And yet- if they had more food they wouldn't be starving- which makes US selfish for wanting our computers and educations and broadband connections and whatnot while there are still people starving.
I can't tell if you're ignoring my points or just don't get them. I think it's some of both. But let me point out that if you can't convince somebody who agrees with you on the end goal (that none should be deprived of basic necessities), you'll never convince the much larger percentage of people who don't give a damn.
You said that you deserve a broadband connection to the net and a computer
No, I didn't. I just checked. If you're having problems telling what I said, I'm not sure this conversation will be very productive. But I'll give it one more try.
Tell that to the people of Niger- they certainly don't think that the world already produces enough food.
Yes, but it still does. My point is that it's not a labor problem, and sending people off to farm won't help. It will hurt.
Look at population growth rates for people in developed countries versus subsistence farmers.
Did it ever occur to you that could possibly be due to free trade?
I can't quite follow you here. You're telling me the reason developed-world population growth is flat or negative is because they have free trade? Then wouldn't it follow that we should get those subsistence farmers some free trade as well?
If we achieve no more need- then I have no problem with people pursuing their wants. It will cost more to pursue their wants, of course, but that's why those are WANTS and not NEEDS.
Actually, in the long run it should cost less. People not getting what they need have their economic productivity stunted due to malnutrition, poor education, and so on. Fix that and you can make everybody richer. That's one of the keys to the virtuous circle that allowed the developed nations to pull so far ahead of the undeveloped ones.
Note also that you're asking a lot of people to abandon their wants to pursue something you want. I happen to share those wants, but I don't quite see why I should be able to use force to impose my wants on others.
What makes you think that YOU deserve to live a life of luxury built upon the poverty of others?
Funny, I missed where I said that.
Look, I think your apparent goals (e.g., that nobody should live in obvious want) are laudable. And it turns out, I share them. But I think your proposals are ludicrously simple-minded.
The world already produces enough food. This is not because we've devoted a large portion of our labor to that. Instead, it depends upon a lot of advanced technology, careful research, and strong education. All of those take educated specialists and special infrastructure to produce.
If we followed your everybody-back-to-the-land approach, agricultural productivity would plummet. The net result would be more starvation, not less. And as a kicker, population would shoot up, causing more starvation still. Don't believe me? Look at population growth rates for people in developed countries versus subsistence farmers.
And of course, people would put up with your approach for about 30 seconds before pursuing their wants again. Which means you'll need a large police force and a large bureaucracy to impose your will on everybody else. And as police forces and bureaucrats always end up favoring the rich and powerful, you won't be doing many favors for the poor and weak.
I'm all for improving the lot of the poor, and I'm excited to see what comes next after capitalism. But if you can't figure out how to get there without making a lot of people miserable, it won't happen. And if you aren't fomenting realistic change, then there are more productive ways for you to help the poor.
Oh, yeah, right- like you failed 2nd grade social studies. The basic human needs are: food, clothing, shelter, clean water, and medical care. Every 7 year old on the planet knows that one.
Oh! Well then, as a representative of the world's 7-year-olds, perhaps you can be a bit more precise.
It turns out that each one of those allegedly obvious needs comes in a variety of forms. If we're going to optimize the economy for needs only, we'll need to know exactly where to draw the dividing line between needs and wants. Which foods, exactly, should we be producing? And how much medical care is the amount you really need?
Personally, I'm going to miss education, human contact, and intellectual stimulation. And, come to think of it, clean air. And to be economically productive, I thought I needed a computer and some bandwidth. But if all the world's seven-year-olds (plus at least one Marxist) are sure I don't need those, how can I argue with that?
I've seen this a lot- but there is at least one economic theory, distributism, that claims that human wants are just the mortal sin of greed and human NEEDS are what we should be focused on satisfying- and human needs are indeed finite.
When you figure out how to get six billion people to agree on exactly what qualifies as a need, let me know.
You are still missing my point. I'm not saying that it's impossible, or even hard to write transactionally aware code. I'm saying that Prevayler doesn't provide that functionality for you, and it is a mistake to say that it does.
As I said before, the food taster stuff provides exactly the functionality you want, which is rollback on errors in the code you write. If you have some code that demonstrates a problem with it, post it to the Prevayler list.
