One of the things I like about Discourse is that they have a business model – they will be maintaining, improving and supporting the software for years to come because they are making a living doing it. How do you plan to keep up maintenance without inevitably burning out?
I use it (both when consulting and for personal projects such as https://plforums.org), and I work on the features I need. Other people use it for their businesses/projects and add the features they need (as plugins or to core), and everyone benefits from that. Some of these people incidentally learn the internals and become maintainers.
I know plenty of people who disagree but I believe open source software does not need to make money nor be copyleft to be successful.
Was this a response to Discourse in any way? It seems to hit on a few pain points with Discourse such as the boring deployment. For a non-Rails dev, how hard is it to add a custom authentication provider to Thredded so that users from another site don’t need to have a separate logon?
Was this a response to Discourse in any way?
Yes. :)
For a non-Rails dev, how hard is it to add a custom authentication provider to Thredded so that users from another site don’t need to have a separate logon?
That depends. Integrating it with thredded is very easy (you only need to provide a current_user method) but writing it might not be.
What is the caching strategy here? It doesn’t matter until it does, but I was curious if this has been run at a load where Redis or Memcached start to make sense. (This isn’t to imply that it’s useless w/o it, I’m just curious). One of the benefits of Discourse is the overflow of SO talent at running quick websites.
What is the minimum size server that would hold up to, say, 20 active users? Can I run this on a VPS and have it play nice with other websites w/o the server requiring 1GB of RAM? (1 GB seems to be the min for discourse)
You can use any cache backend supported by Rails (that includes Redis and Memcached). If you use a backend that supports multi-get (e.g. Memcached), all the posts will be fetched in a single get call. 20 active users is very little, you can likely have a single worker, so around ~300 MiB of RAM last time I checked.
Glad to hear that! Please feel free to make an issue on our GitHub project if you have any questions / issues.
I’d love to contribute, but I don’t know much Ruby. Do you have any outlined milestones so I can keep my eye on the project though?
We do, we use GitHub milestones https://github.com/thredded/thredded/milestones
This is awesome, but it doesn’t have any algorithms yet. Their client-side stack is described here https://github.com/algomation/apis. The other day I wanted to visualize an algorithm (http://codepen.io/glebm/pen/83967288b0e673d07dc6237379a0c6b8) and hand-coded it with D3, wish I’d known about dagre (https://github.com/cpettitt/dagre).
There is also http://sigmajs.org/ for drawing directed graphs.
This is an excellent improvement to Rails i18n! But as a regular user of gettext, I’m a bit surprised nobody figured this out before. Analyzing the application source code to extract translation strings is a standard feature of gettext and people have been using it for years. Anyway, kudos to the authors for porting this to the i18n gem!
Compared to gettext, doing this for i18n is less trivial. I’ve never used gettext, but a few things that gettext does not have to deal with, but this gem does are:
Sorry for answering so late. I’ve just read your reply. You’re right, i18n was not designed with static analysis in mind, unlike gettext. Now I understand why implementing this gem was not trivial at all. Thanks for sharing your insights about this; it would probably be helpful for a JS project I’m working on!
Yeah, the stream-of-consciousness thing really works as a way to show seasoned pros what it’s like to be a newbie again. As a maintainer this kind of thing would be really valuable for uncovering usability issues in my own software.
If you reply to a post notification, will it post your response to the corresponding thread? I was looking at GroupServer a while ago to set up a hybrid mailing list-web forum, but the setup process is rather involved.
Almost: the email previews linked from the homepage don’t display without JS enabled.
No. That’s something we’d be interested in merging in case you’re interested in contributing.
The email previews linked from the homepage are not user-facing, they are a developer tool. The only user-facing feature that does not yet degrade gracefully (that I can think of) is the recipients list input in private messages.
The recipient list input in private messages also degrades gracefully as of v0.13.0+