Are forums dying?

[Fraser]

I asked the same question as the OP regarding the Thermal Imaging area of this forum. Over the years I have tried to keep that forum active during quiet periods by posting, what I hoped were, interesting equipment teardowns and starting discussions.

https://www.eevblog.com/forum/thermal-imaging/is-this-thermal-imaging-forum-dying-if-so-what-to-do-about-it/

What I have noticed is the number of people who join this forum to ask me a question and then we hear no more from them. I am now regularly approached in a manner not dissimilar to asking AI for the answer to a question. I have become tired of being treated like ChatGPT so no longer respond to such “one off” post forum members.

A fun fact, I specialise in thermal imaging related technical topics and have posted a lot on this forum. AI often regurgitates my posts, almost word for word ! So, as has been stated, Forums are a very rich hunting ground for information and AI makes great use of it. If the forums disappear or AI is blocked from accessing them, the AI engines will be starved of the wealth of information shared on such. Do I care if that happens ? Me personally, no. I belong to a generation that is happy to do research and does not require “spoon feeding” answers. That said, AI can be a useful shortcut to obtaining a fast answer, but such AI answers still need to be verified for accuracy.

Fraser

[factory]

As a forum person, i would say AI answers probably cut a good piece off the traffic that directed users to forums, but in the end people try to solve a problem or search for entertainment, which is what gets them to where they wanted. A potentially flawed AI giving you an answer is not always the best way to get a solution. It does not give you fresh topics to read or other peoples insight. Google does forward you to different websources if no answer comes out. Haven´t found it to forward to a forum, though.

Discord was and is unaffected by search traffic (the ones i know are outlets for YT-creators i watch or a teamspeak/voicechat replacement), Reddit did a partnership to deliver it´s content for AI training purposes.

To me the (small) forum is still one of the best things the web has to offer.

Google search giving results on "forums" now seem to be buried in a separate tab at the top of the page, example attached.

David

[factory]

Discord's biggest issue for me is its inherently closed nature.
You need a discord account, and you need to use their software to even see anything, let alone participate.

I dislike closed forums. I spend a lot of time helping people, especially in the data recovery sphere. Instead of taking my knowledge and experience to the grave, I share it with the Internet community. I can't stomach the idea that all my effort would benefit only a select few.

Along with the forums that restrict attachments/images etc to members only, which there are many, I also hate the fact some images & info is locked behind forums & other sites that are closed to new members, I've come across a few forums/other sites that are still active/online but closed in this way.
It also works the other way, for example HiFiengine being closed for new & lapsed members, is a great excuse to avoid having to repair coworkers audio equipment, which is usually far too time consuming to repair. 

David

[abeyer]

Discord is great - and it even has features for community forum options and threads within servers if people prefer a thread/post style of dialogue.

Not just once, they went "threads are nice, let's do them thrice!" There are (at least) three different ways of doing thread-like things, without good controls over them, and they don't play nice together. Without a small disciplined community deciding on one, it invariably just degrades into a mess that's worse than no threading at all. And that's just one glaring example of their awful broken UX. It's like someone read the CADT post and thought it was a how-to guide.

At the end of the day, though, I often resign myself to dealing with bad software. What I don't tolerate is malicious software. Their history of intentionally stealing user data in violation of their own terms, lying about it, getting caught in their lies, and then later just rewriting the terms to say "we're gonna do it anyway, so there" puts them beyond hope of ever being trustworthy in my eyes. And they forbid using any third-party client software that might alleviate those concerns.

[fzabkar]

Many years ago there were numerous forums which blatantly pilfered their content from Usenet newsgroups. They would falsely identify the Usenet posters as their "guests". My name used to turn up as a contributor to forums that I had never heard of. Without this stolen Usenet content, many of these forums would have been ghost towns.

[Bud]

What I have noticed is the number of people who join this forum to ask me a question and then we hear no more from them

I would not care about it unless I am a narcissist which i hope i am not. People have priorities, work, family. Cant blame them for lack of feedback.

. I am now regularly approached in a manner not dissimilar to asking AI for the answer to a question. I have become tired of being treated like ChatGPT so no longer respond to such “one off” post forum members.

This is the nature of AI. Perhaps you can always add to your posts " Always check with Fraser to validate this information" kind of thing. Then AI should add that to its blurbs.

[Randy222]

The topic thread title is interesting.

I wonder if you can convince the bot users in Moltbook to kill off their conversations?

