Yes, for hard drives and memory. What about when your WiFi card dies? Optical drive? What if you're using a MBP (as I am) which has a thoroughly inaccessible hard drive?
I love the system for quite a number of reasons, but accessibility of hardware for repairs is absolutely not one of them. Which is probably for the better, as I've had then spent money I had no intent to spend on a hard drive upgrade at the very least.
Perhaps. I think that if anything, they'd try to rework their server software so it'll be more useful in the educational sector. I've called Apple about business inquiries and they are apparently able to arrange a 5% discount, if your business rep ever calls you back (which obviously didn't happen in my case). However, you can get a larger discount when dealing with their education side, often on more products - hell, they knock 50% off the cost of Leopard Server if you get it as a BTO option on a Mac Pro (which are already $200 off).
I doubt they'd ever abandon business entirely, but it seems like the kind of thing where they'd want to sub-contract out the maintenance aspect of things (even if it's some sort of internal thing, an Apple for Business, Inc, if you will). I'd put my money on them actually putting a lot of development efforts into a long-term business architecture while they focus almost entirely on getting people to the platform for the next few years. Let's not forget how they've gone and positioned the iPhone after that roadmap event - they're definitely looking to penetrate more into the business market. I have no reason to think that they wouldn't want to do the same in the desktop/notebook market.
You mean like, say, Google? Just as Kleenex = facial tissue, Xerox = photocopy, and Hoover = vacuum (at least in the UK), Google = search.
Now I'm not sure how much this would really apply to them given the number of people who would go to Google to search for Google (I did test out of curiosity though), but it could damage them too for the same reasons.
Maybe, but I think they know their limits. And of course they make it clear enough what's an ad and what isn't, unlike. I was slightly amused when I did a search for SugarCRM and got an ad for SalesForce. But to reiterate, the top ads show up in a yellow box that clearly separates them from the actual search results, and are limited to one to two (plus quite a few crappy ones on the right). Yahoo! has four on top plus six on the side (and two oddities: no salesforce ad, and the rather broken English "sponsor results" box), plus two more on the bottom I just spotted. MSN has three and four on top and right respectively, though has a total of under 3000 results as compared to Google's 1.59 million.
Of course plenty of people have no idea that "Sponsored Links" means advertisements, but it's those kind of people that the ads are targeting in the first place.
What do I make of this? Not a whole lot, to be honest. Regardless of their advertising policies, I find Google's results to be better, more relevant, and presented in a slightly cleaner way than either MSN or Yahoo, so I'll stick with them.
Yeah, but the ontology problem is going to apply to any system. Searching may be the way to go, but tagging is also quite effective, and the combination of the two is probably the best without making board@organization.org send out to ten private accounts in a small list of sorts.
Lucky Gmail combines tags with great searching, and it sounds like it's already on the table. The good thing with tagging is that since it's not stuck to a single folder, duplicate tags (if you will) aren't going to clog things up too badly or throw off someone else's system - unless a term means two different things to two different people, in which case you really do have to split it into multiple inboxes.
That's really what I was getting at. Of course my post is almost entirely wild speculation, but given the simplicity of twitter (I expect that most experience coders could replicate at least 90% of it in an afternoon), I think it's unlikely that Rails is the entire problem here.
That said, my own experiences with Rails have not been good, and I've got a friend trying to deploy a Rails blog (not using the screencast that everyone was talking about, mind you) and is having a hell of a time between whatever webserver app his host is running, complete removal of functionality rather than deprecation between versions, and an extremely unclear architecture. Of course I think the latter is true for most frameworks and isn't specific to Ruby/ROR, but the whole semi-compiling thing that Rails seems to do doesn't help much for performance given that it dumps the cache after five minutes of inactivity and has to rebuild the entire site - not at all unlike recompiling a.NET site after an IIS restart, except somehow much slower and much more frequent.
