In Illinois employers must give workers time off to vote, so asking for a simple verification that they did so seems fair.
Maybe, but it's not actually necessary, is it? We're not talking about them taking time off any time they want it, after all.
But the main point is that there are people actually suggesting that voters should get a receipt of *how they voted*, to act as a paper trail. I'm not saying that's what happened here, I'm saying that this kind of incident demonstrates that the argument against it isn't just a paper tiger.
Eliminating the grace period also eliminates front-running. If you read the transcript, NSI had in fact indicated that they would roll back front-running if they lost the grace period.
The more important part, how the brain can sublimate operating complex machinery so that it doesn't require conscious thought to operate, isn't explained here.
Same way the body can sublimate operating the body itself so it doesn't require conscious thought. Solve that problem, and you've solved this one.
I mean, look at how babies learn. Look at how you learn any skill. It all comes down to contracting muscles and getting feedback of a successful result... what this is demonstrating is that whether the levers the muscles are pulling are inside the body or outside is not a major factor.
And I doubt very much whether this capability was something that developed in primates or mammals or birds... any species that learns how to use its own body probably has this ability.
My friend nearly got beat up. Not by the poll workers, but by the voters. Many of them were city workers, who were afraid that if they didn't bring their voting receipt to their boss that they would be fired.
This is why all the schemes involving a "receipt" rather than a ballot as a "paper trail" are not just flawed but actually reduce the validity of a vote. When you leave the polling place you should not be permitted to carry anything that you didn't have when you entered it... even so much as a "how to vote" card.
I forget the brand, but it required a key card, and had a paper tabulation of your votes that was visible behind a clear window for your approval. After approving the paper backup, it scrolled up out of sight.
Two problems:
(1) Since the records are a tape, then by recording the order of voters in the precinct records and the order of records on the tape they can determine who voted which way to a fair degree of certainty.
(2) If the tape was treated as the primary ballot (that is, it was physically transferred to a counting station and read) then this would address many of the problems of an electronic ballot. However not only is it not treated as the primary ballot, but as a result in some cases such tapes were discarded as trash after the elections.
Text messages are sent in the overhead: even if they didn't implement text messages at all they'd still be building all that infrastructure for voice. In fact that's what they DID build it for. SMS is a side effect, piggybacked on other traffic: it doesn't cut into bandwidth for voice at all.
The marginal cost for text messages is 0.
Turning that around, if they were just setting up a text message service, they would need fewer and simpler cells because they wouldn't be trying to send nearly as much data over the airwaves. Ten seconds of talking requires as much bandwidth as hundreds of messages, and because voice conversation is real-time they have to send it synchronously and even a second's latency is noticable. A text-only network could delay a minute waiting for a slot to open to send a message and most people wouldn't notice most of the time.
Based on TOTAL cost recovery, as you suggest, text messages should at most be something like 1/10,000th the cost per minute of voice. Not 2 cents (that's gotta be a wild-assed guess), or even 0.2 cents: the phone company would still be making a hell of a profit off them at that rate.
I can see why they would want an authorization system, because they have already expressed their worries about iPhone malware.
That's their excuse. But that's not how effective malware typically gets into mobile devices... not that there's much malware for mobile devices out there at all, but what there is tends to be good old backdoors and buffer overflows, not crocked installers, because you don't typically download and install software directly to these devices so there's no way for malware to propagate from one device to the next.
I've been pointing this out since the antivirus companies started really pushing AV for Palm and Pocket PC several years back. There's no viral ecosystem for these devices, because there's no device-to-device transmission path that supports execution of code, with or without social engineering being involved. You don't install software directly on your handhelds (PDAs or phones), you do it on your desktop or laptop and download it to the handheld from there. And the iPhone is no different.
So there's no technical reason for this, and the security argument is devastatingly weak. It's all about control. Malware is just the excuse.
Please. I can't buy Mac hardware without buying MacOS X.
You mean you can't run Mac OS X without buying Apple hardware. While I'm sure there are a few people who buy into the Mac chic and buy Macs to run Windows or Linux on, the vast majority of people are buying Macs to run OS X. The Windows Tax is the extra cost of an OS you don't want to get the hardware you do. The Mac Tax is the extra cost of the hardware you have to buy to run the software you're buying it for.
