@UnfairMode Yes gemini openai use outdated information, but you can try use ChatGPT - n8n Docs Assistant which check the current docs, works alot better. I always cross reference the doc sites further if am unsure.
I did check your profile but I donāt see the post yet, but yes please do create a separate post in questions, this thread is different. Have a great day.
Samuel
@King_Samuel_David Sorry for the delay in responding.
As I recently created my account, I was waiting for it to come out of HOLD status so I could create a post. I had been having difficulties with the Adobe API and ConvertAPI for almost 2 weeks. While I was waiting for my account to come out of HOLD status, after much pain, frustration and suffering, I decided to change my approach and asked Gemini: āHey, why donāt you use the same approach we use with the pandoc API?ā Which is responsible for the automation to convert Word files. Gemini made the process simpler, as it started using the poppler API. However, for no obvious reason, the process did not identify PDF documents in the same way that pandoc identifies Word documents. At that time, I discovered that when the automation created by Gemini sent the data to poppler, it looked for information in the fileName, whereas the entire path was in the āpathā. Even though I sent the complete Java to Gemini, it did not realize this, but when I realized it, I asked it to make changes to the codes considering the āpathā instead of the fileName and the magic happened. The problem was not with Adobeās API, nor with ConverAPI or Poppler. In the end Iām using Poppler for PDF and Pandoc for Word. All self-hosted. Fast, free process and it met my needs. Finally n8n will be able to be an integral part of my workflow. But thanks for asking!
I just posted here how I came to know n8n because I thought this knowledge would be relevant because of the topic of the thread. I was going to ask the question itself in the appropriate topic. But, before that happened I managed to solve it.
@UnfairMode That is great, thank you for sharing and glad n8n is meeting your needs, I love it, and used pandoc before, but might have to try in n8n too good idea. Well, have a nice day and week, and if you have any questions am sure the community will be more than happy to help.
Many thanks,
Samuel
ā1 Month of Cloud Proā might not be that attractive as a prize for self-hosted users like me
@saedyousef Am still yet to use mine, I donāt like cloud tbh, much much prefer selfhost, but then again I donāt mind handling the infrastructure and issues that come with it.
Might do a giveaway soon if itās okay ekk
Samuel
Great initiative for new learners like me!!
Kudos
@King_Samuel_David I had the same idea as giveaway! Wasnāt sure how tho. Bart confirmed it would be okay.
2 posts were split to a new topic: Encountering an issue with my RAG agent
Excited to join the n8n community! Iām here to grow my developer skills, explore the latest features and updates, and collaborate to solve real-world problems together.
@ThinkBot Yeah I like that idea bro, maybe letās run with a mix of the āmind-blowing workflowā and āremixā challenge. We can ask people to drop their most clever or unexpected n8n workflows, shortlist the top ones together, and then open it up for the community to remix them. Winner gets the free month of n8n Cloud. Could be a fun way to show off creativity and help others learn at the same time
Any ideas?
Hey team, @bartv
First off, big thanks for running the Answer & Earn program. I genuinely enjoy contributing and helping others in the community.
That said, Iāve been feeling quite demotivated lately. Iāve personally solved issues in 15+ threads this month alone (not even counting last month or resolved under my name), yet in many of those, the original poster either didnāt mark my response as the solution or marked their own āthank youā reply instead.
This makes it tough to stay motivated, especially now that only marked solutions count toward the reward system. Iāve flagged a few of these cases, but there doesnāt seem to be a feedback loop or any clear way to appeal them.
Would it be possible to:
I want to keep supporting the community. Iāve enjoyed it a lot, but itās disheartening to see my genuine help go unrewarded.
Hereās one example I reported and was told was being looked into, but Iāve had no follow-up:
N8n (v1.95.2, Docker) - Workers not processing jobs from Redis queue in queue mode - #7 by Alexseyka_Smirnov
Thanks for understanding,
hi
I donāt remember how to set decisions on messages
I couldnāt change it from my mobile now
I think itās an automatic label or Iāll look at it from my PC later
@King_Samuel_David yeah fair enough. I donāt have a great solution. Best I can think of is to provide an FAQ post to explain how to mark something as solved, would that work? Or is this mostly about abandoned posts?
