Great post @erlend_sh!

And love the site list, gotta get some ideas for my own instance of gaming forum.

Shoot! I forgot to nudge one of my favourite new (non-customer) Discourse forums about the surevey, https://boards.faeria.com/. If anyone wants some styling inspiration, that one's a must-see.

I find it a bit curious that one of the features mentioned as awesome by the communities polled is the badge system.... not because it isn't good, I think it is AWESOME and we use it extensively on our own forum http://forum.combustionpunks.co.uk/badges
but because there doesn't seem to be much custom use on the forums surveyed.

We are really wanting to push the use of the badges on our forum and are looking at how other forums utilise the system as per my post here https://meta.discourse.org/t/who-is-extensively-using-the-badge-system/46770 and I am looking for inspiration from how other communities use the badges.

Anyone know of any forums pushing the badge system?

That's a good observation. We can continue this discussion in your dedicated topic.

this is pretty awesome! I might try this out on a new project, but just a heads up, the github links are dead. looks like the repo was refactored, so now the links are here:

wiki.js

wiki.php - now a module
https://github.com/osmc/website/tree/master/server/modules/wiki

is it possible to get Sam (or someone from OSMC) to put together a little how-to on using their Wiki code?

Hi. Yeah I've been refactoring a bit .

All the server side code is now here:
https://github.com/osmc/website/tree/master/server/modules/wiki

Wiki.js is just a very basic search feature you can see live on https://osmc.tv/wiki (at this point all the wiki data is being served server-side).

It's a bit of a convoluted/hack'ish setup, so it's probably a bit tricky to follow without knowing the flow. I'll make a guide soon

awesome - thanks @marktheis! as an aside - is the site css based off a framework, or completely custom? I noticed you guys use express/ghost/normalize.css - wasn't sure if the rest was built from scratch or used something a la Susy or Bourbon as a starting point. either way, I love the design!

css is completely custom. Frameworks are nice, but I always end up fighting the framework. And they're usually too big for my taste.

Yes apart from express, ghost, and a few js/css modules everything (server/client) is built from scratch.

With a website that depends on a lot of external services (discourse for wiki and comments, woocommerce for shop, another server for diskimages) I think that node with express and handlebars templates works really well.

Btw. I'm also currently building an email system that takes posts from discourse, and turns them into ready and processed emails via Foundation for Emails. To be consumed by phplist. Yes I really don't like phplist's html editor

Do these custom scripts handle interlinking, categorization and tagging in any way?

Interlinking: if you mean canonical urls then no. We find that google prioritises osmc.tv/wiki over discourse for us, so I haven't made any changes there. (It's not that important for us atm).

categorization: Yes each wiki post is in a category our discourse and site.

Tagging: No. We don't use the tag plugin.

Way to go Team Discourse! This release looks amazing! Thank you for all of your hard work!

Discourse is an AMAZING piece of software!

Some great stuff there

Typo:

Crazy in Love — Used 50 lines in a day 20 times

Courtesy of @DeanMarkTaylor, here's an interesting stat addendum.

Counting activity in all the "discourse/*" repos during the period of 22. September 2015 - 01. April 2016:

• 136 users made 356 Pull Requests (across 17 repos)
• 89 users had 221 of those PR's merged (across 16 repos)
• In addition other pull requests were used as the basis for coding changes added to the core directly by the team.

Big thanks to the following contributors on GitHub:

