Finish up and release

General discussion of development ideas and the approaches taken in the 3.x branch of phpBB. The next feature release of phpBB 3 will be 3.3/Proteus.
Forum rules
Please do not post support questions regarding installing, updating, or upgrading phpBB 3.2.x. If you need support for phpBB 3.2.x please visit the 3.2.x Support Forum on phpbb.com.

If you have questions regarding writing extensions please post in Extension Writers Discussion to receive proper guidance from our staff and community.
User avatar
Erik Frèrejean
Registered User
Posts: 207
Joined: Thu Oct 25, 2007 2:25 pm
Location: surfnet
Contact:

Finish up and release

Post by Erik Frèrejean » Mon Nov 14, 2011 12:03 pm

After an internal team discussion started off after the "phpBB Falling Behind?" topic on .com, Oleg asked me to post this here for further discussion.

At this point the development of phpBB 3.1 is pretty much at a stand still, and the new development strategy as introduced back in February 2010 is pretty much out of the window at this point of time. With the last feature being close to two years ago a feature release should be made sooner than later. Having this much of development time scares away community members (hay bb solution 'x' has a feature release every 'y' time and phpBB doesn't evolve. I'll be using 'x'), and makes it for contributors less interesting to contribute. As I wrote in our internal discussion
Erik Frèrejean wrote:One cause for the lack of external contributors is IMO the lack of releases. I personally lost interest in contributing code for 3.1 because I feel that I can spend my time better than writing code for a release that might be released somewhere in the future. I'm aware that this doesn't help the cause but I can't find the motivation to do so, and I'm not surprised if other team/community members have the same feeling.
having more regular feature releases with a relative small change set (the "new" strategy was based upon this concept), allows users to get new features more regularly and motivates contributors and developers to spend time working on new features because they know their efforts will be available to the community at a wide within the foreseeable future, sparking their interest to contribute of which the community at large greatly benefits.

With this in mind I want to propose the following changes to make sure that phpBB 3.1 gets finished sooner than later.
  • Feature freeze! Officially there have been a feature freeze since july 2010, but according to various development team members this is more like an "RFC freeze". However 13 RFC's have been posted and some have been accepted as well after this freeze (nearly 25% of all 3.1 RFC's have been posted after the cut off date). To enforce this, all non-merged RFC topics will be moved to the 3.2 RFC forum to ensure that they reflect the correct feature release for which they are projected. Because some of those RFC's are actively being developed and this change comes pretty suddenly, there will be a transitional phase. Any 3.2 feature that is finished before the first beta should be moved forward and be merged into Ascraeus. Any feature missing this deadline will be postponed to Arsia.
  • Holy hooks, making the phpBB core more hook able for easier MOD installation has been a priority for a long time. Unfortunately hooks are complicated and the hooks RFC basically is the one stalling new feature releases. The hooks RFC will be pushed back to Arsia, allowing a bigger time frame to correctly implement them without them blocking the whole development process. 3.1 will not be cut off of being more hookable than 3.0. The current core already has a hook system (yes it has limitations but it is quite handy if you work around some of the limitations), I've started to implement the requested hooks with the current system. Ticket
  • Fully implement the new development cycle. At the day 3.1 is released the feature freeze for 3.2 should be announced, this probably should be something around 10 months. This feature freeze should be set in stone, features not finished on this date will be pushed back to 3.3. After the freeze there will be some time needed to finish up things an to go through the release cycle. All together this will result in a feature release being made roughly every 12/14 months, which is a very acceptable cycle. Doing so will ensure that changesets growing to big and development taking up more and more time and nobody seeing the result of the hard work. I think that this will increase developer/contributor activity (and motivate one. If I want feature 'x' to ship with the next release I'll be a lot more motivated to get things done when I know that I might miss the date), as well keeping users because as a user you know that new features will come at a somewhat regular pace. This hopefully will also increase the amount of feature requests/RFC because posting one will more likely result in the feature being included.
All together this will quite quickly cut off the current development cycle and may look like rushing a release out for the sake of having a release. I personally however feel that this is the only way to break through the status quo that currently keeps the progress of the project hostage.
Last edited by Erik Frèrejean on Mon Nov 14, 2011 10:13 pm, edited 3 times in total.
Reason: Added link to the new hooks ticket
Available on .com
Support Toolkit developer

victory1
Registered User
Posts: 5
Joined: Tue May 17, 2011 2:29 am

Re: Finish up and release

Post by victory1 » Mon Nov 14, 2011 9:49 pm

This is a brilliant plan. I think we are moving in the right direction. Last year, I did start a poll at a major admin site asking for advice on the 3 softwares that I was considering for my board (phpBB3, myBB, and SMF) and I have to say myBB won by a mile. So why did I choose phpBB3 after asking for advice and it was not the #1 choice by the poll participants? Because after downloading all 3 software and playing with them for about a week, I thought phpBB3 interface suited me way better. The major complaints were the installation of modifications. Also how phpBB3 does not have a few features that should have been built-in that forces the non-technical admin into installing complicated modifications if they want some of these features. So just addressing this a little is a great accomplishment.

