What you want is a CERTIFICATE, not a degree. Undergraduate degrees from reputable (accredited) institutions, by their very nature, involve not just some level of mastery of a field, but also a broad education to make one a better member of society, not just a whiz at one skill.
Moreover, if you think that "these fields... will not contribute to making me better at my job", then (a) you are dead wrong, and (b) you probably need more exposure to "those fields" more than the average college student, so for you to skip them would be an ESPECIALLY bad idea.
So: either go get some comp certificate from a community/technical college, then maybe supplement it with a few more core CS classes from a university (and try to compete in the world without a bachelors degree), or else just buckle down and fulfill whatever requirements the school asks you to do.
I also wouldn't get too caught up in BA vs BS. You'll do fine either way. If you really want to go in-depth in your education, continue on for a MS. Otherwise, realize that you'll probably learn a lot more by actually working in your field than you do in school anyway (given a suitable foundation).
I hope Mozilla gets hurt enough by this to re-think what they are doing. Among other things, their quality does not seem to be improving as a result of their new numbering. FF4 has some notable regressions from 3.6, which are for the most part NOT fixed in FF5 (which has its own additional regressions). So I don't see FF5 as FF 4.1, but more like FF 4.0 beta 2, compared to the FF 4.0 beta 1 they released as "Firefox 4".
And I'm with others - I don't care how they number things, or whether FF5 is just a minor security update to FF4. But a minor security update does NOT break existing plug-ins, as FF5 apparently does. So they want to have their cake and eat it too, which means they are getting lazy, which means that Google and MS will eat Mozilla's cake instead.
Fortunately, the combined market share of Chrome, Safari and various mobile browsers - all WebKit brethren - means webkit market share is getting competitive with Firefox market share (or IE8 market share), so I may feel comfortable making Chrome my primary dev/test browser going forward. And good riddance to Firefox's massive memory overusage.
Maybe this isn't the use case of the original poster, but here is the situation that I expect most corporations face: - Need to share files across the company or workgroup - Files on individual workstations need to be accessible when offline from server - Files on individual workstations need to be encrypted (in case laptop stolen, etc.) - Need various group-based permissions for files belonging to different projects etc. (then add users to appropriate groups) - History, version control, etc. - Secure data transmission between client and server - Server repository encrypted (and if on a hosting service, needs separate key from other customers); preferably able to install server software on own hardware for full control/privacy (and possibly keep within VPN) - Client software fully supported on Windows and Mac (and preferably Linux, smartphones) - Dead easy to use - just open and save files in the mirrored drive/directory and they sync/version automatically - Bonus: shareable links (to files or folders), maybe sends an email containing the link. (Best to have tracking of when, by whom and with whom the file is shared, and policy control over whether allowed for certain files/folders/users, plus maybe expiration time on the link) - Another bonus - web service to view files where you don't have the application installed locally
I haven't really evaluated them, but Spideroak and Sycplicity might be the closest that I've seen.
Many content management systems can handle links like this without any human intervention (let alone expert human intervention), automatically updating all references within the system to a resource if it moves or changes names (while still being able to use descriptive names that can be changed at will). Some will also automatically add redirects when things move, so that external links coming in to the old URL can continue working as well. If some references need to point to, say, the latest resource in a given category, which can change at any time, that can be handled automatically also.
Not sure that addresses all your issues, but it doesn't really sound like static documents are the only (or even best) way to handle this.
If you are editing your Facebook privacy settings, there is a button to "Preview My Profile", which will show "how your profile looks to most people on Facebook".
You can also see how it looks to a particular person.
Hmm, yes, did anyone else look at the pictures in the article? Did anonymous hack securityweek.com to doctor the pictures?? I wonder if the edited the article a little, too? It seems surprisingly favorable to their POV...
It does none of those things and evidently costs the city money annually on top of that.
1) I'm sure they do increase revenues, but since they also likely increase expenses they may or may not increase net income, but that's not the same thing.
