To view parent comment, click here.
To read all comments associated with this story, please click here.
loss of internet is mostly worked around by html5 local storage - which will take another 2 years or so for widespread use.
But I actually expect chromeos, firefox, etc to push this forward.
For backup, downloading a zip with all data is probably the way to go.
HTML5 local storage is insecure[1] and ZIP files aren't as easily searched as a client with built in search tools against indexed content.
Of course, there will always be ways you can work around the limitations of the web; Javascript runtimes have come a long way and many web apps these days are pretty sophisticated. Plus I'm not ignorant to some of the unique benefits they offer (eg easy to roll out bug fixes). But native apps have their own benefits as well; they're generally more responsive, can be better secured for offline content and they integrate with the host OS better (this is even the case on many platforms that are internet orientated devices; eg smartphone and tablet OSs).
So while I do appreciate that you're trying to demonstrate that you can replicate some functionality in HTML5, it's really not at a stage where it's ready to replace native binaries entirely.
[1] http://www.youtube.com/watch?v=RbOaRle4Pw8
Edited 2013-01-29 10:01 UTC
One caveat though, local storage has a size limits of a sort.
http://diveintohtml5.info/storage.html#limitations





Member since:
2007-03-26
It helps you getting your data from everywhere, and to everyone _you_ want to give it.
True. I do run some services myself (eg Subsonic, my own hosting photo gallery, etc).
Even then though, I still dislike OSs that push processing away from native binary clients. eg webmail is great - possibly the best example of the 'cloud' in fact - but I still want a binary client that I can run locally. If just in case of emergencies (loss of internet, backing up stuff from the cloud, etc)
Edited 2013-01-29 09:32 UTC