Timmy cheats at Tic-Tac-Toe. Timmy throws like a girl. Timmy hates little kids, dogs and old people. Timmy cries during Lassie. Timmy eats yellow snow. Timmy cries over spilled milk. Timmy prefers the metric system. Timmy breastfed until he was 13. Timmy likes interpretive dance. Timmy rarely bathes. Timmy is a vegetarian. Timmy is worried about the "military-industrial" complex. Timmy can't wait for the dawning of the age of Aquarius. Timmy thinks hard work is for suckers. Timmy can't tell time. Timmy thinks Errol Flynn is dreamy. Timmy buys things not made in the USA. Timmy pees sitting down. Timmy breathes through his mouth. Timmy wants to live in the south of France. Timmy has never heard of Google. Timmy calls hot dogs frankfurters. Timmy eats tofu. Timmy needs to diet. Timmy practices making out with his hand. Timmy has a pet turtle named Shelly. Timmy wets the bed. Timmy is too scared to watch the Wizard Of Oz. Timmy talks to his mom on the phone 3 hou... Timmy watches the radio and listens to TV. Timmy never brushes his teeth. Timmy uses medicated deodorant. Timmy fears commitment. Timmy gives himself dutch ovens. Timmy drinks paint thinner. Timmy goes to be at 7:30. Timmy stole the Lindbergh baby. Timmy is a COMMUNIST!
The phrase "long tail" has a couple of related, but different meanings - both from the shape of a graph of sales (or popularity) on the Y-axis.
In this case, the X-axis is a listing of different titles, sorted by popularity. There are a small number of extremely popular items, tailing off to huge numbers of less popular items.
In the other use, the X-axis is time. Something can be hugely popular for a few days/weeks/months, but will continue to have greatly diminished sales for many years to come.
The concept of the "long tail" is that the combined popularity or profitability of the items in the tail can be a significant part of the overall profitability of the entire market, particularly when internet sized economies of scale make it efficient to stock and sell these items.
Certainly DNS keeps you from needing to use IP addresses for most day to day usage of any network.
The issue is the lower level configuration that occurs before DNS will work - configuring DHCP servers, routers, etc. All the times that I would need to directly key in an IPv4 address now. It doesn't seem that v6 would require entering any fewer address than v4.
And I can't remember individual v4 addresses, either. I only need to remember one address for the network at hand, and then I can derive the address of machine 1, machine 2, etc. But I don't think it's practical to remember even one v6 network address. But then, I find it hard to dial 10 digit phone numbers without making a mistake, so maybe I'm not the best example.
I'm not really asking if it's impossible to do this, clearly it can be done, but the question is more about whether the inconvenience is a factor in slow adoption.
I submitted this as an Ask Slashdot some time ago and it was rejected, but I'd really like opinions on it:
How much of a problem/obstacle to adoption is the need for humans to deal with a 128 bit address?
I can deal with xxx.xxx.xxx.xxx in my head where most of the x's are the same all the time, but yyyy:yyyy:yyyy:yyyy:yyyy:yyyy:yyyy:yyyy is simply too much.
Is it such a pain to deal with such long addresses that admins who would be configuring v6 "just because" don't? Those of you who have v6 networks, are there automated tools that keep you from ever having to key in an address, do you have the address range printed on your t-shirt, or what?
Would it have been better to use a smaller (40? 48?) bit range, and perhaps supplement that with an "extension" mechanism that could be appropriately sized for the network involved?
Neither the antenna software nor the CAD program is something I care enough about to have a windows box for. If I didn't have my work laptop, I wouldn't use either one.
And as far as being a pirate, I paid for the antenna software, and the CAD program was a high-end piece of professional software that cost well in excess of $2000 something like 14 years ago when I was doing drafting part-time during college. (And yes, the version I have is that old.)
There is no lost sale involved, except maybe some as yet to be determined $100 Mac CAD package, but probably not even that. If I had been able to find a replacement program that represented an value to me, I would have sent in my CC number. And I've looked.
I'll concede, you may have a point about the BSA - but then, isn't that a pretty good reason in itself to ditch windows?
Yes, that was a snarky comment, but here are the facts:
I used to be a windows programmer. I got so sick and tired of spending more time dealing with figuring out why windows isn't working correctly than the fun part of programming, which is getting the machine to sing and dance at your command, that I worked my way into a non-programming position at my company, and have moved farther, and farther from software at every chance I've gotten.
I bought a Mac for home use, and I don't have any working windows machines here. I keep a couple of boxes around with Linux on them to play with once in a while, but it's just much less effort to use the Mac.
I have a laptop at work that runs XP. I use it there because I have to. I have a couple of personal programs that I run on it - a very, very old piece of pirated CAD software, and some antenna modeling software. These two programs are the *ONLY* thing I ever use personally that would tie me to windows. I would replace the CAD program with a non-pirated Mac program, but I worked with it for so many years at my old job that I just don't want to learn something completely different. The antenna software is PC only, but I might try to get it working under WINE someday.
