Slashdot Mirror
LiveDrive vs GDrive vs Personal Data Storage?
ozmanjusri asks: "At a blogger's breakfast prior to the opening of Tech.Ed in Sydney, Microsoft Australia technical specialist John Hodgson has confirmed that Microsoft will introduce its LiveDrive online storage system which can be mapped directly as a Vista drive. The service will offer 2GB of space free, with additional capacity available at a cost. Earlier this year, rumors surfaced regarding a similar scheme from Google, the GDrive. There are already hacks to do this with GMail, but Google's goal with GDrive appears to be infinite storage, accessible from anywhere. Meanwhile, the price of portable USB flash drives has been falling to the point where 2GB drives are cheap enough for every day storage purposes. Is this the start of a new era of (nearly) free online storage, thin clients and OS independent services? Will data storage which is tightly integrated to the OS be more attractive to the average user, or will we prefer to have our information stored on a physical media we can put in our pockets?"
It's not going to work for me for a number of reasons:
1. I'm in Australia, and bandwidth is expensive in Australia. Cable ISPs offer plans like 10gb per month, and some DSL ISPs offer up to 60-70gb per month. Some are upstream + downstream added together. It's not much when you're considering storing your stuff on the net.
2. I'm on cable, and the upstream bandwidth is terrible. 64k if I'm lucky. I really don't want to wait hours to store my files on somebody elses server.
3. I'm sure plenty of people will make statements like "What about the privacy!? I don't want google looking at pictures of my kids!". I don't really care, but it's certainly an issue.
Anything is possible, except skiing through revolving doors.
echo YOUR_OPINION >
The choice of storage for an individual depends on the nature of the data, the amount of data, the available bandwidth, the availability of a connection at all, what they are most comfortable with and what their idea of convenience is.
For example, you might not trust Microsoft or Google with your data even if it is encrypted. If you are in a competing business, you wouldn't want to store your business data on their servers. Alternatively, you may not trust them to provide you with the level of availability you desire. It doesn't help you if you can't access your data when you want it.
If you have a few hundred GB of data, you aren't going to want online storage. To access your data is going to take too much time. Even with decent bandwidth, anything more than a couple GB is going to give some serious delay. If you want to access the data at your grandparent's house and they use dial-up, online isn't an option.
Finally, if I am not comfortable with the online option, or I'm not comfortable with keeping my data in a single physical location, I'm not going to choose those options. Personally, I like having it on physical media that I can carry around. I like the bandwidth I get from a USB device and I don't have to worry about getting an online volume properly mounted.
On a side note, I don't trust the idea of "free" or even "cheap" online storage. The money for the hardware, bandwidth and administration have to come from somewhere. If I'm not paying for it directly, where is the money coming from? Either the company is getting some benefit from it - such as Google analyzing it for keywords to target advertising, or they are selling some sort of information about my data, or they are making it up in indirect costs (add $25 to the price of Vista). I would rather pay the direct costs so I know how much it is costing me; but that may be personal preference.
Reading code is like reading the dictionary - you have to read half of it before you can go back and understand it.
Wait, multiple companies tossing out storage like it's nothing, and the base price is a broadband connection?!?
Yet flash drives are comming down in price?
I mean, why would you need to waste bandwidth (which as noted can be expensive in some civilized nations) to pull in and work on files, when you can plug in a USB thumb drive with all your files in, or that spare 20 gig 2.5" drive, or a portable 3.5" drive, and work off of that? There are already distros out there that will boot off of USB drives. Why bother getting online when you don't need to be? What if you can't get online (no Wifi hotspot in the Nevada desert, and you forgot your EVDO card, plus Iridium is too expensive)?
Forget the Internet. Let's build up the Sneakernet.
--
# Canmephians for a better Linux Kernel
$Stalag99{"URL"}="http://stalag99.net";
The best thing I have found that exists in a ready-to-be-used state (as opposed to the countless hordes of maybe-someday projects) is vtfileman. Its available at http://vtfileman.sf.net/ and there are at least two instances of it running at universities (http://filebox.vt.edu/ is the original and http://filer.case.edu/ is the one that I run). I have some implementation notes on installing Filer at http://filer.case.edu/wiki/filer/notes
It does have a lot of other requirements though, such as an LDAP server for accounts, Apache to serve the HTTP and WebDAV pages, Apache Tomcat for the JSP interface and proftpd if you want FTP access. However, it is pretty sweet once its running.
If thats too complicated, you may be better off just making WebDAV shares individually for different groups. Personally I like that with vtfileman, people can set up their own accounts with little to know interaction with the system administrator.
Absolutely. As long as your files are in a TrueCrypt volume.
http://www.truecrypt.org/
Slashdot anagrams to "Sad Sloth"
My ideal setup would be one in which my local device is just a write-thru cache for my network storage. The "network computer" notion of fetching your applications on a JIT basis is attractive enough that it will one day succeed, but I'm talking about my personal data, not apps.
Back up should be a server responsibility, not a client responsibility. The client should be responsible for passing data/documents through to the *real* storage location ASAP (ideally, as the data is entered into the client). This wouldn't be considered backup any more than saving from RAM to a disk file now is considered backup. Saving to the server should just be "saving". And pros keep the server backed up, of course.
Since before long all of us will have multiple networked clients capable of serious work (our old laptop, our new laptop, our phone, etc.) and we'll want to be able to move transparently from device to device and keep working, and not lose data when we lose hardware, having our "one place" for data be a server somewhere, with the clients functioning as local caches, seems the natural way to go.
Whoever gets the usability right ought to have a huge hit on their hands. Will it be Microsoft, with their control over such a high percentage of "serious" client OSes? It would make sense to build this in as a transparent feature of every PC/device OS from MS, increasing the attractiveness of MS OSes on devices if that's what you use on your PC. Or will it be Google, with their openness to all clients, regardless of vendor?
Or will they miss the local cache idea altogether and just create an offsite network drive?
"Those who have never entered upon scientific pursuits know not a tithe of the poetry by which they are surrounded."
I'm glad Microsoft is finally introducing a product in this space, because they're the company I look to for reliable and secure software.
TrueCrypt supports two levels of plausible deniability to combat this though, and it's something that set it apart from other utilities like this.
TC can put a hidden "inner" volume inside your encrypted volume, that is simply mountable with a different password. But there's no way to prove that such a volume exist (TC volumes are indistinguishable from random data, and even file system and unused data is encrypted) and that you have more passwords than one for the "outer" regular encrypted volume.
Beware: In C++, your friends can see your privates!