Didn't their change their licensing or something and now folks are leaving? I've seen a few ex-N8n'ers coming over to the Node-RED forum, hence the question.
Yet more evidence that venture capital is basically incompatible with open source software. It's just a matter of time before any VC-backed open source company betrays its users.
n8n was never open source. They started out using Apache with a non-free clause added, but they called themselves open-source incorrectly. Then when this became more widely known, they stopped calling themselves open-source.
Why is it okay to just accept that the hyperscalers are 100% closed, but the minute a smaller player tries to play "open-ish" with "fair source" we crucify them?
Fair source is amazing. You get the code. You can modify the code. You can redistribute the code. You just can't take their business from them and compete with them head-to-head with their product. You can reformulate it into something else, but you can't take their labor and cut their knees off with it. You have literally every other freedom.
Why in the hell are so many people against that?
Fair source is sustainable and equitable. You get everything except for that one little right with which you could compete with them.
Everyone gushes over Obsidian - that's not even "source available" or "fair source". It's completely closed.
I bet if Obsidian went "fair source", and you could download the code and compile it yourself, there would be hundreds of voices crawling out of the woodwork to call them the devil for not being pure OSI-approved open. How dare they keep one little freedom to themselves for all the hard work they've done?
I'm half convinced the anti-"fair source" voices are deep industry folks who want big tech to own the OSI definition. Who can use it strategically as they sit atop hundreds of billions of dollars in cash flows. Open source to the FAANGMAGMA Gods is just a way of outsourcing labor and dangling trinkets in front of underpaid labor.
n8n isn't doing a "VC rug pull." They're trying to be sustainable. The only thing that's been "pulled" is the wool over the eyes of every engineer satisfied with FAANG OSI-approved serfdom. Some of the pieces are "open" all right, but they own pretty much everything the sunlight touches.
If n8n can't build a business then it just becomes some side project hobby. What exactly do we think every other company is doing?
Please be more pragmatic and realize that "fair source" is sustainable. It rewards the innovators and you get almost everything for free. You just can't shoot them with their own gun.
Do you not want the ability to download the source and potentially tweak the software? To have a copy you can hold onto forever? To be able to analyze it for telemetry. To be able to potentially submit patches (if you're feeling generous)?
Don't you want them to make money? Rather than beg for Github donation scraps?
Be especially mindful if you're typing a retort on a Macbook Pro or iPhone and deploying software to GKE or AWS.
I care that software I depend upon survives long term. In consequence, I care to some extent that the principal company or organization building that software also survives... But that finite concern is balanced against the simultaneous concern that the software itself can survive the organization's potential dissolution. I don't want to have to make a sudden, stressful, expensive migration because corporate buy-outs, bankruptcies, or mergers result in rug-pulls for software I use.
"Fair" source does absolutely nothing to allievate those concerns. The four software freedoms do.
It is true that for a quarter century software developers have naively licensed their projects under MIT and BSD style permissive licenses, and have then felt robbed when big companies come in and eat their lunch. Except it was never their lunch because they didn't actually use licenses that would have asserted their rights to that lunch.
Thankfully, that naivete is finally, slowly dying but, rather than use robust solutions that have been around for decades (e.g. GPL-or-AGPL/proprietary dual license), some developers have invented a new, largely untested concept and named it by dressing up their own sour grapes and spite as "fairness".
"Fair" source software is a non-solution in search of a problem.
Free software and strong copyleft licenses already exist. Dual-licensing already exists. Viable, profitable, healthy, ethical companies built on these strategies already exist.
I used to think Stallman was a crank and a fundamentalist, and he absolutely is, and thank goodness for that. I now think "open-source" is exactly as diluted as free software advocates initially pointed out, and the emergence of "fair" source is part of the damage it has done.
Authors using Fair want to share their code while getting protections for themselves. Strong copyleft doesn't care about authors and is all about protecting the end users.
So Fair fullfills an actual need or desire not covered elsewhere, thus is not a non-solution.
It might be not the appropriate or the best solution to solve the exact concerns of people using it, that's debatable and a different topic, akin to using the wrong tool for the job. Strong copyleft is the wrong tool, too; obviously competitors can just deploy without modifications and offer it as a service.
> Authors using Fair want to share their code while getting protections for themselves. Strong copyleft doesn't care about authors and is all about protecting the end users.
