The future of FluxBB

xable · 2008-12-24 19:56:56

Good to see a thread like this...

On reflection I agree with bumping the major version number.

I don`t agree with working on 1.2 beyond security updates, at least not until 1.3 is final and only if there was enough demand. (I`m not convinced there would be)

A template system should come after final and waiting for Paul`s markup revision to be completed and working on everything else in the mean time would be the best way forward imo. That said, it may be better in the long run if a template system was implemented sooner rather than later, depending on what the consequences would be from haveing implementing it later.

AracornRed · 2008-12-25 04:40:28

This is my first post on the FluxBB forums. I've been a member of the PunBB forums since before "the split" and run a heavily modified version of PunBB 1.2.~~. I was just on the brink of moving to v1.3 until I learned of the complicated development and political situation behind both branches of the forum.

As a user who has invested a lot of time into v1.2 modifications, I don't want to upgrade to a major new version unless I'm sure I won't have to do so again for a long, long time. Once the development situation on 1.3 gets sorted out (and it definitely looks like more help is needed somewhere) it looks to be well on its way to being that ideal forum. The implementation of hooks is a major step, but IMO a comprehensive templating system is the other essential element. I want to be able to add/delete/move specific elements - from the pagination links to the number of thread replies to the poster's web site link - without touching the core php files.

How that applies to the future version path - Security and performance upgrades should continue to be applied to v1.2. I would avoid any feature enhancements, but if utf-8 is a highly requested feature and can be done without major modifications, then go for it. I say leave the other new features for v1.2 to mods. I also think upping the version number is a bad idea. Too confusing and I can think of no reason you'd do so other than to "out-version" somebody else.

Work should be continued on v1.3beta with the stipulation that it is the development version, in that continuing updates will make it progressively more stable but with the possibility that future updates will render stylesheets or extensions broken. I would keep v1.3 in beta and progressively up the version number until the hooks are completely sorted, a comprehensive templating system is in place, and all the major new features are implemented. Then make it the official v2.0 and end development on 1.2.

Some of the new features in v1.3 look really nice (proper "topic read" handling, move posts to other topics, extensions!) but I would prefer to see it fleshed out completely before official release. Of course people who have already moved to v1.3 for live forums probably feel otherwise.

ridgerunner · 2008-12-25 06:41:21

1.2 is the culmination of years of punBB evolution and incremental refinement. It is the most stable, solid and widely implemented code base here. I think the old saying applies: If it ain't broke, don't fix it! Of course in the unlikely event it does break, do fix it (and apply security fixes as well), but its features should be frozen to keep the many current 1.2 users running smoothly.

1.3 should not be released until it is really truly finished (i.e. its requirements are met). Call it 2.0 if you must (and with all the major changes, this is probably a good idea), but I see no problem with staying the course and keeping it 1.3. (After all, there is historical precedent in the project's version numbering system - yes?) But please, take whatever time is necessary to: Do it right the first time! The developers will need to put up with some impatience from the community (some of whom may even leave), but once it is done, and it is good, all will be forgiven and happiness will return.

But beware of feature creep. It can kill a project - (it killed the last company I worked for due to delayed release). When working on a major project (like FluxBB), it is very easy to think up new improvements that can be made, but one must not succumb to the temptation to incorporate them right away if they require significant design changes. Thus, to be successful, it is critical that the software requirements be well defined in advance and once those requirements are met, then it is done. Period.

For me, the extension system is the main requirement for 1.3/2.0. And I don't mind waiting for it, but I would like to be informed of a ball park release date, because if it is realistically 6 months to a year (which seems to be the case), then I'll just go ahead and (happily) use 1.2 in the meantime. (One thing I've learned in 30 years of programming is that software always takes longer than your first estimate - frequently: 5-10 times longer!)

Cheers and keep up the good work. Good thread BTW.

Dr.Jeckyl · 2008-12-25 09:03:10

I'm here til the end but I have some small input.

1.3 -> 2.0 not a bad idea.

1.2 should have a set, hard and fast EOL date. Say, July 15th, 2009. That way you guys can concentrate solely on 2.0.