@Alexseyka_Smirnov you marked your own thank-you note as the solution I switched it to @King_Samuel_David 's post.
Thanks @bartv appreciate the quick reply and also for switching that thread over 
I totally agree, itās not always intentional, and I understand that most users arenāt aware of the rewards system. So yeah, a simple FAQ post could help raise awareness, Iāll think further into this and let you know if I think of anything further.
Even just a monthly review of flagged threads would go a long way. I totally get that this isnāt an easy thing to manage.
Happy to help and @Alexseyka_Smirnov Thanks for checking too hope everything is going well always happy to help.
Samuel
I like the idea, but I donāt think this should be an issue as long as the answer is on topic and helpful. Why? I usually write my answers quickly in my own language; itās much faster than writing them in English. Then, I ask AI to translate my answer into English and formulate it more nicely. So, the answer is mine, but itās formatted by AI.
Also, because I can think and write faster in my native language, I can help more people in the community.
Is this also a good opportunity to find a job or get a job in the future?
I have some thoughts around this if you want to connect at some point. I was a forum ambassador and moderator for a different software called Tableau. We had systems in place were volunteer moderators monitored long open questions to help the questions get answered, and attribute the answer to the correct answer. I dont remember ever changing a solution if it was marked wrong, but we did try to reach out to abandoned posts and marked solutions when the answer looked correct and the post was abandoned. Glad to help out if you are looking to find solutions to issues like this.
@yefangjie2024 course, having this on your cv / linkedin can help alot, especially if new in the industry. Depends on what youāre trying to get into, if itās technical support, or solutions, or even becoming a dev. There are many more roles, but employers like to see knowledge and past experience. Remember, though, having certs helps a lot too. I would even recommend having some projects on GitHub showing your skills, which you can showcase to employers, too. hope that makes sense, and n8n did mention previously being part of the community is a good way to get an eye on you for future opportunities within n8n itself too
@lvalics I agree, as long as itās not spammy and the post is genuinely helpful, I think it should be allowed. The issue is when people go overboard and flood threads with overly long or off-topic AI content, making it harder for others to find the actual fix. A balance is definitely needed in this AI-driven era.
@rbreen @bartv Thatās great insight, Iāve heard good things about Tableau. Your experience would definitely be valuable here. That said, we need to make sure weāre not unintentionally limiting or discouraging contributors. It can be disheartening when people like myself put in time to solve issues, only for the original poster to mark something else, or nothing at all, as the solution.
Iāve worked in tech support for years and even monitored and reported stats for TSEs. Thereās nothing more frustrating than being discredited for the exact solution you provided, especially when it gets overlooked or reassigned. In this forum, anyone can work on any thread, so the least that could be done is reviewing and assigning the solution to the rightful responder, especially now that thereās a competitive element involved.
That lack of recognition creates a frustrating environment. Personally, Iāve slowed down my contributions compared to the previous month, partly due to this and a few personal reasons, but I should be back in full swing soon.
At the end of the day, it just depends on what level of support you want to build within the community. If you want to keep good contributors active and engaged, the system has to be fair. Otherwise, itās easy to lose people and hard to get that quality support back.
Itās 2025, and adaptation is key. New innovations create success, but only if the support system evolves with them.
Implementing a flag feature for threads that are clearly solved but left unresolved could really help. It would make it easier for others to find relevant answers quickly, and also help tidy up the community by letting active users submit those threads for review and proper attribution.
You have a significant lack of onboarding materials. EVERY NODE should have a link from the docs to a insertable workflow of examples. Split Node for instance, should have a workflow with a least a dozen examples just based on how powerful and unintuitive it can be. The workflows with Split Node in it are too complex, doing hard to follow tasks, making inserting them to look at how its used frustrating.
Ive recently submitted one of these for Form node, which is also very powerful and yet very little documentation.