I bitch and rant because for every aware practitioner like yourself who examines the technology carefully and can fill in the gaps, there are ten or more people who take it at face value, stick some code in, go live without testing, and then wonder what they did wrong.
If you go talk to pretty much any sailing instructor, they'll tell you to learn in the smallest, most responsive boat you can find. The sooner you make mistakes, the sooner you learn from them, and the better you learn how to handle making mistakes.
I feel the same way about code. If the problem is shitty coders who don't test, the solution isn't to make it so that 90% of the time they don't notice. By doing that, you enable them to write nine times as much crappy code before they stop and say, "Wait, something's not right here!" And now they have nine times as many bad habits to unlearn.
Let's take as a given that there are people out there who will grab a radically new technology, stick some code in, and go live, only to be somehow shocked that there might be issues. (Personally, I doubt this. They've certainly never turned up on the Prevayler list.) Is bitching about Prevayler really the best way for you to help them? While we're at it, why is it your job to save them from themselves? And do you find your approach is reducing the percentage of people acting like idiots?
Personally, I prefer to wait until they're in the "wondering what they did wrong" stage. People are much better at listening to advice on how to drink responsibly when they're in the middle of the hangover.
Prevayler does not provide ACID transactions. Saying "you don't need rollback or undo" is all well and good, but you can't call it a transaction. It's not atomic, and it's not consistent.
Used properly, that's incorrect. It provides all of the ACID properties.
So if you use Prevayler, do you have "reliability" and "transactional integrity"? No. It's something you have to code yourself. Among other things, you have to make sure no code in a transaction ever throws a RuntimeException:
I'd agree that if you don't know whether you can write your code to be reliable, then maybe you're not ready to use Prevayler. That wasn't an issue for us. We did our validation well before we reached the point of making a change. That's why we didn't need to write any rollback logic.
But in Prevayler 2.0, which has been out for quite a while, there's something for people who are afraid they'll be writing buggy code. I believe they call it the "Royal Food Taster". You pay in extra memory usage and a performance hit during rollback. I wouldn't recommend it for people for whom NullPointerExceptions are a way of life, but it's a reasonable solution if you want an extra safety belt. In our case, we eventually turned it off.
Of course, if a shop is not sure they can write reliable code, I don't know that any product will really help. The places I've seen like that still managed to make a hash of their persistent data because their troubles coding everything else also meant they had trouble drawing their transaction boundaries, handling in-transaction failures, and properly dealing with edge cases.
I'm not, by the way, saying that there's no use for the RDBMS approach to things. On a current project we just decided to use a typical SQL database because it was the best match for the business problem we're solving. But I am saying that you can build systems of equally high quality with either approach.
If you don't know how to use Prevayler to achieve that, fine, say so. But enough of this "It's impossible!" and "Mod him down!" stuff. I've used it and it's fine. If you want to learn how we did it, go ahead and ask. But if you're just looking to bitch and rant, I've got better things to do.
Interesting! I'll check it out. It's funny that my post got two responses: one saying I should take it a step further and one saying the approach was completely insane.:-)
By the way, Prevayler doesn't require you to keep everything in a Map. But it does require that there be one root object for the tree. Generally I make that a custom object (often called Domain) that holds the rest of the goods.
For somebody who has never tried it and admits not understanding how it works, I think "Mod Parent Down" is a little strong. Just because you're scared of new ideas doesn't mean you have to take it out on my karma.
Prevayler is not a reliable system. It does not have any transactional integrity. You can't run "entire systems" from it because you are limited to the maximum heap size of the JVM.
Look. One of us has built production systems using Prevayler. And it isn't you.
It indeed does have transactional integrity: all changes are done through transactions, which are completely reliable. If your application dataset fits in RAM (and given the price of RAM, an awful lot of them do) then Prevayler works just fine. In my experience it's very reliable: one production application has been up 11 months now with no downtime except a couple of planned 5-minute upgrade windows. And, of course, no data corruption.
Prevayler isn't suitable for every application, but there are plenty of applications where it works just fine. And given how lightweight it is (runs in the JVM, no external tools, no schema that duplicates your object model, a code base you can patch yourself, no concurrency worries), I felt development with Prevayler was much easier and more pleasant. The whole team walked away happy with the experience.
Spille's a solid, no-bullshit kind of guy, and his technical calls are right on every time I've had independent knowledge, so I trust him on this.
