Any "enterprise" that isn't at a minimum strongly exploring mobile is missing the boat big time. We're at a point now where everything in enterprise should be considering mobile..Net development is far from "button pusher". The fact is C#/.Net is really Java, only done right. MS had the benifit of seeing all the major ways Java screwed up and were able to avoid all that baggage and effectively leap frog it.
The raw ignorance you're expressing is honestly incredible.
If they don't show I, I really don't give a flying what they paid for their medallion (and most Taxi drivers in SF are contract workers, with the medallion being owned by the taxi company; the driver is just an employee with no benefits who has to follow radio orders).
That's just it, in SF the medallions are not owned by the taxi company. They're privately owned. Sure, most of the drivers are contracted...but they are contracted to the medallion owners (individually), not the taxi companies. Or more correctly, the drivers rent the medallions and so aside from paying that rent not even the medallion owners care what the driver does.
The taxi companies have no leverage over the drivers, they can't fire them or otherwise demand they actually do what they're told.
The sad fact is they don't have to follow radio orders, ever, from anyone.
If they were anywhere else they'd get hailed (almost instantly) and have to drive. And they can't just turn the service light off and not take riders, they wouldn't be "working".
In San Francisco "taxi services" do not own medallions generally, individual drivers do. This is a distinction that matters a lot.
It means that a customer can call a taxi service and request a cab...and the service will put the call out on the air...and no cabs will care to pick it up (bad neighborhood, too out of the way, whatever). The service can't hold the drivers accountable (the drivers own the medallions) and the customer just gets to stand on the corner forever at times...completely unable to get a cab from any of the "5 pages" of taxi services (since they're all asking largely the same pool of medallion-carrying drivers). -The "services" get kick backs from the drivers for routing them calls, that's how they make money: The drivers are their customers, not the passengers.
And...Most drivers however, aren't even medallion holders. I know, I just said they were, but they're not. Drivers with medallions rent their medallions out to other drivers (thus ensuring the expensive medallion is making money 24/7 for the owner). Those sub-sub-contracted drivers are most of the actual drivers in the city.
The actual owner is required to "drive" some number of hours each week to maintain the medallion. But they're lazy...they own a medallion (read: free money), so why should they actually work? So they do their hours by sitting in the taxi waiting line at the airports, watching YouTube on their phones.
DOZENS of them...all in a line for ages at the airports...transporting no one...for hours at a time. Meanwhile the actual CITY of San Francisco suffers a severe lack of taxi service.
-----
And no one is accountable. Not the "taxi services" (they can't fire or refuse to "hire" a given cab). Not the medallion holders (who largely see their medallion as a free-ride and so long as the sub-contracted driver pays their rent who cares). And definitely not the consumers (who have no ability to select or rate drivers...and it wouldn't matter if they could because again, most drivers aren't medallion owners and the owners don't give a damn).
And it's been this completely fucked up for at least half a century, probably longer.
It depends on the class of service (delivery time guarantees, etc), if signatures are required, and the neighborhood (UPS/FedEx isn't rolling a truck deep into the suburbs to deliver one or two packages unless the customer paid extra for guaranteed delivery dates).
For a lot of folks that don't live in dense cities with lots of online shoppers (ie, still most of the country), it's extremely common for USPS to handle the last mile. FedEx alone sends nearly two billion dollars of business annually to USPS for last mile service. USPS is far more of a business partner with "private" shipping companies than it is a competitor.
There are plenty of private carriers that deliver physical things. UPS, FedEX, DHL, others. Typically the private carriers do a better job. In my opinion that makes private service worth paying more for. I haven't used USPS to send things for years.
Aaaaand... all those "private" carriers contract with the USPS for much of their final delivery, especially residential delivery.
Blame the design (especially the migration design), where the real fault lies. Not with Agile.
For example, switchable code, which is where a lot of the industry is going. Got a new Group Home Page implementation? That's great! Make it configurable.
Who controls the configuration that turns off the old and on the new? That depends. It could be an opt-in beta (user controlled). It could be a forced-beta (only 1% of users or groups, randomly selected). It could be invite only.
