2.0: Joining forces with Flarum

Franz · 2015-03-16 09:15:25

It is time for an update on the status of FluxBB 2.0. Development has been progressing, albeit very slowly. This is, in part, caused by time constraints, but there are many other things that could be better: I decided to develop the forum using components from the Laravel framework, a tool I am very comfortable with, but which none of the other developers have used before. I could have done better in making it clear which tasks can be taken up by volunteers, where I need help, from things that are easy to pick up to complex architectural decisions. Communication, basically. Things have to get better.

Therefore, I am delighted to announce that FluxBB 2.0 will from now on be part of the Flarum project. Flarum is being developed by Toby Zerner, the creator of the esoTalk forum software.

What is this Flarum thing?

The vision for Flarum is very similar to the one we have for FluxBB 2.0: provide a very basic forum software that brings all the essential features a forum needs, and nothing else. Users should be able to customize their forums to their liking, using extensions that are easy to install and can modify and extend the forum core. Toby outlined his ideas for Flarum in a post on his blog. Flarum is currently at a point in its development that's very similar to that of FluxBB 2.0. Its backend is built on the same foundation (the Laravel framework), so we can truly merge some of our code.

What does this mean for FluxBB?

First of all, nothing should change for you. We will continue to support the 1.5 branch, providing regular maintenance releases as we used to. When Flarum hits beta, we will start providing an upgrade path for users of older FluxBB versions. But don't worry, we won't leave you alone with your FluxBB forums. As long as a significant fraction of users still run FluxBB, we will continue to support the latest version, at least with security fixes.

We believe that this decision is for the very best of the FluxBB community. To be honest: seeing the previous pace of development, FluxBB 2.0 would have still been a long time away. With Toby taking time off college to develop Flarum and determined to implement his roadmap, you will likely be able to test-drive a new forum soon. Also, our skill sets are quite complementary: Toby will work his frontend magic to deliver a pleasant user experience, while I will focus on developing the backend and making that as fast as possible.

I encourage you to check out a very early demo of Flarum to see what it's all about.

If you're interested in contributing (code or wise thoughts) to the development of Flarum, head over to the development forum and post away.

Please let us know any feedback you have - we want to hear your comments, questions and concerns.

fgreinus · 2015-03-16 09:51:59

Does that mean, that there'll be a new "product" or are you just merging fluxbb2 code into flarum?

Don't really like, that flarum uses ember.js for frontend-things. For simple "install-and-use" cases this might be a nice choice, but for integrating it into an existing project it will possibly be pain . What's your thoughts of making it easily integratable?

But, don't get me wrong, i like the fact of joining forces!

Franz · 2015-03-16 10:14:55

I'll see which ideas from the 2.0 code I can get into Flarum.

Toby predicted there would be worries about ember, so he wrote another blog post explaining the decision.
Integration is an important issue for us, so we will definitely target that. Feel free to contribute to the direction this will take in the development forum.

tobscure · 2015-03-16 10:23:51

Integration is an issue I forgot to address in the Ember blog post, but as Franz said it is definitely important and we'll make sure there's a good solution

- Toby (Flarum guy)

Last edited by tobscure (2015-03-16 10:24:08)

fgreinus · 2015-03-16 10:41:08

Thank you two guys for your super fast replies.

My concerns are about one specific case. We already have a ui based on bootstrap and some other extensions, but basically css based with some ajax-plays in it. So integration the forum software into our portal- and system that has already been built around the forum while using another ui "system" for it is really risky. Possibly you create a super fancy ui for the flarum app and it all works super nice together. But a, in my opinion, more realistic case is, that you get into trouble when maintaining two different ui-sets and two different ways of doing things.

But I'm really eager to see what you'll come up with .

muli · 2015-03-16 15:42:10

I'll just hope, that the new version isn't a kind of Vanilla-Board, because the demo has the same "flow" of it (demo.flarum.org) - means tons of JS and so on.

But still looking forward and be curious for the new version!

Last edited by muli (2015-03-16 15:43:53)

zaher · 2015-03-16 16:48:48

I looked at it, I dislike it.

Eh, I will start my own one , but not a forum, I dreamed about community for developers, markdown instead BBCode, Simple fast like FluxBB, no framework. etc.

Franz · 2015-03-16 16:51:11

What do you dislike? Nothing is set in stone yet, we're experimenting and looking for feedback.

cyberman · 2015-03-16 17:37:54

muli wrote:

I'll just hope, that the new version isn't a kind of Vanilla-Board, because the demo has the same "flow" of it (demo.flarum.org)

Thought the same ... and I'm not really inspired yet (like on FluxBB some years ago) .

