So what? At the time of scaling, which for video games has to be the time of playing in realtime, the output resolution is known. Sure, we can all come up with fairy-tale scenarios of 8-bit video games that will be recorded and re-displayed on big screens and watches, but in real life that just doesn't happen enough to be statistically significant.
You can't understand the difference between SCALING and VECTORIZING? Really? Scaling algorithms increase the number of pixels, but you're fundamentally still dealing with a raster image. A vectorized drawing is a whole different beast.
Your words are all true, but the reality of the situation is that the images are going to be displayed on a pixel-based device. The input is just pixels, and the eventual output is also just pixels. This approach, that internally uses vector graphics, appears to work incredibly well - but they're final results aren't better simply because they use vectors.
I'm saying I won't pay $50 for it. Slicing it up into minutes, days, months, years is a meaningless rationalization. Divide any number up over enough units and it appears small.
I prefer it strongly, but to save $50 I will cope with glossy.
I don't see how its meaningless - and I'm interested to hear why you do. Something that annoys (or pleases) you for 2 years, to my mind, has to be different than something that annoys or pleases you for 2 minutes, or even 2 hours - including that factor in a comparison is only reasonable.
The last time I used IntelliJ was for a 6 week gig, so I may not have given it enough attention - eg, I'm not sure how much was reconfigurable. At the time though, what I really despised was the way that it would automatically save every change I made to every file I touched. That in and of itself was a significant enough change in my preferred work habits that I never got used to it.
You sound like the car salesman who fought with me on my last car over $3000 on a $30k car, arguing it's "not that much", he lost the sale (and his commission) to his competition. $50 is $50, $3000 is $3000. $50 will buy me a piece of software, several DVDs, a steak dinner for my wife and I. I don't care how much you divide it up over a lifetime, money is still money.
And time is still time. If you're saying that you don't prefer matte to glossy enough to spend ten cents a day on it, even though you use it all day long, then I'd say that you don't significantly prefer matte to glossy in any meaningful way. That's not an unreasonable conclusion to make, especially when you state that you'd prefer a single (cheap) steak dinner for two to years of using a matte screen. When comparing two objects objectively, money makes a pretty handy ruler.
You can accomplish some of that through training. I hate external keyboards on laptops (just a personal preference - I prefer having the same environment wherever I am) but I have some wrist pain with traditional typing. I simply hold my hands at about a 30-35 degree angle, elbows out a bit, with a little rotation so that my thumbs are slightly higher than my pinkies.. My "home row", if you want to call it that, would be... checking... "qsdfjkl;" - pretty normal. It took a bit of retraining but I now touch-type just fine, on any random keyboard, with no issues. It might help that I have long fingers. On my MacBook, F6 is the only key that I actually have to move my wrists to reach.
An ideal ergonomic keyboard that maintained the current layout would be folded in half, with half the keys facing left and half facing right, on a triangular stand (blank face towards the typist. That just doesn't seem to be happening, so I'll stick with a reasonably versatile re-learning.
I want to observe that 'their models suggest that the planet is habitable'. Don't get all excited until their models are validated, verified, and well-tested. Until then, it could indeed turn out to just be that trick of curved space-time that brought in a few funny photons in the right place.
Even then, while they're prettier to look at, you're better off listening to the scientists rather than their models.
You're a system admin and your answer to security in the cloud is to obfuscate your filenames? Ye gods...
Its like locking your car doors. There are so many juicy targets out there that all you have to do is not be the low-hanging-fruit. Will obfuscating filenames stop a dedicated inspection of your data? Of course not. Will it stop a bored sysadmin looking for porn (the original example)? Probably, because there will be thousands of obvious targets to go after instead of yours. He's not interested in your porn, but rather some illicit customer porn.
Not everywhere needs to be Fort Knox to be reasonably safe from casual penetration.
Hershel Robinson is a long-term member of the CiviCRM community, runs a specialty hosting business for CiviCRM hosting called CiviHosting, and is also a freelance web developer specializing in Drupal and CiviCRM development.
Sounds like the perfect candidate to write an honest, hard-hitting book review:)
Have you seen a lot of white Christian suicide bombers and terrorists lately? Because aside from Tim McVeigh and a few nutbags bombing abortion clinics, I haven't seen many of them in the last 20 years....