That would increase the base cost of Internet access by 25-30%, if prices in Canada are in line with the USA. That seems out of line when compared with other compulsory licenses.
Grog watch video. Grog see Tufte try to cram full page of information into index card. Grog think Tufte missing point. Grog think Tufte not realize what good for printed page not good for small interactive gadget.
First of all, the straight line from upper left to lower right is inherent in the graph. The way it's defined, no points can be above that line.
Second, information storage technologies almost inevitably have a limited lifetime, and the longer the lifetime the more expensive it is. To store information like "when was fire first used as a tool" required hundreds of thousands of individual controlled fires to be kindled for there to be enough remains to survive to mark the time (yes, the people creating the fires were not deliberately creating a record of the invention, but the result and the effort are the same as if they were). For some astronomical events, the cost of recording them easily exceeds the lifetime energy production of any conceivable human civilization. The survival of durable records, once the immediate loss of data that was not initially preserved, follows an exponential decay curve.
So that graph doesn't really say anything about the rate of important events, it says something about the viewpoint of the people making the lists. There may have been billions of events as important as the discovery of fire or slood that we have no idea about, because no record of them exists, and so they won't skew the curve.
Hell, we haven't even rediscovered slood yet ourselves.
You're being ridiculous. If 17 cubic inches (roughly the volume of a wall-wart or a pair of socks) is that critical, there's something wrong with your packing.
I use a toshiba r10 which isn't an ultra portable but I find switching to tablet mode makes it easier to work in a cramped space.
Well now, if the Macbook Air had a tablet mode, we wouldn't be having this conversation. But it's not.
And I do remember the tray tables from the last time I travelled, and the time before that, and missing my Libretto because my Thinkpad was big enough I couldn't fit a glass of juice on th tray as well.
So you agree that because it's space savings eliminates the need for a larger briefcase that it is not a pretender ultra portable?
No, because we were talking about 1/4 of an inch.
There are limits to what you can take as a carry on bag you see.
i think the point is to think small here... the idea is to allow the smallest set of changes that work to kernels and code that still works and increases real security.
That's how we got in this mess in the first place.
There's a better patch that's even simpler, and that is quit pretending that restricting ports is a useful security feature in the first place.
In FreeBSD, that's done with "sysctl -w net.inet.ip.portrange.reservedhigh=0". I'm sure there's an equivalent in Linux.
That has the advantage that it's OS independent and it doesn't break existing installations that are already using groups for other things. The "reserved port" model made sense when computers that had network interfaces and could run IP were expensive, and access to a LAN required physical access to secure areas, but it's long since become little more than an annoyance.
So your point is that displaying a summary and then letting to user drill down to the information he wants is better than displaying all information into the first page.
No, that's not the point at all. The point is that if you can't fit more information on the first page, as in Tufte's example, you don't cram it in and demand the user interactively (there's that word again) zoom and scroll to see it all, you change the presentation, reduce the amount of information being presented (in the paper, one example was the location of the theatre) or change the way it's presented (showing time graphically rather than a number in a table), and so on. The information you're not including in the summary still exists. In the theatre example, after you've decided what show you're going to see, is that the end of it? No, you're going to buy tickets, or maybe you need to call the theatre to see if they have a changing table in the men's bathroom so you need the number, or you're on the phone talking to someone who doesn't know where the theatre is so you need the street address, but whatever it is, THAT is what you're drilling down for. In Tufte's stock market case, it's the comparison between two stocks.
Of course, now you simply assume that people only use the summary in 99% of all cases, which wasn't clear from your original post.
Yes, I understand that you didn't find that clear, which is why I am trying to clarify it.
But that implies that the summary already contains all the information which the user wants to see
That's the goal, yes.
I really don't know why you're even arguing with me.
Is that what you call it?
Possibly because you didn't watch the video, and jumped on my posting without knowing the context? I don't know, you tell me.
There is no interaction needed in 99% of all cases.
What does that have to do with anything?
