4 ms·

Anyone know how this compares to Huginn? I've built out some self hosted automation on Huginn and it works great once it's set up, but it's just a bit janky an

Anyone know how this compares to Huginn? I've built out some self hosted automation on Huginn and it works great once it's set up, but it's just a bit janky and difficult to use. And getting initial flows set up can be frustrating as the diagnostics aren't always great.

tonyhb 3y ago
RE. initial flows and diagnostics, I totally feel your pain. Part of the trouble with these systems is that they're too different from our average SDLC we have every day: - Write regular code - Test locally - Then deploy This is one of the key aspects we're fixing at https://www.inngest.com https://www.inngest.com — you can write regular TS (or Go, or Elixir) code as a function, create simple workflows, test locally, and have your functions automatically triggered by events. I still think there's room to go with improving automations here, and it's going to be an interesting few years as things develop.
- cpursley 3y ago
  Elixir, really? That sounds amazing. So many of these tools are TS or Python only.
  - tonyhb 3y ago
    https://github.com/darwin67/ex_inngest https://github.com/darwin67/ex_inngest. We'll officially support this soon — it's written by one of our engineering team. The SDKs are lightweight, work in any language (we only support a few right now), and you can live-migrate long running functions across languages at any time. Long running state is automatically transferred across those SDKs. One of our foundational aspects is to make long running state, queues, events really easy, and that means we need to make multi-cloud and multi-lang migrations easy, too.
    - thelastbender12 3y ago
      Transferring state across sdks sound very cool, how do you achieve that? With all the workflows as code frameworks (like say, Temporal), this is what I imagine to be a shortcoming. With regular services + state in database implementations, you can query the database separately. While with a temporal workflow, you can't cleanly figure out where a particular workflow run is atm since the state is an append only stream of events?
  - giovannibonetti 3y ago
    For Elixir, you might be interested in Broadway [1], described in their website as "Build concurrent and multi-stage data ingestion and data processing pipelines with Elixir." It is maintained by Jose Valim's consultancy company, so it should be pretty good. I'm looking forward to using it in a project at work soon. [1] https://elixir-broadway.org/ https://elixir-broadway.org/
- fnikacevic 3y ago
  Sorry for the loaded question, but can you share examples of companies using inngest in production and roughly what scale? Your product appears to fill the exact gaps in Supabase's current offering.
  - tonyhb 3y ago
    Yeah, for sure. Individual companies run tens of millions of functions a month across anything as basic as braze-like customer lifecycle emails (https://www.inngest.com/blog/lifecycle-emails-with-resend https://www.inngest.com/blog/lifecycle-emails-with-resend) to running complex AI workflows at scale. The basic premise is that you can send Inngest events (for free) then start writing functions hosted in your own API at any time, with local testing, branch deploys, etc.
iamwil 3y ago
What do you usually automate? I had some before when I was working with a friend, but now I struggle to find stuff for myself.
- BaseballPhysics 3y ago
  A few things. My favourite is a flow to pull HTML-formatted newsletters out of my email and push them to Wallabag. I then use Wallabag's RSS feature with Calibre to turn them into an ebook I can read on my Kindle. My other major use case is centralized notifications. For a lot of things (e.g. backups), I have a webhooks in Huginn that then push out a notification via Gotify to my phone. I've also used Huginn to scrub/clean up RSS feeds, since it can pull the feed, apply transforms to it, and then republish it.
  - gettodachoppa 3y ago
    This sounds fine but what is this doing that a bash script + cronjob doesn't? Isn't all of this functionality available in CLI tools (and in fact you're likely already using them, e.g. making calibre convert the HTML newsletter to .mobi). I ask because I'm also curious what I can use these sort of tools for. I installed Node Red before and other than visual smart home automation, was at a loss of ideas. As someone who can write bash scripts (and nowadays ChatGPT does all the scripting for me), how can they make my life easier?
    - BaseballPhysics 3y ago
      I think you're underestimating the effort involved in writing something to poll an IMAP folder, checking for new emails, pull the right mime attachments, etc, etc, and then turning around and post them to the right place. Yeah I could write some Ruby or Python or Perl to do the job but I can also spend way less time and use something like Huginn to do it for me. Honestly, your argument could apply to virtually anything you might do with something like Zapier or equivalent. Nothing they do can't be done via scripting. The point is these tools simplify the grunt work so you can go from idea to implementation much faster, with proven code that you don't have to write and maintain yourself.
nylonstrung 3y ago
This is an straightforward alternative to Make.com or Zapier vs Hugginn has AI agent stuff
rubenfiszel 3y ago
windmill.dev is a modern take on Huginn: optimal performance, great DX, scale to any cluster, work with python, typescript, go, bash, and flows that are more powerful. n8n is great because their UX is simple and they have many well-made integrations, including an automatic way of setting up the webhooks in your github account for instance to trigger automation on changes to your repo. Windmill is less about having many integrations, and more about being able to write real code-based workflows, running on your infra and with potentially long-running jobs that you can assign easily to bigger nodes. It's a compromise between n8n and temporal for the workflow part. Anyway, n8n was a precursor in many aspects and a great inspiration for all of us building workflow engines.
- c0brac0bra 3y ago
  I'm loving windmill currently. I've been able to automated all sorts software painful manual stuff with it. For a developer, it seems like a great low code tool with the right flexibility to write whatever script you need.
- robertlagrant 3y ago
  Note: previous commenter is a windmill dev, I think.
input_sh 3y ago
I haven't tried Huginn in a while, but from my experience n8n can get a bit janky at times as well. I feel like the sweet spot it hits is neither developer- nor newbie-friedly, but somewhere in between where it's equally bad for both target groups. Workflows get out of hand very quickly, making me need more blocks than lines of Python I'd have to write to achieve the same thing. You need to really understand their logic gates to do stuff, and the docs you'll be relying on range from not great to very bad, incomplete, or just plain incorrect. You'll have to rely on their (Discourse) community forum quite a bit. If you're one of those people that are strict about open source definitions (I'm personally not one of them), then it's not open source, as you'd need the enterprise license to run it in production for commercial purposes. But, to make it worse, you'll reach some of those enterprise features in your personal projects as well: global variables, Git version control, bash scripts, custom nodes, workflow history (so that you can revert if you mess something up)... all behind an unobtainable license for personal use. With that out of the way, it is the closest thing to what I want out of such a tool, so I keep on using it despite its flaws. Managing credentials is very easy, you can make custom HTTP requests, sharing and importing workflows is very straightforward, debugging them is easy (most of the time), it's easy to self-host using Docker, it supports many of the APIs I want to use that other similar tools don't (okay some of the cloud ones do, but they're pretty unobtainable . Overall, I'd say 6/10? It's good, but it could be so much better.
- cutemonster 3y ago
  You were describing n8n not Huginn, right? (Just to avoid misunderstandings. "It" in "sweet spot it hits..." is not crystal clear)
  - input_sh 3y ago
    Yes, I'm talking about n8n, sorry for the confusion!

