I can sort quasi-spam (ads from companies I've placed orders from, for example) far more effectively with body filters.
For that use case, you might try tagged addresses. Thanks to qmail and postfix, I subscribe to things with addresses like
bob-vendor.com@example.com
Where "vendor.com" is the name of the company I'm giving the address to and "example.com" is my domain. When somebody misuses or leaks an address, I just add a filter for it.
Yes, but that's arguably a mistake, and certainly no excuse.
MTAs are there to present user-friendly interfaces on top of protocols. For example, the way SMTP works is to require authentication every time you connect. But that's no reason for a mail client to act the same way.
People like trash cans. If that's how they think of deletion, there's no reason to train them to use some other model. Software exists for the convenience of its users, not vice versa.
I confess it has been a few years, but that's why I walked away from Evolution. That and some fuckwad Ximian developers who were utter dicks on the mailing lists. I'm sure most of them must have been nice, but all I remember about Evolution is the sound of my teeth grinding like somebody trying to start a running car.
Thunderbird's not what I call bulletproof, but it never loses my mail, and I have come to terms with most of its quirks.
You are assuming that "trade" is automatically equitable and of benefit to both parties.
I don't have to assume it. I can prove it. Every time you buy something, you're engaging in trade. When you buy something, it's because you think you'll be better off with the thing than the money. The seller thinks that they'll be better off with the money than the thing. Sure, sometimes you're wrong. But you're usually right, or you'd stop trading.
That's all trade is. People deciding to trade one thing for another because they think it's good for them. If you think it's a bad idea, stop buying things for a few weeks.
Interestingly enough, truly "free" trade among nations invariably results in one or more parties being completely screwed by another.
Nice assertion. Got facts?
I do. If you look at the rise of global trade versus pretty much any useful metric over the last 300 years (standard of living, lifespan, health, education, wealth) they are positively correlated with trade. America is one of the most open nations to trade, and one of the most wealthy. Hong Kong's another great example. A lot of the places most resistant to trade are poor, miserable places to live.
Put it this way: free trade is like depending, for your very life, on another man's better nature.
Hardly. The guy at the corner store who sells me milk doesn't do it because he's nice. It's because he makes his living that way. The same thing for the guy delivering the milk, the ones processing it, and the ones with the cows. None of them are in it because they'd be sad to think of me eating dry cereal in the mornings.
Trade only depends on other people behaving in their own interest. That's much more reliable than somebody's good nature. Trade binds us together into networks of mutual benefit.
Of course, I don't expect to convince you, but let's move the discussion ahead a few steps. If you're going to explain why trade is bad, start off by explaining why whatever effect you're going to point at is large enough to overwhelm comparative advantage. And if you don't know what comparative advantage is and how it works, then you should probably stop now. Arguing about trade without understanding that is like arguing about physics without understanding atoms: possibly entertaining, but unlikely to be useful.
how is America's participation in the Global Economy of benefit... to America?
Your focus on trade is a red herring. Trade makes both participants richer. Otherwise, they wouldn't trade. Artificially limiting trade makes people poorer.
If you're concerned about inequality, then focus on the causes of inequality. Get upset about deficit funding of tax cuts for the rich. Get upset about poor education. Get upset about large corporations, who take most of the value generated and give it to a few people at the top. Get upset about crony capitalism. Get upset about the way rich people and large companies buy the laws they want.
But don't get in the way of trade. Since trade makes people richer, you can take part of that extra money to help everybody else.
Do you think the factory workers, or even the management at Lenovo have anything to do with China's military decisions?
Nice rant -- if you were right. But Lenovo is 30% state-owned, so its success benefits China's gerontocracy, who are behind a lot of smashed skulls in that part of the world.
AFAIK Ennis does not sound like a paranoid lunatic to me
No, it's another known category of people I don't take very seriously: paranormal promoters.
I used to look into these things with some vigor. You never know, right? I got tired of it, as every time I got to the bottom of it, it was bunk. The problem is that it's much easier for somebody to fool themselves than it is for an outside observer to figure out how they screwed up. And that's leaving out entirely the question of fraud, which it ten times harder to prove.