So, if you ignore the vast majority of US terrorism activities over the last 20 years, including one of the most deadly (although not the most deadly) one, then there's not a lot of white Christian suicide bombers.
In case you hadn't noticed, we don't actually have that much terrorism in the US. Hell, for a population of just under 7 billion there really isn't that much worldwide. There wasn't 10 years ago, there wasn't 50 years ago, there probably won't be 10 years from now.
Reinventing every standard just because of a single act of terrorism is stupid. Ignoring the 2nd-most-lethal act of US terrorism completely because "everything changed" on 9/11 is also stupid.
Oddly enough I have the opposite view. If we're going to have security theater, it should at least be effective security theater, without having gaps in it a mile wide. If you're doing random searches "Except for well known group of people xyz" then there's no point in having them at all.
By the same token, if you're going to try to convince me that my accidentally-packed full-size toothpaste is dangerous, don't just drop it from a height of five feet into a big bucket of similarly "dangerous" items you'll stand next to comfortably for the rest of your shift. At least pretend.
Show me a profile of a modern terrorist that would have caught Timothy McVeigh, arguably one of the most lethal domestic terrorists ever. Just remember, he was white, had some college, got an honorable military discharge, Christian...
In all fairness, you clicked on a link which caused a big popup window to appear stating, "{APPNAME} wants to learn about all your stuff, and your friends, and write on your wall, before showing you what kind of beaver mustache you are. Mmmmkay?" to which you had to very explicitly say "APPROVE!!!" Its not like they're making it a big secret. How would you handle it, exactly?
They already have email access. In fact, their FAQ states that if your phone is b0rked you can authorize a new computer through an email process.
Besides, if you're logging on to Facebook through a new computer, maybe you don't want to pull up your email on the same new computer? Not everyone has webmail, you know. Besides, that also removes one of the two factors - instead of a password and a device, you now need two passwords. Very different.
So... rather then provide a fob or phone app to provide a "one-time" number that constantly changes, they'll SMS it to your phone. Well, it's not exactly instant and depending on network load can take a while (ok the 4 hour delays at new year are a bit of an exception from the norm). It seems to me that the "one-time" number has to remain valid for quite a while and every second would increase the vulnerability.
Meh. Simply adding the requirement - even if the codes never expired - would decrease the ability of a "password guesser" to gain access by a factor of several thousand (probably much more). Expiring the codes after a day would be just fine. Worrying about being 1,000,000 times more secure vs. only 10,000 times more secure is a silly reason to not do it the simple way.
Its because most people already have a mobile phone, and thus they can offer this for free. They already have email verification though the "I forgot my password" process, so that wouldn't be newsworthy. What's the alternative, sending everyone a SecureID card? Should every website make you carry a keyfob to use it?
Right. And maybe 1% of developers out there write code that is that critical. *Maybe* 1%. Most are writing web sites or internal applications that have nowhere near that kind of criticality.
Agreed. But fewer than 1% are writing solutions that require code-creativity. Most problems are well-understood, and even in systems that are creatively solving new problems, most code follows established patterns - or should, especially if you want it to still be working next year, or five years from now.
I think you nailed it. The only thing I would add to your list is a dedicated QA team. And perhaps the ability for senior developers to veto management decisions on features and functionality.
Vetoing features and functionality is often needed. What cannot happen with any success is vetoing business requirements. Admittedly those requirements should never be presented directly as functionality requests; still, writing beautiful elegant solutions to the wrong problems will put you out of business a lot faster than nasty crufty solutions that actually apply to the problem at hand.
Business users get to define the problem. Technical staff work with them to define an acceptable solution, and then implement it. Business staff then gets to explain the solution to the customers and find out what the next pain points are. Lather, rinse, repeat.
"What creativity is needed to code a crystal clear requirement or specification? Sorry I don't get it."
I sure as hell wouldn't hire you then - you sound like a standard issue glue coder doing lego brick style programming. Algorithms and solutions to tricky problems don't just invent themselves, someone with creative flair and intelligence needs to dream them up. Thats obviously not you.