He may be smart generally, but here he misses the point entirely. If your transactions execute in a couple of milliseconds or less, you don't need locking or concurrency. Or rather, you need exactly one lock, which is what Prevayler has. And on a modern box, if everything is in RAM, a couple of milliseconds is a really long time: most of our Prevayler transactions were well under a millisecond.
It's true there's no rollback/undo. Our team never needed that. And had we needed it, we could have coded it. If you can't figure out how to write programs without rollback and also can't code what you need, then by all means you shouldn't use Prevayler. Of course, you also shouldn't write the next Microsoft Word, Quake, or Google: none of those apps use an SQL database, either, and all of them run in RAM.
Yay. I feel so safe.
If you want to feel safe with it, try writing some tests against it. Before we committed to it, we did extensive load testing and did our very best to break it. We couldn't. If you can, both I the Prevayler team would love to hear from you. Until then, maybe you should stick to ranting about things you understand, eh?
Before that talk I never new you could run entire systems directly from RAM. Wild.
Yes, it's very cool. Done right, response times are dumbfounding. And if you take an approach like Prevayler you can still have reliability and transactional integrity.
(To be fair, I haven't actually tried this. My bias is entirely from some hideous globs of tag soup that a Dreamweaver user has handed me. It might be a PEBKAC issue rather than a Dreamweaver issue.)
How is evolution involved in "hardwiring" the brains of media consumers to prefer mindlessness to rationality? The adaptive aspects of evolution suggest that species improve. Brakdown in rational thought is a devolution.
Only from the human perspective, which is a bit species-centric. There's relatively little evidence that rationality is long-term adaptive. We humans might only be a flash in the pan, and have come reasonably close to using our own "rationality" to extinguish all animal life, ourselves included.
Even without a dramatic ending, it's not clear that rationality is all that adaptive. It gave us birth control, for example. I know a lot of smart people who have decided not to have kids, not realizing that this will mean future generations will have fewer people like them. Evolutionary success is determined by what genes produce the largest number of ancestors, not whether they match human aethetic judgements.
My point: It's not possible to entirely divorce laws from religion. What laws we consider acceptable are based on our notion of the "good", and our notion of the "good" is partially informed by religious (or a-religious) beliefs.
Wrong! My notions of "good" are entirely underived from religious belief. Your notions might be, but understand that others don't need that.
The most recent attempts that come to mind for government rule based on athiesm are the Third Reich and the government of China. Not really the kind of governments I'm keen to live under.
That was government imposition of athiesm, mainly as a way to destroy competing power bases. That has exactly zero to do with the question of whether athiests can live moral lives. I note that Christians are much more likely to end up in prison than athiests are. And let's not even bring up all the horrific crimes committed in the name of religion.
Intelligent Design is a collection of holes in evolutionary theory.
Not really. Evolutionary biology journals are a collection of holes in evolutionary theory. That's how science progresses, and has for centuries.
Intelligent Design is a big set of out-of-context quotes and facts that have been assembled to give the impression that there's a reasonable alternative to the theory of evolution. I'm not saying that such a theory couldn't arise, but ID is theology and cultural manipulation in the guise of science.
Don't believe me? Try suggesting the designer is a group of space aliens who tinkered with terrestrial biology at key points. Actual scientists would admit that as one possible explanation for the holes the claim to have found. But ID proponents flip out. Heck, even try suggesting it's a group of gods, rather than one god that did it. Scientifically, both hypotheses are equally tenable, but you'll never find the ID people talking about "god(s)".
It's not just the media choosing to run only sensationalist stories, but media consumers, who mainly buy crap and leave thoughtful articles unread.
And it's not just the media consumers, but the brains of media consumers, which were hardwired by evolution to prefer gossip, political posturing, and photos of pretty girls and cute babies eating ice cream over thoughtful, rational discourse.
So nobody believes in evolution because of evolution. Which to my mind pretty much proves that there's no intelligent designer involved in this process.
Now, I've never used IDEA for a prolonged period of time - I couldn't get into it, and was happy enough with Eclipse not to worry. (The fact that Eclipse is free helps - it would be difficult to persuade my company to pay for loads of licences for IDEA when Eclipse is perfectly all right and free.)