Switchable code lets you "ship early and often" and leverage the best parts of Agile (early, fast feedback) without blowing up the entire thing with a stupid "all or nothing" release. You can test the waters (user feedback, performance, data integrity) on a limited scale but still in an actual real world environment that increasingly is impossible to accurately simulate.
---
But hell, even without switchable code if they were doing Agile "right" the shipped increments of code would have each been so small that the chance of an user revolt like this is nearly impossible. It really sounds like Yahoo wasn't using Agile, rather they're using Scrummerfall...the worst of all worlds.
Anyone who wants to keep the information out of the public (and away from competitors) without legally acknowledging wrong doing. ie, a shell corporation for the companies he claims stole his IP.
Anyone who wants to keep what are very likely current survalance protocols hidden, "2 decades old tech" or not. ie, the US government.
Anyone who wants to subvert that technology. ie any non-US government as well as plenty of multinational corporations who routinely engage in corporate spying on each other and governments.
Any inquisitive tech multimillionaire/billionaire. ie, Zuckerberg, any of the Google folks, etc.
All of which makes the eBay auction an absolutely brilliant legal maneuver:
To shut down the eBay auction Lockheed would have to claim ownership of the IP...which would mean admitting to fraud and making it legally liable to a hell of a lot more then just his 10% share (especially given the claims of continued use of his IP in other projects), likely including serious criminal charges.
If they don't take the bait...he's potentially $10M richer and still retains IP rights and can still use that to pursue the use of his tech in other projects (drones, etc).
It's by far better legal leverage than he'd ever see in a court room.
Even worse for sales tax rates...many small cities are too small for their own zip code (intended only for assisting the post office, not the IRS), even if they are "big enough" for their own legislature and thus big enough to have their own sales tax rates.
And the rates change somewhere in the country every single day.
Never, ever try to implement your own sales tax lookup system, it is absolutely doomed to fail. I don't care how good you are or how much money you throw at it, you will fail, badly.
If you're ever faced with this nightmare, hire one of the sales tax rate services to do the lookups. Only if your entire business is specifically about keeping sales tax records straight will they actually be correct, so hire someone for whom that is there only business.
You do realize many of these options are comparable to compile time options in Apache and/or PHP?
It's extraordinarily rare that they need tweaking...but when they do, I'll take a registry key (trivially managed via PowerShell btw) over a complete reconfigure and recompile from source of Apache and/or PHP, etc.
Sadly, it's probably a side effect of Android largely being rebranded Linux, which has never really solved its horrific sound system issues...:-(
But yep, I see Androids everywhere...except the music (eg, Guitar Center, Sam Ash, etc) stores. Absolutely everything there is strictly iDevice if it's anything.
The overview and management of a tool like iTunes is indispensable when you have a large music library... I have 24 k items
Except for the fact iTunes can't sanely handle large libraries.
When I had an iPhone I tried to scan my own collection into iTunes, some 30k+ tracks at the time. Two full days later it was still scanning, I shit you not.
WinAmp scans the same collection in about 5 minutes and is far superior in every possible way when trying to actually manage such huge, diverse collections sanely (WinAmp's Smart Filters are far better then iTunes's insanity). Windows Media Player scans about as fast, although the library management functionality isn't much better then iTunes (but it's still better).
And WinAmp manages any portable device. Well, except iDevices...because they're the only ones on the planet with that retarded proprietary syncing nonsense rather then "just copy it to a folder".
The only users I've ever heard praise iTunes were Mac users. I'm honestly not sure if that's because the Mac version of iTunes is wildly better (not hard to believe) or of Mac user's expectations in general are just lower (also, not hard to believe).
Bash can't pickup a random.so shared library and start instantiating objects, calling methods, as if it was just any normal bash object. That's the kind of power we're talking about with PowerShell: Any DLL on the system is at your command, no need for any "glue".
Perl can't do it either, you'll need an XS wrapper. Python I believe is the same. Ruby I'm not sure of, but I highly doubt it.
Not that PowerShell isn't without its warts. It's a great idea, but with one of the worst implementations I've ever had the displeasure of dealing with. String quoting rules are all over the place and incredibly error prone. The pipe everything obsession with syntax is insane. Wildcard handling is so horribly broken that most commands have both -Path and -LiteralPath options.