This is a one-sided assertion of fairness, and therefore an abuse of the concept. Free software offers the same rights to both authors and users. That is fairness.
If you want to reserve rights to yourself that are withheld from end-users, that's fine. Arguably even still within a broader conceptual realm of fairness. But naming your personally preferred arrangement of rights "fair" and in so doing implying most or all other arrangements are unfair is just plain arrogant.
Strong copyleft cares equally about authors and end-users. It doesn't disregard authors. Some authors just want to co-opt the general notion of fairness to mean their own licensing preferences.
this is a solid argument against the open-source purists that I've never heard before and really appreciate. Many people would like to make money on the things they spend their time building. Some percentage of those people also love FOSS. Fair source is the middle ground where they can share their work, open it up for criticism, issues, and fixes, but ensure that their business is secure. It's a solid middle ground that shouldn't be demonized just because it isn't open source.
I don't know/care about "fair source," but the "open source" label in particular has ALWAYS had a well-known and specific meaning in the software world. Attaching it to a proprietary product for the marketing and social media good feels is actively deceitful, no matter if the company is USA trillion-dollar Big Tech or a single developer from a marginalized demographic in a developing country.
Software authors have the right to choose whatever license they wish for the project/product, just don't try to lie to users about what it really is.
The person I was replying to seemed to be under the impression that n8n used to be open source and now is not. This is not the case, so I pointed that out.
Did you have anything to say about my actual comment, or are you just attaching your rant to a random part of the thread?
+1 for Node-RED. If we've learned anything from elasticsearch/redis/bitnami/and dozens of others, it should be "don't build important things on code that isn't enshittification-resistant"
Those guys did all the work and AWS and Google get to collect a cool hundred million each quarter for doing next to nothing. And the original authors can't compete with that.
Are we all gonna go sign up to "Enterprise Redis Cloud" to support them? I don't think so. They have nothing left. They got picked dry.
If they're preserved a "no managed offerings" freedom for themselves, they could collect a few hundred million from each hyperscaler each quarter and put that directly into product. And their engineers would be extra nicely compensated.
But that's not how this story plays out. Big tech just takes things and finds out how to monetize them in ways they don't have to give back.
And we've been trained (by big tech?) to yell at the little guys that try to carve out a space for themselves.
Both are billion dollar companies, we as individuals have nothing in common with them. Enshittification happens due to market conditions that apply to small and large companies alike. Redis and elasticsearch aren't underdogs fighting for the little guy, they are just a smaller scale version of the same shit.
I'd rather have a software commons and have tech be owned by the workers and not soul-sucking corporations, no matter the size.
I would recommend Node-RED but that's a lot more work than a Zapier or N8n but you can do more with it - Node-RED abstraction level is closer to a programming language than a SaaS combinatorial tool.
Depends on what you want to achieve and how much effort you want to apply.
The other licensing shoe is going to drop.
And when it does, it will be same as always. Important features will be cut of from “community” edition, like admin for minio, and pricing will be predatory for full product, like Taipy. Seriously, screw
Taipy. Got quoted 200k for an open license if I wanted to self host it, with no limit on number users.
I got hit by the Minio admin change in the console when I upgraded my installation recently, and I found https://github.com/huncrys/minio-console which adds it all back in. It works as expected so far.
I find this kind of rug pull behavior so hostile I will be looking to replace Minio as soon as possible in my homelab. To be clear, I would pay for a license if the prices weren't impossible to afford as an invidual who uses Minio for non-business reasons.
When I set up n8n last year I was thoroughly confused at the lack of state management. I ended up storing as JSON in a remote blob “database”, a horrible hack.
Node-RED has three state access levels: global one across all flows, a flow based state which is only applicable to one flow and node-based state/storage that is linked to single node.
Any node can access global state, only nodes contained in the flow can access flow state and nodes that have their own state can solely access that storage.
It looks to me like n8n has suddenly taken over to the point where Automation is now almost synonymous with n8n for most people.
As someone who builds workflows using my own agent system which is based on checklists, subtasks and tool calling with new features or tools based on plugins in Python, it is harder and harder to find an "AI Automation" project where people haven't predetermined that I have to use n8n.
It's ridiculous.
I actually think that defaulting to creating workflows in raw code is not an ideal outcome though because it feels inaccessible to non-programmers.