If I want a vanilla board, I can use it already yet and must not wait to FBB2.0/Flarum.

Is there a backend demo available too? What template engine will be used?

By the way - I accept the your decision cause I know (too) much about working on open source software. But I hope your influence will give it the FluxBB feeling .

Last edited by cyberman (2015-03-16 17:44:26)

Studio384 · 2015-03-16 20:39:39

So... FluxBB is coming to an end?

chris98 · 2015-03-17 09:30:40

I'm quite disappointed about this, I took a look at Flarum and I dislike it. For one, it's far too heavy on JavaScript, and the style isn't very pleasing either - it doesn't look like a forum to me, more of a quick, simple AJAX chat script.

I spent ages trying to find the perfect forum software, when I finally find it, it seems to vanish

GWR · 2015-03-17 11:04:05

I like the things like "auto append" of new posts, I also like that kind of "scrollbar" on the right - but this seems all. Such things could get appended even to a 1.5.x fluxbb (using some ajax magic and getters/api) - of course it is nicer to have it more tightly integrated (performance wise) but you also add it to people who do not want such a feature.

As much as collaboration helps gaining "power" (read potential "work hours") it also adds hours and days of fiddling around how to join two worlds (does not matter how similar they are).

For me, who is not into "timelines", "twitter" etc. it is not the step I want fluxbb to move into. There might be users out there who like that way, but I am sure we (who still play SNES games sometimes ) like to have start menues, gnome 2.0 (in the likes ) or other things we are used to have. And same counts for forum software. But like said, things like auto-append and ajaxification are things _I_ would like to see (or eg. warnings if posts were done inbetween).

bye
Ron

Franz · 2015-03-17 13:12:39

Hello folks, thanks for the feedback! Let me try to address your concerns:

muli wrote:

I'll just hope, that the new version isn't a kind of Vanilla-Board, because the demo has the same "flow" of it (demo.flarum.org)

I don't think the two will be very similar. Are you referring to having a list of recent discussions front and center? Not sure how or whether this will be baked in right away, but I'm sure there will be a simple way to start with categories like we always did with FluxBB.

Please remember that nothing is set in stone yet. The demo can give you a glimpse of what's probably coming, but things may change completely at any point. So please give us a chance!

Studio384 wrote:

So... FluxBB is coming to an end?

No. It will be around for a while in its current version, even when Flarum is released. At that point, we'll try to convince people to upgrade to 2.0 / Flarum. We'll see how long that takes, and then slowly end support for the 1.x versions, just like what is normal for old versions. Nobody will be abandoned.

chris98 wrote:

I'm quite disappointed about this, I took a look at Flarum and I dislike it. For one, it's far too heavy on JavaScript, and the style isn't very pleasing either - it doesn't look like a forum to me, more of a quick, simple AJAX chat script.
I spent ages trying to find the perfect forum software, when I finally find it, it seems to vanish

Well, obviously simplicity is a central goal with Flarum, performance is not yet very good, but we'll focus on that. Regarding JavaScript: This is still something that concerns me, and I'm not totally happy yet, either. I do think, though, that such interaction-heavy systems as forums can benefit greatly from client-side code.

What is it that you like most about FluxBB if you call it the "perfect" forum software? (Nice to read that.) Again, things are not totally set in stone.

GWR wrote:

For me, who is not into "timelines", "twitter" etc. it is not the step I want fluxbb to move into. There might be users out there who like that way, but I am sure we (who still play SNES games sometimes ) like to have start menues, gnome 2.0 (in the likes ) or other things we are used to have. And same counts for forum software. But like said, things like auto-append and ajaxification are things _I_ would like to see (or eg. warnings if posts were done inbetween).

Again, we still have categories, it's mainly a thing of layout to change back to the way FluxBB handled them. I'm sure this is something that can be discussed and I'm looking forward to your input in this area.

To all of you: please feel free to discuss your concerns. Join the discussion at the development forums and make your voice heard. There's a strong vision, but it's still a community project.

XAOS-Eric · 2015-03-17 16:32:46

Studio384 wrote:

So... FluxBB is coming to an end?

Not quite yet, I have a lot of code yet to be uploaded to github for 2.0, so I emailed Franz with a solution that might work for all of us. I would be willing to take over the development of FluxBB to keep it alive.

