You have literally no idea what the performance or UX of this service will be like, but you've decided to have a pop at it anyway. That's not a very endearing quality. You'll recall of course the similar short-sighted comments when Dropbox first debuted…
Have a look at some of the technology - there is some interesting stuff in there - before writing it off.
Normally I'd agree, but there's a pretty substantial history of vendors trying to solve these problems in the filesystem context, and it's pretty clear that it's hard problem from many different perspectives. It's very reasonable to the skeptical.
Microsoft gave up in OneDrive (where they used stubbing for awhile), and moved to a model where they, in Windows 10 deliver a filesystem replacement that looks and smells like a traditional filesystem to end users, but is in fact a Dropbox-like mobile client.
I'm sure that when I get to use it (I'm merely a chump paying $100/year for Dropbox) it will be great for 90% of my needs. But there are many edge cases where things will break for users or applications that don't handle high latency file operations well.
If anyone has a chance to pull this off with a UX that people can grok, it will be Dropbox.
Comments
This sounds like NFS for the .js age
Slower, buggier and less reliable (and I mean, yes, compared to NFS)
You have literally no idea what the performance or UX of this service will be like, but you've decided to have a pop at it anyway. That's not a very endearing quality. You'll recall of course the similar short-sighted comments when Dropbox first debuted…
Have a look at some of the technology - there is some interesting stuff in there - before writing it off.
Normally I'd agree, but there's a pretty substantial history of vendors trying to solve these problems in the filesystem context, and it's pretty clear that it's hard problem from many different perspectives. It's very reasonable to the skeptical.
Microsoft gave up in OneDrive (where they used stubbing for awhile), and moved to a model where they, in Windows 10 deliver a filesystem replacement that looks and smells like a traditional filesystem to end users, but is in fact a Dropbox-like mobile client.
I'm sure that when I get to use it (I'm merely a chump paying $100/year for Dropbox) it will be great for 90% of my needs. But there are many edge cases where things will break for users or applications that don't handle high latency file operations well.
If anyone has a chance to pull this off with a UX that people can grok, it will be Dropbox.
UX has to deal with real-life and technological limitations
That's the main problem in this case. And NFS. And CIFS. And in other similar things
NFS never did selective local and offline syncing. If you are going to compare it to anything, compare it to AFS.
what about AFS or Coda?