Re:Chiropractic treatment worked for me
on
Trick or Treatment
·
· Score: 1
Every person I have ever known who has started going to a chiropractor has had to _continue_ going to a chiropractor for the rest of their lives.
That is not a selling point.
Re:The author is wrong about accupuncture
on
Trick or Treatment
·
· Score: 1
Except that it's the responsibility of the _vendor_ to prove efficacy, not me to prove it's bunk.
That paragraph is for the _second_ deal they are offering.
The company is also launching another interesting scheme with its Esprimo range, offering a complete refund of the original sales price if the customer needs to send the notebook back to Fujitsu Siemens for any repairs.
Of course, the author used hardcover instead of paperback pricing for books. If I'm buying a hardcover of a book, it's because I want to put the book on the shelf.
Otherwise, I'm buying a paperback, which is typically cheaper than the equivalent Kindle price.
The Bricking was caused by a bug in the unlock code. The unlock code resulted in a bad security zone. When the baseband was later upgraded, the security zone would fail verification and the phone was locked up.
However, it wasn't "bricked", there were methods to recover from the bad upgrade, and later unlocks fixed the problem from the first one.
There were other problems with various unlocks. One jailbreak resulted in every iPhone using that tool having the same MAC address.
2 assignments, 100% = 2% first test, 100% = 15% 0 on remainder = (100-17)*(50/100) = 41.5
So, we've got a 58% just with that. Do 2 more assignments, and we've got a pass, without even showing up for class for the majority of the tests, or the final exam.
The problem occurs regardess of makeup and layout. I encourage you to model your own course layouts and see what minimum level of effort would be needed to obtain a pass under these rules.
Replace actually doing the test with not showing up for class that day.
It's a manufactured example, with extremes to demonstrate the point.
The majority of students will decide at the beginning of the term what grade they need to achieve what they want, and then once they've earned that mark, stop. Allowing a zero to earn the 50 will exacerbate that problem - the student will be able to begin coasting earlier in the term.
Actually, under existing systems, the instructor would invoke their own ability to scale your mark and give you a mark they think properly reflects your indicated level of knowledge.
This is precisely what happened to me, where I got grades that were mathematically impossible to earn based on the work I had handed in.
However, codifying such behaviour removes the incentive to do the work in the first place.
Of course it is just as possible to game your system as it is with the one in Pennsylvania.
How many marking periods are there? 2? 3? 4?
Let's say there are 2 marking periods of equal weight. What mark is needed to pass?
Goal: 65% First term: 75% Second term: 0% converted to 55% Final grade: 65%, pass.
Did that student really earned a pass?
If you think that your students haven't figured this out... Consider this, it only takes _one_ student to figure it out and they will explain it to the rest of them.
With facebook/email/instant messaging, do you honestly think that this sort of information wouldn't go viral?
"The 'E' is to be recorded no lower than a 50 percent, regardless of the actual percent earned. For example, if the student earns a 20 percent on a class assignment, the grade is recorded as a 50 percent," said the memo from Jerri Lippert, the district's executive director of curriculum, instruction and professional development, and Mary VanHorn, a PFT vice president.
The problem is actually caused by having assessments too late in the course. The problem would be more easily handled by having more assessments, starting earlier. That way, "tanking" on any one item doesn't greatly affect your mark, and provides early feedback to the teacher that there is a problem to be solved.
Has anyone thought of what this actually means? Mathematically?
For example, let's say there are 5 assignments and 2 tests. The tests are worth 25% of your final mark. The assignments are worth 10% each.
Additionally, let's go with the ABCDE scheme, and the student needs a 60% to pass with a D.
What's the minimum mathematical grade needed to pass?
First the tests: 0% on either test.
We've now got 25% on the course.
Then the assignments:
3 assignments: 0%
We've now got 40% on the course.
2 assignments: 100%
We've now got our 60%, D grade for the course.
That means even though the student received a mathematical 20% when their entire coursework is taken into account, they would receive a D.
That is definite grade inflation.