Unfortunately, you have to use it for a prolonged period of time to see the difference. It also helps if you pair with somebody who's comfortable with it, as you can pick up the tricks they use. It's roughly equivalent to the difference between the Windows and Mac UIs. IDEA is polite, deferential, and wants to help you get work done. Eclipse has the same checklist of features, but doesn't have the same attention to usability.
I think this is especially valuable for people building UI code, be it web, GUI, or whatever. I can certainly get things done in Eclipse, but I have to crank up my pain tolerance to do it. Keeping the ambient level of pain down makes it easier to see UI problems in my own apps, and makes me more willing to fix them.
Of course, I'd still take Eclipse over any other IDE I've used. Last I did a comparision, it was still far superior to products like Borland's JBuilder, which a friend rightly called RageBuilder.
As many as possible. Use PHP for the front end, Perl for input parsing, [...]
I disagree. You should use as many as necessary, but as few as possible.
Maintenance cost is a huge portion of any ongoing software project. Requiring a maintainer to be an expert in 6 or more languages is a lot of training. What you're more likely to get is somebody who's an expert in one or two and might be willing to dabble in the rest.
Instead of picking the most appropriate language for the task, a large-scale web app has to build the team most appropriate to the project. Sometimes the cost of a new language is worth it. But the costs can be huge: in many shops there's a huge chasm between the database guys and the rest of the developers, and a lot of productivity goes down that chasm. You see similar problems, although not usually as dramatic, in places that do XML->XSLT->HTML.
I strongly agree with some of this, especially giving everything clear names, which he calls code commenting. (I hate Hungarian notation, though.) I love TODO comments: they are a great for an author to record thinking about future improvements. And if I'm publishing an API, in-code documentation like JavaDoc are a great way to keep your API docs.
But he doesn't mention unit tests at all, which I think are one of the best places to define what code is for. Unlike regular comments, unit tests can be verified by the computer, so they're much less likely to be misleading or out of date. If you have good unit tests (and especially if you do test-first development), I think a lot of traditional comments become less useful.
When somebody starts telling me that they used 5 different patterns in their program and they're proud of it - then I know the code is crap. Most of the pattern zealots I've seen write bloated, inefficient code.
I mostly agree with that. I have slogged through too much code that was written for architectural coolness. Zealots for anything are, almost by definition, hopeless at anything useful. This applies to pattern zealots, object zealots, lisp zealots, and perl zealots. It even applies to computer zealots: we all know people who will spend ages creating software for a problem they could solve in 15 minutes with common household objects.
But don't throw the baby out with the bathwater. Patterns are neat, and I think anybody serious about building software should study them. They are great for a common vocabulary. And they're handy when you're struggling with a design issue: a flip through the GoF book will make you consider design approaches you might otherwise miss. Used lightly and sparingly, design patterns are very helpful.
Apparently, it's not quite all in Japanese. Zooming in on the maps, there is some weird stuff in roman characters there, including LIGHTPUBLICITY, SUZUEBAYDIUM near Seduce something or other, Costello RIVULET (I'm still looking for Abbott's body of water), and THEFOOTBALLCAFE. I have no idea what any of this means.
Only if they do a respectable job. The ones that abuse their situation are a horrible burden on humanity. Ditto with dirty cops.
I agree completely. Perhaps I should have said that they're entitled to respect until they demonstrate otherwise.
Gotcha. Toe the line, don't speak unless spoken to, do so carefully and quickly. Do all this right and you'll probably avoid unfair treatment.
Well, I personally am willing to take it farther than that. I've certainly disagreed successfully with cops before. But you have to visibly and continuously demonstrate respect for the cop's authority. And you should never, ever give a hint of the me-big-monkey routine: no yelling, no getting angry, no chest-puffing, no challenging looks.
Given that, I've actually gotten pretty unfair treatment... in my favor. My girlfriend is amazed at some of the tickets I've gotten out of just by being respectful, agreeable, and honest. It's crazy, but I swear it works.
Slashdot Mirror
And yet- if they had more food they wouldn't be starving- which makes US selfish for wanting our computers and educations and broadband connections and whatnot while there are still people starving.
I can't tell if you're ignoring my points or just don't get them. I think it's some of both. But let me point out that if you can't convince somebody who agrees with you on the end goal (that none should be deprived of basic necessities), you'll never convince the much larger percentage of people who don't give a damn.