Now my take is that it is incumbent upon the person with the wild result or theory to come up with the clear proof if they want me to take them seriously. Yes, this means that I won't be the first person to confirm that invisible pink unicorns really exist. I'm ok with that. If I had infinite time, I'd pay the fringe types more attention. But I've only got so long before they put me in a pine box, and I've got better things to do.
He had one of the best rails programmers there was. It wasn't a problem with not knowing rails.
No, it was a problem of trying to use Rails to work just like he did in PHP. He wanted to use his existing database layout. He wanted to write SQL queries. The Rails attitude is that your database should derive from your OO model, and you should generally trust Rails to get your data for you.
So you could sorta say he knew it. But until he tries it the way the authors meant it to be used, he still won't really get what its strong points are.
Looking at Ruby on Rails, it seems like a cheap clone of the old NeXT WebObjects
It has been a long time since I did WebObjects work, but I don't think they're particularly similar. The spirit is certainly very different.
So, why do people use Ruby?
I think there are two big crowds. One is the smart OO guys who have been suffering through Java for years. Smalltalk is not commercially viable, but Ruby is. Suddenly, they can escape all the Java idiocy. For the ones doing web stuff, and in particular the ones who have dealt with the absurdity of trying to program via large XML framework config files, Rails is a similarly big relief.
The other crowd comes from PHP and other low-rent web development. Ruby + Rails lets them get something up as quickly as in PHP, but provides them better long-term tools and something much closer to what I'd call object-oriented development. Of course, a lot of those people quickly get in over their heads in Rails, as a lot of the shoddier Rails plugins demonstrate.
So maybe it would be better if Rails were written in Smalltalk or Lisp. But Ruby it is, and it's not so bad. If Java's half-way between C++ and Lisp, then Ruby's cut the distance in half again.
Similarly, why care about the pledge of allegiance? It's one small line, and you can omit it if you wish.
You need to work on your thinking skills a little.
The whole point of the pledge of allegiance is to care about something, and to do it with everybody else. As far as I'm concerned, bringing up somebody else's imaginary god in the middle is like putting a screen door in an aquarium.
The truth is that OLPC was largely unaware of the difficulties this kind of project would face. OLPC set an unreasonable goal for the price, and now they're coming to terms with the reality of the situation. Initially OLPC had said that the market wouldn't produce an inexpensive laptop because the profits weren't there. It turns out that the market wasn't making them because it's not possible.
Evidence?
In the earliest OLPC speeches I heard, the $100 was always a goal, and they didn't expect to hit it with their first systems. They expect to get there over time.
And as you can see elsewhere in the thread, ASUS's upcoming Eee seems to be exactly the kind of inexpensive laptop that you claim is impossible. The point wasn't to hit some magic $100 goal; the point was to make a radically cheaper laptop build for the needs of third-world students rather than western consumers.
You're behind the times. The base Eee has dropped to 256 MB RAM, the storage to 2 GB, and they dropped the camera. See the Eee Wikipedia entry for more.
And what you're missing is that they only reason the Eee exists is the OLPC program. Which was partly the point. Nobody believed it was possible or reasonable when they started, as everybldy was focused on high-end laptops. Now a major manufacturer is trying to match them. That's not a problem, that's success.
Re:Where's the evidence Ennis is incompetent?
on
Science vs. Homeopathy
·
· Score: 2, Insightful
But because of the circumstances, I doubt most scientists would risk their careers investigating "cold fusion" or homeopathy, or even be able to get funding to do so in the first place.
Plenty of people investigated cold fusion. Some people found something. Most people didn't. The people who found something couldn't reliably replicate their results. Research continues.
What's clear, though, is that some well-meaning people ended up doing bad science because they fell for the hype and let it influence their results. That is an entirely justified black eye for them, as guarding against that is pretty much the point of science.
I think we shouldn't dismiss her (or the entire field of homeopathy) just because of that experiment.
A homeless guy in my neighborhood is convinced that there is a sinister, far-reaching conspiracy against him, orchestrated by his estranged and abusive father. Now sure, it could be true. We probably shouldn't dismiss him or his theories just because he sounds like every other paranoid lunatic and has no proof. But let's just say I'm not rushing to investigate, either.
Some things are just provably wrong and don't need experimentation.
You realize that there's no difference between your reasoning and that of most pseudoscience followers, right?