Based on my behaviour in high school, I would have most definitely gotten 100% on the first two assignments, and then skipped the rest of the term, walking out with my 60%. Would I have known the material? Definitely not. Would I have known 60% of the material? Definitely not.
The price per a la carte SMS has gone up, however there are plenty of cheap plans replacing them.
So, you've got people shifting to the plans, typically your high-quantity users. The customers remaining are unwilling to pay US$10/mo for an SMS plan. These people will be the ones who don't send many per month.
Since the cost to bill an SMS is fixed (the network has been purchased, but not paid for), and as the number of a la carte SMS that the network delivers drops, that component of the per-SMS price goes up. That will result in upwards price pressure.
Even better, SMS bundles are cheaper to administer and bill than a la carte. That provides the carrier with an additional incentive to convince people to shift to a plan.
BTW, customers like the certainty that comes with a fixed-price unlimited bundle.
Also, I've seen figures saying that about 2/3..3/4 of a carrier's operating costs revolve around billing.
Yes, in the US, it costs the same wherever you are, but it still costs you. In Europe they have a "caller pays" system, where it is generally free to receive a call, unless you aren't on your home network. Then you pay insane prices.
For example, it costs me US$3.10 to place a call to a US number when I am roaming in the US. To call back to NZ, it costs NZ$3.23 (on AT&T).
You can realize some pretty amazing savings if you can avoid roaming.
As for 3G coverage, since the VoIP app isn't allowed to use 3G data, I don't think 3G coverage matters. 2G coverage is what is important, not that it is any better (in my experience). 3G UMTS phones will generally fall back to 2G GSM, if the 3G network isn't visible.
1) Americans/Canadians pay to receive calls. 2) When roaming, you can pay insane charges to receive a call, even on "caller pays" networks. 3) VoIP calls can use higher bandwidth codecs, resulting in better sound quality. 4) Coverage. If your carrier doesn't have coverage where you are (in a mall/building/airport), but you do have WiFi, you can still make and receive calls.
Having just come from the US and Canada, their GSM coverage was amazingly bad (compared to NZ/UK). I was in San Francisco, and I kept losing the signal!
I think it depends on your definition of "seamless".
I don't think it requires carrier interaction.
When transferring to VoIP, it can be done seamlessly, the application is launched and the call handed off to VoIP. Transferring out will result in a ring (needing to be answered) on the device, but it can still be done without loss of packets in the call.
If the application is smart, it could possibly track the strength of the WiFi network and transfer the call pre-emptively.
I would expect that their application is using a VoIP server with an entirely different number from your mobile number. That way they are able to perform their own call control, complete with multiple B-legs without needing any carrier agreement.
It would be great when roaming.... Imagine sitting in your hotel room, receiving phone calls on your mobile phone and not paying the cell phone company a dime in international roaming charges.
This would need you to take a new phone number, much like Grand Central.
Then, when the call arrives, the SIP Invite is forwarded to the application (if running), and the user is prompted to decide on delivery mechanism.
If the app isn't running, the call is connected. If at a later point, the user starts the application, the app registers with the service, and, if desired, the call is dropped from the mobile connection and sent to the VoIP link using a reinvite (probably).
This can be probably be done using Asterisk on the server side. The nifty bit is the VoIP client on the iPhone. Other than that, the service looks pretty bulk standard.
This definitely wouldn't need anything other than the standard APIs.
What they aren't doing is using the built-in Mobile Phone Application and intelligently re-routing outgoing calls based on the presence of a WiFi connection, the way that TruPhone was going last September.
I think they would have some pretty extreme problems constructing a business case around selling this through the AppStore. Apple's current billing and charging limitations pretty much kill it instantly.
Privacy? How about protection from bad data?!?
on
Understanding Privacy
·
· Score: 0, Offtopic
I just had an interesting experience on the F-line in San Francisco. A woman was having a discussion in the seat behind me. It seemed that her husband had been stopped by the police in a parking lot and told that his license had been rescinded. That meant that if he got into his car, he was driving illegally. The police took his license, told him not to get in the car and left.
