Can you explain why you feel speed is a particular problem with PHP compared to other web languages? Pointing out Facebook's improvements/hacks isn't very relevant considering they are one of the highest-traffic sites on the entire internet. Most Java container servers have simply never been used for anything approaching the traffic Facebook deals with. And even then, Facebook is just optimizing to cut costs (reduce total server count+power usage), not to make their site load faster for users. It's trivial to add servers to a cluster, and the only real downside is cost, not speed.
The real bottleneck you run into LONG before hitting application server problems is your datastore. That's the one that requires real thought and hard work to deal with once you've pushed "buy a bigger database server" to its limits. Your failure to mention this or acknowledge how little application speed matters makes me wonder how experienced you really are with high-traffic websites.
Looking at an 'average' website, the ones that are most consistently slow or downright bad are Java sites. Having seen this in development first-hand, it's because Java gives you more than enough rope to hang yourself with if you're a clueless off-shored developer (as so much java software is written by). Too many Java websites do a horrific job handling session scoping, cookies, caching (lets cache the whole database in the app layer and loop on it, right?), and so on. By comparison, your standard php, python, or ruby website will be "defaulted" to doing things the correct, scalable way, unless you go out of your way to screw things up. Ruby is the most immature of the 3 from what I've seen, but PHP and Python are great choices over any compiled language for the vast majority of websites. And with PHP, unlike any of those other languages, you already know it can scale to running the biggest site on the internet!;)
I would agree with PHP being a bad language comparable to Perl if you're looking at PHP 4. When Magic_quotes and register_globals and the like were on by default, it was encouraging things to be done Very Wrong by default. Unless you have a penchant for language-enforced whitespace, I don't see any reason with PHP5 or 6 that PHP is any worse than Python.
I know you're about the dozenth post like this on Slashdot, and while the answer is obvious, I'd say that demonstrates the shift in Slashdot's audience and quality over the past decade.
The answer is that Slashdot is an open source and linux targeted website. The audience is assumed to have interest in development and software beyond your basic "I use whatever has a monopoly". It's long been impossible to avoid using Windows because of all their software and hardware lock-in thanks to their desktop and office monopolies. As a hub for the linux community, a goal common to Slashdot, its users, and its articles, was getting Linux to the point where it was a valid second operating system option for desktop computers. Your post helps demonstrate that while Linux may have succeeded on mobile devices with Android, the Windows monopoly is stronger than ever on the desktop.
Still, it's sad to see that Slashdot's community is now the type to respond to "What is still preventing you from having a 2nd choice in operating systems" with "But my 1 and only option is my 1 and only option, why would I want to use anything else?" Can't you imagine a world with multiple operating systems available to choose from, and real competition in features, quality, and design? Or is the stockholm syndrome of Windows really too strong to even consider there could be a better ecosystem for software?
I used to work at a company where we used Linux on our development desktops, and actually work at another company in the same boat today. At both places the boss had us try to use OpenOffice, AbiWord, etc., before finally admitting that despite them functionally doing everything we want, they're not compatible with Microsoft Office's formats and thus unusable.
Virtual machines work great though, and that was our solution to do word processing at both companies. You just have to buy a $250 retail copy of Windows and a $220 copy of Microsoft Office. Next install and setup your virtual machine software of choice with document shares (can be tricky). Now you can, amazingly, for only $470, read and write text documents that are compatible with other businesses.
By the way, isn't it cute how Newegg calls the category "Operating Systems" even though Windows is the only option?
By "de facto standard" you meant "monopoly" right? Isn't it great how they've managed to retain a monopoly on something as basic as text documents.
Remember back when there were a dozen programs you could choose between for word processing? Now Microsoft is the only paid product, despite there being 4+ free alternatives. You have to pay $200+ for Microsoft Office anyway, simply because of monopoly vendor lock-in with the.doc and.docx formats.
It's too bad we don't have some kind of governing body to prevent this kind of abuse!
Often there are already comments proving it's wrong (with links) and that's HOW you know it's wrong. All that's left is to mod the incorrect post down so it isn't wasting readers' time anymore.