tonyhb 3y ago

RE. initial flows and diagnostics, I totally feel your pain. Part of the trouble with these systems is that they're too different from our average SDLC we have every day: - Write regular code - Test locally - Then deploy This is one of the key aspects we're fixing at https://www.inngest.com https://www.inngest.com — you can write regular TS (or Go, or Elixir) code as a function, create simple workflows, test locally, and have your functions automatically triggered by events. I still think there's room to go with improving automations here, and it's going to be an interesting few years as things develop.

cpursley 3y ago
Elixir, really? That sounds amazing. So many of these tools are TS or Python only.
- tonyhb 3y ago
  https://github.com/darwin67/ex_inngest https://github.com/darwin67/ex_inngest. We'll officially support this soon — it's written by one of our engineering team. The SDKs are lightweight, work in any language (we only support a few right now), and you can live-migrate long running functions across languages at any time. Long running state is automatically transferred across those SDKs. One of our foundational aspects is to make long running state, queues, events really easy, and that means we need to make multi-cloud and multi-lang migrations easy, too.
  - thelastbender12 3y ago
    Transferring state across sdks sound very cool, how do you achieve that? With all the workflows as code frameworks (like say, Temporal), this is what I imagine to be a shortcoming. With regular services + state in database implementations, you can query the database separately. While with a temporal workflow, you can't cleanly figure out where a particular workflow run is atm since the state is an append only stream of events?
- giovannibonetti 3y ago
  For Elixir, you might be interested in Broadway [1], described in their website as "Build concurrent and multi-stage data ingestion and data processing pipelines with Elixir." It is maintained by Jose Valim's consultancy company, so it should be pretty good. I'm looking forward to using it in a project at work soon. [1] https://elixir-broadway.org/ https://elixir-broadway.org/
fnikacevic 3y ago
Sorry for the loaded question, but can you share examples of companies using inngest in production and roughly what scale? Your product appears to fill the exact gaps in Supabase's current offering.
- tonyhb 3y ago
  Yeah, for sure. Individual companies run tens of millions of functions a month across anything as basic as braze-like customer lifecycle emails (https://www.inngest.com/blog/lifecycle-emails-with-resend https://www.inngest.com/blog/lifecycle-emails-with-resend) to running complex AI workflows at scale. The basic premise is that you can send Inngest events (for free) then start writing functions hosted in your own API at any time, with local testing, branch deploys, etc.

iamwil 3y ago

What do you usually automate? I had some before when I was working with a friend, but now I struggle to find stuff for myself.

