I’ve been talking a little more with some Wave users and playing with Wave more and I thought I’d elaborate a bit more on Wave. The one thing I think is pretty cool about Wave is, as I’ve said before, the multimedia collaborative edited document. I can certainly see how that would be useful for collaborative planning, etc.; my beef with Wave is with the thought, or lack thereof, that went into the context in which that multimedia collaborative document lives.
First of all, it should be pointed out that a lot of the use cases for Wave are already possible using Google Docs and Spreadsheets; i.e., you can edit in real-time things like to-do lists, plans, etc., at the same time, and even chat about it using a separate chat interface. For example, if you could inject videos, etc., into a Google document, then it would start to have a lot of the same power as Wave.
What could have and should have set Wave apart were the ways in which you could use, structure, share, and link the wave; not only whether it is possible to do these things but how intuitive and easy it is to do these things. You could summarize my complaint about Wave this way: instead of basing it on email and adding wiki, social networking, and IM functionality to that, they should have started with either wikis or social networking/twitter and added functionality that allowed email-like use cases.
For example, sharing a wave with a group of users is apparently possible using a hack via Google Groups — but this isn’t easy or intuitive. As far as I know there’s no way, yet, to create a collection of waves and share them automatically with everyone in the group, so that edits to the waves automatically show up and new waves created in that collection are also automatically shared (i.e., with friends, Twitter-style followers, coworkers, etc.) Linking waves is cumbersome, forcing you to enter the “Wave ID”, which is bound to cause confusion as where one finds the wave ID is not immediately apparent, and furthermore when you’ve linked the wave this doesn’t give the viewer permission to see the linked wave unless they’ve been explicitly added to it, or it is totally public. Features in Wave are not easily discoverable or evident. There’s no way to control the types of edits people can make; for example, one of the best things about blogs is that the main content is only editable by the author, but other users can add comments; this tends to damp down on the flame wars that can sometimes pop up in “flat” threaded discussion forums. There are many other structural problems I’ve already commented on with Wave.
Of course, going back to the strength of Google, Google Wave is open source. This means a lot of these problems can be fixed, in theory. However, looking at the history of open source, since it is very engineer-driven there really haven’t been many if any open source projects with a particularly usable interface. The most usable open source or free software projects tend to simply copy existing successful commercial designs; i.e., KMail is a clone of Eudora, Evolution is a clone of Outlook, KDE and Gnome are similar to Windows, etc. In other words, open sourcing it may allow Wave to eventually patch a lot of the holes I mention above, but I’m not clear on whether or not it will ever make this functionality intuitive and accessible to users. For a lot of these problems to be solved, Google would also have to make massive changes to the basic addressing and sharing metaphor of Wave (i.e., which is based on email), so the new collaborative features would actually be usable everywhere, seamlessly — and I somehow doubt this will occur, at least on Google’s servers, because it might involve rewriting a lot of stuff at the core of Wave… we’ll see. Without that, some small corner of users or some subset of features or people using some customized Wave branch on a small set of servers might be able to get past some of the limitations of Wave’s email-based design, but if it isn’t fluid and universal it will make for a frustrating experience.
Looking at Wave I am inspired, that is I can imagine all sorts of amazing things that could be possible with some future communications tool, but the number of built-in roadblocks to these things in Wave make it seem a daunting task to actually implement it on top of Wave the platform, given that many of these problems are built into the originating metaphor of Wave: email. Rather than taking email and blowing it out into something big, they should have started with people: communication, and worked backwards to engineering that would or could support these new modes of communication.permalink |