No, it's not an exclusive-or choice. Gehry is just simply a shitty architect. He's making big sculptures instead of what a good architect should do - innovative new buildings that look good, make a statement AND make a pleasure to use the building through the master architect's good solutions.
We need a good micropayment system. It won't fucking happen though until Windows exists with their million botnet mess, because noone in their right mind would put out a micropayment system where someone can just syphon millions of dollars by having a large botnet.
One guy who I know from IRC sent a postcard when he was visiting the USA, to his german friend. The post was delivered by the german police to his friend. Funny eh.
Re:Average people don't need PCs
on
The Dying PC Market
·
· Score: 2, Insightful
People referring to average people tend to undershoot what average people need and underestimate the average person. The sooner we stop dealing with averages, the better.
There is a quite good talk that summarizes this in another context. It is worth watching in my opinion. The relevant gist of it is that we shouldn't cater for the average or "the biggest group" because the average is usually, only a relative majority of the market.
What we should be doing is to look for clusters of users, not just the biggest cluster you can generally find, and say: "Hey! If we try to follow the mythical average, a lot of people won't like it! They'll be left unhappy and their needs aren't catered to, even though they are not small groups. Let's find out what our users really want, take the top 4-5 clusters and market to those groups independently. This will cover 90-95% of the people on the market instead of 45-65%!".
I think murder and genocide or running a dictatorship - which I do consider a crime against humanity - are quite separate categories. Murder threatens individuals and a "crime against humanity" threatens societies.
If someone would take it's two minutes in order to check out the article, then it would be quickly realised that the _abstract_ of the actual paper is more detailed than the whole article linked in the summary and it is also free of the stupid sensationalization.
I want fugitives, whereever they may be and however long time has passed, to fear that some day they'll be found out and brought to justice. Within a reasonable balance of catching them, bringing them to trial and making sure they don't escape in the first place, that is.
I want all crimes except genocide or crimes against humanity to expire in 20 years tops. "Fugitives" are humans most of all and if they managed to keep themselves out of the hands of law enforcement for 20 years and didn't commit any crime apart from the one that expired (and I would say jailbreak is not a continous but singular event), then you can say that pursuing those crimes is probably a colossal waste of money and time.
What would be the justification for not allowing crimes to lapse? It is hardly a deterrent, to say that if I didn't catch you for 20 years, I will try and catch you later. The single fact that this woman was caught so late and it makes the news is an indication of how rare this event is. However, in 20 years or more worldviews, laws, court practice changes and people change. If someone committed a crime 20 years ago, but has lived a law abiding life since (apart from being a fugitive of course), then I see no reason why to waste money and resources, plus ruin a person's life. "Justice" is not about revenge: it should be about rehabilitation and deterring people from committing crimes.
Revenge would dictate to haul someone's ass back into prison whenever they are caught, but that serves no purpose whatsoever apart from revenge. I say, that if 20 years from now on someone is still free and there is no record of him ever committing a crime again, we should just let him/her enjoy his freedom. The law enforcement failed here and a crime is not the most defining quality of a human being, so why should we be punishing someone much later in his life - just because the law enforcement failed to do it's duty and apprehend him/her?
I agree with what you've said, just not the site you've linked. Crichton is not a scientist and spreads FUD about climate change, just from the opposite blindsided viewpoint.
I would've attacked the issue of terrorism the exact opposite. I'd tell everyone to grow some balls, carry a fucking gun, ask suspicious people questions, and be vigilant. And.. everyone can carry whatever the hell they want on a plane. I'd like to see a terrorist just try to hijack a plane when who knows how many people are armed...
True, that would probably stop a few aspiring terrorist the first few times, but of course 19 times out of 20 you'd get people shooting at other perfectly normal people for whatever reasons. Increased anxiety is well documented when you're packed with a lot of other people on an airplane.
I was a bit vague and this is why I stated it's a corner case. It is entirely plausible that someone uses a reasonably full hard drive and updates some files a lot. Consider the scenario of simple atime updates, databases, logs, package management, emails...
This is what I ment by a corner case. I was actually quite generous since I didn't presuppose utilities that check whether a file has changed every second. The bottom line is, I can see the drive failing under relatively uncommon but normal operating conditions.
Let's imagine a daemon that updates a file every 5 seconds. That's 518400 writes a month. Most flash based storage devices are rated between 100k to 1 million writes.