2) You're right, they probably don't decrease police workload. (I'm not sure how they could, unless previously there were officers sitting at all those intersections 24 hours a day that can suddenly be re-assigned.)
3) They probably do make intersections significantly safer, if you measure by injuries, deaths, and total dollars of damage (but maybe not if you measure only by number of minor fender-benders). While potentially increasing the number of low-speed rear-end collisions (which rarely result in serious injuries since (a) they usually involve two cars, (b) they usually involve the bumpers/crumple zones of each car and (c) the relative speed between them is usually fairly low since they are traveling in the same direction and the trailing car should have some opportunity to brake), they also tend to reduce the number of (often fatal) "T-bar" side-on collisions or collisions with pedestrians/bicyclists. You may be pissed off that you messed up your bumper or crumpled your hood, or got an expensive ticket and maybe an insurance increase; I'd be far more pissed if you ran over and killed a mother pushing a stroller (in the crosswalk, on a walk signal) with her infant child in it...
The only things that they do seem to do is
1) Make money for operators 2) Piss off everyone else. 3) Erode your rights
1) True.
2) Really only those who were habitually running red lights, usually at high speed (at great hazard to other drivers, bicyclists and pedestrians) and are annoyed that they either can no longer do so or got an expensive ticket for doing it, or those who get caught unaware that they really were driving dangerously, who typically drive more carefully in the future (I know I do). (Yes, in a few cases the light is broken or an emergency vehicle was trying to get through or whatever, but they can be contested in those cases, most likely successfully.)
BTW, I suspect that drivers who have either had red-light camera experiences elsewhere or who are afraid of such tickets ARE tending to run red lights less, even if the cameras aren't present at a particular intersection, town or even state. I wouldn't be surprised if that were a significant factor (though surely not the only one) in overall reductions even when cameras aren't present.
3) How? What right do you have to violate traffic laws while operating a motor vehicle on a public road, regardless of whether someone is watching? Especially violations that can easily lead to tragedy because red-light runners tend to be going full speed (often speeding by 10mph or more) and often actually ACCELERATING in an attempt to speed through the intersection, and what they are likely to hit will be essentially stationary and completely unprotected (ie, a pedestrian, or hitting a car directly in its side), often leading to serious injuries or fatalities?
What I DO agree with (which you didn't address) is the abuse (real and potential): It should be required (and even maybe added to state laws) that use of red-light cameras be contingent on using the data to actively make IMPROVEMENTS to signal timing or intersection design in an attempt to reduce accidents, injuries, etc. (which would also tend to reduce tickets); instead, we've all heard of cases (though probably a much smaller minority than some suspect) where they seem to tinker with timings to try to INCREASE tickets (which results in making the intersection more dangerous). The latter especially should be banned and/or cities should be at least partly liable for any accidents (and maybe even tickets) that occur after they've made the intersection more dangerous.
Hmm, on second thought, this actually seems fairly insightful - Apple may be betting that many people who would actually go buy a new computer just to access iCloud have a very good chance of selecting a Mac for their next system over a PC... They probably figure that this possible sales boost outweighs the over 50% market share of XP among Windows users that they're missing out on currently (since they may also be projecting that this figure will start to drop like a rock over the next couple of years as people replace aging PCs and get Windows 7 or 8 automatically in the process).
It is news when over 50% of Windows users are still using XP (a figure that would be a lot higher if people were given a choice of what to install on new computers bought over the last several years - though corporate buyers have actually demanded, and exercised, the right to downgrade Vista/7 installs on new computers back to XP at no extra charge).
Seriously, I have Windows 7 at home (for the last 1.5 years) and use XP at work, and NOT ONCE while at work have I thought "gee, I wish I were using my home computer, because of [x] feature that's really nice/better on Windows 7". Windows 7 is different from XP, but it's just plain not any better (at least in any way that affects actual usability).
