".
In closing, the "theory" of evolution is actually more of a set of interconnected theories that successfully aligns several hypothesized mechanisms with the empirically observed differentiation of species, geological age of the earth, and direct observations on the microscopic timescale of genetic drift. The "theory" of creation has no empirical support, and when confronted with actual disconfirmatory evidence its supporters either cover their ears and say "lalala" or they wave their hands and say "just a theory". BLAM. Dead theory.
There's nothing wrong with believing in a creation myth. I personally am a pastafarian (my heaven is waaaay better). But don't conflate irrational clinging to a belief with "proof" that that belief fits scientific models. And don't conflate the status of evolution as "theory" with "uncertainty". All the theory tag attaches is the notion that the entity it describes includes mechanistic and/or causal reasoning and appeals to logical reasoning in addition to empirical observations.
tl;dr: You're wrong, but you have a very common misconception about the scientific method. Plenty of people think that because you can never prove something true in all ways that any claim will do (since you can't prove that you're right, you can't prove that I'm wrong). This is known as "radical relativism", and is a dead end in reasoning.
You know, I was just going to post something like this myself. Except you beat me to it. So hey, I'll just post here instead. Curse you computers for taking my brains away or something.
http://www.misterbg.org/AppleProductCycle/
It's continuously relevant.
It's also trivially easy with a few python scripts to strip the DRM from any kindle book you've purchased legally (think about it- the kindle has to be able to decrypt the book, and it's running on a pretty small chip). All you need to do is extract your decrypt key from the kindle, which turns out to be a function of the kindle's serial number.
I've done this for all the books I've bought for the kindle, to save a "just in case" version. It's also worth noting that the majority of piratebay books are pretty lousy OCR scans of books, with lots of markup and text errors. All the harry dresden books, for example, decided to be in a bold fond in the version I downloaded. Makes purchasing them a LOT more worthwhile (which I ended up doing for the first few, until I decided to give up on the series, but that's another story).
Still, I recognize that purchasing from amazon/bn/whoever is just supporting the business model of DRM, even if I strip out the DRM later. Would be nice to get somebody who didn't use platform lockin techniques, but that's probably unlikely in the near term.
Well lookit that. I guess I could have figured that out by clicking the homepage link in your comment page. I must say, I wouldn't be keeping current on developments in Mary Worth if it weren't for your blog over there.
Are you also the same Josh Fruhlinger that writes the Comics Curmudgeon blog? Name seems familiar, and although I at first thought it was just a coincidence, the connection is understandable. After all, plenty of people are interested in programming, blogging, and comics. Like me.
Even better. "Who is suing whom". Because grammar matters.
WHOM.
There is a great iPhone application programming course on iTunesU, which includes an introduction to ObjC and rapidly moves into powerful programming techniques for iOS. Better yet, it uses a lot of the examples from the SDK as the course material (there's no book), and the slides are provided as part of the program. It's all free to boot. Definitely worth watching for a good introduction to the platform. Plus, it's all free.
It's the Stanford University CS 193P, iPhone Application Programming, Spring 2010 course that I watched.
Seems like we're at:
"An obscure component manufacturer somewhere in the Pacific Rim announces a major order for some bleeding-edge piece of technology that could conceivably become part of an expensive, digital-lifestyle-enhancing nerd toy"
http://www.misterbg.org/AppleProductCycle/
There are two problems with the premise of this article. The first is inherent- while there are clearly better and worse ways to teach (and lectures in particular are usually pretty poor), there is no One Right way to do it. Teaching is like Perl. Different students require different kinds of experiences to learn stuff, and different material lends itself to different forms of teaching. Requiring "tech" is stupid. Teaching is a complex activity requiring specialized knowledge and skills, not just "tech". Requiring that faculty keep abreast of best practices in the field and implement those best practices is probably a good idea. Of course, that's not their job, which brings me to the other problem.
Faculty at prestigious institutions are hired for their understanding of the advanced content in their field, their ability to carry out publishable research in that field, and their ability to secure funding to do more research in that field. In other words: faculty are judged by how well they raise the profile of the institution. Faculty teaching ability is weighted at or near to last place when considering the "quality" of the professor. Sure, a good teacher is a good idea, but anybody spending lots of time on their teaching is taking away time from their research and grantwriting, which are the activities that will lead to tenure, promotion, and other forms of recognition from the broader community.
In short: as long as we keep selecting faculty based on their academic achievements, we will have faculty that's good at research but not good at teaching. No amount of anything will change that- teaching is a difficult job, and simply layering on requirements to use courseware or videoconferencing will not help.
Software patch cannot fix signal attenuation from a hand.
Actually, a software patch *can* help when it changes how the software-tunable capacitors in the antenna system respond. Not that there's anything about doing that in this particular press release, but you're being a bit under-optimistic here.
I had heard of a scam wherein hackers change your outgoing voicemail message to be "I accept the charges", and then call you collect from one of those strange high-priced calling codes. Effectively, you end up responsible for a huge phone bill, some percentage of which goes to the hackers.
This could be one of those urban legends too- it's late and I'm too tired to confirm it right now, but one can at least see how this isn't necessarily a non-issue.