What you're missing is that the vast majority of all problems arenot tricky. At least, not unless someone decides to make them so to satisfy their creative urges. Most companies are, indeed, solving very much the same problems as hundreds or thousands of other companies. The twists in business logic that (may) make them unique are often not even visible at low levels of software development.
This is the same urge that caused someone at a company I worked at (for a whole 6 weeks) to spend several months re-designing the way that a Java-based web application handled translation. Creative? Sure. Solved the company's technical problem slightly better than a "stock" solution would have? Yep. Worthwhile? Absolutely not. Nobody else knew how to deal with it, no other libraries would work with it, and it probably ended up costing the company around $100,000 in the end in lost time and productivity - while saving them a few nickels a day in unneeded efficiency.
Most problems are not unique. Learning to recognize patterns and apply conventional solutions to most things is a very valuable skill. Knowing where to apply conventional patterns to unconventional problems is even more helpful. Correctly identifying the 2-3 times a year you actually need a solid, creative solution to a really unique problem is priceless - and very hard to do if you haven't made them stand out from the sea of normal boring problems by glorifying those into things they didn't need to be.
But it also bars those who wish to tinker from ever doing so, for the sole reason of padding their profit margins.
Did you realize that if you develop apps for the iPhone you can install them on your iPhone without jailbreaking it? Without going through the appstore?
Admittedly it does stop some random 3rd person from easily installing code that some other random person wrote that uses undocumented APIs, thus avoiding the problem that Microsoft has experienced ever since Adobe bought so many copies of Charles Petzold's _Undocumented Windows_ back in the day... but it doesn't stop any true tinkerers from tinkering. Script kiddies, but not tinkerers.
I use LastPass because I want access to my passwords at work and Dropbox is blocked. LastPass does the same thing as KeePass+Dropbox, and I can access it from anywhere.
And now, apparently, so can everybody else! That is convenient.
Depends. Were you planning on doubling the number of wheels (actually the size of the contact patch, but similar) as the weight doubled? If so, then yes, double weight translates into double wear. If not then you now have a much more concentrated weight, which does much more damage to the roads.
Slashdot Mirror
So what? At the time of scaling, which for video games has to be the time of playing in realtime, the output resolution is known. Sure, we can all come up with fairy-tale scenarios of 8-bit video games that will be recorded and re-displayed on big screens and watches, but in real life that just doesn't happen enough to be statistically significant.
You can't understand the difference between SCALING and VECTORIZING? Really? Scaling algorithms increase the number of pixels, but you're fundamentally still dealing with a raster image. A vectorized drawing is a whole different beast.
Your words are all true, but the reality of the situation is that the images are going to be displayed on a pixel-based device. The input is just pixels, and the eventual output is also just pixels. This approach, that internally uses vector graphics, appears to work incredibly well - but they're final results aren't better simply because they use vectors.
I'm saying I won't pay $50 for it. Slicing it up into minutes, days, months, years is a meaningless rationalization. Divide any number up over enough units and it appears small.
I prefer it strongly, but to save $50 I will cope with glossy.
I don't see how its meaningless - and I'm interested to hear why you do. Something that annoys (or pleases) you for 2 years, to my mind, has to be different than something that annoys or pleases you for 2 minutes, or even 2 hours - including that factor in a comparison is only reasonable.
The last time I used IntelliJ was for a 6 week gig, so I may not have given it enough attention - eg, I'm not sure how much was reconfigurable. At the time though, what I really despised was the way that it would automatically save every change I made to every file I touched. That in and of itself was a significant enough change in my preferred work habits that I never got used to it.
Eclipse is certainly no pleasure to use, though.
You sound like the car salesman who fought with me on my last car over $3000 on a $30k car, arguing it's "not that much", he lost the sale (and his commission) to his competition. $50 is $50, $3000 is $3000. $50 will buy me a piece of software, several DVDs, a steak dinner for my wife and I. I don't care how much you divide it up over a lifetime, money is still money.
And time is still time. If you're saying that you don't prefer matte to glossy enough to spend ten cents a day on it, even though you use it all day long, then I'd say that you don't significantly prefer matte to glossy in any meaningful way. That's not an unreasonable conclusion to make, especially when you state that you'd prefer a single (cheap) steak dinner for two to years of using a matte screen. When comparing two objects objectively, money makes a pretty handy ruler.