All its other under-the-hood improvements are counterbalanced by increased anti-piracy/DRM technology, which I strongly prefer to do without. (Even though I rarely if ever pirate anything, I'm philosophically opposed to allowing such things.) Not that Apple is much better on that count...
I have no desire whatsoever to "upgrade" my work computer to Windows 7.
From what I've dug up, there are several sources of potential speedup: 1) Acts as a CDN (with 5 data centers - 3 US, 1 Europe, 1 Asia) to cache static files (such as images, js, css) from a location on average nearer to most visitors, plus the cache servers are fast and well-connected. I read a claim somewhere that based on total traffic going through their system, they would be the 10th busiest site on the web (unverified). 2) Filters out enough "bad" traffic, which it never sends on to the site's originating server (while occasionally challenging a legitimate user with a CAPTCHA page in the process...), and the cached traffic of course, to noticeably reduce server load on the originating server (which is also where all non-static content comes from so the most likely to get overloaded) 3) Their internal network is claimed to be good enough and well enough connected/peered that (they claim) sites often see fewer hops and lower pings going through their network (even when not using the cache) than by a more "direct" normal route. 4) I'm sure they've enabled all the usual Apache performance enhancements, like gzip etc. (which may or may not be optimized on your average web server) 5) For the paid accounts, it uses Javascript to pre-load static files from cache for pages linked from the current page that it's determined (algorithmically through experience) a visitor is likely to go to next. Thus when you click a link, many of the needed files might already be in your browser's cache.
Other benefits: 1) Apparently (not sure if for both free and paid accounts, or for dynamic content also) can build a cached "picture" of your site (including HTML, which it normally does not serve from cache) so if your server goes down, it can still serve up its copy from cache. (Conversely, if most of its services fail, it's supposed to just pass the traffic through to your server.) 2) Unlike other CDNs, no need to modify ANYTHING on your site. 3) Can add apps such as Google Analytics or Pingdom to all pages on your site easily from their control panel (it inserts it into your HTML as it passes through their servers/filters)
Its security filtering goes both ways: incoming, it looks for various signs of malicious requests and blocks them (ie, potential SQL injection code in get/post data or other suspicious activity) and outgoing it can detect any email address and replace with a javascript, to deter harvesting of addresses. I'm not really sure what else.
They claim no bandwidth limits, but their caching probably selects some subset of the most frequently requested files to save space, and the dynamic content goes straight through to your server anyway.
The rest (aside from the filtering) is mainly DNS, and here is both the good and bad part: they become your DNS provider. In the sense that this provides an instant, zero configuration CDN (with caching and proxying) and security filtering for your site, it's great. In the sense that if their DNS has problems, your site disappears, well, this might not be ready for critical production sites that need constant uptime. And apparently they've had less-than-perfect reliability so far. (I don't know if it's terrible, or not bad but with a few bad stories, or what.)
For most of their services, failure just means traffic flows through to your site rather than gaining the benefits of their services (if it's well-designed), which is a nice mode of failure, but if the main DNS fails or some other problem makes your site inaccessible, well, that portion needs to be super-reliable and I'm not sure if they're quite there yet.
Another thing to consider is the impact on Google Analytics etc. - your server will likely see a noticeable drop in total traffic/page views (due to the filtering out of "bad" requests), but the quality of that traffic may go up (due to the filtering, and possibly a little due to your site performing better so people don't bounce as much in frustration). It's also possible Google will notice the DNS change etc. and that could also affect things.
Overall, it looks like a potentially very cool service, but I wouldn't want to use a critical/high-traffic site as a guinea pig either.
While this point of view was quite popular 5-10 years ago, now it just betrays your ignorance. We know better now. Douglas Crockford has been the most active promoter of a more mature understanding of Javascript: http://javascript.crockford.com/
Some of Javascript's repuation is, like PHP's, historical baggage based on immature early versions that lacked features or had bugs which have since been corrected.
