From: Caleb Clausen <vikkous@...>
Date: 2010-04-12T12:02:36+09:00
Subject: Re: Google Wave: A new type of "Ruby Quiz" ?

On 12/14/09, Rick DeNatale <rick.denatale@gmail.com> wrote:
> So far, in my experience Google Wave has not worked very well.
>
> Part of the problem is that it's a confusing muddle of email, and
> wiki.  They claim to have (underlying?) version control, but it's not
> obvious other than the ability to 'replay' the history of the wave,
> with no obvious way to recover a previous state.
>
> A wave is really a document which in the wiki fashion, 'anyone can
> edit', but I've found that users don't really understand that when
> they edit a wave, they are affecting every wave participant's 'copy'
> of the wave, since there really is only one copy.
>
> I set up a wave for a couple of technical groups, one accrued some
> interesting contents, until someone, thinking of it as 'email' decided
> to clean things up by deleting everything HE had already read.  Which
> deleted it for everyone.
>
> So I'm seeing a lot of people playing with wave with no clear picture
> of how it is intended to be used (I'm including myself in this), and
> Google hasn't as far as I can see given such a picture.  I'm not
> really sure that they have one themselves and that wave is still a big
> social experiment to try to figure out what it really SHOULD be.
> There are a few resources going out like Gina Trappani's book on wave,
> but right now, it seems to be worse than the wild wild west, or the
> unexplored sea. It's hard to know where the bandits and dragons are.
>
> By the way, there are ways to have waves searchable by members of a
> google group (e.g. the google group which follows ruby talk). You can
> actually add the group as a participant using the groups email
> address.  Wave will seem to complain about it not being a wave
> address, but it will work, and then members of the group can search
> for the wave with the group: prefix to the search.  (I'm doing this
> from memory, so there might be some variations from what I just said,
> but the function is, or at least was, there).

I read a description of the design of network protocols once which
noted that simple, to-the-point, successful protocols are succeeded by
ornate, overcomplicated, overdesigned protocols. Examples: slip was
followed by ppp, bootp by dhcp, rip by ospf/bgp/isis. (Actually, I
like both dhcp and ppp myself, and consider them improvements on what
came before.)

Clearly the same is true of other areas of technology. Gmail was a
very successful (or at any rate, popular) service for google, so they
said, "Ok, good. Now let's turn it up to 11."