Hello all.

I've long been in the search for the Holly Grail of opensource Groupware, quite unsuccessfully.

To give you an idea of how long, I was there eagerly watching when Postpath (an opensource drop-in replacement for Exchange Server) was bought by Cisco (year 2008), and later killed by them. I was also thrilled by the news of Novell's Hula project being opensourced (year 2005); but then it was abandoned. So yes, I've been watching the opensource groupware scene since way back in time -- and I can say there is no opensource project yet that can match Microsoft Exchange, not even by a long shot (disclaimer: although I am a fan of opensource who discovered Linux with Slackware 3.0 running Linux kernel version 1.2 and has loved it since, one of the many hats I carry at the workplace is that of senior Exchange consultant for customers with Exchange problems --i.e., I do know well Microsoft Exchange inside out).

Yes, I know Zimbra, which can be argued is the closest thing in functionality to Microsoft Exchange, and yet I do not consider them opensource --because you need a proprietary connector for it to work fully with Outlook, and without Outlook you cannot use Zimbra to its full potential, which is sad. Also I am aware of Zentyal.

The problem with projects like Zimbra and Zentyal is that they attack the "opensource groupware problem" from the point of view of emulating Exchange to the point of aiming for protocol compatibility with Exchange/Outlook at the protocol (MAPI) level. This is both the appeal for those projects, and ultimately the reason for their demise: you cannot aim to emulate a proprietary protocol of a monopolistic closed source company, constantly playing catch-up with them (now RCP-over-HTTP, then EWS for free/busy discovery and OOF, later MAPI-over-HTTP, etc.), and get anything but grief and pain in return. This fact of life has long being theorized in the computer industry, please see the "Fire And Motion" blog post by Joel Spolsky (http://www.joelonsoftware.com/articl...000000339.html).

What I feel the opensource world needs in the realm of groupware is an integrated solution which can stand against Microsoft Exchange in features (a subset of Exchange's features would be "good enough", really), but discard the crazy target of being Outlook-compatible. Because Outlook is a moving target, and you will be forever playing catch-up with Microsoft if that course of action is to be followed.

I could move customers out of Microsoft Exchange if I had an open source solution which stored Mail, Calendars and Contacts server-side, where those three categories of data were accessible with all of (1) a desktop fat client, (2) a mobile device like a phone/tabled, (3) and through a classic web interface like OWA.

Lets analyse points 1, 2 and 3 above:

1. I would not need for that solution to be Outlook-compatible, if Thunderbird could store Contacts and Calendar in such a cloud-like solution; but currently Thunderbird just cannot do it. Thunderbird can do Calendars with and add-on, but it cannot do server-stored Contacts. So there --end of game.

2. The mobile client I think would be easiest problem to overcome, at least in Android. Android can do IMAP email just fine, and I should not be hard to have an app to sync Android's built-in Calendar to a CalDAV server, and to sync Android's built-in Contacts to a CardDAV server.

3. Opensource webmail front-ends like RoundCube and Squirrelmail have the same problem as Thunderbird: they can do IMAP just fine, but then they store their Contacts and Calendars in a locally-private way which cannot be shared with your desktop fat client or with your PIM-related mobile apps.

I've put together this diagram with my thoughts about this. http://s18.postimg.org/hedig00jd/Group_Ware_mindmap.png

Feel free to flame me!

Without trying to match or understand your exact wishes or requirements, perhaps Open-Xchange meets a lot of a lot of your criteria already by way of it's community edition? Citadel is another open source solution worth looking at.

The requirements which would fit the bill for "good enough" are spelled in this diagram (note that those functionalities marked there explicitly as "optional" are, obviously, not required): http://s18.postimg.org/hedig00jd/Group_Ware_mindmap.png

Even if Open-Xchange and/or Citadel where up to the task regarding the mobile access and the web access side of things, we are still missing a fat desktop client which can do Contacts against a CardDAV server (Thunderbird cannot, currently).

It is the Mozilla Foundation which in my opinion has dropped the ball big time failing to take Thunderbird to the next level (it can store Calendars server side with the Lighting add-on, but there is no add-on for CardDAV support in Thunderbird). See this link with Thunderbird users clamoring for CardDAV support, for years: https://bugzilla.mozilla.org/show_bug.cgi?id=546932

Yes, there are hacks and a non-official add-on which can bring CardDAV support to Thunderbird, but it only works with very specific old releases of Thunderbird and therefore it is not dependable:
https://help.hover.com/hc/en-us/arti...or-Thunderbird

So in my opinion the opensource groupware problem is not solved (for enterprise use), after all this years (unless you want to remove the "fat desktop client" out of the equation, in which case it can be argued that solutions like the Open-Xchange and Citadel could fit the bill). I still cannot recommend to my clients an opensource groupware solution to move them away from Exchange/Outlook.

Ah, got it. Thanks for clarifying further. I agree, a more complete solution with email client integration would be a good thing.

Hi VamosBaby,

Regarding Thunderbird CardDAV support, have you looked at the SOGo Connector lately? I've used it very successfully in the past, with ownCloud as back end. They currently support ThunderBird v38 (current) and v31 (The enterprise/extended support release until recently a.f.a.i.k.).

Alternately, have you considered an LDAP back end for the address book?

Regards,

Clifford

Yes, I know about the SoGo connector for Thunderbird, it's talked about in the Mozilla's Bugzilla link I gave above, and in the other link in my same post, too. The problem is that 1) that Thunderbird add-on is not in the official repository of Thunderbird add-ons, 2) its authors tend to lack behind newer Thunderbird releases, 3) CardDAV should be built-in into Thunderbird or at the very least be an officially supported add-on in lockstep with Thunderbird releases.

Because of those reasons, you just cannot bring the SoGo connector in the enterprise as part of a enterprise grade solution for groupware.

About LDAP, it is useful to build a Global Address List, and that is nice to have but not a strict requirement to put a foot into the enterprise groupware scene (please, see the diagram I posted in my previous message).

Yes, integrated email-client support for CardDAV is what is required for enterprise support. I'm surprised that after all this time that Mozilla hasn't integrated this support, or provided their own officially-supported extension.

Does anyone know about any GUI-based desktop email client for Linux which can do IMAP, CalDAV and CardDAV, out of the box or with easy tweaking?

I don't know of any Linux email client that supports all of those directly. However, I've had good results in the past with using DavMail as an Exchangre gateway. It was trivial to set up and worked seamlessly. I had it running on a couple of laptops along with my preferred email client without issue.

Check out Zimbra.

Yes, Zimbra is currently the closest thing to cover all the bases.

Also, perhaps the new version of SoGo fits the bill: IMAP, CardDav, CalDav, ActiveSync, and a webmail frontend which interfaces with CardDav and CalDav. Also, they make the "SoGo connector add-on" which brings CardDav support to Thunderbird (versión 38 only, at the moment).

More on SoGo frontends here.

SoGo also packages something called the "OpenChange connector" for their backend, which offers native MAPI connectivity for Outlook -- but I really do not care about that, provided their Thunderbird support is good enough.

I think I really need to try SoGo!