Your example usage and my example usage are the two corner cases, the endurance problem and it's effect depends on what the drive is used for. Still, I just wanted to illustrate that your scenario is an optimistic one.
...is like worrying about "vulnerabilities" exploitable only by root. It's cosmetics. They can revoke this law and bring a new one that taxes the net if they wish so. Banning themselves from doing something at a later time is totally futile when they can just change their mind.
Seconded. The Nobel Prize in Economics is NOT a real Nobel, and is awarded by the socialist Swedish central bank. Their awards are biased.
Honestly any economist who doesn't recognize the value of creating and protecting intellectual property rights in an information economy is a POORLY trained economist. Hernando de Soto has pegged a lack of real property rights as the primary issue that prevents wealth from being created in the third world (agricultural economies). It follows that in economies (such as the US/Europe) which derive their wealth, more and more, from intellectual property, that the ability to protect those rights is ultimately to our benefit.
Not to rain on your parade but you seem pretty biased or to be more accurate parroting certain right-wing dogmas. Intellectual property doesn't exist. It is an umbrella term at best, but mostly it is a word designed to mislead and cause confusion. The value created by creating and protecting patents and to a lesser extent copyrights is seriously outweighed by the bad effects of giving a monopoly on specific information or methods to private companies or persons.
Copyright and patents do not create wealth, they make all of us poorer, because they tend to inhibit progress as the professor who got the Nobel has shown in the specific case of patents and the IT industry.
Also, quoting wikipedia:
The Bank of Sweden Prize in Economic Sciences in Memory of Alfred Nobel (Sveriges Riksbanks pris i ekonomisk vetenskap till Alfred Nobels minne), usually called the "Nobel Prize in Economics", is a prize awarded each year for outstanding intellectual contributions in the field of economics. The prize is generally considered the most prestigious honor in economics.
Componentize them: turn them into shared libraries and load them into a single process. All panels, widgets, and other add-ons should be loaded into the window manager (or even the X11 server).
I originally wanted to include a paragraph about why separate processes/threads are good, but I forgot and when I realised I didn't want to reply to my own post. You've given me the excuse to hilight a few things about it.:)
Modularizing all boils down to multitasking and separation. The separation part is trivial in libraries and separate packages. Multitasking should be left for the OS to handle, it is very efficient at it, much more so than an in-process solution (not talking about OS supplied threading mind you). Threads inside a single process are not good when a library possibly needs/should have multithreading support, it is not that maintainable. Processes are cheap, the OS takes care of them, which is a good enough reason to use them for background processes.
The library method, where multiple libraries would work together and share and access common data structures in itself is a bit of a paradox. Libraries are commonly generalized pieces of code especially for the reason to factor out common elements from programs. They don't usually share that much common data structures with other libraries, ESPECIALLY when you're talking very general things, like using a common daemon for a lot of wildy different background processes. Creating a master background daemon would require a lot of glue code, with a very small common denominator between the different libraries.
Basically you'd code up the system in a few years and slowly optimize it, and one day you realise that you've reinvented process management in userspace. Matrioshka playing with low level OS functions no good is (try to pronounce the last sentence with a thick russian accent).
Slashdot Mirror
Congratulations! You've just found a page widening bug in slashcode! Now quick, bugreport it for credit.
No, it's not an exclusive-or choice. Gehry is just simply a shitty architect. He's making big sculptures instead of what a good architect should do - innovative new buildings that look good, make a statement AND make a pleasure to use the building through the master architect's good solutions.
We need a good micropayment system. It won't fucking happen though until Windows exists with their million botnet mess, because noone in their right mind would put out a micropayment system where someone can just syphon millions of dollars by having a large botnet.
One guy who I know from IRC sent a postcard when he was visiting the USA, to his german friend. The post was delivered by the german police to his friend. Funny eh.
People referring to average people tend to undershoot what average people need and underestimate the average person. The sooner we stop dealing with averages, the better.
There is a quite good talk that summarizes this in another context. It is worth watching in my opinion. The relevant gist of it is that we shouldn't cater for the average or "the biggest group" because the average is usually, only a relative majority of the market.