If you can show the information the user wants to see, do it. Don't hide it.
Can you go back and show me where I suggested hiding information that the user wants to see?
Seriously?
The POINT of a summary is that it contains the information you need in most cases. If it isn't, then you're not doing a good job of summarizing. Tufte was sticking a bunch of extraneous detail into the page, more detail than you need at a glance, like the animated cloud map and his tens of thousands of numbers in his spark charts. My point is that if that's not information you need all the time (and it's not) it shouldn't be crammed into an index-card sized sheet of paper as if it was a sheet of foolscap, instead of cluttering up the display or making the user drag it around like a magnifying glass over a phone book, you take advantage of the interactive nature of the device. Let the user ask for more detail for the 1% of the time they need it.
Like, you know, the subway map, or trains leaving three hours from now, or bookmarks, or alarms.
I've only been talking about building real tables, for displaying tabular data within the flow of a document. Not using tables for layout.
But if you want to get into CSS Zen Garden, and all the trickery you need to do to make CSS layout actually work, well... sheesh.
I've used a variety of applications and tools for doing programatic layout of documents over the years, with a variety of levels of functionality, some good, some bad, and CSS is one of the worst. It seems that the people who designed CSS hadn't any actual experience with more than one tool, if that.
In CSS you only have two layout mechanisms: absolute, with absolute scaling, and relative, with absolute scaling. If you want to create a web page containing three columns, you should be able to do it one of three ways:
(1) Define three elastic containers, and glue them to each other, right side of the first to the left of the second, right of the second to left of the first, and specify some basic constraints on each container. Adding a fourth column, you simply change where they're glued together. It would just happen, you wouldn't need to do a google search and look for a four-column model that's close enough to your existing three column code that it'll mostly work.
(2) Start with the full space, and pack containers into it, starting with one side, with constraints specified as they're packed. Pack the first on the left, the second on the left of the remaining space, and the third on the left of the remaining space. Adding a fourth column, you simply pack it in at the right place. It should just happen, you wouldn't need to do a google search and... oh, I mentioned that already.
(3) Specify a grid, with constraints on the grid, and place the containers in the appropriate parts of the grid.
No, instead, in CSS, you have to go back and readjust all the column widths, mentally allowing for the extra space you need for padding. You have a 4 column model, and you add a column, and it wraps, so you go back and look for the place where the column widths are specified, and reduce them all by about 20%, and then it's a little narrow because you forgot to add a spacer, so you do that, and it wraps again, so you reduce the spacer widths a bit, and it's narrow again,...
Here's an example of how onother tool I use does four column layout:
pack [frame.f1] -side left -expand 1 -fill both pack [frame.f2] -side left -expand 1 -fill both pack [frame.f3] -side left -expand 1 -fill both pack [frame.f4] -side left -expand 1 -fill both
pack [label.f1.l -text col1] -side top pack [label.f2.l -text col2] -side top pack [label.f3.l -text col3] -side top pack [label.f4.l -text col4] -side top
Adding another column is obvious, and it just works. Adding more text just works. Padding is an attribute of the layout, not extra invisible objects.
This is already more powerful than CSS, but sometimes it's not enough, so there's a third layout manager, the grid layout manager. Because sometimes you really do need to specify the layout in rows and columns like a table, which is why people still use tables for layout.
The only thing I'd like to add would be the idea of "space", so that you could have a space spread over two or more containers (frames) and have the content flow from one to the next when it fills up.
Think about what you would benefit more from, a 1/2 inch off the sides or 1/4 inch off the top.
A half inch off the sides, for sure. I'm only carrying one laptop, no matter how much other paper I have (and I've had plenty). It's a lot easier to get a carryon that's 1/4 of an inch fatter than a tray table that's 1/2 an inch wider. It's easier to get a briefcase that's 1/4 inch thicker than a desk or bench that's 1/2 inch wider.
Slashdot Mirror
In Illinois employers must give workers time off to vote, so asking for a simple verification that they did so seems fair.
Maybe, but it's not actually necessary, is it? We're not talking about them taking time off any time they want it, after all.