[wilfred]

I asked the same question as the OP regarding the Thermal Imaging area of this forum. Over the years I have tried to keep that forum active during quiet periods by posting, what I hoped were, interesting equipment teardowns and starting discussions.

https://www.eevblog.com/forum/thermal-imaging/is-this-thermal-imaging-forum-dying-if-so-what-to-do-about-it/

What I have noticed is the number of people who join this forum to ask me a question and then we hear no more from them. I am now regularly approached in a manner not dissimilar to asking AI for the answer to a question. I have become tired of being treated like ChatGPT so no longer respond to such “one off” post forum members.

A fun fact, I specialise in thermal imaging related technical topics and have posted a lot on this forum. AI often regurgitates my posts, almost word for word ! So, as has been stated, Forums are a very rich hunting ground for information and AI makes great use of it. If the forums disappear or AI is blocked from accessing them, the AI engines will be starved of the wealth of information shared on such. Do I care if that happens ? Me personally, no. I belong to a generation that is happy to do research and does not require “spoon feeding” answers. That said, AI can be a useful shortcut to obtaining a fast answer, but such AI answers still need to be verified for accuracy.

Fraser

Your post encapsulates the dichotomy quite well. Are forums a place to gather and discuss things or are they a place to simply get information? I think they are both. But with AI substituting adequately for most of the time answering questions, it starves the forum of enough people to have a vibrant community.

It is a pet hate of mine on the other side when a thread takes off it inevitably devolves into a very small number arguing about smaller and smaller details on tangential issues. But the effect is the same in that it diminishes the community when those who contributed along the way get swamped by the noise and retreat to the sidelines.

As I get older I see the futility in shouting at clouds and have become more accepting of the state of affairs.

[fzabkar]

What I have noticed is the number of people who join this forum to ask me a question and then we hear no more from them

I would not care about it unless I am a narcissist which i hope i am not. People have priorities, work, family. Cant blame them for lack of feedback.

I think people should make time for feedback, whatever their priorities. It's the least they can do to acknowledge the help they've been given.

[Bud]

If you feel (not referencing you specifically, just in general) you are entitled for feedback, do not post.

[fzabkar]

If you feel (not referencing you specifically, just in general) you are entitled for feedback, do not post.

The sort of feedback I expect is a report whether a suggested approach worked or didn't work. The original poster is not always the only person who benefits from the advice and solutions.

It's nice to get a thank you, but I won't be annoyed too much if it's not forthcoming. I hate the upvotes and downvotes at places like Reddit. I much prefer the "Say Thanks" feature in this place. It's quick and efficient and requires no additional posting.

[golden_labels]

Fraser didn’t mention “feedback” in their post. Bud interpreted it this way in their reply. Sure, it is one of possible ways to read Fraser’s words. To me it’s not the obvious one, and the second part of the reply suggests further misreading of Fraser’s words.

I believe Fraser speaks of something that also irks me. I don’t care too much about 👍/👎 responses to my posts (about that later). But if I’m with somebody on a phone, I don’t expect hearing a sudden silence mid-conversation. They don’t even hang up, which would at least indicate no will to continue the talk. They just remain silent, while on my end it’s a call that hangs for days. It gets worse in group chats, like fora. Not only multiple people get confused that way, each of them is still talking to the missing person. And each of them is notified about that, still donating their time to an already dead thing.

While the intentions may differ, to the community the effect is not far from a common trolling. A single person ties up multiple members, both using their invididual time and redirecting attention from other people that need help, over time exhausting people and eroding their will to engage. And that is combined with the issue of asymmetric effort.


Regarding rating systems, it is motivating to receive a positive feedback. I certainly like seeing that my effort is actually useful. At the same time we refused to introduce any such thing on my forum. Our aim were good answers, not measuring emotional response of the OP. In our case the latter often being negative, when good answers appeared and the asking party was a beginner. So it all depends on circumstances, the kind of the community, topics discussed.


On some fora the solved threads are expected to be marked as such. Which solves the first problem without talking, and is a kind of a feedback.

[EEVblog]

I much prefer the "Say Thanks" feature in this place. It's quick and efficient and requires no additional posting.

I think it works well. Very early on with the forum we experimented with a downvote system and also a form of karma points system, but most people hated it.

[EEVblog]

Your post encapsulates the dichotomy quite well. Are forums a place to gather and discuss things or are they a place to simply get information? I think they are both.

Even if a forum starts out as a pure information resource, it will inevitably end up as a discussion forum if it has that ability.
This forum did start out as a way to discuss my videos, better than Youtube comments at the time, and also general chat. I think it was just weeks before the topics expanded, and it's been a discussion forum ever since.
Because it's highly indexed ad ranked on Google a lot of external people do see it as a way to ask a question and get a single answer and then vanish.

