Could AI coding "duplicate" Xenforo like it was done for Adobe products, but alot easier because of the open code aspect of the product ?

Digital Doctor

Digital Doctor

Well-known member
Licensed customer
Could Xenforo almost be "forked" ?
Is this type of issue a reason why Xenforo as a Cloud only product might be inevitable ?
@JustinHawk said he is releasing less Xenforo addons as they just get gobbled up by AI, isn't this extra true for Xenforo Inc ?
 
.... hmmmmmm ..... maybe that's part of the reason for the "Xenforo pause" situation ?
:)

all situations have 3-100 reasons, of course.

FWIW, I'm not asking.
 
Is it possible that Xenforo cloud could "bring in" addon coders, so their code isn't exposed ?

or will Xenforo + AI, just crank out the addons "itself" ?

hmmmmm.

The s--- seems to have gotten real lately. Claude helped me crank out the AutoHotKey script I wanted. It also helped me write an SQL query that would have normally taken me 8 hours - and not always successful, I did it in 30 minutes.

May you live in .....
 
Pereto Principal

I could build something that does 80% of what XenForo does in a day or two.

The problem is that the value in a platform like XenForo is in the other 20%, which will take a long time and a lot of effort to get right.

You see it all the time these days - people building something new really quickly which looks great on the surface - but what you don't get to see is all of the missing functionality that will take a very long time to perfect.

For example - a bunch of well known PHP devs recently built a social network from scratch in PHP for fun in only 48 hours - heavily utilising agentic coding. The product they built was amazingly functional for such a quick turnaround.

To view this content we will need your consent to set third party cookies.
For more detailed information, see our cookies page.

The problem is that it's nowhere near complete and there are already nearly 1,000 pull-requests and hundreds of issues that have been logged - it's going to take them years to get caught up with all the functionality, even with agentic help.

github.com

GitHub - dev-null-sh/dev-null.sh: a social media for devs

a social media for devs. Contribute to dev-null-sh/dev-null.sh development by creating an account on GitHub.
github.com github.com

Either way, it reached over 10,000 members in under 5 days, which is pretty impressive. It's also open source, so you could take it, fork it and make your own social network for your own niche if you wanted to.

The trick with software like XenForo is that it has to cater for everyone - so there's a bunch of features and functionality that some people will never use.

If you were to build your own custom site from scratch, you could leave out all the stuff you didn't need and get something up and running pretty quickly.

But the benefit of a widely used open source platform or a well supported and mature product like XenForo, is that you know it's going to be robust and has been well tested across hundreds or even thousands of sites with many many different clients - so the problem you'll inevitably face with your custom built system have already been found - and most of them solved - by a mature platform like this.
 
Sim said:
For example - a bunch of well known PHP devs recently built a social network from scratch in PHP for fun in only 48 hours - heavily utilising agentic coding. The product they built was amazingly functional for such a quick turnaround.
Click to expand...
I had a stab at this and well 99% of the time and cost is getting through legal regulatory requirements for various countries for 'social media' :LOL:o_O
 
  • Wow
Reactions: Sim
Sim said:
If you were to build your own custom site from scratch, you could leave out all the stuff you didn't need and get something up and running pretty quickly.
Click to expand...
This would be why I would consider doing it. We are a small site that uses only a fraction of the newer features from 2.2 and later. We could run on a fairly simple forum system and I have eyeballed things like phpBB but dread the work of importing. Vibe coding something that uses the existing XF data structures eliminates that import and I could also mainline some features I have added with add-ons. So basically Wondercafe2 becomes both the site and software. But I don't have the time or, really, the skills for it. We're building a new system at work with AI assistance but I have other people doing the work and I am just the PM and liaison with senior leadership.
 
Digital Doctor said:
Could Xenforo almost be "forked" ?
Is this type of issue a reason why Xenforo as a Cloud only product might be inevitable ?
@JustinHawk said he is releasing less Xenforo addons as they just get gobbled up by AI, isn't this extra true for Xenforo Inc ?
Click to expand...
Paul B said:
It has already happened.

We are aware of at least 2 products which have been derived from XF.

