My impression is that the problem never was the scheduler. It's the memory management. Load a linux app. Stop using it for a few minutes in a system that's been running for a while. Linux will page out part of the app, replacing it with crap as meaningless as cached web pages or cached files.
This is why, if you have more than a gig of ram, you should not use a swapfile under linux or a pagefile under windows. Shut them off, disable them, set the size to zero, whatever. There is absolutely no sane reason under the sun to use them once you have enough main memory for the applications you use.
With a gig of ram, you'll still be providing an enormous amount of space for the OS to cache files and content, and you won't ever have to wait for a program to page back into memory.
Photoshop 7 used to complain that I should have a pagefile and that I was making some sort of dramatic mistake. Photoshop CS2 and CS3 don't complain any more. Looks like they got with the times.
You still missed it. The other jobs don't exist. They're already fully marginalized.
For someone to commit suicide in the face of their farm not being profitable says more about their whininess than about a fundamental injustice.
You didn't quite follow what "unprofitability" means to these people. They are one famine away from starvation at the best of times. An unprofitable farm means the death of themselves and their extended family.
But I'm thrilled that you're even thinking about the situation of the desperately poor in other places. It gives me (a little) hope for the future. Work on that ignorance. Become a world traveler, and I guarantee that your perspective on many issues will change.
Ok I'll bite. Why exactly? What makes a farmer any different than a butcher, doctor, or mutual fund investor? I don't know who made my computer, who made my car, or who picked my tomatoes. I don't really see why I should, the information is not relevant to the quality of the good.
Our current food practices are unsustainable. And what I mean by "unsustainable" is that we can not continue with our current practices for much longer without dramatic and catastrophic failure of the system of food production. For example, the amount of fuel energy (whether fossil or not) required to deliver one calorie of food energy to your table is usually on the order of hundreds or thousands to one. As in, for most US households, fossil fuels containing the same energy as two gallons of gasoline were consumed for each plate at the dinner table.
Caring about where you get your food means valuing things that agri-business wants to pretend are irrelevant (because they make their money only when those issues are irrelevant). Agri-business values transportability, shelf-life, year-round availability, and shelf appearance over flavor, nutritional value, clean water, and living oceans. Farms used to be able to use animal waste as fertilizer and a hedge against erosion. Modern farms buy nitrogen derived from fossil fuels while CAFO's (concentrated animal feeding operations) produce unending streams of nitrogen, phosphorus, and antibiotic contaminated poisonous sludge that can't be used as a fertilizer and must be buried as toxic waste by people wearing moon suits.
WTF? Committing suicide because you can't make a profit as a farmer? Sorry, that's just stupid. Get another job!
Your statement represents an absolutely breathtaking ignorance. I strongly recommend you travel to India to correct it. Go to Goa during the high season (October-January), as you'll find it's friendly to westerners and the culture shock will be greatly lessened.
Now, once you've settled in and are feeling somewhat comfortable, hire a cab and head out away from the "wealthy" spice farms of Goa. Just head east for about two hours. When you see a temple in good repair, stop, stretch your legs, make an offering. Walk for a bit and take a long look around you. Pay close attention to what you see.
Head back to the city, stop where some road work is being done. Get out and talk to the road workers about the tarp and cookfire at the side of the road. Find out how many family members live under that tarp. Find out how many family members are doing back-breaking labor on the road construction project. Find out the age of the youngest family member doing road construction at that site. The oldest.
In the next traffic jam, try to figure out the job of all of the people you see walking on and near the street. Your goal, should you choose to accept it, is to figure out how many of them are beggars.
India is a paradox contained in a country. There are enormously more unskilled people than unskilled jobs. Even the measures are so dramatically different as to defy comparison: the poverty level in India is just barely this side of survival. Telling a failing farmer to just "get a job" without knowing the first thing about what kind of jobs exist (and how many) is spectacularly ignorant.
American consumers also suffer from the "OMG I *must* be a farmer" mentality: it got us the blight known as HFCS and ethanol subsidies.
Your causation is mistaken. Corn subsidies got us HFCS and processed food products that slowly poison us. Government policy intended to guarantee a surplus of commodity foods, even in lean times, is what got us the corn/soy subsidies.