Whatever you guys decide to do I'm sure is for the better. I'm just sick of the negativity as of late. It doesn't help one bit and can possibly hurt progress.

Last edited by Dr.Jeckyl (2008-12-25 09:03:44)

elbekko · 2008-12-25 12:59:17

ridgerunner wrote:

But beware of feature creep. It can kill a project - (it killed the last company I worked for due to delayed release). When working on a major project (like FluxBB), it is very easy to think up new improvements that can be made, but one must not succumb to the temptation to incorporate them right away if they require significant design changes. Thus, to be successful, it is critical that the software requirements be well defined in advance and once those requirements are met, then it is done. Period.

We're well aware of that, and we've done that. But we also realise that sometimes a feature is too essential to keep neglecting, like the templating system.
Let me assure you, don't worry about feature creep. If anything, features get removed more often than added

ridgerunner · 2008-12-25 14:30:30

elbekko wrote:

... But we also realise that sometimes a feature is too essential to keep neglecting, like the templating system. ...

Yes, if templating truly is important (and it sounds like it is), and incorporating it requires major tweaking to the design of the core (and these changes are intertwined with the extension system), then I would say: by all means take whatever time is necessary to do it right (and it should probably be called v2.0). Personally, I have no problem waiting a year or more - (This will just give me the opportunity to get to know 1.2 up close and personal ). But in the big picture, I'd recommend working only one active development track (1.3/2.0), but continue to support the 1.2 track in a "bug fix only" maintenance mode.

p.s. Ho, Ho, Ho!

Franz · 2008-12-25 17:59:25

What kind of changes would make this software v2.0, if not those we are experiencing now?

horus · 2008-12-26 09:09:30

ridgerunner wrote:

1.3 should not be released until it is really truly finished (i.e. its requirements are met).

I quote this sentence to remark the "requirements" word. Reading home and about page on this site the big requirements of 1.3 version are Capablility of being served as application/xhtml+xml, One click install extension system and Full support for UTF-8. I think these requirements are already here, I (and maybe other peolple) am waiting for these requirements to do a step forward, to make free from comparing files at every release. In addiction there is a tableless markup that I like very much and many improvements that perhaps I neither know.
So I would like to see a 1.3 version with these "simple" (and quite already done) requirements, then a 2.0 version with the complete markup revision, templating system and other important features. In my opinion make all changes in one step only is too much.

Christian · 2008-12-26 09:51:28

Release the last working 1.3 as it is (the few style bugs can be fixed quite easily). And leave anything else for 2.0. This would also give you more feedback on the new 1.3 features which then can be considered while working on 2.0!

Strofanto · 2008-12-26 09:58:23

Christian wrote:

Release the last working 1.3 as it is (the few style bugs can be fixed quite easily).

At this point I definetely agree, Paul stated he's done with the markup, if it is so the community can help with the CSS. That would make the 1.3 the so wanted beta 3 and the devs can fix the long known bugs.

Meow · 2008-12-26 11:42:46

I absolutely agree Cornor's plan.

1.2 has existed for over 3 years. The structure now became so old, why not just release a new version which is based on it to end the life of 1.2? So there should be 1.4.

I don't think 1.3 should be released. If it were released, people still had to spend much time on upgrading their forum to 1.3. Why not only release 2.0? People just need to spend much time on upgrading their forum once, not twice.

1.3 looks like a small improvement, and 1.3 is fresh no longer. Why not do a larger improvement such as developing 2.0? 2.0 would be a revolution of FluxBB, and FluxBB will rewrite the history of Internet forums.

I hope there would be more people who want to support Cornor's plan. His plan is absolutely the best plan for FluxBB.

Strofanto · 2008-12-26 13:52:32

Meow wrote:

Why not only release 2.0?

Because 2.0 could take another year or more, 1.3 might be released in a few weeks or months as soon the markup is done and a few bugs are out. Or if not so, it might be still useable when the markup is committed.

Meow wrote:

People just need to spend much time on upgrading their forum once, not twice.

Upgrading the forum takes an FTP transfer, a php script run, a few options to change in the admin panel. Can't take more than a hour.

