I know about this, and I use it, it's how I bring up /etc. Still, a royal pain in comparison to say, Windows 95 B.

The thing about these UNIX based system in the modern era - system requirements can actually go down between releases. Draw lines where they need to be, with real specs, not age. I'm running a five year old system at home more powerful than a lot of new stuff, it just takes a little more electricity to run that quad core 64bit Athlon 64 with 8 GB of RAM than if I were to build it today. Apples hardware in the Mac Pro line of the same era had the same type of power my home built system does. Age of hardware makes little to no sense, you're just a fanboy defending your religious icon.

Dare I say it - they could adopt something along the lines of the Windows Score Microsoft has been using. Run the "score card" app on your hardware before upgrading to see if upgrading is for you.

Otherwise regular USB would be "good enough" like it has been for the past decade.

I don't know how you can look at a Mini USB 3.0 and say that. It's nearly as wide as the old style Apple connector, no USB really did need a Thunderbolt or better treatment, and I hope they stick with it for a while.

It's a common courtesy to give the slow kids a head start.

My biggest problem with Qt is the same as my biggest problem with Java. It's okay for in-house test tools, but it should never be used for actual end-user apps. You'll invariably end up with an application that doesn't quite behave like a native app, doesn't perform as well as a native app, and can't take advantage of advanced features of each OS without losing portability. In short, cross-platform apps written in this way almost invariably take advantage of only the least-common-denominator functionality common to all OSes, resulting in an app that sucks equally everywhere.

There's only one right way to create a cross-platform application: by separating the core logic from the UI, and then hiring people who actually know how to write software for each target OS to designing a custom UI that conforms to the way that users on that particular platform expect an app to behave. Anything less will always be substandard on one platform at best, and on all platforms at worst.

Now you could write that core logic in Java if you really want to (though you'll run into portability problems if anybody ever asks for an iOS port), but chances are, you're better off writing it in C or C++. Either way, using Qt or Java for the UI is IMO invariably a mistake unless you're just writing an app to meet your own personal needs.

I've got a VERY large Java code base. We use a number of the tools you mention and they are quite powerful and useful.

Yes, and I have some relatively large Perl scripts that are readable and maintainable. These are the exceptions that prove the rule.... :-)

The GP's point was that Starcraft is possible to play on a server unlikely to get DDoS'd. When my friends and I play Starcraft, it's over a LAN with no internet access at all. If you wish to DDoS my game server, you'll have to trespass to do it.

Targeting any given company's game servers doesn't affect the titles who don't require that players be online to play them.

That's already been done.

The challenge DropBox faces with a self-hosted iteration of its software is that it stops being 'simple'. Existing Dropbox clients would have to be completely rewritten to go from asking "username and password, please" to "username, password, server address, and port, please". Even if we hand-wave away that problem by assuming that users can either correctly type a server name and port number, or that Dropbox will still have 'accounts' but essentially become a DynDNS clone and simply handle network traversal and matching users to their data repositories, we then have to deal with the Dropbox Server software. There may be a market for Dropbox to sell drives like these, but I don't see Western Digital wanting to partner with Dropbox to provide redundant functionality to their existing apps, and I don't see consumers paying more for a Dropbox branded drive if they're already in the "self-contained NAS" market - a handful might, but now Dropbox, for all intents and purposes, finds itself with all the challenges of being an external hard drive vendor...with the added bonus of directly competing with the vendors from whom they're sourcing their parts.

The obvious alternative to this would be for them to sell their software and let it run on a LAMP/WAMP stack, on whatever hardware is on hand, and market it to the enthusiast/enterprise market, like UnRAID or Nexenta. That might be a short term win, especially if they do some fancy stuff with LDAP/Active Directory integration. Conversely, I see it potentially being a support nightmare based on how it deals with storage. Will it install on an Ubuntu desktop containing a hodgepodge of hard disks? Would it be more like FreeNAS where it makes its own software RAID, but requires hardware to be dedicated (or its own VM)? Even at that, how do they bill for the software? One-time use seems like it wouldn't be a good long-term plan, but I don't see too many users being okay with Dropbox charging them an annual fee to use their own hard drives. CALs could be a useful method (arguably the most workable one), but they'd have a hard time managing their consumer-friendly image on one hand with Oracle-style licensing on the other.