It will become commonplace.
Click to expand...
zappaDPJ said:
It's practically a given that Xenforo will become largely obsolete, replaced by an AI created, functionally similar but cheaper product perhaps even before v3.0 sees the light of day.
Click to expand...
Seems my time frame was a little off, sorry about that.
 
philmckrackon said:
Building a site does not have belonging or users. That takes time.
Click to expand...
On the other hand, being able to build a site that is tailored to your users, perhaps with input from them incorporated into the process, would help with that belonging, wouldn't it? That's part of why we are building our own EHR at work. We can precisely tailor it to how our users work and what we as a company require.
 
Sure, you "can" build your own forum software with AI. That doesn't mean you "should", however.

These days, it's pretty easy for anyone to create just about anything they want with AI. That's not really the problem. The problem comes when you start running into edge cases, security vulnerabilities, and other issues that go over the head of the "developer".

AI might be perfectly capable of fixing those issues, sure, but if the person using it doesn't understand what's going wrong, how are they going to know whether AI is actually fixing the problem or just putting a band-aid on it? I run into situations like this regularly when using AI to help me code.

There's also the issue of efficiency. Just because AI can write code that works doesn't necessarily mean it's writing code that works efficiently, especially when you're continuously adding new functionality to an existing codebase. Without someone who understands how everything works together, it's entirely possible to end up with software that requires significantly more server resources to accomplish the exact same tasks.

And then there's security. If you're running forum software that handles user accounts, private conversations, and other sensitive information, do you really want to be responsible for identifying and fixing every potential security vulnerability yourself?

I'm not saying AI can't build a viable alternative to XenForo. I absolutely believe it can. But building the software is only part of the equation. Maintaining it, securing it, and keeping everything running efficiently over the long term is a whole other animal. That's where having an experienced development team behind the software becomes valuable.

Considering how relatively inexpensive a XenForo license is, I just don't see the benefit of taking on all that responsibility yourself.
 
mjda said:
I'm not saying AI can't build a viable alternative to XenForo. I absolutely believe it can. But building the software is only part of the equation. Maintaining it, securing it, and keeping everything running efficiently over the long term is a whole other animal. That's where having an experienced development team behind the software becomes valuable.
Click to expand...
Exactly. For our project at work we have two of my team plus a couple engineers from our corporate group's AI team.
 
mjda said:
Sure, you "can" build your own forum software with AI. That doesn't mean you "should", however.

These days, it's pretty easy for anyone to create just about anything they want with AI. That's not really the problem. The problem comes when you start running into edge cases, security vulnerabilities, and other issues that go over the head of the "developer".

AI might be perfectly capable of fixing those issues, sure, but if the person using it doesn't understand what's going wrong, how are they going to know whether AI is actually fixing the problem or just putting a band-aid on it? I run into situations like this regularly when using AI to help me code.

There's also the issue of efficiency. Just because AI can write code that works doesn't necessarily mean it's writing code that works efficiently, especially when you're continuously adding new functionality to an existing codebase. Without someone who understands how everything works together, it's entirely possible to end up with software that requires significantly more server resources to accomplish the exact same tasks.

And then there's security. If you're running forum software that handles user accounts, private conversations, and other sensitive information, do you really want to be responsible for identifying and fixing every potential security vulnerability yourself?

I'm not saying AI can't build a viable alternative to XenForo. I absolutely believe it can. But building the software is only part of the equation. Maintaining it, securing it, and keeping everything running efficiently over the long term is a whole other animal. That's where having an experienced development team behind the software becomes valuable.

Considering how relatively inexpensive a XenForo license is, I just don't see the benefit of taking on all that responsibility yourself.
Click to expand...
Spot on.
It's about us using the AI stuff along with the license in it's correct way.
On my forum we use AI as part of our games and humour board.
I let my members use ChatGTP and any other AI software they want as long as they keep it to that board.
I'm waiting on an official XF AI app that helps with registration. I suspect that might take a while though. Still the idea is there and it's only going to be that way.
 
Itworx4me said:
Embrace AI today, or risk falling behind tomorrow.
Click to expand...
Nah. I'll tell you why.
It's being forced down our throats everywhere and therefore misused everywhere.
Misused as in AI bots scraping your work and putting it up as their own is not on.

Phone calls are done by AI bots wanting to get your voice involved with their scams.
 
Mendalla said:
On the other hand, being able to build a site that is tailored to your users, perhaps with input from them incorporated into the process, would help with that belonging, wouldn't it? That's part of why we are building our own EHR at work. We can precisely tailor it to how our users work and what we as a company require.
Click to expand...
Toche, it is a double edged sword.
 
Last edited:
You must log in or register to reply here.
Back
Top Bottom