• cpradio
• dandv
• davidgnavas
• fantasticfears
• gschlager
• xfalcox
• fefrei
• mikhailvink
• pvalexander
• rriemann
• abbasfaisal
• adamcapriola
• adrapereira
• andyw8
• angusmcleod
• aryanraj
• bgr11n
• BYWallace
• carsonreinke
• chapel
• darix
• dasnixon
• dereckson
• devonestes
• discoursehosting
• djtye
• Dorthu
• dtchau
• DukeofRealms
• ebernhardson
• emanuelet
• fearlessfrog
• featheredtoast
• gdpelican
• getabetterpic
• gpaumier
• GrantStreetGroup
• Grexy
• gwwar
• halfstrik
• httvncoder
• humzashah
• huulbaek
• iamntz
• jamesfid
• jamesgecko
• jamielinux
• JaredReisinger
• Jdesk
• jeremylan
• jfoclpf
• joebuhlig
• jomaxro
• JSFernandes
• jslew
• karies
• keepcosmos
• kellec
• khoa-le
• LeoMcA
• lumenlunae
• maclover7
• manelvf
• markbiegel
• Martyn96
• mcwumbly
• MichaelMarner
• Mooash
• motte215
• phw
• pra85
• RalfJung
• rchanley
• RomeoPapaSK
• rubo77
• ryantm
• sammyd
• scossar
• seth-reeser
• sethdeckard
• sethherr
• sghebuz
• simplystuart
• Stealthii
• tnorthcutt
• tomasibarrab
• uppfinnarn
• xfix
• zigomir

Maybe we should start a tradition of including this exhaustive list as part of the #releases topics?

Is there any way to provide Google Gmail (SMTP) or Google Apps (SMTP-relay) as one of standard option in Discourse admin panel - rather updating app.yml?

Between, is there any post on new version 1.5 release? I mean change log summary?

Other than this?

wow. that's big list Thanks @Falco

Setting up SMTP has to be done by updating app.yml, sorry.

Using google for SMTP has been discussed at length.. try the search. Here's one.

I personally tried it and gave up on it in favor of mandrill, and now have switched to mailgun. It's easy and "set and forget" once it's working.

Is a mail-tester score of 7.5 out of 10 expected? To me it looks like everything is fine, except maybe there's a broken link there ( "Test your DKIM record" ) you guys could fix.

I had deliverability issues with Mailgun that I fixed by adding MX records for our Mailgun subdomain:
https://documentation.mailgun.com/quickstart-receiving.html#add-receiving-mx-records

It's a bit counterintuitive, because we're not receiving emails through Mailgun. But here's what Mailgun's support told me:

The error (which is called Sender Address Verification) that you are seeing is due to the recipient server performing a MX lookup on the envelope.mail-from address and failing to find the records for the domain listed. Some email servers check the domain in this envelope.mail-from field for MX records, and if the records do not exist, the email servers reject the message. Therefore, by adding Mailgun's MX records to your domain, the MX lookups of those email servers should succeed, and the servers will accept your messages.

Hope this helps other frustrated folks!

As a 10-year veteran of GSoC, I believe at least one pre-proposal code contribution (i.e., involvement with the project before applying) is the biggest predictor of student success. It's not a guarantee, and it's possible to be successful without it, but as mentioned in the blog post, it's extremely valuable.

1. Mentors and org admins have some level of assurance that you are up for the challenge.
2. They also see that you care enough about the project to learn a little bit about it and take on a bug. (Even if you've done a pull request and it doesn't get merged!)

Many GSoC organizations require a contribution before the proposal, but even if they don't, it's always a good idea.

Good luck to all students!

I guess he's talking about doing Passion Work on your free time.

I have a full time job (8h a day + plus commute) and I try to help Discourse at night when I get home or on Weekends (between some Dota 2 games ).

My wife uses her free time doing charity work, I do open source, my friend study foreign languagues, etc. It's all about following your passion .

Just to clarify -- if your project is on GitHub and you don't quite meet the threshold criteria of …

• 2000+ stars
• 30+ contributors

… feel free to contact us via the form. We'll work with you.

The purpose of the thresholds isn't to arbitrarily keep projects away, but rather to ensure that there's enough critical mass for a solid discussion community to form around your project. We guesstimate it takes about 5 people actively participating in the discussions each day for your discussion community to "work".

That's great news!

Your users can log in with GitHub logins

Are only GitHub logins supported, or will Twitter, Facebook, Google+ and plain e-mail accounts also work?

Also, I understand custom imports would take too much time, but importing from an existing Discourse instance can be done manually, right?

All of those will also be available.

