Obviously you've never been on the freeway with several pre-occupied cell phone jockeys, otherwise known as drivers.
I could care less until it becomes a matter of my safety. When I get calls while I'm in the car, I let them know I'm in the car and that I'll call them back.
I know that cops have a stigma, especially with the age group that primarily frequents slashdot, but it's really not that hard to work with them.
I used to fight it too - just realize that if a cop feels that something needs to be enforced it is his job to do it - more often than not, "feeling" has less to do with it and "continued employment" has more effect on his decisions.
If you get hostile with a cop, remember that he has to deal with this crap every hour of every day he's at work - just like those of you who work in tech support having your occasional moments of anger, the cop is human just like you and is no different - he merely wants to shut you up, and will do anything reasonable within his power (read: something that won't get him canned) to do it. I know when I used to work tech support, "accidental transfers to a dial tone" were not uncommon when I was having a particularily bad day.
Understanding things like that will go a long way with a cop - being up front and honest goes a LOT farther that some of you might think. I've been pulled over several times when I was younger for not having insurance, and was honest every time - only when the officer pulled me over for having expired tags for the 3rd or 4th time did I get a ticket for it, because it was quite obvious to them that I wasn't going to resolve the issue.
Seriously, just ask the police officer, if permission from the library is a problem, if you may attempt to get permission - more often than not, you'd be surprised at the result: as long as you don't dawdle about it, he'll be more than happy to comply - a lot of it having to do with the fact that if he knows it's ok, he doesn't have to bother anyone else - meaning he can go sit in the car and wait for something more important to go wrong.
This is a prime example of what real respect is, and frankly, cops have a very negative stigma to them because they're the equivalent of the father/mother family where the father is always doing the spanking and the mother is always giving you cookies - cops aren't there to give you cookies.
That said, there are always going to be bad apples, and in a goverment job this is easily expanded because getting rid of the bad apple is much harder. Do your best and keep your chin up, if you get one of these types, stooping to their level is really only going to make it much, much worse for you, as you've given them justification, not them.
It's trivial to make any language unreadable - yes folks, even python. As long as there are named variables, comments are optional, and interfaces aren't required to follow a pattern, this will always be the case. And I wouldn't have it any other way.
Some great examples: the ioccc, "duff's device" - those are both in C, the most widely used programming language in the world. And if you want to have more fun, look at real world programs that abuse the crap out of cpp - it's not readable in the slightest.
In many ways what separates a good programmer from a bad one is how they use the language, not which language they use. A language is merely a means to an end - in unix, one can write a program to eject your cd-rom in bourne shell without using external programs just as much as they can do it with ioctl() C. That doesn't mean the former is the right or best way to do it, it's just that it's possible. That's not the best example, but a turing-complete language has the ability to do just about anything. This applies to your use of perl (or realistically, any other language) just as easily - there are many ways to do things, that doesn't mean that most of them are the best, readable or "right". Take a look at the OSS world and the 200 some-odd widget libraries - obviously, code reuse is not a problem specific to perl.
Percieved "lack of structure" is almost funny - perl gives you many, many, many ways to give structure to your programs.. More than any other language I have seen. Sometimes an AUTOLOAD routine fits the bill, and sometimes a class fits the bill. Sometimes you just want to write a basic shell script to get the job done. Of course, anyone who's written in bash, visual basic, basic, and assembler knows those languages don't require any structure, either. In the case of the last two there is nothing that/provides/ structure (only deviants and extended versions of these do). I know nothing of matlab and "excel", as padding my resume is not my style.
Perl borrows a lot from list processing concepts and I think this is where a lot of people start to miss the "readability" factor. It did wonders for my ability to manipulate perl to learn Lisp. Nowadays, nothing requires magic variables and CPAN seems to refute your reusability argument rather well - granted, I have another 90k LOC to back me up on that one that's at work here.
I'm more inclined to believe you don't know how to program in perl than to believe you are knowledgable enough to judge it.
I actually like the first play on words - often when a case is pro bono is has less to do with being a "good samaritan" and more to do with making headlines.
For starters, it's easy to preach and hard to prove.
XSLT can do great transformations but it is no match for a program that actually knows how to interpret the content - a badly designed XSD will not transform well.
Here's an example of a horrible but all too common schema problem:
XSLT simply cannot predict that there are going to be 10 of those prices, nor can it predict that there is an upper bound of 50 of them.
