Jesus, chief, get up on the wrong side of the bed? I read "analyst" as someone who tracks the industry, analyzes trends, and writes reports. You seem to have more of an issue with your manager.
I disagree. If I have a good idea, for which I have received a patent, why should I be prevented from simply selling the patent to a corporation that is prepared to enforce it? This gives me the money up front if I need it now instead of later.
In some ways, this is like a discussion of mining rights. A lot of effort is put in up front to figure out where there is a high probability of finding a valuable mineral, but until you sink a mine there's no proof anything is there. Some of the comments here suggest that in this case, you shouldn't be able to claim any value until you build a working mine. I disagree. There is value in doing intellectual leg work up front, and a person should be compensated for doing that work.
Instant runoff voting would be a huge step forward. I'd love to see that.
The other thing I'd like to see, is the ability to vote for different people to represent me on different subjects since, for example, I tend to side with Republicans on gun control issues but with Democrats on abortion and foreign policy. Imagine being able to vote for person A to represent you on education, person B on the military, person C on foreign affairs, person D on budgets,...
To be clear, I'm talking about a theoretical goal here, Since I know of no way to make it actually work in the real world. There are obvious problems with this, such as the fact that whoever controls the money controls everything.
Yes, generally companies will require some sort of ID and credit check to purchase a phone line, but they don't require that the person who pays for the line be the person who is going to use it. Lot's of people pay for a phone line for their aging parent who has gotten forgetful about paying bills ontime, or for their child in school who cannot afford the line.
I am simply paying for a phone line for my good friend Vladimir Macallan, and so it is his name that should be published in the phone book. Make up a suitably unrealistic name by mixing cross cultural first and last names, and you're pretty much guaranteed you won't accidentally get calls for people who really are looking for someone who happens to have that name.
Under NAFTA, don't the US and Canada have the same policy towards each other?
I'm a Canadian who came to the US under the Canada-US free trade agreement (the precursor to NAFTA) and all I had to do was show that I was a Canadian citizan, that I had a degree in one of the listed areas, and that I had a job offer from a US company. I paid a $50 processing fee at the border, and was given a one year work visa renewable indefinitely.
Admittedly, I came to the US 14 years ago, but I was under the impression things were pretty much the same under NAFTA.
Is it more difficult than that for US citizens to work in Canada?
Many pages are cluttered with navigational junk and ads that detract from the interesting content. Take a look at www.cnn.com, for example. The story text is in the middle, and that's what I'm interested in, but all the buttons, ads, additional information, etc. takes up a lot of space.
Automatically identifying the main content of a page, and fading everything else out a bit would be very helpful.
Some sites take an article and break it up into several pages. It would be useful to automatically recognize that, fetch the continuation pages for the article, and pull the relevent content back into the original page.
I don't think anyone is disupting that people need down time. A big value of having WiFi on the train is for people who choose to live a long distance from work in order to have a larger home (or in some cases to be afle to afford any home) to raise their families in. By being able to reclaim the commute time for useful work, it's possible to spend more down time with the family rather than talking to a random stranger on the train for an hour or two in each direction.
That makes a lot of sense to me. I live and work in the San Francisco Bay Area, and I'd love to be able to live someplace less densely populated but still work here.
I'm hoping we're going in this direction, though. There has been talk recently of passing a law requiring cable and satelite companies to offer individual channels a la carte. I'd love to see this, because I never tune to most of the channels I'm forced to buy from DirecTV in order to get the few I am interested in.
Similarly, as you and others have commented, it would be great to get individual shows without being required to buy the entire channel/network. I don't think it's as bad as you say, however. It would just bring things more in line with the movie industry, where you choose individual movies to watch but aren't forced to buy a subscription to all of Disney's movies just because you want to watch one of them.
I dream of the day I can buy a subscription to a few networks that consistently produce good content (HBO, Discovery, History), grab just a few shows I'm interested in from some other networks (FX, NBC, ABC), and ignore the rest. And all of this would be delivered through my existing internet connection. The money I save due to not having to buy content I'm not interested in can be put towards a faster pipe.
Still, this model has the downside of further contributing to the "winner take all" structure that is becoming more and more prevalent in entertainment and business these days.
