XF 3 General Information & Discussion

Has everyone now given up on 3.0, it's seemingly vapourware?
No, it is not. They have been quite clear that it is still planned and in development but that they are releasing 2.4 first. In my world, vaporware is software that never appears or takes years of rumours and speculation to come out. XF 3.0 isn't even close to that standard yet

Given 2.4 is probably happening early in 2025, it makes sense to focus on it for now.
 
They have been quite clear that it is still planned and in development but that they are releasing 2.4 first
That clarity came after they were quite clear that 3.0 would be next after 2.3. Changing plans, moving goalposts, missed timeframes, inserted versions doesn't equal clarity.

18 months ago we given clarity that 3.0 was all but done, it just needed a migration system for existing styles, and that 3.0 would follow 2.3 likely shortly after. We were also given clarity that major XenForo style producers would be shortly contacted to help them get ready for the 3.0 style.

12 months ago we were given clarity that 3.0 will be 2.3 with a new style. But now we have 2.4, and further clarity that 3.0 is a seperate and simultaneous development pathway kept in functional parity with 2.3 and 2.4.

Customer's planning and scheduling for upgrade paths cannot realistically occur because of this ever moving clarity. I can only imagine what it's like for style and add-on developers. Instead, it's just reactive and/or waiting for what may or may not come within any approx. timeframe. I paid for subscription renewal with the clarity that 3.0 would be released in 2024. Instead it was only 2.3. A good reminder that you should really only pay for what is available now.
 
That clarity came after they were quite clear that 3.0 would be next after 2.3. Changing plans, moving goalposts, missed timeframes, inserted versions doesn't equal clarity.

18 months ago we given clarity that 3.0 was all but done, it just needed a migration system for existing styles, and that 3.0 would follow 2.3 likely shortly after. We were also given clarity that major XenForo style producers would be shortly contacted to help them get ready for the 3.0 style.

12 months ago we were given clarity that 3.0 will be 2.3 with a new style. But now we have 2.4, and further clarity that 3.0 is a seperate and simultaneous development pathway kept in functional parity with 2.3 and 2.4.

Customer's planning and scheduling for upgrade paths cannot realistically occur because of this ever moving clarity. I can only imagine what it's like for style and add-on developers. Instead, it's just reactive and/or waiting for what may or may not come within any approx. timeframe. I paid for subscription renewal with the clarity that 3.0 would be released in 2024. Instead it was only 2.3. A good reminder that you should really only pay for what is available now.
@Chris D told us that they are bringing some features early on in the piece.
As 3.0 is the next major release they're trying to work with a new editor earlier. So wait for early 2025.
Let the fellas have their feative break.
Stop sooking and renew that license of yours.
 
Customer's planning and scheduling for upgrade paths cannot realistically occur because of this ever moving clarity.
I only ever plan around whatever version is current; anything new is just a bonus. I don't worry about it, to be honest.

If anything, new versions (even minor ones) just mean more work for me. So I have split opinions about upgrades in general. 👍 to new features and appearance; 👎 for the work needed to adapt my forums to the new versions. Honestly, if we could stay on 2.3 forever, I'd be happier.

I'd say that 99% of members don't even care about new features (many are unaware of features we've had for a dozen years now!), and I'd also say that we've had more complaints when layouts/functionality changed than simple upgrades. So while I'm anxious to see what 3.0 holds, I also forsee all the complaints from the handful of members who don't like change.
 
I only ever plan around whatever version is current; anything new is just a bonus. I don't worry about it, to be honest.

If anything, new versions (even minor ones) just mean more work for me. So I have split opinions about upgrades in general. 👍 to new features and appearance; 👎 for the work needed to adapt my forums to the new versions. Honestly, if we could stay on 2.3 forever, I'd be happier.

I'd say that 99% of members don't even care about new features (many are unaware of features we've had for a dozen years now!), and I'd also say that we've had more complaints when layouts/functionality changed than simple upgrades. So while I'm anxious to see what 3.0 holds, I also forsee all the complaints from the handful of members who don't like change.
This!

What I do is make sure that my licenses including my cloud license is upto date.
Just so i can get the upgrades.
It does not matter when they show up. Once they're done they're released.
What i do is discuss things like the new editor we're about to get in 2.4
There's no harm in doing that as long as nobody is abusing the staff here.
It gives your crew an idea on what might look different.
Think of it like you do with your meals you eat.
It might taste different due to what vegies are in it.
Then the aftermath might be bad.
Especially if someone put in brussel sprouts!
 
