XF2.3 feedback and comments

[whynot]

I am genuinely concerned that something awful has happened to someone in the dev team.

Hopefully not lightning again.

 

[Alpha1]

Personally I find it ironic that the XenForo team has opted to essentially do an incremental rewrite and spread the team thin over writing code and supporting 3 different versions (2.2, 2.3, 3.0) instead of just skipping 2.3 and focusing on XF3. I know that would probably not go over well with customers to say the least, but its the fastest way to the finish line. It seems somewhat logical that 2.3 and 3 are going to take 'forever' so to speak.
There have been times that @Kier and Mike quit their jobs at vbulletin over the decision to do an incremental rewrite. The old schoolers here all know what happened to that incremental rewrite.

 

[Mouth]

In just a few days, it will have been 6 months since the first 2.3 HYS and 3 months since the last 2.3 HYS.

 

[Wing]

 

[Chris D]

Personally I find it ironic that the XenForo team has opted to essentially do an incremental rewrite

I think you're misunderstanding. There is no "incremental" rewrite.

XF 3.0 is essentially the same code base as 2.3. Anyone expecting the kind of changes we introduced between 1.5 and 2.0 will be disappointed (though a fair few will be relieved!)

We're not even contemplating any major reengineering of the underlying code in any sense that would quantify a rewrite. We're happy with that and the choices we made quite a while ago.

But the front end needs to evolve and so that's the sole focus of 3.0.

3.1 will focus on new features. Perhaps if there is a 4.0 one day that will be closer to a rewrite but there's not a compelling reason for that in the near term.

 

[Chris D]

Issues

• BB Code present in post content
  

We have rolled out a candidate fix for this. We're not entirely confident as we aren't able to reliably reproduce all parts of this and there is inconsistency in the behaviour depending on operating environment but let us know if you run into any of the following:

• Issues with [COLOR] tags in raw BB code containing colours or CSS variables
• Issues with dark text on dark backgrounds
• Issues with light text on light backgrounds

 

[Digital Doctor]

This is beta software. It is not officially supported.
We do not recommend running it in production.

But … we need you to install it in a production environment to bug fix !

 

[Mendalla]

But … we need you to install it in a production environment to big fix !

No, they want you to install it in a TEST environment, which is how any major change, including upgrades, should be done. Then you and maybe a couple of your mods go into the test environment and try stuff to see what breaks. I am in IT and we would never, ever run beta code in production except in very specific circumstances (e.g. it fixed a critical issue we were having) and then only after testing.

 

[Suzanne O]

Glad i'm not a beta tester!
Waiting on the full stable release which is going to be huge!

 

[James]

Ok decided to have a bit of a play.

Upgraded it seems to missing some tick boxes when upgrading the media and resource addons

Glad i'm not a beta tester!
Waiting on the full stable release which is going to be huge!

You are a beta tester if you upgraded.

 

[Suzanne O]

You are a beta tester if you upgraded.

Haven't upgraded. I used a test site that went sour when i started using it.
Had a lot of bugs in it.

 

[Mendalla]

Haven't upgraded. I used a test site that went sour when i started using it.
Had a lot of bugs in it.

Which is pretty much the dictionary definition of beta testing. Trying it out to see what works and what doesn't. Doesn't matter whether you do it in production or dev (though I advocate the latter per my post above), just whether you've tried the beta code.

 

[Digital Doctor]

No, they want you to install it in a TEST environment

People don't.
The majority of people don't have a test environment.

of course it is alot safer to install in a test environment.

I suspect there could be some code that would PREVENT beta releases from being installed in NON-Test environments (or make it alot harder), but I don't suspect that code exists.

 

[Suzanne O]

Which is pretty much the dictionary definition of beta testing. Trying it out to see what works and what doesn't. Doesn't matter whether you do it in production or dev (though I advocate the latter per my post above), just whether you've tried the beta code.

When i went to have a look at it i had to get it up on a test site. After looking at it i knew there were too many bugs in it to proceed.

 

[Mendalla]

People don't.
The majority of people don't have a test environment.

of course it is alot safer to install in a test environment.

I suspect there could be some code that would PREVENT beta releases from being installed in NON-Test environments (or make it alot harder), but I don't suspect that code exists.

Anyone who installs a beta 1 of anything in a production environment is taking a big risk. If you don't have a test environment and feel you really must use a pre-release version, my advice is to at least wait for a later beta or the release candidates. I am holding for the RCs anyhow. Don't have the time or energy for beta testing right now.

 

[Mendalla]

Oh, and please, please take a backup of both files and db so you can roll back.

 

[Mr Lucky]

I have it on my test site BUT the biggest thing holding me back is the bug that makes it very difficult to carry out the first task - merging templates.

I suspect anyone who finds it to be too buggy is attempting to use it with addons and/or 3rd party sales enabled. It will take some days or weeks for many addons to work hence a lot of the apparent bugs will be related to addons or styles not yet compliant.

Having said that there are some glaring issues on here that were not yet dealt with before releasing the beta. I think in some respects it may have been better to resolve those unfixed reported bugs before the beta release but at least it being out sooner will give addon/style devs the chance to deal with it.

I suspect it will be months before this is ready for release - best not to rush it.

I hope the people who have been impatient for the release are helping by beta testing...

 

[Forsaken]

I have it on my test site BUT the biggest thing holding me back is the bug that makes it very difficult to carry out the first task - merging templates.

I suspect anyone who finds it to be too buggy is attempting to use it with addons and/or 3rd party sales enabled. It will take some days or weeks for many addons to work hence a lot of the apparent bugs will be related to addons or styles not yet compliant.

Having said that there are some glaring issues on here that were not yet dealt with before releasing the beta. I think in some respects it may have been better to resolve those unfixed reported bugs before the beta release but at least it being out sooner will give addon/style devs the chance to deal with it.

I suspect it will be months before this is ready for release - best not to rush it.

I hope the people who have been impatient for the release are helping by beta testing...

I'm waiting for next week when they start complaining about something else.

 

[badmonkey]

I installed in a test environment with usually about a half dozen or so online at any given time.

If installing in a test environment worked, they'd do it themselves.

 

[Mendalla]

I installed in a test environment with usually about a half dozen or so online at any given time.

If installing in a test environment worked, they'd do it themselves.

Where do you think Xenforo does their development work? Right on the live site? I have no insight into their environment, but based on my own, I would bet on them having multiple environments for dev, QA, etc. But at some point, you need to let users have at it because (a) they are highly likely to find stuff you missed and (b) user environments are going to be much more complex than any dev or QA can be. Custom work, different combinations of add-ons, and so on.