I think the slew of 37 Signals/BaseCamp stuff is also RoR powered, and I don't remember ever hearing about scalability problems for them. I think the issue with Twitter is that every time someone tweets, a mail message goes out to tens to tens of thousands of people, and mailservers tend to not really care for that much activity especially when real time messaging is expecting. Architecturally, a tweet is just an INSERT query followed by a SELECT with some not especially complex JOINs - it's just that every row of the SELECT result requires a message in some format (primarily email, assuming they're using the phonenumber@carrier.com approach for texting, though they also have an AIM service - AFAIK all of the other clients are pull services rather than actually having content pushed to them).
You try sending out several hundred emails every time someone clicks a button on your site when there are thousands of users clicking that button several times a day and try not having scalability issues on any platform. My own experience with Ruby/RoR has been minimal but unpleasant, but unless they're doing something god-awful with their databases like I did with my very early code (read: foreach loops in the scripting language instead of a single join in the sql), I think the issue lies outside of the language/framework.
I would set it up just like this, except perhaps some sort of reasonable limit on connection time or frequency to the system ("legit" use should never have more than one connection per phone number, for example).
Will some people abuse it? Yes, though I wouldn't consider it abuse. It's really a long-term value add. People see that iPhone users get free Wifi at these places from their iPhone, and a few know how to get it on their laptop too. That's a tremendous additional value to using AT&T over competitors if you spend any amount of time at those businesses. People can get on sometimes using their friend's phone number for the time being, but realize how damn awesome the access is that comes for free because they're paying the same bill they'd be paying already, and make the switch when it comes time to replace their existing phone.
How much can bandwidth cost them, anyways? I know it's not free, but it's not costing them anywhere near the $20/mo that they charge for the iPhone unlimited data plan either. So long as customers aren't maxing out the connection for the entire month (and even if they do, it's probably still pretty profitable), it's in their best interest to get as many people paying for the service even if it means slightly slimmer margins by giving them "free" access at other places.
I agree, except this was really just a reference book on the series. Rowling, to my understanding, had very little objection to it being created. The issue came from the fact that it was to be published for profit (which even people who spend all day on thepiratebay tend to take issue with, since stealing software at least doesn't result in the thief making money), and more importantly that she was planning to do the same in the future and donate all profits to charity.
Uh oh, someone had better call the waaaaahmbulance.
Times have changed, and everyone certainly has their own opinion. You'll note that Rowling's work gets quite as much attention today (if not more so, thanks to the ease of distribution) as Shakespeare's ever has. Say what you want about both sides, but I've read a number of Shakespeare's works and thought they sucked and are well deserving of my own title of "most overrated crap of all time".
What are your standards for the comparison?
Appeal? Both had very widespread appeal for their time, though of course only time will tell whether Rowling's work continues to be admired by many centuries from now. Given the effect it's had on so many people today, I expect it won't disappear by any stretch of the imagination.
Grammar? Standards have changed tremendously between then and now, so it's hard to make a fair comparison. I can't read old English well enough to comment on the grammatical structure of Shakespeare, but there's nothing wrong with Rowling's syntax. Diction has likewise changed over time, but one can't deny that the Potter series is aimed towards the younger crowd. Is this a fault? I don't see why it should be, but that's mostly opinion.
Monetary value? Rowling wins, hands down. Shakespeare sure as hell wasn't a billionaire, even adjusted for inflation. Quite irrelevant to me, but that's what some people may use as a standard.
Effect on society? I can't claim to know enough about Shakespearean-era life to give an informed response, but whether you love or hate the Potter series you can't deny its marked effect on millions of people. I'd use this as the primary indicator of literary value. Again, time will tell.
Or screw the mouse, since you're hand-coding. Taking your hands off the keyboard slows you down a surprising amount.
I'm partial to a widesceeen monitor in portrait mode with Coda's split view live previewing. Keep the eyes on the code and the preview, and generally no manual refreshing.
Just because I'm hand-coding doesn't mean I have to waste my time using inadequate tools. At work I use Dreamweaver occasionally in split view for the same reason, though only because there's no good equivalent to Coda for Windows (and our CMS controls integrate with DW and I don't really feel like learning dozens of parameters and risking stupid typos)
Interestingly, you can only do so much damage with PHP's handling of SQL statements. Namely, you can only run one statement per mysql_query() so while you could bypass a login with a ' OR 1=1 ' deal, Bobby Tables couldn't completely kill your DB. Doesn't stop code injection in the slightest which would be the easiest to prevent, but it's a start. Unfortunately it would be rather tricky to write software that knows when to escape characters and when I'm using variables in the statement safely which would break things if their contents were escaped.