But I think within X months there will be a lot of people who find out how bad the licensing issues are with n8n and migrate to something similar to my system where workflows are run by agents that have a delegate_subtask command or commands to manage checklists etc. Because most of the workflows can be managed easily by strong models and just described in natural language if the agents have the right tool commands and the system has a scheduling/trigger system.
But then give it another X months or a year or so later and many will start using general purpose computer use agents that they just treat quite similarly to human employees. Because one of the biggest gaps regardless of how you do it is with the inconvenience of setting up OAuth 2 and the operational and bottom line issue of running all of your API requests through some centralized service like n8n.
So we will see people who have agent systems like myself (mindroot on GitHub) start building in computer and browser use capabilities and recipes for accessing websites and creating API keys etc. for their users.
Also there inevitably is going to be something along the lines of OAuth or similar that will allow agents to sign up for services and create credentials on behalf of users to solve this type of problem.
But one of the big advantages n8n has with users right now is that they have OAuth set up with literally everything.
> Also there inevitably is going to be something along the lines of OAuth or similar that will allow agents to sign up for services and create credentials on behalf of users to solve this type of problem.
I agree with you, but the real solution to this is an API.
On the other hand LangGraph and LangChain are an absolute pain to use and you end up spending most of your effort in boilerplate hell if you want to stitch together anything useful
I recently had to build a production-ready workflow in N8N - it ended up being a spaghetti flow of custom code nodes and custom http requests (because none of the provided connectors did exactly what we needed) that I was left wondering if this wouldn't be easier to code up in Cursor.
I already felt the need to persist data between different nodes and workflows, so I made a bespoke API with dedicated endpoints that perform CRUD tasks (Yes, it defeats the purpose of a no-code platform but I can code, so...).
I have to say I love N8n so far, it is a great productivity tool.
They fit different niches, IME; node-red is designed for IoT workloads, and so is a great fit for high volume messaging; n8n on the other hand is more workflow automation focused -like Zapier - and so has higher level abstractions and is less focused on performance efficiency.
Can they both fit some of the same use cases? Definitely.
What is the point of this... platform? Also you can't scroll down to the footer because some wonderfully experienced designer (or PM or even engineer) decided the infinite scroll list was a good idea.
it'll be interesting to see what happens with n8n over the next few years.
On one hand, a ton of people are using it to build AI workflows.
On the other, it's never been easier to build a similar workflow in Python with Claude Code or other tools where you get infinite flexibility, version control, and more flexibility on hosting.
We're betting on the later at https://www.definite.app/. Our agent writes pure Python with pre-built integrations into stuff like Hubspot, Salesforce, Attio, Stripe, Postgres, etc.
That pricing model is insane and confusing as hell. Your FAQ clears it up slightly but was hard pass from me almost instantly due to the conflicting information in the pricing section.
Didn't their change their licensing or something and now folks are leaving? I've seen a few ex-N8n'ers coming over to the Node-RED forum, hence the question.
You can also try https://github.com/autokitteh/autokitteh. Python based, completely open source. Got inter workflow storage as well
Hadn't heard of Node-RED but this is really cool. Thanks for mentioning it.
Node-RED is cool but I didn't realise it could do the AI workflows that n8n is aiming for?
For the AI workflows, we (windmill.dev) added AI Agent steps as first class primitives very recently.
Node red is better (imo) but harder for non technical users to use
Yet more evidence that venture capital is basically incompatible with open source software. It's just a matter of time before any VC-backed open source company betrays its users.
n8n was never open source. They started out using Apache with a non-free clause added, but they called themselves open-source incorrectly. Then when this became more widely known, they stopped calling themselves open-source.
Why is it okay to just accept that the hyperscalers are 100% closed, but the minute a smaller player tries to play "open-ish" with "fair source" we crucify them?
Fair source is amazing. You get the code. You can modify the code. You can redistribute the code. You just can't take their business from them and compete with them head-to-head with their product. You can reformulate it into something else, but you can't take their labor and cut their knees off with it. You have literally every other freedom.
Why in the hell are so many people against that?
Fair source is sustainable and equitable. You get everything except for that one little right with which you could compete with them.
Everyone gushes over Obsidian - that's not even "source available" or "fair source". It's completely closed.