I'm also a Treo 600 user, and I really like it. I'm using AT&T, and although their customer service has pretty consistantly underwhelmed me, their coverage seems to be decent (San Francisco Bay Area). I've been using AT&T for many years, and switching from TDMA to GSM when I bought the Treo definitely reduced my coverage. Still, between the Cingular merger and migrating the TDMA cells to GSM, the service is getting better and should continue to improve.
Okay, back to the phone. The big selling point for me is merging the phone and organizer into a single unit. I had a Palm V before, but had stopped using it because I hated dragging it along with me everywhere. I'd invariably be out somewhere and friends would want to schedule another get together, but I'd left the Palm at home.
The todo list is really useful as well. For work, it's the list of all the things I need to get done, and whenever I think up something new, it goes into the phone so I won't forget. This is a huge help when you need to prepare for your yearly review or update your resume, because all those todo items remind you of all the things you forgot you'd done over the year. Similarly, my personal todo list keeps track of all the things I need to do around the house once the weekend comes.
Something else it's great for is attaching notes to people's address book entries. There are people I see only once every year or so, and putting directions into the phone is really handy since I can never remember them a year later. Once they're in the organizer, you'll never forget.
I'm still undecided on the value of the wireless access. Yes, there are lots of useful things you can do with it, but is it really worth $30/month for 10M? I use it for stuff like checking the train schedule, retrieving recipes so I know what to buy at the grocery store, googling phone numbers, checking weather forecasts, but there are less expensive ways of doing all those things.
Anyway, I see that I have a todo item that says "Stop reading slashdot, and get back to work" so I better take care of that.
Re:Contents of the Letter (playfair.txt)
on
Update on Playfair
·
· Score: 1
[This is a bit of a long winded followup, that's more than just a response to your article. Bear with me and hopefully I'm not ranting too much.:-)]
So are you saying you think a business should not be allowed to negotiate a contract that restricts what you do with information or content they sell to you?
That seems to be all the Apple is doing here. They're offering to sell you access to content, with the restriction that it's locked into Apple's hardware. The reason they're willing to sell you content at the price they're selling it at, is that it drives more iPod sales. You're buying the content, and in doing so agreeing to the restrictions they've requested, but then breaking your part of the contract.
Don't get me wrong; I'd want to do the same as you. But you're changing the business model unilaterally, and if the bulk of consumers were to do what you want to do, Apple's response might need to be to raise music prices.
The general feeling here on Slashdot seems align into three groups.
Software, music, etc. should always be given away for free, since it costs nothing to duplicate.
It's okay to sell software, music, etc., but once I've paid for it, I get to do anything I want with it. The producer of that content must not be allowed to restrict my use in any way.
It's okay to sell software, music, etc., and to place controls on how it is used after sale.
The first group is whacked, in my opinion, and don't understand the realities of how things get produced. As someone who makes a living from writing software, I believe what I do has value and I should have the option of making the decision whether what I produce is given away for free, or sold as a product.
Personally, I fall somewhere in the second and third groups.
When I buy a CD, I expect to be able to rip MP3s and put them on a variety of machines from which I will listen to them. That means my home server, my portable player, my car stereo, and a directory on my work machine at the office. I don't download music, because it's already compressed at a lower quality setting than I desire. I buy CDs because I can rip them, and as compression technologies advance I can rip them again later to get a higher quality, smaller file.
I choose not to buy copy protected CDs, or DRM protected downloads, because I don't agree to the restrictions they include. But that's just my choice, and I support other people's right to sell in this way if they choose. If there is enough of a market to support that business model, then so be it.
And yes, I expect the government to help in enforcing rights, through the court system and the police if necessary. That's part of why we have a government.
DIVX was an attempt at an alternative licensing scheme, and it became clear quickly that there was not much of a market for DVDs that could only be watched a few times. I hated it, and I'm glad it died, but there was nothing wrong with it having been tried.
The video/DVD rental market is also a good example of an alternative licensing. Instead of paying $20 to own a copy of a DVD that you can watch as many times as you like, you're paying $3 to have the right to watch it for only a few days. You don't have the right to make a copy and add it to your personal library. If you want that, buy the $20 copy. If everyone were to buy the $3 rental and burn a copy, pretty soon you'd only have $20 rentals and we'd all lose out.
