Moving On - The Future of phpBB Development [split]

[Obsidian]

No! None of it is true!

Well, ok, so all of it is true IMO.
It's really amazing how just about everybody I know who was really into phpBB has really lost interest in the project even though it's development has opened up with github.

Indeed, I'm honestly looking at how hard it would be to expand my current microforum's codebase, clean it up a bit, and release it. it's not the most feature-stuffed forum script available, but it's straightforward, which is what I've had my members complement me on.
Only issue is that I don't have that much time right now to do it; most of the backend's nice and smooth, but I've got upstream library changes underway that'll break just about everything again as soon as I finish up and put out the new major framework release.

I suspect a lot of this would relate to how the project's been run (into the ground) - PR for the teams is a nightmare, with poor attitudes all over and egos aplenty to step on, it's hard for anyone to get a decent support experience on .com. And by support, I don't just mean software support, but development support. There's a lack of mutual respect between the team members, all of them, and the userbase. [sarcasm]I wonder why...[/sarcasm]

I think we're all moving on to more advanced software because phpBB is getting to be ancient at this point.

Let's also be honest, it wasn't the most well written script either. There's a lot of fudges all over and unorganized code smashed in together within a few files (lol functions.php) which makes it a nightmare to try and work with, without having to rip chunks out and replace them entirely.

EDIT:

No! None of it is true!

Well, ok, so all of it is true IMO.
It's really amazing how just about everybody I know who was really into phpBB has really lost interest in the project even though it's development has opened up with github.
I think we're all moving on to more advanced software because phpBB is getting to be ancient at this point.

I agree and think its really sad.
They lost me with the word Symfony anyway. They just needed to address the codes state and move on to improvements and keeping up with the times.
Not adding a ton of overhead like an external framework.

Anyway, I just spent the last 2 hours tracing the topics and reading through them. (Eg. from the Gist link posted above.)
Reading some of the responses to the topics and people involved only proves to be more upsetting.

Well, moving to a framework would actually reduce a good bit of code duplication, wtfs, and oddities IF it was a decent framework (things like an ACL, post parsing, input validation are all fairly universal and can cover multiple software platforms with ease). If anything it'd probably reduce the number of issues we'd see with the software - if you find a library that does something, does it well, does it safely, does it fast, why not leverage it?

From my reading though, there's not been much of an attempt at 4.0. It's just stagnant all around, 3.0.x, 3.1.x, 4.x.

[Handyman]

Indeed, I'm honestly looking at how hard it would be to expand my current microforum's codebase, clean it up a bit, and release it. it's not the most feature-stuffed forum script available, but it's straightforward, which is what I've had my members complement me on.

But is it easy to add on to? If so, users can create libraries/plugins/modules for just about anything to make it as advanced as they want it. The base just needs to be solid, which phpBB isn't.

[Obsidian]

Indeed, I'm honestly looking at how hard it would be to expand my current microforum's codebase, clean it up a bit, and release it. it's not the most feature-stuffed forum script available, but it's straightforward, which is what I've had my members complement me on.

But is it easy to add on to? If so, users can create libraries/plugins/modules for just about anything to make it as advanced as they want it. The base just needs to be solid, which phpBB isn't.

The base itself is an MVC with a Front-end controller, so some things are easy to tack on (routers are <3). Others, not so much - it's something I skipped when I was trying to make the site launch deadline. Easy enough though, as at its core it's build around the principles of event-oriented design.
'sides, I've got a library for handling an "addon" structure of loading and running things which I am currently using in Yukari. https://github.com/emberlabs/materia It's worked for other things I've used it in so far.

[arch stanton]

I haven't been on here for a while but this thread does make gloomy reading.

Do people still feel this negative about the future of phpBB? I've seen mention of more advanced projects but I'm not aware of any better forum solutions that are open source. The lack of integration with social media is a serious downside of it for me - the developers don't seem to be very forward-looking at all.

[Highway of Life]

Do people still feel this negative about the future of phpBB? I've seen mention of more advanced projects but I'm not aware of any better forum solutions that are open source. The lack of integration with social media is a serious downside of it for me - the developers don't seem to be very forward-looking at all.

It's a little sad, but yes... I still do feel this negative about phpBB. As far as open source (or even proprietary) Bulletin Board systems go, phpBB is still the best out there that I'm aware of. BB's are just not a fast moving development industry at the moment.

phpBB 2.0.0 was released on April 4, 2002, 5 years and 8 months later, December 13, 2007, phpBB 3.0.0 was released. This "gap" in development was considered at the time to be unnecessary. Development was to progress much faster for phpBB 3.2, 3.4 and 4.0. It has now been 5 years and 19 days and we've yet to see the next major phpBB release. It's rather sad, for such a popular piece of software for development to progress so slowly. I left the phpBB team in 2009 after things just never seemed to progress. People were working hard, but the project didn't have good direction and would change direction quite frequently.

