bfarber

If you can reproduce a bug on this site, it definitely should be reported. A lot of times, especially skin/javascript issues, it's because we fiddle with the skins here and they're not always up to date) but it should be noted nevertheless.

Yes and this has been a very long beta process which is winding down. There are 3 bugs on the public side. Three. But we have to keep waiting due to skin issues piling up more and more each day. I don't care really, but someone whined that I shouldn't be testing beta 3 and should be testing beta 4. Well bfarber hinted that beta 4 could be skipped because beta 5 was going to be unusually quick and it hasn't been.... it's been 2 weeks and I think there were only 3 weeks or less between 3 and 4.

Well, things change I'm afraid. ;) We expected to get beta 5 out really quickly, but decided to hold it back internally to take care of other things (and to fix bugs we've found...e.g. the skin system bugs I reported yesterday).

I can't go into details for obvious reasons, but we have plans internally, and sometimes they change. I said we *expected* a much quicker turn around time between beta 4 and 5 (based on those plans) and that was the reason we didn't plan to refresh beta 4 with the language XML files. I DID post those language XML files for anyone who installs beta 4 to import, so it's not like beta 4 is not testable.

And keep in mind - our primary goal here is to get testing done so we can release a solid and stable product. That's our number one goal. Giving modders time to work on stuff to get ready for release and so forth is a nice side effect where possible, but that's not going to drive our release schedule.

I'm afraid you're simply missing something. ;) For instance, your site might be using the class_db_mysqli_client.php file - in which case the edit wouldn't do anything.

You really need to post in the peer to peer forums to get help with this sort of thing.

Right after the last } in your code you posted you'd add

mysql_query( "SET NAMES utf8", $this->connection_id );

For the record, I don't think Rikki is keeping the light blue bar. ;) Guess we'll have to wait and see. I know he plans on keeping the CSS class in place, however, so skinners can duplicate what you see now without modifying the HTML code.

You have to modify either class_db_mysql_client or class_db_mysqli_client (depending on which one your server uses) so that right after mysql_connect or mysqli_connect you run either mysql_query( "SET NAMES utf8", $this->connection_id ) OR mysqli_query( $this->connection_id, "SET NAMES utf8" ), depending on the file.

While we may test some things in beta releases of browsers, we can never properly support a browser beta release. They have their own bugs which we shouldn't be attempting to correct for.

When FF 3.1 or 3.5 or whatever version they decide to slap on it is released final, we will of course correct any problems within IPB that affect functionality in the new release. Until then, we have to code for Firefox 3.0, as it's the current stable release.

As for why skin bugs aren't worked through as quickly as others - there are several reasons

1) There's one designer and 3 backend developers
2) A lot of the remaining skin bugs are much more challenging to correct than earlier bug reports. Sometimes one bug can take several hours to correct, or requires huge amounts of code to be refactored to correct.
3) Rikki has other work to attend to simultaneously (such as the lofi skin) which may not be listed in the bug tracker, but still needs to be done and takes time.

You have to run a query after the mysql connection is established.

SET NAMES utf8

In IPB3 you can edit conf_global.php to tell it to do this - in 2.3 you need to modify the database driver to do this.

aurora, look past the lines. That's not what I'm saying. I'm simply saying that there are probably a lot of client of IPS that are upset over the fact that all of the renewals have been paid without a software update to the next version.

Brandon, what I've been trying to clarify is that every summer, we would see a release 2.1 (2005), 2.2 (2006), 2.3 (2007) and so on. What I'm trying to explain is that some are wondering why it took two years for the new version. A new version of the software wasn't announced until sometime this past summer ... I think some are just concerned why it took so long to get the new version out.

2.1 -> 2.2 was a point release.
2.2 -> 2.3 was a point release.
2.3 -> 3.0 is a *major version* release

That's what's different.

The underlying code architecture for 2.1 -> 2.2, and 2.2 -> 2.3 did not change much. We had an ipsclass.php file with most of the functions. Folder structure was the same. The biggest changes were various features and bug fixes.

3.0 is virtually a complete rewrite, with an entirely rewritten architecture, and as such it takes longer than a minor point release.

We have never given any specific expectation as to when new versions will be available, outside of official date estimates (which are estimates only). I apologize if you believed there would be a summer release this past summer of 2.4 or 3.0, however there's nothing we've done specifically to give you that expectation. :)

No, why would it?

This is a BIG update, not a minor point release. It takes a lot more time to develop such a release. A lot of what we have done with 3.0 is specifically worked to make it easier to build updates and new features quicker in the future.

What's going to force the IT departments to upgrade however? If web developers continuing working overtime to try to make these fancy new apps work on IE6, IT department has no incentive to ever upgrade their browser status.

Unfortunately, it's going to take a push from everywhere....but it seems like everyone doesn't want to be "the first" (and we're hardly the first...)

Exactly. killswitch - I see you telling someone else not to skim your posts, but perhaps take some of your own advice and not skim mine. :) I never said release would be in June. In fact, if we released tomorrow (which we're not) it would be before June, and my statement would be perfectly accurate.

