Thanks for the link -- the space.com images are very low resolution (even when you click on the "full size image" link); the real images are MUCH better!
Yes, this type of inaccurate naming is a common practice -- it does not make it a good one.
Any programmer working for me that writes "toIndex" and actually means "uptoButNotIncludingIndex" will be instructed to write what they mean, not what they can be bothered to type.
To not do so makes the code unnecessarily cryptic, and is the cause of many bugs.
I recently came across a flag that had been named "mustDoX", which actually meant "canDoX". If I believed the meaning of the name, the code was wrong. Turned out the name was wrong. Getting this sort of thing right is really important. If you're happy to have names that are "conveniently" related to English, rather than "accurate", the resulting source is poor, deserving the moniker "code". Good quality source does is not "code" to be decrypted!
The code might be "correct" but the terminology used in its interface does not match what the code does. Abusing the English language because its too hard to come up with unambiguous accurate descriptors makes for lousy code. Just because there are loads of APIs that follow this pattern doesn't make them right, if they are using words that mean something else entirely. There's nothing wrong with the "last exclusive one" pattern, but you need to make it obvious in your choice of words that this is what is intended. There are a loads of APIs that have misleading interfaces. Take STL's std::numeric_limits<int>::min() -- returns the minimum value representable by an int, a negative value. What do you think std::numeric_limits<float>::min() returns? Surprisingly, it is not negative. Misleading junk!
private static void rangeCheck(int arrayLen, int fromIndex, int toIndex) {
if (fromIndex > toIndex)
throw new IllegalArgumentException( "fromIndex(" + fromIndex + ") > toIndex(" + toIndex+")" );
if (fromIndex < 0)
throw new ArrayIndexOutOfBoundsException(fromIndex);
if (toIndex > arrayLen)
throw new ArrayIndexOutOfBoundsException(toIndex); }
This is crap. Either 'toIndex' actually means 'onePastToIndex', or the final if test should be if (toIndex >= arrayLen) . Try calling this function with arrayLen=10, fromIndex=1, toIndex=10 -- no exception, despite the zero-based array's last element being index 9. If you ask somebody to count from 1 to 10, do they stop at 9? If you see the following prototype: int randomValue( int fromValue, int toValue), what is the highest value you would expect could be returned?
Identifiers should be named accurately! Bad identifiers like this make code unnecessarily difficult to read, and makes it easier for bugs to go unnoticed.
If the allegation is that Google copied this code, and they've replicated the bad names along with the trivial algorithm, it does seem likely they authored their version via cut & paste...
Oh please! It should be twice the speed if it has half the latency, not half the speed. Speed and latency are related, but not interchangeable synonyms! If a cache has "half the speed" of your uncached memory, you need to disable that cache ASAP!
At 1.6GHz, L4 latencies are 50-60ns which is significantly higher than the L3 but just half the speed of main memory.
Hmmm. L4 cache runs at half the speed of main memory? That doesn't seem right Why bother reading these summaries? The people posting them certainly don't
Anybody know an easy way to get Slashdot to filter out *all* patent related stories? They are always ridiculous, even when they're accurate!
I had enough of this crap when I wasn't allowed to make a protocol I was implementing work as efficiently as it could, because Motorola had a patent on the concept of "Pardon? Could you repeat yourself please?" in this particular context. Utterly freaking ridiculous.
The current patent system does not encourage innovation -- it encourages taking out patents.
I remember trying DVD rentals when these shiny new DVD thingies came out. Trouble was, 1/3 of the movies I rented were ruined due to the discs being scratched to death. In my experience VHS tape survives typical renter mistreatment a lot better than optical discs. Not to mention all the unskippable shit. I presume there are now DVD players that ignore the 'do not skip' flag. Ugh. DVD rental stores drove me away long ago.
Maybe your friend should try renting movies on VHS, and don't forget betamax! DVDs are cheap and plentiful, whereas new release movies on betamax are hard to come by. Supply and demand, you see. Now all you need to do is convince people that watching movies stored on magnetic tape is really the only way to truly enjoy them.
Magnetic tape? What am I thinking!? Your friend should rent out the latest release movies on spools of film.
You also get to rent out the projectors, screens, piano-players, etc, that you need, remembering that all of this is a loss-leader to sell organic popcorn.
While this seems like a small victory for common sense, don't forget that Conroy wants the ISPs to store years of traffic for every customer. Rather than prevent a site from working (via blocking against a secret government list), they want to be silently collecting "evidence" instead.
A proper weather satellite would control the weather, rather than simply observe it. Then I could write my name in snow, across an entire continent. Muhahahaha.