The romance of farming is largely gone in the US. Personally, I believe that we would be better off as a culture if we knew who had grown or raised our food items, and the restoration of the local farmer to significance (and the near elimination of the food processor) would be a great benefit to all of us as a nation, a state, and a people.
And yet when someone misprints an ad, they are often liable to provide the goods for the advertised price, even when the price is clearly in error.
I don't see it so cut and dry as you do. I think it's just a simple case of seller beware. If you enter into an agreement where you could have gotten a better deal, it is not always the state's role to allow you to correct the situation post hoc. Not to say that there aren't laws to do exactly that, but IMHO, most are bad laws.
On the earth, it basically impossible to stay in the air for more than half a second. An athlete's excellent standing jump imparts about 1kJ (100kg elevated by 1m) to his body. A jump on the moon with the same force will last about three seconds and put you six meters off the moon's surface. During those three seconds and twelve meters of vertical travel, there is a lot more opportunity to not land on your feet. Landing on your helmet on the moon is a very, very risky proposition.
Which is what I meant by "a possibly more dangerous height".
A pound is a unit of force, but the pound is commonly used as a mass unit (would be a technically "lb-mass") because of it's familiarity to US readers. Everyone in the US then knows what you mean, and when you need to be precise about your units, it's not very hard to remove the acceleration component.
The inverse happens with the kilo in the metric system. You could convert to newtons whenever you need force, but kilogram-force will be easier for most metric-thinking people to relate to. Basically, when communicating with lay-people, scientists and engineers have learned to be flexible.
Didn't the extra mass come in handy to keep people from flying away?
Actually, you can't get to escape velocity on the moon using your feet. You can jump to a possibly more dangerous height without the extra mass/weight, but you'll quickly learn not to.
Aforementioned entry-level disks certainly do have the property of failing frequently
According to Google, consumer grade hard drives fail at the same rates and with the same predictability as enterprise grade drives.
Outside of a few specific cases (death star's, for instance), buying a bunch of the same disks will work just fine, and going to great lengths to get drives from different manufacturers and manufacturing lots doesn't seem to be worth the effort.
according to Murphy they do so in the worst possible moment you can imagine, a RAID rebuild.
And this is why you should still have a backup of your RAID. The ability of a RAID to tolerate one failure is at least partially offset by the fact that there are more parts that can fail.
I use LiPos on my RC helicopters. If the battery is broken or punctured, they tend to start a fire...
Actually, I think you've got it a bit backwards. Puncturing a lipo is how you deactivate them when you want to throw them into a landfill (though you really should recycle them instead of tossing them). That's completely safe as long as the cell isn't fully charged, and even then, it's not that dangerous.
When a lipo cell has an overcurrent failure, part of the cell vaporizes (puffed cell failure). In particularly bad failures after heavy use (low voltage, abusive current loads, etc.), the vapors can ignite from the heat of the cell. When this happens, the cell wall is guaranteed to fail, and then you get the green flames of death.
But just poking a hole in your lipo won't start a fire. I also fly RC helis (T-Rex 450).
A grad, or gradian, is 1/400 of a full rotation and is nearly useless for angles since it's not evenly divisible by 3, 6, 9, or 12, with 12 being the most important divisor in that list (even division by 12 means that you can represent all of the angles in a bisected equilateral triangle (30-60-90 triangle) using integer values).
The grad first appeared in France, and was only ever adopted in a few countries for specialized tasks (artillery, surveying) and except for French artillery, appears to be a dead unit.
It was believed that an angle unit with right angles of 100 units would make mental math easier (if you're facing 115 grad clockwise from due North, then you're 15 grad clockwise from due East). However, if you're frequently doing navigation or other cartographic-related tasks, it turns out that you quickly adapt to the mental math required to understand 90 degree right angles and angle division is much easier.
WTF? If you don't have requirements, then what the hell do you expect to test?
You misread something. I never said I didn't want requirements. I just don't think that big specs are even slightly effective ways to communicate requirements.
For instance: on my current project, we use a wiki combined with an issue tracking system (http://trac.edgewall.com/) to articulate and discuss requirements. Usually (always?) the developers need to refine the requirements as the features are developed; removing inconsistencies, adding details, etc. and since it's a wiki, they can do that fairly easily.
