I don't think mapping group ID directly to port is a good idea. For one thing, it's not generally extensible to other numeric domains. You should map permissions on an object visible in the file system to access to the port. Ideally, this would be handled by replacing crazy overlaid in_addr objects with filenames.
That is, instead of calling socket and bind and so on, you'd call 'sock = open("/net/tcp/listen/25", O_RDWR)'. Normal file system protection on/net/tcp/listen/* would provide access control. Similarly, to make an outgoing connection, you'd call 'sock = open("/net/tcp/connect/10.0.0.2/25", O_RDWR);'. You wouldn't need to modify your code to connect to IPV6 or even OSI TP4/CONS or DEC LAT ports... or UNIX domain sockets... or named pipes...
The whole Berkeley Socket design is second only to System V IPC in terms of missing the whole point of UNIX.
I suspect you misunderstood the article you directed me to.
The widget he built is highly interactive. It only shows arrivals, departures, and trip times, and hides details like the map and other schedules until you request them.
The article is about selecting the information that you provide in the overview so that you don't have to drill down most of the time. Not about cluttering the overview with extraneous details the way Tufte was.
Clutter can be fixed not by throwing away information but by changing the design.
But his video doesn't show that. It shows more clutter.
Look at his proposed weather interface. It's more cluttered than the original.
Look at his proposed stock interface. It's not only cluttered it's actually got LESS useful information than the original. It really doesn't have the thousands of numbers he thinks it does.
There's a classic Dilbert cartoon that shows some sales guy trying to sell Dilbert a non-interactive computer. It's easy to use. "There's only one button, and we push it at the factory." "What does it do?" "Whoa! Let me get an expert!"
If you just want to make phone calls, the iPhone itself is a waste of time. I want a phone that has only a keypad and a small monochrome display, with no games, no internet access, no maps, no weather, no music, no ringtones, just a damned phone. I have a separate PDA that isn't tied in to a phone company, upon which I can install any software I want, for that kind of thing. I can leave my PDA behind when I'm going someplace that's hazardous to digital devices, and still carry my phone. I can leave my phone behind without losing my personal information.
So I will agree, if you don't need a lot of capability, you don't need an interactive device.
But the iPhone isn't for someone who doesn't want a lot of capability. It's for someone who wants more than you can fit on an index card. You have three choices: limit the amount of information, add clutter, or let the user ask for more information. The first and third choices are both valid. What Tufte's arguing for is the second.
your example password lacks mixed-case and punctuation.
It also only has one misspelling and contains real words.:)
It's an example of something I think the average user might be willing to put up with.
The more you can memorize, the better, but if someone's got rainbow tables big enough to cover "squam1sh666oss1frage" then you're better off with "tquamish666ossifrabe", since there's 25 possible substitutions for each symbol, rather than 3 or 4 (eg, "a" to one of [A@4]). Once you go to non-mnemonic symbol substitutions, you might as well pick something like "abgflynt6yta" or "qe5t-agxe-596p-epea-guxv-6gre" or even "550e8400-e29b-41d4-a716-446655440000"... because most people simply aren't going to be arsed trying.
Even getting people to use something like "3 words 2 numbers" (elvish17puce24capital) or even "2 words 1 number" (squeamish46ossifrage) would be infinitely better than "one word and one number if you're lucky" most people use.
Yah, and he's not the only one who's come up with a neat idea that isn't really as widely applicable as he thinks. He's also really not understanding the capabilities of interactive interfaces... rather than throwing all the information on one page, you drill down from the summary into detail.
For example on the stock market page, drag stocks over each other to compare them, dragging a stock all the way to the top of the page would give you more information on that stock and let you drag the screen left or right to get other stocks, flip it sideways to get the graphs, and drag left and right to compare with other graphs.
On the weather page, use the same approach, and flip sideways to get the weather map, drag up and down to see different maps.
A video screen isn't a static device, and you don't need to cram data into a single static view. Data clutter is as bad as administrative clutter.
Which is hardly an advantage on Linux because everybody can su to root.
OK, check ALL your other users.
Limiting it to group wheel is not a particularly big hurdle. There are enough applications setuid to root to find execution exploits in that one more is not much of a barrier.
The common UNIX implementation whereby group 0 or UID 0 membership acts as the gateway to such unobvious capabilities as opening low ports really needs to be readdressed. Traditionally, you would have group access to devices (eg,/dev/tcp/25) acting as the required capability, rather than having to become superuser for such a common operation.
I think Tufte would agree with you. His opening remarks at least (I'm still waiting for the video and illustrations to make it here against the slashdot stream) seem quite favorable to the device.
