Outline vs Discourse

We also have good experiences with testing

as a searchable, collaborative team notebook: It has OIDC login, exports to Markdown and has real-time capabilities.

I moved this post to a new topic because of its proprietary license, and the fact most (if not all) of its integrations are with proprietary software.

Its license states:

The Business Source License (this document, or the ā€œLicenseā€) is not an Open Source license. However, the Licensed Work will eventually be made available under an Open Source License, as stated in this License.

IMO what Outline does, Discourse can do. Iā€™d be curious, since you have tested this softwareā€”which I do not recommend for the reasons stated aboveā€”, to see a comparison of Outline and Discourse, to see where Discourse could become better.

I can see a bit already:

  • / commands in the editor
  • beautiful design for published documents

For the rest I donā€™t know, Discourse seems very well suited to create, maintain, and publish excellent documentation. Maybe it would be stronger at it if it supported Asciidoc in addition to Commonmark.

1 Like

I agree with the licencing notion, that as a Librehoster it is hard to argue for offering Outline as a service. Itā€™s licence makes it incompatible with our value to support and encourage Free Software. While the license also states:

Additional Use Grant: You may make use of the Licensed Work, provided that
                      you may not use the Licensed Work for a Document
                      Service.

                      A ā€œDocument Serviceā€ is a commercial offering that
                      allows third parties (other than your employees and
                      contractors) to access the functionality of the
                      Licensed Work by creating teams and documents
                      controlled by such third parties.

Which is a precise definition of our use case, because its business licence will have a blind spot on alternative realities. Here we are presented with the motivations one can follow when using Outline, and when not to. Especially when we are not selling it as a service, e.g. per user account or organisational third-party, but using it as a tool to grow a community of informal or not-for-profit activities, it may still remain deployed within the above constraints.
For me the question is more to justify actual development efforts and contributions in filing issues, tracking down bugs, providing example code changes, or testing merge branches, if not as a personal hobby.


In our establishing community which grows with an Outline knowledge base attached, we already witness certain community dynamics, which might have occurred differently in another setting, such as Discourseā€™s. While Outline excels technically, its code base is clean, modular, documented, typed, and progressively, large-scale refactorings are given priority over small feature improvements (migration to Passport.js for authentication, to TypeScript for code quality, to Y.js for collaborative editing), we might not be able to acknowledge fully for its limitations.

  • Marginalia discussions or Comments below are missing, which is dispersing the conversation around a text into side-channels. And there are plenty.
  • E-Mail notifications about a change of a text do not contain the difference (yet?).
  • Its hierarchical tree of documents hides the actual visibility of documents to varying audiences. A groups feature is present in a rudimentary form, and only indirectly accessible as an administrative means, not as a visible and palpable structure. This may be due to the aforementioned assumptions shaping the platform, which constitutive forces nevertheless turn out to be constraining the applicability of its design in diverse environments.

What we are grateful are, as you mention:

  • A very useful graphical editor, that makes usage of Markdown a breeze. Which doesnā€™t mean it works well with advanced usages of the markup.
  • And also, it is a convenient place to quickly deposit information in a semi-attributable, collective environment where ā€˜the sum is more than its partsā€™.
  • The search interface also allows non-technical people to quickly access their repository.

I think these are also the points, which make it a more suitable place for collective writing than Discourse is. As a regular human, it can only be confusing, why there are three instances of a given text, e.g. that I am wiki editing, if I have managed to find the right button: one that I wanted to edit, and another one that is a mirror of yet another one that looks slightly different than the original text and a bit like code sometimes. What did I want to write?", is where the story ends.

While Discourse can be extended easily through plugins and themes, it cannot be an All-in-One-Tool for every use case. It turned a bit into the ā€œWordPressā€ of community engines, while still serving a mainly technical community. Its strengths come into play, when adopters heavily tweak the design and reduce the visual clutter to minimise the presented options to the absolutely neccessary.

It seems, eventually, more software than expected turns itself into wikis with attached conversations. Itā€™ll be beautiful to watch how many more of these digital gardens converge on humane, yet well integrated interfaces that prioritise their accessibility and community aspects.

From the same closed community and their Outline instance we can learn a lot more of accompanying projects and argumentations. For direct comparison, here we have the same content thrice: once in Discourse and CodiMD each, and twice in Outline itself.

This documentation has been extracted/evacuated/shared from the closed team space for supporting a not-for-profit initiative, which is using Discourse within the educational sector, more precisely for schools, their administrative personnel, teachers, parents and children alike. @toka, itā€™s maintainer, is a great fan of Discourse, like yours, and curiously makes up new use cases of a platform that has: variable authentication around content, a multi-layered notification system, email support, an extended search function, a useful content moderation and curation scheme, as well as a ā€˜neutralā€™ interface, as in not so exciting. Which can be good, I guess. Yet no App - whoā€™s ever installed it to their home screen and allowed the web worker to drain your battery for sending native notifications to your device? - and on the Web, which is an unusual encounter for people these days. You should talk to each other another day, and see how your interest for Discourse as a platform for multiple use cases can be accommodated, e.g. by suitable community moderation guidelines and practice, or varying sets of plugins that work well in certain combinations.

I think Outline only struck, because it is dead simple, and, in a very tempting way, puts content first. But itā€™s as easy to get lost in deep, long and intermingled argumentation, as it is anywhere else. If there was only transclusion readily available for the rescue ā€¦

:eyes: Looking at you, LogSeq.

ā€¦ incoming links as marginalia right on the spot, an actual @-mention notification system, the integration of other Chat platforms than Slack and the publishing of whole site sections and not only individual documents aggregated into ā€œbooksā€, export/sync to Git(-ea,-Lab,-Hub) plus shared document drafts for collective editorial before publishing could be a beginning.