But the idea and the power that idea offers is absolutely incredible. The closest equivalent in the open world would be something like what Groovy script is to Java.
You can backup IIS's config just the same. It's just an XML file (and a surprisingly easy to read/understand one at that).
You can also do your config by editing it as well, although typically you'll use something like AppCmd or more modernly PowerShell.
It's frankly easier to reliably automate/script IIS configuration changes than Apache. Apache's configuration system is incredibly powerful and at times that's needed, but that power also means it's effectively impossible for a random admin script to make sense of it enough to modify. Such a tool must intrinsically know not just Apache's config system...but your specific implementation with it. AppCmd and PowerShell can pretty reliably walk into nearly any IIS setup, no matter how convoluted, and safely make additions, tweaks, etc.
Frankly I'm first and foremost an Apache fan, have been since it was literally A Patchy Server. And I still deploy it more often than not, often in front of IIS to get some clever hack done that just isn't practical in IIS.
But that said...I'm warming up to IIS, especially as C#/.Net gains major traction in the wake of Oracle's kiss of death to Java.
Why use "email" at all, why not Usenet or a web forum? Eliminate the entire point-to-point tracking problem.
Just throw your message (encrypted and signed of course) onto whatever forum you feel like, from whatever location, via whatever server (especially nice with Usenet). Only the receiver can tell if it's actually for them (their decrypt key works) and from you (signature matches). Use bots to mildly spam other nonsense messages encrypted and signed to add noise to the signal.
You could do something similar with torrents too, but for that both sides would need a steady supply of attractive content to act as the carrier. Game of Thrones episodes wouldn't get you very far. You'd need something like The Simpsons.
Here's the problem: Most all of what makes a good lawyer or doctor isn't at all about finding creative, novel ways to solve problems. Much the opposite in fact; Creative application of law or medicine is most likely to get you disbarred or thrown in jail. It's much easier to create a quantifiable exam when the subject matter is so well defined and creativity is shunned.
In software however, it's completely the opposite. Creative, novel application of existing technology and/or the invention of entirely new technology, is a good software developer's bread and butter. It's a big part of the essence of what makes them a good developer rather then a coding drone.
So how do you create an exam to quantify a good developer? By the very nature of what you're looking for the only "right" answers are "wrong" ones. But which wrong answer is right? That's completely subjective in an exam setting, however in the real world it's much easier to quantify: Your shit works and works well or it doesn't.
No, they aren't. With few exceptions they are all 100% company owned (at least in the 'States).
Starbucks is not a franchise, which is why they've been able to push out so many local coffee shops. With the weight of the entire corporation able to be brought to bear at any single location, they've been able to run locations at a loss for years right next to mom & pop shops. The "genius" here is that they don't need to do better then the mom & pop to win...they simply need to reduce the income of the mom & pop shop enough to make them unprofitable...and then wait them out until they leave. Then with the location clear of competition they get all the business.
NOT being locally owned is the key to Starbuck's domination.
Not necessarily. In a recent benchmark, Firefox beat Chrome.
Then I've no idea what they were benchmarking.
Chrome is at least an order of magnitude faster the Firefox in nearly everything. At least in real time (which is all that actually matters). It doesn't take "benchmarks" to see clearly how snappy Chrome is and how much of a total dog Firefox has become.
Granted, it's probably the JavaScript engine more then anything else. But on today's web, JavaScript is where 90% of sites spend 90% of their time. And Chrome's JS engine is miles ahead of Firefox in performance.
And god forbid you fire up Firebug these days as a developer...it'll take even the most tricked out machines to their knees in a second. Conversely you can run with Chrome's (far superior) debugger running and not even feel it.
That's just it, it's much more then even VPSs. Sure you can just use them as VPSs, and a lot of folks find that's the easiest first step into the cloud since it's all largely familiar. They still have "servers", they still connect the dots in the same way with the same tools and same terms.
That said, almost no one has much interest in VPSs. Not the providers, not the developers. It's all the same work (and shortcomings) as before, with a slight cost savings. That technology isn't advancing, no one really cares.