There is no a priori knowledge. Evolution has given you a variety of biases and some mechanisms for picking up plenty more. You can either let your biases run away with you, or you can roll up your sleeves and do some science.
If you're not doing the science and you run around telling people not to do the science, you're basically cheerleading for a return to a medieval paralysis of knowledge. To which I say: fuck that! If nothing else, people were much smellier.
I own a retail store in California, and have made it my business to know the law. Here the store would win any lawsuit hands down. It wouldn't even make it to a trial; the defending attorney would quote all the case law that has already decide the issue and the judge would throw the suit out in pre-trial motions.
Don't you have to have probable cause? E.g., seeing somebody steal something? I'm pretty sure that you can't arbitrarily detain people just because they've been in your store.
If I'm wrong please cite the laws you've been reading, as I'd like to see the details.
On the one hand, this is great. The more a law enforcement officer can get done with their time, the better. Plenty of crime goes unaddressed because it is "too small". The FBI, for example, won't talk to you about interstate computer crime unless you can prove a minimum of $10k of losses. And because they're busy, the effective threshold is much higher.
On the other hand, the US government has recently been a little cavalier about my rights, and there are historical periods where they've been a fair bit worse, like the Second Red Scare. It's enough to make me want to get a bunch of disposable prepaid cellphones, just in case.
I'd feel a lot better if there were some rule about public posting of eavesdropping records. E.g., that within 10 years after any eavesdropping incident, the government is obliged to publish who ordered it, why, and who accessed any data from it. As Brin points out in The Transparent Society, the problem isn't so much surveillance as secret, one-way surveillance.
You used eleven and a half words to communicate a concept that could have been done with two: being reactive and proactive. I know the latter has been sullied by its use in jargon bingo, but we are still capable of understanding the concept.
When I talk to MBAs, I use words like "reactive" and "proactive". But here I wouldn't because a) people would think I'm some fuckwad MBA and ignore me, and b) those who have just heard the jargon terms but haven't thought about what they really mean would have skated on by my message.
You know the terms and their real meaning, so you thought about what I said and sorted it out. Fine, have a cookie. For everybody else, I wanted them to do similar thinking whether or not they really got the terms.
And yes, of course people are capable of understanding the concept. That's why I focused on the concept, rather than on some popular labels for the concept. So thanks for your reactive copyediting, but maybe you could run along and proactively contribute to the discussion?
No. They did what companies do, make money for their shareholders.
Yes. "Ve vere just following orders, mein herr." Always a perfectly good excuse.
Wait, no, I forgot. That isn't a good excuse.
Because we are all citizens and human beings first, and employees second. Capitalism is the game we agree to play with one another because it's the best way we've figured out to make a good world. If you are making the world worse, whether you're getting paid or just doing it for kicks, you're still an asshole.
Saying, "Well, I got my money. Fuck 'em if they can't look out for themselves," isn't an excuse we'd accept from a drug dealer who dropped out of junior high. I don't know why you'd accept it from somebody in a suit and tie making millions.
All I can see is capitalism at its very best. Not very pretty but nothing surprising.
I'm hoping that last bit was ironic, as I completely agree with you otherwise.
I'm an energetic entrepreneur and a big fan of capitalism when it involves small and medium-sized businesses. Competition keeps everybody sharp and mostly honest. And I think it will work even better as the internet becomes more pervasive, as the internet helps reduce the information asymmetries that can lead to consumer exploitation.
Large companies, though -- especially ones large enough to be dominant in a market -- are often a parasitic scourge. Having both dealt with them extensively and worked inside them some, my belief is that they don't really practice what I'd call capitalism. Instead, they behave like medieval feudal kingdoms. Internally they're rigidly hierarchical and frequently wasteful, and externally they're often soulless monsters that crush the weak and toady to the strong. Well, that or they get fat and lazy.
In short, I'm a big fan of open markets and open competition, both in ideas and in goods -- just as long as it is truly open and power is distributed widely. Of course, SBC/AT&T knows fuck-all about that model, which is why we broke them up in the first place.
then it isn't really your realm to measure something like productivity
I disagree for two reasons.
First, I want to always be getting better at my job. I like to measure things to see if my experiments actually make things better.
Second, I want to be transparent about what and how I'm doing. If people don't know how I spend my time and why it's valuable for the company, it's easy to start suspecting that I'm not pulling my weight. I want them to trust me, and being open with my colleagues really helps that.