I bet if Obsidian went "fair source", and you could download the code and compile it yourself, there would be hundreds of voices crawling out of the woodwork to call them the devil for not being pure OSI-approved open. How dare they keep one little freedom to themselves for all the hard work they've done?
I'm half convinced the anti-"fair source" voices are deep industry folks who want big tech to own the OSI definition. Who can use it strategically as they sit atop hundreds of billions of dollars in cash flows. Open source to the FAANGMAGMA Gods is just a way of outsourcing labor and dangling trinkets in front of underpaid labor.
n8n isn't doing a "VC rug pull." They're trying to be sustainable. The only thing that's been "pulled" is the wool over the eyes of every engineer satisfied with FAANG OSI-approved serfdom. Some of the pieces are "open" all right, but they own pretty much everything the sunlight touches.
If n8n can't build a business then it just becomes some side project hobby. What exactly do we think every other company is doing?
Please be more pragmatic and realize that "fair source" is sustainable. It rewards the innovators and you get almost everything for free. You just can't shoot them with their own gun.
Do you not want the ability to download the source and potentially tweak the software? To have a copy you can hold onto forever? To be able to analyze it for telemetry. To be able to potentially submit patches (if you're feeling generous)?
Don't you want them to make money? Rather than beg for Github donation scraps?
Be especially mindful if you're typing a retort on a Macbook Pro or iPhone and deploying software to GKE or AWS.
---
https://faircode.io/
https://fair.io/
I care that software I depend upon survives long term. In consequence, I care to some extent that the principal company or organization building that software also survives... But that finite concern is balanced against the simultaneous concern that the software itself can survive the organization's potential dissolution. I don't want to have to make a sudden, stressful, expensive migration because corporate buy-outs, bankruptcies, or mergers result in rug-pulls for software I use.
"Fair" source does absolutely nothing to allievate those concerns. The four software freedoms do.
It is true that for a quarter century software developers have naively licensed their projects under MIT and BSD style permissive licenses, and have then felt robbed when big companies come in and eat their lunch. Except it was never their lunch because they didn't actually use licenses that would have asserted their rights to that lunch.
Thankfully, that naivete is finally, slowly dying but, rather than use robust solutions that have been around for decades (e.g. GPL-or-AGPL/proprietary dual license), some developers have invented a new, largely untested concept and named it by dressing up their own sour grapes and spite as "fairness".
"Fair" source software is a non-solution in search of a problem.
Free software and strong copyleft licenses already exist. Dual-licensing already exists. Viable, profitable, healthy, ethical companies built on these strategies already exist.
I used to think Stallman was a crank and a fundamentalist, and he absolutely is, and thank goodness for that. I now think "open-source" is exactly as diluted as free software advocates initially pointed out, and the emergence of "fair" source is part of the damage it has done.
Authors using Fair want to share their code while getting protections for themselves. Strong copyleft doesn't care about authors and is all about protecting the end users.
So Fair fullfills an actual need or desire not covered elsewhere, thus is not a non-solution.
It might be not the appropriate or the best solution to solve the exact concerns of people using it, that's debatable and a different topic, akin to using the wrong tool for the job. Strong copyleft is the wrong tool, too; obviously competitors can just deploy without modifications and offer it as a service.
> Authors using Fair want to share their code while getting protections for themselves. Strong copyleft doesn't care about authors and is all about protecting the end users.
This is a one-sided assertion of fairness, and therefore an abuse of the concept. Free software offers the same rights to both authors and users. That is fairness.
If you want to reserve rights to yourself that are withheld from end-users, that's fine. Arguably even still within a broader conceptual realm of fairness. But naming your personally preferred arrangement of rights "fair" and in so doing implying most or all other arrangements are unfair is just plain arrogant.
Strong copyleft cares equally about authors and end-users. It doesn't disregard authors. Some authors just want to co-opt the general notion of fairness to mean their own licensing preferences.
Call it a non-compete license and be forthright.
this is a solid argument against the open-source purists that I've never heard before and really appreciate. Many people would like to make money on the things they spend their time building. Some percentage of those people also love FOSS. Fair source is the middle ground where they can share their work, open it up for criticism, issues, and fixes, but ensure that their business is secure. It's a solid middle ground that shouldn't be demonized just because it isn't open source.
