>> Sending the message that "good" engineers are the ones who'll stay all night is
>> exactly what keeps people who value life balance out of fields like engineering
>The word you're looking for is "capitalism".
No, the word is "stupid".
Focus and intensity - keeping on task, not reading slashdot when you should work - these are what increase your output. Lots of extra hours at your desk per week isn't it - unless you're charging by the hour and your client will pay for 12 hour days, in which case your client is the stupid one. There are lots of studies on this stuff: if you want to be a capitalist I'd recommend you hit the books and read some.
I'm a successful capitalist: a partner in a little firm that builds systems for financial derivatives, which pays very nicely. We take 8 weeks leave a year, we don't work nights or weekends (except occasionally, if some interesting problem really needs solving or some small crisis has blown up). We are very happy, we expect to retain our staff well, and we are very, very productive because when we work, we work.
Working a lot of late nights means a failure in project management - any senior manager who finds his staff are doing that should be very concerned and starting seeking the scalp of whatever middle-manager is at fault.
A lot of people are talking a lot of shite about how different men and women are.
Individuals vary a lot. Gender differences are small.
Yes, when you look at large groups there are statistically significant differences in average risk-taking behaviour. But the averages aren't very different and it's a big bell curve. That means that about 45% of women take more risks than the average bloke, and about 55% of women take less risks.
As for the claim that this is chromosones instead of culture: if so how come the percentage of women who are top mathematicians has quadrupled in the last 30 years, and how come the risk-taking differences vary so wildly when psychologists repeat the experiments cross-culturally?
And please spare us the ill-informed evolutionary just-so stories - or go climb the wall of a zoo-cage and tell a lioness that she's "not aggressive outside her group".
Raise the topic in conversation sometime, over by the coffee machine.
See what they say.
I'm serious. You Yanks don't notice how bad the INS is: it'd do you good.
As a foreign grad student and foreign worker hanging out with others in the same boat, chatter about the nastiness of the INS was really, really common. Of course, if your company is that big you may well employ someone whose full-time job is dealing with the INS on behalf of your workers (I'm not kidding, I knew someone whose mum had the job for a chemicals company). If so your workers will be doing better not because the INS is competent but because you're paying someone full-time to understand and pander to its incompetence.
"I'm also sure that there are lots and lots of people who enter America legally as visitors, workers, students, and so on."
Well yes. That's what I did. I didn't say the INS failed to function: my claim is that it's really unpleasant to deal with.
"My mother is a Canadian citizen and was a resident alien for 25 years"
I'm happy to believe that someone who has been married to an American for decades, has lived in America for decades, is from the country that is your closest neighbour and best buddy, and has a son who is an American, is *probably* not hassled by the INS much.
But you're hardly talking the case of the typical foreigner there.
So let's modify his claim: you Americans treats almost all non-citizens like shit.
If you can get even one tenth of the "tons" of foreign nationals in your dept to say anything good at all about your INS I'll be gobsmacked. After I'd lived in the USA for 4 years I was sick of my wife and I being dumped on by the govt all the time and we left.
USA: Good, mostly. US govt's INS: Bad, completely.
If Cyc doesn't get that a chrismas tree is a tree, then it becomes much less useful: it can't infer things like "a christmas tree has branches" from "a christmas tree is tree".
If it does get that a christmas tree is a tree, then it will infer things that might be wrong like "needs air and water" or even "is a plant".
By having a simplistic ontology is either infers too little, or too much.
The real problem is that Cyc is built on clear-cut ontologies: sets, supersets, subsets.
But 'natural' ontologies are not clear-cut, they are very messy: an oak tree is a more typical tree than a christmas tree, a cabbage tree, a family tree or a binary tree. They are still trees of a type: they still share features with trees, you can still make some valid inferences about them that transfer from them being trees. But they're not trees in quite the same way.
(Philosophers waste thousands of pages on whether 'natural' ontologies are 'natural' in the world or just 'natural' in our heads or just 'natural' in our language. But for judging Cyc all we need to say is that natural ontologies are bloody useful.)
No, the problem is contextualization. The solution is something CYC doesn't come close to.