No, it is not. They have been quite clear that it is still planned and in development but that they are releasing 2.4 first. In my world, vaporware is software that never appears or takes years of rumours and speculation to come out. XF 3.0 isn't even close to that standard yet

Given 2.4 is probably happening early in 2025, it makes sense to focus on it for now.

or we could reach 2.10 😎
XF 3.0 delayed, but that’s the entire reason 2.4 exists and why 2.5 may exist. Who knows? Maybe there’ll be a 2.6 and 2.7. Maybe a 2.8 and a 2.9. We’ve still not run out of numbers, there could be a 2.10!

I think the important thing is we continue delivering what we can until it’s ready. And it being ready is absolutely critical because we’re not going to release something before it’s ready just to appease an arbitrary release date. This is a good thing.
 
I would be fine with that. I really don't like the prospect of redoing my style again for 3.0 so will stay on the 2.x line as long as it remains viable. And might just hold on whatever is the final release for 2.x. XF really has what we need as it is now. Even us going to 2.3 mostly happened to stay in support, not because anything there was really anything compelling for us in it. 2.4 I am actually seeing some value in, though.
 
Curious to see how it can be made even better. Removing the JQuery dependency and moving to WebP was very important. But otherwise I don't see much room for improvement from the current state of web development.

Making the layout heavier/complicated would be a no-go, as everything is going in the opposite direction. As mobile devices have now reached certain resolutions, it works very well with the current design. So I am not sure what @Paul B has in mind.

Otherwise it probably has too many settings already.

I think it will all move to AI features and especially AI moderation. Otherwise maybe App support at this point.
 
You're kidding right? Or just forming an opinion based on ignorance?
Most browsers and operating systems have supported it for 4 years now.

FIrefox since 2019, and especially Safari since the end of 2022 and very limited ! And where is the big gain Avif vs. WebP? Even if a better compression <=> its also more CPU load. In my opinion, even JPEG with AI compression can lead to very good results, close to both, if its just serving images.

So yes, there is a lot of support, but for large projects you would rather go "legacy" than have 2-3 year old hardware that is not supported. So adding/moving to WebP makes sense at this point, but going further probably not.
 
Last edited:
  • Like
Reactions: rdn
You're kidding right? Or just forming an opinion based on ignorance?
Most browsers and operating systems have supported it for 4 years now.
Edge literally didn’t support it until last year. MacOS Safari only had it for the last couple of years.

The caniuse database (which is what web developers tend to trust instead of Wikipedia) defines it as “available across all major browsers as of 2024”

Which means, realistically, it might be usable now but giving it another year or two isn’t actually a bad idea. Especially as not all server OSes that are currently shipping PHP 8.1 or 8.2 are shipping libavif yet.
 
  • Like
Reactions: Mbx
I would give AVIF +2 years. Its still too new and offers no big gains to begin with.
The major browsers have been supporting the format for a while now, so in that particular respect it is not too new. As for gains, “big” is relative. After conducting some experiments, we discovered that within the quality range of 60–70 %, AVIF achieves a similar quality to WEBP at 90 %, but with a significantly smaller data size (≈ twice as small). In our case, that is a big difference.
 
Which means, realistically, it might be usable now but giving it another year or two isn’t actually a bad idea.
You are considering the server side of things, wisely so. However, having the feature in XF 3.0 for those who want to implement and enjoy it, would still be a good move, methinks.
 
You are considering the server side of things, wisely so. However, having the feature in XF 3.0 for those who want to implement and enjoy it, would still be a good move, methinks.
Sure, it is not really a big deal to add the imageavif to the image methods. If you need it, it would probably take minutes to replace the imagewebp stuff with imageavif as example and there you go...
 
The major browsers have been supporting the format for a while now, so in that particular respect it is not too new. As for gains, “big” is relative. After conducting some experiments, we discovered that within the quality range of 60–70 %, AVIF achieves a similar quality to WEBP at 90 %, but with a significantly smaller data size (≈ twice as small). In our case, that is a big difference.
Edge is not a major browser now? It hasn’t even had the format a year yet.
 
Sure, it is not really a big deal to add the imageavif to the image methods. If you need it, it would probably take minutes to replace the imagewebp stuff with imageavif as example and there you go...
That depends on your hosting company. imageavif requires libavif, which isn’t in every major distribution yet.
 
It supports it, doesn’t it? Any specific issues you have discovered with Edge’s AVIF implementation?
It supports it now. But claims of “Major browsers supporting it for years” are definitely not entirely the case.

I haven’t seen many uses of AVIF in the wild to be honest, but when I have seen them, I have more trouble with them on iOS then I ever do on desktop.
 
Back
Top Bottom