I think it was at a similar time to http://michaeldaw.org/news/news-221206 this, using TinyURL for storage (although I don't think tinyurl had the preview functionality back then?)
This isn't particularly original research, though. I forget the specifics a long time ago (and also the sources, I'm afraid!) but I remember seeing a piece of research years ago talking about how you could treat various systems as short-to-long term storage (a TCP packet aimed at a refusing source can let you store a tiny amount of information for a very short period of time, a url shortening service can store some data for you, a wiki, a guestbook, a slashdot comment, an image hosting site (steganographically insert some content in your image).
The only difference is that the people doing this research actually wasted some peoples bandwidth abusing their websites rather than just positing the concept
They're just automating the process of deciding what's new and interesting. Their sources are blaring it out onto the internet... if Google's only using them to form a consensus of what's interesting then I don't see what the problem could be
Probably, but when 1000 people want to play game X at one time you still need 1000 licenses at that time.
I'd imagine you'd get a licensing agreement with the publisher to charge whatever they want for the game, then subsidise that with a small cut of the subscription revenue based on how much people play that game. That way it's a major win for publishers if their games are successful (and it's basically free money, since most people using Future Cloud Gaming Service wouldn't otherwise be able to play the games because they lack the cash for the hardware
For the cost of the game, there's a huge cut of it which is (pointless and wasteful, IMO) marketing, production, having to cover the cost of returns (from game retailers when they can't sell all the copies they bought). They could probably charge half of the price of the game to subscribers and the publisher would be better off.
The other great thing is that you could offer demos which were the start of the game - 60 minutes playing Crysis, after which time your game is saved and you're asked if you want to buy it. Much cheaper than building a demo. And you're building some idea of progress into the experience - if they pay you, they can continue from where they were. If they don't, they don't get any further. The problem is for games where 60 minutes is just longer than they can hold a players interest with their mechanics...
(by the way, I'm against OnLive as with lots of other digital distribution systems - I want to own something. I want to be able to lend it to my friends. I want to sell it to someone if I want. I know publishers and developers hate the used games market but you've got to factor it into your sales and live with it. Book publishers seem to be able to survive, after all)
At work we have a pretty good experience with Pound - it's easy to set up & it load balances and will detect when one of your servers is down and stop sending traffic there. You can get hardware load balancing from people like F5 too.
If you're just starting out you'll probably want to start with software and then, if the load demands it, move to hardware
Machine-wise, we use cheap & not overly powerful 250 GBP, 1u servers with a RAID1; they'll die after a few years (but servers will need to be refreshed anyway) and they provide us with lots of options. They're all plugged into 2 gigabit switches
I'd like to start by pointing out that I'm not a music pirate - I've bought the sum total of 2 CDs in my life and I own coming up to 3000 songs bought from iTunes. I write code for a living so I'm a firm believer that it's Wrong[tm] to copy stuff without paying for it.
That said, if you've been on slashdot for 10 years you must know that correlation is not causation. You need to do research where you change one variable and observe the effect on the other.
Side-bar: in a psych class we were shown the compelling correlation between number of churches in an area and crime in that area. Does this prove that churchgoers are criminals? It ended up being the affluence of the area (of which churchgoing was an indication). There are other examples like number of bananas imported into a country and the level of bad mental health (bananas go up, mental health goes up).
Reasoning people are simply not going to listen to you when you bring up a correlation as your 'proof' (normal people on the street might, but certainly not someone who has a scientific background). Has it occurred to you that things like Rock Band and Guitar Hero are giving people what they want in their homes? It's not traditional karaoke but it's fulfilling a similar need. They've sold millions of copies & they let you get together in a group now
Even their low-cost machines like Macbooks and Mac Minis are pretty beautiful. Compare it to the design of a lot of other small PCs - there's too much going on. The mac mini design is so elegant and minimalist - like all Apple's stuff, as the AC beside me says, they spend a lot of time throwing away stuff you don't *need*. Boiling things down to their essence is a difficult job.
Indeed - if you strip the metadata stream of identical songs purchased by 2 different people then you get identical bitstreams. Removing the information is even more trivial - you can hex edit the e-mail address and timestamp.
Unfortunately the files aren't lossless, though, they're high quality (256kbit) AAC.