You can accomplish some of that through training. I hate external keyboards on laptops (just a personal preference - I prefer having the same environment wherever I am) but I have some wrist pain with traditional typing. I simply hold my hands at about a 30-35 degree angle, elbows out a bit, with a little rotation so that my thumbs are slightly higher than my pinkies.. My "home row", if you want to call it that, would be ... checking ... "qsdfjkl;" - pretty normal. It took a bit of retraining but I now touch-type just fine, on any random keyboard, with no issues. It might help that I have long fingers. On my MacBook, F6 is the only key that I actually have to move my wrists to reach.
An ideal ergonomic keyboard that maintained the current layout would be folded in half, with half the keys facing left and half facing right, on a triangular stand (blank face towards the typist. That just doesn't seem to be happening, so I'll stick with a reasonably versatile re-learning.
I want to observe that 'their models suggest that the planet is habitable'. Don't get all excited until their models are validated, verified, and well-tested. Until then, it could indeed turn out to just be that trick of curved space-time that brought in a few funny photons in the right place.
Even then, while they're prettier to look at, you're better off listening to the scientists rather than their models.
You're a system admin and your answer to security in the cloud is to obfuscate your filenames? Ye gods...
Its like locking your car doors. There are so many juicy targets out there that all you have to do is not be the low-hanging-fruit. Will obfuscating filenames stop a dedicated inspection of your data? Of course not. Will it stop a bored sysadmin looking for porn (the original example)? Probably, because there will be thousands of obvious targets to go after instead of yours. He's not interested in your porn, but rather some illicit customer porn.
Not everywhere needs to be Fort Knox to be reasonably safe from casual penetration.
Sounds like the perfect candidate to write an honest, hard-hitting book review :)
Have you seen a lot of white Christian suicide bombers and terrorists lately? Because aside from Tim McVeigh and a few nutbags bombing abortion clinics, I haven't seen many of them in the last 20 years. ...
So, if you ignore the vast majority of US terrorism activities over the last 20 years, including one of the most deadly (although not the most deadly) one, then there's not a lot of white Christian suicide bombers.
Gotcha.
In case you hadn't noticed, we don't actually have that much terrorism in the US. Hell, for a population of just under 7 billion there really isn't that much worldwide. There wasn't 10 years ago, there wasn't 50 years ago, there probably won't be 10 years from now.
Reinventing every standard just because of a single act of terrorism is stupid. Ignoring the 2nd-most-lethal act of US terrorism completely because "everything changed" on 9/11 is also stupid.
Oddly enough I have the opposite view. If we're going to have security theater, it should at least be effective security theater, without having gaps in it a mile wide. If you're doing random searches "Except for well known group of people xyz" then there's no point in having them at all.
By the same token, if you're going to try to convince me that my accidentally-packed full-size toothpaste is dangerous, don't just drop it from a height of five feet into a big bucket of similarly "dangerous" items you'll stand next to comfortably for the rest of your shift. At least pretend.
Show me a profile of a modern terrorist that would have caught Timothy McVeigh, arguably one of the most lethal domestic terrorists ever. Just remember, he was white, had some college, got an honorable military discharge, Christian...
In all fairness, you clicked on a link which caused a big popup window to appear stating, "{APPNAME} wants to learn about all your stuff, and your friends, and write on your wall, before showing you what kind of beaver mustache you are. Mmmmkay?" to which you had to very explicitly say "APPROVE!!!" Its not like they're making it a big secret. How would you handle it, exactly?
They already have email access. In fact, their FAQ states that if your phone is b0rked you can authorize a new computer through an email process.
Besides, if you're logging on to Facebook through a new computer, maybe you don't want to pull up your email on the same new computer? Not everyone has webmail, you know. Besides, that also removes one of the two factors - instead of a password and a device, you now need two passwords. Very different.
So... rather then provide a fob or phone app to provide a "one-time" number that constantly changes, they'll SMS it to your phone. Well, it's not exactly instant and depending on network load can take a while (ok the 4 hour delays at new year are a bit of an exception from the norm). It seems to me that the "one-time" number has to remain valid for quite a while and every second would increase the vulnerability.