So: I don't *NEED* windows. I don't *LIKE* windows. I don't want to *ACTIVATE* a bunch of DRM riddled garbage. I don't think the ms has done anything worthy of my money. I'm not going to give them any more chances to get their products right, technically or ethically.
For many, many years, utilities were generally run by municipalities (some very well, others, not so much), or were very heavily regulated private monopolies.
I suppose my opinion isn't very popular anymore, but I always thought that regulated monopolies normally worked pretty well.
Competition in utilities is generally a farce anyway. It almost always means competition among a bunch of resellers and middle-men who don't really add much value, and boils down to passing money along to the people who actually own the pipes and wires, who are generally still regulated monopolies, but without enough protection to reasonably finance proper maintenance and growth capacity. Everybody lines up with their checkbooks to get in on "The Next Big Thing", but once it's more or less matured, the focus switches to this quarter.
We see how well power deregulation worked in California, and the news has been talking about how much electric rates have gone up in Virginia since deregulation took effect. Have you tried to get a problem fixed on a T1 circuit lately? The smartjack loops, it's your wiring. No, then maybe the local carrier's loop is having a problem. Uhh, maybe it's an issue with our upstream provider...
People like to talk about all the innovation since the Bell System break up. What did we get? Cell phones? Bell invented them. Cheap long distance? Long distance is now technically trivial, but I'm still paying for it. It is technically trivial because of all the hard work Bell Labs put into cutting its cost. The Internet? Data circuits were out there for a long time, and the ARPA net was going along pretty well for its intended purpose until the www "Killer App" came along, at which point, ISDN would have been great - except it died because of interoperability problems and clueless pricing. (See Scott Adams for commentary on ISDN.) A regulated monopoly may still have had clueless pricing, but at least the interop problems would be non-existent.
Somewhat off topic, but what I'd really like to see is Apple implement their Cocoa API and development tools on Windows and Linux, or at least provide cross-platform development tools.
This could be a big boon to Apple converts who would like to develop software on their favorite platform but have a market ($ or FAIB) of a reasonable size.
The benefit to Apple (and everyone else) would be enormous, as it could make the Mac the preferred first target, greatly increasing the software available for it, and allowing the various OS's to compete on their merits rather than the size of their existing installed base.
(I know there are some workable cross-platform tools around, but most are either interpreted languages that are not suitable for a lot of purposes, or they produce applications that always look decidedly out of place on all but their primary OS. I guess what I really want is for Apple to take a stab at doing it better than anyone else has so far.)
A meeting of the minds between Google, Yahoo, and Microsoft resulting in an agreement to not index these idiotic domains could kill this quick before it gets out of hand.
*Will it happen - doubtful.
Can you or I do anything about it - probably not.
But I can dream.
They also question how NATCA gets away with constantly stating that the US air traffic control system is the safest in the world, when it is actually behind Europe when it comes to safety metrics (Europe has.032 hull losses per 1 million departures vs..049 in North America).
They get away with it because it's a tradition that practically no-one questions. All you have to do is say "The US is the bestest in the world when it comes to [x]" and few people bother checking. (Except "America-haters", of course.)
Perhaps because the ratio of hull losses related to ATC is better in the US.
I don't know that, just sayin'...
Weird - Seems pretty silly to me, even a little disturbing.
But to be accurate, I followed the first link and read 4 of the 13 laws listed there. In each case, it was necessary for the statement made to be *FALSE* to trigger the law, so these are indeed actually "Food Libel" laws.
The first thing that popped into my head when I saw it was to submit a very long request of some kind.
Overflowing an index with simply a lot of characters might be tough, but if you had knowledge (or at least a concept) of the inner workings of the web app, it might be possible to take advantage of a multiplying effect, where each item of a request could result in some index being incremented not by one, but by 1000 or something.
As an example, maybe a search application would create several entries in a table for each search term (an entry per matching item, perhaps). It might work ok for 10 terms, but 10 million would make it blow up.
Of course, this scenario would indicate a bug in the app, but using a language like Ruby should prevent that kind of error from allowing arbitrary machine code to execute.
This is all sort of vague and hand wavy, but that is because it is just to outline a concept and would vary from case to case.
Slashdot Mirror
Timmy cheats at Tic-Tac-Toe.
Timmy throws like a girl.
Timmy hates little kids, dogs and old people.
Timmy cries during Lassie.
Timmy eats yellow snow.
Timmy cries over spilled milk.
Timmy prefers the metric system.
Timmy breastfed until he was 13.
Timmy likes interpretive dance.
Timmy rarely bathes.
Timmy is a vegetarian.
Timmy is worried about the "military-industrial" complex.