[jpanhalt]

Chat v. Information
I don't do a lot of chatting, except with family.

In April, 2025, I posed a question question about how EEVBlog was going to compete with AI as a source for technical information (https://www.eevblog.com/forum/blog/is-ai-an-ultimate-threat-to-eevblog-and-similar-sites/msg5897192/#msg5897192).

While vacationing from here, I have used AI extensively from three sources: Google (Chrome is my default browser), Grok, and ChatGPT.  Google is hard to have a dialog with, but Grok and ChatGPT are excellent in that regard.  That's particularly evident when one points out inconsistencies and errors.  Not once has any of those sources, referred to MAGA, the Orange Man, or TACO.  I've gotten almost through two projects: 1) PIC 16F1xxx Assembly code for using the MAX3421E as USB host; and. 2) Installing Starlink with the antenna on a barn that is on a separate electric meter with  Point-to-Point link to my home about 200 ft away.

Can EEVBlog do anything differently to thrive in an AI world and distinguish itself from other technical forums? 

AI provides answers that always sound right, but there is no inherent check on that.   EEVBlog is a resource of enormous human experience that can tell right answers from wrong.  Rather than appear to fight AI, what about building on that resource and encourage answer seekers to post AI results for constructive comments?

I do not disagree with the decision to block AI crawlers because of demands on your servers.

[Kilrah]

Rather than appear to fight AI, what about building on that resource and encourage answer seekers to post AI results for constructive comments?

No universal solution, but on the places i do support for people who come after they tried to solve a problem using AI are the worst, had they come and asked their question straight away it'd have been a 2-line answer, but since they spent a couple of hours with AI going nowhere it's a lot more effort to first make them forget/ignore all the crap they were told before actually getting to solve the real problem.

Probably works better for "general theory" kind of stuff, but on specific projects the result is usually very poor.

[SparkyFX]

Can EEVBlog do anything differently to thrive in an AI world and distinguish itself from other technical forums? 

a) Automatically respond to new topics with an AI answer by a designated AI user (marked as such and all) for everyone to read
Pro:

• would streamline both approaches into a thread, allows people to consent or dissent with AI right away
• guarantees an answer
• fixes the problem that many people might ask the same/a similar question, each individually hassling with the AI to squeeze a solution out of it. Would make the forum an even better resource  because you can see other peoples prompts

Con:

• which model?
• Would need to be free
• Might need protections against all kinds of abuse
• the moderators would need to be OK with that (more effort)

[David Hess]

What I have seen recently is forums being increasingly dominated by AI generated posts intended to either deliberately disrupt, or collect views, which lowers the forum's signal to noise beyond usability.

[PlainName]

Your post encapsulates the dichotomy quite well. Are forums a place to gather and discuss things or are they a place to simply get information? I think they are both.

Even if a forum starts out as a pure information resource, it will inevitably end up as a discussion forum if it has that ability.
This forum did start out as a way to discuss my videos, better than Youtube comments at the time, and also general chat.

I am sure that's what keeps this place going, but it is only because the moderation allows significantly diverged chat. Often, some unexceptional starting topic will digress into something completely unrelated but nevertheless more interesting than the initial topic. Of course, not always  But those whining about off-topic stuff sometimes, I think, miss the cart for the horse.

And, importantly, this place is generally not real-time. Often I will read a reply to a comment and then just think about it for a bit before coming back and dealing with it. With IM-like services other people know when you are there or not, and then wonder why you are not instantly and always responding. This is one reason why I like to hide my online status.

And, a long stream of real-time chat hides stuff if you're not there all the time. Go off for a shag and when you get back any relevant info is probably many pages off-screen and disjointed. Reddit also suffers this, in that it is really hard to see new comments and follow what's going on. In fact, with Reddit I tend to drop into a topic and then that's done - never go back to it. But, OTOH, for threads that have been and gone for a while, it's great to read top to bottom to pick up what it was about and extract any info.

I may be biased, having run QuickBBS and been a Fidonet node and hub, but ISTM that whilst EEVBlog isn't perfect it is vastly more appropriate for online knowledge and social interaction than other current places. Sure, we get members whose sole purpose seem to be to provoke arguments, but over on Reddit when I started there I posted three comments and two got pretty instantly downvoted and spitefully tossed off by fuckwits. I just check it out for info via Google now.

[jpanhalt]

If one looks at new thread numbers from Dave, you see a healthy year to year increase even before COVID.  2017 and 2018 were the highest years with 18,426 and 18,329, respectively.  In 2023 after peak COVID, the number was 14,597.  That dropped a little in 2024 to 14,364, but the big drop (12.7%) was 2024 to 2025. 