Instead the cloud has taken SOA to the next logical step. Everything is a service and all of them can be swapped in or out as needed. Send an email, transcode a media file, lookup a tax rate, whatever. They're all moving to services, tied together with cloud APIs. In that model a "web server" is just another service. What is breaking down is the idea of "a server"; The high end development taking place has gone way past the traditional idea of "a server" or "servers".
I may want a search index service (solr, etc) integrated into my site, but do I really need or care about the particulars supporting it? Not really. I want it fast, reliable, scalable, etc. All things someone else far more skilled then I in all things indexed has already figured out how to do very well. So instead of firing up a bunch of VPSs of my own to run solr, configure them all, get clustering working, worry about DR, etc. In stead of all that costly, error-prone crap I can hire a company that runs solr as a service. I'll pay for as much (or as little) as my site actually makes use of that service.
The cloud is quickly becoming an API for web development very akin to OS level APIs for local application development. It becomes a matter of gluing the features you want together to make your app, not coding each feature up almost from scratch. The web is moving far too fast to be reinventing the wheels all the time and 95% of every site is the same exact set of features and problems. Spend your time and money on the unique 5% of your product, rent the common widgets from someone who does it better already. And trust me, there's always someone who does it better then you ever will.
That's the thing; Most (all?) the "cloud" providers are baking in DR. And redundancy (of everything). And load scaling. And CDN. And all at a fraction of the cost (especially upfront) of traditional hosting and management.
If you think "cloud" is just a rebranding of colocation or even managed hosting, you really have a lot of learning to do. Just because it's hyped up doesn't mean there's nothing real there. Cloud hosting is a sea change.
Manned aircraft don't suffer from the communications issue. You can explain to a pilot what you want him to do, send him up in an armed aircraft and wait for him to come back. He can manage the task if the target moves or shows up in the wrong place. He can react to unforeseen circumstances and modify how he executes his task and still achieve the goals. You don't have to watch what he's doing to make sure the mission continues and you don't have to talk to him along the way. You can send him in a stealth aircraft and not need to put a RF source on it too.
The fact is all of this applies to modern drone technology as well and in many ways far better then a human. The days of drones as glorified R/C model airplanes is long, long gone.
Slashdot Mirror
Electric cars require only a fraction of the maintenance of traditional cars.
Also it's trivial for the car and road to track the car's maintenance schedule and simply refuse an ill-maintained car access to the express lane.
Technologically it's a non-problem.
Wow...
Any "enterprise" that isn't at a minimum strongly exploring mobile is missing the boat big time. We're at a point now where everything in enterprise should be considering mobile. .Net development is far from "button pusher". The fact is C#/.Net is really Java, only done right. MS had the benifit of seeing all the major ways Java screwed up and were able to avoid all that baggage and effectively leap frog it.
The raw ignorance you're expressing is honestly incredible.
That's just it, in SF the medallions are not owned by the taxi company. They're privately owned. Sure, most of the drivers are contracted...but they are contracted to the medallion owners (individually), not the taxi companies. Or more correctly, the drivers rent the medallions and so aside from paying that rent not even the medallion owners care what the driver does.
The taxi companies have no leverage over the drivers, they can't fire them or otherwise demand they actually do what they're told.
The sad fact is they don't have to follow radio orders, ever, from anyone.
If they were anywhere else they'd get hailed (almost instantly) and have to drive. And they can't just turn the service light off and not take riders, they wouldn't be "working".
In San Francisco "taxi services" do not own medallions generally, individual drivers do. This is a distinction that matters a lot.
It means that a customer can call a taxi service and request a cab...and the service will put the call out on the air...and no cabs will care to pick it up (bad neighborhood, too out of the way, whatever). The service can't hold the drivers accountable (the drivers own the medallions) and the customer just gets to stand on the corner forever at times...completely unable to get a cab from any of the "5 pages" of taxi services (since they're all asking largely the same pool of medallion-carrying drivers). -The "services" get kick backs from the drivers for routing them calls, that's how they make money: The drivers are their customers, not the passengers.
