2014 roadmap

[albummaster]

It's that time of year again to take stock of where BEA is going and where it's been. The site is now entering its ninth year! At the start of 2013, there were about 30,000 albums on the site, and there are now about 45,000 (a 50% increase in the size of the site during 2013). The number of active members on BEA also increased by 33% year on year, so things are going in the right direction. The servers were upgraded yet again in 2013 to cope with the extra load and we now have a platform which should scale into 2014 (but it's also becoming more and more expensive to host the site!)

The number of outstanding (uncompleted) features on my 'to do' list is about 230 (with a few still to log) (2013: 200 items), so the 'to do' list is increasing despite my best efforts to keep on top if it. The changes are mostly harvested from the forums. Numerous items have been ticked off and added during 2013 but I'm struggling to keep up.

2013
In terms of the features that I expected to deliver from the 2013 roadmap, these didn't materialise in the way that I expected due to the extra work that I was having to do to keep a bigger site running (lots of fire fighting behind the scenes, including recovering from a server meltdown a few months ago).

Despite this, in 2013 aggregated year and decade charts were increased in size to 1,000 entries, the overall chart was increased to 10,000 entries (& CSV exports increased correspondingly) & the top artists chart increased to 1,000 entries. The ability was also added to sort the charts on BEA by a variety of metrics, & a new site logo was added (thanks to Puncture Repair). A lot of smaller changes were also added, but most of the changes to BEA during 2013 were in the background to address performance and scalability issues and provide a platform for growth in 2014.

Time issues
As the site has grown, the amount of time it takes to look after it has increased (not surprisingly) which has squeezed the amount of time left to develop new features. The site now requires more individual communication via PM and so forth e.g. making sure users get issues resolved and trying to keep some order on the site. Also, as the site has grown, it now takes longer and longer to test each release (due to the increase in the amount of regression testing). BEA also has to ensure that there are no major bugs on whichever browsing device that people are using (& more platforms are being used than ever before). In fact, development time is now being squeezed so much by so many things that there is hardly any time left to build new features!

In fact, 2013 saw less frequent releases of new functionality than in previous years, even though my time on the site probably increased. My presence is probably not noticed by many people here, except maybe for moderators or those people involved in data moderation. Most of my time on the site is spent on the background aspects of running BEA (keeping the servers running, fixing issues, responding to queries etc).

I should have seen this coming a few years ago, but I'm realising now that to keep developing the site mostly by myself is a bit of a losing battle. No matter how much work I put in, there is always something else that I need to do & it gives me no rest (the site runs 365 days a year).

The list of changes that require implementing on my 'to do' list is somewhere around 230 and a few of the larger features *by themselves* might take several months to implement! (e.g. tagging, recommendations etc), so the big changes tend not to get done even though some of them would make a huge difference to the site. I worked it out earlier, and the list probably contains in the region of 5 man years worth of work (& growing), and I maybe only have 10 days a month spare for BEA (depending on other commitments).

The site isn't going to grow much further unless I can get more help on the design and development side of things. I simply can't deliver everything that is required in the free time that I have available and things are being 'rushed' & fudged just to get them off my list. Puncture Repair has helped this year by designing a new logo and I think the whole 'experiment' of sharing the work has proven quite successful.

Ideally, I'd like one or two experienced developers to help me out. One person would be somebody with strong UX experience (html, js & css) to help improve the layout of BEA and address some of the usability & design issues.

The other person would help back-end to re-factor the existing code into a better platform. I need somebody with strong knowledge of PHP and MySQL, object oriented design, knowledge of templating, caching & scaling big sites who can help optimise performance and improve scalability. Locally, in Spain, I don't know anybody so maybe someone on BEA has some spare time and would like to contribute. Some people have already approached me independently of this message, so I'm just opening this up to the wider community if anybody else would like to be involved.