Not just the headlines and articles, but the user-posted summaries too. Too many are blatantly biased, and more importantly, biased towards idiocy. Often items that are disregarded on every other site end up posted on Slashdot. Two examples that come to mind are:
1) Google reading SSIDs from WIFI and capturing some plaintext-over-the-air info by accident. This deserved 1 post, maybe a 2nd about the (equally retarded) government looking into it. I swear this non-issue must have been posted to the front page at least a half dozen times, every time with horrendous biased writeups by ODP. This is a tech site full of tech people who are smart enough to know that anyone broadcasting plaintext unencrypted wifi is an idiot and deserves what they get, which in this case was NOTHING because there was zero evidence Google ever did anything with the info or even knew it was there. Non-issue, fear-mongering, posted a half dozen or more times, insulting technically literate users of a tech site every single time.
2) Similar and probably far worse, an article has made it to the front page at least twice about the fact that stolen credit cards are occasionally used to buy things on iTunes. The horror! Once again an absolute non-story, worded sensationally and fear-mongering, which doesn't hold up whatsoever under the slightest scruitiny, yet made it to the front page at LEAST twice.
With stories this bad eventually you'll run out of technically competent users of this site to post thoughtful comments about how idiotic these posts are. I finally pulled Slashdot from my RSS feeds after I found the same stories with smarter comments showing days sooner on my other tech feeds.
Slashdot's strength has always been posting smart tech stories with smart technical comments. Dumb stories with smart comments are still a waste of everyone's time. If you can't compete on timely news, try to compete on rigorously screening stories and promoting the smart technical comments as the site's value.
I don't see this working unless people started modding good posts as Offtopic -1. If people were doing that instead of modding +1 Insightful then the problem wouldn't exist in the first place.
This would be pretty extreme, but why not randomize thread order for each person viewing the post? Or, alternately to support caching a little bitter, randomize thread order every few minutes.
I completely agree on the Google+ name, and their weird attempt to push "+1" as a substitute for "like".
Interface design however, I can't agree on. They may not be up to Apple's standards, but they're generally head and shoulders above the rest of their competitors. Google+ is still new and growing, and the interface can and will change easily as they experiment and look at feedback and usability. On the other hand, fixing the poor name and +1 will be harder the longer the wait.
Declaring G+ dead already, really? How hard is it to use both G+ and Facebook, why would you have to "switch"? Facebook and Twitter are already conjoined in the way most people talk about them. And how many other social networks are you using simultaneously? A hell of a lot of them. Let's see?
Your e-mail address book. IM (probably gchat and AIM, if not also MSN and Yahoo). Skype. Steam, Bnet, Xbox Live, PSN. Eventually Nintendo might make a serious online platform and you'll add WiiNet to the list. Foursquare. Phone+texting address book. My RSS feed reader (google reader) has its own independent social network too. I have at least two more e-mail and IM networks through work. I "should" have two more phone networks if I actually used the phone numbers work provides me.
The list goes on and on. Most blogs still have their own internal user list, as do a huge number of other applications and websites. Every one of these is a social network with its own subset of functionality, much of it directly overlapping. In real terms, using G+ for me meant adding it on my phone, occasionally checking it when I'm bored, and otherwise only noticing if I get a notification from it. That's what I do with Facebook to, with the exception that while I don't post to G+ much yet, I never post to Facebook and never will until they do friend groups (circles) right. And when they do (they're trying now, at least) it will only be because of pressure from G+.
Strongly typed languages are great for the enterprise, because you know (and Intellisense knows too) at compile time what to expect from objects.
After moving to Java and.NET development for the past few years, no, I don't buy it. The runtime XML errors are as frequent as their stacktraces are indecipherable. Not that XML errors are the only runtime errors you get either. All that being strongly typed and compiled seems to buy me is long compile and deploy times compared to interpreted languages. Either way I have to test well to determine if there are serious runtime errors.
It doesn't even buy me platform independence, apparently. The company I'm at is using Weblogic *8*. They can't even upgrade to a newer Weblogic, let alone move to a different application server, let alone a different server platform. I'm sure that's because "they're doing it wrong" by using app server or platform specific logic, but that's still what it boils down to in the real world.
I'm not saying Javascript in particular is the answer to that, but rather that I've found all interpreted languages to be dozens of times faster and more effective to develop with.