Franz · 2015-03-17 16:40:56

FluxBB 2.0 gets a new name, that's all. It's not coming to an end and it is most certainly not dying. Flarum's codebase is fairly similar to the code I had written for 2.0, and its goals are similar.

All that changes is that we suddenly have someone with a lot of frontend experience and talent working on the next version, too.

chris98 · 2015-03-17 18:28:14

What is it that you like most about FluxBB if you call it the "perfect" forum software?

I'll start with the main points that are so bad about the most popular forum software:

phpBB
1. Heavy and cluttered
2. Hard to navigate in (esp. the admin and moderator control panels)
3. Stupid and un-needed parameters (such as f=1&t=2 when id=2 would do fine) - this is done even when the forum id isn't needed.
4. Horrible styles
5. Unusual database structures
6. Things don't work (i.e. forcing signatures to x lines through admin)
7. Attracts massive, massive amounts of spambots
8. Certain words you're not allowed to search for because they are 'too common'
9. Two pages for profiles - one for editing your own and one for viewing others
Invision Power Board
1. Too heavy and cluttered full of rubbish - which makes navigation difficult. The features are good (on the most part, but have been implemented poorly)
2. Very Slow, often receiving host unreachable or page can't be displayed errors.
3. Database structure is not very well optimised at all, and is even worse than phpBB - one example alone being a 'banned' column in the users table rather than a banned table like in FluxBB.
4. You get errors in your error log - yes, although it may seem that because it's commercial there would be no errors, but this is simply not true, there are more than in FluxBB. When installing, you get database driver errors in your error log that aren't shown in the browser.
5. Commercial software should be 100% SQL Injection proof, and use prepared statements, which it doesn't
6. Uses $_REQUEST on everything rather than $_GET and $_POST as needed.
7. Uses far too much JavaScript
8. Two pages for profiles - one for editing your own and one for viewing others
9. No ability to create categories - they are instead forums themselves which is pointless.
VBulletin
1. Not very well optimised
2. Heavy and cluttered, and similarly to IP.Board navigation is cluttered.
3. Styling is horrible
4. Admin panel navigation is particularly bad to use
5. Full of unnecessary settings that forums don't need (once found after several minutes of searching) - i.e. ability to login from social networking accounts
6. Two pages for profiles - one for editing your own and one for viewing others
7. Stupid URL names and GET parameters
SMF
1. Is only compatible the deprecated MySQL_ extension and Postgre SQL
2. No group permissions are in place - instead, users take permissions from the forum permissions table
3. Has ridiculous GET parameters which will require an implode() to be used on them to be useful
4. Hard to search in, and has two pages for viewing profiles like the other software.

Really, for me, the fact that FluxBB is perfect is because all these things aren't a part of FluxBB. But also because no matter how much you add, it never gets slow, it always stays fast and always stays optimised. It's very easy to alter and doesn't have stupid classes like IP.Board (not that classes are stupid, but the way they are set out in IP.Board are)

FluxBB is very easy to alter, it's reliable and secured. While I would personally rather it use prepared statements, it isn't vulnerable to SQL Injection (apart from when using the default, deprecated MySQL extension). It's incredibly easy to use, everything is set out right in front of you so you have no problems navigating, it's not cluttered no matter how much stuff you add to it and it's now got a very good extensions system in 1.5.9.

Yes, there are a few flaws, like unsalted passwords using sha1 - but these can be fixed on a per forum basis, and don't drastically alter the behaviour of the forum. Next to the other software, FluxBB is a million miles ahead.

Everything in FluxBB makes sense, from the database structure to the code, and it seems intelligent the way it's been programmed - and that's why I'm so interested in it. Even the name has been cleverly thought out. That's all I can really say

It will be around for a while in its current version, even when Flarum is released. At that point, we'll try to convince people to upgrade to 2.0 / Flarum. We'll see how long that takes, and then slowly end support for the 1.x versions, just like what is normal for old versions.

So will you be adding new stuff to the 1.x branch now or just security fixes? I think it would be a shame to stop adding features to the old branch.

I may be just being critical, but I have a nagging suspicion that merging with Flarum isn't the right thing to do and is a mistake. Still, best of luck

Last edited by chris98 (2015-03-17 18:40:33)

benjawi · 2015-03-17 19:02:16

From what I've seen of Flarum, I don't like it. Doesn't look like a forum and I don't agree that you need javascript to make anything look nice.

I liked what we'd seen of Fluxbb 2.0 up until now and was really looking forward to it - would happily wait. But I have no interest in anything javascript heavy and believe that should be reserved for essential bits, like making a responsive menu.