I don't know/care about "fair source," but the "open source" label in particular has ALWAYS had a well-known and specific meaning in the software world. Attaching it to a proprietary product for the marketing and social media good feels is actively deceitful, no matter if the company is USA trillion-dollar Big Tech or a single developer from a marginalized demographic in a developing country.
Software authors have the right to choose whatever license they wish for the project/product, just don't try to lie to users about what it really is.
"Open" AI, "Open" Router, and the like typically get a pass.
n8n has never called themselves "open". Their first commit:
https://github.com/n8n-io/n8n/blob/9cb9804eeec1576d935817ecd...
There's so much misinformation and animosity about fair source licenses.
> n8n has never called themselves "open"
Trivially debunked by looking at their homepage in 2019: https://web.archive.org/web/20191009000026/https://n8n.io/
The person I was replying to seemed to be under the impression that n8n used to be open source and now is not. This is not the case, so I pointed that out.
Did you have anything to say about my actual comment, or are you just attaching your rant to a random part of the thread?
+1 for Node-RED. If we've learned anything from elasticsearch/redis/bitnami/and dozens of others, it should be "don't build important things on code that isn't enshittification-resistant"
What did elasticsearch/redis do so wrong?
AWS and Google stole their only revenue source.
Those guys did all the work and AWS and Google get to collect a cool hundred million each quarter for doing next to nothing. And the original authors can't compete with that.
Are we all gonna go sign up to "Enterprise Redis Cloud" to support them? I don't think so. They have nothing left. They got picked dry.
If they're preserved a "no managed offerings" freedom for themselves, they could collect a few hundred million from each hyperscaler each quarter and put that directly into product. And their engineers would be extra nicely compensated.
But that's not how this story plays out. Big tech just takes things and finds out how to monetize them in ways they don't have to give back.
And we've been trained (by big tech?) to yell at the little guys that try to carve out a space for themselves.
Both are billion dollar companies, we as individuals have nothing in common with them. Enshittification happens due to market conditions that apply to small and large companies alike. Redis and elasticsearch aren't underdogs fighting for the little guy, they are just a smaller scale version of the same shit.
I'd rather have a software commons and have tech be owned by the workers and not soul-sucking corporations, no matter the size.
Try activepieces, they never did any bait and switch.
https://www.activepieces.com/
Looks like they have tables too
Keep to know how people are using it inspite of the licensing issues. Langflow seems good?
Langflow was more painful than a wisdom tooth extraction to use. This was back in early 2025 tho
I would recommend Node-RED but that's a lot more work than a Zapier or N8n but you can do more with it - Node-RED abstraction level is closer to a programming language than a SaaS combinatorial tool.
Depends on what you want to achieve and how much effort you want to apply.
ActivePieces is also an option
The other licensing shoe is going to drop. And when it does, it will be same as always. Important features will be cut of from “community” edition, like admin for minio, and pricing will be predatory for full product, like Taipy. Seriously, screw Taipy. Got quoted 200k for an open license if I wanted to self host it, with no limit on number users.
I got hit by the Minio admin change in the console when I upgraded my installation recently, and I found https://github.com/huncrys/minio-console which adds it all back in. It works as expected so far.
I find this kind of rug pull behavior so hostile I will be looking to replace Minio as soon as possible in my homelab. To be clear, I would pay for a license if the prices weren't impossible to afford as an invidual who uses Minio for non-business reasons.
Great to see this land.
When I set up n8n last year I was thoroughly confused at the lack of state management. I ended up storing as JSON in a remote blob “database”, a horrible hack.
Node-RED has three state access levels: global one across all flows, a flow based state which is only applicable to one flow and node-based state/storage that is linked to single node.
Any node can access global state, only nodes contained in the flow can access flow state and nodes that have their own state can solely access that storage.
What percent of agent devs are using tools like this versus writing code to define their systems?
Seems like a lot, and I find it surprising that people would build real world workflows on platforms like this.
It looks to me like n8n has suddenly taken over to the point where Automation is now almost synonymous with n8n for most people.
As someone who builds workflows using my own agent system which is based on checklists, subtasks and tool calling with new features or tools based on plugins in Python, it is harder and harder to find an "AI Automation" project where people haven't predetermined that I have to use n8n.
It's ridiculous.
I actually think that defaulting to creating workflows in raw code is not an ideal outcome though because it feels inaccessible to non-programmers.