It sounds like code injection is also at fault here, not just SQL injection. Which is to say that while the above method protects attacks directly on the database (and I use a very similar method), it doesn't stop someone from sticking a malicious script tag in an input field. You need to make sure that you run htmlentities() on the output or else I can screw with your page formatting (Name: "" type things) just as easily as linking to a heinous javascript file.
You could also get three 500GB drives, giving you a terabyte of RAID5 storage. Or just JBOD for the same 1.5TB, which you're still better off doing than RAID0 unless you absolutely need the extra speed (meaning HD video capture, and almost nothing else). Of course 750GB drives are starting at $120 these days, so you can head to 2.25TB for $360. Blah blah blah.
The only really nice thing about having that kind of speed in a single self-contained drive is not having to futz with RAID support at an OS level, data integrity and security be damned. For day to day use the seek time of any 10K RPM drive is going to provide a decent improvement over your standard 7200RPM unit, though it's not too noticeable in my experience if you keep your drive defragged. The very tiny decrease in seek time of THIS unit over a normal Raptor is pretty minimal, and not likely worth the premium.
Maybe relative tracking can't be done by simple means since it requires participation on everyone's part, but absolute local tracking is trivially easy on any server that supports server-side scripting and has some sort of database access. A couple lines of code at the bottom of your page to insert a new row on a page load and you've got nearly perfect visitor logs that can easily go beyond your standard server logs.
Again, useless for relative popularity unless you have everyone's data. But it still tells you how popular your site is which is great for ego boosting and advertiser stats if nothing else.
(I'd suggest that Google Analytics is going to be a lot more useful in the long run and at least has the potential to provide relative data in addition to the absolute, but anything that relies on client-side scripting is going to give less accurate numbers since clients can disable or screw around with scripting)
Sen. Joe Biden (D-Del.) said he was under the impression it's "pretty easy to pick out the person engaged in either transmitting or downloading violent scenes of rape, molestation" simply by looking at file names.
Does that quote suggest to you that he's technical enough to even know what a MAC address is, let alone know that it's spoofed with five seconds worth of work in the command prompt? People have of course screwed with filenames just to see what happens (much more in Kazaa/Limewire than Bit-torrent), probably to test Rule 34, with interesting results; more importantly, it shows that filenames are completely meaningless if you're looking to actually reveal content. In any case, just ticking the "encryption required" box in your torrent client solves that problem - and you can bet your ass that any sort of unique serial number would be the much more industry-standard (and equally absurd) IP address. You know - take the RIAA approach.
That may be true, but it's still a pretty meaningless number if you're not actually consuming gasoline. I could measure the power consumption based off the number of candles it takes to power a steam-based generator, but that would be equally stupid. Like the ongoing "Libraries of Congress" thing.
If you're going to use an indirect unit of measure to determine the vehicle's efficiency, surely you'd be better off with pounds of CO2 as a waste-product of the energy production to power the vehicle for an hour. Or something like that.
Yes, the tracker is public. However even to browse to many of the older torrents, you need to be a member. So many of their torrents end up on other sites so it barely matters, but it's still not as wide-open as TPB.
I'm sure you won't like the example, but Facebook is almost entirely powered by MySQL. Granted, it's very heavily modified (at least according to their jobs pages) to provide better support for pretty much everything that'll get mentioned in all of these comments, but they say that most or all of those changes will be released back into the public for future revisions.
I believe that Google also uses MySQL heavily, or at least did at one point. However, that's just some vague recollection and could be totally wrong.
That certainly sounds like a pretty reasonable approach, if you must have the traffic camera in place. If - and only if - the yellow lasts long enough to provide sufficient stopping time.