Now, the interesting thing is that his license was not invalid. There was a problem at the DMV. He went down to the DMV (taking a cab, he can't drive remember) waited in line for a couple of hours, got it sorted out, got a letter and went back to the police.
They still refused to give him back his license. He has to wait for a new one to appear in the mail. So, what does that tell you? A bureaucratic error resulted in lost hours, lost income and hassle. Was he guilty of anything? No. Caused by a loss of privacy, allowing buggy data to be accessed in real-time by the police.
Oh, perhaps you haven't heard of automatic license plate recognition systems? The police only have to drive by a car, and the computer pops up data on it.
Somehow, I doubt that Apple will release a 3G iPhone straight to Europe and NOT have it available in the US. In order to be used in the US, it needs FCC approval. The FCC operates an open approvals process, with the devices being displayed on their web site for everyone to see and comment on.
That approvals process takes time. I somehow doubt that it only takes 6-8 weeks.
Of course newspapers are still relevant. Even if all they do is repackage AP stories.
Newspapers provide context, aggregation and community. They have provided a location for discussion (letters to the editor) for pretty much their entire life. They are the proto-social news organisation.
I expect newspapers to change, but they will still exist.
I expect newspapers to change to offer a service where the reader is allowed to click through to the context behind a story. Stories are grouped by subject area, providing a living timeline on the subject. This will allow the reader to explore the story, becoming more informed as they go. Additionally, since stories will have subject areas, it will be possible to track an issue as it develops.
Tracking individual story issues will allow people to create their own personal newspaper, following the subjects that most interest them.
Add to that detailed discussion (also on a per issue basis), will bring in the community aspects. Imagine what slashdot would look like if, instead of starting each discussion with a blank slate, we started with the +5 comments from the previous time this subject had been reported on?
Finally, newspapers will still be able to excel at providing context (local/national/international), providing the reader with the information surrounding the events that might be missed if they just received the facts.
Slashdot Mirror
Every person I have ever known who has started going to a chiropractor has had to _continue_ going to a chiropractor for the rest of their lives. That is not a selling point.
Except that it's the responsibility of the _vendor_ to prove efficacy, not me to prove it's bunk.
That's all we need, second hand smoke that makes you high. No thanks.
I think you should re-read the article.
That paragraph is for the _second_ deal they are offering.
Of course, the author used hardcover instead of paperback pricing for books. If I'm buying a hardcover of a book, it's because I want to put the book on the shelf.
Otherwise, I'm buying a paperback, which is typically cheaper than the equivalent Kindle price.
That was changed, it's now open to third parties requesting the source. Otherwise, the secondary market (used equipment) is locked down again.
The Bricking was caused by a bug in the unlock code. The unlock code resulted in a bad security zone. When the baseband was later upgraded, the security zone would fail verification and the phone was locked up.
However, it wasn't "bricked", there were methods to recover from the bad upgrade, and later unlocks fixed the problem from the first one.
There were other problems with various unlocks. One jailbreak resulted in every iPhone using that tool having the same MAC address.
Then you haven't taken enough courses. :)
However, feel free to change the problem to:
1) 10 assignments worth 10% of total
2) 4 tests worth 60% of total
3) 1 final worth 30% of total
What's the minimum work? You might be surprised.
10 assignments, 0% = 5%
2 tests , 100% = 30%
2 tests , 0% = 15%
1 final , 0% = 15%
Total 65%
Just as silly.
How about we do everything to the first test.
2 assignments, 100% = 2%
first test, 100% = 15%
0 on remainder = (100-17)*(50/100) = 41.5
So, we've got a 58% just with that. Do 2 more assignments, and we've got a pass, without even showing up for class for the majority of the tests, or the final exam.
The problem occurs regardess of makeup and layout. I encourage you to model your own course layouts and see what minimum level of effort would be needed to obtain a pass under these rules.
Replace actually doing the test with not showing up for class that day.
It's a manufactured example, with extremes to demonstrate the point.