Levie is right; 'free' isn't a business model. Dropbox's 2GB number is only sustainable because they're betting that a certain number of those users will go for a paid tier. Either every Dropbox customer will pay, or they start advertising, or they data mine. To my knowledge, those are the three business models that have sustained themselves on the internet. 'Everyone Pays' may be a viable model if Dropbox can do things like sell gift cards for their service (for users unable/unwilling to fork over their Mastercard) and come up with the right formula of how much customers are really willing to pay for storage+ubiquity+simplicity. Although Levie must certainly be feeling the pinch from Microsoft's 1TB of OneDrive for $60/year, the one client we attempted to migrate to that service went back to dropbox VERY quickly because the desktop client was utter crap; I'm left to believe that Dropbox's simplicity still has an edge just yet. Conversely, I don't think that $50/month for 500GB is worthwhile, either - That's only slightly less than it'd cost to buy a 500GB hard disk outright from Newegg every month.

Dropbox is still a well-recognized brand that I'm certain many consumers are still willing to pay a premium for, and Microsoft and Google are competing not only with more storage for less money, but with integration as well - editing a spreadsheet in Sheets or Excel and seamless saving of attachments is not the kind of thing that Dropbox can effectively compete with. Dropbox's best bet right now, in my half-asleep opinion, is to see how much value-add they CAN provide to their existing tiers. I can't quite fathom what that is (a trivial example off the top of my head would be an IM client add-on), but one thing is for sure: they can't easily compete against companies who sell their own gigabytes by selling someone else's gigabytes.

I use Apple products, but I'm by no means a fanboy, as my signature suggest (fanboys should NEVER have mod-points). I support Apple products at work, use a 27" iMac at work (which I rather like, and I've put TotalTerminal and other utils on to make it more Linux-like and comfortable for me), and I've got a work iPad 2, all of which I like.

I'm actually a Linux/Android guy.

Why I like Apple better under cook:

Less lawsuits. They're slowly settling/arbitrating old ones and filing less new ones. I developed a deep hatred of Apple under Jobs due to his temper-tantrums and deep ingrained need to shit in everyone else's punch bowls.

I'm seeing less new intenentional handicaps of their own products, and some of the old ones are getting less rigid (iOS is becoming slightly less user-hostile).

They've finally declared hardware the source of their profits and allowed free upgrades to the OS. (I refuse to use the nomenclature of "Free Operating System" that's been used here on Slashdot too damned many time to describe Mavericks since it's still tied to a mandatory purchase to run it)

What Apple still needs to work on:

Drop all user hostility - make so people can release source code for iOS apps they write. Stop attempting to strong-arm exclusivity out of the iOS platform.

ADOPT FRIGGIN NORMAL CABLES FOR YOUR IOS DEVICES
USB-C connectors are on their way, go with those. All the advantages of your Lightning cables but not "just ours".

Give me an editable path bar I can enable (it can be off by default) like every other OS. As a tech moving around yoru file system is more of a pain than it's worth. Don't spout anything at me about using muCommander or something, I'm a tech, I support other peoples stuff and I don't want to install crap or run utilities that have to be imported somehow every time I sit at a different system.

Drop the artificial restrictions on OS updates "when it was manufactured" isn't a good yard-stick for install eligibility and everyone knows it. Those Mac Pros that are six months too old to run Mavericks are more than capable of doing so and everyone knows it, it just makes you look like a bunch of pricks by barring install.

So we can thank you for the success of the Linux desktop then? Fine.

Do you think that your carefully crafted arguments convinced him, or anybody? (Outside of 4chan)

Apparently anybody that disagrees with you is a fool, liar, and a paid stooge of ${ORG}.

Linux would be better off if it had fewer poison pen advocates like you.

It's public record who passed the law. The Democrats own that law, lock, stock, and barrel, for better or worse.