Another example would be if a CD was sold at $5, with the restriction that it could only be listened to for a year instead of forever. I believe a company should be able to do that, and some customers might choose that licensing for pulp pop music that they know won't stand the test of time. Personally, I wouldn't buy it (because it's pulp pop music, and because the DRM would probably get in the way of having the music available in the ways/places I want it), but other people should have the option if it works for them.
Why would I ever buy a 100 GB hard drive if it was going to fail before I could fully use that capacity?
I definitely agree that the current 1 yr warranties are ridiculous.
Why, when hard drive speed is the single largest factor affecting perceived system performance, do manufacturers insist on improving storage capacity at the expense of speed and reliability?
It depends what you're using the disk for. If you're doing video editting, 4200rpm IDE drives are fine, so why pay a premium for the faster IDE or SCSI drives? Ditto if you're using the drive for an mp3 collection or as a backup medium. In all these cases (and others), faster drives aren't going to make a difference, so cheaper/slower drives are a better choice.
Portability is nice, but you can get that conveniently with USB. An external USB disk enclosure costs about $30, and you can put whatever IDE hard drive you want into it.
If you want something small, notebook hard drives are around $100/40G plus the $30 enclosure. That's more expensive than this Jaz drive, but if you go with standard desktop drives the $/G is much better.
And of course there's also the fact that most of the slashdot crowd probably have a few extra drives laying around on a shelf somewhere, such that the only real cost is the $30 enclosure.:-)
How would you encrypt your emails with a service like this? If google can't decrypt it, then their web interface is going to only present you with a stream of encrypted data, which you'll then have to copy/paste into another window to decrypt it. This seems like way too much of a hassle to me.
I suppose you could create a front end proxy to do the encryption automatically for you, but at that point you've spent so much time and effort on the problem that you might as well just run your own imap or web mail server.
Why on earth would anyone lock in a single service provider for 100 years? Would you sign a 100 year cell phone plan? I sure wouldn't, because it removes any incentive for the company to do a good job.
It also removes any opportunity for you to vote with your dollars if you don't like the behavior of the company. I used to have my domain registered with Network Solutions, but when their parent company, Verisign, started hijacking NXDOMAIN responses and sending people to their advertising instead, I moved my domain to GoDaddy as a protest.
I don't want my dollars going to a company that seems to want to use their ICANN granted monopoly over the root database to "innovate" themselves as much profit as the can, at the expense of everyone else on the internet.
Besides, the discounted 100 year price from Network Solutions is the same as GoDaddy's everyday price, so it's not even a bargain. I've also found GoDaddy's support to be much easier to get to, and much more responsive and knowledgable than Network Solutions. I'm definitely very happy I made the move.
I think it's important to do both, actually. I used to just talk with the candidate at the high level about design and OO issues, but then I had a guy slip through who'd completely snowed me.
Since then, I still hit on high level understanding, but I also insist on covering some of the nuts and bolts. I have a list of language features, and I make the candidate describe each feature, and where they've used it in one of their own designs.
I also make them walk through some example code, and explain what's going on (e.g. C++ constructor/destructor ordering, virtual function behavior, exceptions, memory allocation leaks, etc.). This isn't "reverse the linked list" sort of stuff, it's just demonstrating a knowledge of the capabilities of the language, and ensuring they can identify some rookie mistakes in the code.
What do you need in the next few years? Put that inside a conduit. Make sure the conduit is large enough that it'll be easy to add and/or replace cables down the road.
I think fiber doesn't like to be bent 90 degrees, so build the conduit such that it makes gradual rather than sharp corners.
Houses last a very long time relative to networking technologies, so you *will* be changing the cabling down the road. In 10, 20, or 30 years, 100T and probably fiber will be about as useful as RS232 is now.
Of course, if current trends towards wireless continue, you'll end up abandoning the cables anyway.:-)
SCO's embarassment of an operating system didn't noticably change between the first time I had to use it on a 286 machine in the late '80s and the last time I had to use it on a pentium in the late '90's.
What the hell are you talking about? I worked at SCO in Santa Cruz from '90 to '93, and I can remember lots of significant changes made to the OS.
transition from Xenix to Unix
transition from curses to X
inclusion of a networking stack
support for kerberos and AFS
tons of system administration software
graphical mail application
inclusion of browsers
lots of driver work.
lots of performance work to make Oracle run faster
switched to using stream
performance analysis software
There's lots more work that was done, but that's the stuff I can quickly remember off the top of my head. You really have no idea what you're talking about.