And what about tests as proxies for detailed requirements? We try to create automated tests for everything. Unit tests (a few), functional tests (a lot), integration tests (a lot), load/scalability tests (a few). Contrary to XP, we generally find the best tests are written along with the interfaces and implementation, not before. On our team, QA is responsible for challenging developers on test adequacy. Do the tests really exercise the code? (QA also helps to create automated UI tests, which they base off of the trac wiki, usually after primary development is complete.)
But I find it almost as funny that agile developers believe their process to be somehow novel.
That's an interesting but flawed observation. What's novel about agile processes isn't the process. Iterative processes are "waterfalls writ small" and have been useful for many, many years. What's novel is the perspective that, at the end of the day, agile teams find that skilled people are more important than devotion to any particular process.
As a final note, the overlapping of the phases results (as is happening in a RUP-managed project I'm developing-for at the moment) in previously-accomplished work being changed later, after more work has been done than would've been in the waterfall process, because of a failure to think through the project's needs thoroughly enough.
I see a part of the problem right here. You believe that it's possible to fully understand the requirements of a system before the system is built. For systems that have been built before: Sure. That's fairly easy. But if this is something new (and therefore likely to be more valuable), my 14 years of software/hardware development experience tell me that getting up front requirements correct is not possible. I've seen a lot of smart people try, and to be honest, I'm tired of being right on this topic.
Iterative processes (not just agile processes) work around this issue by spending some time developing something that works, and the using the experience from that exercise to inform subsequent projects to develop something that's excellent. As for "more work" being done on the rework of iterative improvement, the whole team, including the requirements writers, has to learn what's actually needed. To pretend that that learning is possible without something you can put in front of users (the previous iteration, perhaps) is either ego or delusion.
You've got it in your head that there's some way to be perfectly efficient in software development, wasting no time on incorrectly described needs, poorly developed features, etc. Everything I know about commercial software development says you're barking up the wrong tree. Being effective at satisfying needs and providing value must be more important than being efficient. IME generally, and in this case specifically, I find that efficiency is the enemy of effectiveness. There's simply too much learning to be done to pretend that once a moderately complex need has been identified, that it's possible to up front correctly specify a solution in any detail.
But if you want to keep trying, good luck to you. If you actually find a way, you should write a book on it. It's a guaranteed best seller.
I think part of the program is that most professional programmers tend to think in terms of systems with 1" requirement specifications.
One would think that "agile development" had never been proposed as a radically superior alternative to the waterfall process. In my experience, the number of programmers asking for thick specs are vanishingly few. Also, all of the ones I've observed were among the worst programmers I've ever seen, and wanted the spec as some sort of CYA exercise.
Most of the time it's management who wants the big spec, usually as some sort of control/power game.
Big specs are a sign of either 1) a badly broken development process or 2) an incredibly high risk project. One or more people are pretending that they know all of the trade-offs and can keep all of the details straight before doing any coding. The reality is that 99% of the time, the spec writers can't get close on the details or the trade-offs, and the big spec is already a SNAFU before it's half done.
On high risk projects (space shuttle control software), the spec is the project and the team will have processes to incorporate development-time learning back into the spec. Personally, I find those kinds of projects don't suit my personality (I thrive on agile teams with effective requirements management).
VB makes the easy stuff easy; but the hard stuff is simply impossible.
The other thing too, is that, the right wing is a diverse group, just as much as the left wing is. For us, we have a coalition of religious types and free market libertarian types. I fall into the latter.
This is what I find interesting. In your previous post, you label yourself a Bush supporter, and in this post, you label yourself a libertarian. I also label myself a libertarian, and I find Bush's corporatist factionalism, authoritarian leanings and complete lack of any fiscal responsibility to be completely contrary to anything like libertarian principles.
Basically, if you're:
In support of the establishment of fundamental Christianity as the official Church of the USA
In favor of letting moneyed corporations write market regulations
In support of finishing the conversion of the USA into a police state
In favor of radically reduced privacy and personal freedoms
In favor of unlimited deficit spending to pursue any of the previous four goals
then I can see why you're a Bush supporter.
But how do you support even one of those things and call yourself a libertarian? Now, clearly, in this message, you're moderating that position, but I have found it utterly astonishing since 2003 for any non-fundamentalist person to express any support whatsoever for Bush since he is so obviously against the interests of you and I. How do you reconcile your statement about being a Bush supporter with his blatant antipathy for you and your wife (since her business is not a large and established contributer towards the Republican party)?