But I think within X months there will be a lot of people who find out how bad the licensing issues are with n8n and migrate to something similar to my system where workflows are run by agents that have a delegate_subtask command or commands to manage checklists etc. Because most of the workflows can be managed easily by strong models and just described in natural language if the agents have the right tool commands and the system has a scheduling/trigger system.
But then give it another X months or a year or so later and many will start using general purpose computer use agents that they just treat quite similarly to human employees. Because one of the biggest gaps regardless of how you do it is with the inconvenience of setting up OAuth 2 and the operational and bottom line issue of running all of your API requests through some centralized service like n8n.
So we will see people who have agent systems like myself (mindroot on GitHub) start building in computer and browser use capabilities and recipes for accessing websites and creating API keys etc. for their users.
Also there inevitably is going to be something along the lines of OAuth or similar that will allow agents to sign up for services and create credentials on behalf of users to solve this type of problem.
But one of the big advantages n8n has with users right now is that they have OAuth set up with literally everything.
> Also there inevitably is going to be something along the lines of OAuth or similar that will allow agents to sign up for services and create credentials on behalf of users to solve this type of problem.
I agree with you, but the real solution to this is an API.
On the other hand LangGraph and LangChain are an absolute pain to use and you end up spending most of your effort in boilerplate hell if you want to stitch together anything useful
I recently had to build a production-ready workflow in N8N - it ended up being a spaghetti flow of custom code nodes and custom http requests (because none of the provided connectors did exactly what we needed) that I was left wondering if this wouldn't be easier to code up in Cursor.
I don’t get all the n8n hype
I already felt the need to persist data between different nodes and workflows, so I made a bespoke API with dedicated endpoints that perform CRUD tasks (Yes, it defeats the purpose of a no-code platform but I can code, so...). I have to say I love N8n so far, it is a great productivity tool.
I love Node-RED instead of n8n. But it's biggest problem is that does not have the concept of an "execution". Which sucks.
They fit different niches, IME; node-red is designed for IoT workloads, and so is a great fit for high volume messaging; n8n on the other hand is more workflow automation focused -like Zapier - and so has higher level abstractions and is less focused on performance efficiency.
Can they both fit some of the same use cases? Definitely.
Slightly off topic question , but do you pronounce it n eight n, or Nathan ? I got blank stares when saying Nathan from my colleagues :')
It's awesome for quick projects, but that 50MB limit makes me think I'll just end up migrating to Supabase or Airtable down the line anyway.
Also, take a look at windmill.dev. It’s a beast
I think windmill got the entities right (script, flow, app). The last I tried it was a bit more complex to setup than n8n.
We’ve similarly had a lot of success with self-hosting Windmill.
What is the point of this... platform? Also you can't scroll down to the footer because some wonderfully experienced designer (or PM or even engineer) decided the infinite scroll list was a good idea.
Classic.
You're looking at the community forum, which uses Discourse (https://www.discourse.org/).
Check out the n8n home page (https://n8n.io/), which is actually great at clearly communicating the point of the platform.
The n8n space has been crowded for a long time - Tray io, Zapier Enterprise, Workato, Make, UIPath, and many others
N8n seems to be standing out thanks to their open source roots (and not deliberately hampering the OSS variant too much).
DataTables will make a big difference again - being able to store some level of "state" has been a need in almost every workflow I've built.
n8n doesn’t have open-source roots. It’s never been open-source and doesn’t have an OSS variant.
It’s smart for N8N to add DataTables. Had to of been the biggest dev need.
it'll be interesting to see what happens with n8n over the next few years.
On one hand, a ton of people are using it to build AI workflows.
On the other, it's never been easier to build a similar workflow in Python with Claude Code or other tools where you get infinite flexibility, version control, and more flexibility on hosting.
We're betting on the later at https://www.definite.app/. Our agent writes pure Python with pre-built integrations into stuff like Hubspot, Salesforce, Attio, Stripe, Postgres, etc.
Might be easy to build, but impossible to maintain.
That pricing model is insane and confusing as hell. Your FAQ clears it up slightly but was hard pass from me almost instantly due to the conflicting information in the pricing section.
Yeah, we're working on it. I'd prefer to by compute hours, but it's hard for people to know how many hours they'll need.