We don't know exactly when we will release. Contrary to what some here believe, we also do not base our release dates on our competition. We are setting goals for ourselves to release the software in a stable, working, reliable and secure way for our customers. If that means holding back the release for "6 months" from our original plan to ensure that everything is fully functional, stable and secure - so be it. This is a huge release, a big rewrite from the 2.3 series, and we are aiming to do it right, not "fast".

nooo, june is too late allowing vb to catch up.

Should have stuck to the core forum rather than running off with all the plugins

Uh, you misread my post. I said before June, not in June.

Additionally, we absolutely cannot release the forum upgrade without the components. Customers have purchased these components from us and it is our responsibility to ensure there are compatible versions available for the new version of the forums.

I'll simply say that our target release date is before June. Targets are hit sometimes, and missed others, but we're winding down now with bug reports. Largely the bug reports we've been getting lately are simply duplicates.

IPB 3.0 will not feature tagging. A future version likely will.

As far as I know (though admittedly I haven't researched), vBulletin is the only forum that does what you are suggesting.

I think it's safe to say that the consensus amongst the developers here (myself included) is that we do not wish to change any of the links to load the post separately. It is very easy to cause confusion when a post is shown by itself, and outside the context of the thread in which it was made. Overall we prefer the linking functionality we have in place as it is.

IE6 is still very high in Developing countries such as India and Pakistan, majority of users have a dial up connection and it is very hard for them to download IE7.

We also can't develop our software solely around the capabilities of developing countries. While we try to cater to as many people as possible, the fact is that most of our customers are from countries where the ability to download IE7 is not an issue.

I still remember downloading Win XP SP1 over dialup - it can be done if someone has the desire. ;)

1) You can assign a stylesheet using special comments in templates that IE reads

include ie6 stylesheet here <![endif]-->

<!--[if lt IE 7]>

2) IPB3 supports assigning a skin to a user based on their user agent already. No need for a modification or anything there.

3) IPB3's user agent support is greatly enhanced - IPB already stores the browser properties (so the detection part is done).

As a topic starter I will clarify my words - I was asking about "joinable usergroups"

phpbb - has this feature far ago from 2.X branch

vbulletin - has this feature since 3.7.X

IPB has this only as a mod Clans - and great thanks for them.

Users can:

- create a group

- join / leave group

- group leader can create forum category for group members

this feature is a must - I don't understand why ipb developers don't get this.

It's a "must" for a relatively small niche set of forums. I have yet to *use* a forum of any software that truly utilized such a feature tbh.

That said, we've simply had other priorities. We may or may not add such a feature in the future, but it wasn't considered a must-have for 3.0.

NBC:

http://boards.nbc.com/nbc/index.php?act=idx

Nvidia:

http://forums.nvidia.com/index.php?act=idx

For NBC, this is the stock 2.x skin with a simple bit of html in the wrapper around it to "integrate it" into the rest of the site. For Nvidia it is nearly stock entirely. If they did this with IPB 3.x it would have the problems I discussed with IE6.

Firstly, I assure you, boards.nbc.com is pretty heavily customized all in all. Trust me - I've done a lot of work on it. ;)



At the end of the day, pushing support for IE6 for another year does JUST that, and we would be having this same discussion then (just like this discussion came up a year ago, and when IE7 first came out - those points we did agree it was too early to move away from IE6).

For larger sites this is an issue and I feel like IPB is not considering customers in this situation. I am sure their corporate and larger customers such as NBC and others who use IPB will take ill to this decision.

Absolutely none of our corporate customers use our default skin, so this argument is irrelevant. It is possible to make a consistent skin if you want to (albeit likely without all the flashiness in the current skin).

I'm surprised another topic has been opened about this tbh. :unsure:

The reason people don't bother upgrade most of the time is become nothing forces them to. It was the same way with PHP5 - had tons of nice features and functionality we've been wanting to put to use, but because hosts didn't want to upgrade because nothing forced them to, nobody releasing an application available to a customer base could really make use of the new features for a very long time.

IE8 is almost out. IE6 will be EOL very soon. At some point you have to move on. There are potential alternatives for IE6 users, but we can't concentrate on supporting 10 year old technology. It's just unrealistic. The bulk of the big issues will be fixes as Charles has stated. But it's about time to move on, and the longer people (including us) cater to the outdated browsers, the longer people will cling to them, since there's no reason not to.

This is a setting in IPB3. It's already added. You can opt to turn it on or off at your own will.

It's a hook. Users can create and edit hooks at will. I think recent topics is more valuable to most busy sites (replies would simply mean the list wouldn't sit still long enough to view much of interest usually), but that's of course an opinion and I'm sure it's one that differs from many.

That's the beauty in the hooks system.

Your quote content doesn't seem to really be a bug - you have incomplete url tags (so they didn't parse), though I don't know how you submitted the content originally to compare.

There are a plethora of ban options in IPB3...

From the warn panel you can "ban" indefinitely, or for a specific time period. This does not change their group - it works identical to IPB2.

From the ACP you can access the ban panel for a user where you can "ban" the account (this sets a flag on the account, banning them from the site, without changing their group or anything else), and add their data to the ban filters (IP address, username and email address). You can also, of course, still move a user to a banned user group should you wish, however it is not necessary (if for instance you didn't want anyone to know the user was banned).

There is a blog entry from middle of last year in our blog on this if you are interested in reading about it more.