What happened in 2024?  In May, 2024, Google Chrome rolled out the first phase of integrated AI to its searches.  I suspect other free AI also became more popular.  Google just completed the second phase in July, 2026. (According to a google search.) A year over year drop of 12.7% may be small compared to the drop from 2017, but it would still get my attention.  We've completed 6 months of 2026, and if that date can be extrapolated to year over year, what does it show?  If 5,509 is for 6  months, and if it is legitimate to double that, 11,018 is a drop of 10.8%.

As for AI users seeking answers and not accepting them, that's common for people who don't use AI too.  I suspect a lot of them know they will be shamed if they admit using AI.  What I am suggesting is to remove that stigma and facilitate fact checking AI.  So far, AI cannot replace the knowledge and experience of many members of EEVBlog.  I won't name any for fear of omission, but there are a lot whose posts I consistently read regardless of the topic.

[PlainName]

I suspect a lot of them know they will be shamed if they admit using AI.

It's not the using of AI that is a turn off, it's the posting comments directly from AI. Someone asks a question and the comment is: "AI says ....blah...". That's no different to "Google search gave: ... blah...". It's a waste of space and time - anyone could just ask Google and AI directly, so what does the poster gain from duplicating wasteful stuff? It make them feel big - 'hey, I can answer your question' (but, of course, they can't because they had to ask AI to do it).

That's the problem.

[tszaboo]

Rather than appear to fight AI, what about building on that resource and encourage answer seekers to post AI results for constructive comments?

No universal solution, but on the places i do support for people who come after they tried to solve a problem using AI are the worst, had they come and asked their question straight away it'd have been a 2-line answer, but since they spent a couple of hours with AI going nowhere it's a lot more effort to first make them forget/ignore all the crap they were told before actually getting to solve the real problem.

Probably works better for "general theory" kind of stuff, but on specific projects the result is usually very poor.

In my estimate, when a coworker tries to do my job with AI, it's creates 5-6 times more work for me than doing the work myself.
Because I need to convince them that the AI workslop is wrong, find the issue in it. Which often times illogical bs, which no human with an engineering degree would make. And then I have to do the work. And then do this again in a few months when they forgot the interaction with me (too complicated), and they only remembered the workslop and I have to do the same thing again.
I don't know what the solution is, but I hope the whole AI infrastructure will bankrupt themselves soon, hopefully before we all go mad because of it.

[SiliconWizard]

I suspect a lot of them know they will be shamed if they admit using AI.

It's not the using of AI that is a turn off, it's the posting comments directly from AI. Someone asks a question and the comment is: "AI says ....blah...". That's no different to "Google search gave: ... blah...". It's a waste of space and time - anyone could just ask Google and AI directly, so what does the poster gain from duplicating wasteful stuff? It make them feel big - 'hey, I can answer your question' (but, of course, they can't because they had to ask AI to do it).

Yes, but I think people are going to have an increasingly hard time seeing the difference between information that strictly comes from their own thoughts and experience and information gotten via LLMs. They're increasingly seeing that as an extension of their own brain and that's also exactly how it's being marketed to them.

So they're not necessarily doing that just to "feel big", but actually because they can't see the difference. And as to your remark "anyone could just ask Google and AI directly" - sure, but many now probably assume that if you are asking on a forum, that's not because you want expert opinions from real people, but that's because you don't know how to ask "AI". So, they'll do it for you.

[PlainName]

hey're increasingly seeing that as an extension of their own brain and that's also exactly how it's being marketed to them.

So we'd be doing them a favour by reminding them they're being taken over 

but many now probably assume that if you are asking on a forum, that's not because you want expert opinions from real people, but that's because you don't know how to ask "AI".

That's a good point. Or it would have been last week, but I think nowadays the ones not using AI are pretty rare. Hell, if you can type into a search engine you'll probably be using AI, even unintentionally.

[Kilrah]

And as to your remark "anyone could just ask Google and AI directly" - sure, but many now probably assume that if you are asking on a forum, that's not because you want expert opinions from real people, but that's because you don't know how to ask "AI". So, they'll do it for you.

That's the misunderstanding. I at least want to believe that if people come to forums it's because they want real human input and they at least did a search first, which will have included AI results. Often confirmed by the people just posting "I don't know but AI said..." being shot down quick as unhelpful. And often wrong too because they don't know the subject and the AI hallucinated their reply and they had zero knowledge to validate it.

Most times I see it it's older people who want to think they understand things (or that AI allowed them to), but actually don't. Or want "to feel useful" with the least amount of effort.