We're aiming to get out a beta build that supports file syncing by the end of the month. Initially, it will require data storage with WebDAV support (provided by a university, MobileMe, a web host, etc.).
Gmail does not provide WebDAV access to storage. There are programs that let you use your allocated space as a network share, but I don't know of any that make it WebDAV-accessible.
Online storage is a commodity. At the moment you could use Amazon S3 (which us used by services like Smugmug and many others). Similar services will probably be available from Google and Microsoft in the near future as these companies are building massive data centers to support cloud computing. Microsoft, for example, is currently adding 10,000 servers a month to its infrastructure and plans to ramp up to 20,000 a month.
For more on S3 see:
http://www.amazon.com/s3/
http://jeremy.zawodny.com/blog/archives/007641.html
http://en.wikipedia.org/wiki/Amazon_S3
Yes, we'll likely be adding S3 support as well. Initially, however, one goal behind supporting WebDAV is to allow people affiliated with institutions to use fast, local-network-based storage.
S3 is also not WebDAV-based. Since the "supported OSes" allow mounting WebDAV locally, perhaps the "PDF sync tool" should also support "local" storage (which could be WebDAV, an external drive, a mapped network drive, or something mapped to S3/gmail/etc.)?
No, but it's REST-based, so adding support should be pretty straightforward once we have WebDAV support.
Since the "supported OSes" allow mounting WebDAV locally, perhaps the "PDF sync tool" should also support "local" storage
Yeah, this was suggested on another thread as well, and it seems like a good idea. (Using a USB drive as a sneakernet between two locations—while still keeping full copies on each machine—was suggested as a possible use.) Native support for WebDAV and S3 has some interesting potential future benefits, though, in terms of authentication (particularly with S3's auto-expiring URLs).
I'm glad there will be S3 support. S3 appears to provide cheap, scalable, and flexible storage on a well-tested system.
Jungledisk (http://www.jungledisk.com) appears to provide a webdav interface to Amazon S3. Don't know whether a Zotero with webdav support could use Jungledisk to access S3.
Note that there are some interesting comments about cross-platform webdav support on their blog:
http://blog.jungledisk.com/2007/12/19/linux-updates-coming-soon/
http://blog.jungledisk.com/2007/08/21/the-road-ahead/
Greetings! I am attempting to sync to a WEB dav at a sakai server provided by my institution. I have no problem synchronizing the records. However, I have a problem with synchronizing the attached PDF documents. Zotero creates files on my WEB dav server. When I connect with another computer, I get only the records but when I click on ths PDFs I get a message: "The attached file could not be found. It may have been moved outside of Zotero".
Has anyone seen this? Can you sugest a solution? Thanks in andvance!
When I connect with another computer, I get only the records
Is "Enable file syncing" enabled in the Sync pane of the Zotero prefs on the second machine, and have you entered the necessary information and clicked Verify Server?
Dan, This might be a beginner's question but after I launch Firefox in the debug mode how do I send you the output. I tried highlighting it and copying but that did not work. Any suggestions?
I just sent the debug output for error -15. Thank you so much for your help. I really love zotero. I already have 2 of my grad students using it - for now without synchronization. If we can get this to work this will be awesome! Thank you again for your help.
I downloaded and installed the latest trunk - I understand that it might contian bugs. I no longer get the "-15" error. However, now I am getting a message saying tha the Dav .href does not match path in Zotero .sync.Storage.....
I installed the new trunk. Still no success.... Is this something with my web dav server? Now I am getting an error something like invalid response from the storage server...... I just sent the debug file to you.
Yes, it is. It appears that Sakai wrote their own Java-based WebDAV server, and from my review of the code, it doesn't appear to support setting properties on files, which—until a few minutes ago—was required by Zotero.
We've now removed the PROPPATCH dependency in the latest trunk build, which should allow Zotero to work with even incomplete WebDAV implementations. Please give it a try and let us know if it works for you.
A note for existing storage sync users: The latest trunk will reset the storage sync history, causing Zotero to attempt to re-upload all storage files in all profiles. If you have a library with many files, you may want to (back up and) remove the 'storage' directory in all but one profile and let it re-download the files from the storage server after the first profile has re-uploaded them all.
I think we are almost there. Now I am cleaning up my database. Now when I do a purge Deleted storage files or purge orphaned files I get an error: Unexpected href '/dav/%7E42058/zotero/' in Zotero.Sync.Storage.purgeOrphanedStorageFiles()
See the end of the attached debug report i just sent.
ttiborko: Both of these issues should now be fixed in the latest build. You'll need to select "Reset Storage History" from the Actions menu and sync to fix the mod times.
Upgrade Storage
For more on S3 see:
http://www.amazon.com/s3/
http://jeremy.zawodny.com/blog/archives/007641.html
http://en.wikipedia.org/wiki/Amazon_S3
Jungledisk (http://www.jungledisk.com) appears to provide a webdav interface to Amazon S3. Don't know whether a Zotero with webdav support could use Jungledisk to access S3.
Note that there are some interesting comments about cross-platform webdav support on their blog:
http://blog.jungledisk.com/2007/12/19/linux-updates-coming-soon/
http://blog.jungledisk.com/2007/08/21/the-road-ahead/
Has anyone seen this? Can you sugest a solution? Thanks in andvance!
I re-clicked it now. Now I am getting error -15.
Actually, it seems like I am getting the same error with both the installs. However, i see that files did get created on the Webdav server.
Do you think it is something with the web dav server?
I just sent the debug output for error -15. Thank you so much for your help. I really love zotero. I already have 2 of my grad students using it - for now without synchronization. If we can get this to work this will be awesome! Thank you again for your help.
I wil be sending a debug shortly.
Tibor
Tools->Add-ons->Find Updates
I installed the new trunk. Still no success.... Is this something with my web dav server? Now I am getting an error something like invalid response from the storage server...... I just sent the debug file to you.
We've now removed the PROPPATCH dependency in the latest trunk build, which should allow Zotero to work with even incomplete WebDAV implementations. Please give it a try and let us know if it works for you.
A note for existing storage sync users: The latest trunk will reset the storage sync history, causing Zotero to attempt to re-upload all storage files in all profiles. If you have a library with many files, you may want to (back up and) remove the 'storage' directory in all but one profile and let it re-download the files from the storage server after the first profile has re-uploaded them all.
I think we are almost there. Now I am cleaning up my database. Now when I do a purge Deleted storage files or purge orphaned files I get an error: Unexpected href '/dav/%7E42058/zotero/' in Zotero.Sync.Storage.purgeOrphanedStorageFiles()
See the end of the attached debug report i just sent.
Now synch works like a charm. Thank you!