Timestamps can be changed. So, you are saying that one could distribute and get away with it because, though they left their email address in the file, they were able to alter the date/time stamp?
Moot point, of course, because if you were distributing your purchased music and wanted to get away with it (and had the knowedge of how to change the timestamp) then you'd just blank the details out and never even be a suspect in the first place
IANAL, obviously, but I suspect since it's not verifiable it'd be virtually worthless in court - anyone could have edited the tags at any point to say anything (or copied it from your machine without your knowledge/permission)
perhaps insurance companies will require that procedures over a certain value or with a certain risk factor have their (digital) files sent to another doctor (across the internet) for a consult... perhaps saving lives and certainly reducing the risk of malpractice claims for unnecessary procedures.
Or, more likely, the 2nd opinion doctors will realise the HMO calls on them a lot more often when they say "I don't see a tumor... probably just a smudge", giving them another excuse to delay/prevent another expensive payout.
I've mentioned it elsewhere but songs are also encoded with the purchase timestamp. So if you've no access to someone's files then you've essentially zero chance of getting the purchase timestamp right, even if you get the songs they own right.
I diff'd 2 non-DRM iTunes songs a while ago... they just list your e-mail address and purchase date in the metadata. And I suspect that's done client-side to simplify their cache system.
Memory management isn't a Hard Problem but it is a problem - making sure you free all your various allocations can be difficult. Knowing that someone magically frees your memory lets you keep your focus on what the code's *really doing*
The fundamentally hard thing in programming - understanding the problem, making models to reason about it & architecting a solution - is essentially unchanged. It's just black-boxing another boring aspect of programming.
You may get a few more incompetent programmers now that the barrier for entry is lower (why, you might say it's the same as the Assembly to C days! Real men code in assembly, after all...) - but they're pretty easy to weed out if you ask an architecture or process question that's obvious to someone who's accustomed to thinking things through.
Rather than contract work, where you're self-employed, temp agencies are essentially a CV/Resume bank for companies who want to hire someone for a relatively short period of time (usually a few months); on the whole they tend to be for low-pay low-skill positions - so I wouldn't imagine you'd have [m]any skilled coders/architects signed up since they'd know they can get a much better price by going to the companies directly to sell themselves as contractors or permanent employees
For Yahoo OpenIDs, when prompted for your OpenID just enter "yahoo.com". Yahoo then takes care of figuring out who you are and telling the relying party. I don't think you need to enable it, either (although I could be wrong)
Forcing the player to restarting huge segments at the smallest error is a very cheap way to make something "difficult".
well yes, except it has to be compared to the previous PoP mechanic, where if you died you could simply hold a button to reverse time - a much nicer mechanic. In the new Prince of Persia I often die on some ridiculously easy jump after not having had two feet on solid ground (which sets a new checkpoint) for a while, so I have to re-do a long running section again. I'd much rather just rewind time a few seconds and try that particular jump again (and, as a result, have the game's jumps be generally harder). Time rewinding is a cool mechanic (and I don't really care how they justify it in the context of the game - maybe they don't even have to justify it).
Braid did it quite nicely with rewinding time - there was no limit on how much you could rewind so you didn't have the worry that you'd run out of the 'time buffer'.
The one thing I do like about the new Prince of Persia is the removal of multi-enemy combat. I'd much rather have all combat removed, actually - it's just "block-block-hit-jump-magic-magic-magic-[goto START]"
So then you'd have to prove you don't watch Prime Minister's questions, BBC news, football, etc. ...
I suspect even now you'd get harassed and made to prove you never watch live broadcasts online if you opted not to pay the licence and got rid of your TV.
Actually, not so much in my experience. I'd heard of people having huge trouble with the TV licensing people but I sent them a letter saying that I didn't watch TV, I have a TV with games consoles plugged into it and they were welcome to come and see if they wanted - they sent me a letter confirming they'd received my letter & marked me as not having a TV. I've heard nothing since.
(Also, I believe it's their responsibility to prove you're watching TV without a license, not your responsibility to prove that you don't watch TV)
Except if your organisation works in Java they probably want their Java developers to be able to modify and extend your code... so having it written in a language they've never seen before, even if it's binary-compatible with Java, probably isn't what they want.