At some point in the future, I plan for the front-end of BEA to be completely rewritten (in html5) and be sympathetic to mobile devices. This is a big undertaking and one I haven't got time to do on my own. There are lots of other things I'd like to do with the site as well, but haven't got time to do on my own.

The bigger features such as tagging are unlikely to be implemented unless i can get some help delivering them. I'd also like BEA to move to a more service oriented architecture to allow the future adding of a public api, so basically the whole BEA 'stack' needs re-engineering as it has grown organically and the scope of the site has changed considerably since it first started and is heading in a direction that I never envisaged.

Thanks
I'd like to thank here everyone that has helped out this year with the site. So, a big thanks (particularly) to the moderators (past and present): an_outlaw, joyofdivision, Mancsoulsister, Necharsian, Romanelli, ShaneSpear, 19loveless91 & Jhereko who have all been involved and been a great help. I'd also like to thank the people that are helping in the background to moderate the data, which is a vital task that often goes unnoticed. So, thank you to everybody that contributes, but particularly (again) to Romanelli, Mirek, harkan & The Poe as they are the biggest contributors, because without these people, the site couldn't function. On the forums, I'd like to thank Jhereko, drakonium, harkan & swedenman for running some great tournaments this year (apologies to any i have missed), and thank you to everybody else on the site for their continued contributions. The community here are most of the reason that the site is still running, and some of the people here have been with us a number of years now which makes this place very special to me.

2014 roadmap:

2014 most important features:

• Premium membership - Carried over since 2013, pay a small fee, bypass the need to earn member points, and gain access to some exclusive premium member features.
  
• Charts by tag - Carried over since 2011, a tagging system will allow people to generate lists filtered by 'tag' e.g. filter the overall chart for the best soul/r&b/metal albums etc, or generate a chart of the best 1970s soundtrack albums. Supporting this feature, people will be able to assign tags to anything across the site, and the tag can then be used as a filter for any of the charts.
  
• BEA Mobile - Carried over since 2013, make BEA more friendly on mobile devices and tablets (probably means html5 rewrite of the entire site and changing the design).

2014 other features:

The following seem the most likely other features to be implemented in 2014 (out of the 230 on the current list & excluding bugs which will always have priority):

• Ability to make/receive recommendations
  
• Add an option to show more items on a single page, e.g. 10, 25, 50, 100
  
• Recent 'activity stream' per member or object (or whole site)
  
• Improve handling of non-Latin alphabets
  
• Allow users to comment/rate the aggregated charts
  
• Chart update notifications - receive an email when a chart that you subscribe to is updated
  
• Add a stats page showing risers/fallers in the overall chart in last month/week
  
• Improve the average rating algorithm (complaints that it favours items with less ratings)
  
• Add a BEA 'most wanted' list - (items appearing most frequently in people's wish lists)
  
• Improve the chart editor - better suggestions based on albums that appear in your other charts, copying of rank comments between charts and ability to add an album into a chart from outside of the chart editor e.g, add an album from another chart page

Again, right now, my day job is still taking up a lot of time, which is limiting the amount of time I can spend on BEA, so the above is just a general guide.

Thanks to everybody for your continued support and apologies for a very long-winded post.

albummaster

* The time-scales have the same caveats as ever: BEA is not a 'job'. It is a hobby. I'd like to spend more time on BEA, but I also need to make a living. Being able to deliver 'on time' depends on other commitments outside of BEA.

[Guest]

Super cool. You're the best, AM. I look forward to every update. I hope you find the help you need.

[Patman360]

[Norman Bates]

Yeah, same here. Unfortunately I can't help with coding, but let's hope someone can give a hand.

[GeevyDallas]

[antiloopje]

Well, I can only admire the fact that this great site was set up by one person. That's an awesome achievement. Let's hope you find some people with programming skills to help you.

[Precedent]

[Defago]

Thanks a lot, AM! I can't code beyond VBA, but if you need any other sort of misc. help let me know.

[junodog4]

This fantastic site is a tribute to all your hard work. Bravo.
_________________
Finnegan was super bad-ass.