And if you haven't seen that, I'm sure you've seen the same principle applied to tag attributes, which supplies the same problem.
To make a "simple" thing more complicated, you get into fun things like character sets and fonts (combined this can be a very nasty mess - "webdings" is a great example of how your character set means nothing), and then to insure the complete compatibility of the document the rendering engine will most likely need to be swapped or reconfigured depending on the format you're reading or signature "quirks" of the "common" format - we already see this in HTML - Opera and Firefox both have multiple rendering engines to better support IE-only pages.
For a great example, take a look at the w3c specs - they are very simple and fairly well-defined nowadays, and still, no browser renders everything properly. How do you expect a clone of a proprietary program with a closed format to fare any better?
Re:Far off, but going to happen
on
Palmtop Nirvana?
·
· Score: 1
And if you want to see anything useful, that watch is going to be slightly larger that a gideon's bible.
Personally, I've done the PDA thing and even though I held off for quite a long time being a skeptic of "gadgetitis", I gave in and bought one and frankly, I'm back to square one - maybe I just don't have as many appointments as other people, or I'm just better organized or something, but I rarely find it more than an extra to be reminded of my appointments since I'm normally aware they're coming.
Other than the calendar and address book, I really don't see a need to have yet another thing in my pocket - then again, I'm the guy who turns his phone off when going to the movie theatre and leaves it in the car when we go camping. Don't get me wrong, I'll bring a laptop camping too, but if I open it up chances are no one else is awake or I had some spur of the moment idea that I needed to jot down really quick.
The constant attachment to tech really bothers me - I hate having a cellphone because people get angry when they call it and I don't answer - yet when people call my home phone and I don't answer they just assume I'm out.
Likewise are the people who feel the need to constantly remind me to read my email - I'll read it when I'm ready - if it was that important you could have taken the "read my email to you" time to explain the problem.
Of course, there are extremes to everything - it's very frustrating to write up a design at work only to check with people two weeks later, that criticize the topic of the design but haven't read the actual article describing it. Or to have my boss get PO'd at me because I "didn't do something" that's been sitting in unread (or even more ironically, marked read) status in his email box.
Re:The device isn't usually the issue for me
on
Palmtop Nirvana?
·
· Score: 1
That last comment frustrates me the most about the Mozilla distribution - they don't take advantage of the system it's residing on at all.
Thunderbird on Windows and Mac do not use the supplied address book software - less important on windows but when insignificant software like the AIM client on the mac does, perhaps it's time to "Get with the program" and use what every other app does.
Firefox doesn't share bookmarks with the rest of the system - another very annoying "feature". Being able to test on multiple browsers becomes a lot easier when you don't have to bookmark everything 4+ times. I'm quite aware this is not a problem unique to firefox yet it frustrates me just the same.
As I understand it (I'm not really a network geek, so I could be wrong here), arp poisoning is an easy tactic to start getting data from any machine with a signal strength that can reach you.
The negotiation is done in hardware, so if drivers are implemented correctly all the OS sees is another ethernet device with a possible extra set of status information and twiddles.
This is how some hardware SSL accelerators work as well.
Although you are correct in the fact that the encryption standards are not compatible with each other.
IANAL, but I think the thing that really bit them in the rear here is that they went for rights to reproduce the format first. This bolsters the Best Brains side of the case that they are copying the format and not doing a parody.
I agree - if the cop felt that a law was being broken, and he had no choice on how to enforce it, he could have taken it much farther.
But that gets to another point - instead of whining on the internet about it... Knowing it's the library's signal, why not just go in and see what the AUP or whatever the rules are, and bring the cop in for good measure? More often than not (there are exceptions), the cop will be happy to find out if the library staff really cares or not.
While I'm sure there are a few out there, your average ground-pounding police officer will never read this blog entry or the slashdot article that links to it. He won't be informed of the issues, and certainly isn't going to listen to anyone that he has no reason to believe as an authority on the issue. This is a better practice than you think - I'd rather be getting told that something is wrong because someone with authority told him to tell me that than some "ordained priest" with a WiFi connection told him it was ok.
Out of a team of 7, 3 of our coworkers have worked at other jobs with him - to which all of them were hired without question... All of our standard testing and interview practices were thrown out the window.
As a result, one of them (the only one hired before me) really had a grasp on our primary programming language (perl) when they were hired. Guess who got to teach them basic things like "no, while() requires a block in perl - that's just how it is". Realistically, only one more of them knows the language at an acceptable level to write production code - very slow production code that uses more tricks that he learned from MJD at OSCON than pragmaticism.