I'm a huge fan of developing code in solving problems in the language that's most appropriate but this just breaks down when not all programmers have the same level of experience / fexibility
If you were around to be told wouldn't you have the same chance of knowing about its release? I know I'm overanalysing the joke and breaking it but...:-P
Slashdot Mirror
I think it was at a similar time to http://michaeldaw.org/news/news-221206 this, using TinyURL for storage (although I don't think tinyurl had the preview functionality back then?)
This isn't particularly original research, though. I forget the specifics a long time ago (and also the sources, I'm afraid!) but I remember seeing a piece of research years ago talking about how you could treat various systems as short-to-long term storage (a TCP packet aimed at a refusing source can let you store a tiny amount of information for a very short period of time, a url shortening service can store some data for you, a wiki, a guestbook, a slashdot comment, an image hosting site (steganographically insert some content in your image).
The only difference is that the people doing this research actually wasted some peoples bandwidth abusing their websites rather than just positing the concept
They're just automating the process of deciding what's new and interesting. Their sources are blaring it out onto the internet... if Google's only using them to form a consensus of what's interesting then I don't see what the problem could be
I'd imagine you'd get a licensing agreement with the publisher to charge whatever they want for the game, then subsidise that with a small cut of the subscription revenue based on how much people play that game. That way it's a major win for publishers if their games are successful (and it's basically free money, since most people using Future Cloud Gaming Service wouldn't otherwise be able to play the games because they lack the cash for the hardware
For the cost of the game, there's a huge cut of it which is (pointless and wasteful, IMO) marketing, production, having to cover the cost of returns (from game retailers when they can't sell all the copies they bought). They could probably charge half of the price of the game to subscribers and the publisher would be better off.
The other great thing is that you could offer demos which were the start of the game - 60 minutes playing Crysis, after which time your game is saved and you're asked if you want to buy it. Much cheaper than building a demo. And you're building some idea of progress into the experience - if they pay you, they can continue from where they were. If they don't, they don't get any further. The problem is for games where 60 minutes is just longer than they can hold a players interest with their mechanics...
(by the way, I'm against OnLive as with lots of other digital distribution systems - I want to own something. I want to be able to lend it to my friends. I want to sell it to someone if I want. I know publishers and developers hate the used games market but you've got to factor it into your sales and live with it. Book publishers seem to be able to survive, after all)
16It really would!
At work we have a pretty good experience with Pound - it's easy to set up & it load balances and will detect when one of your servers is down and stop sending traffic there. You can get hardware load balancing from people like F5 too.
If you're just starting out you'll probably want to start with software and then, if the load demands it, move to hardware
Machine-wise, we use cheap & not overly powerful 250 GBP, 1u servers with a RAID1; they'll die after a few years (but servers will need to be refreshed anyway) and they provide us with lots of options. They're all plugged into 2 gigabit switches
I'd like to start by pointing out that I'm not a music pirate - I've bought the sum total of 2 CDs in my life and I own coming up to 3000 songs bought from iTunes. I write code for a living so I'm a firm believer that it's Wrong[tm] to copy stuff without paying for it.
That said, if you've been on slashdot for 10 years you must know that correlation is not causation. You need to do research where you change one variable and observe the effect on the other.
Side-bar: in a psych class we were shown the compelling correlation between number of churches in an area and crime in that area. Does this prove that churchgoers are criminals? It ended up being the affluence of the area (of which churchgoing was an indication).
There are other examples like number of bananas imported into a country and the level of bad mental health (bananas go up, mental health goes up).
Reasoning people are simply not going to listen to you when you bring up a correlation as your 'proof' (normal people on the street might, but certainly not someone who has a scientific background). Has it occurred to you that things like Rock Band and Guitar Hero are giving people what they want in their homes? It's not traditional karaoke but it's fulfilling a similar need. They've sold millions of copies & they let you get together in a group now
Even their low-cost machines like Macbooks and Mac Minis are pretty beautiful. Compare it to the design of a lot of other small PCs - there's too much going on. The mac mini design is so elegant and minimalist - like all Apple's stuff, as the AC beside me says, they spend a lot of time throwing away stuff you don't *need*. Boiling things down to their essence is a difficult job.