Best of luck in sorting it out, though.
No, I didn't. I just checked. If you're having problems telling what I said, I'm not sure this conversation will be very productive. But I'll give it one more try.
Tell that to the people of Niger- they certainly don't think that the world already produces enough food.
Yes, but it still does. My point is that it's not a labor problem, and sending people off to farm won't help. It will hurt.
Did it ever occur to you that could possibly be due to free trade?
I can't quite follow you here. You're telling me the reason developed-world population growth is flat or negative is because they have free trade? Then wouldn't it follow that we should get those subsistence farmers some free trade as well?
If we achieve no more need- then I have no problem with people pursuing their wants. It will cost more to pursue their wants, of course, but that's why those are WANTS and not NEEDS.
Actually, in the long run it should cost less. People not getting what they need have their economic productivity stunted due to malnutrition, poor education, and so on. Fix that and you can make everybody richer. That's one of the keys to the virtuous circle that allowed the developed nations to pull so far ahead of the undeveloped ones.
Note also that you're asking a lot of people to abandon their wants to pursue something you want. I happen to share those wants, but I don't quite see why I should be able to use force to impose my wants on others.
What makes you think that YOU deserve to live a life of luxury built upon the poverty of others?
Funny, I missed where I said that.
Look, I think your apparent goals (e.g., that nobody should live in obvious want) are laudable. And it turns out, I share them. But I think your proposals are ludicrously simple-minded.
The world already produces enough food. This is not because we've devoted a large portion of our labor to that. Instead, it depends upon a lot of advanced technology, careful research, and strong education. All of those take educated specialists and special infrastructure to produce.
If we followed your everybody-back-to-the-land approach, agricultural productivity would plummet. The net result would be more starvation, not less. And as a kicker, population would shoot up, causing more starvation still. Don't believe me? Look at population growth rates for people in developed countries versus subsistence farmers.
And of course, people would put up with your approach for about 30 seconds before pursuing their wants again. Which means you'll need a large police force and a large bureaucracy to impose your will on everybody else. And as police forces and bureaucrats always end up favoring the rich and powerful, you won't be doing many favors for the poor and weak.
I'm all for improving the lot of the poor, and I'm excited to see what comes next after capitalism. But if you can't figure out how to get there without making a lot of people miserable, it won't happen. And if you aren't fomenting realistic change, then there are more productive ways for you to help the poor.
Oh, yeah, right- like you failed 2nd grade social studies. The basic human needs are: food, clothing, shelter, clean water, and medical care. Every 7 year old on the planet knows that one.
Oh! Well then, as a representative of the world's 7-year-olds, perhaps you can be a bit more precise.
It turns out that each one of those allegedly obvious needs comes in a variety of forms. If we're going to optimize the economy for needs only, we'll need to know exactly where to draw the dividing line between needs and wants. Which foods, exactly, should we be producing? And how much medical care is the amount you really need?
Personally, I'm going to miss education, human contact, and intellectual stimulation. And, come to think of it, clean air. And to be economically productive, I thought I needed a computer and some bandwidth. But if all the world's seven-year-olds (plus at least one Marxist) are sure I don't need those, how can I argue with that?
I've seen this a lot- but there is at least one economic theory, distributism, that claims that human wants are just the mortal sin of greed and human NEEDS are what we should be focused on satisfying- and human needs are indeed finite.
When you figure out how to get six billion people to agree on exactly what qualifies as a need, let me know.
You are still missing my point. I'm not saying that it's impossible, or even hard to write transactionally aware code. I'm saying that Prevayler doesn't provide that functionality for you, and it is a mistake to say that it does.
As I said before, the food taster stuff provides exactly the functionality you want, which is rollback on errors in the code you write. If you have some code that demonstrates a problem with it, post it to the Prevayler list.
I bitch and rant because for every aware practitioner like yourself who examines the technology carefully and can fill in the gaps, there are ten or more people who take it at face value, stick some code in, go live without testing, and then wonder what they did wrong.
If you go talk to pretty much any sailing instructor, they'll tell you to learn in the smallest, most responsive boat you can find. The sooner you make mistakes, the sooner you learn from them, and the better you learn how to handle making mistakes.