Slashdot Mirror
Yes, for hard drives and memory. What about when your WiFi card dies? Optical drive? What if you're using a MBP (as I am) which has a thoroughly inaccessible hard drive?
I love the system for quite a number of reasons, but accessibility of hardware for repairs is absolutely not one of them. Which is probably for the better, as I've had then spent money I had no intent to spend on a hard drive upgrade at the very least.
Perhaps. I think that if anything, they'd try to rework their server software so it'll be more useful in the educational sector. I've called Apple about business inquiries and they are apparently able to arrange a 5% discount, if your business rep ever calls you back (which obviously didn't happen in my case). However, you can get a larger discount when dealing with their education side, often on more products - hell, they knock 50% off the cost of Leopard Server if you get it as a BTO option on a Mac Pro (which are already $200 off).
I doubt they'd ever abandon business entirely, but it seems like the kind of thing where they'd want to sub-contract out the maintenance aspect of things (even if it's some sort of internal thing, an Apple for Business, Inc, if you will). I'd put my money on them actually putting a lot of development efforts into a long-term business architecture while they focus almost entirely on getting people to the platform for the next few years. Let's not forget how they've gone and positioned the iPhone after that roadmap event - they're definitely looking to penetrate more into the business market. I have no reason to think that they wouldn't want to do the same in the desktop/notebook market.
You mean like, say, Google? Just as Kleenex = facial tissue, Xerox = photocopy, and Hoover = vacuum (at least in the UK), Google = search.
Now I'm not sure how much this would really apply to them given the number of people who would go to Google to search for Google (I did test out of curiosity though), but it could damage them too for the same reasons.
Maybe, but I think they know their limits. And of course they make it clear enough what's an ad and what isn't, unlike. I was slightly amused when I did a search for SugarCRM and got an ad for SalesForce. But to reiterate, the top ads show up in a yellow box that clearly separates them from the actual search results, and are limited to one to two (plus quite a few crappy ones on the right). Yahoo! has four on top plus six on the side (and two oddities: no salesforce ad, and the rather broken English "sponsor results" box), plus two more on the bottom I just spotted. MSN has three and four on top and right respectively, though has a total of under 3000 results as compared to Google's 1.59 million.
Of course plenty of people have no idea that "Sponsored Links" means advertisements, but it's those kind of people that the ads are targeting in the first place.
What do I make of this? Not a whole lot, to be honest. Regardless of their advertising policies, I find Google's results to be better, more relevant, and presented in a slightly cleaner way than either MSN or Yahoo, so I'll stick with them.
Yeah, but the ontology problem is going to apply to any system. Searching may be the way to go, but tagging is also quite effective, and the combination of the two is probably the best without making board@organization.org send out to ten private accounts in a small list of sorts.
Lucky Gmail combines tags with great searching, and it sounds like it's already on the table. The good thing with tagging is that since it's not stuck to a single folder, duplicate tags (if you will) aren't going to clog things up too badly or throw off someone else's system - unless a term means two different things to two different people, in which case you really do have to split it into multiple inboxes.
That's really what I was getting at. Of course my post is almost entirely wild speculation, but given the simplicity of twitter (I expect that most experience coders could replicate at least 90% of it in an afternoon), I think it's unlikely that Rails is the entire problem here.
.NET site after an IIS restart, except somehow much slower and much more frequent.
That said, my own experiences with Rails have not been good, and I've got a friend trying to deploy a Rails blog (not using the screencast that everyone was talking about, mind you) and is having a hell of a time between whatever webserver app his host is running, complete removal of functionality rather than deprecation between versions, and an extremely unclear architecture. Of course I think the latter is true for most frameworks and isn't specific to Ruby/ROR, but the whole semi-compiling thing that Rails seems to do doesn't help much for performance given that it dumps the cache after five minutes of inactivity and has to rebuild the entire site - not at all unlike recompiling a
But how will we then kill the poor servers?