I think it means that the beautiful design of the machine is hidden away in the basement :)
Indeed - if you strip the metadata stream of identical songs purchased by 2 different people then you get identical bitstreams. Removing the information is even more trivial - you can hex edit the e-mail address and timestamp.
Unfortunately the files aren't lossless, though, they're high quality (256kbit) AAC.
Moot point, of course, because if you were distributing your purchased music and wanted to get away with it (and had the knowedge of how to change the timestamp) then you'd just blank the details out and never even be a suspect in the first place
IANAL, obviously, but I suspect since it's not verifiable it'd be virtually worthless in court - anyone could have edited the tags at any point to say anything (or copied it from your machine without your knowledge/permission)
Or, more likely, the 2nd opinion doctors will realise the HMO calls on them a lot more often when they say "I don't see a tumor... probably just a smudge", giving them another excuse to delay/prevent another expensive payout.
I've mentioned it elsewhere but songs are also encoded with the purchase timestamp. So if you've no access to someone's files then you've essentially zero chance of getting the purchase timestamp right, even if you get the songs they own right.
I diff'd 2 non-DRM iTunes songs a while ago... they just list your e-mail address and purchase date in the metadata. And I suspect that's done client-side to simplify their cache system.
Memory management isn't a Hard Problem but it is a problem - making sure you free all your various allocations can be difficult. Knowing that someone magically frees your memory lets you keep your focus on what the code's *really doing*
The fundamentally hard thing in programming - understanding the problem, making models to reason about it & architecting a solution - is essentially unchanged. It's just black-boxing another boring aspect of programming.
You may get a few more incompetent programmers now that the barrier for entry is lower (why, you might say it's the same as the Assembly to C days! Real men code in assembly, after all...) - but they're pretty easy to weed out if you ask an architecture or process question that's obvious to someone who's accustomed to thinking things through.
Rather than contract work, where you're self-employed, temp agencies are essentially a CV/Resume bank for companies who want to hire someone for a relatively short period of time (usually a few months); on the whole they tend to be for low-pay low-skill positions - so I wouldn't imagine you'd have [m]any skilled coders/architects signed up since they'd know they can get a much better price by going to the companies directly to sell themselves as contractors or permanent employees
For Yahoo OpenIDs, when prompted for your OpenID just enter "yahoo.com". Yahoo then takes care of figuring out who you are and telling the relying party. I don't think you need to enable it, either (although I could be wrong)
well yes, except it has to be compared to the previous PoP mechanic, where if you died you could simply hold a button to reverse time - a much nicer mechanic. In the new Prince of Persia I often die on some ridiculously easy jump after not having had two feet on solid ground (which sets a new checkpoint) for a while, so I have to re-do a long running section again. I'd much rather just rewind time a few seconds and try that particular jump again (and, as a result, have the game's jumps be generally harder). Time rewinding is a cool mechanic (and I don't really care how they justify it in the context of the game - maybe they don't even have to justify it).
Braid did it quite nicely with rewinding time - there was no limit on how much you could rewind so you didn't have the worry that you'd run out of the 'time buffer'.
The one thing I do like about the new Prince of Persia is the removal of multi-enemy combat. I'd much rather have all combat removed, actually - it's just "block-block-hit-jump-magic-magic-magic-[goto START]"
Actually, not so much in my experience. I'd heard of people having huge trouble with the TV licensing people but I sent them a letter saying that I didn't watch TV, I have a TV with games consoles plugged into it and they were welcome to come and see if they wanted - they sent me a letter confirming they'd received my letter & marked me as not having a TV. I've heard nothing since.
(Also, I believe it's their responsibility to prove you're watching TV without a license, not your responsibility to prove that you don't watch TV)
It'd be funny if it wasn't so true :-(
...you mean I can't just give all my users root access to shared machines? How will they survive?!
I don't know that they care too much - I submitted all work (code+text) to my university under open source licenses for this purpose
Except if your organisation works in Java they probably want their Java developers to be able to modify and extend your code... so having it written in a language they've never seen before, even if it's binary-compatible with Java, probably isn't what they want. I'm a huge fan of developing code in solving problems in the language that's most appropriate but this just breaks down when not all programmers have the same level of experience / fexibility
If you were around to be told wouldn't you have the same chance of knowing about its release? I know I'm overanalysing the joke and breaking it but... :-P