Your "vampire" example is a typical AI researcher's example: it's too trivial to show the real problem. That's because with "vampire" you can get the context of "fiction" from the word. So let's take a more typical word: "tree"
Basic ontology: A tree is a plant.
Basic fact: A plant requires air and water to live.
Have you watered your red-black binary tree today? How about your boxed christmas tree? Your family tree? Your oak tree?
You can solve this by saying that only *some* trees are living things. But then you lose the power CYC's ontology-and-logic combination was supposed to give you because you can no longer reach userful conclusions based on the fact that this palm tree is a tree.
Or you can solve it by deciding the problem is english and its foolish use of the word 'tree' for several different things, so you invent your own words tree(1) tree(2) tree(3), etc. But that just moves the problem of understanding the world out of your system and onto your users. Your clear logical rules and ontology become an unmaintainably complex hodge-podge of exceptions. And worse, it misses the fact that all these trees really are trees even if they're not real trees. It's not an accident that we call a christmas tree a tree.
I'm not aware of any existant video of a live adult giant squid. If Greenpeace can get any that would be a major coup. But it's gonna be bloody hard.
It's also gonna be boring. The video we have of juvenile Giant Squid shows that they are passive drifters that just hang out.
Now, heading down to the sub-antarctic waters south of New Zealand to catch some Collosal Squid on video - that'd be much better! Typical adult Collosal Squid are about 18 meters long, they've got these cool swivelling hooks on their tentacles, they're aggressive hunters that maul sperm whales. Best of all, they're pretty much transparent while alive so they're almost completely invisible when hunting.
They're not quite invisible. The eyes and beak aren't transparent, and the rest of the body will reflect some light like a translucent jellyfish does.
(PS: Bizzare Squid Fact #28: squid are very stretchy. Which is why you get various different sizes given for how big they grow - non-scientists tend to pull their tentacles out as long as they can and measure that).
Another company that hired only A-level talent, robustly avoided B-level talent, ran strong internal competitions to try and attract other employees onto your star project, and talked a lot about "darwinian" processes running between internal projects was:
Enron
Like Google now, Enron back in the day had management consultants writing magazine articles about the wonders of their "fluid" structure, the way petty beaurocrats were kept out of people's way, and their hiring practices. It was The Way Of The Future. Enron was the best, was going to take over and Rule Supreme. Like Google, Enron was proud that it didn't just keep to one boring idea of what they did, the company could perpetually reinvented itself.
Those standard management structures exist for a reason. If google finds a way to work without them long-term, then good for them. But it's harder than it might seem.
We don't really have a crisis yet, now do we, let's be sensible here, you're doing too much modelling on not enough data, let's gather more evidence implementing those measures have real economic costs.
OR
It's too late now, such small baby steps that take decades to take affects cannot save us, we're all doomed.
Most posters on Slashdot threads have moved from "too early" to "too late" on global warming in the last year. Sigh.
Physical distribution of your systems carries security and political/legal risk.
For example: Chavez nationalizes the 8 Google containers in his country.
Or: The Chinese Committee For Political Stability nationalize the 100 Google containers in their country.
OR: The US's quaint anti-monopoly laws, built to stop railway barons, are ideally designed to stop monopolies that provide a federal service and break them up into several companies each serving only a a few states.
Microsoft isn't like that: their distribution presence is largely virtual, not physical. The Chinese govt can't nationalize Microsoft China and end up with a usable software company.
Mate,
You're making clear that you don't use, or know much about, Java.
Unlike, say, C++, choice of JVM and JRE is _not_ a mess of compatibility issues. It's backwards compatible and cross-platform compatible as all hell. Y
Your claim that 'different JREs support different features' is just surreal. I've never
seen any other language that is as good at backwards compatibility as Java (certainly not C++).
Of course new features from Java 5.0 don't exist if someone runs in a JRE1.2. But the only way around that would be to freeze the language and never add a new feature ever.
I spend a painful amount of time dealing with C++ issues with different compilers, I miss the ease of Java development.
The doomsayers cry out that the end is nigh, but so far humans have adapted remarkably well to changing climactic conditions.