What we should be doing is to look for clusters of users, not just the biggest cluster you can generally find, and say: "Hey! If we try to follow the mythical average, a lot of people won't like it! They'll be left unhappy and their needs aren't catered to, even though they are not small groups. Let's find out what our users really want, take the top 4-5 clusters and market to those groups independently. This will cover 90-95% of the people on the market instead of 45-65%!".
I think murder and genocide or running a dictatorship - which I do consider a crime against humanity - are quite separate categories. Murder threatens individuals and a "crime against humanity" threatens societies.
If someone would take it's two minutes in order to check out the article, then it would be quickly realised that the _abstract_ of the actual paper is more detailed than the whole article linked in the summary and it is also free of the stupid sensationalization.
What would be the justification for not allowing crimes to lapse? It is hardly a deterrent, to say that if I didn't catch you for 20 years, I will try and catch you later. The single fact that this woman was caught so late and it makes the news is an indication of how rare this event is. However, in 20 years or more worldviews, laws, court practice changes and people change. If someone committed a crime 20 years ago, but has lived a law abiding life since (apart from being a fugitive of course), then I see no reason why to waste money and resources, plus ruin a person's life. "Justice" is not about revenge: it should be about rehabilitation and deterring people from committing crimes.
Revenge would dictate to haul someone's ass back into prison whenever they are caught, but that serves no purpose whatsoever apart from revenge. I say, that if 20 years from now on someone is still free and there is no record of him ever committing a crime again, we should just let him/her enjoy his freedom. The law enforcement failed here and a crime is not the most defining quality of a human being, so why should we be punishing someone much later in his life - just because the law enforcement failed to do it's duty and apprehend him/her?
I agree with what you've said, just not the site you've linked. Crichton is not a scientist and spreads FUD about climate change, just from the opposite blindsided viewpoint.
Yeah exactly, there are 10000 cells created in the average body per second, even if DNA degradation is only a percent, it's very bad news.
Source?
The guy who leads the iPlayer project was responsible for WMP while he was working at Microsoft Europe.
I reckon we could do that.
I was a bit vague and this is why I stated it's a corner case. It is entirely plausible that someone uses a reasonably full hard drive and updates some files a lot. Consider the scenario of simple atime updates, databases, logs, package management, emails...
This is what I ment by a corner case. I was actually quite generous since I didn't presuppose utilities that check whether a file has changed every second. The bottom line is, I can see the drive failing under relatively uncommon but normal operating conditions.
Let's imagine a daemon that updates a file every 5 seconds. That's 518400 writes a month. Most flash based storage devices are rated between 100k to 1 million writes.
Your example usage and my example usage are the two corner cases, the endurance problem and it's effect depends on what the drive is used for. Still, I just wanted to illustrate that your scenario is an optimistic one.
Hurricanes are driven by an awful lot of energy, so I'm wondering whether this could work even if we're talking about strategic points...
...is like worrying about "vulnerabilities" exploitable only by root. It's cosmetics. They can revoke this law and bring a new one that taxes the net if they wish so. Banning themselves from doing something at a later time is totally futile when they can just change their mind.
To produce a product. In a broader sense to compete in order to get income and turn a profit, that is incentive enough.
Firefox did take over IE in a couple of European countries and it is above in 40%+ marketshare in a lot of others.
Copyright and patents do not create wealth, they make all of us poorer, because they tend to inhibit progress as the professor who got the Nobel has shown in the specific case of patents and the IT industry.
Also, quoting wikipedia:
Modularizing all boils down to multitasking and separation. The separation part is trivial in libraries and separate packages. Multitasking should be left for the OS to handle, it is very efficient at it, much more so than an in-process solution (not talking about OS supplied threading mind you). Threads inside a single process are not good when a library possibly needs/should have multithreading support, it is not that maintainable. Processes are cheap, the OS takes care of them, which is a good enough reason to use them for background processes.
The library method, where multiple libraries would work together and share and access common data structures in itself is a bit of a paradox. Libraries are commonly generalized pieces of code especially for the reason to factor out common elements from programs. They don't usually share that much common data structures with other libraries, ESPECIALLY when you're talking very general things, like using a common daemon for a lot of wildy different background processes. Creating a master background daemon would require a lot of glue code, with a very small common denominator between the different libraries.
Basically you'd code up the system in a few years and slowly optimize it, and one day you realise that you've reinvented process management in userspace. Matrioshka playing with low level OS functions no good is (try to pronounce the last sentence with a thick russian accent).