But the real problems with Javascript are threefold: 1. Being embedded in the browser, it can only operate on the elements the browser provides; since those elements differ from browser to browser, the naive assumption is that it is the language (or at least its implementation) causing these differences. 2. Javascript may appear to be a simplified version of Java, but it's not. Crockford contends it is "Lisp in C's clothing". Most professional programmers know Java-like languages much better than Lisp, so they don't really understand Javascript's power and elegance, 3. Most people writing Javascript aren't even programmers, and those that are, are often not very good ones. While this is not unique to Javascript at all, it is certainly the case that a good chunk of the Javascript in existence is of the form of "knew just enough of the language to use it in the worst possible way". (This is often true of PHP, COBOL, and many other language as well - the easier to use and more accessible the language is, the more likely this is the case.)
Touche. I think the use of the words "backwards" and even "civilized" were not particularly appropriate or helpful; however, the point that this was the strongest earthquake to have hit a densely-populated urban area appears to be correct. To address your response:
The earthquakes in Alaska and Chile happened about 50 years ago, when those areas were much less built up than today. Valdivia Chile has a much less impressive skyline than Santiago even today, and the epicenter of that earthquake was over 400 miles from Santiago. The 2004 "Indonesian" earthquake struck off the west coast of Sumatra; Jakarta is on a sheltered side of Java about 1000 miles away.
If OS upgrades were free (and didn't involve re-installing all your software), I might be inclined to agree with you. As it is, no, I can't agree with you at all. I think people can complain all they want, though I'd much rather they just switch to Firefox or Chrome than complain. (And I expect that those that would complain will do exactly that.)
Re: pissing off content developers. You're right: as a web developer, I'm kind of doubting I will ever have a reason to test webites in IE9 or 10. I doubt IE9 or 10 will ever achieve much market share (they'll probably never break 10% each), and may be the beginning of the end of even testing for IE at all. Once IE6-8 all drop below about 3% market share each, there won't be much reason to test for old IE compatibility, and by that time time they'll have IE 11 or 12 or even 15, with increasing standards compliance, but no version with more than about 5-8% market share. If a site runs in current versions of Firefox and Chrome, AND in IE8, I'll just figure IE9+ must be AOK.
Dropping XP support in IE9 is not understandable; it is even more idiotic than dropping Win2K/98/Me support in IE7 was (a big reason for there being more IE6 users today than IE7 users), or than dropping Vista/Server 2008 support in IE10 will be.
My prediction: neither IE9 nor IE10 will ever pass 10% market share globally.
Not to argue with your point about the validation, but the chances that Epsilon had anything to do with implementing that Sears.com login page are virtually nil.
As a member of the society that has granted copyright holders their limited, temporarily monopolistic usage rights, I am at this moment reporting the misappropriation of the terms "theft", "ownership" and "property" with respect to these rights and the content they cover. (They are violations because copyright is a social contract between creators of intellectual property and society, and any contract has two sides. The MPAA et al are violating their side of the contract.) I discovered these violations on their very website! Anyone else want to complain?
Give yourself RSI all you want, when whatever you type gets into a professionally-printed publication, it will likely have one space between sentences.
Ms. Atkins is OK with the "valuable life lesson" being "I only learned 1/2 year worth of material this year, compared to 1 to 1-1/2 with a good teacher"?
Here is a better listing of specific capabilities for different browser versions: http://caniuse.com/
It's still unclear how competitive IE9 will be at the time it comes out, but it will clearly be a huge leap forward from IE8. (The big problem will be getting all the existing IE6, 7, and 8 users to migrate to IE9 or other modern browsers.)
Slashdot Mirror
Any place that is accredited in the US will have some level of GE requirements, regardless of the name.
What you want is a CERTIFICATE, not a degree. Undergraduate degrees from reputable (accredited) institutions, by their very nature, involve not just some level of mastery of a field, but also a broad education to make one a better member of society, not just a whiz at one skill.
Moreover, if you think that "these fields ... will not contribute to making me better at my job", then (a) you are dead wrong, and (b) you probably need more exposure to "those fields" more than the average college student, so for you to skip them would be an ESPECIALLY bad idea.