Meow wrote:

1.3 looks like a small improvement

A small improvement to what? If you're talking of 1.2, I wouldn't call that a small improvement.

SuperMAG · 2008-12-26 14:21:43

as i said, if paul wants more time give him more time by just doing these stuff:

paul do mini css fixes the small important problems and then release beta 3. (should be release in a week at most)
and when paul complete his markup css, release RC1. (should be released at most 2 or 2.5 or 3 months)
and then small bugfixes in RC2. (should be done in 3 weeks)
Then Final. (give it 2 more weeks of no bugs)

in MY openion that is the best strategy, or if you want to change the 1.3 to 2.0, then just change it.

Thinking (unless the plan of this topic is to get away from angry posts lol)

Last edited by SuperMAG (2008-12-26 14:27:21)

jmp · 2008-12-26 15:52:33

Rich Pedley wrote:

1.2 shouldn't be developed any more - sure maintain it for now with security releases but don't do more with it. Keeping it going will only dilute what resources you have.

I disagree. I think the 1.2 branch is the most robust, stable and in general best piece of forum software I've ever used. For me (and I don't think Smartys is wrong when saying that there are others who share the same view), it's more than enough. I don't need a one-click extension system or a very advanced template system. Apart from UTF-8 support, it pretty much contains all the crucial elements a web forum needs, in my opinion.

The point about resources is probably a non-issue too, if the solution Smartys suggested (having separate developers working on the 1.2 branch) works. Both branches would have their own devs, and neither would have to worry too much about the other branch.

As for bumping the version number to 2.0, I think it's a good idea.

Pedro · 2008-12-26 16:06:32

Becko, i mean no offense with this. It sounds like part of the plan is trying to fix some dealy in the release schedule with version strategy rearrangements.Though, I believe there's is absolute no intention to do so.

So it turns out that a more propper name for 1.3 would be 2.0. I defentely agree with that. However i don't think upgradding the version numer now, when there is already betas out there, is necessary, at all. It's too late for that.

Fluxbb already has a template system, but it is used in a too minimalist way. All the contents should go individual placeholders, including the ones in the post design which needs to be looped.
This would IMO be a fairly good improvment and it would make site integrations a breeze.
Now that we already have 1.3 betas, why not using a full blown template system as 2.0 biggest change? After all it would be THE release with a clean core, zero html.

As for the extension system... what's missing on the current (1.3) one? Looks awsome to me...

MattF · 2008-12-26 20:07:05

Jarkko wrote:

Rich Pedley wrote:
1.2 shouldn't be developed any more - sure maintain it for now with security releases but don't do more with it. Keeping it going will only dilute what resources you have.
I disagree. I think the 1.2 branch is the most robust, stable and in general best piece of forum software I've ever used. For me (and I don't think Smartys is wrong when saying that there are others who share the same view), it's more than enough. I don't need a one-click extension system or a very advanced template system. Apart from UTF-8 support, it pretty much contains all the crucial elements a web forum needs, in my opinion.
The point about resources is probably a non-issue too, if the solution Smartys suggested (having separate developers working on the 1.2 branch) works. Both branches would have their own devs, and neither would have to worry too much about the other branch.

Pretty much sums it all up perfectly in a nutshell.

All of you people on here who are obsessing over running 1.3* and keep suggesting that 1.2* ought to be allowed to die off ought to remember that there are people with the opposite train of thought, namely that 1.2* is perfectly fine and preferable to 1.3*.
At the end of the day, 1.2* is tried and proven. 1.3* will not be in such a position for a long time.

That is also without taking any modifications/extensions into the equation. Cleaning other peoples sloppy and insecure coding up is not something I intend to start doing all over again, just so that one can run 1.3* for no tangible benefit or gain.

Last edited by MattF (2008-12-26 20:09:28)

yann · 2008-12-27 00:05:52

Hello Connor, I have been running punbb for more than 4 years, seen a lot of work done and sent several patches myself.

I have to strongly disagree to your proposition though, and to some comments that were made before "get it right the first time". I believe you can't get anything right the first time, and am a fervent believer of the "Release early, release often". Read about the agile development model!