Timmy can't wait for the dawning of the age of Aquarius.
Timmy thinks hard work is for suckers.
Timmy can't tell time.
Timmy thinks Errol Flynn is dreamy.
Timmy buys things not made in the USA.
Timmy pees sitting down.
Timmy breathes through his mouth.
Timmy wants to live in the south of France.
Timmy has never heard of Google.
Timmy calls hot dogs frankfurters.
Timmy eats tofu.
Timmy needs to diet.
Timmy practices making out with his hand.
Timmy has a pet turtle named Shelly.
Timmy wets the bed.
Timmy is too scared to watch the Wizard Of Oz.
Timmy talks to his mom on the phone 3 hou...
Timmy watches the radio and listens to TV.
Timmy never brushes his teeth.
Timmy uses medicated deodorant.
Timmy fears commitment.
Timmy gives himself dutch ovens.
Timmy drinks paint thinner.
Timmy goes to be at 7:30.
Timmy stole the Lindbergh baby.
Timmy is a COMMUNIST!
The phrase "long tail" has a couple of related, but different meanings - both from the shape of a graph of sales (or popularity) on the Y-axis.
In this case, the X-axis is a listing of different titles, sorted by popularity. There are a small number of extremely popular items, tailing off to huge numbers of less popular items.
In the other use, the X-axis is time. Something can be hugely popular for a few days/weeks/months, but will continue to have greatly diminished sales for many years to come.
The concept of the "long tail" is that the combined popularity or profitability of the items in the tail can be a significant part of the overall profitability of the entire market, particularly when internet sized economies of scale make it efficient to stock and sell these items.
That seems fairly reasonable.
Ok, now I'll change my question - are people like me running around with horrible misconceptions causing slow adoption of v6?
Certainly DNS keeps you from needing to use IP addresses for most day to day usage of any network.
The issue is the lower level configuration that occurs before DNS will work - configuring DHCP servers, routers, etc. All the times that I would need to directly key in an IPv4 address now. It doesn't seem that v6 would require entering any fewer address than v4.
And I can't remember individual v4 addresses, either. I only need to remember one address for the network at hand, and then I can derive the address of machine 1, machine 2, etc. But I don't think it's practical to remember even one v6 network address. But then, I find it hard to dial 10 digit phone numbers without making a mistake, so maybe I'm not the best example.
I'm not really asking if it's impossible to do this, clearly it can be done, but the question is more about whether the inconvenience is a factor in slow adoption.
I submitted this as an Ask Slashdot some time ago and it was rejected, but I'd really like opinions on it:
How much of a problem/obstacle to adoption is the need for humans to deal with a 128 bit address?
I can deal with xxx.xxx.xxx.xxx in my head where most of the x's are the same all the time, but yyyy:yyyy:yyyy:yyyy:yyyy:yyyy:yyyy:yyyy is simply too much.
Is it such a pain to deal with such long addresses that admins who would be configuring v6 "just because" don't? Those of you who have v6 networks, are there automated tools that keep you from ever having to key in an address, do you have the address range printed on your t-shirt, or what?
Would it have been better to use a smaller (40? 48?) bit range, and perhaps supplement that with an "extension" mechanism that could be appropriately sized for the network involved?
First he says don't be a dinosaur, then he starts talking about tapping Ethernet cables.
The last time I tapped an Ethernet cable, my buddy was throwing 9-track tapes at the dinosaurs to keep them away!
I thought it was just me that hated the new system.
I've not seen anyone comment on it.
No, I don't *NEED* windows.
Neither the antenna software nor the CAD program is something I care enough about to have a windows box for. If I didn't have my work laptop, I wouldn't use either one.
And as far as being a pirate, I paid for the antenna software, and the CAD program was a high-end piece of professional software that cost well in excess of $2000 something like 14 years ago when I was doing drafting part-time during college. (And yes, the version I have is that old.)
There is no lost sale involved, except maybe some as yet to be determined $100 Mac CAD package, but probably not even that. If I had been able to find a replacement program that represented an value to me, I would have sent in my CC number. And I've looked.
I'll concede, you may have a point about the BSA - but then, isn't that a pretty good reason in itself to ditch windows?
A. ELNEC - which I believe is written in VB6, I think the author mentions it on his website.
B. Never heard of it. I'll see what google turns up.
Dear Mr. Flamebait Modder,
Yes, that was a snarky comment, but here are the facts:
I used to be a windows programmer. I got so sick and tired of spending more time dealing with figuring out why windows isn't working correctly than the fun part of programming, which is getting the machine to sing and dance at your command, that I worked my way into a non-programming position at my company, and have moved farther, and farther from software at every chance I've gotten.
I bought a Mac for home use, and I don't have any working windows machines here. I keep a couple of boxes around with Linux on them to play with once in a while, but it's just much less effort to use the Mac.