So: either go get some comp certificate from a community/technical college, then maybe supplement it with a few more core CS classes from a university (and try to compete in the world without a bachelors degree), or else just buckle down and fulfill whatever requirements the school asks you to do.
I also wouldn't get too caught up in BA vs BS. You'll do fine either way. If you really want to go in-depth in your education, continue on for a MS. Otherwise, realize that you'll probably learn a lot more by actually working in your field than you do in school anyway (given a suitable foundation).
I hope Mozilla gets hurt enough by this to re-think what they are doing. Among other things, their quality does not seem to be improving as a result of their new numbering. FF4 has some notable regressions from 3.6, which are for the most part NOT fixed in FF5 (which has its own additional regressions). So I don't see FF5 as FF 4.1, but more like FF 4.0 beta 2, compared to the FF 4.0 beta 1 they released as "Firefox 4".
And I'm with others - I don't care how they number things, or whether FF5 is just a minor security update to FF4. But a minor security update does NOT break existing plug-ins, as FF5 apparently does. So they want to have their cake and eat it too, which means they are getting lazy, which means that Google and MS will eat Mozilla's cake instead.
Fortunately, the combined market share of Chrome, Safari and various mobile browsers - all WebKit brethren - means webkit market share is getting competitive with Firefox market share (or IE8 market share), so I may feel comfortable making Chrome my primary dev/test browser going forward. And good riddance to Firefox's massive memory overusage.
Maybe this isn't the use case of the original poster, but here is the situation that I expect most corporations face:
- Need to share files across the company or workgroup
- Files on individual workstations need to be accessible when offline from server
- Files on individual workstations need to be encrypted (in case laptop stolen, etc.)
- Need various group-based permissions for files belonging to different projects etc. (then add users to appropriate groups)
- History, version control, etc.
- Secure data transmission between client and server
- Server repository encrypted (and if on a hosting service, needs separate key from other customers); preferably able to install server software on own hardware for full control/privacy (and possibly keep within VPN)
- Client software fully supported on Windows and Mac (and preferably Linux, smartphones)
- Dead easy to use - just open and save files in the mirrored drive/directory and they sync/version automatically
- Bonus: shareable links (to files or folders), maybe sends an email containing the link. (Best to have tracking of when, by whom and with whom the file is shared, and policy control over whether allowed for certain files/folders/users, plus maybe expiration time on the link)
- Another bonus - web service to view files where you don't have the application installed locally
I haven't really evaluated them, but Spideroak and Sycplicity might be the closest that I've seen.
Many content management systems can handle links like this without any human intervention (let alone expert human intervention), automatically updating all references within the system to a resource if it moves or changes names (while still being able to use descriptive names that can be changed at will). Some will also automatically add redirects when things move, so that external links coming in to the old URL can continue working as well. If some references need to point to, say, the latest resource in a given category, which can change at any time, that can be handled automatically also.
Not sure that addresses all your issues, but it doesn't really sound like static documents are the only (or even best) way to handle this.
As others have said, we don't even need to do it, because it's already done.
a) it's in a shared footer file
b) it's using PHP or another language so it already reads the current year
(All of which can be cached to static pages to keep the performance high.)
The other benefit is, what if you WANT something to read "©2010" and not change to 11, somewhere within a page? Your method is, unfortunately, unable to handle such a distinction.
Plus, we can use sed etc. too, when we really need to. (Not so easily in a database, but there are ways there also.)
If you are editing your Facebook privacy settings, there is a button to "Preview My Profile", which will show "how your profile looks to most people on Facebook".
You can also see how it looks to a particular person.
Hmm, yes, did anyone else look at the pictures in the article? Did anonymous hack securityweek.com to doctor the pictures?? I wonder if the edited the article a little, too? It seems surprisingly favorable to their POV...
Since the entire system was pitched to them as a way to
1) Increase revenues
2) Reduce police workload
3) Make intersections safer
It does none of those things and evidently costs the city money annually on top of that.