Agnosticism and atheism are orthogonal categories. I am both. I currently lack a belief in invisible friends in the sky (atheist) and am also unwilling to assert that there are not one or more invisible friends in the sky (agnostic).
I simply find it improbable to the point that it's not worth considering in my philosophy or daily life.
The religious wish to claim that everyone is slightly religious by making the definition of atheist as indefensible as their own belief system (thereby requiring it's own kind of "faith") and then claiming agnostics as potential believers. I intend to resist this rhetorical game by using the definition that makes the most etymological sense: atheist - a-theist - lack of theism - lack of belief in deities.
An earlier poster provided a better term for "hard atheism" -- antitheism. I think I'm going to begin using that as a retort to those who claim I'm not an atheist or who would pretend that atheism is "anti-Christianity".
If someone feels so strongly about work structures I suggest they attempt to start their own business, be all wavy-gravey, and try not to act too surprised when your customers hate your "in your face service" and your employees come in wearing stinky beach-wear over their obese bodies.
Interesting that you conflate casual work dress with bad attitudes. In my experience, they are entirely separate (though the bad attitudes are usually the worst of the dressers).
The plan is definitely to start my own enterprise. The dress code? Probably not very different from where I work now (at a Fortune 50 company) where I currently wear hiking boots, blue jeans, and t-shirts most days. Customer service? Professional, courteous, and helpful. However, since 99.9% of our customer service will happen over email or on the phone, who cares how the customer service people are dressed?
As for the bad attitude: that's a significant culture fit problem. If you're going to work for me, you have to think of yourself as a craftsman, be able to work well with others, and have a pretty tight work ethic. If you don't have that sort of an attitude, I'm sure that there will be other employment opportunities that will be a better match for you, sorry that it didn't work out here.
IT nerds found themselves temporarily immune to such things due to the explosive growth of computers/networking/etc. in the business world, and the seemingly magic nature of the internet and all of that. But those days were a fluke, they're mostly over.
Are you sure about that? Where I work (a Fortune 50 company), we dress in such a way as to maximize our productivity: comfortable. T-shirts from Think Geek are fairly common. Bare feet and flip flops are fairly common. I've only had to take my button-down shirts to the cleaners once in the past year. They just don't get used any more. I wear hiking boots, blue jeans, and tasteful t-shirts or polos. So does the local VP of product development.
I still keep in touch with friends who work at other high-tech companies near me (Los Angeles). Not too many of them have to wear button-down shirts either. My impression is that "professional" attire has permanently shifted to a more casual and comfortable wardrobe. Good thing, too.
Who actually killed those troops? You are blaming Bush, but he didn't kill them. Blaming the president takes away the responsibility from the people who actually planted the roadside bombs and blew themselves up.
The leader who put them in harm's way for a lie is also responsible. Funny thing about responsibility, there are many different people who are responsible for an irreparable harm like death. Nobody is entirely responsible, but lots of people are responsible.
Don't try to isolate responsibility, as if everyone else is no longer responsible once a "responsible party" has been identified. Don't buy into that sort of partitioning. It's false and destructive.
Both versions are wonderful, but Soderberg spends more of the movie with Kelvin, while Tarkovski spread out the emphasis quite a bit more, specifically spending a lot more time on/with the planet itself. Apparently Lem wished that Soderberg had done the same (spent more time exploring Solaris).
Personally, my favorite thing about the Soderberg version is that the pacing is so deliberate. The earth scenes (mostly rainy) make me want to wrap up in a blanket and start sipping some cocoa. Almost like I'm nice and cozy, reading a great book in a comfortable chair in front of a huge bay window on a rainy Sunday... mmm...
"Do Androids Dream of Electric Sheep", on which Bladerunner is theoretically based, contains many times the depth and probably only takes you the same couple hours to read.
I disagree, sorta. They're such different stories, with such different protagonists, themes, and antagonists, I don't think the comparison is apt. An earlier post said it pretty well: there's no way of saying whether a movie that closely followed the book would be great. The movie "Blade Runner" is great, so let's enjoy it for what it is.
As for one being deeper than the other... personally, I find the movie's resolution of the synthetic/authentic dichotomy more satisfying. The book says that the synthetic is never as "good" as the authentic. The movie says it can be.