I feel the same way about code. If the problem is shitty coders who don't test, the solution isn't to make it so that 90% of the time they don't notice. By doing that, you enable them to write nine times as much crappy code before they stop and say, "Wait, something's not right here!" And now they have nine times as many bad habits to unlearn.
Let's take as a given that there are people out there who will grab a radically new technology, stick some code in, and go live, only to be somehow shocked that there might be issues. (Personally, I doubt this. They've certainly never turned up on the Prevayler list.) Is bitching about Prevayler really the best way for you to help them? While we're at it, why is it your job to save them from themselves? And do you find your approach is reducing the percentage of people acting like idiots?
Personally, I prefer to wait until they're in the "wondering what they did wrong" stage. People are much better at listening to advice on how to drink responsibly when they're in the middle of the hangover.
My USPS "toilet" probably wouldn't go over too well either . . .
Wow! When the box is full, where do you ship the stuff?
Prevayler does not provide ACID transactions. Saying "you don't need rollback or undo" is all well and good, but you can't call it a transaction. It's not atomic, and it's not consistent.
Used properly, that's incorrect. It provides all of the ACID properties.
So if you use Prevayler, do you have "reliability" and "transactional integrity"? No. It's something you have to code yourself. Among other things, you have to make sure no code in a transaction ever throws a RuntimeException:
I'd agree that if you don't know whether you can write your code to be reliable, then maybe you're not ready to use Prevayler. That wasn't an issue for us. We did our validation well before we reached the point of making a change. That's why we didn't need to write any rollback logic.
But in Prevayler 2.0, which has been out for quite a while, there's something for people who are afraid they'll be writing buggy code. I believe they call it the "Royal Food Taster". You pay in extra memory usage and a performance hit during rollback. I wouldn't recommend it for people for whom NullPointerExceptions are a way of life, but it's a reasonable solution if you want an extra safety belt. In our case, we eventually turned it off.
Of course, if a shop is not sure they can write reliable code, I don't know that any product will really help. The places I've seen like that still managed to make a hash of their persistent data because their troubles coding everything else also meant they had trouble drawing their transaction boundaries, handling in-transaction failures, and properly dealing with edge cases.
I'm not, by the way, saying that there's no use for the RDBMS approach to things. On a current project we just decided to use a typical SQL database because it was the best match for the business problem we're solving. But I am saying that you can build systems of equally high quality with either approach.
If you don't know how to use Prevayler to achieve that, fine, say so. But enough of this "It's impossible!" and "Mod him down!" stuff. I've used it and it's fine. If you want to learn how we did it, go ahead and ask. But if you're just looking to bitch and rant, I've got better things to do.
Interesting! I'll check it out. It's funny that my post got two responses: one saying I should take it a step further and one saying the approach was completely insane. :-)
By the way, Prevayler doesn't require you to keep everything in a Map. But it does require that there be one root object for the tree. Generally I make that a custom object (often called Domain) that holds the rest of the goods.
For somebody who has never tried it and admits not understanding how it works, I think "Mod Parent Down" is a little strong. Just because you're scared of new ideas doesn't mean you have to take it out on my karma.
Prevayler is not a reliable system. It does not have any transactional integrity. You can't run "entire systems" from it because you are limited to the maximum heap size of the JVM.
Look. One of us has built production systems using Prevayler. And it isn't you.
It indeed does have transactional integrity: all changes are done through transactions, which are completely reliable. If your application dataset fits in RAM (and given the price of RAM, an awful lot of them do) then Prevayler works just fine. In my experience it's very reliable: one production application has been up 11 months now with no downtime except a couple of planned 5-minute upgrade windows. And, of course, no data corruption.
Prevayler isn't suitable for every application, but there are plenty of applications where it works just fine. And given how lightweight it is (runs in the JVM, no external tools, no schema that duplicates your object model, a code base you can patch yourself, no concurrency worries), I felt development with Prevayler was much easier and more pleasant. The whole team walked away happy with the experience.
Spille's a solid, no-bullshit kind of guy, and his technical calls are right on every time I've had independent knowledge, so I trust him on this.
He may be smart generally, but here he misses the point entirely. If your transactions execute in a couple of milliseconds or less, you don't need locking or concurrency. Or rather, you need exactly one lock, which is what Prevayler has. And on a modern box, if everything is in RAM, a couple of milliseconds is a really long time: most of our Prevayler transactions were well under a millisecond.