1) I'm sure they do increase revenues, but since they also likely increase expenses they may or may not increase net income, but that's not the same thing.
2) You're right, they probably don't decrease police workload. (I'm not sure how they could, unless previously there were officers sitting at all those intersections 24 hours a day that can suddenly be re-assigned.)
3) They probably do make intersections significantly safer, if you measure by injuries, deaths, and total dollars of damage (but maybe not if you measure only by number of minor fender-benders). While potentially increasing the number of low-speed rear-end collisions (which rarely result in serious injuries since (a) they usually involve two cars, (b) they usually involve the bumpers/crumple zones of each car and (c) the relative speed between them is usually fairly low since they are traveling in the same direction and the trailing car should have some opportunity to brake), they also tend to reduce the number of (often fatal) "T-bar" side-on collisions or collisions with pedestrians/bicyclists. You may be pissed off that you messed up your bumper or crumpled your hood, or got an expensive ticket and maybe an insurance increase; I'd be far more pissed if you ran over and killed a mother pushing a stroller (in the crosswalk, on a walk signal) with her infant child in it...
The only things that they do seem to do is
1) Make money for operators
2) Piss off everyone else.
3) Erode your rights
1) True.
2) Really only those who were habitually running red lights, usually at high speed (at great hazard to other drivers, bicyclists and pedestrians) and are annoyed that they either can no longer do so or got an expensive ticket for doing it, or those who get caught unaware that they really were driving dangerously, who typically drive more carefully in the future (I know I do). (Yes, in a few cases the light is broken or an emergency vehicle was trying to get through or whatever, but they can be contested in those cases, most likely successfully.)
BTW, I suspect that drivers who have either had red-light camera experiences elsewhere or who are afraid of such tickets ARE tending to run red lights less, even if the cameras aren't present at a particular intersection, town or even state. I wouldn't be surprised if that were a significant factor (though surely not the only one) in overall reductions even when cameras aren't present.
3) How? What right do you have to violate traffic laws while operating a motor vehicle on a public road, regardless of whether someone is watching? Especially violations that can easily lead to tragedy because red-light runners tend to be going full speed (often speeding by 10mph or more) and often actually ACCELERATING in an attempt to speed through the intersection, and what they are likely to hit will be essentially stationary and completely unprotected (ie, a pedestrian, or hitting a car directly in its side), often leading to serious injuries or fatalities?
What I DO agree with (which you didn't address) is the abuse (real and potential): It should be required (and even maybe added to state laws) that use of red-light cameras be contingent on using the data to actively make IMPROVEMENTS to signal timing or intersection design in an attempt to reduce accidents, injuries, etc. (which would also tend to reduce tickets); instead, we've all heard of cases (though probably a much smaller minority than some suspect) where they seem to tinker with timings to try to INCREASE tickets (which results in making the intersection more dangerous). The latter especially should be banned and/or cities should be at least partly liable for any accidents (and maybe even tickets) that occur after they've made the intersection more dangerous.
Hmm, on second thought, this actually seems fairly insightful - Apple may be betting that many people who would actually go buy a new computer just to access iCloud have a very good chance of selecting a Mac for their next system over a PC... They probably figure that this possible sales boost outweighs the over 50% market share of XP among Windows users that they're missing out on currently (since they may also be projecting that this figure will start to drop like a rock over the next couple of years as people replace aging PCs and get Windows 7 or 8 automatically in the process).
It is news when over 50% of Windows users are still using XP (a figure that would be a lot higher if people were given a choice of what to install on new computers bought over the last several years - though corporate buyers have actually demanded, and exercised, the right to downgrade Vista/7 installs on new computers back to XP at no extra charge).
Seriously, I have Windows 7 at home (for the last 1.5 years) and use XP at work, and NOT ONCE while at work have I thought "gee, I wish I were using my home computer, because of [x] feature that's really nice/better on Windows 7". Windows 7 is different from XP, but it's just plain not any better (at least in any way that affects actual usability).