On the flip side, look at projects like Drupal, Joomla!, Wordpress, PyroCMS, FuelPHP, other OSS projects progress significantly faster, with both smaller and larger development teams. That is where I put my effort in now, I like to see fast moving software which is able to release quickly and develop with the times and the needs of the community. phpBB doesn't seem to be meeting that need at the moment. There are a lot of features coming in the 3.2 release, but it needs to get here soon, the phpBB worldwide community has been shrinking for years and they need to give it a boost with a new release.

I was very disappointed to hear they went with the Symfony framework, which is not an easy to use or easy to learn framework. The majority of third-party developers for phpBB are not expert level and need something that's simple and easy to work with. CodeIgniter would have been a much better framework to go with. If I were to do it now, I'd say FuelPHP, but that wasn't around back then.

[topdown]

phpBB has a lot of huge hurtles missed that has caused a lot of its community involvement.

1. phpBB has no backing, completely community developed.
This is a huge issue because its very limiting in core contribution. Developers can only dedicate so much free time before than need to work for an income.
And you can not expect them to.
2. phpBB is a forum system, simply servers one function.
CMS's server just about any function because the decent ones are plug-able or modular.
Thats not to say that phpBB can't be used for other things, its just not designed or meant for it.

Examples.
CodeIgniter not a forum but a OS framework has ExpressionEngine and the creators ellislabs backing it.
WordPress has Automatic the owners backing it.

Both of these create new opportunities for the project including funding.

Another issue is creating an income from phpBB is very difficult because the market is small in comparison to other options like CMS's that do everything.
This posses a major issue because if you can make a good income from an OS software, you generally will contribute back.

Example, WordPress again, thousands of people and companies generate an income using WordPress.
A lot of them dev companies, the big ones all contribute back, because they can, their income is supported by the product they are contributing to.

Then we have the development cycles of phpBB.
Consumers want features noooowww. Its always been this way and always will be.
Developers want to be able to understand the code they are working with and be able to work with it efficiently.

So the dev cycles are so far behind to satisfy the consumer that they are/were leaving at a rapid rate, also because they realize that phpBB only solves one thing they need for their site.

On the developer side, the devs spent all of their focus integrating a bloated framework that will not make enough improvement in the later mentioned issues to make it worth dealing with as an outside developer. Also as HoL said, its a difficult and very large framework to learn (just to do side stuff for fun ). Not to mention the time wasted on PSR compliance. 98% involved outside of phpBB's development don't even know what that is, or care! Servers no purpose to waste time on it.

Instead they should have focused on what it is and refactored, saving a ton of framework files and un-necessary bloat.

As Highway of Life mentioned its all sad, because phpBB was and still is a great product for what it does.
But I fear that these issues have been quit damaging and the fact that the market is so slim that it will never change.

phpBB as a product should always focus on the consumers needs and wants. Thats all it has, consumers..

[Highway of Life]

For #1, I agree, it is a big issue... they need to get a corporation specifically for commercial support of phpBB which would give it a huge boost not only in as far as development is concerned, but also with enterprise level companies.
However, there are many other OSS without backing moving significantly faster. Again, the problem is the rapid changes of direction and poor project management. This was the case when I was on the teams. I didn't realize it so much back then, but do now with a lot more experience. phpBB just won't be able to compete in the big leagues without some sort of corporate backing.

In addition to your examples, Drupal has Acquia backing.

After working with some of these systems (Joomla, Wordpress and Drupal, specifically), phpBB has much better quality code. A much better, but severely lacking API. We were always missing components that would really support 3rd party development well. You had to hack too much code. That's not necessary in Wordpress or Drupal, there's no code hacking there, and that's part of what makes those better platforms even with uglier code and poor-practice API.

For #2, if it were me, I would be working with other groups, Drupal, Joomla (we have to a limited extent), Wordpress, and other up and coming CMS systems on integrating the environments so that you become very plug-and-play. Drupal alone has a massive community that phpBB could tap into if they were to release a Drupal phpBB module. But phpBB isn't really designed for that. It would take quite a bit of effort and a rather nasty looking bridge. (It would regardless, Drupal code is very ugly)

This posses a major issue because if you can make a good income from an OS software, you generally will contribute back.