If I invest years of my own time and hundreds of thousands of my own money, why shouldn't I be able to arrange things to recoup an actual profit for that?
Doesn't that assume that charging an yearly fee is the only way to achieve that? On what is that based?
I don't follow. I was just trying to point out that I should have a right to use technology to defend my revenue stream. I don't see it as any different to having staff and a cash register in a shop! It's designed to protect my business model. Whether circumventing that protection should be legal or not is another matter entirely.
I'm sorry, what indignation? Who said it was wrong?
...
What's your point?
Apologies for the misunderstandings!
I'm sure you can think of plenty of money-saving software that doesn't come with an yearly fee.
Sure, and I could point out that the yearly fees for this software also include training, support, requested improvements, and so on, so not too different to you!:-) Software companies charging for 'upgrades' each year or two is pretty common, too. I'm not saying it's right or wrong.
This solution works for us - it doesn't mean I'm an expert in this issue.
Expert or no - thanks for sharing your perspective. I appreciate it and applaud you making a living with open source. FWIW I now consider the crowd-funded-GPL model more viable than I did yesterday.
I can't disagree with that, but that's a *really* narrow definition of fair. Also, there have been cases in Australia recently where "the trade did not happen", and subsequently the farmer committed suicide.
Any pointers on how to convert "what we've got" into a "free market economy where nobody gets screwed?"
You don't seem to get it, frankly.....you could use a different business model.
Yes, that's why I'm trying to discuss different business models with you. I am not espousing a position here - I am exploring the space and trying to learn new things!
If you could show how the "very-expensive-tool-it-took-20-man-years-to-make-for-a-small-audience-that-they-didn't-know-they-wanted-it-until-we-could-show-them-how-much-money-it-would-save" project could be reworked into a different business model that doesn't rely on artificial scarcity, I would be most appreciative.
Once the package you are writing GPL modules for becomes 'perfect' (or more likely 'good enough'), you've got no more income. What is the scaling factor between cost-of-producing-a-module and the one-time-single-fee you get for writing it? Do commissioned modules get added to the standard set available to everybody? Why pay for a feature when I can wait three months for somebody else to buckle and pay for it? Please - I am not trying to have an argument I am genuinely interested.
And why should people pay you annual fees?
Why shouldn't they? Why should banks pay interest on deposits? If I invest years of my own time and hundreds of thousands of my own money, why shouldn't I be able to arrange things to recoup an actual profit for that? If these algorithms allow other business to save themselves millions ever year, what's so wrong about trying to get some share of that? It's a win for both parties (and the environment) -- why the indignation? If you don't like the price - don't buy it, but also - don't use it. I guess we'll see a rise in building such applications directly into purpose built hardware -- certainly some of our competitors do build & sell hardware, and would love to have our technology built in - we will hopefully be licensing some of this soon. We couldn't compete against *any* of these companies if we had GPL'd our goodies, and there's no way we've got deep enough pockets to win a patent war. Lets not even talk about software patents.
computer saves me a lot of time and money every year, but I don't pay a yearly fee to HP.
I thought we were talking about alternatives to artificial scarcity - last time I looked computers were physical items.
Piracy doesn't kill game studios, lack of customers does.
Indeed - concisely put. The studio I was referring to didn't die due to piracy - actually the GFC and the high australian dollar that followed had a lot more to do with it - the "new management" were just the final nail in the coffin..., but that's a different story entirely. It would be interesting to see how these businesses would have played out without piracy - but this is impossible and unmeasurable so its moot anyway.
Again, different business models are needed...
Yes, that's what I'm saying, and looking for!! I'm NOT saying artificial scarcity is the one true business model or that nobody has a right to try to undermine it.
Double Fine Adventure got $3.3 million...
It's great to see kickstarter type projects gaining some traction - getting development totally funded before release is definitely a good model, if you can pull it off. Not applicable to my expensive-tool example above, though. Celebrities with a single hit under their belts though, should be able to crowd source their funding nicely.
Also seems like a good way to throw up a dozen potential ideas and see which ones gain interest before you've had to build anything. I can see Apple getting pissed at developers that crowd source funding for free iOS releases -- developer gets paid, everybody gets the App, Apple doesn't get their 30%. I don't see how this is different to any other pay-a-developer-to-make-a-free-app model that is extremely common, so I can't see Apple doing anything about it, ei
I don't see how this could be applied to any of the projects I've ever worked on while still putting food on the table, but hey, good on you for making it work!