Why wasnt this patent written so that in X number of years in became extinct
ALL patents and in fact all the temporary monopolies known in layman's terms as "intellectual property" are already term-limited. The problem is that the terms have been progressively extended over the decades since they were introduced so that, for all practical purposes, they might as well be permanent.
That is, "year X" just plain takes too long to get here.
A secondary issue that is closer to what you're thinking of is the terms of the licenses for the patents. These should have been made permanent and irrevocable. Another possibility is that the IEEE could have been granted a permanent and irrevocable right to sublicense the technology under the original terms for the purpose of fulfilling IEEE standards.
This shouldn't be a big deal unless root also has an easily guessed password.
* Have good passwords.
Absolutely. Try "squam1sh666oss1frage" instead of "susan". Check your other users too, particularly people in group wheel. I had an account used as an attack because it was set up with an easily guessed name and password and I was never actually given the password - I always sshed in with a DSA key.
I would add:
* Have a string password validator, but don't force people to cycle passwords... that encourages easily guessed passwords because easily remembered ones are easily guessed. The best password validator is to run the best password cracking tools you can lay your hands on against your own password file.:)
The Toshiba only weighs in at 1.7lbs if, AFAIK, if you go with a solid state drive.
The Macbook Air with the solid state drive is still three pounds.
The macbook air is NOT a pretender. It's got the 1.7 inch drive, it's got the ultra small profile cpu
Those are cool in a geeky sense, but they're not something that directly benefit the user unless they actually make it smaller and lighter. Thinness doesn't really benefit the user either, beyond a certain point... and the plain old Macbook is thin enough that it's really a matter of style to go thinner. The main characteristics of an ultraportable are overall footprint and weight. Remember the old 12" Powerbook ads? They didn't go on about how thin the 12" powerbook was... which is a good thing, because it was actually thicker than the 15". They advertised the footprint.
Jobs said that they made thinness a higher priority than weight, and it's got the same footprint as the plain old Macbook (it's even a smidgen wider).
you get 13.3inch screen rather than 12.1inch
Which makes it an inch and a half wider and half an inch deeper. I don't think any other laptop shooting for the label of ultraportable uses a screen larger than 12.1". The Fujitsu Lifebook is 10.6", the smallest VAIO is 11.1". My Libretto was 7.2".
They'd have done better copying the hackintosh guys who got OS X running on an Eee PC.
I don't put the Macbook Air in the same category as that Toshiba. Portability to me means: how much mass you have to haul to get the functions you need. Compare the weight budget, and you'll see the difference between the Macbook Air and the Toshiba is the difference between a real ultralight notebook and a pretender.
If you get the Toshiba configured similarly to the Macbook Air is 1.7 lbs. That's close to half the weight. Fully loaded you get a computer that's still lighter and still has all the missing features that you can't get integrated into the Macbook Air for any price. That's a real "ultraportable".
The Macbook Air might have been an ultraportable ten years ago... it's comparable in weight to my old Libretto, though it's not as convenient to carry... but the goalposts have moved in the past decade.
I'm talking about "align=". "style=" was the obvious workaround. There are others, like "
". They all end up bloating the size of the document for no good purpose.
But your scheme of creating a separate style for each table cell won't work for the kinds of situations that align= or style= is typically used for.
The problem is that each class is a separate name. This leads to an exponential explosion in the number of classes you need, and it completely breaks free-standing components.
First, if you have a table that has (say) three different cell styles, and two different data types, you now need to define six classes instead of three. Assuming they're going to remove other attributes like align, and you have (say) two other ways tables can differ, in (say) three ways, you need 54 classes instead of three. And you end up with...
Second, when you have software-generated HTML, every place you have to expose details of one components classes to another you get more opportunity for things to break, so it makes sense to have typographic details set separately.
Apple hasn't quite caught up with WinXP, but they're unfortunately getting there.
Windows Media Player 9, which was introduced shortly before XP, included a kernel component that gave it privileged access and control, to support DRM. Later versions of WMP, XP, and now Vista have even stronger restrictions. This kind of chichanery from Microsoft is why my Wintendo is still based on Windows 2000.
I had honestly expected that Apple would implement similar restrictions to Microsoft's when the iTunes Music Store came out... and I've been pleasantly surprised by the lack of these kinds of DRM-friendly features in OS X up to now.
My fallback position is switching back to other operating systems based on Free UNIX. I don't want to do that, because I'd rather have access to applications that don't suck, but I'm not naive enough to assume I won't have to.
All this means is that instead of using "td align=...", you'll get people using "td style='text-align:...;'". Where's the win, other than making documents larger?
