This is an idea that's been percolating in my head for awhile - if there's a theory why it can't be done I haven't thought of it yet, though I can think of a dozen practical reason it might not work, mostly due to the computing overhead becoming intolerable inside a browser or edits on the file causing versioning issues. I can imagine either simultaneous edits being a problem, or edits being made in isolation and needing to be correlated after the fact.
That said - I've been using BTsync for various cloud backup/syncing applications for about eight months now, and I can't help but wonder if something conceptually similar could be used in TiddlyWiki. For the uninitiated, BTsync uses the bittorrent protocol to advertise/transmit the contents of a folder on the internet with a (IIRC) 40Byte Encrpytion key - anyone with a matching key can have a read write or read-only (in effect) copy of the folder automatically synced. Put one on a free AWS server and you've got your own dropbox account (I put all my ebooks there). It seems like something that *could* work with a single file as opposed to a folder. It would require implementing the same kind of btsync concept in Javascript of course, a non-trivial obstacle I'm going to just glide on by thank you very much. Given that that can be done, I see two obvious problems, both involving versioning. One would be that in order guarantee updates at least one other copy would have to be open at any given moment - which one should be unimportant, as long as there is a 'chain of provenance' at any given moment - when a new tiddly wiki is opened it would update. The other would involve two or more people simultaneously editing a single tiddler; Some variation of 'diff'' logic would seem suitable. It would be . . . awfully neat . . . if the first item could somehow be managed completely within the confines of tiddlywiki, but I concede I'm seeing no strategy for doing that doesn't involve someone magically making sure there is at least one copy open somehow. I can however see it being integrated with BTsync in some way that a copy in a BTsync folder with the same key is somehow set as a master copy, with BTsync keeping it updated when not in use. It's not entirely lacking in elegance to be sure. The second is probably more difficult than it looks, but doesn't seems terribly hard at first glance. The result be a bit of html the would be (within certain limitations) genuinely 'in the cloud' as it were - a website without any pure client server relationship. Does this strike people as something that could actually be done, or is it a neat idea that wouldn't work in practice? -- You received this message because you are subscribed to the Google Groups "TiddlyWiki" group. To unsubscribe from this group and stop receiving emails from it, send an email to [email protected]. To post to this group, send email to [email protected]. Visit this group at http://groups.google.com/group/tiddlywiki. For more options, visit https://groups.google.com/d/optout.