Thanks for the reminder, wish this was up as most of the comments miss this. I didn't think about it either and I actually wrote a trucking fuel tax application >.
Probably beating a dead horse here since you already said you believe in one of the hundreds of creation myths, so facts are unlikely to matter. But just in case, evolution has been observed repeatedly. As has speciation. Perhaps you meant abiogenesis? They're still filling in a few of the remaining gaps in that one, so there's still a little room for your mythology to hide. For now.
I, personally, don't believe that either creationism OR evolution belong in the realm of science
start concentrating on things that really matter, like finding a cure for cancer, cure for HIV
You think medical science would progress without teaching evolution? Heck, we wouldn't even have flu shots without evolution.
One of the reasons is probably their comparatively sparse product lineup. I was looking at Sony Vaio laptops, recently, and they have a 10", 11", 12"... everything up to 17" laptops. Apple has 3, 13", 15", and 17". There are a hundred android handsets with every imaginable combination of hardware, but there only 2 iphones, the gsm and cdma ones. When apple changes their product lineup people know what product they're talking about even if they don't own one. It's not "the 12" version of the SXA laptop model", its just "the macbook".
They also (generally) maintain their price point across product revisions, so there's extra interest in when a new launch is going to arrive. Waiting a month might mean significant hardware improvements for the same cost. The Macbook Pro is still $1800, but now you get an extra 2gb of ram, an SSD, a newer CPU, new battery tech, and whatever else they toss in the new model. Or maybe it's a shrug of an upgrade, just a minor cpu update, and you'd rather go ahead and get one now than wait 3 months?
With so few product models and infrequent changes, it makes people more interested than they would otherwise be. Any tech company could probably do this. But they have to be willing to put in the legwork (ie: NDAs, lawyers, and paced product updates) to make it happen.
It's also worth noting how much more impact Apple's launches have when they aren't significantly leaked. That is Apple's goal (or at least Steve Jobs'), so I doubt they regularly leak details on purpose. Remember the bomb Apple dropped on Microsoft when they released updated iPods along with a price drop shortly before the Zune's launch? By keeping it secret up until launch they left Microsoft's head spinning. MS was caught competing with last year's model and prices, rushing to play catchup.
I've seen servers fail to come up for a reboot because of hardware failures before. I'm not sure if I would recommend an annual reboot or what, but scheduling a reboot to test for that sort of issue lets you pick the time for a potential recovery effort. If an unexpected power or software failure causes the reboot, you could find yourself dealing with multiple issues at once, which can make troubleshooting and recovery much harder.
I've also seen unix boxes start having issues after being up too long (again, multiple years) and having certain integers wrap. The kernel hacker at my company had seen the issue multiple times before and recognized it, I was just the on-site monkey at the time, so I don't recall more specific details.
I finally removed flash from my main browsers (Safari and Firefox). I don't even using FlashBlock anymore since that lies to sites ("I want flash content!") and sometimes prevents html5 video from appearing. I use Chrome as my backup when I really need Flash, ie: to watch Fringe on Hulu once a week.
Whenever I do run Flash, it inevitably kicks one of my Core2Duo's cores to 100%. My laptop heats up, the fans kick on at full volume. Then I close down Chrome (which didn't have flash in the foreground or operating anyway, but maybe in a background tab) and the fan shuts off, cpu drops back down to 2-4% usage.
Having flash on my phone sounds about as good as towing a yacht with my bicycle. DO NOT WANT.
It can be annoying on occasion, but sad as it may be, Google is almost always right. And when it isn't, I've always had the "Click here if you really wanted to search for "teh internet" not "the internet"", which is easy and reasonable enough for me. That said, I absolutely despise no-interaction auto-correct systems in general, so I'm still on the fence with google's work even though they've managed to not annoy me yet.
AFAIK from implementing an AVS system at a previous job, the zip codes aren't required because there are so many older systems that don't support them. The retailer gets a discounted credit card processing rate for each transaction (or maybe if X% of transactions include zip) to encourage them to use zipcode verification.
I'm not sure what the point of asking is if they aren't going to reject it anyway, but they may use the zipcode along with other info (eg. location of the purchase, retailer's sketchiness quotient, purchase amount, etc.) to selectively reject transactions.