It's true there's no rollback/undo. Our team never needed that. And had we needed it, we could have coded it. If you can't figure out how to write programs without rollback and also can't code what you need, then by all means you shouldn't use Prevayler. Of course, you also shouldn't write the next Microsoft Word, Quake, or Google: none of those apps use an SQL database, either, and all of them run in RAM.
Yay. I feel so safe.
If you want to feel safe with it, try writing some tests against it. Before we committed to it, we did extensive load testing and did our very best to break it. We couldn't. If you can, both I the Prevayler team would love to hear from you. Until then, maybe you should stick to ranting about things you understand, eh?
Before that talk I never new you could run entire systems directly from RAM. Wild.
Yes, it's very cool. Done right, response times are dumbfounding. And if you take an approach like Prevayler you can still have reliability and transactional integrity.
4) Now clean up Dreamweaver's HTML.
(To be fair, I haven't actually tried this. My bias is entirely from some hideous globs of tag soup that a Dreamweaver user has handed me. It might be a PEBKAC issue rather than a Dreamweaver issue.)
How is evolution involved in "hardwiring" the brains of media consumers to prefer mindlessness to rationality? The adaptive aspects of evolution suggest that species improve. Brakdown in rational thought is a devolution.
Only from the human perspective, which is a bit species-centric. There's relatively little evidence that rationality is long-term adaptive. We humans might only be a flash in the pan, and have come reasonably close to using our own "rationality" to extinguish all animal life, ourselves included.
Even without a dramatic ending, it's not clear that rationality is all that adaptive. It gave us birth control, for example. I know a lot of smart people who have decided not to have kids, not realizing that this will mean future generations will have fewer people like them. Evolutionary success is determined by what genes produce the largest number of ancestors, not whether they match human aethetic judgements.
My point: It's not possible to entirely divorce laws from religion. What laws we consider acceptable are based on our notion of the "good", and our notion of the "good" is partially informed by religious (or a-religious) beliefs.
Wrong! My notions of "good" are entirely underived from religious belief. Your notions might be, but understand that others don't need that.
The most recent attempts that come to mind for government rule based on athiesm are the Third Reich and the government of China. Not really the kind of governments I'm keen to live under.
That was government imposition of athiesm, mainly as a way to destroy competing power bases. That has exactly zero to do with the question of whether athiests can live moral lives. I note that Christians are much more likely to end up in prison than athiests are. And let's not even bring up all the horrific crimes committed in the name of religion.
Intelligent Design is a collection of holes in evolutionary theory.
Not really. Evolutionary biology journals are a collection of holes in evolutionary theory. That's how science progresses, and has for centuries.
Intelligent Design is a big set of out-of-context quotes and facts that have been assembled to give the impression that there's a reasonable alternative to the theory of evolution. I'm not saying that such a theory couldn't arise, but ID is theology and cultural manipulation in the guise of science.
Don't believe me? Try suggesting the designer is a group of space aliens who tinkered with terrestrial biology at key points. Actual scientists would admit that as one possible explanation for the holes the claim to have found. But ID proponents flip out. Heck, even try suggesting it's a group of gods, rather than one god that did it. Scientifically, both hypotheses are equally tenable, but you'll never find the ID people talking about "god(s)".
It's not just the media choosing to run only sensationalist stories, but media consumers, who mainly buy crap and leave thoughtful articles unread.
And it's not just the media consumers, but the brains of media consumers, which were hardwired by evolution to prefer gossip, political posturing, and photos of pretty girls and cute babies eating ice cream over thoughtful, rational discourse.
So nobody believes in evolution because of evolution. Which to my mind pretty much proves that there's no intelligent designer involved in this process.
Now, I've never used IDEA for a prolonged period of time - I couldn't get into it, and was happy enough with Eclipse not to worry. (The fact that Eclipse is free helps - it would be difficult to persuade my company to pay for loads of licences for IDEA when Eclipse is perfectly all right and free.)
Unfortunately, you have to use it for a prolonged period of time to see the difference. It also helps if you pair with somebody who's comfortable with it, as you can pick up the tricks they use. It's roughly equivalent to the difference between the Windows and Mac UIs. IDEA is polite, deferential, and wants to help you get work done. Eclipse has the same checklist of features, but doesn't have the same attention to usability.