Speaking as someone on a pacific island: humanity would survive a 5m rise in sea level just fine, but for everyone around me, and a billion or so other coast-dwellers world wide, it will suck.
Your argument that global warming isn't a human species extinction event is hardly an argument against those who are saying it's bad.
And your argument that we shouldn't act until we've studied it to death is just silly. There's a balance between acting in complete ignorance and doing nothing until you've studied something to death. You don't seem to be willing to accept that.
Every single analysis of 'sustainability' by the folks preaching doom and gloom over the Earth's carrying capacity assumes that *technology will never advance beyond what we have now*.
Untrue. Most explicitly state that they're assuming future technological change will not change the amount reduce resource use. But they usually make clear that is an OPTIMISTIC assumption, not a PESSIMISTIC one.
Because Tech change almost always leads to MORE resource use, not LESS. For example: horses used a lot of resources, but cars use a lot more. If we just did with cars what we used to do with horses (which were too expensive for most city dwellers to own, and didn't go more than a few miles a day) then cars would use less resources. But cars enable us to move more people further, so we do - and we eat up more resources doing it.
For you to be right that tech changes will reduce, not increase, resource use you have to assume that people will start saying: "Here's a fantastic new technology! But lets not use it to do cool and awesome new things, lets just use it do what the old technology did!"
But hey, maybe suddenly all people will spontaneously decide that they'll only use environmentally friendly technologies from now on. And given some Genetic Engineering maybe the world will fill with flying pigs, too.
You're attacking a straw man by by claiming people "keep saying this is just as good as Hubble at 1/10 the cost".
What they're saying is that this is BETTER THAN Hubble at MANY THINGS at 1/10th the cost.
Not better at everything (UV absorbtion by the atmosphere, for example, is just a killer for ground-based UV on some wavelengths). But ten times the resolution of Hubble is nothing to sneeze at.
And the people saying this are responding to a naive horde who keep saying "but why not spend money on Hubble instead coz its out of the atmosphere". Hubble does have unique advantages, and being is space is very cool, but given its horrificly large cost its very hard to justify Hubble in terms of bang per buck.
For simple (ie largely single-threaded) apps the dual core chips will reduce further the speed difference between app and C and C++ and apps written in Java (or other high-level languages).
Because automated garbage collection is the big performance difference: and for modern runtime environments automated garbage collection runs in a parallel thread.
Two (completely different) reasons to have a design doc:
1) As part of the process of figuring out what to build.
2) To document what we have built.
If (1) then it's a part of the build process - you write it to get your ideas concrete for yourself and so others can review them, you may mention options you aren't taking and why, the results are fluid.
If (2) then you'll want references to the code that you've built to give examples, and also pointers of to key bits of code and a mention of what bits ended up more complex than expected and why, what bits of code to read and pay attention to, etc.
Many bad processes try to conflate these two documents. That wastes a lot of developer hours and results in crappy documentation.
If your aim is (1) then don't pretend you're producing good documentation for what the final system is like. Also realise that anything written before you coded and tested will be inaccurate by the time you go live because plans shift and change.
If your aim is (2) then you can write it much faster and better after you've finished coding.
Do you believe that the earth goes round the sun instead of the other way around? Do you believe that Einstein's theory of relativity is true? Do you believe in evolutionary theory in biology? Do you believe the DNA is what carries human genetic material?
Have you done the experiments or analysis to prove any of these?
Or do you believe them "simply because it's the consensus view among scientists"?
The fact that some unicode characters look exactly like ascii (or whitespace) also means you can break java source code (java bytecode accepts unicode) by replacing a few dozen 'a's with 'a's. It can have developers weeping with frustration trying to figure out why the code isn't working.
Slashdot Mirror
Focus and intensity - keeping on task, not reading slashdot when you should work - these are what increase your output. Lots of extra hours at your desk per week isn't it - unless you're charging by the hour and your client will pay for 12 hour days, in which case your client is the stupid one. There are lots of studies on this stuff: if you want to be a capitalist I'd recommend you hit the books and read some.