For example -- I helped to create a tool that saves users >$500K for every project they use it on. If this tool wasn't artificially scarce, we would find it difficult to convince people to renew their annual licence fees -- and the industry involved isn't exactly known for giving people money out of the goodness of their hearts. When your business model is based on the usefulness of your secret source -- I don't see an alternative to artificial scarcity.
It's pretty annoying when you've just spent two years working on a console game to see it show up on torrent sites days before it is even available in the stores. As a salaried employee I was paid to make it just the same, and I do appreciate that a pirated copy doesn't equal a lost sale - it's a complex issue - but I also appreciate that the game studio I was working for no longer exists...
This is why there's been a huge rise in subscription models that rely on the internet to work -- e.g. World of Warcraft doesn't have a problem with piracy -- there are loads of other problems that arise from that, but freely copying the client software around isn't a problem for them. I am not looking forward to these silly high-latency high-bandwidth cloud gaming services that are in the pipe... ugh! There has to be a better way!
a free market economy IS an example of a capitalist economy where no one gets screwed
That's the funniest thing I've seen all day! This is some pretty impressive system that guarantees that every single trade is completely fair and equitable -- unlikely in the real world, don't you think? When one man gets paid 10,000 times as much as another, somebody is definitely getting screwed!
Capitalising the word "IS" isn't much of an argument to warrant a "Therefore", really.
I'm surprised at all the surprise?! I thought it was pretty common knowledge that Huawei and ZTE were run and funded by the Chinese Military. They have been using their financial muscle to undercut and bribe their equipment into as many countries telecoms infrastructure as they possibly can for over five years now.
Slashdot Mirror
of course the guy that put up the street signs and the house numbers are responsible for what goes on inside the buildings; makes perfect sense to me.
Also, fuck beta.
They've basically written off slashdot as worthless
Should be cheap to buy it off them, then!
Nah Hitler's singular plum was British propaganda; Chairman Mao on the other hand...
http://en.wikipedia.org/wiki/H...
http://en.wikipedia.org/wiki/M...
The Nazis would've won if the Allies used Colossus Beta.
Beta Sucks Hilter's Balls.
Yep, Sadly the current Atari is simply Infogrames rebadged.
Infogrames bought the name, trademarks, and other IP.
Thanks for the link -- the space.com images are very low resolution (even when you click on the "full size image" link); the real images are MUCH better!
Yes, this type of inaccurate naming is a common practice -- it does not make it a good one.
Any programmer working for me that writes "toIndex" and actually means "uptoButNotIncludingIndex" will be instructed to write what they mean, not what they can be bothered to type.
To not do so makes the code unnecessarily cryptic, and is the cause of many bugs.
I recently came across a flag that had been named "mustDoX", which actually meant "canDoX". If I believed the meaning of the name, the code was wrong. Turned out the name was wrong. Getting this sort of thing right is really important. If you're happy to have names that are "conveniently" related to English, rather than "accurate", the resulting source is poor, deserving the moniker "code". Good quality source does is not "code" to be decrypted!
The code might be "correct" but the terminology used in its interface does not match what the code does.
Abusing the English language because its too hard to come up with unambiguous accurate descriptors makes for lousy code.
Just because there are loads of APIs that follow this pattern doesn't make them right, if they are using words that mean something else entirely.
There's nothing wrong with the "last exclusive one" pattern, but you need to make it obvious in your choice of words that this is what is intended.
There are a loads of APIs that have misleading interfaces.
Take STL's std::numeric_limits<int>::min() -- returns the minimum value representable by an int, a negative value.
What do you think std::numeric_limits<float>::min() returns? Surprisingly, it is not negative. Misleading junk!
From the TFA:
private static void rangeCheck(int arrayLen, int fromIndex, int toIndex) {
if (fromIndex > toIndex)
throw new IllegalArgumentException( "fromIndex(" + fromIndex + ") > toIndex(" + toIndex+")" );
if (fromIndex < 0)
throw new ArrayIndexOutOfBoundsException(fromIndex);
if (toIndex > arrayLen)
throw new ArrayIndexOutOfBoundsException(toIndex);
}
This is crap. Either 'toIndex' actually means 'onePastToIndex', or the final if test should be if (toIndex >= arrayLen) .
Try calling this function with arrayLen=10, fromIndex=1, toIndex=10 -- no exception, despite the zero-based array's last element being index 9.
If you ask somebody to count from 1 to 10, do they stop at 9?
If you see the following prototype: int randomValue( int fromValue, int toValue), what is the highest value you would expect could be returned?