This analysis is consonant with my impression of Penrose re: AI's potential. Penrose says we can't simulate intelligence using Von Neumann computers because intelligence relies on quantum-mechanical nondeterministic computation to evade Godel's incompleteness theorem. I say that Penrose has made at least three significant errors: 1) his argument that human intelligence does successfully evade Godel's incompleteness theorem is pure speculation; 2) simple electrochemical models of brain operation include nondeterministic elements (neurotransmitter diffusion, etc.), without any need for quantum-level effects; and 3) that it would be difficult to add probabilistic operations to Von Neumann systems if nondeterministic elements were found to be necessary to simulate intelligence.
Don't get me wrong. I love reading PKD's stuff and am a huge fan. I just happen to disagree with his thesis in that story ("Do Androids Dream of Electric Sheep"), and that disagreement leads me to be more satisfied with Ridley Scott's variation on the story.
Um, I generally dislike Bush, but I think your sensitivity to the subject of Bush criticism has led you to miss the point.
The point of this lawyer's letter is not to demonize Bush or his daughters. It's to show how utterly and completely ridiculous the statutory damages for copyright infringement are, along with the RIAA rhetoric on the subject.
Strange, then, how these lawyers who created the laws seem to require so much research and time to prepare a case based on them. If they'd lobbied for them, one might think they'd be more familiar with the contents.
To solve this mystery, follow the money. What is the lawyer paid for? Time spent working on your case. So it's to your lawyer's benefit to not only have a legal system so complex that it excludes all but the professionally trained, but to have a legal system so complex that he has to spend a lot of time preparing.
Mysteries of this type are usually solved by understanding who gets paid for what and then closely examining the relationship between those things. Often the conflicts of interest become plainly obvious with very little effort.
It's a systemic problem, not an individual choice problem. Our streets, stores, homes, and cities are quite simply not even slightly bicycle friendly. They're bicycle-hostile, if anything. What makes me really sad is that we have built cities that are so good for cars that they're absolutely horrible for every other mode of transportation.
I've seen bicycle friendly cities and I want to find a way to move there because they seem so much more human friendly. Every time I think about my last trip to Amsterdam, I start trying to figure out how to move my family there. Every time I visit Europe I start apartment hunting.
Fair enough. I think I see where I misunderstood your argument, though you have to admit that it was a rather loaded statement of yours that I misunderstood.
Slashdot Mirror
With a gig of ram, you'll still be providing an enormous amount of space for the OS to cache files and content, and you won't ever have to wait for a program to page back into memory.
Photoshop 7 used to complain that I should have a pagefile and that I was making some sort of dramatic mistake. Photoshop CS2 and CS3 don't complain any more. Looks like they got with the times.
But I'm thrilled that you're even thinking about the situation of the desperately poor in other places. It gives me (a little) hope for the future. Work on that ignorance. Become a world traveler, and I guarantee that your perspective on many issues will change.
Regards,
Ross
Caring about where you get your food means valuing things that agri-business wants to pretend are irrelevant (because they make their money only when those issues are irrelevant). Agri-business values transportability, shelf-life, year-round availability, and shelf appearance over flavor, nutritional value, clean water, and living oceans. Farms used to be able to use animal waste as fertilizer and a hedge against erosion. Modern farms buy nitrogen derived from fossil fuels while CAFO's (concentrated animal feeding operations) produce unending streams of nitrogen, phosphorus, and antibiotic contaminated poisonous sludge that can't be used as a fertilizer and must be buried as toxic waste by people wearing moon suits.
Barbara Kingsolver says it better than I can in "Animal, Vegetable, Miracle". Other books that make a similar argument from a different perspective include "The Omnivore's Dilemma" by Michael Pollan and "Much Depends on Dinner:..." by Margaret Visser. If you're interested, head to the library.
Basically, that system can't last, and by opting out of that system (eating local), you do yourself, your neighborhood, and your planet a huge favor.
Regards,
Ross
Now, once you've settled in and are feeling somewhat comfortable, hire a cab and head out away from the "wealthy" spice farms of Goa. Just head east for about two hours. When you see a temple in good repair, stop, stretch your legs, make an offering. Walk for a bit and take a long look around you. Pay close attention to what you see.