And...Most drivers however, aren't even medallion holders. I know, I just said they were, but they're not. Drivers with medallions rent their medallions out to other drivers (thus ensuring the expensive medallion is making money 24/7 for the owner). Those sub-sub-contracted drivers are most of the actual drivers in the city.
The actual owner is required to "drive" some number of hours each week to maintain the medallion. But they're lazy...they own a medallion (read: free money), so why should they actually work? So they do their hours by sitting in the taxi waiting line at the airports, watching YouTube on their phones.
DOZENS of them...all in a line for ages at the airports...transporting no one...for hours at a time. Meanwhile the actual CITY of San Francisco suffers a severe lack of taxi service.
-----
And no one is accountable. Not the "taxi services" (they can't fire or refuse to "hire" a given cab). Not the medallion holders (who largely see their medallion as a free-ride and so long as the sub-contracted driver pays their rent who cares). And definitely not the consumers (who have no ability to select or rate drivers...and it wouldn't matter if they could because again, most drivers aren't medallion owners and the owners don't give a damn).
And it's been this completely fucked up for at least half a century, probably longer.
The medallion system is completely indefensible.
"It depends"
It depends on the class of service (delivery time guarantees, etc), if signatures are required, and the neighborhood (UPS/FedEx isn't rolling a truck deep into the suburbs to deliver one or two packages unless the customer paid extra for guaranteed delivery dates).
For a lot of folks that don't live in dense cities with lots of online shoppers (ie, still most of the country), it's extremely common for USPS to handle the last mile. FedEx alone sends nearly two billion dollars of business annually to USPS for last mile service. USPS is far more of a business partner with "private" shipping companies than it is a competitor.
Aaaaand... all those "private" carriers contract with the USPS for much of their final delivery, especially residential delivery.
Blame the design (especially the migration design), where the real fault lies. Not with Agile.
For example, switchable code, which is where a lot of the industry is going. Got a new Group Home Page implementation? That's great! Make it configurable.
Who controls the configuration that turns off the old and on the new? That depends. It could be an opt-in beta (user controlled). It could be a forced-beta (only 1% of users or groups, randomly selected). It could be invite only.
Switchable code lets you "ship early and often" and leverage the best parts of Agile (early, fast feedback) without blowing up the entire thing with a stupid "all or nothing" release. You can test the waters (user feedback, performance, data integrity) on a limited scale but still in an actual real world environment that increasingly is impossible to accurately simulate.
---
But hell, even without switchable code if they were doing Agile "right" the shipped increments of code would have each been so small that the chance of an user revolt like this is nearly impossible. It really sounds like Yahoo wasn't using Agile, rather they're using Scrummerfall...the worst of all worlds.
Anyone who wants to keep the information out of the public (and away from competitors) without legally acknowledging wrong doing. ie, a shell corporation for the companies he claims stole his IP.
Anyone who wants to keep what are very likely current survalance protocols hidden, "2 decades old tech" or not. ie, the US government.
Anyone who wants to subvert that technology. ie any non-US government as well as plenty of multinational corporations who routinely engage in corporate spying on each other and governments.
Any inquisitive tech multimillionaire/billionaire. ie, Zuckerberg, any of the Google folks, etc.
All of which makes the eBay auction an absolutely brilliant legal maneuver:
To shut down the eBay auction Lockheed would have to claim ownership of the IP...which would mean admitting to fraud and making it legally liable to a hell of a lot more then just his 10% share (especially given the claims of continued use of his IP in other projects), likely including serious criminal charges.
If they don't take the bait...he's potentially $10M richer and still retains IP rights and can still use that to pursue the use of his tech in other projects (drones, etc).
It's by far better legal leverage than he'd ever see in a court room.
Even worse for sales tax rates...many small cities are too small for their own zip code (intended only for assisting the post office, not the IRS), even if they are "big enough" for their own legislature and thus big enough to have their own sales tax rates.
And the rates change somewhere in the country every single day.
Never, ever try to implement your own sales tax lookup system, it is absolutely doomed to fail. I don't care how good you are or how much money you throw at it, you will fail, badly.