Sorry, but that's not the way the statistics work.
The probability of a failure on a single drive is a cumulative distribution function. The longer the drive has been running, the higher the probability of a failure. Also, it's not linear. There are usually a few failures early in life, then relatively few for a long period of time, and then a bunch of failures again clustered around some point in time. It's kind of like a poisson distribution, but with a long head instead of a long tail.
When the manufacturer reports MTBF, I suspect they're talking about where the mean point is on this curve (i.e. at what point in time have 50% of the drives failed). I don't work in the storage industry, so this is just an educated guess. Someone will probably correct me on this.
Now, if you want to figure out the cumulative distribution function for a bunch of disks, you can't simply divide the MTBF by the number of disks.
Instead, the probability of at least one drive failure is calculated as one minus the probability that none of the drives have failed.
So, if there's a 10% chance that a single drive fails within the first year, the probability of at least one failure in a 4 drive box within that same year is 1 -.9^4 =.6.
Sorry, but that's not the way the statistics work.
The probability of a failure on a single drive is a cumulative distribution function. The longer the drive has been running, the higher the probability of a failure. Also, it's not linear. There are usually a few failures early in life, then relatively few for a long period of time, and then a bunch of failures again clustered around some point in time. It's kind of like a poisson distribution, but with a long head instead of a long tail.
When the manufacturer reports MTBF, I suspect they're talking about where the mean point is on this curve (i.e. at what point in time have 50% of the drives failed). I don't work in the storage industry, so this is just an educated guess. Someone will probably correct me on this.
Now, if you want to figure out the cumulative distribution function for a bunch of disks, you can't simply divide the MTBF by the number of disks.
Instead, the probability of at least one drive failure is calculated as one minus the probability that none of the drives have failed.
So, if there's a 10% chance that a single drive fails within the first year, the probability of at least one failure in a 4 drive box within that same year is 1 -.9^4 =.6.
If you like Macallan 18, then you should also try Glen Farclas 21 and Springbank 21. If you want to spend the bucks on something *really* nice, look into a Mortlach 25. I compared the Mortlach to the Springbank a few years ago, and found that in comparison to the Mortlach, the Springbank was thin, watery, and lacking in complexity. That's saying something if you've ever enjoyed Springbank.
Slashdot Mirror
Jesus, chief, get up on the wrong side of the bed? I read "analyst" as someone who tracks the industry, analyzes trends, and writes reports. You seem to have more of an issue with your manager.
I disagree. If I have a good idea, for which I have received a patent, why should I be prevented from simply selling the patent to a corporation that is prepared to enforce it? This gives me the money up front if I need it now instead of later.
In some ways, this is like a discussion of mining rights. A lot of effort is put in up front to figure out where there is a high probability of finding a valuable mineral, but until you sink a mine there's no proof anything is there. Some of the comments here suggest that in this case, you shouldn't be able to claim any value until you build a working mine. I disagree. There is value in doing intellectual leg work up front, and a person should be compensated for doing that work.
Clinton didn't use email either, for the same reasons.
Instant runoff voting would be a huge step forward. I'd love to see that.
The other thing I'd like to see, is the ability to vote for different people to represent me on different subjects since, for example, I tend to side with Republicans on gun control issues but with Democrats on abortion and foreign policy. Imagine being able to vote for person A to represent you on education, person B on the military, person C on foreign affairs, person D on budgets, ...
To be clear, I'm talking about a theoretical goal here, Since I know of no way to make it actually work in the real world. There are obvious problems with this, such as the fact that whoever controls the money controls everything.
Yes, generally companies will require some sort of ID and credit check to purchase a phone line, but they don't require that the person who pays for the line be the person who is going to use it. Lot's of people pay for a phone line for their aging parent who has gotten forgetful about paying bills ontime, or for their child in school who cannot afford the line.
I am simply paying for a phone line for my good friend Vladimir Macallan, and so it is his name that should be published in the phone book. Make up a suitably unrealistic name by mixing cross cultural first and last names, and you're pretty much guaranteed you won't accidentally get calls for people who really are looking for someone who happens to have that name.
Strangely enough, no. Besides, with the federal registry in place these days, I rarely get any sort of telemarketing calls anymore.