I'm a successful capitalist: a partner in a little firm that builds systems for financial derivatives, which pays very nicely. We take 8 weeks leave a year, we don't work nights or weekends (except occasionally, if some interesting problem really needs solving or some small crisis has blown up). We are very happy, we expect to retain our staff well, and we are very, very productive because when we work, we work.
Working a lot of late nights means a failure in project management - any senior manager who finds his staff are doing that should be very concerned and starting seeking the scalp of whatever middle-manager is at fault.
A lot of people are talking a lot of shite about how different men and women are.
Individuals vary a lot. Gender differences are small.
Yes, when you look at large groups there are statistically significant differences in average risk-taking behaviour. But the averages aren't very different and it's a big bell curve. That means that about 45% of women take more risks than the average bloke, and about 55% of women take less risks.
As for the claim that this is chromosones instead of culture: if so how come the percentage of women who are top mathematicians has quadrupled in the last 30 years, and how come the risk-taking differences vary so wildly when psychologists repeat the experiments cross-culturally?
And please spare us the ill-informed evolutionary just-so stories - or go climb the wall of a zoo-cage and tell a lioness that she's "not aggressive outside her group".
Raise the topic in conversation sometime, over by the coffee machine.
See what they say.
I'm serious. You Yanks don't notice how bad the INS is: it'd do you good.
As a foreign grad student and foreign worker hanging out with others in the same boat, chatter about the nastiness of the INS was really, really common. Of course, if your company is that big you may well employ someone whose full-time job is dealing with the INS on behalf of your workers (I'm not kidding, I knew someone whose mum had the job for a chemicals company). If so your workers will be doing better not because the INS is competent but because you're paying someone full-time to understand and pander to its incompetence.
"I'm also sure that there are lots and lots of people who enter America legally as visitors, workers, students, and so on."
Well yes. That's what I did. I didn't say the INS failed to function: my claim is that it's really unpleasant to deal with.
"My mother is a Canadian citizen and was a resident alien for 25 years"
I'm happy to believe that someone who has been married to an American for decades, has lived in America for decades, is from the country that is your closest neighbour and best buddy, and has a son who is an American, is *probably* not hassled by the INS much.
But you're hardly talking the case of the typical foreigner there.
So let's modify his claim: you Americans treats almost all non-citizens like shit.
If you can get even one tenth of the "tons" of foreign nationals in your dept to say anything good at all about your INS I'll be gobsmacked. After I'd lived in the USA for 4 years I was sick of my wife and I being dumped on by the govt all the time and we left.
USA: Good, mostly.
US govt's INS: Bad, completely.
Your claims about water vapour being ignored in current climate models are 100% false.
We have quite a lot of people doing a lot of work to measure and model both its insulative and reflective effects.
" it is so poorly understood that it can not be included "
Untrue. Completely untrue. If you leave water vapour out the resulting model is gibberish - you get freezing temperatures in New York in July.
If Cyc doesn't get that a chrismas tree is a tree, then it becomes much less useful: it can't infer things like "a christmas tree has branches" from "a christmas tree is tree".
If it does get that a christmas tree is a tree, then it will infer things that might be wrong like "needs air and water" or even "is a plant".
By having a simplistic ontology is either infers too little, or too much.
The real problem is that Cyc is built on clear-cut ontologies: sets, supersets, subsets.
But 'natural' ontologies are not clear-cut, they are very messy: an oak tree is a more typical tree than a christmas tree, a cabbage tree, a family tree or a binary tree. They are still trees of a type: they still share features with trees, you can still make some valid inferences about them that transfer from them being trees. But they're not trees in quite the same way.
(Philosophers waste thousands of pages on whether 'natural' ontologies are 'natural' in the world or just 'natural' in our heads or just 'natural' in our language. But for judging Cyc all we need to say is that natural ontologies are bloody useful.)
The solution is contextualization.
No, the problem is contextualization. The solution is something CYC doesn't come close to.
Your "vampire" example is a typical AI researcher's example: it's too trivial to show the real problem. That's because with "vampire" you can get the context of "fiction" from the word. So let's take a more typical word: "tree"
Basic ontology: A tree is a plant.
