I'm always imagining a world of honesty and integrity, so I came up with what I came up with.
Welcome to germany!
Well, not really ofcourse, but at least we have the tax thing nailed here. It's very simple: Whenever you offer anything (good or service) to a consumer then your advertised price must include all taxes.
It's been that way for as long as I can remember (at least 30 years) and our economy doesn't seem to have died from it. In contrast the american system sounds really fucked up to me, what a constant rip-off!
Furthermore grandparent seems to forget that taxes apply to everyone. So when *you* (as a company) are forced to swallow taxes and raise your price by $10 then your competitor will have to do the same. And if he doesn't then he *really* offers the better deal. No magic here.
Yes, the paypal procedure comes close. Except that paypal is a third party, taking a cut from each transaction and with a dubious track record of various problems. Furthermore the paypal integration in all places that I have seen was lacking; I still had to create an account and enter my billing and shipping details at the merchant's site.
The reason something like this didn't get set up, is because it means the retailer has to do a little more work to get things set up initially.
No, I think the real reason is sheer incompetence, negligence and a vested interest in keeping up the established foodchain. From a technical point of view such a system could be even easier to set up for a retailer than the kludges that we have today. Furthermore it would be more secure and comfortable for the customer because he wouldn't have to deal with a zillion different shop-accounts [ending up using the same username/password everywhere] and a zillion different checkout procedures with various third parties involved (gaypal & friends).
The ideal checkout would be "enter your VISA/Mastercard/whatever userid" - and done. The user gets a signed "request to authorize payment" in a standard format that he can actually understand (what am I buying and from where) and after authorization the merchant gets a signed confirmation including the user's shipping address.
Why not make it a friggin config-option in the [c|k]ontrol-center under the accessibility-tab, and be done?
"Where to put the Cancel-button in dialog-windows?"
[ ] Always right-most [ ] Always left-most [ ] Random position [ ] Careful mode: Enable OK-button only after a delay of [ 5 ] seconds [ ] Super-careful mode: Ask for [ 3 ] confirmations before proceeding with any action [X] Toggle right/left when mouse-cursor hovers [ ] Random position and randomly swap the labels of OK/Cancel [ ] Random position and pick a random language for the word "Cancel" [ ] Real-Man mode (Hide the cancel button) [ ] Wimpy mode (Turn all dialog buttons into "Cancel"-buttons) [ ] Gambler mode (Present only a single button that performs a random action) [ ] Mental mode (don't display any dialog buttons) [ ] Vintage mode (display a thumbdial instead of buttons) [ ] George W Bush mode (perform the most destructive action without even displaying a dialog) [ ] Schroedinger mode (display two dialog windows, neither of which is clickable, at 50% opacity) [ ] Nerdsex mode (display normal dialogs but afterwards ask: "Was it good for you, too?") [ ] Hollywood mode (don't click me, I click you!) [ ] Rorschach mode (let the user draw a picture, then determine action from that) [ ] CAPTCHA/Turing mode (perform action only after user has proved that he is a human) [ ] Windows95 mode (display normal dialog, then freeze no matter which button was clicked) [ ] Windows98 mode (display normal dialog, then bluescreen no matter which button was clicked) [ ] WindowsXP mode (display normal dialog, then download and run malware no matter which button was clicked) [ ] WindowsVista mode (remember all dialogs that have ever been displayed and repeat them at random intervals while slowing down to a crawl)
Well, apart from the usual beaurocracy nonsense (re-licensing after 8 years...) I still don't see how that's unreasonable. Working with children is a big responsibility and I find a criminal background check adequate in that case as long as the terms are clear upfront.
Ofcourse nobody should be denied to work with children because they stole a lipstick as a teen or somesuch. But if you have *real* crimes on record then you probably shouldn't be in the business of educating children, don't you think? I really fail to see what the big deal is in this particular case...
What exactly do these guys look for? Past legal records about sexual offense or so?
I wonder why you're talking so negative about it. To me that sounds like a measure that actually makes a little sense. I mean, sure it may be overkill but hey, at least for once they target the real world! Having to apply for childwork sounds sensible to me. And if you want to work with children then why should you mind a background check?
Disclaimer: I'm not one of the "think of the children" treehuggers at all. In fact I find the noise that little kids make rather denerving;-)
You are right but every layer of security helps and at least this raises the bar for large scale traffic analysis. Today's hardware can easily "grep" on a large number of concurrent, non-encrypted connections - without breaking a sweat. The same task suddenly becomes very expensive when you have to maintain a MITM for each connection.
