In addition to what you said, population growth (of this whole planet) is much slower than bandwidth growth... so long term, we'll all have ``unlimited'' (for all practical definitions of ``unlimited'') bandwidth.
So all these "uh, oh, in the FUTURE, the Internet will implode under the load" is just bull.
This isn't a "resource" that needs to be protected behind a price (it shouldn't be more expensive than the electricity to move those bits around).
This probably isn't the best solution for relational databases...
Actually... that's exactly the architectures many enterprise scale relational databases take (not the transactional type ones, but data warehousing ones); which conceptually are mostly SQL interfaces to map-reduce...
Eh? GPL Perl module is not OK? How so? It's not something that YOU modify or distribute. You just distribute your script, that just happens to use the library that's already installed on the system.
That's like saying you won't use Linux, 'cause your non-GPL program might run on it.
Consider having Gmail shut down (or become too expensive for non-enterprise customers), and having to move to say... YahooMail, yet maintaining all contact lists, past emails, settings, etc., as they are. It's ridiculous until it happens... And yes, it's -possible- to back your email, somehow. But having it easily search-able within your email ``client'' is no longer an option (how do you import old emails into some random 3rd party webmail app?)
Exactly what I was thinking. Googling for the answer would provide the question (heck, `I'm feeling lucky' would probably get you the correct response in the title of the returned page).
You're right. But it's not even malice. I'd imagine most folks in finance don't even understand how those algorithms work.
For example, a programmer may not understand the financial basics of what they're programming (for example, they'd use normal distribution as opposed to lognormal distribution), while a financial dude may not understand what the program is doing (they may not have the ability to verify it's actually lognormal instead of normal distribution). QA folks are clueless in either case.
Separate the two groups by several years of job changes/promotions/lay-offs, etc., and you got a situation where nobody exactly knows how things happen at a financial corp (they just know that until recently "it worked" and now it doesn't).
Also, mis-pricing risk has the upside of higher short-term returns---which is what folks evaluating different models really care about. (you build a model, run it over the last 10 years, and... one of them comes out on top; chances are, the one that came out on top doesn't take care of long-term risk properly).
Universities have a big incentive to ensure that their graduates live up to expectations.
Long living universities,... maybe. Short lived online-universities, definitely not. They care about number of students. You cannot fail students, or else they drop out. And you wouldn't want a paying customer to stop paying, would you? So you fix the system to ensure you have as many students as possible---make a school even an idiot can graduate from... 'cause even idiots have money!
Also, education is -huge- business. Kinda like healthcare. Just when you think it can't possibly get more expensive, it just does, even when it defies logic about who can afford it. So online or not, education costs will go up.
In fact, I think the ``value'' of some networks can be negative. Just picture the jump in world productivity (added-value) if some networks (WoW? etc.) are shut down.
Desktop, I agree with you. But laptop... well, I'm still "stuck" with XP on the laptop due to boot speed (using laptop on subway, you don't want it to take 60 seconds to come out of hibernate---or to do other stupid things when turning on).
Emulating motives that are abstract enough to eventually lead back to our demise are quite complex
No it's not. Ignoring the engineering aspect (ie: to build a robot in the first place), it seems pretty easy to ask the robot to "solve world hunger" or "global warming", and have it figure out that "killing all humans" is the quickest solution.
Plus, it also hugely lowers the barriers to entry for other programmers.
Exactly! That's pretty much why I'm very excited about this. Especially since I've learned quite a bit from the original code around 1996-ish... this will be my 2nd learning from the wolf code... this time mostly the iphone related stuff.
Slashdot Mirror
Exactly!
In addition to what you said, population growth (of this whole planet) is much slower than bandwidth growth... so long term, we'll all have ``unlimited'' (for all practical definitions of ``unlimited'') bandwidth.
So all these "uh, oh, in the FUTURE, the Internet will implode under the load" is just bull.
This isn't a "resource" that needs to be protected behind a price (it shouldn't be more expensive than the electricity to move those bits around).
This probably isn't the best solution for relational databases...