Meh. Simply adding the requirement - even if the codes never expired - would decrease the ability of a "password guesser" to gain access by a factor of several thousand (probably much more). Expiring the codes after a day would be just fine. Worrying about being 1,000,000 times more secure vs. only 10,000 times more secure is a silly reason to not do it the simple way.
Its because most people already have a mobile phone, and thus they can offer this for free. They already have email verification though the "I forgot my password" process, so that wouldn't be newsworthy. What's the alternative, sending everyone a SecureID card? Should every website make you carry a keyfob to use it?
Right. And maybe 1% of developers out there write code that is that critical. *Maybe* 1%. Most are writing web sites or internal applications that have nowhere near that kind of criticality.
Agreed. But fewer than 1% are writing solutions that require code-creativity. Most problems are well-understood, and even in systems that are creatively solving new problems, most code follows established patterns - or should, especially if you want it to still be working next year, or five years from now.
I think you nailed it. The only thing I would add to your list is a dedicated QA team. And perhaps the ability for senior developers to veto management decisions on features and functionality.
Vetoing features and functionality is often needed. What cannot happen with any success is vetoing business requirements. Admittedly those requirements should never be presented directly as functionality requests; still, writing beautiful elegant solutions to the wrong problems will put you out of business a lot faster than nasty crufty solutions that actually apply to the problem at hand.
Business users get to define the problem. Technical staff work with them to define an acceptable solution, and then implement it. Business staff then gets to explain the solution to the customers and find out what the next pain points are. Lather, rinse, repeat.
"What creativity is needed to code a crystal clear requirement or specification? Sorry I don't get it."
I sure as hell wouldn't hire you then - you sound like a standard issue glue coder doing lego brick style programming. Algorithms and solutions to tricky problems don't just invent themselves, someone with creative flair and intelligence needs to dream them up. Thats obviously not you.
What you're missing is that the vast majority of all problems arenot tricky. At least, not unless someone decides to make them so to satisfy their creative urges. Most companies are, indeed, solving very much the same problems as hundreds or thousands of other companies. The twists in business logic that (may) make them unique are often not even visible at low levels of software development.
This is the same urge that caused someone at a company I worked at (for a whole 6 weeks) to spend several months re-designing the way that a Java-based web application handled translation. Creative? Sure. Solved the company's technical problem slightly better than a "stock" solution would have? Yep. Worthwhile? Absolutely not. Nobody else knew how to deal with it, no other libraries would work with it, and it probably ended up costing the company around $100,000 in the end in lost time and productivity - while saving them a few nickels a day in unneeded efficiency.
Most problems are not unique. Learning to recognize patterns and apply conventional solutions to most things is a very valuable skill. Knowing where to apply conventional patterns to unconventional problems is even more helpful. Correctly identifying the 2-3 times a year you actually need a solid, creative solution to a really unique problem is priceless - and very hard to do if you haven't made them stand out from the sea of normal boring problems by glorifying those into things they didn't need to be.
Yes but the federal highway system doesn't have the problem that, if you store it in a tank made of 3 inches of solid steel, ...
You've got to admit, that'd be a really cool tank. Not sure where we'd put it though.
But it also bars those who wish to tinker from ever doing so, for the sole reason of padding their profit margins.
Did you realize that if you develop apps for the iPhone you can install them on your iPhone without jailbreaking it? Without going through the appstore?
Admittedly it does stop some random 3rd person from easily installing code that some other random person wrote that uses undocumented APIs, thus avoiding the problem that Microsoft has experienced ever since Adobe bought so many copies of Charles Petzold's _Undocumented Windows_ back in the day... but it doesn't stop any true tinkerers from tinkering. Script kiddies, but not tinkerers.
On the whole, its a fairly reasonable balance.
I use LastPass because I want access to my passwords at work and Dropbox is blocked. LastPass does the same thing as KeePass+Dropbox, and I can access it from anywhere.
And now, apparently, so can everybody else! That is convenient.
Depends. Were you planning on doubling the number of wheels (actually the size of the contact patch, but similar) as the weight doubled? If so, then yes, double weight translates into double wear. If not then you now have a much more concentrated weight, which does much more damage to the roads.
And how much pixel-based display output was going on in 1963, exactly?