Could We Have Real Collaboration, Please?

Melo dropped me a link to Zimbra, an Open Source "collaboration" suite that weighs in at 150MB of code.

Zimbra has a lot of the "right" keywords: it's Open Source, it uses tags, its front-end seems to employ more Ajax features than Botox injections in the average movie star, etc. It can even run on Linux (and, in fact, it has Fedora packages for download).

And, to give it that zesty "sign of the times" look, it even uses brushed metal. All it needs is a GTD wizard and documentation on Ruby bindings to truly be a 2005-style web application.

But while watching the Flash demo, I started wondering where all the collaboration was.

Don't get me wrong, it looks like a kick-ass Ajax mail client, but tagging and creating search folders is something I can already do in plenty desktop clients.

Conversation view is neat, too, but Gmail's looks a bit better. The auto-linking tricks were a bit more interesting, but Microsoft has already done those too (they call them "smart tags", if I'm not mistaken).

I haven't seen their calendaring yet (more on why later), but their search function has nothing on Spotlight or MSN Search (both of which I use daily to much the same effect).

In a nutshell, and at first glance, their key selling point is delivering a unified Ajax UI to a lot of pretty common functionality - which is a pretty good selling point in this era of web-based, remotely accessible interfaces, but that falls short of the term "collaboration" as I see it.

Which leads me to the questions I keep asking myself about CSCW:

• What should collaboration vendors be focusing on?
• Why do they keep re-inventing the wheel?
• What is their added value on the bits they did not re-invent?

My answer to the first question is, as you may have read before, a decent way to do asynchronous collaboration in a corporate environment - preferably in a platform-agnostic way.

I have no ready answer for the other two. I can squarely blame the second one on the usual process of conducting "market research" without involving any actual users, and the third depends a lot on what a particular company/initiative is aiming for.

I can make suggestions, however. For instance, my main gripe against such heavyweights as Documentum eRooms (of which I am an everyday, advanced user) is that they're stuck to two 1990's design principles that ought to be taken out and shot:

• That everybody uses IE on Windows (I can't do rich text editing on Firefox - not even in Windows - or Safari).
• That folder trees and tables are the best way to organize information.

So full cross-platform functionality and non-linear classification (either via Wiki-style hyperlinks or tags) are definitely on my shopping list for my next collaboration tool.

And you may ask: "So, that's all fine and dandy, but what do you use?"

Well, disregarding collaboration, you're looking at it. This Wiki covers 90% of my off-work activities (in terms of information, not actual effort). The rest is mostly tracked by e-mail.

At the office, where I have exponentially higher needs for tracking and re-using information, I use Outlook (which, let's face it, is still the best thing out there) and eRooms (not entirely my choice, but quite usable).

None of those (at either home nor office) can cater to my collaboration needs.

Confluence gets a lot of it right, but it can't cater to my personal needs just yet (and I am still wary of Java-based applications, no matter how good and cleverly designed they are). I might have a simplistic view of things, but if it says it can index my e-mail, it should be able to index all 4GB of it (for my work inbox alone).

Exchange calendaring works great, of course - and instant messaging is so damned useful that we would not get anything done without it these days (like many, I find it amusing that NetMeeting is still pretty much useless, and that we can do mostly the same using VNC and IM with much less hassle). But what I'd really like is something like a Wiki based on Groove - if (and it's a very big "if", considering their recent acquisition by Microsoft) it was cross-platform.

There are a lot of other options out there (such as the ever-elusive Chandler, which is still plodding along its roadmap), and even Hula has made noises about adding some collaboration options in the future (something that jwz has warned them against, and with good reason), but, at least for now, Zimbra isn't one of them, at least for me.

I'll be trying out their other demos (which don't work in Safari, only Firefox) and keeping track of their progress, but, at least for now, they fall short of delivering something truly new.