I've always gotten around the charge for an unlisted number by simply giving them a bogus name to publish in the book.
I'm not sure I know what you mean by "free trade by example". Can you elaborate and give some specific examples?
Under NAFTA, don't the US and Canada have the same policy towards each other?
I'm a Canadian who came to the US under the Canada-US free trade agreement (the precursor to NAFTA) and all I had to do was show that I was a Canadian citizan, that I had a degree in one of the listed areas, and that I had a job offer from a US company. I paid a $50 processing fee at the border, and was given a one year work visa renewable indefinitely.
Admittedly, I came to the US 14 years ago, but I was under the impression things were pretty much the same under NAFTA.
Is it more difficult than that for US citizens to work in Canada?
Many pages are cluttered with navigational junk and ads that detract from the interesting content. Take a look at www.cnn.com, for example. The story text is in the middle, and that's what I'm interested in, but all the buttons, ads, additional information, etc. takes up a lot of space.
Automatically identifying the main content of a page, and fading everything else out a bit would be very helpful.
Some sites take an article and break it up into several pages. It would be useful to automatically recognize that, fetch the continuation pages for the article, and pull the relevent content back into the original page.
I don't think anyone is disupting that people need down time. A big value of having WiFi on the train is for people who choose to live a long distance from work in order to have a larger home (or in some cases to be afle to afford any home) to raise their families in. By being able to reclaim the commute time for useful work, it's possible to spend more down time with the family rather than talking to a random stranger on the train for an hour or two in each direction.
That makes a lot of sense to me. I live and work in the San Francisco Bay Area, and I'd love to be able to live someplace less densely populated but still work here.
I'm hoping we're going in this direction, though. There has been talk recently of passing a law requiring cable and satelite companies to offer individual channels a la carte. I'd love to see this, because I never tune to most of the channels I'm forced to buy from DirecTV in order to get the few I am interested in.
Similarly, as you and others have commented, it would be great to get individual shows without being required to buy the entire channel/network. I don't think it's as bad as you say, however. It would just bring things more in line with the movie industry, where you choose individual movies to watch but aren't forced to buy a subscription to all of Disney's movies just because you want to watch one of them.
I dream of the day I can buy a subscription to a few networks that consistently produce good content (HBO, Discovery, History), grab just a few shows I'm interested in from some other networks (FX, NBC, ABC), and ignore the rest. And all of this would be delivered through my existing internet connection. The money I save due to not having to buy content I'm not interested in can be put towards a faster pipe.
Still, this model has the downside of further contributing to the "winner take all" structure that is becoming more and more prevalent in entertainment and business these days.
I'm also a Treo 600 user, and I really like it. I'm using AT&T, and although their customer service has pretty consistantly underwhelmed me, their coverage seems to be decent (San Francisco Bay Area). I've been using AT&T for many years, and switching from TDMA to GSM when I bought the Treo definitely reduced my coverage. Still, between the Cingular merger and migrating the TDMA cells to GSM, the service is getting better and should continue to improve.
Okay, back to the phone. The big selling point for me is merging the phone and organizer into a single unit. I had a Palm V before, but had stopped using it because I hated dragging it along with me everywhere. I'd invariably be out somewhere and friends would want to schedule another get together, but I'd left the Palm at home.
The todo list is really useful as well. For work, it's the list of all the things I need to get done, and whenever I think up something new, it goes into the phone so I won't forget. This is a huge help when you need to prepare for your yearly review or update your resume, because all those todo items remind you of all the things you forgot you'd done over the year. Similarly, my personal todo list keeps track of all the things I need to do around the house once the weekend comes.
Something else it's great for is attaching notes to people's address book entries. There are people I see only once every year or so, and putting directions into the phone is really handy since I can never remember them a year later. Once they're in the organizer, you'll never forget.
I'm still undecided on the value of the wireless access. Yes, there are lots of useful things you can do with it, but is it really worth $30/month for 10M? I use it for stuff like checking the train schedule, retrieving recipes so I know what to buy at the grocery store, googling phone numbers, checking weather forecasts, but there are less expensive ways of doing all those things.
Anyway, I see that I have a todo item that says "Stop reading slashdot, and get back to work" so I better take care of that.