Yep, that's entirely doable. Correct me if I'm wrong @team, but we do prefer to have the Discourse backup sent to us so that we can apply it ourselves.

Yes, you can configure this yourself, no problems. Additionally we plan to roll out "automatically" configured incoming email in the upcoming months.

Oh my bad, I didn't realise you were replying to the post right above. No, you will not need to create a separate gmail account for email. Speaking of which, we are gradually rolling out this feature for our customers now. We're starting in small increments, so the official announcement is still a little ways off.

My comment might be slightly off-topic, but I wanted to say that I'm amazed at how Discourse team markets their product and applies various strategies to target primarily larger organizations and companies. There is a lot to learn from them - and this topic is just another excellent example of how to make people tempted to move to Discourse. Absolutely amazing stuff. Read it twice - first, as a Discourse (potential) user. Second - as a marketer.

Happy Birthday!

Happy birthday! I feeling old coming from the red/blue pill days.

It's good to have the .com domain name, but I'd argue the general cultural expectation is for an open source project to run under a .org domain.

Congratulations to the entire team!

Congrats guys!

Congratulations to the discourse team.

In this case they could be more like WordPress and move the hosting portion over to the .com

p.s. I would have liked to include some proper GitHub statistics e.g: "unique contributors to Discourse (+ related projects) from Jan 1 2015 to Dec 31 2015" and so forth, but I can't work out how to get this from the GitHub API.

If someone would like to lend a hand with this, let me know and I'll be sure make good use of it, starting with our upcoming v1.5 release!

I think we should consider doing this with plain ol' git rather than github.

The script that parses the logs to create release notes might be a good place to start. There might also be prior art out there.

Had a little play with BigQuery, first time I have so you might want to check for errors:

This is just for the github.com/discourse/discourse project.

Pasted image879x306 8.41 KB

Pasted image1002x410 14.8 KB

SELECT
count(*) as count,
type as type,
STRFTIME_UTC_USEC(TIMESTAMP_TO_USEC(created_at), "%Y-%m-%d") as date
FROM (
  TABLE_DATE_RANGE([githubarchive:day.events_], 
    TIMESTAMP('2015-01-01'), 
    TIMESTAMP('2015-12-31')
  )) 
WHERE repo.name = "discourse/discourse"
AND NOT type contains "Comment"
GROUP BY date, type
ORDER BY date, type

I can provide a Google Sheet or CSV perhaps if needed.
Google Sheet here

You can provide a range of dates (see URL parameters)
https://github.com/discourse/discourse/graphs/contributors?from=2015-01-01&to=2015-12-31&type=c

Thanks @erlend_sh it was fun to put thought into this.

I was going to add that I'm still keeping my eye on the topic about warning on external links, e.g. bad guys posting phishing links like this:

[totally safe and normal link](http://evil-phishing-site.com/bad-link)

It might be best to warn (if enabled) on TL0 posted links, like so:

"Improving SEO" must be one of the best and often technically overlooked discourse features.

For one, www.discourse.org have no mention of "SEO", "engine" or "google" (in this sense) after a quick CMD + F.

Very good word choices for the title!

Yep. "Reply as linked Topic" should do the trick nicely

@john_mardlin incredibly helpful to read about your SEO strategy and results.

Also, I am intrigued by how you link up your knowledge base and Discourse as that's a balance we're trying to strike as well and we'll probably just move forward, copying your approach with "feedback" buttons on the bottom that link to hidden docs.

@codinghorror I'd love to see more blog posts about how people use Discourse, this was really helpful.

Go nuts.

It was a lot of work at first, but we eventually found some tricks to simplifying the process. The key was instead of manually creating a new topic, we made use of the /newtopic URL feature, and built a spreadsheet of all our KB articles:

Screenshot 2016-01-14 at 9.51.14 PM.png2525x530 132 KB

Hopefully that's enough to help you reverse engineer the process.

To @cregox, SEO was actually more of a pleasant side effect. I typically think of SEO as a marketing tactic, although for us it's almost entirely SEO for people to help themselves. Though really the lines behind customers service and marketing are blurring, and both are good things to have.

