Fuck the Cloud
ascii.textfiles.com
ascii.textfiles.com
I must say, I'm impressed; the comments on here don't really descend into the ad hominem, design crazy, check-out-how-hardcore-geek-I-am mess it did over at reddit.
By link bait, I assume you mean for attention instead of money, since I don't do adwords/banners or stuff like that. In that definition, guilty as charged. A lot of it is to just knock people out of the safety zone of thinking of services they don't pay for and just "use", as being their big ol' buddy. And to not fall for "cloud computing" things like software rental, all-actions-on-web and so on.
As for length, well, some people will always want what a weblog posting to say to be three lines, but they're people who don't read weblogs all that much either.
When I got to the paragraph that starts with "There was a time when we gave the Cloud (before it was a Cloud) a big pass" I thought it might start to get interesting. Instead I found this convoluted moon-laser analogy. That's when I finally realized the article was going nowhere and considered the whole thing a waste of time.
I'm not sure who your intended audience is but it clearly is not me.
I think solving this issue could/should/will be done by whatever ends up really nailing item 18 of Paul Graham's "Startup Ideas We'd Like to Fund" (http://ycombinator.com/ideas.html):
"18. The WebOS. It probably won't be a literal translation of a client OS shifted to servers. But as applications migrate to servers, it seems possible there will be something that plays a central role like an OS does. We've already funded several startups that could be candidates. But this is a big prize, and there will probably be multiple winners."
I'm imagining something (begin hand-waving) like a VM image hosting service that lets you take your VM from hosting site to hosting site; which knows about your friends' VM images; and which is to VM images as ZumoDrive is to FUSE fileystems.
This article also completely glosses over the fact that most people don't keep their data backed up, or in a secure place (even people that know better).
It doesn't even consider that fact that... if you could accept that not all people offering cloud services are incompetent/misguided/evil... Cloud services have the potential to help people (!).
Consumers don't backup because it's messy. The could takes away the messiness: backups (and all the other sysadmin stuff) happen automatically. It is an improvement over the current situation.
There is a similar case with businesses: businesses do backup because they have to, but it is still messy. Businesses like to outsource messy stuff that is not core to what they do to others, and these others are in the cloud again because it is cost effective. So you may use Google Apps and back up your mail on Amazon S3 if you are that paranoid about losing your data.
I read halfway before I understood what he meant by "cloud." It's now officially an overloaded term, in danger of becoming meaningless like "web 2.0."
By cloud, he means all those services where you put your personal data, thoughts, and ideas -- potential black holes in the sky from which your lifework may never return. Flickr, LiveJournal, Twitter, Delicious, etc. fit in this bucket, and I couldn't agree more with his assessment that blindly using these services is stoopid.
Lately, on the other hand, I've been using the term "cloud" to refer to services that allow developers to dynamically allocate computing resources. I suppose this is better referred to as "utility computing" -- y'know, the stuff that Amazon (AWS), Google (AppEngine), Microsoft (Azure), and Rackspace (Mosso) are big into these days. That kind of cloud? Awesome.
It opens up a useful business model for companies to get into a person's "cloud" and back things up for them, either into another cloud or into some hard model (DVD).
But then again, nobody thinks about backups until they really need one.
I also think using the 'cloud' can actually make our data much safer. I mean my home PC is much more vulnerable than a monster data mainframe in googles basement. As long as your password is sufficiently complex, google Docs is more secure for a 'secret' document than my home PC is.
2. Wasn't Google running on commodity (x86 comp.) GNU/Linux clustered machines?
Guess it would be too expensive atm ;-)
*I know it was a joke etc, just making some point(s)
Also, hard drive recovery services already exist and cost serious money. I don't think it would necessarily be evil to charge quite a bit for backing up internet content. After all, the costs for doing the backing up are probably significant.
Looks like somebody's bought it now though, and hopefully have designs on getting it going again. I meant to follow the auction more closely, but does anybody know what it actually sold for?
On the other hand I will not dump a hard fetch technology on a big company cloud or an important piece of my application on some cloud no matter how soft. Given that there’s no real proprietary/ restitution agreement between you and cloud owners it doesn’t seem Wright. The last point I want to make is the Mainframe to PC comparison which appear to be shifting the other way around; It was proven unfit for evolution and probably will be again.
One advice to startup, store only space hungry data on the cloud (huge non tech files, images, music etc ..) we’ll see what the future brings next.
It's better to have them in both places than in just either, privacy concerns aside. But if I were my mom, I'd be way better off having them in at least some pieces of the cloud (the ones likely to survive, and to give you your stuff if they are about to not survive?) than on my machine.
There's no real reason to use the cloud for data backup anyway, since uploading large files over DSL is a pain, and hard drives are holding steady (not!) at $100/TB.