[This is a bit of a long winded followup, that's more than just a response to your article. Bear with me and hopefully I'm not ranting too much. :-)]
So are you saying you think a business should not be allowed to negotiate a contract that restricts what you do with information or content they sell to you?
That seems to be all the Apple is doing here. They're offering to sell you access to content, with the restriction that it's locked into Apple's hardware. The reason they're willing to sell you content at the price they're selling it at, is that it drives more iPod sales. You're buying the content, and in doing so agreeing to the restrictions they've requested, but then breaking your part of the contract.
Don't get me wrong; I'd want to do the same as you. But you're changing the business model unilaterally, and if the bulk of consumers were to do what you want to do, Apple's response might need to be to raise music prices.
The general feeling here on Slashdot seems align into three groups.
The first group is whacked, in my opinion, and don't understand the realities of how things get produced. As someone who makes a living from writing software, I believe what I do has value and I should have the option of making the decision whether what I produce is given away for free, or sold as a product.
Personally, I fall somewhere in the second and third groups.
When I buy a CD, I expect to be able to rip MP3s and put them on a variety of machines from which I will listen to them. That means my home server, my portable player, my car stereo, and a directory on my work machine at the office. I don't download music, because it's already compressed at a lower quality setting than I desire. I buy CDs because I can rip them, and as compression technologies advance I can rip them again later to get a higher quality, smaller file.
I choose not to buy copy protected CDs, or DRM protected downloads, because I don't agree to the restrictions they include. But that's just my choice, and I support other people's right to sell in this way if they choose. If there is enough of a market to support that business model, then so be it.
And yes, I expect the government to help in enforcing rights, through the court system and the police if necessary. That's part of why we have a government.
DIVX was an attempt at an alternative licensing scheme, and it became clear quickly that there was not much of a market for DVDs that could only be watched a few times. I hated it, and I'm glad it died, but there was nothing wrong with it having been tried.
The video/DVD rental market is also a good example of an alternative licensing. Instead of paying $20 to own a copy of a DVD that you can watch as many times as you like, you're paying $3 to have the right to watch it for only a few days. You don't have the right to make a copy and add it to your personal library. If you want that, buy the $20 copy. If everyone were to buy the $3 rental and burn a copy, pretty soon you'd only have $20 rentals and we'd all lose out.
Another example would be if a CD was sold at $5, with the restriction that it could only be listened to for a year instead of forever. I believe a company should be able to do that, and some customers might choose that licensing for pulp pop music that they know won't stand the test of time. Personally, I wouldn't buy it (because it's pulp pop music, and because the DRM would probably get in the way of having the music available in the ways/places I want it), but other people should have the option if it works for them.
Why would I ever buy a 100 GB hard drive if it was going to fail before I could fully use that capacity?
I definitely agree that the current 1 yr warranties are ridiculous.
Why, when hard drive speed is the single largest factor affecting perceived system performance, do manufacturers insist on improving storage capacity at the expense of speed and reliability?
It depends what you're using the disk for. If you're doing video editting, 4200rpm IDE drives are fine, so why pay a premium for the faster IDE or SCSI drives? Ditto if you're using the drive for an mp3 collection or as a backup medium. In all these cases (and others), faster drives aren't going to make a difference, so cheaper/slower drives are a better choice.
Portability is nice, but you can get that conveniently with USB. An external USB disk enclosure costs about $30, and you can put whatever IDE hard drive you want into it.
If you want something small, notebook hard drives are around $100/40G plus the $30 enclosure. That's more expensive than this Jaz drive, but if you go with standard desktop drives the $/G is much better.
And of course there's also the fact that most of the slashdot crowd probably have a few extra drives laying around on a shelf somewhere, such that the only real cost is the $30 enclosure. :-)
How would you encrypt your emails with a service like this? If google can't decrypt it, then their web interface is going to only present you with a stream of encrypted data, which you'll then have to copy/paste into another window to decrypt it. This seems like way too much of a hassle to me.
I suppose you could create a front end proxy to do the encryption automatically for you, but at that point you've spent so much time and effort on the problem that you might as well just run your own imap or web mail server.
Why on earth would anyone lock in a single service provider for 100 years? Would you sign a 100 year cell phone plan? I sure wouldn't, because it removes any incentive for the company to do a good job.