Thus, at the very least this would throw big brother back to picking his targets one by one...
IPSEC *may* be very well engineered but few of us would want to touch it even with a 10ft pole. Especially those of us who *had* to work with it in the past. It should be possible to implement IPSEC without the warts. Hell, IPSEC could be zero-configuration out of the box (linklevel encryption only) with only minimal configuration for peer certificates.
Good Crypto doesn't have to be painful, see OpenSSH, OpenVPN (commonly chosen instead of IPSEC), GnuPG.
I just don't see what this has to do with P2P at all? Solution looking for a problem? When the ISPs can't sniff our traffic anymore they'll just connect to the trackers and look at the offerings.
But then I again I never understood the legal fuzz about P2P in first place. To me the key is plausible deniability. Store your shared content on an encrypted drive and that's it.
Sorry, I cannot take anyone seriously who proposes to "standarize developement tools". As others have pointed out that's such an obvious productivity killer - you're in the wrong business if you even think about proposing such a thing.
Everyone knows the ancient proverb that says it all: Right tool for the job. Period.
If java is the right tool for everything in your company then fine, standarize on that. But in reality it normally makes more sense to write the application in java, write the admin-scripts in perl (or whatever scripting language), script the webserver in lua, and so on.
Trying to use "the one hammer" for everything simply doesn't work, never did. Furthermore: Productivity can only suffer if you force your team to use tools that they don't like.
Your best developer probably prefers Vim or Emacs to write his code while everybody else is using eclipse. If you try to force an IDE down his throat he'll soon be your former best developer. Trying to standarize on editors or IDEs is a no-go. No sane developer will follow if you rule out his preferred tool.
Well, my point is that the solution to the bus-factor is not to constantly reduce your team's productivity with "rotation" (gasp!), excessive pair-programming or similar mindless measures for the sole purpose of "preparing for when it happens".
The solution is to enforce good coding and documentation practices so that you don't need any advance preparation.
I don't need to stay up to date with my peer's code constantly to know that "if it happens" I'd be able to replace any of them after some warmup time. I know that because we share common coding practices and don't hire people who write bad code that would turn into a liability after they leave.
Again: If your team is in good shape then some mysterious bus-factor will be the least of your worries. Simply because it happens very rarely (developer exodus would point to bigger problems) and in a healthy team it's a non-issue to assign the orphaned tasks to someone else.
Oh yea, the mythical "bus-factor". That's one of the most retarded business memes ever, yet I keep hearing it.
If you think the "bus-factor" matters in your project then you have bigger problems to start with!
Ask yourself: Why can't an an equally skilled member of the team pick up the work of the lost employee in reasonable time? The answer to that question is your real problem, kthxbye.
Interesting definition of "high performance cluster" you have there. Gigabit interconnect? Running your simulations in java?
I have no idea what kind of simulations you're running there but I seriously question the price/performance ratio of Opteron hardware for such an application. Sounds like you could save a ton of money by building your cluster from off-the-shelf hardware.
Well, you're making it sound like the better qualified person (by what metric, business school degrees?) will always make the better decisions. IMHO it's a bit more complicated than that.
Yes, you need a certain mix of knowledge, social skills and elbow grease to qualify for a CEO position and never did I claim that "anyone" could be a CEO. The problem is that you cannot measure these skills or really determine whether someone is 5% more competent than someone else. You can only measure the bottom line and in most cases the difference will be surprisingly small when you swap out a CEO or other high level exec for someone equally qualified (in terms of: knowing his craft).
This is what I mean when I say: no individual can really constantly make such a big difference. Especially not in a big company - and we're talking about big when we talk about 7-digit salaries.
My (idealistic) proposal still stands: Pay them huge bonuses when their decisions bring success to the company. Pay them high salaries if you must but tie them to measurable success, nothing else. Ofcourse that's not how our system works and will never become reality. Greed is rooted very deeply in human nature even when the only difference between owning one yacht or a whole fleet is "do i have more than the other guy?".
It's not uncommon for a CEO to earn 10x or even 100x the salary of an average employee. The reason is not that he's adding 10x or 100x more value, the reason is because he can. He's worked himself up (or got born into) the top of the food chain and that's his privilege: he can fire you, you can't fire him. He can demand ridiculous salaries, you can not. He can sink your company but still get the golden parachute, you can't.
This is the common pattern, admittedly quite a bit simplified. Nonetheless my point is: no single person can add >1000% value above average to a company constantly. As far as I am concerned: Pay them big bonuses when they strike a hot deal. But seven digit "salaries" are a [known] bug in our system.