The irony of all this is that 3 of them were hired later than me and they're all getting paid more because they started later - great reward system. 30% of the reasonably large code bsae came from my fingertips... And since they have been in the primary design roles in the last year and refuse to listen to me as I try to plan ahead, we have appromately 2 months to double our site performance (200 h/s to 400 h/s). To add icing to the cake, I am the only member on my team willing to work nights and weekends to make up lost time for everyone writing design documents which involve less than 100 lines of code.
I quit once this month already, and due to no change in the environment (a condition I demanded upon returning) I left resignation last thursday and was told to wait until today to discuss it with my vacationing manager. To put it simply: we are screwed, and I want nothing to do with it - I put up an idea on something that might fix it but 2 weeks have now passed with no support from management or the team, because our other "designers" don't like it (read: Not Invented Here).
Apologies for the rant, but I know there are those who have been in this kind of situation before and I'm sure know how frustrating it is to take pride in your work, only to have it stomed on by a bunch of nieve "good ol' boys".
My first IBM PC in the home was a 286/16, somewhere between '86 and '88 - I can't remember. It came with a Logitech 3 button mouse.
That mouse was with all my machines until around '96 or '97. A roommate's cat gave it a bath by knocking it in a soda cup. Sadly enough, I remember how angry I got over that.:)
Naturally, given the quality of the mouse I had, I figured I'd buy another logitech. That thing didn't last 2 weeks. At least I got my money back for that one.
2-3 years down the road, I was in the market for yet another mouse, so I figured I'd get another Logitech - that thing lasted about 3 months.
After going through tons of mice since around '96, I finally settled on something that I like: a shitty Microsoft Optical that cost me $20. Combine it with a nice $25 mouse pad and it plays better than or equal to your MX<insert number of the week here>. I have recieved several unsolicited comments about how smooth it works and how well it tracks.
MX1000 sounds very neat but I'll be sure to wait for the corded version - I've done my tour with wireless mice and now the only thing that bothers me about them is that my wife complains because I pawned the $120 mouse off to her to take the $20 mouse I have now.:)
What I don't get - when microsoft first started selling mice, you could look on the bottom of the mouse and it would say something to the effect of "manufacturered by logitech". If they can manage to make MS mice better than their own....sigh.
If you can buy into the whole "performance mouse pad" thing (it took a lot of convincing from other people before I could), check out the func1030. It really is worth it, the surface is slicker than any desk or other mouse pad I have used and one side has more tack. For FPS games, I am superbly satisfied and the fact that my desk surface doesn't have a huge rubbed off spot on it makes me happy too.
I've had mine for about a year now with no complaints. The pad is nice and large and comes with a silly gimmick that is actually handy for people w/ corded mice - a clip that keeps your cord from moving... Give it a little slack and you never have to worry about dragging that cord again.
We have a guy at work (that I'm "lucky" enough to sit next to) that has a phone that supports polyphonic ring tones.
After hearing a really bad midi version of "The Final Countdown" for a good 9 months, at full volume, having the majority of the people in proximity to him complain about it, he replaces it with some bullshit spewing out of William Shatner's mouth. Meanwhile, the rest of us leave our phones with the generic stock tones (polyphonic or not), and those that get calls often set their phones to just vibrate.
Granted, this is the guy who is proud of writing 40k LOC of useless code that we replaced with a 100 line perl module and doesn't even work because he never finished a working build system for it. I guess that explains a lot.
Systems that use the Open Sound System have this problem - basically, most free unix derivatives.
Basically, the Open Sound System provides no mixing whatsoever - so any program that binds to the OSS interface blocks it - and other programs waiting for the device to become available again come to a screeching halt.
Creative hackers have built daemons which do the talking to the OSS interface and then provide an API for programs to use it - you might have heard of esd or arts, which is how GNOME and KDE do their sound mixing. Of course, if any program gets the OSS interface before those programs, guess what happens.:) Or, if you have a program that tries to bind to it later, guess what happens to that program?
There is one problem in your rationale - generally, executive officers in particular not only make decisions that can easily "make or break" the company, but are figureheads that people get behind - namely the stockholders, but also some of the customers. Take Bill Gates or Steve Jobs for example.
To put it bluntly, outsourcing these "stars" to a bunch of no-names that may or may not even speak the language of the stock holder is a very, very bad idea.