Of course, I work exclusively with startups and small companies these days. Larger companies have a lot more room for politicking and perverse incentives, so YMMV in applying this approach.
That's a good list. I'd add a little more, though.
Personally, I split sysadmin work up into two categories: doing something and making it so you don't have to do anything. The second is much more important, but much harder to quantify.
For the first category, you can definitely count things for managers. E.g., X accounts created, Y support requests handled. Be very careful quantifying things like this, though, or you create perverse incentives. If I make a system that's hard to use, I can receive and satisfy a lot of support requests. Or if I concentrate power rather than distributing it, then I get to look busy and important.
The other category is much trickier. Long ago I worked for a financial trading company. About 80% of the working day, the head clerk would just loiter on the trading floor, reading the paper and shooting the shit with clerks and traders. And that was exactly what his bosses wanted: they correctly saw that as a sign he kept things running smoothly. And then when problems popped up, he could give them his full attention while the rest of the operation kept running.
So I'd add two items to your list: user satisfaction, measured through surveys, and crisis preparedness, measured by speed and quality of response during drills (and actual crises, of course, but you can't wait for those to find out how ready you are).
If you insist on remaining at a large company, please follow this handy instructional brochure. I think a lot of large companies are negatively productive, so following that plan will help us all.
Up until a couple of months ago, I had gear in 365 Main. They are charging for ultra-reliable collocation, not the kind of hosting I can get off my home DSL line. Further, I only moved my gear to the other side of the block, and it has been up solid all day.
I guess they'll have to change it to 364 Main now.
Erlang/Mnesia would have been better fit for what they're doing, it was literally "made for it".
I should have known better than to listen to an AC. As a Prevayler fan, I think Mnesia's neat, but they make pretty clear that Mnesia isn't good for large amounts of data. And given that Twitter's well north of 50 million messages and growing absurdly, it's well outside of Mnesia's envelope. Sure, you could build something federated, but you could do that nicely with all sorts of things, including Ruby.
Slashdot Mirror
I can sort quasi-spam (ads from companies I've placed orders from, for example) far more effectively with body filters.
For that use case, you might try tagged addresses. Thanks to qmail and postfix, I subscribe to things with addresses like
bob-vendor.com@example.com
Where "vendor.com" is the name of the company I'm giving the address to and "example.com" is my domain. When somebody misuses or leaks an address, I just add a filter for it.
Uh, "delete, then purge" is the way IMAP works.
Yes, but that's arguably a mistake, and certainly no excuse.
MTAs are there to present user-friendly interfaces on top of protocols. For example, the way SMTP works is to require authentication every time you connect. But that's no reason for a mail client to act the same way.
People like trash cans. If that's how they think of deletion, there's no reason to train them to use some other model. Software exists for the convenience of its users, not vice versa.
No way. Thunderbird is stable, Evolution is not.
I confess it has been a few years, but that's why I walked away from Evolution. That and some fuckwad Ximian developers who were utter dicks on the mailing lists. I'm sure most of them must have been nice, but all I remember about Evolution is the sound of my teeth grinding like somebody trying to start a running car.
Thunderbird's not what I call bulletproof, but it never loses my mail, and I have come to terms with most of its quirks.
You are assuming that "trade" is automatically equitable and of benefit to both parties.
I don't have to assume it. I can prove it. Every time you buy something, you're engaging in trade. When you buy something, it's because you think you'll be better off with the thing than the money. The seller thinks that they'll be better off with the money than the thing. Sure, sometimes you're wrong. But you're usually right, or you'd stop trading.
That's all trade is. People deciding to trade one thing for another because they think it's good for them. If you think it's a bad idea, stop buying things for a few weeks.
Interestingly enough, truly "free" trade among nations invariably results in one or more parties being completely screwed by another.
Nice assertion. Got facts?
I do. If you look at the rise of global trade versus pretty much any useful metric over the last 300 years (standard of living, lifespan, health, education, wealth) they are positively correlated with trade. America is one of the most open nations to trade, and one of the most wealthy. Hong Kong's another great example. A lot of the places most resistant to trade are poor, miserable places to live.
Put it this way: free trade is like depending, for your very life, on another man's better nature.