If you're worried about an extra pound, the Macbook Air is pretty heavy for a subnotebook. It weighs a pound more than my Toshiba Libretto did *in 1998*, and laptops have gotten better over the past 10 years.
And I used the replacable battery in my Libretto on practically every trip, despite its 5 hour battery life, it was great not having to fight over power strips in airports and seminars. And I'd have been happy to accept an extra pound for a port replicator with a CDROM built in.
I'm sure the Macbook Air is great for a lot of people, but it's got real shortcomings thanks to Apple's style-over-substance design. For a lot of us Mac users, Apple's "cool" design has been nothing but a pain in the cervical vertebrae.
I've never replaced the battery in the laptop that I've had for over 2 years. If I'm going somewhere, I bring the appropriate charger.
I really liked the separate battery pack on my Toshiba Libretto. It meant I could actually take notes on my laptop for a whole day at a class or presentation without having to be one of the geeks sitting at the side of the room fighting over the available power sockets. I didn't even need to take the charger out of my hotel room: with a fully charged battery pack in the Libretto and another in my pocket, I was good for 10 hours.
The plain old Macbook is more portable than the Macbook Air.
It's the same width, and you're not going to carry your Macbook Air in a manilla envelope, or even a badded jiffy bag, you're going to carry it in a bag, briefcase, or backpack with about an inch of padding... just like any other laptop... and with the optical drive shoved down into another pocket (so you don't scratch up the 'book's shell).
Yes, the optical drive.
No road warrior is going to go out to a customer without the ability to read anything they get handed on a disc.
So you end up with it taking just as much real space, in a less convenient form.
No, even I, a notorious defender of the Mac against the people who insist it's just style, have to say that this time it really is just style.
Slashdot Mirror
I don't think mapping group ID directly to port is a good idea. For one thing, it's not generally extensible to other numeric domains. You should map permissions on an object visible in the file system to access to the port. Ideally, this would be handled by replacing crazy overlaid in_addr objects with filenames.
/net/tcp/listen/* would provide access control. Similarly, to make an outgoing connection, you'd call 'sock = open("/net/tcp/connect/10.0.0.2/25", O_RDWR);'. You wouldn't need to modify your code to connect to IPV6 or even OSI TP4/CONS or DEC LAT ports... or UNIX domain sockets... or named pipes...
That is, instead of calling socket and bind and so on, you'd call 'sock = open("/net/tcp/listen/25", O_RDWR)'. Normal file system protection on
The whole Berkeley Socket design is second only to System V IPC in terms of missing the whole point of UNIX.
I suspect you misunderstood the article you directed me to.
The widget he built is highly interactive. It only shows arrivals, departures, and trip times, and hides details like the map and other schedules until you request them.
The article is about selecting the information that you provide in the overview so that you don't have to drill down most of the time. Not about cluttering the overview with extraneous details the way Tufte was.
Clutter can be fixed not by throwing away information but by changing the design.
But his video doesn't show that. It shows more clutter.
Look at his proposed weather interface. It's more cluttered than the original.
Look at his proposed stock interface. It's not only cluttered it's actually got LESS useful information than the original. It really doesn't have the thousands of numbers he thinks it does.
There's a classic Dilbert cartoon that shows some sales guy trying to sell Dilbert a non-interactive computer. It's easy to use. "There's only one button, and we push it at the factory." "What does it do?" "Whoa! Let me get an expert!"
If you just want to make phone calls, the iPhone itself is a waste of time. I want a phone that has only a keypad and a small monochrome display, with no games, no internet access, no maps, no weather, no music, no ringtones, just a damned phone. I have a separate PDA that isn't tied in to a phone company, upon which I can install any software I want, for that kind of thing. I can leave my PDA behind when I'm going someplace that's hazardous to digital devices, and still carry my phone. I can leave my phone behind without losing my personal information.
So I will agree, if you don't need a lot of capability, you don't need an interactive device.
But the iPhone isn't for someone who doesn't want a lot of capability. It's for someone who wants more than you can fit on an index card. You have three choices: limit the amount of information, add clutter, or let the user ask for more information. The first and third choices are both valid. What Tufte's arguing for is the second.
your example password lacks mixed-case and punctuation.
:)
:(
It also only has one misspelling and contains real words.
It's an example of something I think the average user might be willing to put up with.