I think this is especially valuable for people building UI code, be it web, GUI, or whatever. I can certainly get things done in Eclipse, but I have to crank up my pain tolerance to do it. Keeping the ambient level of pain down makes it easier to see UI problems in my own apps, and makes me more willing to fix them.
Of course, I'd still take Eclipse over any other IDE I've used. Last I did a comparision, it was still far superior to products like Borland's JBuilder, which a friend rightly called RageBuilder.
As many as possible. Use PHP for the front end, Perl for input parsing, [...]
I disagree. You should use as many as necessary, but as few as possible.
Maintenance cost is a huge portion of any ongoing software project. Requiring a maintainer to be an expert in 6 or more languages is a lot of training. What you're more likely to get is somebody who's an expert in one or two and might be willing to dabble in the rest.
Instead of picking the most appropriate language for the task, a large-scale web app has to build the team most appropriate to the project. Sometimes the cost of a new language is worth it. But the costs can be huge: in many shops there's a huge chasm between the database guys and the rest of the developers, and a lot of productivity goes down that chasm. You see similar problems, although not usually as dramatic, in places that do XML->XSLT->HTML.
I strongly agree with some of this, especially giving everything clear names, which he calls code commenting. (I hate Hungarian notation, though.) I love TODO comments: they are a great for an author to record thinking about future improvements. And if I'm publishing an API, in-code documentation like JavaDoc are a great way to keep your API docs.
But he doesn't mention unit tests at all, which I think are one of the best places to define what code is for. Unlike regular comments, unit tests can be verified by the computer, so they're much less likely to be misleading or out of date. If you have good unit tests (and especially if you do test-first development), I think a lot of traditional comments become less useful.
When somebody starts telling me that they used 5 different patterns in their program and they're proud of it - then I know the code is crap. Most of the pattern zealots I've seen write bloated, inefficient code.
I mostly agree with that. I have slogged through too much code that was written for architectural coolness. Zealots for anything are, almost by definition, hopeless at anything useful. This applies to pattern zealots, object zealots, lisp zealots, and perl zealots. It even applies to computer zealots: we all know people who will spend ages creating software for a problem they could solve in 15 minutes with common household objects.
But don't throw the baby out with the bathwater. Patterns are neat, and I think anybody serious about building software should study them. They are great for a common vocabulary. And they're handy when you're struggling with a design issue: a flip through the GoF book will make you consider design approaches you might otherwise miss. Used lightly and sparingly, design patterns are very helpful.
Technically, according to the CAN-SPAM Act, spamming is legal
We here answer to a higher law. Ever since the ascension of Jon Postel, the Internet is now not subject to mundane terrestrial regulation.
Apparently, it's not quite all in Japanese. Zooming in on the maps, there is some weird stuff in roman characters there, including LIGHTPUBLICITY, SUZUEBAYDIUM near Seduce something or other, Costello RIVULET (I'm still looking for Abbott's body of water), and THEFOOTBALLCAFE. I have no idea what any of this means.
And as a bonus, a bunch of swastikas. (I was hoping I had discovered secret Nazi bases, but it's probably just Buddhist temples. Drat!)
Only if they do a respectable job. The ones that abuse their situation are a horrible burden on humanity. Ditto with dirty cops.
I agree completely. Perhaps I should have said that they're entitled to respect until they demonstrate otherwise.
Gotcha. Toe the line, don't speak unless spoken to, do so carefully and quickly. Do all this right and you'll probably avoid unfair treatment.
Well, I personally am willing to take it farther than that. I've certainly disagreed successfully with cops before. But you have to visibly and continuously demonstrate respect for the cop's authority. And you should never, ever give a hint of the me-big-monkey routine: no yelling, no getting angry, no chest-puffing, no challenging looks.
Given that, I've actually gotten pretty unfair treatment... in my favor. My girlfriend is amazed at some of the tickets I've gotten out of just by being respectful, agreeable, and honest. It's crazy, but I swear it works.
The correct answer would be: "I believe I was doing the speed limit officer, is there a problem?"
That's only the correct answer if it's plausible. If it's on a highway where everybody is doing 15 over, it's a quick way to earn a ticket.
Slashdot is going downhill faster and faster.
That's completely unjustified! It's always been kinda half-assed.