I've always thought of autorun as a ridiculously obvious attack vector. But you have a good point, in that once upon a time the only CDs available were software (that you were going to execute anyway) and music CDs (whose creators were not malevolently-inclined towards their users). Around 1998 or 2000 when CD-R drives were in every computer, that's when MS should have removed autorun or done this prompt-ever-time fix. I wonder how many corporate systems were infiltrated simply by leaving an unlabeled CD-R on someone's desk, or mailing them one.
When you put a music CD in your drive do you expect it to run executable code (that potentially roots your windows install)? Of course not. Media has long been used more for file storage than for executables/installs, even moreso now that installation discs are out of date by the time they ship.
How did you turn it off, out of curiosity? There seems to be about a dozen different ways, which I found quite frustrating after finding an SD card that had a windows virus on it. (luckily found that out when plugging the card into my mac, not my PC)
When Apple made (only) crappy products, you had a point. But welcome to the year 2011, you can retire 1995's memes. Apple makes some great but expensive products. They are being copied by other manufacturers left and right because of their design, not because of their marketing.
I'm gonna believe them, sorry. Woz is neither infallible nor an expert on automobiles or their electrical systems. Woz is talking about a repeatable cruise control issue, not a brake failure. My car does the same thing all the time, especially if you turn cruise on while going uphill. You're decelerating rapidly so your car immediately drops below the cruise speed and the car slams on the gas to try to catch up, and then stops just as suddenly once it hits that speed.
I drive a Honda btw, and cruise control does the same thing on every other model car I've driven.
Also if this is just due to driver error, why it is only happening mainly on Toyotas? Answer me that!!
I heard on the radio last year that the rates for Toyotas were almost identical to the rates for every other manufacturer. There are a lot more Toyotas out there than most brands so you hear a lot of Toyota anecdotes. But percentage-wise, Toyota was unremarkable. The other car manufacturers are/were scared shitless that there would be a high-profile incident of driver mistake in their cars that caught the media's attention.
Feel free to hunt down the stats and correct me if I'm misremembering.
Slashdot Mirror
Can you explain why you feel speed is a particular problem with PHP compared to other web languages? Pointing out Facebook's improvements/hacks isn't very relevant considering they are one of the highest-traffic sites on the entire internet. Most Java container servers have simply never been used for anything approaching the traffic Facebook deals with. And even then, Facebook is just optimizing to cut costs (reduce total server count+power usage), not to make their site load faster for users. It's trivial to add servers to a cluster, and the only real downside is cost, not speed.
The real bottleneck you run into LONG before hitting application server problems is your datastore. That's the one that requires real thought and hard work to deal with once you've pushed "buy a bigger database server" to its limits. Your failure to mention this or acknowledge how little application speed matters makes me wonder how experienced you really are with high-traffic websites.
Looking at an 'average' website, the ones that are most consistently slow or downright bad are Java sites. Having seen this in development first-hand, it's because Java gives you more than enough rope to hang yourself with if you're a clueless off-shored developer (as so much java software is written by). Too many Java websites do a horrific job handling session scoping, cookies, caching (lets cache the whole database in the app layer and loop on it, right?), and so on. By comparison, your standard php, python, or ruby website will be "defaulted" to doing things the correct, scalable way, unless you go out of your way to screw things up. Ruby is the most immature of the 3 from what I've seen, but PHP and Python are great choices over any compiled language for the vast majority of websites. And with PHP, unlike any of those other languages, you already know it can scale to running the biggest site on the internet! ;)
I would agree with PHP being a bad language comparable to Perl if you're looking at PHP 4. When Magic_quotes and register_globals and the like were on by default, it was encouraging things to be done Very Wrong by default. Unless you have a penchant for language-enforced whitespace, I don't see any reason with PHP5 or 6 that PHP is any worse than Python.
I know you're about the dozenth post like this on Slashdot, and while the answer is obvious, I'd say that demonstrates the shift in Slashdot's audience and quality over the past decade.