The more you can memorize, the better, but if someone's got rainbow tables big enough to cover "squam1sh666oss1frage" then you're better off with "tquamish666ossifrabe", since there's 25 possible substitutions for each symbol, rather than 3 or 4 (eg, "a" to one of [A@4]). Once you go to non-mnemonic symbol substitutions, you might as well pick something like "abgflynt6yta" or "qe5t-agxe-596p-epea-guxv-6gre" or even "550e8400-e29b-41d4-a716-446655440000"... because most people simply aren't going to be arsed trying.
Even getting people to use something like "3 words 2 numbers" (elvish17puce24capital) or even "2 words 1 number" (squeamish46ossifrage) would be infinitely better than "one word and one number if you're lucky" most people use.
No smiley. I'm unfortunately serious.
Yah, and he's not the only one who's come up with a neat idea that isn't really as widely applicable as he thinks. He's also really not understanding the capabilities of interactive interfaces... rather than throwing all the information on one page, you drill down from the summary into detail.
For example on the stock market page, drag stocks over each other to compare them, dragging a stock all the way to the top of the page would give you more information on that stock and let you drag the screen left or right to get other stocks, flip it sideways to get the graphs, and drag left and right to compare with other graphs.
On the weather page, use the same approach, and flip sideways to get the weather map, drag up and down to see different maps.
A video screen isn't a static device, and you don't need to cram data into a single static view. Data clutter is as bad as administrative clutter.
I have exactly the same combination on my highly secure luggage. Well, I did, anyway. Thanks a lot, asshole.
y0ure666welc0me
Which is hardly an advantage on Linux because everybody can su to root.
/dev/tcp/25) acting as the required capability, rather than having to become superuser for such a common operation.
OK, check ALL your other users.
Limiting it to group wheel is not a particularly big hurdle. There are enough applications setuid to root to find execution exploits in that one more is not much of a barrier.
The common UNIX implementation whereby group 0 or UID 0 membership acts as the gateway to such unobvious capabilities as opening low ports really needs to be readdressed. Traditionally, you would have group access to devices (eg,
I think Tufte would agree with you. His opening remarks at least (I'm still waiting for the video and illustrations to make it here against the slashdot stream) seem quite favorable to the device.
It's too true to be "+1 funny". :(
Why wasnt this patent written so that in X number of years in became extinct
ALL patents and in fact all the temporary monopolies known in layman's terms as "intellectual property" are already term-limited. The problem is that the terms have been progressively extended over the decades since they were introduced so that, for all practical purposes, they might as well be permanent.
That is, "year X" just plain takes too long to get here.
A secondary issue that is closer to what you're thinking of is the terms of the licenses for the patents. These should have been made permanent and irrevocable. Another possibility is that the IEEE could have been granted a permanent and irrevocable right to sublicense the technology under the original terms for the purpose of fulfilling IEEE standards.
How many different types of point-like particles do we need to explain things?
"Everything should be made as simple as possible, but not simpler." -- Albert Einstein
* Don't allow root to ssh into your machine.
:)
This shouldn't be a big deal unless root also has an easily guessed password.
* Have good passwords.
Absolutely. Try "squam1sh666oss1frage" instead of "susan". Check your other users too, particularly people in group wheel. I had an account used as an attack because it was set up with an easily guessed name and password and I was never actually given the password - I always sshed in with a DSA key.
I would add:
* Have a string password validator, but don't force people to cycle passwords... that encourages easily guessed passwords because easily remembered ones are easily guessed. The best password validator is to run the best password cracking tools you can lay your hands on against your own password file.
The Toshiba only weighs in at 1.7lbs if, AFAIK, if you go with a solid state drive.
The Macbook Air with the solid state drive is still three pounds.
The macbook air is NOT a pretender. It's got the 1.7 inch drive, it's got the ultra small profile cpu
Those are cool in a geeky sense, but they're not something that directly benefit the user unless they actually make it smaller and lighter. Thinness doesn't really benefit the user either, beyond a certain point... and the plain old Macbook is thin enough that it's really a matter of style to go thinner. The main characteristics of an ultraportable are overall footprint and weight. Remember the old 12" Powerbook ads? They didn't go on about how thin the 12" powerbook was... which is a good thing, because it was actually thicker than the 15". They advertised the footprint.
Jobs said that they made thinness a higher priority than weight, and it's got the same footprint as the plain old Macbook (it's even a smidgen wider).
you get 13.3inch screen rather than 12.1inch
Which makes it an inch and a half wider and half an inch deeper. I don't think any other laptop shooting for the label of ultraportable uses a screen larger than 12.1". The Fujitsu Lifebook is 10.6", the smallest VAIO is 11.1". My Libretto was 7.2".
They'd have done better copying the hackintosh guys who got OS X running on an Eee PC.