Ofcourse, royal fuckups happen all the time. In that particular case I guess they either didn't consult someone with a clue or they actually applied some of that sick "Total War" strategy. Y'know the part where it says "we must patrol that area but don't have proper armor available, ah we'll do it anyways".
There are many many places where one group sets a strategy, and another sets the tactics.
Bizarre imagination you have there. Yes, upper mgmt decides for a strategy but usually *after* consulting with the tacticians whether that strategy is viable at all.
That's partly the reason why not so many companies aim for world domination although upper management in almost every company would certainly love that. Your business-reference is not so far off nonetheless - your mindset reminds me of the typical MBA/manager caricatures (dilbert).
why should I bother making a fool of myself by making rediculous suggestions that I know little about
Congrats, you did just that.
Strategy is an easier to understand game since studying history does well to suggest what works and what doesn't. The effectiveness of strategy changes very little over the centuries, unlike tactics.
This oversimplification is so idiotic, I don't even know where to start. Strategies only make sense in context, after evaluating the benefits and drawbacks of a certain strategy in the given situation.
Claiming that a given strategy applies optimally to any given situation is well, to use your words and your spelling: rediculous.
You're proposing a strategy ("total war") as the only and ultimate strategy - but without the slightest idea of how to implement it? Great call, but I'll suggest "Make love, not war" instead.
You didn't answer the question and Iraq doesn't remotely compare to any of these countries or the WWII situation.
You made it sound like you have the patent recipe for winning a war - comfortably from your couch, too. So what does your ultimate strategy suggest? Nuke over baghdad?
The only way success in war can actually be achieved is by the complete submission of the enemy population (which also includes those not hostile).
And how do you achieve that Einstein? Or should I say Adolf? I wouldn't consider this post worth responding to if it hadn't been modded "Insightful" - my ass. Mods didn't get their medis again?
So, what's your definition of "complete submission"? Is Iraq in "complete submission"? If it is not then what "total war"-steps do you propose to enforce "complete submission" there? Drop some nukes? Kill their children? Starve them?
Now you may go back to watching TV, just please keep your brainfarts to yourself next time.
Slashdot Mirror
For the non-technies among us: On the internet nobody cares that you're a dog!
As a python programmer I find this discussion refreshingly amusing! ;-)
Welcome to germany!
Well, not really ofcourse, but at least we have the tax thing nailed here.
It's very simple: Whenever you offer anything (good or service) to a consumer then your advertised price must include all taxes.
It's been that way for as long as I can remember (at least 30 years) and our economy doesn't seem to have died from it.
In contrast the american system sounds really fucked up to me, what a constant rip-off!
Furthermore grandparent seems to forget that taxes apply to everyone. So when *you* (as a company) are forced to swallow taxes and raise your price by $10 then your competitor will have to do the same. And if he doesn't then he *really* offers the better deal. No magic here.
Yes, the paypal procedure comes close. Except that paypal is a third party, taking a cut from each transaction and with a dubious track record of various problems. Furthermore the paypal integration in all places that I have seen was lacking; I still had to create an account and enter my billing and shipping details at the merchant's site.
Businesses, by definition, don't care who you are. They care for the money that you give to them, be it your own or someone elses.
No, I think the real reason is sheer incompetence, negligence and a vested interest in keeping up the established foodchain.
From a technical point of view such a system could be even easier to set up for a retailer than the kludges that we have today. Furthermore it would be more secure and comfortable for the customer because he wouldn't have to deal with a zillion different shop-accounts [ending up using the same username/password everywhere] and a zillion different checkout procedures with various third parties involved (gaypal & friends).
The ideal checkout would be "enter your VISA/Mastercard/whatever userid" - and done.
The user gets a signed "request to authorize payment" in a standard format that he can actually understand (what am I buying and from where) and after authorization the merchant gets a signed confirmation including the user's shipping address.
2008-1, 2008-2, 2008-3 makes sense to me.
I couldn't care less whether a kernel was released in january or april,
so simple numbering would be my favorite.
That scheme is easy to memorize, you immediately know which one is older and it keeps the number short.
Why not make it a friggin config-option in the [c|k]ontrol-center under the accessibility-tab, and be done?
"Where to put the Cancel-button in dialog-windows?"