If you're ever faced with this nightmare, hire one of the sales tax rate services to do the lookups. Only if your entire business is specifically about keeping sales tax records straight will they actually be correct, so hire someone for whom that is there only business.
You do realize many of these options are comparable to compile time options in Apache and/or PHP?
It's extraordinarily rare that they need tweaking...but when they do, I'll take a registry key (trivially managed via PowerShell btw) over a complete reconfigure and recompile from source of Apache and/or PHP, etc.
Sadly, it's probably a side effect of Android largely being rebranded Linux, which has never really solved its horrific sound system issues... :-(
But yep, I see Androids everywhere...except the music (eg, Guitar Center, Sam Ash, etc) stores. Absolutely everything there is strictly iDevice if it's anything.
Except for the fact iTunes can't sanely handle large libraries.
When I had an iPhone I tried to scan my own collection into iTunes, some 30k+ tracks at the time. Two full days later it was still scanning, I shit you not.
WinAmp scans the same collection in about 5 minutes and is far superior in every possible way when trying to actually manage such huge, diverse collections sanely (WinAmp's Smart Filters are far better then iTunes's insanity). Windows Media Player scans about as fast, although the library management functionality isn't much better then iTunes (but it's still better).
And WinAmp manages any portable device. Well, except iDevices...because they're the only ones on the planet with that retarded proprietary syncing nonsense rather then "just copy it to a folder".
The only users I've ever heard praise iTunes were Mac users. I'm honestly not sure if that's because the Mac version of iTunes is wildly better (not hard to believe) or of Mac user's expectations in general are just lower (also, not hard to believe).
In general actually.
Bash can't pickup a random .so shared library and start instantiating objects, calling methods, as if it was just any normal bash object. That's the kind of power we're talking about with PowerShell: Any DLL on the system is at your command, no need for any "glue".
Perl can't do it either, you'll need an XS wrapper. Python I believe is the same. Ruby I'm not sure of, but I highly doubt it.
Not that PowerShell isn't without its warts. It's a great idea, but with one of the worst implementations I've ever had the displeasure of dealing with. String quoting rules are all over the place and incredibly error prone. The pipe everything obsession with syntax is insane. Wildcard handling is so horribly broken that most commands have both -Path and -LiteralPath options.
But the idea and the power that idea offers is absolutely incredible. The closest equivalent in the open world would be something like what Groovy script is to Java.
You can backup IIS's config just the same. It's just an XML file (and a surprisingly easy to read/understand one at that).
You can also do your config by editing it as well, although typically you'll use something like AppCmd or more modernly PowerShell.
It's frankly easier to reliably automate/script IIS configuration changes than Apache. Apache's configuration system is incredibly powerful and at times that's needed, but that power also means it's effectively impossible for a random admin script to make sense of it enough to modify. Such a tool must intrinsically know not just Apache's config system...but your specific implementation with it. AppCmd and PowerShell can pretty reliably walk into nearly any IIS setup, no matter how convoluted, and safely make additions, tweaks, etc.
Frankly I'm first and foremost an Apache fan, have been since it was literally A Patchy Server. And I still deploy it more often than not, often in front of IIS to get some clever hack done that just isn't practical in IIS.
But that said...I'm warming up to IIS, especially as C#/.Net gains major traction in the wake of Oracle's kiss of death to Java.
Why use "email" at all, why not Usenet or a web forum? Eliminate the entire point-to-point tracking problem.
Just throw your message (encrypted and signed of course) onto whatever forum you feel like, from whatever location, via whatever server (especially nice with Usenet). Only the receiver can tell if it's actually for them (their decrypt key works) and from you (signature matches). Use bots to mildly spam other nonsense messages encrypted and signed to add noise to the signal.
You could do something similar with torrents too, but for that both sides would need a steady supply of attractive content to act as the carrier. Game of Thrones episodes wouldn't get you very far. You'd need something like The Simpsons.
Here's the problem: Most all of what makes a good lawyer or doctor isn't at all about finding creative, novel ways to solve problems. Much the opposite in fact; Creative application of law or medicine is most likely to get you disbarred or thrown in jail. It's much easier to create a quantifiable exam when the subject matter is so well defined and creativity is shunned.
