Civilized Discourse Construction Kit

Occasionally, startups will ask me for advice. That's a shame, because I am a terrible person to ask for advice. The conversation usually goes something like this:

This is a companion discussion topic for the original blog entry at:

I’ll certainly be checking this out. I agree that forum software sucks. The best I’ve ever found, even to this day, is not open source or even downloadable. Delphi Forums. When you need ideas for innovation… look there :wink:

I know they’re completely separate ecosystems and should probably stay that way but it is tempting to think of a “Migrate to Discourse” button for certain questions on stackexchange sites like arqade that would actually be suitable for continued discussion.

The web… a new frontier. These are the voyages of the startup Discourse. Its five year mission: to explore strange new customers… to seek out new thoughts and new deliberations… to boldly go where no forum dared go before!

(When you said “five year mission”, I couldn’t resist.)

Ruby on Rails and Postgress? shudders Why, Jeff, why?

Very cool. I’ll definitely take a look. As far as I’m concerned, integration will be key. Forums, as such, are part of the answer even in a lot of places where they suck today. Even the chatrooms connected to SE serve this purpose from time to time. I’d like to see a way for the chaos of forums to seek nice, dark, quiet corners, while the good stuff – the “sticky posts” or whatever – can float to the top and be promoted to something more dignified than a “post”.

I’d bet you could come up with a pretty good list of requirements just by listing “forum antipatterns” that you want to break:

  • The troll.
  • The "haven't we done this seventeen times already?" thread.
  • The "you're not following the rules for this category" thread.
... and so on.

Should be interesting. Thanks for the contribution.

“it is tempting to think of a “Migrate to Discourse” button for certain questions on stackexchange sites like arqade that would actually be suitable for continued discussion.”

I dislike this as some actual signal might escape to the other thread, and similarly it’d encourage people to start noisy discussions where noise is less tolerated, as they know it would be a jumping off point.

You mentioned frequently finding useful information on forums when doing web searches. What does it mean for search engines that the client is entirely JavaScript (and has such high browser requirements)?

Another question: I feel like the high success of WordPress is largely due to the ubiquity of its platform, i.e. PHP/MySQL. What effect do you think the relative scarcity of affordable Ruby/Postgres hosting will do to the adoption rate of Discourse?

This isn’t 2007, guys. Hosting Rails is cheap and easy now.

What I’m surprised about it that it’s not .NET, given Jeff’s frequent comments about how amazing it is.

Coupla things. First, “mailing list” - if you mean listservs - are NOT awful. Best forum/list I was ever on was for ultramarathoners. It was amazing - you could post a question and receive an answer from a top South African ultrunner. Of course, the group was tightly focused around a single topic - had highly respected participants - and a population of about 1000.

That said, I’ll be looking at Discourse very seriously as a platform for our organization with 40,000-plus members worldwide.

And THAT said, you could have mentioned at the start that it’ll be TWO YEARS before you release it. And that “open source, completely free” doesn’t mean, well, free as in beer.

It feels awesome to use, I guess the only gripe I have is with the way it has been released, i.e. as a big code dump (similarly to Android, which is infamous for it). I hope to see a more collective effort in the future development of this platform.

I would love to see Persona/BrowserID support built in here.

Just found out the source for your “let a million discussions bloom:”

The first part of the phrase is often remembered in the West as "let a thousand flowers bloom". It is used to refer to an orchestrated campaign to flush out dissidents by encouraging them to show themselves as critical of the regime, and then subsequently imprison them.

We're also serious about this eventually being a viable open-source business, in the mold of WordPress.

The way it looks now, you’re not engaging in open-source in the same way as WordPress does.
The GPL is enough, you don’t need a contributor license agreement. What it actually means is that you are keeping an exit door if you ever want to close the source again in the future: contributors agree to license you their copyright for whatever future happens.

Probably lots of open-source projects use this method, but it sure is not reassuring. Please don’t compare yourself again to WordPress, though. Nuance matters.

Andrew: I’ld also like to see that. Perhaps you could add it, and submit a pull request? :slight_smile:

(I would, but I’m not really a Ruby/Ember kinda guy. I’m more Node/Backbone. :wink:

That’s so absolutely awesome, I’m exicted as I was last time when Google Chrome came out.

20 years of suffering is over!

The best forum software ever created was Usenet. Unfortunately, it never made the leap to the web.

Discourse looks nice, but will it be better than tin

You have to do something about SEO. I read the article in Zite on the way home on my iPad…came home and googled “discource” and got…nothing remotely resembling the site. Had to come to codinghorror to find this article and the URL…

Also: Where in discourse is the refresh button to load new comments while you’re reading? Or does it automagically update? If it does, I haven’t noticed it…

I’m looking forward to this!

I’ve cursed forum software for years and a recognized Stackexchange and Quora as important, thoughtful, successful attempts to do something about the problem. I am glad to see that you are continuing to iterate on the problem, and even happier that it is open source.

I think though that Wan raises an excellent point about the licensing though. I can appreciate the desire to be able to re-license, but I hope you can appreciate why potential contributors would be reluctant to participate. I’ll also point out that there is nothing to stop people from creating a fork right now and accepting commits without the encumbrance of the licensing agreement.

Wordpress/Automattic managed to pull-off its achievement without having a paid staff and a paid product offering in place until well after the open source project took off. So, you are in many ways starting off way ahead of where Automattic started in being able to capture value from Discourse. You could quickly loose that leading position and favored momentum if the contributor agreement drives an early fork.

Actually, the more I think about it, the more obvious it is to me that the contributor agreement is almost a moot point. You’ll never be able to incorporate contributions from a fork without nullifying the flexibility of the contributor agreement, but any fork will be able to pull anything you publish to your repo.