I think the slew of 37 Signals/BaseCamp stuff is also RoR powered, and I don't remember ever hearing about scalability problems for them. I think the issue with Twitter is that every time someone tweets, a mail message goes out to tens to tens of thousands of people, and mailservers tend to not really care for that much activity especially when real time messaging is expecting. Architecturally, a tweet is just an INSERT query followed by a SELECT with some not especially complex JOINs - it's just that every row of the SELECT result requires a message in some format (primarily email, assuming they're using the phonenumber@carrier.com approach for texting, though they also have an AIM service - AFAIK all of the other clients are pull services rather than actually having content pushed to them).
You try sending out several hundred emails every time someone clicks a button on your site when there are thousands of users clicking that button several times a day and try not having scalability issues on any platform. My own experience with Ruby/RoR has been minimal but unpleasant, but unless they're doing something god-awful with their databases like I did with my very early code (read: foreach loops in the scripting language instead of a single join in the sql), I think the issue lies outside of the language/framework.
I would set it up just like this, except perhaps some sort of reasonable limit on connection time or frequency to the system ("legit" use should never have more than one connection per phone number, for example).
Will some people abuse it? Yes, though I wouldn't consider it abuse. It's really a long-term value add. People see that iPhone users get free Wifi at these places from their iPhone, and a few know how to get it on their laptop too. That's a tremendous additional value to using AT&T over competitors if you spend any amount of time at those businesses. People can get on sometimes using their friend's phone number for the time being, but realize how damn awesome the access is that comes for free because they're paying the same bill they'd be paying already, and make the switch when it comes time to replace their existing phone.
How much can bandwidth cost them, anyways? I know it's not free, but it's not costing them anywhere near the $20/mo that they charge for the iPhone unlimited data plan either. So long as customers aren't maxing out the connection for the entire month (and even if they do, it's probably still pretty profitable), it's in their best interest to get as many people paying for the service even if it means slightly slimmer margins by giving them "free" access at other places.
I agree, except this was really just a reference book on the series. Rowling, to my understanding, had very little objection to it being created. The issue came from the fact that it was to be published for profit (which even people who spend all day on thepiratebay tend to take issue with, since stealing software at least doesn't result in the thief making money), and more importantly that she was planning to do the same in the future and donate all profits to charity.
Uh oh, someone had better call the waaaaahmbulance.
Times have changed, and everyone certainly has their own opinion. You'll note that Rowling's work gets quite as much attention today (if not more so, thanks to the ease of distribution) as Shakespeare's ever has. Say what you want about both sides, but I've read a number of Shakespeare's works and thought they sucked and are well deserving of my own title of "most overrated crap of all time".
What are your standards for the comparison?
Appeal? Both had very widespread appeal for their time, though of course only time will tell whether Rowling's work continues to be admired by many centuries from now. Given the effect it's had on so many people today, I expect it won't disappear by any stretch of the imagination.
Grammar? Standards have changed tremendously between then and now, so it's hard to make a fair comparison. I can't read old English well enough to comment on the grammatical structure of Shakespeare, but there's nothing wrong with Rowling's syntax. Diction has likewise changed over time, but one can't deny that the Potter series is aimed towards the younger crowd. Is this a fault? I don't see why it should be, but that's mostly opinion.
Monetary value? Rowling wins, hands down. Shakespeare sure as hell wasn't a billionaire, even adjusted for inflation. Quite irrelevant to me, but that's what some people may use as a standard.
Effect on society? I can't claim to know enough about Shakespearean-era life to give an informed response, but whether you love or hate the Potter series you can't deny its marked effect on millions of people. I'd use this as the primary indicator of literary value. Again, time will tell.
Or screw the mouse, since you're hand-coding. Taking your hands off the keyboard slows you down a surprising amount.
I'm partial to a widesceeen monitor in portrait mode with Coda's split view live previewing. Keep the eyes on the code and the preview, and generally no manual refreshing.
Just because I'm hand-coding doesn't mean I have to waste my time using inadequate tools. At work I use Dreamweaver occasionally in split view for the same reason, though only because there's no good equivalent to Coda for Windows (and our CMS controls integrate with DW and I don't really feel like learning dozens of parameters and risking stupid typos)
Interestingly, you can only do so much damage with PHP's handling of SQL statements. Namely, you can only run one statement per mysql_query() so while you could bypass a login with a ' OR 1=1 ' deal, Bobby Tables couldn't completely kill your DB. Doesn't stop code injection in the slightest which would be the easiest to prevent, but it's a start. Unfortunately it would be rather tricky to write software that knows when to escape characters and when I'm using variables in the statement safely which would break things if their contents were escaped.