Hardly. The guy at the corner store who sells me milk doesn't do it because he's nice. It's because he makes his living that way. The same thing for the guy delivering the milk, the ones processing it, and the ones with the cows. None of them are in it because they'd be sad to think of me eating dry cereal in the mornings.
Trade only depends on other people behaving in their own interest. That's much more reliable than somebody's good nature. Trade binds us together into networks of mutual benefit.
Of course, I don't expect to convince you, but let's move the discussion ahead a few steps. If you're going to explain why trade is bad, start off by explaining why whatever effect you're going to point at is large enough to overwhelm comparative advantage. And if you don't know what comparative advantage is and how it works, then you should probably stop now. Arguing about trade without understanding that is like arguing about physics without understanding atoms: possibly entertaining, but unlikely to be useful.
how is America's participation in the Global Economy of benefit ... to America?
Your focus on trade is a red herring. Trade makes both participants richer. Otherwise, they wouldn't trade. Artificially limiting trade makes people poorer.
If you're concerned about inequality, then focus on the causes of inequality. Get upset about deficit funding of tax cuts for the rich. Get upset about poor education. Get upset about large corporations, who take most of the value generated and give it to a few people at the top. Get upset about crony capitalism. Get upset about the way rich people and large companies buy the laws they want.
But don't get in the way of trade. Since trade makes people richer, you can take part of that extra money to help everybody else.
Do you think the factory workers, or even the management at Lenovo have anything to do with China's military decisions?
Nice rant -- if you were right. But Lenovo is 30% state-owned, so its success benefits China's gerontocracy, who are behind a lot of smashed skulls in that part of the world.
AFAIK Ennis does not sound like a paranoid lunatic to me
No, it's another known category of people I don't take very seriously: paranormal promoters.
I used to look into these things with some vigor. You never know, right? I got tired of it, as every time I got to the bottom of it, it was bunk. The problem is that it's much easier for somebody to fool themselves than it is for an outside observer to figure out how they screwed up. And that's leaving out entirely the question of fraud, which it ten times harder to prove.
Now my take is that it is incumbent upon the person with the wild result or theory to come up with the clear proof if they want me to take them seriously. Yes, this means that I won't be the first person to confirm that invisible pink unicorns really exist. I'm ok with that. If I had infinite time, I'd pay the fringe types more attention. But I've only got so long before they put me in a pine box, and I've got better things to do.
He had one of the best rails programmers there was. It wasn't a problem with not knowing rails.
No, it was a problem of trying to use Rails to work just like he did in PHP. He wanted to use his existing database layout. He wanted to write SQL queries. The Rails attitude is that your database should derive from your OO model, and you should generally trust Rails to get your data for you.
So you could sorta say he knew it. But until he tries it the way the authors meant it to be used, he still won't really get what its strong points are.
Looking at Ruby on Rails, it seems like a cheap clone of the old NeXT WebObjects
It has been a long time since I did WebObjects work, but I don't think they're particularly similar. The spirit is certainly very different.
So, why do people use Ruby?
I think there are two big crowds. One is the smart OO guys who have been suffering through Java for years. Smalltalk is not commercially viable, but Ruby is. Suddenly, they can escape all the Java idiocy. For the ones doing web stuff, and in particular the ones who have dealt with the absurdity of trying to program via large XML framework config files, Rails is a similarly big relief.
The other crowd comes from PHP and other low-rent web development. Ruby + Rails lets them get something up as quickly as in PHP, but provides them better long-term tools and something much closer to what I'd call object-oriented development. Of course, a lot of those people quickly get in over their heads in Rails, as a lot of the shoddier Rails plugins demonstrate.
So maybe it would be better if Rails were written in Smalltalk or Lisp. But Ruby it is, and it's not so bad. If Java's half-way between C++ and Lisp, then Ruby's cut the distance in half again.
Similarly, why care about the pledge of allegiance? It's one small line, and you can omit it if you wish.
You need to work on your thinking skills a little.
The whole point of the pledge of allegiance is to care about something, and to do it with everybody else. As far as I'm concerned, bringing up somebody else's imaginary god in the middle is like putting a screen door in an aquarium.
