On Friday, May 22, 2015 at 9:45:54 PM UTC+2, Michael Wiktowy wrote:
Looking at the
beta.etherpad.org site, it would be pretty cool if Tiddlywiki could be loosely integrated with
Etherpad software as an optional backend for handling simultaneous tiddler editing.
Nice idea. The iframe viewer is simple to implement. ... but there is no communication other than copy / paste.
With etherpad, the backend is not trivial and public free services are not very reliable atm. If you really want simultaneous editing the sync mechanism is relatively complex.
If the editing of a tiddler could be optionally switched from editing "Draft of Tiddler Name" to "
https://custom_etherpad_site.org/p/Tiddler_Name" and the result returned as a export option in etherpad,
that would take care of a lot of issues.
Yes. And creates a lot of new ones. Access rights, Group management, User management, .... TW has no UI for this atm, but etherpad needs it that way.
see the doc. I'm not sure if etherpad has a clear API for doing all of this.
There is a relatively simple API, that lets you read and write whole tiddlers. Group, User, access rights management is there too. But as I looked the last time, imo the
authentication system doesn't work well with a web app like TW.
eg:
Also have a look at this request example:
http://pad.domain/api/1/createAuthorIfNotExistsFor?apikey=secret&name=Michael&authorMapper=7
which can be found here: http://etherpad.org/doc/v1.5.6/#index_http_api
see: http:// and ?appkey=secret .... nobody should send plain text keys over http:// !
They may have improved the possibilities, so more reading would be needed. May be someone else may jump in here.
As I wrote, the "simple read the whole document API" is far away from simultaneous editing.
For simultaneous editing, you basically need to send a packet every view key strokes. This API is much more complex, since you can have more than 2 editors at the same time.
I see it has been discussed
in the past but I think in the context of TWC. Just a thought to throw out there as the new "tiddler in an independent window" stuff coming down the pike might lend itself easily to something like that.
IMO if we just want simultaneous editing the new native browser based WebRTC mechanism would be preferable. It creates direct connections without the need of a server. Except for establishing the connection.
Latest browsers have online chat built in:
https://support.mozilla.org/en-US/kb/firefox-hello-video-and-voice-conversations-onlinealso see:
http://www.webrtc.org/web-apis/firefoxjust my 2 cents
mario