Announcing the new Erlang Forums!

[elixirforum]

Hello everyone

We have some exciting news, the brand new Erlang Forums has just gone live!

This is great news not just for the Erlang-specific community but for the entire BEAM world - we’re on the cusp of an Erlang evolution and we’re really excited about what this could mean not just for Erlang but for Elixir and all the other languages that run on the Erlang VM too.

Please read the launch announcement to see why we’re so excited and then come back here to see what all this could mean for Elixir

Read it? Excited!? Great!!

Ok, let’s continue…

Firstly, don’t panic, this forum is staying here While it would have been easy to move it there it wouldn’t have been fair on Elixir users, nor Erlang/other BEAM lang users because it would have just felt like an Elixir forum full of Elixir content, and this could have made everyone else feel like they’re secondary or imposing on someone else’s territory. Besides, the community here has shaped this forum’s personality and helped forged its path over the last 5 years and the new forum and the community that develops there deserves the opportunity to do the same. So this forum is not only staying here, but we even made some changes recently to help with Elixir’s own future growth, such as with our new Elixir Ecosystem Chat sections.

The reason why the new forum matters is because it is aiming to play a part in the progress and success of Erlang, and the success of Erlang is important to every language that runs on the BEAM. A strong and healthy Erlang means a strong and healthy BEAM world, a dead or dying Erlang would almost certainly spell the same for all other BEAM languages. But that’s not the only reason to support Erlang.

Erlang has done so much for Elixir (and other BEAM languages) so much so that they simply wouldn’t exist as we know them today without it. Not only that, but many Erlang VIPs including the Erlang/OTP team and Erlang’s creators Robert and Joe, have been openly supportive of Elixir on places like this very forum. In fact Robert has been on our admin team here since pretty much day one and Joe joined it just before he passed away. They have shown us a lot of love and kindness, and so it feels right that we do the same for them. Now that Elixir is off to a great start and past its most fragile stages, we can definitely turn some of our attention to helping Erlang in a similar way.

If the above isn’t good enough reason in itself, there are actually a few more things worth considering.

As many of you know Elixir is no longer seen as a new language and you could argue that we’re now way past attracting early adopters. This means we’re more likely to see different kinds of users interested in Elixir moving forward, and perhaps more importantly, we’re going to encounter different expectations - and these are likely to be ‘higher’ expectations. This is because the shift in users progresses from mostly interest based, to needs based. While it may feel like a small detail it’s an important one. Take a look at the chart we posted recently to see this in a little more detail.

Something we’re also more clearly beginning to see are the segments that Elixir has been carving out for itself - the areas where it is excelling, as well as the areas where it’s not doing quite as well. We’re seeing a lot of start-ups use Elixir for instance - which is very cool! On the flip side, we don’t see much interest from the Enterprise - which is ok too. It would be near impossible to be all things to all users and this is one of the reasons why a cohesive and cooperative BEAM community working in harmony benefits us all.

Did you know that when vending machines were first introduced soft drink manufacturers would insist machines from a competitor could not be placed on the same site? They were worried they would lose sales. One day this was put to the test, where they placed vending machines for soft drinks of two competing brands right next to each other. Can you guess what happened? Did they lose sales? Sell as many as they were when there was only one machine? Nope - sales doubled! The choice went from “do I want a can of cola?” to “which can of cola do I want?” Fascinating, right!?

Benefits to the BEAM world can go beyond that - because if there is relative equality and good and meaningful BEAM interop it could expand the potential userbase of each language by the potential userbase of each individual language.

Imagine a BEAM world where a few decades worth of Erlang users suddenly decide to use Phoenix for their web layer, or Nx for ML, or Gleam because they need static typing, or Caramel or LFE or one of the other BEAM languages because they excel in some area that is an important part of their application. We already see some of these benefits on other platforms like the JVM and we’ve seen some of it with the BEAM - but we can go much further.

We’re stronger together.

A strong, healthy Erlang brings numerous other benefits too. Some of this was covered in the forum’s launch announcement but is worth repeating here. Did you know the Erlang core team consists of almost 20 people? And that they’re all employed by Ericsson? We didn’t either!