User avatar
Lurttinen
Registered User
Posts: 78
Joined: Mon Feb 20, 2006 7:57 am
Contact:

Re: Finish up and release

Post by Lurttinen » Mon Nov 14, 2011 10:29 pm

Too much planning? One could just polish phpBB 3.0 and call it 3.1
part of the job is already done and is waiting in the MOD database. (as some of the phpBB3 features were back in the phpBB2 days)
Just get around the usual "Not everyone wants feature X " cry. Might even ignore it and look other applications which use things like highslide, password strength checker, etc... And grabs it for phpBB3.1
After all it is just 3.1, not a complete rewrite as phpBB3 was for phpBB2.

wGEric
Registered User
Posts: 521
Joined: Wed Jun 11, 2003 2:07 am
Contact:

Re: Finish up and release

Post by wGEric » Mon Nov 14, 2011 10:39 pm

This is great :)
Eric

User avatar
RMcGirr83
Registered User
Posts: 357
Joined: Fri Mar 09, 2007 1:51 am
Contact:

Re: Finish up and release

Post by RMcGirr83 » Mon Nov 14, 2011 10:59 pm

The hooks RFC will be pushed back to Arsia, allowing a bigger time frame to correctly implement them without them blocking the whole development process.
So don't plan on having hooks for the next few years when many other BB software already provides hooks (properly btw)? Granted I have no idea how long it will be till Arisa is released but I can't imagine it would occur prior to 2014ish.
Do not hire Christian Bullock he won't finish the job and will keep your money

User avatar
Pony99CA
Registered User
Posts: 986
Joined: Sun Feb 08, 2009 2:35 am
Location: Hollister, CA
Contact:

Re: Finish up and release

Post by Pony99CA » Mon Nov 14, 2011 11:04 pm

Erik Frèrejean wrote:All together this will result in a feature release being made roughly every 12/14 months, which is a very acceptable cycle.
I think 12-14 months for a feature release is still too long; 12 months should be the outside target (it might slip, of course). IMHO, feature releases should be more like 6-8 months in general; 12 months should be for very complex releases (perhaps like hooks or events).

Steve
Silicon Valley Pocket PC (http://www.svpocketpc.com)
Creator of manage_bots and spoof_user (ask me)
Need hosting for a small forum with full cPanel & MySQL access? Contact me or PM me.

User avatar
Erik Frèrejean
Registered User
Posts: 207
Joined: Thu Oct 25, 2007 2:25 pm
Location: surfnet
Contact:

Re: Finish up and release

Post by Erik Frèrejean » Mon Nov 14, 2011 11:16 pm

Lurttinen wrote:Too much planning? One could just polish phpBB 3.0 and call it 3.1
part of the job is already done and is waiting in the MOD database. (as some of the phpBB3 features were back in the phpBB2 days)
Just get around the usual "Not everyone wants feature X " cry. Might even ignore it and look other applications which use things like highslide, password strength checker, etc... And grabs it for phpBB3.1
After all it is just 3.1, not a complete rewrite as phpBB3 was for phpBB2.
I don't think it is "too much" planning, you need to have a clear plan on where development is going. Otherwise you'll end up with a phpBB 2.2 type of situation where things just went out of control. Sure the MODDB has a lot of features, though just throwing them in the core will only cause problems (not to mention the rubbish quality of some of them). However I agree that they can be used to quickly add a feature, but someone has to sit down, clean up the code and properly integrate it with the core.
The RFC system was designed to assure a certain quality of code/features that get implemented in the core, and I see nothing wrong there.
Available on .com
Support Toolkit developer

User avatar
Erik Frèrejean
Registered User
Posts: 207
Joined: Thu Oct 25, 2007 2:25 pm
Location: surfnet
Contact:

Re: Finish up and release

Post by Erik Frèrejean » Mon Nov 14, 2011 11:23 pm

RMcGirr83 wrote:
The hooks RFC will be pushed back to Arsia, allowing a bigger time frame to correctly implement them without them blocking the whole development process.
So don't plan on having hooks for the next few years when many other BB software already provides hooks (properly btw)? Granted I have no idea how long it will be till Arisa is released but I can't imagine it would occur prior to 2014ish.
There *are* hooks, for example this board is a board I maintain. All (but template) modifications there are handled through hooks, the only core edits made is actually the addition of hooks. This is also why I've started to add the requested hooks with the current system (as far as they are possible). The main thing right now how ever is that the new hooks won't be implemented any time soon and certainly not when done properly. So why denying the community new features just for the sake of hooks? Keep in mind that when continuing the current road nobody has a clue when the new hooks will be finished.
This proposal is purely to get the release cycle going, allowing new features to feed into the core product, and with that actually start using a development strategy which will only benefit the community at large.
Available on .com
Support Toolkit developer

User avatar
Lurttinen
Registered User
Posts: 78
Joined: Mon Feb 20, 2006 7:57 am
Contact:

Re: Finish up and release

Post by Lurttinen » Mon Nov 14, 2011 11:31 pm

I think most of the 2.2 problems were in the Teams and changes with key personnel which left the project unable to make decisions. :)

User avatar
EXreaction
Registered User
Posts: 1555
Joined: Sat Sep 10, 2005 2:15 am

Re: Finish up and release

Post by EXreaction » Mon Nov 14, 2011 11:51 pm

I think that selecting wanted RFCs and then doing an RFC freeze and releasing the next version when those are finished is the wrong way to do it, that's the problem we have now where we have a couple of small things that are delaying 3.1 when we have tons of smaller things that could be implemented.

I would argue for a time release instead of strictly for wanted features. We can have a list of features wanted for a certain release, but if they do not make it by the time needed to go into the release, it should be pushed to the next release. This also allows people to create smaller features and have them included in sooner versions instead of being pushed off multiple releases when it's already done. This also gives people interested in helping a timeframe that things need to be done by if they want them included and lets the community know that we're committed to continue improving the software on a regular, scheduled basis.

Post Reply