Actually... that's exactly the architectures many enterprise scale relational databases take (not the transactional type ones, but data warehousing ones); which conceptually are mostly SQL interfaces to map-reduce...
human waste could be -very- dangerous if not properly treated. that's why you don't see anyone fertilizing fields (food stuff) with it.
Eh? GPL Perl module is not OK? How so? It's not something that YOU modify or distribute. You just distribute your script, that just happens to use the library that's already installed on the system.
That's like saying you won't use Linux, 'cause your non-GPL program might run on it.
Consider having Gmail shut down (or become too expensive for non-enterprise customers), and having to move to say... YahooMail, yet maintaining all contact lists, past emails, settings, etc., as they are. It's ridiculous until it happens... And yes, it's -possible- to back your email, somehow. But having it easily search-able within your email ``client'' is no longer an option (how do you import old emails into some random 3rd party webmail app?)
Exactly what I was thinking. Googling for the answer would provide the question (heck, `I'm feeling lucky' would probably get you the correct response in the title of the returned page).
You're right. But it's not even malice. I'd imagine most folks in finance don't even understand how those algorithms work.
For example, a programmer may not understand the financial basics of what they're programming (for example, they'd use normal distribution as opposed to lognormal distribution), while a financial dude may not understand what the program is doing (they may not have the ability to verify it's actually lognormal instead of normal distribution). QA folks are clueless in either case.
Separate the two groups by several years of job changes/promotions/lay-offs, etc., and you got a situation where nobody exactly knows how things happen at a financial corp (they just know that until recently "it worked" and now it doesn't).
Also, mis-pricing risk has the upside of higher short-term returns---which is what folks evaluating different models really care about.
(you build a model, run it over the last 10 years, and... one of them comes out on top; chances are, the one that came out on top doesn't take care of long-term risk properly).
But if we detect they're right handed, we can at least be sure they won't eat us (or rather, won't get any nutritional value from eating us :-)
Universities have a big incentive to ensure that their graduates live up to expectations.
Long living universities,... maybe. Short lived online-universities, definitely not. They care about number of students. You cannot fail students, or else they drop out. And you wouldn't want a paying customer to stop paying, would you? So you fix the system to ensure you have as many students as possible---make a school even an idiot can graduate from... 'cause even idiots have money!
Also, education is -huge- business. Kinda like healthcare. Just when you think it can't possibly get more expensive, it just does, even when it defies logic about who can afford it. So online or not, education costs will go up.
Because an HR drone will discard your resume because you don't have a degree?
They'll prolly disregard online degrees as well, kinda like they do now mostly.
Never thought of carrying along a dictaphone?
But... I'm sooo boring to listen to!
One of the funny things, Unions, if you want to be a member, well, you are free to sign up for it.
Yeah... you're paying the dues anyway... whether you're a member or not.
You have the right to be pissed off.
carnivore?
Two wrongs don't make a right...
But three lefts do! :-)
...and the world goes into slow motion :-)
In fact, I think the ``value'' of some networks can be negative. Just picture the jump in world productivity (added-value) if some networks (WoW? etc.) are shut down.
ie: useless definition.
...some important people close to the king.
Those are the ones who likely curse the king most anyway...
Desktop, I agree with you. But laptop... well, I'm still "stuck" with XP on the laptop due to boot speed (using laptop on subway, you don't want it to take 60 seconds to come out of hibernate---or to do other stupid things when turning on).
Entropy only makes sense in a closed system, which we've never actually observed yet. The universe may be a lot stranger than we think.
and remembering when IBM and Sun were arch-rivals
Indeed. Even product names are anti-Sun... Eclipse, etc.
Emulating motives that are abstract enough to eventually lead back to our demise are quite complex
No it's not. Ignoring the engineering aspect (ie: to build a robot in the first place), it seems pretty easy to ask the robot to "solve world hunger" or "global warming", and have it figure out that "killing all humans" is the quickest solution.
Or the killbot "kill threshold".
Plus, it also hugely lowers the barriers to entry for other programmers.
Exactly! That's pretty much why I'm very excited about this. Especially since I've learned quite a bit from the original code around 1996-ish... this will be my 2nd learning from the wolf code... this time mostly the iphone related stuff.
Urgh. I still use Lotus Notes every freaking day. Urgh!