Yay! @eviltrout just updated us so we officially support Unicode 7.0 emoji, released June 2014, and Unicode 8.0 emoji, released June 2015.

He confirmed all four emoji sets have 144 standard emoji additions and this enhancement will of course be an official Discourse 1.5 feature.

Let the countdown begin. You get a and you get a and you get a ...

Wow! Very nice!

I'm upgrading my instance now because of this .

Ps.: I guess the skin tone interface will be kinda tricky...

This time around I was focused on adding all the missing emoji we had. Just doing that for all 4 sets was quite a bit of work. And @codinghorror I actually added 196 emoji, including all new ones for Unicode 8. The one thing we're missing is skin tones because we'd need to change our interface around for that.

Yes, basically I extracted the current versions of all 4 sets we use. I then wrote a script to get the intersection of emoji that are present in each set that weren't previously in discourse and there were 196.

I'm not sure if we were missing some from Unicode 6 or if all sets happened to include more emoji that is later than unicode 8. Each set had quite different numbers of emoji that weren't in Discourse, so I'm not sure if they speculatively try out new emoji or what, but the good news is we have many more to use!

What about this:

https://github.com/iamcal/emoji-data/blob/master/img-google-136/1f1e6-1f1f9.png

This repo looks quite nice for what you guys do with Emoji. If you would like help adjusting the import I can try and help!

Oh neat! That repo wasn't on my radar. If you want to see the tools we used to extract the emojis they're all in this repo:

https://github.com/discourse/discourse-emoji-extractor

In order to merge them into discourse we need the emojis

1. Present in all 4 sets.
2. Renamed to the emoji.json name
3. Minified using zopflipng

Then they can be put into a PR after they are:

1. copied into the discourse repo
2. updated emoji.json in our repo with what is present
3. Increment emoji version

(It's a lot of work but all the tools are there)

The topic linked in the blog post isn't public.

Thanks Mozilla!

Should be public now.

great. Is there any place to look how email improvements is going to look like ? So we can prepare.
I am dying to build up mail-list upon discource, it will be nice to link posts whit features which describing improvements of e mails.

Yeah, check out the MOSS Roadmap topic:

Not all features are specced out yet, but this list should give you a pretty good idea of what's coming.

The product seems to be working great, so you are off to an amazing start. Best forum platform that I'm aware of.

Weee! Such a blast to be part of this. Here's to the 10 year mission!

Yay! Glad to join the team on its mission!!

Here is a happy cat

Agreed @charleswalter. But it's good to know there are great competitors popping up.

@tgxworld that cat is waving its tail like a happy dog... creepy!

Wait, I always thought the "10 years plan" would always stay 10 years in the future!

What will happen in 2024?

Big fan of Shorewall, been using it for a looong time.

As I mentioned in the blog post -- special thanks to @tgxworld, @Simon_Cossar, @gwwar, and @gerhard who contributed a bunch of stuff to this release.

We will be reaching out to you directly to send you some goodies

Discourse 1.4 represents some Huge Strides to making it the defacto mailing list replacement. I'm excited and encouraged by the development!

Look for the "version" parameter in the app.yml file.

You rock guys! Really

EDIT:

I’m a little behind here, but thank you for the ability to style embedded sections!

I’ve made a few tiny tweaks to the embedded styles, and now our blog comments fit in perfectly with the rest of our site: https://ind.ie/blog/oh-wont-someone-think-of-the-children/index.html#discourse-comments

You can find this here :

Great, just one question. Is it possible to make the poll sticky AND only sitcky to those who have not voted ?

Ah, yeah, you probably don't want HTML output, you likely want to have it send plain text

[poll name=sharebutton_color]
- Blue
- Green
- Other (add comment below)
[/poll]

It would instead push "\n" in replace of the <br> tags, and a space instead of the   items.

No, but sticky posts can be unstickied by the user, or automatically by reading the entire topic.