The answer is that Slashdot is an open source and linux targeted website. The audience is assumed to have interest in development and software beyond your basic "I use whatever has a monopoly". It's long been impossible to avoid using Windows because of all their software and hardware lock-in thanks to their desktop and office monopolies. As a hub for the linux community, a goal common to Slashdot, its users, and its articles, was getting Linux to the point where it was a valid second operating system option for desktop computers. Your post helps demonstrate that while Linux may have succeeded on mobile devices with Android, the Windows monopoly is stronger than ever on the desktop.
Still, it's sad to see that Slashdot's community is now the type to respond to "What is still preventing you from having a 2nd choice in operating systems" with "But my 1 and only option is my 1 and only option, why would I want to use anything else?" Can't you imagine a world with multiple operating systems available to choose from, and real competition in features, quality, and design? Or is the stockholm syndrome of Windows really too strong to even consider there could be a better ecosystem for software?
I used to work at a company where we used Linux on our development desktops, and actually work at another company in the same boat today. At both places the boss had us try to use OpenOffice, AbiWord, etc., before finally admitting that despite them functionally doing everything we want, they're not compatible with Microsoft Office's formats and thus unusable.
Virtual machines work great though, and that was our solution to do word processing at both companies. You just have to buy a $250 retail copy of Windows and a $220 copy of Microsoft Office. Next install and setup your virtual machine software of choice with document shares (can be tricky). Now you can, amazingly, for only $470, read and write text documents that are compatible with other businesses.
By the way, isn't it cute how Newegg calls the category "Operating Systems" even though Windows is the only option?
By "de facto standard" you meant "monopoly" right? Isn't it great how they've managed to retain a monopoly on something as basic as text documents.
Remember back when there were a dozen programs you could choose between for word processing? Now Microsoft is the only paid product, despite there being 4+ free alternatives. You have to pay $200+ for Microsoft Office anyway, simply because of monopoly vendor lock-in with the .doc and .docx formats.
It's too bad we don't have some kind of governing body to prevent this kind of abuse!
Often there are already comments proving it's wrong (with links) and that's HOW you know it's wrong. All that's left is to mod the incorrect post down so it isn't wasting readers' time anymore.
Not just the headlines and articles, but the user-posted summaries too. Too many are blatantly biased, and more importantly, biased towards idiocy. Often items that are disregarded on every other site end up posted on Slashdot. Two examples that come to mind are:
1) Google reading SSIDs from WIFI and capturing some plaintext-over-the-air info by accident. This deserved 1 post, maybe a 2nd about the (equally retarded) government looking into it. I swear this non-issue must have been posted to the front page at least a half dozen times, every time with horrendous biased writeups by ODP. This is a tech site full of tech people who are smart enough to know that anyone broadcasting plaintext unencrypted wifi is an idiot and deserves what they get, which in this case was NOTHING because there was zero evidence Google ever did anything with the info or even knew it was there. Non-issue, fear-mongering, posted a half dozen or more times, insulting technically literate users of a tech site every single time.
2) Similar and probably far worse, an article has made it to the front page at least twice about the fact that stolen credit cards are occasionally used to buy things on iTunes. The horror! Once again an absolute non-story, worded sensationally and fear-mongering, which doesn't hold up whatsoever under the slightest scruitiny, yet made it to the front page at LEAST twice.
With stories this bad eventually you'll run out of technically competent users of this site to post thoughtful comments about how idiotic these posts are. I finally pulled Slashdot from my RSS feeds after I found the same stories with smarter comments showing days sooner on my other tech feeds.
Slashdot's strength has always been posting smart tech stories with smart technical comments. Dumb stories with smart comments are still a waste of everyone's time. If you can't compete on timely news, try to compete on rigorously screening stories and promoting the smart technical comments as the site's value.
I don't see this working unless people started modding good posts as Offtopic -1. If people were doing that instead of modding +1 Insightful then the problem wouldn't exist in the first place.
This would be pretty extreme, but why not randomize thread order for each person viewing the post? Or, alternately to support caching a little bitter, randomize thread order every few minutes.
I completely agree on the Google+ name, and their weird attempt to push "+1" as a substitute for "like".
Interface design however, I can't agree on. They may not be up to Apple's standards, but they're generally head and shoulders above the rest of their competitors. Google+ is still new and growing, and the interface can and will change easily as they experiment and look at feedback and usability. On the other hand, fixing the poor name and +1 will be harder the longer the wait.