All its other under-the-hood improvements are counterbalanced by increased anti-piracy/DRM technology, which I strongly prefer to do without. (Even though I rarely if ever pirate anything, I'm philosophically opposed to allowing such things.) Not that Apple is much better on that count...
I have no desire whatsoever to "upgrade" my work computer to Windows 7.
From what I've dug up, there are several sources of potential speedup:
1) Acts as a CDN (with 5 data centers - 3 US, 1 Europe, 1 Asia) to cache static files (such as images, js, css) from a location on average nearer to most visitors, plus the cache servers are fast and well-connected. I read a claim somewhere that based on total traffic going through their system, they would be the 10th busiest site on the web (unverified).
2) Filters out enough "bad" traffic, which it never sends on to the site's originating server (while occasionally challenging a legitimate user with a CAPTCHA page in the process...), and the cached traffic of course, to noticeably reduce server load on the originating server (which is also where all non-static content comes from so the most likely to get overloaded)
3) Their internal network is claimed to be good enough and well enough connected/peered that (they claim) sites often see fewer hops and lower pings going through their network (even when not using the cache) than by a more "direct" normal route.
4) I'm sure they've enabled all the usual Apache performance enhancements, like gzip etc. (which may or may not be optimized on your average web server)
5) For the paid accounts, it uses Javascript to pre-load static files from cache for pages linked from the current page that it's determined (algorithmically through experience) a visitor is likely to go to next. Thus when you click a link, many of the needed files might already be in your browser's cache.
Other benefits:
1) Apparently (not sure if for both free and paid accounts, or for dynamic content also) can build a cached "picture" of your site (including HTML, which it normally does not serve from cache) so if your server goes down, it can still serve up its copy from cache. (Conversely, if most of its services fail, it's supposed to just pass the traffic through to your server.)
2) Unlike other CDNs, no need to modify ANYTHING on your site.
3) Can add apps such as Google Analytics or Pingdom to all pages on your site easily from their control panel (it inserts it into your HTML as it passes through their servers/filters)
Its security filtering goes both ways: incoming, it looks for various signs of malicious requests and blocks them (ie, potential SQL injection code in get/post data or other suspicious activity) and outgoing it can detect any email address and replace with a javascript, to deter harvesting of addresses. I'm not really sure what else.
They claim no bandwidth limits, but their caching probably selects some subset of the most frequently requested files to save space, and the dynamic content goes straight through to your server anyway.
The rest (aside from the filtering) is mainly DNS, and here is both the good and bad part: they become your DNS provider. In the sense that this provides an instant, zero configuration CDN (with caching and proxying) and security filtering for your site, it's great. In the sense that if their DNS has problems, your site disappears, well, this might not be ready for critical production sites that need constant uptime. And apparently they've had less-than-perfect reliability so far. (I don't know if it's terrible, or not bad but with a few bad stories, or what.)
For most of their services, failure just means traffic flows through to your site rather than gaining the benefits of their services (if it's well-designed), which is a nice mode of failure, but if the main DNS fails or some other problem makes your site inaccessible, well, that portion needs to be super-reliable and I'm not sure if they're quite there yet.
Another thing to consider is the impact on Google Analytics etc. - your server will likely see a noticeable drop in total traffic/page views (due to the filtering out of "bad" requests), but the quality of that traffic may go up (due to the filtering, and possibly a little due to your site performing better so people don't bounce as much in frustration). It's also possible Google will notice the DNS change etc. and that could also affect things.
Overall, it looks like a potentially very cool service, but I wouldn't want to use a critical/high-traffic site as a guinea pig either.
While this point of view was quite popular 5-10 years ago, now it just betrays your ignorance. We know better now. Douglas Crockford has been the most active promoter of a more mature understanding of Javascript: http://javascript.crockford.com/
Some of Javascript's repuation is, like PHP's, historical baggage based on immature early versions that lacked features or had bugs which have since been corrected.