We recently had a thread here where many were disappointed that a billion dollar company which successfully used Elixir to get to where it did, decided to leave Elixir citing one of the reasons that the language they were moving to was being developed by a large company (JetBrains). Well guess what, we have our own billion dollar company already driving the BEAM. Ericsson!

All this time it’s been sitting there right under our noses! Not only are they the type of company that the Enterprise are drawn to, but they are very heavily invested in the BEAM because they created it and because to this day it is still an important part of their business (why else would the team consist of that many people?) They want to see Erlang go from strength to strength just as much as we do, and, so, who knows where a bit of support from us might lead?

Maybe they will employ José and/or some of the Elixir team to help bring macros to Erlang (or whatever else Erlang needs to more easily use cool Elixir tech like Phoenix and Nx), or maybe they will come up with some other way to facilitate interop more easily with Elixir and other BEAM languages. There’s no telling how or where things may go - and we don’t know about you - but we’re ready and willing to find out!

The future of Erlang feels exciting. The future of the BEAM world does too. We hope you agree and we can’t wait to see where - with your help and involvement - all that leads

[tommica]

This is probably the wrong place to ask, but the more you guys spread out the community to different forums, the harder it is to keep up with all of them - is it possible for me to choose which forums would be aggregated in some kind of a personal page - for example let’s say I would want to see new posts for both elixir and Erlang forums in one place? Or is this a job for an RSS reader?

[AstonJ]

This is an easy question to answer Tommica - pick the one that feels most like ‘home’

Although it may not seem like it in the early stages, each community naturally develops its own personality - even when a new one starts with lots of members from another. Sometimes a single person can have a profound effect on the overall tone and feel of a community. This is why I am always saying everyone can make an impact - perhaps more than they may even realise.

So pick whichever one where you feel most at home and the one which you feel you can shape in a positive way

Having said that I do appreciate what you’re saying, and perhaps in the future, once the Erlang and BEAM communities get a chance to catch up to this one then the community platform I am hoping we can develop might be a good opportunity to bring the BEAM community under one roof - but that’s way ahead in the future and it may not even be possible or even advisable. Again it depends on how the communities develop including cross-community relations.

While you’re brought up the topic though I have to say I feel a little on this new forum - it’s full to the brim of super smart people, many of which are VIPs not just within the Erlang space, but dev space overall - with developers from WhatsApp, Klarna, Cisco and a whole host of others. These are some of the biggest and most well known apps and companies in the world - and it’s amazing really, that they’re all using Erlang (We really need to make more of things like this - and I hope the way we have got things like our profiles set up will help draw attention to some of these things).

[derek-zhou]

@tommica

Or is this a job for an RSS reader?

Yes, Discourse has RSS feeds. For example:
[Elixir Programming Language Forum - Latest topics]
has all the latest posts across all categories.
[Elixir News - Elixir Programming Language Forum]
has the topics in one of the categories. You can append any category with .rss postfix to get the feed.

Edit: Discourse is overly smart to mingle the links, please copy the links by right click. Clicking the above most likely will not work in your browser.

[Sebb]

@elixirforum

a dead or dying Erlang would almost certainly spell the same for all other BEAM languages.

Do we have to worry about Erlang? It’s hard to estimate how much Erlang is really used and if its use is declining because its mainly used in closed source.

[josevalim]

No. The point is that it is in our best interest to support a thriving Erlang community in itself.

Congrats on the forum, it looks great!

[tommica]

Thanks for the links, time to update my feeds!

[Rich_Morin]

Looking over the Elixir and Erlang forums, I don’t see as much support for crossover questions as I’d like. How about adding a dedicated section in the Elixir Forum for Erlang and OTP interop? Similarly, the Erlang Forum could have sections for discussions on Elixir, LFE, etc.

-r

P.S. It might also make sense to add a Fly.io section to the Elixir Forum.

[AstonJ]

Hi Rich,

Thanks for the questions!

Part of the reason why I hadn’t posted some of those interop discussions myself (despite mentioning I wanted to) was because I didn’t think it would be fair on José/the Elixir team nor the rest of the BEAM community were they to take place on an Elixir forum. Hence this was another reason why this new forum made sense - it’s much more neutral for those types of discussions.

