YaBB rewrites from the ground up have been attempted 2-3 times with fresh code.  A couple times we got very close, but the team gave up on it for a few reasons:  no speed improvements with the way database implementation was handled (which was in effort to support any type), lack of desire to rewrite every single feature and security measure, disagreement on coding style using OOP, and inexperience by some team members to be able to code in strict more challenging but better coding styles.  In the end, it was decided last year to continue again with the existing original code base (which has been "rewritten" over the years but not from scratch) so it could be done quicker.  With this decision, our development team implemented many desired features like sub boards and a calendar as well as the start to MySQL.  We did not want to release the other major changes as another version 2 but rather reserve them for the big fanfare of MySQL-supporting YaBB 3.  Unfortunately we ran into lack of time on the developers' part to continue working and a stumbling block with the MySQL implementation being poor and not much faster while trying to preserve flat file as a secondary option.  We have just made the decision recently to go ahead and ditch flat file for MySQL only so we can get the show on the road.  Now we are waiting for the developers to find time again to work.

There really isn't anything to prove.  We know a rewrite from scratch or using the existing code can be done.  Why has it taken so long?  Because of the developers having real lives that conflict with time for YaBB and because of people like you saying they can do better on their own.  Don't forget that the team is comprised of random strangers around the world that have never met and are just volunteering time.  It really irks me every time someone comes along and says they can do better than "the team" instead of proving it and bettering YaBB by offering to use that talent and time as part of the team.

Despair is beginning to creep in again, with no real input either here or on the dev site for over a month.  In fact, other than a few words spoken by project leaders and devs on this site a few weeks ago, no progress since February.

I know and fully appreciate we are reliant on busy devs trying to find some spare time, for which they should be applauded, but I can't help feel the devs have lost will to find time since we lost momentum in February.  And absolutely no disrespect intended in my comments, and I have the highest respect for the devs who freely give up their own time.

Just the knowledge that something is happening would be enough of a carrot to keep me patient and loyal 



Selfishly, my current problem is my site is beginning to struggle over the last few months.  I've just spent a bit and thrown new hardware at it, but its only a short term fix, and is putting pressure on me to do something about it.

From testing, AJAX in the current flatfile Next Edition does help a fair bit, which would give extra breathing space until an optimised SQL version is ready.



As I'm backed up in a corner, I need to ask direct questions to Corey. Is there a timetable? If not, have discussions on timetables taken place? Do you honestly foresee Y3 being released, and do you foresee an optimised MySQL version soon?



Thankyou for taking the time to read this long post, and thanks in advance if anyone answers my questions

A very anxious TheBoy

chillipepper wrote on Jul 13^th, 2011 at 6:24pm:

This version always was Y3 (using Y2.5 as the base), and was 'almost' ready back in Jan/Feb time.

However, when I was invited to beta test it, the MySQL side of things was below expectations (I have a fairly large forum).

For me, its virtually good enough to use as-is on Flatfile (accepting the same limitation around searching as 2.5), but I won't as its obvious that extra work needs to be done under the hood to optimise the MySQL side, which might mean data format chances. Thus running a pre-release version with a unique file format is risky. To risky for me.


There was some very strong momentum back then, it was really looking promising, then we hit the issues that I highlighted in my original post. Now it seems (and I'm guessing, as I don't know) there is a lack of direction, a lack of will and a severe lack of developers to continue.


Personally, just to inject something into the staleness, I'd like to see the current Y3 released as Y3 very soon, with a set of fixes shortly after (quite a lot has changed, so will be problems I suspect), followed by a Y3.5 in a few months once MySQL has been optimised.

I know Y3 isn't quite ready for non ASCII sites though.


I've been trying other software, including the converters, nothing quite fits the bill like YaBB. Partly due to the brilliant way YaBB allows splitting of dynamic and static content between different servers (won't affect most YaBB users), partly because no converter maintains the internal links (which will bugger up my Google PR, currently at 4), and partly because all our processes are built around YaBB. Unsurprisingly, SMF is the closest to my needs. But nothing quite matches YaBB.

Like many, I'm wondering if my hand will be forced to migrate though 

Quote:

CJ, I found your post uplifting and depressing.

I am so pleased to hear you feel you (the team) have a goal, and think you know how to get there , but you have confirmed my feelings that nobody is interested any more - you are one of the few left, by your own admission, and are struggling to get new people on board? 



You have no idea how much I want to stay with YaBB.  Both I (from an Admin perspective) and our userbase (from a user persective) generally feel that YaBB is one of the best forums out there.  The feedback I get is they like the simplistic layout, the accessibility of the features, the overall usability, the reliability of the forum as a whole and individual features, the stability, and the speed.

The only real, constant complaint we get is Search. By its nature, its truely awful.  My site, like many forum sites still going, is an information and help site, thus search is critical to it.  Additionally, we are starting to see index corruption more frequently.


Unless YaBB goes fully MySQL optimised, my own issues above will not be resolved. But releasing nothing, MySQL optimised or not, just makes the project feel 'dead'.

Rambling aloud, from your experience, would a release of Y3 in its current guise, attract developers again? Or do you think YaBB has past the point of no return...

Quote:

I like the idea of offering both options.  But as an amateur programmer (sadly not Perl!), I know, given the current Y3 architecture, its nigh on impossible to do properly, without maintaining 2 versions.

Having witnessed the performance of the current SQL implementation - miles worse than the current flat file system - it serves no purpose.  The flatfile is always going to be hard to beat performance wise (except on Sourceforge ) for most operations, but searching through several hundred thousand files is always going to end in failure. Actually, just letting the OS's APIs manage that many files in signal directories isn't going to help.

So flatfile is very convenient - easy to set up, doesn't need database servers, fast. But it doesn't scale.

As per the discussion in Feb on the dev site, there are choices (and it seemed to be deciding which one which stalled the project).  The message back from the devs, Micheal Prager and Jet-Li (both of whom I have the greatest respect for), was it had to go with one back end or the other.

Given that flatfile won't scale up, surely that makes a RDBMS based back end a better option, if it does have to be one method or the other...?

Off-Topic replies have been moved to this Topic.

Skins (templates) are plentiful and MODs to add ads and other features are available at our sister site BoardMOD.org