Take a "for example" - when you read about a promotion at a publicly held company, the focus of the text is always on the previous experience and accomplishments of that person - and stock holders with a lot invested long-term in that company (VC comes to mind) do a lot of research on that person to find the things they didn't talk about - and sometimes have a hand in hiring decisions. This stuff isn't just grandstanding - it's convincing the small guy that their continued investment is worthwhile.
This easily applies to other stars in the company - would you really buy the next game from iD if John Carmack left tomorrow, or would you concentrate on Carmack's next project? A good example of this in reality (although fading) is Richard Garriott, or Gary Gygax.
It's the same reason Hollywood hires big name actors and puts them in movies that would otherwise tank - star power has a lot more to do with the movie's success than the script, and as a result the more prominent actors review their scripts, to protect their reputation (and marketability), as a result.
Once you're at the top, the only way to get back when you fall is to go into rehab.:)
Actually, at least in my mind, all that equates to is an equalized market that pays less for better work - since no workstation tech is going to move to india (or whereever) to further his career, he's going to learn more to keep his job and make the same amount of money - if he loses his job, he's going to go somewhere else, competing with all the other people in the same situation, which leads to an employer's market. Employer's Market = less benefits and lower wages.
The automobile industry is no different than any other - selling the "wide beta" happens all the time, which you can verify on Epinions or other sites that review cars.
Early adopters always get screwed 2 ways: the price and the performance.
DNS returns CNAME to sitefinder.com for any NXDOMAIN that would have been returned.
Put this in your apache conf (sketchy, you might need slightly more)
NameVirtualHost *:80
Then write a stupid php or mod_perl handler to process the Host: header given in the http/1.1 request, and give it to the template engine which generates the page.
Of course on that kind of scale perhaps php or mod_perl (apache?) would be too slow, but you get the idea.:)
Heck, now that I think about it, it could all be done with inetd and a shell script, trivially at that.:)
Slashdot Mirror
Obviously you've never been on the freeway with several pre-occupied cell phone jockeys, otherwise known as drivers.
I could care less until it becomes a matter of my safety. When I get calls while I'm in the car, I let them know I'm in the car and that I'll call them back.
I know that cops have a stigma, especially with the age group that primarily frequents slashdot, but it's really not that hard to work with them.
I used to fight it too - just realize that if a cop feels that something needs to be enforced it is his job to do it - more often than not, "feeling" has less to do with it and "continued employment" has more effect on his decisions.
If you get hostile with a cop, remember that he has to deal with this crap every hour of every day he's at work - just like those of you who work in tech support having your occasional moments of anger, the cop is human just like you and is no different - he merely wants to shut you up, and will do anything reasonable within his power (read: something that won't get him canned) to do it. I know when I used to work tech support, "accidental transfers to a dial tone" were not uncommon when I was having a particularily bad day.
Understanding things like that will go a long way with a cop - being up front and honest goes a LOT farther that some of you might think. I've been pulled over several times when I was younger for not having insurance, and was honest every time - only when the officer pulled me over for having expired tags for the 3rd or 4th time did I get a ticket for it, because it was quite obvious to them that I wasn't going to resolve the issue.
Seriously, just ask the police officer, if permission from the library is a problem, if you may attempt to get permission - more often than not, you'd be surprised at the result: as long as you don't dawdle about it, he'll be more than happy to comply - a lot of it having to do with the fact that if he knows it's ok, he doesn't have to bother anyone else - meaning he can go sit in the car and wait for something more important to go wrong.
This is a prime example of what real respect is, and frankly, cops have a very negative stigma to them because they're the equivalent of the father/mother family where the father is always doing the spanking and the mother is always giving you cookies - cops aren't there to give you cookies.
That said, there are always going to be bad apples, and in a goverment job this is easily expanded because getting rid of the bad apple is much harder. Do your best and keep your chin up, if you get one of these types, stooping to their level is really only going to make it much, much worse for you, as you've given them justification, not them.
Listen,
/provides/ structure (only deviants and extended versions of these do). I know nothing of matlab and "excel", as padding my resume is not my style.
Perl makes it easier. That's it.
It's trivial to make any language unreadable - yes folks, even python. As long as there are named variables, comments are optional, and interfaces aren't required to follow a pattern, this will always be the case. And I wouldn't have it any other way.