I like 1.5.*, it's a fantastic forum system. Building on this instead of moving away from FluxBB would be the ideal way forward. Think I'd rather stick with what I have...

GWR · 2015-03-17 21:46:06

Regarding JScript:

Like said I really like little helpers adding convenience: alerting about new posts since starting to write your reply + appending the posts to the current view etc. I also think that certain moderator/admin actions benefit from ajax responses.
But of course all of this is "doable" without javascript (eg. reload the reply form if something changed inbetween - eg by comparing a simple "creation timestamp" with the "latest post" timestamp ...).

@chris98 list
Very comprehensive. I agree with many things on the list. But of course fluxbb isn't having only positive things on the sheet. Some moderator actions are just "inconvenient" (eg. merging posts from one thread into another one -> needs multiple steps: extract into a new thread, merge that new thread with the target thread). "Styling" in fluxbb isnt that easy as with other boards (I understand that "tight integration" adds to performance... but precompiled templates will do similar things).

Regarding security "fluxbb 2.0" will surely have more flaws than fluxbb 1.5 - this is the nature of new software. This will happen with a "flarum-companionship" too.

If the announcement would be "X joins to overhaul backend/moderator-ui/convenience" I would be for it ... but now it will take a bit of time until I might think differently to it. I am not very into Laravel at all (or other frameworks ... also CodeIgniter and other "Oldies but Goldies" of the framework-era). But this is my personal opionion for "2.0" too - so that does not count in this thread.

bye
Ron

Franz · 2015-03-17 22:39:49

benjawi wrote:

From what I've seen of Flarum, I don't like it. Doesn't look like a forum and I don't agree that you need javascript to make anything look nice.

We're not adding any JavaScript to make things look nice. It's about making things usable and interactive.

In fact, with AJAX requests, JavaScript can improve performance, because only parts of new pages have to be loaded (without the header, menu etc.), and JS and CSS files only need to be loaded once.

Gil · 2015-03-17 22:50:20

Is it possible (or too much difficult?) to imagine a totally common database structure and totally common backend, and two different front-end offers, one with a flarum-like design, and another one nearer the current fluxBB design? I think you also can get some good ideas from Luna (both back-end and front-end).

cyberman · 2015-03-17 23:30:13

Franz wrote:

and JS and CSS files only need to be loaded once.

By the way - this could be effectively done with HTML5 AppCache too ...

Franz · 2015-03-18 00:00:05

Browsers already cache them anyway if the server is configured correctly.

Still, there's lots of potential for speedups when JavaScript is involved, e.g. with only requesting more-or-less raw data from RESTful APIs.

eric235u · 2015-03-18 00:03:47

Thanks for all the hard work Franz! I just signed up to the Flarum forum and will lurk on github.

chris98 · 2015-03-18 09:09:34

I agree with many things on the list. But of course fluxbb isn't having only positive things on the sheet. Some moderator actions are just "inconvenient" (eg. merging posts from one thread into another one -> needs multiple steps: extract into a new thread, merge that new thread with the target thread). "Styling" in fluxbb isnt that easy as with other boards (I understand that "tight integration" adds to performance... but precompiled templates will do similar things).

That list is just a few of the biggest things which I thought off the top of my head, there's probably a lot more. Absolutely FluxBB doesn't only have positive things on the list - but unlike the others, it's the most normal.

XAOS-Eric · 2015-03-18 12:26:21

chris98 wrote:

I agree with many things on the list. But of course fluxbb isn't having only positive things on the sheet. Some moderator actions are just "inconvenient" (eg. merging posts from one thread into another one -> needs multiple steps: extract into a new thread, merge that new thread with the target thread). "Styling" in fluxbb isnt that easy as with other boards (I understand that "tight integration" adds to performance... but precompiled templates will do similar things).
That list is just a few of the biggest things which I thought off the top of my head, there's probably a lot more. Absolutely FluxBB doesn't only have positive things on the list - but unlike the others, it's the most normal.

While I like the idea of Flarum, if Franz doesn't have time to work on FluxBB 2.0, I feel that he should hand over the reigns of FluxBB to someone that can make FluxBB 2.0 the way it was ment to be, made entirely with php and not with ember.js. Sure I see the reasoning behind Franz decision, but I have told him via email that I would be willing to take over development of FluxBB. This is a community, one person alone should be able to decide the future of FluxBB, if someone wants to continue on with FluxBB 2.0 then Franz should let them take over development.

Forums

#1 2015-03-16 09:15:25