But the real problems with Javascript are threefold:
1. Being embedded in the browser, it can only operate on the elements the browser provides; since those elements differ from browser to browser, the naive assumption is that it is the language (or at least its implementation) causing these differences.
2. Javascript may appear to be a simplified version of Java, but it's not. Crockford contends it is "Lisp in C's clothing". Most professional programmers know Java-like languages much better than Lisp, so they don't really understand Javascript's power and elegance,
3. Most people writing Javascript aren't even programmers, and those that are, are often not very good ones. While this is not unique to Javascript at all, it is certainly the case that a good chunk of the Javascript in existence is of the form of "knew just enough of the language to use it in the worst possible way". (This is often true of PHP, COBOL, and many other language as well - the easier to use and more accessible the language is, the more likely this is the case.)
Touche. I think the use of the words "backwards" and even "civilized" were not particularly appropriate or helpful; however, the point that this was the strongest earthquake to have hit a densely-populated urban area appears to be correct. To address your response:
The earthquakes in Alaska and Chile happened about 50 years ago, when those areas were much less built up than today.
Valdivia Chile has a much less impressive skyline than Santiago even today, and the epicenter of that earthquake was over 400 miles from Santiago.
The 2004 "Indonesian" earthquake struck off the west coast of Sumatra; Jakarta is on a sheltered side of Java about 1000 miles away.
If OS upgrades were free (and didn't involve re-installing all your software), I might be inclined to agree with you. As it is, no, I can't agree with you at all. I think people can complain all they want, though I'd much rather they just switch to Firefox or Chrome than complain. (And I expect that those that would complain will do exactly that.)
Re: pissing off content developers. You're right: as a web developer, I'm kind of doubting I will ever have a reason to test webites in IE9 or 10. I doubt IE9 or 10 will ever achieve much market share (they'll probably never break 10% each), and may be the beginning of the end of even testing for IE at all. Once IE6-8 all drop below about 3% market share each, there won't be much reason to test for old IE compatibility, and by that time time they'll have IE 11 or 12 or even 15, with increasing standards compliance, but no version with more than about 5-8% market share. If a site runs in current versions of Firefox and Chrome, AND in IE8, I'll just figure IE9+ must be AOK.
Dropping XP support in IE9 is not understandable; it is even more idiotic than dropping Win2K/98/Me support in IE7 was (a big reason for there being more IE6 users today than IE7 users), or than dropping Vista/Server 2008 support in IE10 will be.
My prediction: neither IE9 nor IE10 will ever pass 10% market share globally.
Not to argue with your point about the validation, but the chances that Epsilon had anything to do with implementing that Sears.com login page are virtually nil.
I notice that the MPAA site has a "Report Piracy" form at http://www.mpaa.org/contentprotection/report-piracy.
As a member of the society that has granted copyright holders their limited, temporarily monopolistic usage rights, I am at this moment reporting the misappropriation of the terms "theft", "ownership" and "property" with respect to these rights and the content they cover. (They are violations because copyright is a social contract between creators of intellectual property and society, and any contract has two sides. The MPAA et al are violating their side of the contract.) I discovered these violations on their very website! Anyone else want to complain?
Give yourself RSI all you want, when whatever you type gets into a professionally-printed publication, it will likely have one space between sentences.
You can always use non-breaking spaces if you insist.
In addition, I worked at one of the biggest printing firms in the world and our typesetting system also stripped out extra spaces, on purpose.
Ms. Atkins is OK with the "valuable life lesson" being "I only learned 1/2 year worth of material this year, compared to 1 to 1-1/2 with a good teacher"?
Here is a better listing of specific capabilities for different browser versions: http://caniuse.com/ It's still unclear how competitive IE9 will be at the time it comes out, but it will clearly be a huge leap forward from IE8. (The big problem will be getting all the existing IE6, 7, and 8 users to migrate to IE9 or other modern browsers.)