Some great examples: the ioccc, "duff's device" - those are both in C, the most widely used programming language in the world. And if you want to have more fun, look at real world programs that abuse the crap out of cpp - it's not readable in the slightest.
In many ways what separates a good programmer from a bad one is how they use the language, not which language they use. A language is merely a means to an end - in unix, one can write a program to eject your cd-rom in bourne shell without using external programs just as much as they can do it with ioctl() C. That doesn't mean the former is the right or best way to do it, it's just that it's possible. That's not the best example, but a turing-complete language has the ability to do just about anything. This applies to your use of perl (or realistically, any other language) just as easily - there are many ways to do things, that doesn't mean that most of them are the best, readable or "right". Take a look at the OSS world and the 200 some-odd widget libraries - obviously, code reuse is not a problem specific to perl.
Percieved "lack of structure" is almost funny - perl gives you many, many, many ways to give structure to your programs.. More than any other language I have seen. Sometimes an AUTOLOAD routine fits the bill, and sometimes a class fits the bill. Sometimes you just want to write a basic shell script to get the job done. Of course, anyone who's written in bash, visual basic, basic, and assembler knows those languages don't require any structure, either. In the case of the last two there is nothing that
Perl borrows a lot from list processing concepts and I think this is where a lot of people start to miss the "readability" factor. It did wonders for my ability to manipulate perl to learn Lisp. Nowadays, nothing requires magic variables and CPAN seems to refute your reusability argument rather well - granted, I have another 90k LOC to back me up on that one that's at work here.
I'm more inclined to believe you don't know how to program in perl than to believe you are knowledgable enough to judge it.
I actually like the first play on words - often when a case is pro bono is has less to do with being a "good samaritan" and more to do with making headlines.
For starters, it's easy to preach and hard to prove.
<price3>14.95</price3>
XSLT can do great transformations but it is no match for a program that actually knows how to interpret the content - a badly designed XSD will not transform well.
Here's an example of a horrible but all too common schema problem:
<price1>12.95</price1>
<price2>13.95</price2>
XSLT simply cannot predict that there are going to be 10 of those prices, nor can it predict that there is an upper bound of 50 of them.
And if you haven't seen that, I'm sure you've seen the same principle applied to tag attributes, which supplies the same problem.
To make a "simple" thing more complicated, you get into fun things like character sets and fonts (combined this can be a very nasty mess - "webdings" is a great example of how your character set means nothing), and then to insure the complete compatibility of the document the rendering engine will most likely need to be swapped or reconfigured depending on the format you're reading or signature "quirks" of the "common" format - we already see this in HTML - Opera and Firefox both have multiple rendering engines to better support IE-only pages.
For a great example, take a look at the w3c specs - they are very simple and fairly well-defined nowadays, and still, no browser renders everything properly. How do you expect a clone of a proprietary program with a closed format to fare any better?
And if you want to see anything useful, that watch is going to be slightly larger that a gideon's bible.
Personally, I've done the PDA thing and even though I held off for quite a long time being a skeptic of "gadgetitis", I gave in and bought one and frankly, I'm back to square one - maybe I just don't have as many appointments as other people, or I'm just better organized or something, but I rarely find it more than an extra to be reminded of my appointments since I'm normally aware they're coming.
Other than the calendar and address book, I really don't see a need to have yet another thing in my pocket - then again, I'm the guy who turns his phone off when going to the movie theatre and leaves it in the car when we go camping. Don't get me wrong, I'll bring a laptop camping too, but if I open it up chances are no one else is awake or I had some spur of the moment idea that I needed to jot down really quick.
The constant attachment to tech really bothers me - I hate having a cellphone because people get angry when they call it and I don't answer - yet when people call my home phone and I don't answer they just assume I'm out.
Likewise are the people who feel the need to constantly remind me to read my email - I'll read it when I'm ready - if it was that important you could have taken the "read my email to you" time to explain the problem.
Of course, there are extremes to everything - it's very frustrating to write up a design at work only to check with people two weeks later, that criticize the topic of the design but haven't read the actual article describing it. Or to have my boss get PO'd at me because I "didn't do something" that's been sitting in unread (or even more ironically, marked read) status in his email box.
That last comment frustrates me the most about the Mozilla distribution - they don't take advantage of the system it's residing on at all.
Thunderbird on Windows and Mac do not use the supplied address book software - less important on windows but when insignificant software like the AIM client on the mac does, perhaps it's time to "Get with the program" and use what every other app does.