In software however, it's completely the opposite. Creative, novel application of existing technology and/or the invention of entirely new technology, is a good software developer's bread and butter. It's a big part of the essence of what makes them a good developer rather then a coding drone.
So how do you create an exam to quantify a good developer? By the very nature of what you're looking for the only "right" answers are "wrong" ones. But which wrong answer is right? That's completely subjective in an exam setting, however in the real world it's much easier to quantify: Your shit works and works well or it doesn't.
No, they aren't. With few exceptions they are all 100% company owned (at least in the 'States).
Starbucks is not a franchise, which is why they've been able to push out so many local coffee shops. With the weight of the entire corporation able to be brought to bear at any single location, they've been able to run locations at a loss for years right next to mom & pop shops. The "genius" here is that they don't need to do better then the mom & pop to win...they simply need to reduce the income of the mom & pop shop enough to make them unprofitable...and then wait them out until they leave. Then with the location clear of competition they get all the business.
NOT being locally owned is the key to Starbuck's domination.
Clearly you haven't tried out Plex yet. The Roku is absolutely fantastic for playing back local content.
Check out the --link-dest option of rsync, it does exactly as you're describing, hard linking to unchanged files.
The dated directories you'll have to code yourself with a trivial amount of shell scripting.
Then I've no idea what they were benchmarking.
Chrome is at least an order of magnitude faster the Firefox in nearly everything. At least in real time (which is all that actually matters). It doesn't take "benchmarks" to see clearly how snappy Chrome is and how much of a total dog Firefox has become.
Granted, it's probably the JavaScript engine more then anything else. But on today's web, JavaScript is where 90% of sites spend 90% of their time. And Chrome's JS engine is miles ahead of Firefox in performance.
And god forbid you fire up Firebug these days as a developer...it'll take even the most tricked out machines to their knees in a second. Conversely you can run with Chrome's (far superior) debugger running and not even feel it.
That's just it, it's much more then even VPSs. Sure you can just use them as VPSs, and a lot of folks find that's the easiest first step into the cloud since it's all largely familiar. They still have "servers", they still connect the dots in the same way with the same tools and same terms.
That said, almost no one has much interest in VPSs. Not the providers, not the developers. It's all the same work (and shortcomings) as before, with a slight cost savings. That technology isn't advancing, no one really cares.
Instead the cloud has taken SOA to the next logical step. Everything is a service and all of them can be swapped in or out as needed. Send an email, transcode a media file, lookup a tax rate, whatever. They're all moving to services, tied together with cloud APIs. In that model a "web server" is just another service. What is breaking down is the idea of "a server"; The high end development taking place has gone way past the traditional idea of "a server" or "servers".
I may want a search index service (solr, etc) integrated into my site, but do I really need or care about the particulars supporting it? Not really. I want it fast, reliable, scalable, etc. All things someone else far more skilled then I in all things indexed has already figured out how to do very well. So instead of firing up a bunch of VPSs of my own to run solr, configure them all, get clustering working, worry about DR, etc. In stead of all that costly, error-prone crap I can hire a company that runs solr as a service. I'll pay for as much (or as little) as my site actually makes use of that service.
The cloud is quickly becoming an API for web development very akin to OS level APIs for local application development. It becomes a matter of gluing the features you want together to make your app, not coding each feature up almost from scratch. The web is moving far too fast to be reinventing the wheels all the time and 95% of every site is the same exact set of features and problems. Spend your time and money on the unique 5% of your product, rent the common widgets from someone who does it better already. And trust me, there's always someone who does it better then you ever will.
That's the thing; Most (all?) the "cloud" providers are baking in DR. And redundancy (of everything). And load scaling. And CDN. And all at a fraction of the cost (especially upfront) of traditional hosting and management.
If you think "cloud" is just a rebranding of colocation or even managed hosting, you really have a lot of learning to do. Just because it's hyped up doesn't mean there's nothing real there. Cloud hosting is a sea change.
The fact is all of this applies to modern drone technology as well and in many ways far better then a human. The days of drones as glorified R/C model airplanes is long, long gone.