The majority of students will decide at the beginning of the term what grade they need to achieve what they want, and then once they've earned that mark, stop. Allowing a zero to earn the 50 will exacerbate that problem - the student will be able to begin coasting earlier in the term.
ref: senioritis.
Actually, under existing systems, the instructor would invoke their own ability to scale your mark and give you a mark they think properly reflects your indicated level of knowledge.
This is precisely what happened to me, where I got grades that were mathematically impossible to earn based on the work I had handed in.
However, codifying such behaviour removes the incentive to do the work in the first place.
Of course it is just as possible to game your system as it is with the one in Pennsylvania.
How many marking periods are there? 2? 3? 4?
Let's say there are 2 marking periods of equal weight. What mark is needed to pass?
Goal: 65%
First term: 75%
Second term: 0% converted to 55%
Final grade: 65%, pass.
Did that student really earned a pass?
If you think that your students haven't figured this out... Consider this, it only takes _one_ student to figure it out and they will explain it to the rest of them.
With facebook/email/instant messaging, do you honestly think that this sort of information wouldn't go viral?
From TFA:
It's not an end-of term grade at all.
The problem is actually caused by having assessments too late in the course. The problem would be more easily handled by having more assessments, starting earlier. That way, "tanking" on any one item doesn't greatly affect your mark, and provides early feedback to the teacher that there is a problem to be solved.
Has anyone thought of what this actually means? Mathematically?
For example, let's say there are 5 assignments and 2 tests. The tests are worth 25% of your final mark.
The assignments are worth 10% each.
Additionally, let's go with the ABCDE scheme, and the student needs a 60% to pass with a D.
What's the minimum mathematical grade needed to pass?
First the tests: 0% on either test.
We've now got 25% on the course.
Then the assignments:
3 assignments: 0%
We've now got 40% on the course.
2 assignments: 100%
We've now got our 60%, D grade for the course.
That means even though the student received a mathematical 20% when their entire coursework is taken into account, they would receive a D.
That is definite grade inflation.
Based on my behaviour in high school, I would have most definitely gotten 100% on the first two assignments, and then skipped the rest of the term, walking out with my 60%. Would I have known the material? Definitely not. Would I have known 60% of the material? Definitely not.
The price per a la carte SMS has gone up, however there are plenty of cheap plans replacing them.
So, you've got people shifting to the plans, typically your high-quantity users. The customers remaining are unwilling to pay US$10/mo for an SMS plan. These people will be the ones who don't send many per month.
Since the cost to bill an SMS is fixed (the network has been purchased, but not paid for), and as the number of a la carte SMS that the network delivers drops, that component of the per-SMS price goes up. That will result in upwards price pressure.
Even better, SMS bundles are cheaper to administer and bill than a la carte. That provides the carrier with an additional incentive to convince people to shift to a plan.
BTW, customers like the certainty that comes with a fixed-price unlimited bundle.
Also, I've seen figures saying that about 2/3..3/4 of a carrier's operating costs revolve around billing.
Yes, in the US, it costs the same wherever you are, but it still costs you. In Europe they have a "caller pays" system, where it is generally free to receive a call, unless you aren't on your home network. Then you pay insane prices.
For example, it costs me US$3.10 to place a call to a US number when I am roaming in the US. To call back to NZ, it costs NZ$3.23 (on AT&T).
You can realize some pretty amazing savings if you can avoid roaming.
As for 3G coverage, since the VoIP app isn't allowed to use 3G data, I don't think 3G coverage matters. 2G coverage is what is important, not that it is any better (in my experience). 3G UMTS phones will generally fall back to 2G GSM, if the 3G network isn't visible.
Several possible reasons
1) Americans/Canadians pay to receive calls.
2) When roaming, you can pay insane charges to receive a call, even on "caller pays" networks.
3) VoIP calls can use higher bandwidth codecs, resulting in better sound quality.
4) Coverage. If your carrier doesn't have coverage where you are (in a mall/building/airport), but you do have WiFi, you can still make and receive calls.
Having just come from the US and Canada, their GSM coverage was amazingly bad (compared to NZ/UK). I was in San Francisco, and I kept losing the signal!