Firefox doesn't share bookmarks with the rest of the system - another very annoying "feature". Being able to test on multiple browsers becomes a lot easier when you don't have to bookmark everything 4+ times. I'm quite aware this is not a problem unique to firefox yet it frustrates me just the same.
As I understand it (I'm not really a network geek, so I could be wrong here), arp poisoning is an easy tactic to start getting data from any machine with a signal strength that can reach you.
Here is a description of what it entails.
The negotiation is done in hardware, so if drivers are implemented correctly all the OS sees is another ethernet device with a possible extra set of status information and twiddles.
This is how some hardware SSL accelerators work as well.
Although you are correct in the fact that the encryption standards are not compatible with each other.
IANAL, but I think the thing that really bit them in the rear here is that they went for rights to reproduce the format first. This bolsters the Best Brains side of the case that they are copying the format and not doing a parody.
I agree - if the cop felt that a law was being broken, and he had no choice on how to enforce it, he could have taken it much farther.
But that gets to another point - instead of whining on the internet about it... Knowing it's the library's signal, why not just go in and see what the AUP or whatever the rules are, and bring the cop in for good measure? More often than not (there are exceptions), the cop will be happy to find out if the library staff really cares or not.
While I'm sure there are a few out there, your average ground-pounding police officer will never read this blog entry or the slashdot article that links to it. He won't be informed of the issues, and certainly isn't going to listen to anyone that he has no reason to believe as an authority on the issue. This is a better practice than you think - I'd rather be getting told that something is wrong because someone with authority told him to tell me that than some "ordained priest" with a WiFi connection told him it was ok.
Lose a Clone. Return to the mission center for re-training.
This might be a little more in line with what woz and jobs were thinking of when they named it the "Macintosh".
I bought it direct.
Especially when you do all of them at the same time.
AvP was a comic book, then a video game before the movie rights were even penned. And even that was many moons ago.
They could have done a lot better, heck, they certainly had the time to do it.
Oh, if you only knew....
Out of a team of 7, 3 of our coworkers have worked at other jobs with him - to which all of them were hired without question... All of our standard testing and interview practices were thrown out the window.
As a result, one of them (the only one hired before me) really had a grasp on our primary programming language (perl) when they were hired. Guess who got to teach them basic things like "no, while() requires a block in perl - that's just how it is". Realistically, only one more of them knows the language at an acceptable level to write production code - very slow production code that uses more tricks that he learned from MJD at OSCON than pragmaticism.
The irony of all this is that 3 of them were hired later than me and they're all getting paid more because they started later - great reward system. 30% of the reasonably large code bsae came from my fingertips... And since they have been in the primary design roles in the last year and refuse to listen to me as I try to plan ahead, we have appromately 2 months to double our site performance (200 h/s to 400 h/s). To add icing to the cake, I am the only member on my team willing to work nights and weekends to make up lost time for everyone writing design documents which involve less than 100 lines of code.
I quit once this month already, and due to no change in the environment (a condition I demanded upon returning) I left resignation last thursday and was told to wait until today to discuss it with my vacationing manager. To put it simply: we are screwed, and I want nothing to do with it - I put up an idea on something that might fix it but 2 weeks have now passed with no support from management or the team, because our other "designers" don't like it (read: Not Invented Here).
Apologies for the rant, but I know there are those who have been in this kind of situation before and I'm sure know how frustrating it is to take pride in your work, only to have it stomed on by a bunch of nieve "good ol' boys".
Your comment got me thinking....
:)
:)
My first IBM PC in the home was a 286/16, somewhere between '86 and '88 - I can't remember. It came with a Logitech 3 button mouse.
That mouse was with all my machines until around '96 or '97. A roommate's cat gave it a bath by knocking it in a soda cup. Sadly enough, I remember how angry I got over that.
Naturally, given the quality of the mouse I had, I figured I'd buy another logitech. That thing didn't last 2 weeks. At least I got my money back for that one.
2-3 years down the road, I was in the market for yet another mouse, so I figured I'd get another Logitech - that thing lasted about 3 months.
After going through tons of mice since around '96, I finally settled on something that I like: a shitty Microsoft Optical that cost me $20. Combine it with a nice $25 mouse pad and it plays better than or equal to your MX<insert number of the week here>. I have recieved several unsolicited comments about how smooth it works and how well it tracks.
MX1000 sounds very neat but I'll be sure to wait for the corded version - I've done my tour with wireless mice and now the only thing that bothers me about them is that my wife complains because I pawned the $120 mouse off to her to take the $20 mouse I have now.
