Communications applications have to be extremely hard to pull off successfully, and God bless Ozzie for trying again.
I'm tempted to spout off on how one would do this thing "right", but I think that would make me the astronaut pot calling the architect kettle black.
“Nobody cared then and nobody cares now, because synchronizing files is just not a killer application. I'm sorry. It seems like it should be. But it's not.”
People were using file syncing for asynchronous collaborative editing. File syncing is not such a big deal now that we have synchronous collaborative editors.
With less double negatives: File syncing has actually turned out to be a pretty big thing, it just took (and this might be a bad list, but off the top of my head) 1. mobile adoption, 2. higher avg bandwidth and 3. someone getting it right. GDrive/Evernote/… does not negate that there is a demand for Dropbox.
Joel assumes the reason why synchronization apps didn't succeed was because there was no market, whereas the success of Dropbox proved the market was there alright, it just waited for the right app at the right time.
I've needed file synchronization for my whole professional life, which started in 1991. At first there was no Internet; then there was Internet but you couldn't connect to it at work (imagine that!) and the rate at home was so slow it wouldn't have made any sense to try to exchange files over the network (even when most Excel files weighted 20k).
I used floppies, and Zip discs (remember those?) and removable hard drives, with bizarre synchronization and copying applications that never worked properly and with which you could very easily erase your own files if used in the wrong way.
The day I found Dropbox was like the Second Coming. It was the day I stopped worrying.
I very much disagree with Joel about architecture astronauts; contrary to what we hear sometimes, being early is not the same as being wrong.
> The people? They love twitter. And flickr and delicious and picasa and tripit and ebay and a million other fun things
I would trade a million Twitters for one Dropbox, and I bet I'm not alone. Dropbox we need; Twitter? not so much.
If I had the resources this is something I'd try to fix. It's 2014 for crying out loud. Live Mesh got me excited at the time, but it just never went as far as it needed to.
Play a movie on your laptop, resume on your device when you leave (given there is an app for it)
Heck, I'm up and working at 4am today because of an AA and his stupid "elegant" data design that is "set in stone" that requires me to provide reams of documentation. I have to make up for the fact that my data is wholly incomprehensible without all kinds of independent documentation. This is all work the AA should have done himself, but instead all he shares are these idyllic, sophomoric examples that NEVER appear in the real world.
It's rich that the system is thought to be "elegant" because one major aspect is simple (there are only something like 8 fundamental data types in the file format to describe the whole system). Elegant? No, perhaps "convoluted" describes it better, maybe "rambling". Data files tend to be veeeery tall and veeeery skinny. Sigh.
So I'm working up to a huge real-world demo and I'm looking forward to canvasing the team in order to create a crisis. "Looking at this realistic example and seeing how much work it took me to get here, what are your impressions of the file format?"