The dark side of software as a service
During the 80s, Sun developed a networked filesystem and it became kind of a fad to use dumb-ish Unix workstations with all of your information stored on the network. At the time, I remember someone describing this trend by saying “it’s like using a normal computer, except that every once in a while, a server you’ve never heard of goes down, and you can’t access your files anymore”.
Fast forward to 2007. I happen to have multiple computers at home, so I’ve been using Google Docs to store some of my critical information so that I don’t have to worry about which computer I’m using. And.. guess what? Google Docs is completely hosed right now (at least for me).
It’s a weird feeling to not be able to access my own data until someone else fixes their problem. Maybe I should have thought about that before putting my data in the cloud…
When I was at Berkeley, all of the workstations were set up with NFS and user home directories were on central servers. Their uptime was actually pretty remarkable; I can’t remember an occasion on which I was hosed due to a downed file server. However, the big discovery in graphics class was that if you copied your source tree to /tmp (mapped to the local HD), you could make compiles go a lot faster.
Hopefully your files come back soon. I remember reading that they are due to add Google Gears shortly to Docs, so at least then you’re not halted when the server does go down or is being upgraded.
Hi Sho,
So Try Buzzword and “Share” -;)
-David
Just for the record, the David above is not me. :-) But he’s got the right idea…
Perhaps there is another edge to this sword.
Aren’t you neglecting the fact that whatever brought down Google (a virus or DOS attack or bad software) could just as easily happen to your desktop? The only difference is that YOU have to do the IT work to fix it, instead of going downstairs and watching Tivo for 20 minutes and coming back to find that somebody else fixed it.
I personally would much rather have Google do my IT work for me.
By the way, when I go to pandora.com, it starts playing music within 7 seconds. When I open iTunes, it takes 15 seconds just to load and do nothing! And that’s after a long painful installation process, and popups appearing every month saying “it’s time to upgrade quicktime AGAIN!”
Long live web apps! Death to the Mac/Win duopoly!
I know it’s not what you want to hear, but I’m not a big fan of the idea of giving my data to Google or anyone else for that matter. Serves you right, Sho. I’m serious that the idea scares me, but teasing you about everything else, of course.
Sorry to hear that the Web v.2 has let you down… and I do hope you get access your docs in a timely manner.
Your commentary on the 80’s pseudo-nix workstation OS and its goofy behavior is something I’m (indirectly) paying the price for now. I do a lot of LISP development for fun (yeah I know that makes me an uber-geek) and the file system API in LISP is absolutely nuts. I did some digging to find out why, in such an elegant language (yes, I said LISP is elegant) something that could be so simple is so twisted – turns out it’s all because of having to support the myriad of quirky file systems that sprung-up from the 1950’s to the 1980’s prior to the rise of the universe where all we care about are Windows and Unix. An interesting bit of LISP history, but nothing much to do with your blog entry.
Well, my data is back. Woohoo!
In response to Harry: I agree to an extent. Having applications online means I don’t have to administer them, and having my data online means that someone else is responsible for backing up my data, etc.
What happened today at Google Docs doesn’t look like a DOS attack to me, based on the error messages I was seeing. To me, it looked like there was a problem with the application itself. Maybe they updated the software on the website, or maybe a database went offline. Who knows?
Problems will happen whether the application is local or remote, but I guess the “out of control” feeling I had was this sudden realization that… criminy! these web-based companies are in the habit of constantly updating their functionality every 2 weeks and using the public at large as beta testers.
Now.. I’m not saying anything specific about Google’s QA procedures. For all I know, they may use more of a desktop software methodology for testing. My point is that with web-based applications, you are not in control of when the application gets updated.
When the software resides on my own machine, I am not going to, say, upgrade to a new beta version of Excel five minutes before printing out my spreadsheet for a meeting.
Simon: I understand the concern about giving all your data to Google. I worry about privacy, as well as longer term “lock-in”. Five years from now, am I going to regret giving my life over to some web-based company?
There is a real lack of clarity around ownership of the data. All the data that you feed into the cloud is yours, in a sense, but is it really? On Ofoto (now called Kodak Easy Share, I think) you can’t get at the original versions of your photos unless you order a CD. What if they take away that feature in the future?
Final thought: what is the equivalent of “back up your data” when you are dealing with a web hosted application?
“these web-based companies are in the habit of constantly updating their functionality every 2 weeks and using the public at large as beta testers”
On the other hand, when I install software out of a box, it’s generally not going to change until I buy the next relea$e, which may or may not be an improvement, and may or may not fix the bug$ of the previous relea$e. On the Web, everything is always in beta, and it always gets better. How could it not, with millions of beta testers working for free?