[ ] Always right-most
[ ] Always left-most
[ ] Random position
[ ] Careful mode: Enable OK-button only after a delay of [ 5 ] seconds
[ ] Super-careful mode: Ask for [ 3 ] confirmations before proceeding with any action
[X] Toggle right/left when mouse-cursor hovers
[ ] Random position and randomly swap the labels of OK/Cancel
[ ] Random position and pick a random language for the word "Cancel"
[ ] Real-Man mode (Hide the cancel button)
[ ] Wimpy mode (Turn all dialog buttons into "Cancel"-buttons)
[ ] Gambler mode (Present only a single button that performs a random action)
[ ] Mental mode (don't display any dialog buttons)
[ ] Vintage mode (display a thumbdial instead of buttons)
[ ] George W Bush mode (perform the most destructive action without even displaying a dialog)
[ ] Schroedinger mode (display two dialog windows, neither of which is clickable, at 50% opacity)
[ ] Nerdsex mode (display normal dialogs but afterwards ask: "Was it good for you, too?")
[ ] Hollywood mode (don't click me, I click you!)
[ ] Rorschach mode (let the user draw a picture, then determine action from that)
[ ] CAPTCHA/Turing mode (perform action only after user has proved that he is a human)
[ ] Windows95 mode (display normal dialog, then freeze no matter which button was clicked)
[ ] Windows98 mode (display normal dialog, then bluescreen no matter which button was clicked)
[ ] WindowsXP mode (display normal dialog, then download and run malware no matter which button was clicked)
[ ] WindowsVista mode (remember all dialogs that have ever been displayed and repeat them at random intervals while slowing down to a crawl)
OSS is about options, right?
TC Phone 1:
1) Start talking
Well, apart from the usual beaurocracy nonsense (re-licensing after 8 years...) I still don't see how that's unreasonable. Working with children is a big responsibility and I find a criminal background check adequate in that case as long as the terms are clear upfront.
Ofcourse nobody should be denied to work with children because they stole a lipstick as a teen or somesuch. But if you have *real* crimes on record then you probably shouldn't be in the business of educating children, don't you think? I really fail to see what the big deal is in this particular case...
What exactly do these guys look for? Past legal records about sexual offense or so?
I wonder why you're talking so negative about it. To me that sounds like a measure that actually makes a little sense.
I mean, sure it may be overkill but hey, at least for once they target the real world! Having to apply for childwork sounds sensible to me. And if you want to work with children then why should you mind a background check?
Disclaimer: I'm not one of the "think of the children" treehuggers at all. In fact I find the noise that little kids make rather denerving ;-)
You are right but every layer of security helps and at least this raises the bar for large scale traffic analysis.
Today's hardware can easily "grep" on a large number of concurrent, non-encrypted connections - without breaking a sweat.
The same task suddenly becomes very expensive when you have to maintain a MITM for each connection.
Thus, at the very least this would throw big brother back to picking his targets one by one...
Parent is spot on.
IPSEC *may* be very well engineered but few of us would want to touch it even with a 10ft pole. Especially those of us who *had* to work with it in the past.
It should be possible to implement IPSEC without the warts. Hell, IPSEC could be zero-configuration out of the box (linklevel encryption only) with only minimal configuration for peer certificates.
Good Crypto doesn't have to be painful, see OpenSSH, OpenVPN (commonly chosen instead of IPSEC), GnuPG.
I just don't see what this has to do with P2P at all? Solution looking for a problem?
When the ISPs can't sniff our traffic anymore they'll just connect to the trackers and look at the offerings.
But then I again I never understood the legal fuzz about P2P in first place.
To me the key is plausible deniability. Store your shared content on an encrypted drive and that's it.
Sorry, I cannot take anyone seriously who proposes to "standarize developement tools". As others have pointed out that's such an obvious productivity killer - you're in the wrong business if you even think about proposing such a thing.
Everyone knows the ancient proverb that says it all: Right tool for the job. Period.
If java is the right tool for everything in your company then fine, standarize on that.
But in reality it normally makes more sense to write the application in java, write the admin-scripts in perl (or whatever scripting language), script the webserver in lua, and so on.
Trying to use "the one hammer" for everything simply doesn't work, never did.
Furthermore: Productivity can only suffer if you force your team to use tools that they don't like.
Your best developer probably prefers Vim or Emacs to write his code while everybody else is using eclipse. If you try to force an IDE down his throat he'll soon be your former best developer. Trying to standarize on editors or IDEs is a no-go. No sane developer will follow if you rule out his preferred tool.
Well, my point is that the solution to the bus-factor is not to constantly reduce your team's productivity with "rotation" (gasp!), excessive pair-programming or similar mindless measures for the sole purpose of "preparing for when it happens".