I have a laptop at work that runs XP. I use it there because I have to. I have a couple of personal programs that I run on it - a very, very old piece of pirated CAD software, and some antenna modeling software. These two programs are the *ONLY* thing I ever use personally that would tie me to windows. I would replace the CAD program with a non-pirated Mac program, but I worked with it for so many years at my old job that I just don't want to learn something completely different. The antenna software is PC only, but I might try to get it working under WINE someday.
So:
I don't *NEED* windows.
I don't *LIKE* windows.
I don't want to *ACTIVATE* a bunch of DRM riddled garbage.
I don't think the ms has done anything worthy of my money.
I'm not going to give them any more chances to get their products right, technically or ethically.
QED
but I still wouldn't buy it.
Why would he need that?
He could have written it himself.
For many, many years, utilities were generally run by municipalities (some very well, others, not so much), or were very heavily regulated private monopolies.
I suppose my opinion isn't very popular anymore, but I always thought that regulated monopolies normally worked pretty well.
Competition in utilities is generally a farce anyway. It almost always means competition among a bunch of resellers and middle-men who don't really add much value, and boils down to passing money along to the people who actually own the pipes and wires, who are generally still regulated monopolies, but without enough protection to reasonably finance proper maintenance and growth capacity. Everybody lines up with their checkbooks to get in on "The Next Big Thing", but once it's more or less matured, the focus switches to this quarter.
We see how well power deregulation worked in California, and the news has been talking about how much electric rates have gone up in Virginia since deregulation took effect. Have you tried to get a problem fixed on a T1 circuit lately? The smartjack loops, it's your wiring. No, then maybe the local carrier's loop is having a problem. Uhh, maybe it's an issue with our upstream provider...
People like to talk about all the innovation since the Bell System break up. What did we get? Cell phones? Bell invented them. Cheap long distance? Long distance is now technically trivial, but I'm still paying for it. It is technically trivial because of all the hard work Bell Labs put into cutting its cost. The Internet? Data circuits were out there for a long time, and the ARPA net was going along pretty well for its intended purpose until the www "Killer App" came along, at which point, ISDN would have been great - except it died because of interoperability problems and clueless pricing. (See Scott Adams for commentary on ISDN.) A regulated monopoly may still have had clueless pricing, but at least the interop problems would be non-existent.
Enough ranting. Who's with me?
Somewhat off topic, but what I'd really like to see is Apple implement their Cocoa API and development tools on Windows and Linux, or at least provide cross-platform development tools.
This could be a big boon to Apple converts who would like to develop software on their favorite platform but have a market ($ or FAIB) of a reasonable size.
The benefit to Apple (and everyone else) would be enormous, as it could make the Mac the preferred first target, greatly increasing the software available for it, and allowing the various OS's to compete on their merits rather than the size of their existing installed base.
(I know there are some workable cross-platform tools around, but most are either interpreted languages that are not suitable for a lot of purposes, or they produce applications that always look decidedly out of place on all but their primary OS. I guess what I really want is for Apple to take a stab at doing it better than anyone else has so far.)
A meeting of the minds between Google, Yahoo, and Microsoft resulting in an agreement to not index these idiotic domains could kill this quick before it gets out of hand.
*Will it happen - doubtful.
Can you or I do anything about it - probably not.
But I can dream.
They also question how NATCA gets away with constantly stating that the US air traffic control system is the safest in the world, when it is actually behind Europe when it comes to safety metrics (Europe has .032 hull losses per 1 million departures vs. .049 in North America).
They get away with it because it's a tradition that practically no-one questions. All you have to do is say "The US is the bestest in the world when it comes to [x]" and few people bother checking. (Except "America-haters", of course.)
Perhaps because the ratio of hull losses related to ATC is better in the US.
I don't know that, just sayin'...
Weird - Seems pretty silly to me, even a little disturbing.
But to be accurate, I followed the first link and read 4 of the 13 laws listed there. In each case, it was necessary for the statement made to be *FALSE* to trigger the law, so these are indeed actually "Food Libel" laws.
Bit of an over-reaction to Oprah, I think.
The first thing that popped into my head when I saw it was to submit a very long request of some kind.
Overflowing an index with simply a lot of characters might be tough, but if you had knowledge (or at least a concept) of the inner workings of the web app, it might be possible to take advantage of a multiplying effect, where each item of a request could result in some index being incremented not by one, but by 1000 or something.
As an example, maybe a search application would create several entries in a table for each search term (an entry per matching item, perhaps). It might work ok for 10 terms, but 10 million would make it blow up.
Of course, this scenario would indicate a bug in the app, but using a language like Ruby should prevent that kind of error from allowing arbitrary machine code to execute.
This is all sort of vague and hand wavy, but that is because it is just to outline a concept and would vary from case to case.