I remember exactly three years ago, it was written "punbb 1.3 will be released before christmas" - that was in 2005... this is also the reason you now want to change the version number to 2.0 instead of 1.3. You keep adding features, without stabilizing the current base.

I believe *nobody* expects a perfect 2.0. People would prefer a rock solid, simple 2.0 (which is all fluxbb is about), with regular releases adding functions, and an easy upgrade system. If not what is most likely to happen is that you will continue to develop without releasing, lose many users, at some point want to rewrite the code you wrote 2 years ago after starting to work for 1.3, and then ultimately get bored as you lost all your users and the workload just became too huge.

Please reconsider your position - you don't have enough people to maintain two branches. Maintain 1.2 with security fixes, release a rock solid, evolutive 2.0, and then add features, regularly, making sure the upgrades are easy.

Pedro · 2008-12-27 02:05:30

Jarko, matt and others, I see you are satisfied with 1.2. In fact noone of us would be here if we wouldn't all think punbb/fluxbb 1.2 is the best forum software ou there.
But allow me to point the goal of this discussion which is the future.

1.2 is a rock solid piece of software that works like a charm. Like any other software, it will become obsolete if it's not further developed.
Are you guys saying that a fully templatable forum and an extension system like 1.3's are not two hugely valuable aditions?

I see this discussion getting more 'political' and less technical and I don't like that. I would invite extensions developers and style designers to come here and give your opinion from a technical point of view.
Having extensions stored in the database across updates and server moves is a very cool thing that worths much more than the label "1 click install extensions".
I am starting to get the feeling that many here are not aware of how big the 1.3 release is. I guess that's the reason behind changing its name to 2.0.

Meow · 2008-12-27 06:17:59

Strofanto wrote:

Because 2.0 could take another year or more, 1.3 might be released in a few weeks or months as soon the markup is done and a few bugs are out. Or if not so, it might be still useable when the markup is committed.
Upgrading the forum takes an FTP transfer, a php script run, a few options to change in the admin panel. Can't take more than a hour.
A small improvement to what? If you're talking of 1.2, I wouldn't call that a small improvement.

1.3 has been developed so long, and it's no longer fresh. Why shouldn't 1.4 use what Paul has done for 1.3? And people could really develop better and better themes and functions for 2.0 in the future.

Most of guys are lazy. If 1.3 existed and 2.0 were also being developed, people wouldn't want to upgrade their forum to 1.3 or support the 2.0 plan. There shouldn't be three projects of FluxBB.

I know 1.3 changed a lot from 1.3 because I've touched 1.3 for over two years. But its "version number" looks like a small improvement, why not use 2.0 to tell people it's really a large improvement?

vnpenguin · 2008-12-27 07:16:44

yann wrote:

Please reconsider your position - you don't have enough people to maintain two branches. Maintain 1.2 with security fixes, release a rock solid, evolutive 2.0, and then add features, regularly, making sure the upgrades are easy.

Agreed 100% !

jmp · 2008-12-27 08:41:41

Pedro wrote:

Like any other software, it will become obsolete if it's not further developed.

I don't think you've been reading carefully what I've said. The way I see it now, there are at least three ideas about this floating around:

Have separate devs for the 1.2 (1.4) branch (as suggested at least by Smartys, Elbekko, me…)
Develop both branches with the same developers (yann, …?)
Stop developing 1.2 (only maintaining security fixes) and focus all efforts on the 1.3 (2.0) branch instead (Rich, ridgerunner)

You seem to be under the impression that I support C, and still want to use 1.2.

Are you guys saying that a fully templatable forum and an extension system like 1.3's are not two hugely valuable aditions?

No one said that. But fact is, many people don't need those features, at least not yet. Things like UTF-8 support are (in my opinion) much more widely needed. And like Smartys said, "The idea is to support 1.2 for the many people who can not or will not upgrade to 1.3 while also developing a more modern version of 1.2". In the long term, using an obsolete forum software would be a bad idea, but like I said, I would support further development of the 1.2 branch, while the current dev team would have less pressure with developing the 2.0 branch.