Head back to the city, stop where some road work is being done. Get out and talk to the road workers about the tarp and cookfire at the side of the road. Find out how many family members live under that tarp. Find out how many family members are doing back-breaking labor on the road construction project. Find out the age of the youngest family member doing road construction at that site. The oldest.
In the next traffic jam, try to figure out the job of all of the people you see walking on and near the street. Your goal, should you choose to accept it, is to figure out how many of them are beggars.
India is a paradox contained in a country. There are enormously more unskilled people than unskilled jobs. Even the measures are so dramatically different as to defy comparison: the poverty level in India is just barely this side of survival. Telling a failing farmer to just "get a job" without knowing the first thing about what kind of jobs exist (and how many) is spectacularly ignorant.Your causation is mistaken. Corn subsidies got us HFCS and processed food products that slowly poison us. Government policy intended to guarantee a surplus of commodity foods, even in lean times, is what got us the corn/soy subsidies.
The romance of farming is largely gone in the US. Personally, I believe that we would be better off as a culture if we knew who had grown or raised our food items, and the restoration of the local farmer to significance (and the near elimination of the food processor) would be a great benefit to all of us as a nation, a state, and a people.
Regards,
Ross
And yet when someone misprints an ad, they are often liable to provide the goods for the advertised price, even when the price is clearly in error.
I don't see it so cut and dry as you do. I think it's just a simple case of seller beware. If you enter into an agreement where you could have gotten a better deal, it is not always the state's role to allow you to correct the situation post hoc. Not to say that there aren't laws to do exactly that, but IMHO, most are bad laws.
Regards,
Ross
My grandmother doesn't type in URL's. So I send her an email with the URL in the email. The :8080 doesn't matter at that point.
She does know not to click on URL's unless she's expecting someone to send her info about something.
Ross
The issue is how long you're off the ground.
On the earth, it basically impossible to stay in the air for more than half a second. An athlete's excellent standing jump imparts about 1kJ (100kg elevated by 1m) to his body. A jump on the moon with the same force will last about three seconds and put you six meters off the moon's surface. During those three seconds and twelve meters of vertical travel, there is a lot more opportunity to not land on your feet. Landing on your helmet on the moon is a very, very risky proposition.
Which is what I meant by "a possibly more dangerous height".
Regards,
Ross
The inverse happens with the kilo in the metric system. You could convert to newtons whenever you need force, but kilogram-force will be easier for most metric-thinking people to relate to. Basically, when communicating with lay-people, scientists and engineers have learned to be flexible.Actually, you can't get to escape velocity on the moon using your feet. You can jump to a possibly more dangerous height without the extra mass/weight, but you'll quickly learn not to.
Outside of a few specific cases (death star's, for instance), buying a bunch of the same disks will work just fine, and going to great lengths to get drives from different manufacturers and manufacturing lots doesn't seem to be worth the effort.And this is why you should still have a backup of your RAID. The ability of a RAID to tolerate one failure is at least partially offset by the fact that there are more parts that can fail.
Regards,
Ross
When a lipo cell has an overcurrent failure, part of the cell vaporizes (puffed cell failure). In particularly bad failures after heavy use (low voltage, abusive current loads, etc.), the vapors can ignite from the heat of the cell. When this happens, the cell wall is guaranteed to fail, and then you get the green flames of death.
But just poking a hole in your lipo won't start a fire. I also fly RC helis (T-Rex 450).
Regards,
Ross
A grad, or gradian, is 1/400 of a full rotation and is nearly useless for angles since it's not evenly divisible by 3, 6, 9, or 12, with 12 being the most important divisor in that list (even division by 12 means that you can represent all of the angles in a bisected equilateral triangle (30-60-90 triangle) using integer values).
The grad first appeared in France, and was only ever adopted in a few countries for specialized tasks (artillery, surveying) and except for French artillery, appears to be a dead unit.
It was believed that an angle unit with right angles of 100 units would make mental math easier (if you're facing 115 grad clockwise from due North, then you're 15 grad clockwise from due East). However, if you're frequently doing navigation or other cartographic-related tasks, it turns out that you quickly adapt to the mental math required to understand 90 degree right angles and angle division is much easier.