This is true, it was our motivation for wanting to start ModGalaxy.
phpBB has had the mentality of being very anti-commercial in nature, so they've generally opposed to the way that many people like to make money. Far more than Wordpress, Drupal, and Joomla have been... this could be a significant reason the product has had a lack of life. I even took on this mentality as a phpBB team member. Anything commercial was very frowned on. I think that's why STG gained a good deal of popularity between 2007 and 2010, we didn't squelch the community nearly as much.

Consumers want features noooowww. Its always been this way and always will be.
Developers want to be able to understand the code they are working with and be able to work with it efficiently.

Yes, plus put a difficult to understand framework in front of them (Symfony) and a challenging code-base, and you're only going to get a small number of contributors.

Instead they should have focused on what it is and refactored, saving a ton of framework files and un-necessary bloat.

Couldn't have said it better myself. It's frustrating enough that it makes you *feel* like you could do better in a shorter period of time, but there's a significant lack of monetary incentive to do take on such a task. Assuming you actually have the interest. phpBB just needs to be *fixed*, and fast...

[stef775]

Said well I believe, and don't hide because people want html post or the mass emails don't work. Tons of messages about it but no answers.

The market is asking for it....

[Highway of Life]

It's better now than it used to be. The mentality of the management team was very much "we do what we want because we really know what's best for the community" and they alienated the community and to a large extent the development community as well. If you can't listen to your fans and incorporate their suggestions, they're not going to feel very welcome, helpful, or useful. So all the smart people who could have contributed a lot more largely weren't there. It was a big turnoff. There were others who either liked the product so much, or wanted to learn, and phpBB was a decent enough place for it... those are the people phpBB attracted. So we had a large number of sub-par 3rd party applications. I say this somewhat with a tongue-in-cheek remark because other OSS doesn't have significantly better experience. Wordpress' modules contain security vulnerabilities and very bad performance practices. Drupal modules have better security but also use very poor programming practices. Although both are due to the platforms and API's that they are running on.

phpBB always had better code, but no API, so we couldn't integrate anything into it. Programming for it was always a pain and due to the level of integration, to a large extent all your plugins absolutely had to be the same license that phpBB was using. It didn't allow for much modularity which would have allowed for (potentially) other licenses. We were left with conflicts and difficulty upgrading sites (like STG) if you installed a lot of mods. At Disney, we use Drupal and have dozens of modules (often 50-70). If that were a phpBB site, you're dead in the water as far as upgrades go. Drupal is an instant, easy upgrade, even for all the modules. Same with Wordpress, which I can instantly update at anytime, same with the modules. PyroCMS? Same thing there... update with no dependence on modules, do them individually. phpBB makes people stuck. So good quality code has no benefit at this point if it's not really allowing people to upgrade. OR it means phpBB has to ship with *everything* you need or you're going to have a headache of a life maintaining it.

[topdown]

The problem with using external framework for a project like phpBB is that phpBB is in-itself a framework (for forums).
Strip the garbage out and what do you have left,
a module system that needs refactoring and abstraction
a template system, session system, DBAL, ACL, language system, cache system, and a couple others.
Dump the rest and create a true separated MVC that is plugable with an actual plugin interface and it would have been great.

Frameworks are adding a ton of functionality, files, functions, classes, abstracts, dependancies that are just not needed or wanted in a specific type software like phpBB.

External frameworks like CI, Yii, Symfony, etc... are designed and meant for building systems for sites quickly.
Like slamming out a custom storefront for a small company.
My opinion anyway.

I spent about a month looking and testing the best way to build a system that I wanted.
Symfony was not an option, to garbled for my liking.
Yii is about the same but with worse documentation.
CodeIgniter is easy, but they require more loaded than I wanted on every page execution.
PyroCMS is the same as it is using CI, but at least it has a module system.
2 other CMS's built on CI that were about the same as PyroCMS.
But these offsprings of CI start getting into 8-9Mb of RAM with just a simple site page. (Not good for scaling large)
Heck, I even stripped phpBB3, leaving just the stuff I mentioned above and tried that out.

So in the end I ended up building my own, no starters, frameworks, bla bla bla.
Just my raw code.
Modular MVC with REST API and ACL using plain old PDO CRUD handler in my models.

I ended up with a system that has database pages loading so fast locally that it seems like the browser did nothing.
And next to no memory.

Does it follow some Johnny Harvard meatheads standard, NO.
It follows My Standard which == Fast, scaleable, every .001 of anything counts, objects when objects are required, procedural when it is required and nothing gets loaded unless its required for the page to function.
Oh, and every line is tested as written when written, not with goofy Unit tests that can fail just as bad as the original code if any little thing is missed or wrong.

My point is if the software does not need all of the other garbage, whats the point in making something bloated and hard to understand.
Write it for what you need and only what is required to make it function properly.
I think phpBB3 had more going for it than not in the codebase.