Declaring G+ dead already, really? How hard is it to use both G+ and Facebook, why would you have to "switch"? Facebook and Twitter are already conjoined in the way most people talk about them. And how many other social networks are you using simultaneously? A hell of a lot of them. Let's see?
Your e-mail address book.
IM (probably gchat and AIM, if not also MSN and Yahoo).
Skype.
Steam, Bnet, Xbox Live, PSN. Eventually Nintendo might make a serious online platform and you'll add WiiNet to the list.
Foursquare.
Phone+texting address book.
My RSS feed reader (google reader) has its own independent social network too.
I have at least two more e-mail and IM networks through work.
I "should" have two more phone networks if I actually used the phone numbers work provides me.
The list goes on and on. Most blogs still have their own internal user list, as do a huge number of other applications and websites. Every one of these is a social network with its own subset of functionality, much of it directly overlapping. In real terms, using G+ for me meant adding it on my phone, occasionally checking it when I'm bored, and otherwise only noticing if I get a notification from it. That's what I do with Facebook to, with the exception that while I don't post to G+ much yet, I never post to Facebook and never will until they do friend groups (circles) right. And when they do (they're trying now, at least) it will only be because of pressure from G+.
Strongly typed languages are great for the enterprise, because you know (and Intellisense knows too) at compile time what to expect from objects.
After moving to Java and .NET development for the past few years, no, I don't buy it. The runtime XML errors are as frequent as their stacktraces are indecipherable. Not that XML errors are the only runtime errors you get either. All that being strongly typed and compiled seems to buy me is long compile and deploy times compared to interpreted languages. Either way I have to test well to determine if there are serious runtime errors.
It doesn't even buy me platform independence, apparently. The company I'm at is using Weblogic *8*. They can't even upgrade to a newer Weblogic, let alone move to a different application server, let alone a different server platform. I'm sure that's because "they're doing it wrong" by using app server or platform specific logic, but that's still what it boils down to in the real world.
I'm not saying Javascript in particular is the answer to that, but rather that I've found all interpreted languages to be dozens of times faster and more effective to develop with.
Thanks for the reminder, wish this was up as most of the comments miss this. I didn't think about it either and I actually wrote a trucking fuel tax application >.
Probably beating a dead horse here since you already said you believe in one of the hundreds of creation myths, so facts are unlikely to matter. But just in case, evolution has been observed repeatedly. As has speciation. Perhaps you meant abiogenesis? They're still filling in a few of the remaining gaps in that one, so there's still a little room for your mythology to hide. For now.
I, personally, don't believe that either creationism OR evolution belong in the realm of science
start concentrating on things that really matter, like finding a cure for cancer, cure for HIV
You think medical science would progress without teaching evolution? Heck, we wouldn't even have flu shots without evolution.
Absolutely right.
One of the reasons is probably their comparatively sparse product lineup. I was looking at Sony Vaio laptops, recently, and they have a 10", 11", 12"... everything up to 17" laptops. Apple has 3, 13", 15", and 17". There are a hundred android handsets with every imaginable combination of hardware, but there only 2 iphones, the gsm and cdma ones. When apple changes their product lineup people know what product they're talking about even if they don't own one. It's not "the 12" version of the SXA laptop model", its just "the macbook".
They also (generally) maintain their price point across product revisions, so there's extra interest in when a new launch is going to arrive. Waiting a month might mean significant hardware improvements for the same cost. The Macbook Pro is still $1800, but now you get an extra 2gb of ram, an SSD, a newer CPU, new battery tech, and whatever else they toss in the new model. Or maybe it's a shrug of an upgrade, just a minor cpu update, and you'd rather go ahead and get one now than wait 3 months?
With so few product models and infrequent changes, it makes people more interested than they would otherwise be. Any tech company could probably do this. But they have to be willing to put in the legwork (ie: NDAs, lawyers, and paced product updates) to make it happen.
It's also worth noting how much more impact Apple's launches have when they aren't significantly leaked. That is Apple's goal (or at least Steve Jobs'), so I doubt they regularly leak details on purpose. Remember the bomb Apple dropped on Microsoft when they released updated iPods along with a price drop shortly before the Zune's launch? By keeping it secret up until launch they left Microsoft's head spinning. MS was caught competing with last year's model and prices, rushing to play catchup.