[…] that worths much more than the label "1 click install extensions". […]

You're making it look like I was diminishing it. I'm not. Some people just don't need it and/or the current methods (hacking the code manually) work well enough.

For people who want an extensible forum as opposed to minimalistic, the extension system will be a good addition. For people who don't need extensions, it's useless. That is my point.

Last edited by jmp (2008-12-27 08:45:42)

Kaboon · 2008-12-27 08:46:18

vnpenguin wrote:

yann wrote:
Please reconsider your position - you don't have enough people to maintain two branches. Maintain 1.2 with security fixes, release a rock solid, evolutive 2.0, and then add features, regularly, making sure the upgrades are easy.
Agreed 100% !

Disagreed 100%!

But seriously, I'm pretty sure this can be done if wanted. There are a few people who'd be willing to continue work on the 1.2 branch besides Paul, Elbekko and Conner who could concentrate on (a followup of) 1.3.

And I'm starting to like the idea of skipping version numbers. How about bumping 1.3 to 2.0 and 1.2 to 1.4? Just to draw a clear line between a 1.x branch and a 2.x branch.

vnpenguin · 2008-12-27 09:23:37

Jarkko wrote:

I don't think you've been reading carefully what I've said. The way I see it now, there are at least three ideas about this floating around:
Have separate devs for the 1.2 (1.4) branch (as suggested at least by Smartys, Elbekko, me…)
Develop both branches with the same developers (yann, …?)
Stop developing 1.2 (only maintaining security fixes) and focus all efforts on the 1.3 (2.0) branch instead (Rich, ridgerunner)
You seem to be under the impression that I support C, and still want to use 1.2.

Yes, I vote for C.

SuperMAG · 2008-12-27 09:43:48

the numbers of 1.2 must not be changed to 1.4, its too late, if you want to release further releases, it should be 1.2.xx

if 1.2 was released as 1.4 with only uf8 support, too many people will get confused.

its the same with 1.3, they were asked long time ago why is that not 2.0 and they didnt changed it then, why now, i mean almost every thing is done.

jmp · 2008-12-27 09:52:07

SuperMAG wrote:

if 1.2 was released as 1.4 with only uf8 support, too many people will get confused.

UTF-8 support would be (in my opinion) a feature big enough to justify the version bump.

[…] i mean almost every thing is done.

No.

Forums

#26 2008-12-24 19:56:56

Re: The future of FluxBB

#27 2008-12-25 04:40:28

Re: The future of FluxBB

#28 2008-12-25 06:41:21

Re: The future of FluxBB

#29 2008-12-25 09:03:10

Re: The future of FluxBB

#30 2008-12-25 12:59:17

Re: The future of FluxBB

#31 2008-12-25 14:30:30

Re: The future of FluxBB

#32 2008-12-25 17:59:25

Re: The future of FluxBB

#33 2008-12-26 09:09:30

Re: The future of FluxBB

#34 2008-12-26 09:51:28

Re: The future of FluxBB

#35 2008-12-26 09:58:23

Re: The future of FluxBB

#36 2008-12-26 11:42:46

Re: The future of FluxBB

#37 2008-12-26 13:52:32

Re: The future of FluxBB

#38 2008-12-26 14:21:43

Re: The future of FluxBB

#39 2008-12-26 15:52:33

Re: The future of FluxBB

#40 2008-12-26 16:06:32

Re: The future of FluxBB

#41 2008-12-26 20:07:05

Re: The future of FluxBB

#42 2008-12-27 00:05:52

Re: The future of FluxBB

#43 2008-12-27 02:05:30

Re: The future of FluxBB

#44 2008-12-27 06:17:59

Re: The future of FluxBB

#45 2008-12-27 07:16:44

Re: The future of FluxBB

#46 2008-12-27 08:41:41

Re: The future of FluxBB

#47 2008-12-27 08:46:18

Re: The future of FluxBB

#48 2008-12-27 09:23:37

Re: The future of FluxBB

#49 2008-12-27 09:43:48

Re: The future of FluxBB

#50 2008-12-27 09:52:07

Re: The future of FluxBB

Board footer