The solution is to enforce good coding and documentation practices so that you don't need any advance preparation.
I don't need to stay up to date with my peer's code constantly to know that "if it happens" I'd be able to replace any of them after some warmup time. I know that because we share common coding practices and don't hire people who write bad code that would turn into a liability after they leave.
Again: If your team is in good shape then some mysterious bus-factor will be the least of your worries. Simply because it happens very rarely (developer exodus would point to bigger problems) and in a healthy team it's a non-issue to assign the orphaned tasks to someone else.
Oh yea, the mythical "bus-factor".
That's one of the most retarded business memes ever, yet I keep hearing it.
If you think the "bus-factor" matters in your project then you have bigger problems to start with!
Ask yourself: Why can't an an equally skilled member of the team pick up the work of the lost employee in reasonable time?
The answer to that question is your real problem, kthxbye.
Interesting definition of "high performance cluster" you have there.
Gigabit interconnect? Running your simulations in java?
I have no idea what kind of simulations you're running there but I seriously question the price/performance ratio of Opteron hardware for such an application. Sounds like you could save a ton of money by building your cluster from off-the-shelf hardware.
Well, you're making it sound like the better qualified person (by what metric, business school degrees?) will always make the better decisions. IMHO it's a bit more complicated than that.
Yes, you need a certain mix of knowledge, social skills and elbow grease to qualify for a CEO position and never did I claim that "anyone" could be a CEO.
The problem is that you cannot measure these skills or really determine whether someone is 5% more competent than someone else. You can only measure the bottom line and in most cases the difference will be surprisingly small when you swap out a CEO or other high level exec for someone equally qualified (in terms of: knowing his craft).
This is what I mean when I say: no individual can really constantly make such a big difference.
Especially not in a big company - and we're talking about big when we talk about 7-digit salaries.
My (idealistic) proposal still stands: Pay them huge bonuses when their decisions bring success to the company. Pay them high salaries if you must but tie them to measurable success, nothing else. Ofcourse that's not how our system works and will never become reality. Greed is rooted very deeply in human nature even when the only difference between owning one yacht or a whole fleet is "do i have more than the other guy?".
I think you're missing relation and context here.
It's not uncommon for a CEO to earn 10x or even 100x the salary of an average employee.
The reason is not that he's adding 10x or 100x more value, the reason is because he can.
He's worked himself up (or got born into) the top of the food chain and that's his privilege: he can fire you, you can't fire him.
He can demand ridiculous salaries, you can not. He can sink your company but still get the golden parachute, you can't.
This is the common pattern, admittedly quite a bit simplified.
Nonetheless my point is: no single person can add >1000% value above average to a company constantly.
As far as I am concerned: Pay them big bonuses when they strike a hot deal.
But seven digit "salaries" are a [known] bug in our system.
Ofcourse, royal fuckups happen all the time.
In that particular case I guess they either didn't consult someone with a clue or they actually applied some of that sick "Total War" strategy.
Y'know the part where it says "we must patrol that area but don't have proper armor available, ah we'll do it anyways".
Yes, upper mgmt decides for a strategy but usually *after* consulting with the tacticians whether that strategy is viable at all.
That's partly the reason why not so many companies aim for world domination although upper management in almost every company would certainly love that.
Your business-reference is not so far off nonetheless - your mindset reminds me of the typical MBA/manager caricatures (dilbert).Congrats, you did just that.This oversimplification is so idiotic, I don't even know where to start. Strategies only make sense in context, after evaluating the benefits and drawbacks of a certain strategy in the given situation.
Claiming that a given strategy applies optimally to any given situation is well, to use your words and your spelling: rediculous.
So, let me get this straight:
You're proposing a strategy ("total war") as the only and ultimate strategy - but without the slightest idea of how to implement it?
Great call, but I'll suggest "Make love, not war" instead.
You didn't answer the question and Iraq doesn't remotely compare to any of these countries or the WWII situation.
You made it sound like you have the patent recipe for winning a war - comfortably from your couch, too.
So what does your ultimate strategy suggest? Nuke over baghdad?
I wouldn't consider this post worth responding to if it hadn't been modded "Insightful" - my ass. Mods didn't get their medis again?
So, what's your definition of "complete submission"? Is Iraq in "complete submission"? If it is not then what "total war"-steps do you propose to enforce "complete submission" there? Drop some nukes? Kill their children? Starve them?
Now you may go back to watching TV, just please keep your brainfarts to yourself next time.
Know what they did with complainers in Guantanamo?
They shouted at them: "RTFM!"