I've seen servers fail to come up for a reboot because of hardware failures before. I'm not sure if I would recommend an annual reboot or what, but scheduling a reboot to test for that sort of issue lets you pick the time for a potential recovery effort. If an unexpected power or software failure causes the reboot, you could find yourself dealing with multiple issues at once, which can make troubleshooting and recovery much harder.
I've also seen unix boxes start having issues after being up too long (again, multiple years) and having certain integers wrap. The kernel hacker at my company had seen the issue multiple times before and recognized it, I was just the on-site monkey at the time, so I don't recall more specific details.
I finally removed flash from my main browsers (Safari and Firefox). I don't even using FlashBlock anymore since that lies to sites ("I want flash content!") and sometimes prevents html5 video from appearing. I use Chrome as my backup when I really need Flash, ie: to watch Fringe on Hulu once a week.
Whenever I do run Flash, it inevitably kicks one of my Core2Duo's cores to 100%. My laptop heats up, the fans kick on at full volume. Then I close down Chrome (which didn't have flash in the foreground or operating anyway, but maybe in a background tab) and the fan shuts off, cpu drops back down to 2-4% usage.
Having flash on my phone sounds about as good as towing a yacht with my bicycle. DO NOT WANT.
It can be annoying on occasion, but sad as it may be, Google is almost always right. And when it isn't, I've always had the "Click here if you really wanted to search for "teh internet" not "the internet"", which is easy and reasonable enough for me. That said, I absolutely despise no-interaction auto-correct systems in general, so I'm still on the fence with google's work even though they've managed to not annoy me yet.
AFAIK from implementing an AVS system at a previous job, the zip codes aren't required because there are so many older systems that don't support them. The retailer gets a discounted credit card processing rate for each transaction (or maybe if X% of transactions include zip) to encourage them to use zipcode verification.
I'm not sure what the point of asking is if they aren't going to reject it anyway, but they may use the zipcode along with other info (eg. location of the purchase, retailer's sketchiness quotient, purchase amount, etc.) to selectively reject transactions.
Also know as: BritneySpears.jpg.exe
I've always thought of autorun as a ridiculously obvious attack vector. But you have a good point, in that once upon a time the only CDs available were software (that you were going to execute anyway) and music CDs (whose creators were not malevolently-inclined towards their users). Around 1998 or 2000 when CD-R drives were in every computer, that's when MS should have removed autorun or done this prompt-ever-time fix. I wonder how many corporate systems were infiltrated simply by leaving an unlabeled CD-R on someone's desk, or mailing them one.
When you put a music CD in your drive do you expect it to run executable code (that potentially roots your windows install)? Of course not. Media has long been used more for file storage than for executables/installs, even moreso now that installation discs are out of date by the time they ship.
How did you turn it off, out of curiosity? There seems to be about a dozen different ways, which I found quite frustrating after finding an SD card that had a windows virus on it. (luckily found that out when plugging the card into my mac, not my PC)
When Apple made (only) crappy products, you had a point. But welcome to the year 2011, you can retire 1995's memes. Apple makes some great but expensive products. They are being copied by other manufacturers left and right because of their design, not because of their marketing.
I'm gonna believe them, sorry. Woz is neither infallible nor an expert on automobiles or their electrical systems. Woz is talking about a repeatable cruise control issue, not a brake failure. My car does the same thing all the time, especially if you turn cruise on while going uphill. You're decelerating rapidly so your car immediately drops below the cruise speed and the car slams on the gas to try to catch up, and then stops just as suddenly once it hits that speed.
I drive a Honda btw, and cruise control does the same thing on every other model car I've driven.
Also if this is just due to driver error, why it is only happening mainly on Toyotas? Answer me that!!
I heard on the radio last year that the rates for Toyotas were almost identical to the rates for every other manufacturer. There are a lot more Toyotas out there than most brands so you hear a lot of Toyota anecdotes. But percentage-wise, Toyota was unremarkable. The other car manufacturers are/were scared shitless that there would be a high-profile incident of driver mistake in their cars that caught the media's attention.
Feel free to hunt down the stats and correct me if I'm misremembering.