Regards,
Ross
For instance: on my current project, we use a wiki combined with an issue tracking system (http://trac.edgewall.com/) to articulate and discuss requirements. Usually (always?) the developers need to refine the requirements as the features are developed; removing inconsistencies, adding details, etc. and since it's a wiki, they can do that fairly easily.
And what about tests as proxies for detailed requirements? We try to create automated tests for everything. Unit tests (a few), functional tests (a lot), integration tests (a lot), load/scalability tests (a few). Contrary to XP, we generally find the best tests are written along with the interfaces and implementation, not before. On our team, QA is responsible for challenging developers on test adequacy. Do the tests really exercise the code? (QA also helps to create automated UI tests, which they base off of the trac wiki, usually after primary development is complete.)That's an interesting but flawed observation. What's novel about agile processes isn't the process. Iterative processes are "waterfalls writ small" and have been useful for many, many years. What's novel is the perspective that, at the end of the day, agile teams find that skilled people are more important than devotion to any particular process.I see a part of the problem right here. You believe that it's possible to fully understand the requirements of a system before the system is built. For systems that have been built before: Sure. That's fairly easy. But if this is something new (and therefore likely to be more valuable), my 14 years of software/hardware development experience tell me that getting up front requirements correct is not possible. I've seen a lot of smart people try, and to be honest, I'm tired of being right on this topic.
Iterative processes (not just agile processes) work around this issue by spending some time developing something that works, and the using the experience from that exercise to inform subsequent projects to develop something that's excellent. As for "more work" being done on the rework of iterative improvement, the whole team, including the requirements writers, has to learn what's actually needed. To pretend that that learning is possible without something you can put in front of users (the previous iteration, perhaps) is either ego or delusion.
You've got it in your head that there's some way to be perfectly efficient in software development, wasting no time on incorrectly described needs, poorly developed features, etc. Everything I know about commercial software development says you're barking up the wrong tree. Being effective at satisfying needs and providing value must be more important than being efficient. IME generally, and in this case specifically, I find that efficiency is the enemy of effectiveness. There's simply too much learning to be done to pretend that once a moderately complex need has been identified, that it's possible to up front correctly specify a solution in any detail.
But if you want to keep trying, good luck to you. If you actually find a way, you should write a book on it. It's a guaranteed best seller.
Ross
Most of the time it's management who wants the big spec, usually as some sort of control/power game.
Big specs are a sign of either 1) a badly broken development process or 2) an incredibly high risk project. One or more people are pretending that they know all of the trade-offs and can keep all of the details straight before doing any coding. The reality is that 99% of the time, the spec writers can't get close on the details or the trade-offs, and the big spec is already a SNAFU before it's half done.
On high risk projects (space shuttle control software), the spec is the project and the team will have processes to incorporate development-time learning back into the spec. Personally, I find those kinds of projects don't suit my personality (I thrive on agile teams with effective requirements management).
VB makes the easy stuff easy; but the hard stuff is simply impossible.
Regards,
Ross
Basically, if you're:
- In support of the establishment of fundamental Christianity as the official Church of the USA
- In favor of letting moneyed corporations write market regulations
- In support of finishing the conversion of the USA into a police state
- In favor of radically reduced privacy and personal freedoms
- In favor of unlimited deficit spending to pursue any of the previous four goals
then I can see why you're a Bush supporter.But how do you support even one of those things and call yourself a libertarian? Now, clearly, in this message, you're moderating that position, but I have found it utterly astonishing since 2003 for any non-fundamentalist person to express any support whatsoever for Bush since he is so obviously against the interests of you and I. How do you reconcile your statement about being a Bush supporter with his blatant antipathy for you and your wife (since her business is not a large and established contributer towards the Republican party)?
Regards,
Ross
Agnosticism and atheism are orthogonal categories. I am both. I currently lack a belief in invisible friends in the sky (atheist) and am also unwilling to assert that there are not one or more invisible friends in the sky (agnostic).
I simply find it improbable to the point that it's not worth considering in my philosophy or daily life.
The religious wish to claim that everyone is slightly religious by making the definition of atheist as indefensible as their own belief system (thereby requiring it's own kind of "faith") and then claiming agnostics as potential believers. I intend to resist this rhetorical game by using the definition that makes the most etymological sense: atheist - a-theist - lack of theism - lack of belief in deities.