But the main point is that there are people actually suggesting that voters should get a receipt of *how they voted*, to act as a paper trail. I'm not saying that's what happened here, I'm saying that this kind of incident demonstrates that the argument against it isn't just a paper tiger.
Thank you, that does make a lot of sense.
Eliminating the grace period also eliminates front-running. If you read the transcript, NSI had in fact indicated that they would roll back front-running if they lost the grace period.
The more important part, how the brain can sublimate operating complex machinery so that it doesn't require conscious thought to operate, isn't explained here.
Same way the body can sublimate operating the body itself so it doesn't require conscious thought. Solve that problem, and you've solved this one.
I mean, look at how babies learn. Look at how you learn any skill. It all comes down to contracting muscles and getting feedback of a successful result... what this is demonstrating is that whether the levers the muscles are pulling are inside the body or outside is not a major factor.
And I doubt very much whether this capability was something that developed in primates or mammals or birds... any species that learns how to use its own body probably has this ability.
In other words, he is an expert on publicizing what most serious researches already know about general security flaws and problems.
And the problem with this is what? Given how badly people misunderstand computer security we don't have enough people doing this kind of job.
My friend nearly got beat up. Not by the poll workers, but by the voters. Many of them were city workers, who were afraid that if they didn't bring their voting receipt to their boss that they would be fired.
This is why all the schemes involving a "receipt" rather than a ballot as a "paper trail" are not just flawed but actually reduce the validity of a vote. When you leave the polling place you should not be permitted to carry anything that you didn't have when you entered it... even so much as a "how to vote" card.
I forget the brand, but it required a key card, and had a paper tabulation of your votes that was visible behind a clear window for your approval. After approving the paper backup, it scrolled up out of sight.
Two problems:
(1) Since the records are a tape, then by recording the order of voters in the precinct records and the order of records on the tape they can determine who voted which way to a fair degree of certainty.
(2) If the tape was treated as the primary ballot (that is, it was physically transferred to a counting station and read) then this would address many of the problems of an electronic ballot. However not only is it not treated as the primary ballot, but as a result in some cases such tapes were discarded as trash after the elections.
So why would Amazon be patenting a client-side component?
(not to mention that they would be total tards for actually implementing it, given how much pain IE causes by intercepting 404 already)
Text messages are sent in the overhead: even if they didn't implement text messages at all they'd still be building all that infrastructure for voice. In fact that's what they DID build it for. SMS is a side effect, piggybacked on other traffic: it doesn't cut into bandwidth for voice at all.
The marginal cost for text messages is 0.
Turning that around, if they were just setting up a text message service, they would need fewer and simpler cells because they wouldn't be trying to send nearly as much data over the airwaves. Ten seconds of talking requires as much bandwidth as hundreds of messages, and because voice conversation is real-time they have to send it synchronously and even a second's latency is noticable. A text-only network could delay a minute waiting for a slot to open to send a message and most people wouldn't notice most of the time.
Based on TOTAL cost recovery, as you suggest, text messages should at most be something like 1/10,000th the cost per minute of voice. Not 2 cents (that's gotta be a wild-assed guess), or even 0.2 cents: the phone company would still be making a hell of a profit off them at that rate.
I can see why they would want an authorization system, because they have already expressed their worries about iPhone malware.
That's their excuse. But that's not how effective malware typically gets into mobile devices... not that there's much malware for mobile devices out there at all, but what there is tends to be good old backdoors and buffer overflows, not crocked installers, because you don't typically download and install software directly to these devices so there's no way for malware to propagate from one device to the next.
I've been pointing this out since the antivirus companies started really pushing AV for Palm and Pocket PC several years back. There's no viral ecosystem for these devices, because there's no device-to-device transmission path that supports execution of code, with or without social engineering being involved. You don't install software directly on your handhelds (PDAs or phones), you do it on your desktop or laptop and download it to the handheld from there. And the iPhone is no different.
So there's no technical reason for this, and the security argument is devastatingly weak. It's all about control. Malware is just the excuse.
Please. I can't buy Mac hardware without buying MacOS X.