What I don't get - when microsoft first started selling mice, you could look on the bottom of the mouse and it would say something to the effect of "manufacturered by logitech". If they can manage to make MS mice better than their own....sigh.
If you can buy into the whole "performance mouse pad" thing (it took a lot of convincing from other people before I could), check out the func1030. It really is worth it, the surface is slicker than any desk or other mouse pad I have used and one side has more tack. For FPS games, I am superbly satisfied and the fact that my desk surface doesn't have a huge rubbed off spot on it makes me happy too.
I've had mine for about a year now with no complaints. The pad is nice and large and comes with a silly gimmick that is actually handy for people w/ corded mice - a clip that keeps your cord from moving... Give it a little slack and you never have to worry about dragging that cord again.
We have a guy at work (that I'm "lucky" enough to sit next to) that has a phone that supports polyphonic ring tones.
After hearing a really bad midi version of "The Final Countdown" for a good 9 months, at full volume, having the majority of the people in proximity to him complain about it, he replaces it with some bullshit spewing out of William Shatner's mouth. Meanwhile, the rest of us leave our phones with the generic stock tones (polyphonic or not), and those that get calls often set their phones to just vibrate.
Granted, this is the guy who is proud of writing 40k LOC of useless code that we replaced with a 100 line perl module and doesn't even work because he never finished a working build system for it. I guess that explains a lot.
Systems that use the Open Sound System have this problem - basically, most free unix derivatives.
:) Or, if you have a program that tries to bind to it later, guess what happens to that program?
Basically, the Open Sound System provides no mixing whatsoever - so any program that binds to the OSS interface blocks it - and other programs waiting for the device to become available again come to a screeching halt.
Creative hackers have built daemons which do the talking to the OSS interface and then provide an API for programs to use it - you might have heard of esd or arts, which is how GNOME and KDE do their sound mixing. Of course, if any program gets the OSS interface before those programs, guess what happens.
Guess what the Flash plugin for linux does?
There is one problem in your rationale - generally, executive officers in particular not only make decisions that can easily "make or break" the company, but are figureheads that people get behind - namely the stockholders, but also some of the customers. Take Bill Gates or Steve Jobs for example.
:)
To put it bluntly, outsourcing these "stars" to a bunch of no-names that may or may not even speak the language of the stock holder is a very, very bad idea.
Take a "for example" - when you read about a promotion at a publicly held company, the focus of the text is always on the previous experience and accomplishments of that person - and stock holders with a lot invested long-term in that company (VC comes to mind) do a lot of research on that person to find the things they didn't talk about - and sometimes have a hand in hiring decisions. This stuff isn't just grandstanding - it's convincing the small guy that their continued investment is worthwhile.
This easily applies to other stars in the company - would you really buy the next game from iD if John Carmack left tomorrow, or would you concentrate on Carmack's next project? A good example of this in reality (although fading) is Richard Garriott, or Gary Gygax.
It's the same reason Hollywood hires big name actors and puts them in movies that would otherwise tank - star power has a lot more to do with the movie's success than the script, and as a result the more prominent actors review their scripts, to protect their reputation (and marketability), as a result.
Once you're at the top, the only way to get back when you fall is to go into rehab.
Actually, at least in my mind, all that equates to is an equalized market that pays less for better work - since no workstation tech is going to move to india (or whereever) to further his career, he's going to learn more to keep his job and make the same amount of money - if he loses his job, he's going to go somewhere else, competing with all the other people in the same situation, which leads to an employer's market. Employer's Market = less benefits and lower wages.
The automobile industry is no different than any other - selling the "wide beta" happens all the time, which you can verify on Epinions or other sites that review cars.
Early adopters always get screwed 2 ways: the price and the performance.
The sick thing is how SIMPLE it was.
:)
:)
DNS returns CNAME to sitefinder.com for any NXDOMAIN that would have been returned.
Put this in your apache conf (sketchy, you might need slightly more)
NameVirtualHost *:80
Then write a stupid php or mod_perl handler to process the Host: header given in the http/1.1 request, and give it to the template engine which generates the page.
Of course on that kind of scale perhaps php or mod_perl (apache?) would be too slow, but you get the idea.
Heck, now that I think about it, it could all be done with inetd and a shell script, trivially at that.
Big money big money big money!
Noooooooooooooooooooo whammies!