Basic fact: A plant requires air and water to live.
Have you watered your red-black binary tree today? How about your boxed christmas tree? Your family tree? Your oak tree?
You can solve this by saying that only *some* trees are living things. But then you lose the power CYC's ontology-and-logic combination was supposed to give you because you can no longer reach userful conclusions based on the fact that this palm tree is a tree.
Or you can solve it by deciding the problem is english and its foolish use of the word 'tree' for several different things, so you invent your own words tree(1) tree(2) tree(3), etc. But that just moves the problem of understanding the world out of your system and onto your users. Your clear logical rules and ontology become an unmaintainably complex hodge-podge of exceptions. And worse, it misses the fact that all these trees really are trees even if they're not real trees. It's not an accident that we call a christmas tree a tree.
> Welcome to Oceania.
I believe some computer error has garbled your message. Clearly you meant to say:
"It is doubleplus good that we have always lived in Oceania."
"I work with C++ every day and I love it because of the sheer power it gives me. "
I work with C++ every day and I hate it because of the sheer power it gave the idiots who wrote the code that I maintain.
Before someone complains that the squid in the link is very visible: they acquire colour when dead.
I'm not aware of any existant video of a live adult giant squid. If Greenpeace can get any that would be a major coup. But it's gonna be bloody hard.
It's also gonna be boring. The video we have of juvenile Giant Squid shows that they are passive drifters that just hang out.
Now, heading down to the sub-antarctic waters south of New Zealand to catch some Collosal Squid on video - that'd be much better! Typical adult Collosal Squid are about 18 meters long, they've got these cool swivelling hooks on their tentacles, they're aggressive hunters that maul sperm whales. Best of all, they're pretty much transparent while alive so they're almost completely invisible when hunting.
They're not quite invisible. The eyes and beak aren't transparent, and the rest of the body will reflect some light like a translucent jellyfish does.
(PS: Bizzare Squid Fact #28: squid are very stretchy. Which is why you get various different sizes given for how big they grow - non-scientists tend to pull their tentacles out as long as they can and measure that).
Another company that hired only A-level talent, robustly avoided B-level talent, ran strong internal competitions to try and attract other employees onto your star project, and talked a lot about "darwinian" processes running between internal projects was: Enron
Like Google now, Enron back in the day had management consultants writing magazine articles about the wonders of their "fluid" structure, the way petty beaurocrats were kept out of people's way, and their hiring practices. It was The Way Of The Future. Enron was the best, was going to take over and Rule Supreme. Like Google, Enron was proud that it didn't just keep to one boring idea of what they did, the company could perpetually reinvented itself.
Those standard management structures exist for a reason. If google finds a way to work without them long-term, then good for them. But it's harder than it might seem.
Either:
We don't really have a crisis yet, now do we, let's be sensible here, you're doing too much modelling on not enough data, let's gather more evidence implementing those measures have real economic costs.
OR
It's too late now, such small baby steps that take decades to take affects cannot save us, we're all doomed.
Most posters on Slashdot threads have moved from "too early" to "too late" on global warming in the last year. Sigh.
Physical distribution of your systems carries security and political/legal risk.
For example:
Chavez nationalizes the 8 Google containers in his country.
Or:
The Chinese Committee For Political Stability nationalize the 100 Google containers in their country.
OR:
The US's quaint anti-monopoly laws, built to stop railway barons, are ideally designed to stop monopolies that provide a federal service and break them up into several companies each serving only a a few states.
Microsoft isn't like that: their distribution presence is largely virtual, not physical. The Chinese govt can't nationalize Microsoft China and end up with a usable software company.
Mate, You're making clear that you don't use, or know much about, Java. Unlike, say, C++, choice of JVM and JRE is _not_ a mess of compatibility issues. It's backwards compatible and cross-platform compatible as all hell. Y Your claim that 'different JREs support different features' is just surreal. I've never seen any other language that is as good at backwards compatibility as Java (certainly not C++). Of course new features from Java 5.0 don't exist if someone runs in a JRE1.2. But the only way around that would be to freeze the language and never add a new feature ever. I spend a painful amount of time dealing with C++ issues with different compilers, I miss the ease of Java development.
dead right re memory space.
the copy collector isn't new (it's been around since 2002). But the big problem for modern JVMs is memory. Paging sucks.