An earlier poster provided a better term for "hard atheism" -- antitheism. I think I'm going to begin using that as a retort to those who claim I'm not an atheist or who would pretend that atheism is "anti-Christianity".
Regards,
Ross
The plan is definitely to start my own enterprise. The dress code? Probably not very different from where I work now (at a Fortune 50 company) where I currently wear hiking boots, blue jeans, and t-shirts most days. Customer service? Professional, courteous, and helpful. However, since 99.9% of our customer service will happen over email or on the phone, who cares how the customer service people are dressed?
As for the bad attitude: that's a significant culture fit problem. If you're going to work for me, you have to think of yourself as a craftsman, be able to work well with others, and have a pretty tight work ethic. If you don't have that sort of an attitude, I'm sure that there will be other employment opportunities that will be a better match for you, sorry that it didn't work out here.
Regards,
Ross
I still keep in touch with friends who work at other high-tech companies near me (Los Angeles). Not too many of them have to wear button-down shirts either. My impression is that "professional" attire has permanently shifted to a more casual and comfortable wardrobe. Good thing, too.
Regards,
Ross
Don't try to isolate responsibility, as if everyone else is no longer responsible once a "responsible party" has been identified. Don't buy into that sort of partitioning. It's false and destructive.
Regards,
Ross
Both versions are wonderful, but Soderberg spends more of the movie with Kelvin, while Tarkovski spread out the emphasis quite a bit more, specifically spending a lot more time on/with the planet itself. Apparently Lem wished that Soderberg had done the same (spent more time exploring Solaris).
:)
Personally, my favorite thing about the Soderberg version is that the pacing is so deliberate. The earth scenes (mostly rainy) make me want to wrap up in a blanket and start sipping some cocoa. Almost like I'm nice and cozy, reading a great book in a comfortable chair in front of a huge bay window on a rainy Sunday... mmm...
Maybe I'm just weird though.
Ross
As for one being deeper than the other... personally, I find the movie's resolution of the synthetic/authentic dichotomy more satisfying. The book says that the synthetic is never as "good" as the authentic. The movie says it can be.
This analysis is consonant with my impression of Penrose re: AI's potential. Penrose says we can't simulate intelligence using Von Neumann computers because intelligence relies on quantum-mechanical nondeterministic computation to evade Godel's incompleteness theorem. I say that Penrose has made at least three significant errors: 1) his argument that human intelligence does successfully evade Godel's incompleteness theorem is pure speculation; 2) simple electrochemical models of brain operation include nondeterministic elements (neurotransmitter diffusion, etc.), without any need for quantum-level effects; and 3) that it would be difficult to add probabilistic operations to Von Neumann systems if nondeterministic elements were found to be necessary to simulate intelligence.
Don't get me wrong. I love reading PKD's stuff and am a huge fan. I just happen to disagree with his thesis in that story ("Do Androids Dream of Electric Sheep"), and that disagreement leads me to be more satisfied with Ridley Scott's variation on the story.
Regards,
Ross
Learn about pecan harvesting. When it's a good investment, the effort will be made.
Regards,
Ross
Um, I generally dislike Bush, but I think your sensitivity to the subject of Bush criticism has led you to miss the point.
The point of this lawyer's letter is not to demonize Bush or his daughters. It's to show how utterly and completely ridiculous the statutory damages for copyright infringement are, along with the RIAA rhetoric on the subject.
Regards,
Ross
Mysteries of this type are usually solved by understanding who gets paid for what and then closely examining the relationship between those things. Often the conflicts of interest become plainly obvious with very little effort.
Ross
I suspect we're pretty much in agreement.
It's a systemic problem, not an individual choice problem. Our streets, stores, homes, and cities are quite simply not even slightly bicycle friendly. They're bicycle-hostile, if anything. What makes me really sad is that we have built cities that are so good for cars that they're absolutely horrible for every other mode of transportation.
I've seen bicycle friendly cities and I want to find a way to move there because they seem so much more human friendly. Every time I think about my last trip to Amsterdam, I start trying to figure out how to move my family there. Every time I visit Europe I start apartment hunting.
*sigh*
Ross
Fair enough. I think I see where I misunderstood your argument, though you have to admit that it was a rather loaded statement of yours that I misunderstood.
Have a good night.
Ross