Identifiers should be named accurately! Bad identifiers like this make code unnecessarily difficult to read, and makes it easier for bugs to go unnoticed.
If the allegation is that Google copied this code, and they've replicated the bad names along with the trivial algorithm, it does seem likely they authored their version via cut & paste...
Oh please! It should be twice the speed if it has half the latency, not half the speed. Speed and latency are related, but not interchangeable synonyms!
If a cache has "half the speed" of your uncached memory, you need to disable that cache ASAP!
At 1.6GHz, L4 latencies are 50-60ns which is significantly higher than the L3 but just half the speed of main memory.
Hmmm. L4 cache runs at half the speed of main memory? That doesn't seem right Why bother reading these summaries? The people posting them certainly don't
Reminds me of the old Goodies' joke: First prize is dinner with X; Second prize is TWO dinners with X
I dutifully flagged this as a dupe in the firehose before it made it to the front page. Lot of good it did!
If we all ignore them, wont they go away?
Anybody know an easy way to get Slashdot to filter out *all* patent related stories? They are always ridiculous, even when they're accurate!
I had enough of this crap when I wasn't allowed to make a protocol I was implementing work as efficiently as it could, because Motorola had a patent on the concept of "Pardon? Could you repeat yourself please?" in this particular context. Utterly freaking ridiculous.
The current patent system does not encourage innovation -- it encourages taking out patents.
Hint -- follow the money.
I remember trying DVD rentals when these shiny new DVD thingies came out.
Trouble was, 1/3 of the movies I rented were ruined due to the discs being scratched to death.
In my experience VHS tape survives typical renter mistreatment a lot better than optical discs.
Not to mention all the unskippable shit. I presume there are now DVD players that ignore the 'do not skip' flag. Ugh.
DVD rental stores drove me away long ago.
Maybe your friend should try renting movies on VHS, and don't forget betamax!
DVDs are cheap and plentiful, whereas new release movies on betamax are hard to come by.
Supply and demand, you see. Now all you need to do is convince people that watching movies stored on magnetic tape is really the only way to truly enjoy them.
Magnetic tape? What am I thinking!? Your friend should rent out the latest release movies on spools of film.
You also get to rent out the projectors, screens, piano-players, etc, that you need, remembering that all of this is a loss-leader to sell organic popcorn.
While this seems like a small victory for common sense, don't forget that Conroy wants the ISPs to store years of traffic for every customer.
Rather than prevent a site from working (via blocking against a secret government list), they want to be silently collecting "evidence" instead.
A proper weather satellite would control the weather, rather than simply observe it.
Then I could write my name in snow, across an entire continent.
Muhahahaha.
If I invest years of my own time and hundreds of thousands of my own money, why shouldn't I be able to arrange things to recoup an actual profit for that?
Doesn't that assume that charging an yearly fee is the only way to achieve that? On what is that based?
I don't follow. I was just trying to point out that I should have a right to use technology to defend my revenue stream. I don't see it as any different to having staff and a cash register in a shop! It's designed to protect my business model. Whether circumventing that protection should be legal or not is another matter entirely.
I'm sorry, what indignation? Who said it was wrong?
What's your point?
Apologies for the misunderstandings!
I'm sure you can think of plenty of money-saving software that doesn't come with an yearly fee.
Sure, and I could point out that the yearly fees for this software also include training, support, requested improvements, and so on, so not too different to you! :-)
Software companies charging for 'upgrades' each year or two is pretty common, too. I'm not saying it's right or wrong.
This solution works for us - it doesn't mean I'm an expert in this issue.
Expert or no - thanks for sharing your perspective.
I appreciate it and applaud you making a living with open source.
FWIW I now consider the crowd-funded-GPL model more viable than I did yesterday.
Thanks!
Fair enough, for some values of the word 'screwed' :-)
If it is that bad, we must fix it.
Absolutely! We must. With what or how, I've no idea...but it will be really obvious in hindsight, if we ever get there.
I can't disagree with that, but that's a *really* narrow definition of fair.
Also, there have been cases in Australia recently where "the trade did not happen", and subsequently the farmer committed suicide.
Any pointers on how to convert "what we've got" into a "free market economy where nobody gets screwed?"
Cheers.
You don't seem to get it, frankly.....you could use a different business model.
Yes, that's why I'm trying to discuss different business models with you.
I am not espousing a position here - I am exploring the space and trying to learn new things!
If you could show how the "very-expensive-tool-it-took-20-man-years-to-make-for-a-small-audience-that-they-didn't-know-they-wanted-it-until-we-could-show-them-how-much-money-it-would-save" project could be reworked into a different business model that doesn't rely on artificial scarcity, I would be most appreciative.