It also removes any opportunity for you to vote with your dollars if you don't like the behavior of the company. I used to have my domain registered with Network Solutions, but when their parent company, Verisign, started hijacking NXDOMAIN responses and sending people to their advertising instead, I moved my domain to GoDaddy as a protest.
I don't want my dollars going to a company that seems to want to use their ICANN granted monopoly over the root database to "innovate" themselves as much profit as the can, at the expense of everyone else on the internet.
Besides, the discounted 100 year price from Network Solutions is the same as GoDaddy's everyday price, so it's not even a bargain. I've also found GoDaddy's support to be much easier to get to, and much more responsive and knowledgable than Network Solutions. I'm definitely very happy I made the move.
I think it's important to do both, actually. I used to just talk with the candidate at the high level about design and OO issues, but then I had a guy slip through who'd completely snowed me.
Since then, I still hit on high level understanding, but I also insist on covering some of the nuts and bolts. I have a list of language features, and I make the candidate describe each feature, and where they've used it in one of their own designs.
I also make them walk through some example code, and explain what's going on (e.g. C++ constructor/destructor ordering, virtual function behavior, exceptions, memory allocation leaks, etc.). This isn't "reverse the linked list" sort of stuff, it's just demonstrating a knowledge of the capabilities of the language, and ensuring they can identify some rookie mistakes in the code.
What do you need in the next few years? Put that inside a conduit. Make sure the conduit is large enough that it'll be easy to add and/or replace cables down the road.
I think fiber doesn't like to be bent 90 degrees, so build the conduit such that it makes gradual rather than sharp corners.
Houses last a very long time relative to networking technologies, so you *will* be changing the cabling down the road. In 10, 20, or 30 years, 100T and probably fiber will be about as useful as RS232 is now.
Of course, if current trends towards wireless continue, you'll end up abandoning the cables anyway. :-)
What the hell are you talking about? I worked at SCO in Santa Cruz from '90 to '93, and I can remember lots of significant changes made to the OS.
There's lots more work that was done, but that's the stuff I can quickly remember off the top of my head. You really have no idea what you're talking about.
[Reposting with better formatting.]
Sorry, but that's not the way the statistics work.
The probability of a failure on a single drive is a cumulative distribution function. The longer the drive has been running, the higher the probability of a failure. Also, it's not linear. There are usually a few failures early in life, then relatively few for a long period of time, and then a bunch of failures again clustered around some point in time. It's kind of like a poisson distribution, but with a long head instead of a long tail.
When the manufacturer reports MTBF, I suspect they're talking about where the mean point is on this curve (i.e. at what point in time have 50% of the drives failed). I don't work in the storage industry, so this is just an educated guess. Someone will probably correct me on this.
Now, if you want to figure out the cumulative distribution function for a bunch of disks, you can't simply divide the MTBF by the number of disks.
Instead, the probability of at least one drive failure is calculated as one minus the probability that none of the drives have failed.
So, if there's a 10% chance that a single drive fails within the first year, the probability of at least one failure in a 4 drive box within that same year is 1 - .9^4 = .6.
Sorry, but that's not the way the statistics work. The probability of a failure on a single drive is a cumulative distribution function. The longer the drive has been running, the higher the probability of a failure. Also, it's not linear. There are usually a few failures early in life, then relatively few for a long period of time, and then a bunch of failures again clustered around some point in time. It's kind of like a poisson distribution, but with a long head instead of a long tail. When the manufacturer reports MTBF, I suspect they're talking about where the mean point is on this curve (i.e. at what point in time have 50% of the drives failed). I don't work in the storage industry, so this is just an educated guess. Someone will probably correct me on this. Now, if you want to figure out the cumulative distribution function for a bunch of disks, you can't simply divide the MTBF by the number of disks. Instead, the probability of at least one drive failure is calculated as one minus the probability that none of the drives have failed. So, if there's a 10% chance that a single drive fails within the first year, the probability of at least one failure in a 4 drive box within that same year is 1 - .9^4 = .6.
If you like Macallan 18, then you should also try Glen Farclas 21 and Springbank 21. If you want to spend the bucks on something *really* nice, look into a Mortlach 25. I compared the Mortlach to the Springbank a few years ago, and found that in comparison to the Mortlach, the Springbank was thin, watery, and lacking in complexity. That's saying something if you've ever enjoyed Springbank.