The truth is that OLPC was largely unaware of the difficulties this kind of project would face. OLPC set an unreasonable goal for the price, and now they're coming to terms with the reality of the situation. Initially OLPC had said that the market wouldn't produce an inexpensive laptop because the profits weren't there. It turns out that the market wasn't making them because it's not possible.
Evidence?
In the earliest OLPC speeches I heard, the $100 was always a goal, and they didn't expect to hit it with their first systems. They expect to get there over time.
And as you can see elsewhere in the thread, ASUS's upcoming Eee seems to be exactly the kind of inexpensive laptop that you claim is impossible. The point wasn't to hit some magic $100 goal; the point was to make a radically cheaper laptop build for the needs of third-world students rather than western consumers.
You're behind the times. The base Eee has dropped to 256 MB RAM, the storage to 2 GB, and they dropped the camera. See the Eee Wikipedia entry for more.
And what you're missing is that they only reason the Eee exists is the OLPC program. Which was partly the point. Nobody believed it was possible or reasonable when they started, as everybldy was focused on high-end laptops. Now a major manufacturer is trying to match them. That's not a problem, that's success.
But because of the circumstances, I doubt most scientists would risk their careers investigating "cold fusion" or homeopathy, or even be able to get funding to do so in the first place.
Plenty of people investigated cold fusion. Some people found something. Most people didn't. The people who found something couldn't reliably replicate their results. Research continues.
What's clear, though, is that some well-meaning people ended up doing bad science because they fell for the hype and let it influence their results. That is an entirely justified black eye for them, as guarding against that is pretty much the point of science.
I think we shouldn't dismiss her (or the entire field of homeopathy) just because of that experiment.
A homeless guy in my neighborhood is convinced that there is a sinister, far-reaching conspiracy against him, orchestrated by his estranged and abusive father. Now sure, it could be true. We probably shouldn't dismiss him or his theories just because he sounds like every other paranoid lunatic and has no proof. But let's just say I'm not rushing to investigate, either.
Some things are just provably wrong and don't need experimentation.
You realize that there's no difference between your reasoning and that of most pseudoscience followers, right?
There is no a priori knowledge. Evolution has given you a variety of biases and some mechanisms for picking up plenty more. You can either let your biases run away with you, or you can roll up your sleeves and do some science.
If you're not doing the science and you run around telling people not to do the science, you're basically cheerleading for a return to a medieval paralysis of knowledge. To which I say: fuck that! If nothing else, people were much smellier.
I own a retail store in California, and have made it my business to know the law. Here the store would win any lawsuit hands down. It wouldn't even make it to a trial; the defending attorney would quote all the case law that has already decide the issue and the judge would throw the suit out in pre-trial motions.
Don't you have to have probable cause? E.g., seeing somebody steal something? I'm pretty sure that you can't arbitrarily detain people just because they've been in your store.
If I'm wrong please cite the laws you've been reading, as I'd like to see the details.
Hey now. He didn't say Republicans.
On the one hand, this is great. The more a law enforcement officer can get done with their time, the better. Plenty of crime goes unaddressed because it is "too small". The FBI, for example, won't talk to you about interstate computer crime unless you can prove a minimum of $10k of losses. And because they're busy, the effective threshold is much higher.
On the other hand, the US government has recently been a little cavalier about my rights, and there are historical periods where they've been a fair bit worse, like the Second Red Scare. It's enough to make me want to get a bunch of disposable prepaid cellphones, just in case.
I'd feel a lot better if there were some rule about public posting of eavesdropping records. E.g., that within 10 years after any eavesdropping incident, the government is obliged to publish who ordered it, why, and who accessed any data from it. As Brin points out in The Transparent Society, the problem isn't so much surveillance as secret, one-way surveillance.
You used eleven and a half words to communicate a concept that could have been done with two: being reactive and proactive. I know the latter has been sullied by its use in jargon bingo, but we are still capable of understanding the concept.
When I talk to MBAs, I use words like "reactive" and "proactive". But here I wouldn't because a) people would think I'm some fuckwad MBA and ignore me, and b) those who have just heard the jargon terms but haven't thought about what they really mean would have skated on by my message.
You know the terms and their real meaning, so you thought about what I said and sorted it out. Fine, have a cookie. For everybody else, I wanted them to do similar thinking whether or not they really got the terms.