You mean you can't run Mac OS X without buying Apple hardware. While I'm sure there are a few people who buy into the Mac chic and buy Macs to run Windows or Linux on, the vast majority of people are buying Macs to run OS X. The Windows Tax is the extra cost of an OS you don't want to get the hardware you do. The Mac Tax is the extra cost of the hardware you have to buy to run the software you're buying it for.
You're right, they have every right to be creepy, and you and I have every right to bitch about it, and put pressure on them to quit being creepy.
Right, convert it into a set of sculpties in SL instead, and add prim parodies of neighbors that bug you.
That would increase the base cost of Internet access by 25-30%, if prices in Canada are in line with the USA. That seems out of line when compared with other compulsory licenses.
Mod parent -1 "no sense of humor".
Grog watch video. Grog see Tufte try to cram full page of information into index card. Grog think Tufte missing point. Grog think Tufte not realize what good for printed page not good for small interactive gadget.
Better?
First of all, the straight line from upper left to lower right is inherent in the graph. The way it's defined, no points can be above that line.
Second, information storage technologies almost inevitably have a limited lifetime, and the longer the lifetime the more expensive it is. To store information like "when was fire first used as a tool" required hundreds of thousands of individual controlled fires to be kindled for there to be enough remains to survive to mark the time (yes, the people creating the fires were not deliberately creating a record of the invention, but the result and the effort are the same as if they were). For some astronomical events, the cost of recording them easily exceeds the lifetime energy production of any conceivable human civilization. The survival of durable records, once the immediate loss of data that was not initially preserved, follows an exponential decay curve.
So that graph doesn't really say anything about the rate of important events, it says something about the viewpoint of the people making the lists. There may have been billions of events as important as the discovery of fire or slood that we have no idea about, because no record of them exists, and so they won't skew the curve.
Hell, we haven't even rediscovered slood yet ourselves.
You're being ridiculous. If 17 cubic inches (roughly the volume of a wall-wart or a pair of socks) is that critical, there's something wrong with your packing.
Eee software is 99.9% hackable. Replace your OS with another distro, or boot off the SD slot.
The 'distro' of my choice isn't legally available for the Eee PC, nor for anyone else in the target market for the Macbook Air.
Why pay the premium for the Mac except to get the Mac software and usability?
Well, that's kind of the point, isn't it? Why would you bother comparing it with anything that's not running OSX?
I use a toshiba r10 which isn't an ultra portable but I find switching to tablet mode makes it easier to work in a cramped space.
Well now, if the Macbook Air had a tablet mode, we wouldn't be having this conversation. But it's not.
And I do remember the tray tables from the last time I travelled, and the time before that, and missing my Libretto because my Thinkpad was big enough I couldn't fit a glass of juice on th tray as well.
So you agree that because it's space savings eliminates the need for a larger briefcase that it is not a pretender ultra portable?
No, because we were talking about 1/4 of an inch.
There are limits to what you can take as a carry on bag you see.
A quarter of an inch? You're being ridiculous.
i think the point is to think small here... the idea is to allow the smallest set of changes that work to kernels and code that still works and increases real security.
That's how we got in this mess in the first place.
There's a better patch that's even simpler, and that is quit pretending that restricting ports is a useful security feature in the first place.
In FreeBSD, that's done with "sysctl -w net.inet.ip.portrange.reservedhigh=0". I'm sure there's an equivalent in Linux.
That has the advantage that it's OS independent and it doesn't break existing installations that are already using groups for other things. The "reserved port" model made sense when computers that had network interfaces and could run IP were expensive, and access to a LAN required physical access to secure areas, but it's long since become little more than an annoyance.
So your point is that displaying a summary and then letting to user drill down to the information he wants is better than displaying all information into the first page.
No, that's not the point at all. The point is that if you can't fit more information on the first page, as in Tufte's example, you don't cram it in and demand the user interactively (there's that word again) zoom and scroll to see it all, you change the presentation, reduce the amount of information being presented (in the paper, one example was the location of the theatre) or change the way it's presented (showing time graphically rather than a number in a table), and so on. The information you're not including in the summary still exists. In the theatre example, after you've decided what show you're going to see, is that the end of it? No, you're going to buy tickets, or maybe you need to call the theatre to see if they have a changing table in the men's bathroom so you need the number, or you're on the phone talking to someone who doesn't know where the theatre is so you need the street address, but whatever it is, THAT is what you're drilling down for. In Tufte's stock market case, it's the comparison between two stocks.