I don't put the Macbook Air in the same category as that Toshiba. Portability to me means: how much mass you have to haul to get the functions you need. Compare the weight budget, and you'll see the difference between the Macbook Air and the Toshiba is the difference between a real ultralight notebook and a pretender.
If you get the Toshiba configured similarly to the Macbook Air is 1.7 lbs. That's close to half the weight. Fully loaded you get a computer that's still lighter and still has all the missing features that you can't get integrated into the Macbook Air for any price. That's a real "ultraportable".
The Macbook Air might have been an ultraportable ten years ago... it's comparable in weight to my old Libretto, though it's not as convenient to carry... but the goalposts have moved in the past decade.
The size bloatedness of tables compared to what?
You think they should make tables with <pre> and little ascii crosses and dashes?
I didn't say *anything* about layout. Tables are also used for, you know, tables...
[speechless]
What's next, no more <div> and <span>?
If your using the style attribute for everything
...
Who said that?
I'm talking about "align=". "style=" was the obvious workaround. There are others, like " ". They all end up bloating the size of the document for no good purpose.
But your scheme of creating a separate style for each table cell won't work for the kinds of situations that align= or style= is typically used for.
The problem is that each class is a separate name. This leads to an exponential explosion in the number of classes you need, and it completely breaks free-standing components.
First, if you have a table that has (say) three different cell styles, and two different data types, you now need to define six classes instead of three. Assuming they're going to remove other attributes like align, and you have (say) two other ways tables can differ, in (say) three ways, you need 54 classes instead of three. And you end up with
Second, when you have software-generated HTML, every place you have to expose details of one components classes to another you get more opportunity for things to break, so it makes sense to have typographic details set separately.
Apple hasn't quite caught up with WinXP, but they're unfortunately getting there.
Windows Media Player 9, which was introduced shortly before XP, included a kernel component that gave it privileged access and control, to support DRM. Later versions of WMP, XP, and now Vista have even stronger restrictions. This kind of chichanery from Microsoft is why my Wintendo is still based on Windows 2000.
I had honestly expected that Apple would implement similar restrictions to Microsoft's when the iTunes Music Store came out... and I've been pleasantly surprised by the lack of these kinds of DRM-friendly features in OS X up to now.
My fallback position is switching back to other operating systems based on Free UNIX. I don't want to do that, because I'd rather have access to applications that don't suck, but I'm not naive enough to assume I won't have to.
All this means is that instead of using "td align=...", you'll get people using "td style='text-align:...;'". Where's the win, other than making documents larger?
If you're worried about an extra pound, the Macbook Air is pretty heavy for a subnotebook. It weighs a pound more than my Toshiba Libretto did *in 1998*, and laptops have gotten better over the past 10 years.
And I used the replacable battery in my Libretto on practically every trip, despite its 5 hour battery life, it was great not having to fight over power strips in airports and seminars. And I'd have been happy to accept an extra pound for a port replicator with a CDROM built in.
I'm sure the Macbook Air is great for a lot of people, but it's got real shortcomings thanks to Apple's style-over-substance design. For a lot of us Mac users, Apple's "cool" design has been nothing but a pain in the cervical vertebrae.
I haven't used a G3 Powerbook, but I am completely in agreement with you about the Thinkpad. I miss the T23 I had as a company laptop a few years ago.
Apple made a laptop in partnership with IBM once, I have a forlorn hope that they might come to some common understanding with Lenovo, some time...
If a student can justify the extra $700 over the plain old Macbook, they're a lot better funded than I ever was at college.
I've never replaced the battery in the laptop that I've had for over 2 years. If I'm going somewhere, I bring the appropriate charger.
I really liked the separate battery pack on my Toshiba Libretto. It meant I could actually take notes on my laptop for a whole day at a class or presentation without having to be one of the geeks sitting at the side of the room fighting over the available power sockets. I didn't even need to take the charger out of my hotel room: with a fully charged battery pack in the Libretto and another in my pocket, I was good for 10 hours.
Two years? You'll learn.
The plain old Macbook is more portable than the Macbook Air.
It's the same width, and you're not going to carry your Macbook Air in a manilla envelope, or even a badded jiffy bag, you're going to carry it in a bag, briefcase, or backpack with about an inch of padding... just like any other laptop... and with the optical drive shoved down into another pocket (so you don't scratch up the 'book's shell).
Yes, the optical drive.
No road warrior is going to go out to a customer without the ability to read anything they get handed on a disc.
So you end up with it taking just as much real space, in a less convenient form.
No, even I, a notorious defender of the Mac against the people who insist it's just style, have to say that this time it really is just style.