José (Elixir), Louis (Gleam), Robert (who holds the record for number of BEAM languages ) and the Erlang team are all on the new forum and I am sure they will all get involved in those discussions as and when they feel it is appropriate to - and all on a neutral platform and without the added pressure of those conversations being inescapable or ‘in your face’. Besides, we have the new Elixir Ecosystem Chat sections now where you could start some of those conversations if you like (though they would be much more Elixir specific or from the Elixir side of things)

With regards to general crossover (and I forgot to mention this in my reply to @tommica) at some point we will have some sort of cross-community notifications - so that could be a ‘Threads on the Erlang Forums’ thread here, and a ‘Threads on the Elixir Forum’ thread there, where we’ll automatically post a link to all publicly viewable threads. That way people who make this forum their ‘home’ will be kept reasonably well updated about what’s going on there, and vice versa. But I need to see how things go first to see what makes most sense for both communities.

Btw, we have dedicated sections for all BEAM languages there, eg, erlangforums.com/lfe and should any of them become popular/busy, we’ll expand them into a larger sub-forum, like what we have here for Phoenix (elixirforum.com/phoenix)

It’s all early days tho - the exciting thing about communities like these is that you just never know where or how they end up (though of course I always hope that is somewhere good/positive)

[frigidcode]

Any chance of dark modes being available across all these forums?

[tristan]

Thanks for the great work @AstonJ and everyone else who made this happen. But I too have to continue to voice agreement with the opinion that it should be a single forum. Not only does it mean 2 separate forums to check for answers to questions and not knowing which might get you the best answer/feedback to your post, but also there has been a lot of effort by some to bridge the communities. There is Code BEAM, the Foundation and multiple library efforts, like GitHub - beam-telemetry/telemetry: Dynamic dispatching library for metrics and instrumentations. and GitHub - open-telemetry/opentelemetry-erlang: OpenTelemetry Erlang SDK – now I have two forums to monitor for questions about OpenTelemetry

[AstonJ]

Sorry about the delay in getting back to you Tristan, as you can probably imagine things have been pretty hectic. Please see below

Not only does it mean 2 separate forums to check for answers to questions and not knowing which might get you the best answer/feedback to your post

Do you mean questions you and others post? If so I would still post Elixir questions here and Erlang or other BEAM lang questions there. Most Elixir folk are here so it’s where you are most likely to get the best answer for Elixir related stuff.

We’ll also be deprecating the Erlang/BEAM sections here as they aren’t needed anymore, although people can still post Elixir related Erlang questions here if they want to and it won’t be a big deal if people post Elixir questions there (a few every now and again may even be beneficial). Most people will find that Elixir questions will have the best chance of getting answered here and Erlang or other BEAM lang questions and discussions will be better suited there.

I expect many people will check both forums at varying frequencies - so Elixir users might check this one daily and that one weekly, and vice versa for Erlang/other BEAM lang users (and as interest in either grows, so too will that frequency). This is a much more gentle easing of people into the wider Erlang community, and perhaps more importantly, it’s on everyone’s own terms. Making it choice helps avoid some potential issues and also results in a much more authentic (and therefore robust) community developing there.

I know one your concerns was that it might fail to get traction, but try not to worry, a major benefit of us managing these forums means a close-relationship can exist between them, and that means we can help with all sorts of things - such as cross-promotion. To begin with, we’ll cross-post threads from there here, and then later, when the forum and community grows, we’ll cross post threads from here there. There are lots of other benefits too, so try not worrying

…but also there has been a lot of effort by some to bridge the communities.

This is of course awesome and we’ve been doing that here from pretty much day one - like when we invited Robert to our Admin team Part of the reason why this made sense was because having a senior Erlang figure on the team would let the Erlang folk see (through them) that there weren’t any behind the scenes conspiracies against Erlang and that we were doing everything possible to be as respectful and supportive of Erlang as we could. Tbh, Robert made it practically impossible not to invite him - because he became one of our most prolific posters early on and so we just had to get him involved! This is actually a big part of why we love Erlang so much - because Erlangers like Robert have shown us so much kindness and positivity through their interactions with us. This has led to things like titles for the Erlang core team, our dedicated Erlang sections, titles for other prominent Erlangers, inviting Joe to our Admin team, cross-posting Erlang news, etc. While we have always been in awe of and grateful for Erlang, it’s the people that drive so much of that respect and admiration, particularly when it comes to things on a community level which of course this forum is most focused on

And so this new forum continues that effort, and in a way that empowers all BEAM languages too. This is especially important for Erlang and the newer/smaller BEAM languages that are currently lagging behind Elixir in terms of adoption/community.

