[10:15] when does desktopcouch sync with ubuntu one? Is it done periodically and is there a way for a client to ask for a sync of a certain database? [12:56] Chipaca, ping... U1 seems to have eaten all my files [13:12] Is there anyway to recover files out of U1 all of mine are missing ... HELP! [13:25] pgraner, perhaps you deleted them from your client? Deletes are synced, so if you delete a synced file from your PC, then it is also deleted from the web and all the other computers that syncs that folder. [13:26] it would be surprising if Canonical kept backups of all the files all users deletes, but you never know. :) [13:26] jo-erlend, they have a way to recover them as I had this happen before [13:27] jo-erlend, I made the mistake of rsyncing my home dir from one box to another and in the it confused U1 and it thought I wanted all my files deleted [13:28] ah, you were logged in on the other computer at the time? [17:17] pgraner: do you still need help recovering your files? [17:46] How close are we to getting the Notes, Contacts and Bookmarks service running again? [18:23] duanedesign, yes please! [20:24] is there any way to share a database with someone? [20:25] I mean like you can with files, by entering an email address and the other user gets an invitation. Is there anything like that for couch? === m_conley_away is now known as m_conley [22:17] jo-erlend: I don't think so, but I haven't done a lot of reading on U1 recently ;) [22:18] jo-erlend: maybe you need to report a feature request ☺ [22:19] actually, I think I'd prefer it if there was a break in adding new features until stuff actually works! :) [22:19] but I think that would be a cool feature. [22:20] feature requests don't mean "implement this now" ;) [22:21] that's true. :) === m_conley is now known as m_conley_away [22:23] to be honest, I still have to see any advantage of using couchdb/desktopcouch except for "sometimes automatic syncing (when it works)" :P [22:25] well... If many applications require the same set of data, then why shouldn't they be able to reuse software? Let's say you sometimes use Firefox, sometimes Chrome and sometimes Opera. Perhaps on different devices. You want to organize your bookmarks. Why should you have to use many different tools for that? Wouldn't it be better if you could just install bookmark managers that would automatically work with all web browsers? [22:26] so sync between computers is not the whole point, although it is a really big advantage. But it makes it very easy to decouple systems and reuse components, which I think is a nice thing. [22:26] jo-erlend: I agree with that idea, but less with the implementation ;) [22:27] oh? [22:30] I think desktopcouch and couchdb in general is one of the coolest technologies I've ever played with. [22:33] I'd prefer a more distributed approach [22:34] hehehe.... Are you serious? One of CouchDBs biggest strength is exactly that it's very distributable. [22:34] and somehow I've seen too many performance issues with using it (although maybe those were because of inexperience with couchdb) [22:35] jo-erlend: what I mean by that is that I can sync e.g. bookmarks between machines on my LAN, without having to go through U1 [22:35] JanC, of course you can. [22:36] which is actually possible, but difficult to find for most people [22:37] hmm. Well, yes, it could be easier. But that's on the roadmap I think. And I don't think that should be a priority. [22:38] I guess I consider sync to my own systems much more important than sync to Canonical's servers ;) [22:39] but a link to file:///home/user/.local/share/desktop-couch/couchdb.html would fix that. I think it's fairly easy to replicate your desktopcouch database compared to most other databases. [22:39] uhm. That was not the right location... (Why doesn't xchat reuse my Firefox bookmarks?! :)) [22:39] hehe [22:40] oh.. It is the right URL, of course, except for the username. :) [22:40] how would you make replication any easier than that? [22:42] couchdb can discover other couchdbs inside a LAN using avahi IIRC [22:42] oh, ok. Yes, now I see your point. [22:43] I think a (secure) GUI way to connect local couchdbs would be cool ;) [22:43] and maybe add remote couchdbs to the "trusted" network too [22:44] that's simply a matter of adding users though? [22:44] which is what U1 currently does for theirs ervice [22:44] it's a matter of adding UI ;) [22:45] at least, AFAIK [22:45] either that, or couchdb lacks the features I want... [22:46] but.. What more is needed than the replicator page? [22:46] the "replicator page" is not user friendly, right? [22:46] how? [22:47] last time I looked at it (and it wasn't exactly easy to find) it was fairly technical? [22:47] choose the database you want to sync from the combobox, check "Remote database" and enter the address. Click replicate. [22:47] see, that's not user-friendly ;) [22:47] oh? [22:48] it's about the same process as navigating to a website? :) [22:49] when people "navigate to a website", they either click a link or enter a search term into a search engine? [22:51] well. It can always be made even easier. But if you compare it with what you would have to do to achieve the same effect with most other dbms... It's not perfect, but it's still very cool. :) [22:52] jo-erlend: I agree with that idea, but less with the implementation ;) [22:55] :) [22:56] sounds like a fairly minor UI issue though. :) [22:59] heh... As long as it works, of course. That has to be a goal. [22:59] jo-erlend: from a developer PoV, maybe, from *my user* PoV, it's important ;-) [23:00] well... I agree that it's comfortable to be able to phone people by tapping their names, but in a crisis, I'm still able to enter their phone number. Important? Perhaps. Critical? Nope. :) [23:03] so, basically, I (almost) don't use desktopcouch/U1 because it's not critical, and inconvenient/uncomfortable? ;) [23:04] I'm not a u1 developer. Not yet, anyway. But it sounds to me like you've set your mind to a solution and anything else is difficult. :) [23:05] not a solution, but a goal [23:07] which is that there should be no central database in your network and that your data should just float between the computers by itself? I agree. That will be nice. [23:09] but I do think the majority of users would still like to have their files available online and probably don't want to get their own server in order to achieve that. So the focus should be on getting u1 to work properly and then to expand it in other directions, like you propose. [23:10] the local sync does not exclude involving an external service [23:11] no it doesn't, and as I understand it, it's on the roadmap. [23:13] * JanC is waiting for seamless operation ;) [23:14] but for now I'm going off-line... [23:14] :) [23:14] slaapwel! [23:14] likewise. :)