Of course, now you simply assume that people only use the summary in 99% of all cases, which wasn't clear from your original post.
Yes, I understand that you didn't find that clear, which is why I am trying to clarify it.
But that implies that the summary already contains all the information which the user wants to see
That's the goal, yes.
I really don't know why you're even arguing with me.
Is that what you call it?
Possibly because you didn't watch the video, and jumped on my posting without knowing the context? I don't know, you tell me.
There is no interaction needed in 99% of all cases.
What does that have to do with anything?
If you can show the information the user wants to see, do it. Don't hide it.
Can you go back and show me where I suggested hiding information that the user wants to see?
Seriously?
The POINT of a summary is that it contains the information you need in most cases. If it isn't, then you're not doing a good job of summarizing. Tufte was sticking a bunch of extraneous detail into the page, more detail than you need at a glance, like the animated cloud map and his tens of thousands of numbers in his spark charts. My point is that if that's not information you need all the time (and it's not) it shouldn't be crammed into an index-card sized sheet of paper as if it was a sheet of foolscap, instead of cluttering up the display or making the user drag it around like a magnifying glass over a phone book, you take advantage of the interactive nature of the device. Let the user ask for more detail for the 1% of the time they need it.
Like, you know, the subway map, or trains leaving three hours from now, or bookmarks, or alarms.
But if you want to get into CSS Zen Garden, and all the trickery you need to do to make CSS layout actually work, well... sheesh.
I've used a variety of applications and tools for doing programatic layout of documents over the years, with a variety of levels of functionality, some good, some bad, and CSS is one of the worst. It seems that the people who designed CSS hadn't any actual experience with more than one tool, if that.
In CSS you only have two layout mechanisms: absolute, with absolute scaling, and relative, with absolute scaling. If you want to create a web page containing three columns, you should be able to do it one of three ways:
(1) Define three elastic containers, and glue them to each other, right side of the first to the left of the second, right of the second to left of the first, and specify some basic constraints on each container. Adding a fourth column, you simply change where they're glued together. It would just happen, you wouldn't need to do a google search and look for a four-column model that's close enough to your existing three column code that it'll mostly work.
(2) Start with the full space, and pack containers into it, starting with one side, with constraints specified as they're packed. Pack the first on the left, the second on the left of the remaining space, and the third on the left of the remaining space. Adding a fourth column, you simply pack it in at the right place. It should just happen, you wouldn't need to do a google search and... oh, I mentioned that already.
(3) Specify a grid, with constraints on the grid, and place the containers in the appropriate parts of the grid.
No, instead, in CSS, you have to go back and readjust all the column widths, mentally allowing for the extra space you need for padding. You have a 4 column model, and you add a column, and it wraps, so you go back and look for the place where the column widths are specified, and reduce them all by about 20%, and then it's a little narrow because you forgot to add a spacer, so you do that, and it wraps again, so you reduce the spacer widths a bit, and it's narrow again,
Here's an example of how onother tool I use does four column layout: Adding another column is obvious, and it just works. Adding more text just works. Padding is an attribute of the layout, not extra invisible objects.
This is already more powerful than CSS, but sometimes it's not enough, so there's a third layout manager, the grid layout manager. Because sometimes you really do need to specify the layout in rows and columns like a table, which is why people still use tables for layout.
The only thing I'd like to add would be the idea of "space", so that you could have a space spread over two or more containers (frames) and have the content flow from one to the next when it fills up.
Think about what you would benefit more from, a 1/2 inch off the sides or 1/4 inch off the top.
A half inch off the sides, for sure. I'm only carrying one laptop, no matter how much other paper I have (and I've had plenty). It's a lot easier to get a carryon that's 1/4 of an inch fatter than a tray table that's 1/2 an inch wider. It's easier to get a briefcase that's 1/4 inch thicker than a desk or bench that's 1/2 inch wider.