You would be forgiven then to think that taking this forum and placing it under a BEAM banner would be the quickest way to ‘fix’ that, but it would be a temporary illusion and fraught with lots of potential problems.

On the Elixir side some people might feel it is a bit of a hostile takeover - the community here has helped build this forum over 5 years, and they have done so in no small part because they’re aware it is one of Elixir’s most notable assets. It helps in all sorts of ways and so losing it could hurt or impact Elixir negatively. This could actually end up hurting the wider BEAM world too as Elixir is currently the biggest gateway into it.

It could also result in an awkward dynamic, where people who are senior figures in the Elixir community would suddenly find they have lost quite a bit of that seniority - not through any kind of weakness, but out of respect for Erlang and the Erlang core team and community. This could have knock on effects and impact all sorts of things, varying from the small to large, from innovation to inter-community issues, personal mental health issues, etc, and all those are just their respective tips of the iceberg. Even it were to only result in impeding Elixir innovation that alone could end up being bad for everyone given Elixir is currently such a big inroad into the BEAM world.

On the Erlang side the resulting community would be very Elixir dominant - and not organically, but through what many will consider has been imposed upon them. This could dictate the path of Erlang (and to a lesser degree, other BEAM langs) in a way which would interfere with what would otherwise be their natural, independent growth and progression. That wouldn’t be doing them justice or acting in their best interest. Thus the new forum encourages a much more authentic and robust Erlang community to develop, and this also applies to the newer/smaller BEAM languages too because of the way it is set up. Yes, it will take a bit longer, but that’s ok - the resulting community will be stronger and more robust - and I feel this would be best for Erlang, Elixir, and the rest of the BEAM world. Keep in mind it took the community here 5 years to get to this point - so it’s not something that will happen overnight, but with enough dedication and hard work (and in all sorts of areas, not just on the forum) it will slowly but surely get there We are already well on the way there, with various threads that are delving deeper into interconnected topics related to things like adoption.

Having said that, are there any circumstances where such a merge might be advisable or worth doing? One that springs to mind is if Ericsson were to buy the rights to Elixir, making Elixir an official dialect of Erlang, and then employing José/the Elixir core team to work on it (like they do the Erlang core team). Btw this is just a possible scenario in which case it ‘might’ be a good idea - and I say might because even then it would need considerable thought and with the knowledge of long terms plans in mind. (Just to be clear, I have zero inside knowledge on how likely or unlikely such a scenario is.)

This is also not to say we won’t ever see the communities move closer together (that’s actually something we’ve all been doing continually over the last 5+ years and this new forum is one step closer again) but as I mentioned above Erlang first needs a little bit of time to find its feet in this modern world and work out how exactly it wants to evolve - lets give the Erlang team and everyone a little bit of breathing space and when it has done that, and has a more robust and independent community (somewhat comparable to what we have on this forum) it will be on a much more level playing field. It’s also probably worth saying that merging/moving to ’just’ another Discourse forum probably wouldn’t be worth it in most cases (other than the one described above) - we have links to threads in all sorts of places, from slides, blogs, videos, articles, books even (and we’d have to consider all the other positives that may be compromised or lost too, such as how the forum helps with SEO, the TIOBE index, how it represents the community, influences culture, the standing of Elixir VIPs, etc). Moving to a platform powered by the BEAM and where each language retains its independence and sovereignty (even if closer together or linked) would be much more interesting, beneficial and widely accepted - and that is something that would probably be most likely to happen with the new community platform we have in mind (but even that may or may not be a good idea - depending on how that platform turns out).

Ok this is getting really long and I could probably write a tonne more, but I hope this helps explain some of the reasons why, right now, an independent forum makes so much sense. Not only does this route help avoid potential problems but perhaps more importantly it’ll help a much more authentic and robust community develop there - one where everyone is invited and joins through choice and their love of and desire to help Erlang be the best it can be