Here's a tip -- you cannot have a space in the poll name. If you do, the poll won't work at all (guess how I found that out, and it was my first poll). So don't do this:

[poll name=My Poll]

Instead use something like this:

[poll name=My_Poll]

Otherwise, great feature!

We'll be removing the name since it's only used internally to differentiate between polls in the same post.

Here's a poll with images. All I did was put * <img src=" ... "> for each option.

how did you have calculated 16.000 active users in a single year? I can't get it

Probably out of Google Analytics or similar.

Those /about page stats mostly represent logged in users, which is only about 25% of overall volume.

You can see the rough breakdown of "pageviews" traffic here, this is based on traffic to Choice of Games forums in the last 30 days:

There's obviously much more IP uniques diversity in anonymous users as well, logged in users will visit more pages for longer.

I should also mention that @erlend_sh put all these questions together and really made this happen.

I had talked with @badevguru a while ago about possibly blogging their story, because open source freakin' underwater robots are awesome, but Erlend was the catalyst that pushed things forward.

We hope to make this perhaps a bi-monthly series to highlight the cool stuff Discourse communities are doing!

Very awesome indeed!

Now, that Auth0 Single Sign On issue there sounds quite big. Did discourse evolve from there since then?

Nice read, enjoyed it.

I can see my feature request got a mention... But still has no likes sad puppy

As far as I know, SSO has not changed since I did the Auth0 bridge. If anyone is interested I can post the bridge code... I have it running on amazon beanstalk nodes.

Excellent post @badevguru. Your story sounds eerily close to what we went through transitioning our site and forums... we have yet to implement the SSO but it's high on the list.

Your theme is close enough to ours that I'm gonna toss ya a line of CSS. This fixes long username overflow on the topic list.

.poster-info {
max-width: 80px;
}

Didn't realise how many shinies there were in this release until now. Woopwoop!

--fixed, back to the point

Congrats, team! Flipping all my servers to stable branch now

if this isn't meant to be a joke

this is quite easy:

cd /var/discourse/containers
nano app.yml

uncomment the last line and change from "tests-passed" to "stable"

exit the yaml file

cd .. 
./launcher rebuild app

things to note:

• assumes docker install
• assumes yaml file is called "app"
• WILL bring you offline while rebuilding ~4-10mins

Great work! That's even more features than I expected

that was not a joke but a real newbie question
thank you

How about making the "New! 1.3" sticker link to the announcement post? It's quite a shiny sticker, I'm sure a lot of people will try clicking on it to learn more about what's new.

Awesome! Congratulations @techAPJ!

From out here, it seems like you've been part of the team for quite some time. Great to see it be made official.

That's awesome, thanks for sharing the story, but... Left me wanting to know more about @techAPJ !

Same here except for me it wasn't "seems", I thought he already was!

And yes it is great [cough] about time [/cough]

Well deserved! Congratulations!

Excited to see even more miracles worked by @techAPJ. Congratulations!

If I don't fulfill every single requirement with excellence, should I even bother applying?

I'm pretty sure this will get plenty of applications and you'll probably filter out based on that first. Right?

The main requirement needs to be strongly met:

Deep Linux sysadmin background, ideally in a hosting environment

This sounds like a calling but I'd be dead weight while learning the ropes of the items underlined at "automation experience" and "tools we rely on". I planned on learning those technologies but again, temporarily unable to do standard due to knowledge limitations.

I am also available on call since I really don't sleep that much, haha.

Best of luck to whoever nabs this dream job.

I would have tried but I'm from Russia

I know someone who might be interested if this position still open, I'll send him the link.

Has this been filled yet? I'm currently hunting a bit and might be interested. I've mainly been software engineer but I always wind up also doing devops/sysadmin for whomever I work for, and I've been running a small webhosting service on the side for over a decade. Unix Sysadmin is where I started...

Is there a product roadmap anywhere? What's the next major release of Discourse? 1.3? 2.0?

Are there plans for a 2.0 release? Would be it in 2015, 2016, or 2017? What's the big feature?

Try the #releases category.