Once the package you are writing GPL modules for becomes 'perfect' (or more likely 'good enough'), you've got no more income.
What is the scaling factor between cost-of-producing-a-module and the one-time-single-fee you get for writing it?
Do commissioned modules get added to the standard set available to everybody?
Why pay for a feature when I can wait three months for somebody else to buckle and pay for it?
Please - I am not trying to have an argument I am genuinely interested.
And why should people pay you annual fees?
Why shouldn't they? Why should banks pay interest on deposits?
If I invest years of my own time and hundreds of thousands of my own money, why shouldn't I be able to arrange things to recoup an actual profit for that?
If these algorithms allow other business to save themselves millions ever year, what's so wrong about trying to get some share of that?
It's a win for both parties (and the environment) -- why the indignation?
If you don't like the price - don't buy it, but also - don't use it.
I guess we'll see a rise in building such applications directly into purpose built hardware -- certainly some of our competitors do build & sell hardware, and would love to have our technology built in - we will hopefully be licensing some of this soon. We couldn't compete against *any* of these companies if we had GPL'd our goodies, and there's no way we've got deep enough pockets to win a patent war. Lets not even talk about software patents.
computer saves me a lot of time and money every year, but I don't pay a yearly fee to HP.
I thought we were talking about alternatives to artificial scarcity - last time I looked computers were physical items.
Piracy doesn't kill game studios, lack of customers does.
Indeed - concisely put. The studio I was referring to didn't die due to piracy - actually the GFC and the high australian dollar that followed had a lot more to do with it - the "new management" were just the final nail in the coffin..., but that's a different story entirely. It would be interesting to see how these businesses would have played out without piracy - but this is impossible and unmeasurable so its moot anyway.
Again, different business models are needed...
Yes, that's what I'm saying, and looking for!!
I'm NOT saying artificial scarcity is the one true business model or that nobody has a right to try to undermine it.
Double Fine Adventure got $3.3 million...
It's great to see kickstarter type projects gaining some traction - getting development totally funded before release is definitely a good model, if you can pull it off.
Not applicable to my expensive-tool example above, though.
Celebrities with a single hit under their belts though, should be able to crowd source their funding nicely.
Also seems like a good way to throw up a dozen potential ideas and see which ones gain interest before you've had to build anything.
I can see Apple getting pissed at developers that crowd source funding for free iOS releases -- developer gets paid, everybody gets the App, Apple doesn't get their 30%. I don't see how this is different to any other pay-a-developer-to-make-a-free-app model that is extremely common, so I can't see Apple doing anything about it, ei
Thanks for that.
I don't see how this could be applied to any of the projects I've ever worked on while still putting food on the table, but hey, good on you for making it work!
For example -- I helped to create a tool that saves users >$500K for every project they use it on. If this tool wasn't artificially scarce, we would find it difficult to convince people to renew their annual licence fees -- and the industry involved isn't exactly known for giving people money out of the goodness of their hearts.
When your business model is based on the usefulness of your secret source -- I don't see an alternative to artificial scarcity.
It's pretty annoying when you've just spent two years working on a console game to see it show up on torrent sites days before it is even available in the stores.
As a salaried employee I was paid to make it just the same, and I do appreciate that a pirated copy doesn't equal a lost sale - it's a complex issue - but I also appreciate that the game studio I was working for no longer exists...
This is why there's been a huge rise in subscription models that rely on the internet to work -- e.g. World of Warcraft doesn't have a problem with piracy -- there are loads of other problems that arise from that, but freely copying the client software around isn't a problem for them.
I am not looking forward to these silly high-latency high-bandwidth cloud gaming services that are in the pipe... ugh!
There has to be a better way!
a free market economy IS an example of a capitalist economy where no one gets screwed
That's the funniest thing I've seen all day!
This is some pretty impressive system that guarantees that every single trade is completely fair and equitable -- unlikely in the real world, don't you think?
When one man gets paid 10,000 times as much as another, somebody is definitely getting screwed!
Capitalising the word "IS" isn't much of an argument to warrant a "Therefore", really.
I'm surprised at all the surprise?!
I thought it was pretty common knowledge that Huawei and ZTE were run and funded by the Chinese Military.
They have been using their financial muscle to undercut and bribe their equipment into as many countries telecoms infrastructure as they possibly can for over five years now.
fine, but they can make the first move...
My plague of self-replicating robot spiders have no interest in your copyright problems. It is in their interest to be copied.