And yes, of course people are capable of understanding the concept. That's why I focused on the concept, rather than on some popular labels for the concept. So thanks for your reactive copyediting, but maybe you could run along and proactively contribute to the discussion?
No. They did what companies do, make money for their shareholders.
Yes. "Ve vere just following orders, mein herr." Always a perfectly good excuse.
Wait, no, I forgot. That isn't a good excuse.
Because we are all citizens and human beings first, and employees second. Capitalism is the game we agree to play with one another because it's the best way we've figured out to make a good world. If you are making the world worse, whether you're getting paid or just doing it for kicks, you're still an asshole.
Saying, "Well, I got my money. Fuck 'em if they can't look out for themselves," isn't an excuse we'd accept from a drug dealer who dropped out of junior high. I don't know why you'd accept it from somebody in a suit and tie making millions.
All I can see is capitalism at its very best. Not very pretty but nothing surprising.
I'm hoping that last bit was ironic, as I completely agree with you otherwise.
I'm an energetic entrepreneur and a big fan of capitalism when it involves small and medium-sized businesses. Competition keeps everybody sharp and mostly honest. And I think it will work even better as the internet becomes more pervasive, as the internet helps reduce the information asymmetries that can lead to consumer exploitation.
Large companies, though -- especially ones large enough to be dominant in a market -- are often a parasitic scourge. Having both dealt with them extensively and worked inside them some, my belief is that they don't really practice what I'd call capitalism. Instead, they behave like medieval feudal kingdoms. Internally they're rigidly hierarchical and frequently wasteful, and externally they're often soulless monsters that crush the weak and toady to the strong. Well, that or they get fat and lazy.
In short, I'm a big fan of open markets and open competition, both in ideas and in goods -- just as long as it is truly open and power is distributed widely. Of course, SBC/AT&T knows fuck-all about that model, which is why we broke them up in the first place.
then it isn't really your realm to measure something like productivity
I disagree for two reasons.
First, I want to always be getting better at my job. I like to measure things to see if my experiments actually make things better.
Second, I want to be transparent about what and how I'm doing. If people don't know how I spend my time and why it's valuable for the company, it's easy to start suspecting that I'm not pulling my weight. I want them to trust me, and being open with my colleagues really helps that.
Of course, I work exclusively with startups and small companies these days. Larger companies have a lot more room for politicking and perverse incentives, so YMMV in applying this approach.
That's a good list. I'd add a little more, though.
Personally, I split sysadmin work up into two categories: doing something and making it so you don't have to do anything. The second is much more important, but much harder to quantify.
For the first category, you can definitely count things for managers. E.g., X accounts created, Y support requests handled. Be very careful quantifying things like this, though, or you create perverse incentives. If I make a system that's hard to use, I can receive and satisfy a lot of support requests. Or if I concentrate power rather than distributing it, then I get to look busy and important.
The other category is much trickier. Long ago I worked for a financial trading company. About 80% of the working day, the head clerk would just loiter on the trading floor, reading the paper and shooting the shit with clerks and traders. And that was exactly what his bosses wanted: they correctly saw that as a sign he kept things running smoothly. And then when problems popped up, he could give them his full attention while the rest of the operation kept running.
So I'd add two items to your list: user satisfaction, measured through surveys, and crisis preparedness, measured by speed and quality of response during drills (and actual crises, of course, but you can't wait for those to find out how ready you are).
How many companies are really driven by passion?
Most of the small ones. Go join or start one.
If you insist on remaining at a large company, please follow this handy instructional brochure. I think a lot of large companies are negatively productive, so following that plan will help us all.
An electrical utility failure is no excuse.
Up until a couple of months ago, I had gear in 365 Main. They are charging for ultra-reliable collocation, not the kind of hosting I can get off my home DSL line. Further, I only moved my gear to the other side of the block, and it has been up solid all day.
I guess they'll have to change it to 364 Main now.
Erlang/Mnesia would have been better fit for what they're doing, it was literally "made for it".
I should have known better than to listen to an AC. As a Prevayler fan, I think Mnesia's neat, but they make pretty clear that Mnesia isn't good for large amounts of data. And given that Twitter's well north of 50 million messages and growing absurdly, it's well outside of Mnesia's envelope. Sure, you could build something federated, but you could do that nicely with all sorts of things, including Ruby.