Previous Entry Share Next Entry
Conversation and CommYou
jducoeur wrote in querki_project
I haven't been posting much here lately -- sorry about that. I've been steaming along towards the first milestone (our wedding Save-the-Date page), which required a lot of heads-down programming, and then the holidays interrupted. (With a couple of weeks of working on the OP Compiler, a personal project I'm doing for the SCA.) I'm gearing back up now, though: Milestone 1 is complete, and I'm starting to work on the parser for the QL programming language (about which more in the coming weeks).

But I woke up this morning with a UI design for the conversations part of Querki's mobile interface, and that's gotten me fired up about Conversations again. This has always been a particular passion of mine (indeed, my sometime professional blog is mainly about the subject), and it's going to be very important to Querki. The programming has slowed down while I read into Parser Combinators for QL, and I have time to write. So let's spend a few posts on the topic. Hang on -- I'm going to be writing quite a bit over the next day or two.

The Rise and Fall of CommYou

Some context is helpful, to understand where this is all coming from.

Once upon a time, I had a previous garage startup, named CommYou. (Pronounced "cuh-MYU" -- one of the company slogans was "The Emphasis is on 'You'".) CommYou was focused on building a truly state of the art conversation system -- taking everything I've observed from 30 years of playing with and working on online discussions and forums, and pulling the best of it together into a single coherent model.

CommYou toddled along for about a year -- I had some really great designs, but quickly came to understand that the state of the art in Web programming wasn't quite where I needed it to be, so everything went slowly. (One reason why I have higher hopes for Querki is that a lot of key technology has come a *lot* further in the past four years.) After about a year, Jane told me that it was time to get a paying job, so I did so but kept CommYou going on 20% time.

And then Wave came along. In one of the most striking pieces of parallel evolution ever, Wave looked bloody damned near identical to CommYou. They had picked up the same threading model (what I call "heavyweight threading", discussed later), many of the same UI details (such as focusing on making it quick to find new comments), and so on. In general, it was so creepily similar that I occasionally wonder if somebody there had looked at my alphas, but I suspect that they simply came to the same conclusions from a very different direction. I was trying to build a system for purposeful conversation, they were trying to build one for document collaboration, but we wound up with the same conclusions about what was needed and how it *had* to work. So I regretfully let the CommYou project go, on the theory that Google was doing it and doing fairly well -- while I didn't agree with every decision (in particular, they *completely* fucked up the social side of it), I figured they'd get it right eventually, and I would at least be able to use it.

And then, a year later, Google killed Wave. To say that I was livid would be a massive understatement.

So I'm left, almost five years after the original inception of CommYou (which started as a few ideas I had to try and save Zingdom, my employer in 2007), and I am still left with the itch. Things aren't completely dire any more -- in particular, Rizzoma has picked up the ball from Wave and is gradually moving that forward -- but I have loads of ideas to play with.

How does this fit in with Querki? I'll get into that in my next post...


Log in

No account? Create an account