It sounds like code injection is also at fault here, not just SQL injection. Which is to say that while the above method protects attacks directly on the database (and I use a very similar method), it doesn't stop someone from sticking a malicious script tag in an input field. You need to make sure that you run htmlentities() on the output or else I can screw with your page formatting (Name: "" type things) just as easily as linking to a heinous javascript file.
Tell that to My "We were just bought for $1,000,000,000" SQL.
You could also get three 500GB drives, giving you a terabyte of RAID5 storage. Or just JBOD for the same 1.5TB, which you're still better off doing than RAID0 unless you absolutely need the extra speed (meaning HD video capture, and almost nothing else). Of course 750GB drives are starting at $120 these days, so you can head to 2.25TB for $360. Blah blah blah.
The only really nice thing about having that kind of speed in a single self-contained drive is not having to futz with RAID support at an OS level, data integrity and security be damned. For day to day use the seek time of any 10K RPM drive is going to provide a decent improvement over your standard 7200RPM unit, though it's not too noticeable in my experience if you keep your drive defragged. The very tiny decrease in seek time of THIS unit over a normal Raptor is pretty minimal, and not likely worth the premium.
Portrait mode is your friend, sir.
Maybe relative tracking can't be done by simple means since it requires participation on everyone's part, but absolute local tracking is trivially easy on any server that supports server-side scripting and has some sort of database access. A couple lines of code at the bottom of your page to insert a new row on a page load and you've got nearly perfect visitor logs that can easily go beyond your standard server logs.
Again, useless for relative popularity unless you have everyone's data. But it still tells you how popular your site is which is great for ego boosting and advertiser stats if nothing else.
(I'd suggest that Google Analytics is going to be a lot more useful in the long run and at least has the potential to provide relative data in addition to the absolute, but anything that relies on client-side scripting is going to give less accurate numbers since clients can disable or screw around with scripting)
Does that quote suggest to you that he's technical enough to even know what a MAC address is, let alone know that it's spoofed with five seconds worth of work in the command prompt? People have of course screwed with filenames just to see what happens (much more in Kazaa/Limewire than Bit-torrent), probably to test Rule 34, with interesting results; more importantly, it shows that filenames are completely meaningless if you're looking to actually reveal content. In any case, just ticking the "encryption required" box in your torrent client solves that problem - and you can bet your ass that any sort of unique serial number would be the much more industry-standard (and equally absurd) IP address. You know - take the RIAA approach.
HTTP spec be damned - has IE taught you nothing?
He's not talking about speed, but rather the energy consumed. MPG is a time-insensitive unit of measure.
That may be true, but it's still a pretty meaningless number if you're not actually consuming gasoline. I could measure the power consumption based off the number of candles it takes to power a steam-based generator, but that would be equally stupid. Like the ongoing "Libraries of Congress" thing.
If you're going to use an indirect unit of measure to determine the vehicle's efficiency, surely you'd be better off with pounds of CO2 as a waste-product of the energy production to power the vehicle for an hour. Or something like that.
Yes, the tracker is public. However even to browse to many of the older torrents, you need to be a member. So many of their torrents end up on other sites so it barely matters, but it's still not as wide-open as TPB.
I'm sure you won't like the example, but Facebook is almost entirely powered by MySQL. Granted, it's very heavily modified (at least according to their jobs pages) to provide better support for pretty much everything that'll get mentioned in all of these comments, but they say that most or all of those changes will be released back into the public for future revisions.
I believe that Google also uses MySQL heavily, or at least did at one point. However, that's just some vague recollection and could be totally wrong.
That certainly sounds like a pretty reasonable approach, if you must have the traffic camera in place. If - and only if - the yellow lasts long enough to provide sufficient stopping time.