Speaking as someone on a pacific island: humanity would survive a 5m rise in sea level just fine, but for everyone around me, and a billion or so other coast-dwellers world wide, it will suck.
Your argument that global warming isn't a human species extinction event is hardly an argument against those who are saying it's bad.
And your argument that we shouldn't act until we've studied it to death is just silly. There's a balance between acting in complete ignorance and doing nothing until you've studied something to death. You don't seem to be willing to accept that.
Every single analysis of 'sustainability' by the folks preaching doom and gloom over the Earth's carrying capacity assumes that *technology will never advance beyond what we have now*.
Untrue. Most explicitly state that they're assuming future technological change will not change the amount reduce resource use. But they usually make clear that is an OPTIMISTIC assumption, not a PESSIMISTIC one.Because Tech change almost always leads to MORE resource use, not LESS. For example: horses used a lot of resources, but cars use a lot more. If we just did with cars what we used to do with horses (which were too expensive for most city dwellers to own, and didn't go more than a few miles a day) then cars would use less resources. But cars enable us to move more people further, so we do - and we eat up more resources doing it.
For you to be right that tech changes will reduce, not increase, resource use you have to assume that people will start saying: "Here's a fantastic new technology! But lets not use it to do cool and awesome new things, lets just use it do what the old technology did!"
But hey, maybe suddenly all people will spontaneously decide that they'll only use environmentally friendly technologies from now on. And given some Genetic Engineering maybe the world will fill with flying pigs, too.
You're attacking a straw man by by claiming people "keep saying this is just as good as Hubble at 1/10 the cost".
What they're saying is that this is BETTER THAN Hubble at MANY THINGS at 1/10th the cost.
Not better at everything (UV absorbtion by the atmosphere, for example, is just a killer for ground-based UV on some wavelengths). But ten times the resolution of Hubble is nothing to sneeze at.
And the people saying this are responding to a naive horde who keep saying "but why not spend money on Hubble instead coz its out of the atmosphere". Hubble does have unique advantages, and being is space is very cool, but given its horrificly large cost its very hard to justify Hubble in terms of bang per buck.
Take off and nuke the site from orbit.
It's the only way to be sure.
For simple (ie largely single-threaded) apps the dual core chips will reduce further the speed difference between app and C and C++ and apps written in Java (or other high-level languages).
Because automated garbage collection is the big performance difference: and for modern runtime environments automated garbage collection runs in a parallel thread.
TANJ
Two (completely different) reasons to have a design doc:
1) As part of the process of figuring out what to build.
2) To document what we have built.
If (1) then it's a part of the build process - you write it to get your ideas concrete for yourself and so others can review them, you may mention options you aren't taking and why, the results are fluid.
If (2) then you'll want references to the code that you've built to give examples, and also pointers of to key bits of code and a mention of what bits ended up more complex than expected and why, what bits of code to read and pay attention to, etc.
Many bad processes try to conflate these two documents. That wastes a lot of developer hours and results in crappy documentation.
If your aim is (1) then don't pretend you're producing good documentation for what the final system is like. Also realise that anything written before you coded and tested will be inaccurate by the time you go live because plans shift and change.
If your aim is (2) then you can write it much faster and better after you've finished coding.
Sean
Aaron,
Do you believe that the earth goes round the sun instead of the other way around? Do you believe that Einstein's theory of relativity is true? Do you believe in evolutionary theory in biology? Do you believe the DNA is what carries human genetic material?
Have you done the experiments or analysis to prove any of these?
Or do you believe them "simply because it's the consensus view among scientists"?
Sean
The fact that some unicode characters look exactly like ascii (or whitespace) also means you can break java source code (java bytecode accepts unicode) by replacing a few dozen 'a's with 'a's. It can have developers weeping with frustration trying to figure out why the code isn't working.
But I wouldn't do that. That would be wrong.
Sean