BaseballPhysics 3y ago
A few things. My favourite is a flow to pull HTML-formatted newsletters out of my email and push them to Wallabag. I then use Wallabag's RSS feature with Calibre to turn them into an ebook I can read on my Kindle. My other major use case is centralized notifications. For a lot of things (e.g. backups), I have a webhooks in Huginn that then push out a notification via Gotify to my phone. I've also used Huginn to scrub/clean up RSS feeds, since it can pull the feed, apply transforms to it, and then republish it.
- gettodachoppa 3y ago
  This sounds fine but what is this doing that a bash script + cronjob doesn't? Isn't all of this functionality available in CLI tools (and in fact you're likely already using them, e.g. making calibre convert the HTML newsletter to .mobi). I ask because I'm also curious what I can use these sort of tools for. I installed Node Red before and other than visual smart home automation, was at a loss of ideas. As someone who can write bash scripts (and nowadays ChatGPT does all the scripting for me), how can they make my life easier?
  - BaseballPhysics 3y ago
    I think you're underestimating the effort involved in writing something to poll an IMAP folder, checking for new emails, pull the right mime attachments, etc, etc, and then turning around and post them to the right place. Yeah I could write some Ruby or Python or Perl to do the job but I can also spend way less time and use something like Huginn to do it for me. Honestly, your argument could apply to virtually anything you might do with something like Zapier or equivalent. Nothing they do can't be done via scripting. The point is these tools simplify the grunt work so you can go from idea to implementation much faster, with proven code that you don't have to write and maintain yourself.

nylonstrung 3y ago

This is an straightforward alternative to Make.com or Zapier vs Hugginn has AI agent stuff

rubenfiszel 3y ago

windmill.dev is a modern take on Huginn: optimal performance, great DX, scale to any cluster, work with python, typescript, go, bash, and flows that are more powerful. n8n is great because their UX is simple and they have many well-made integrations, including an automatic way of setting up the webhooks in your github account for instance to trigger automation on changes to your repo. Windmill is less about having many integrations, and more about being able to write real code-based workflows, running on your infra and with potentially long-running jobs that you can assign easily to bigger nodes. It's a compromise between n8n and temporal for the workflow part. Anyway, n8n was a precursor in many aspects and a great inspiration for all of us building workflow engines.

c0brac0bra 3y ago
I'm loving windmill currently. I've been able to automated all sorts software painful manual stuff with it. For a developer, it seems like a great low code tool with the right flexibility to write whatever script you need.
robertlagrant 3y ago
Note: previous commenter is a windmill dev, I think.

input_sh 3y ago

I haven't tried Huginn in a while, but from my experience n8n can get a bit janky at times as well. I feel like the sweet spot it hits is neither developer- nor newbie-friedly, but somewhere in between where it's equally bad for both target groups. Workflows get out of hand very quickly, making me need more blocks than lines of Python I'd have to write to achieve the same thing. You need to really understand their logic gates to do stuff, and the docs you'll be relying on range from not great to very bad, incomplete, or just plain incorrect. You'll have to rely on their (Discourse) community forum quite a bit. If you're one of those people that are strict about open source definitions (I'm personally not one of them), then it's not open source, as you'd need the enterprise license to run it in production for commercial purposes. But, to make it worse, you'll reach some of those enterprise features in your personal projects as well: global variables, Git version control, bash scripts, custom nodes, workflow history (so that you can revert if you mess something up)... all behind an unobtainable license for personal use. With that out of the way, it is the closest thing to what I want out of such a tool, so I keep on using it despite its flaws. Managing credentials is very easy, you can make custom HTTP requests, sharing and importing workflows is very straightforward, debugging them is easy (most of the time), it's easy to self-host using Docker, it supports many of the APIs I want to use that other similar tools don't (okay some of the cloud ones do, but they're pretty unobtainable . Overall, I'd say 6/10? It's good, but it could be so much better.

cutemonster 3y ago
You were describing n8n not Huginn, right? (Just to avoid misunderstandings. "It" in "sweet spot it hits..." is not crystal clear)
- input_sh 3y ago
  Yes, I'm talking about n8n, sorry for the confusion!