I think it depends on your definition of "seamless".
I don't think it requires carrier interaction.
When transferring to VoIP, it can be done seamlessly, the application is launched and the call handed off to VoIP. Transferring out will result in a ring (needing to be answered) on the device, but it can still be done without loss of packets in the call.
If the application is smart, it could possibly track the strength of the WiFi network and transfer the call pre-emptively.
I would expect that their application is using a VoIP server with an entirely different number from your mobile number. That way they are able to perform their own call control, complete with multiple B-legs without needing any carrier agreement.
It would be great when roaming.... Imagine sitting in your hotel room, receiving phone calls on your mobile phone and not paying the cell phone company a dime in international roaming charges.
This isn't UMA. UMA requires the handset manufacturer and the carrier to cooperate.
Essentially, UMA is tunneling the entire GSM protocol over WiFi. Very much a short-term hack.
This would need you to take a new phone number, much like Grand Central.
Then, when the call arrives, the SIP Invite is forwarded to the application (if running), and the user is prompted to decide on delivery mechanism.
If the app isn't running, the call is connected. If at a later point, the user starts the application, the app registers with the service, and, if desired, the call is dropped from the mobile connection and sent to the VoIP link using a reinvite (probably).
This can be probably be done using Asterisk on the server side. The nifty bit is the VoIP client on the iPhone. Other than that, the service looks pretty bulk standard.
This definitely wouldn't need anything other than the standard APIs.
What they aren't doing is using the built-in Mobile Phone Application and intelligently re-routing outgoing calls based on the presence of a WiFi connection, the way that TruPhone was going last September.
I think they would have some pretty extreme problems constructing a business case around selling this through the AppStore. Apple's current billing and charging limitations pretty much kill it instantly.
I just had an interesting experience on the F-line in San Francisco. A woman was having a discussion in the seat behind me. It seemed that her husband had been stopped by the police in a parking lot and told that his license had been rescinded. That meant that if he got into his car, he was driving illegally. The police took his license, told him not to get in the car and left.
Now, the interesting thing is that his license was not invalid. There was a problem at the DMV. He went down to the DMV (taking a cab, he can't drive remember) waited in line for a couple of hours, got it sorted out, got a letter and went back to the police.
They still refused to give him back his license. He has to wait for a new one to appear in the mail. So, what does that tell you? A bureaucratic error resulted in lost hours, lost income and hassle. Was he guilty of anything? No. Caused by a loss of privacy, allowing buggy data to be accessed in real-time by the police.
Oh, perhaps you haven't heard of automatic license plate recognition systems? The police only have to drive by a car, and the computer pops up data on it.
Somehow, I doubt that Apple will release a 3G iPhone straight to Europe and NOT have it available in the US. In order to be used in the US, it needs FCC approval. The FCC operates an open approvals process, with the devices being displayed on their web site for everyone to see and comment on.
That approvals process takes time. I somehow doubt that it only takes 6-8 weeks.
So, retail availability in June?
Yep, doesn't work for me here with Firefox on OSX.
Oh well.
Of course newspapers are still relevant. Even if all they do is repackage AP stories.
Newspapers provide context, aggregation and community. They have provided a location for discussion (letters to the editor) for pretty much their entire life. They are the proto-social news organisation.
I expect newspapers to change, but they will still exist.
I expect newspapers to change to offer a service where the reader is allowed to click through to the context behind a story. Stories are grouped by subject area, providing a living timeline on the subject. This will allow the reader to explore the story, becoming more informed as they go. Additionally, since stories will have subject areas, it will be possible to track an issue as it develops.
Tracking individual story issues will allow people to create their own personal newspaper, following the subjects that most interest them.
Add to that detailed discussion (also on a per issue basis), will bring in the community aspects. Imagine what slashdot would look like if, instead of starting each discussion with a blank slate, we started with the +5 comments from the previous time this subject had been reported on?
Finally, newspapers will still be able to excel at providing context (local/national/international), providing the reader with the information surrounding the events that might be missed if they just received the facts.