2019 user survey feedback

[albummaster]

First, a huge thanks to everyone who took part in the recent survey. It's always extremely interesting and insightful to hear your candid opinions about the site (warts and all). I'm a bit overwhelmed with the number of great ideas that people came up with. I'd like to also thank everybody that leaves feedback here on the forums, particularly the regulars that are active on the suggestions forum (and for your patience if i can't do things straight away - everything is logged, so it'll get done in a round robin fashion once I'm working on a related feature). I always do my best to work on the site as much as possible in my spare time (but this is not always possible as I still have a 'day job' which sponsors my living expenses and other commitments - such as a life away from here! ).

Over the last couple of years, I've spent literally thousands of hours behind the scenes refactoring the code that runs this site, first to allow the site to use a responsive design (so that it works better on mobile devices) and secondly so that the code itself is modernised so that it can be ported to a newer platform so that the site can become more app-like (- but not there yet ). The next stage in this process is to upgrade the forum software which will then allow the rest of the site to be upgraded. This process is using up 75% of my time in the background (so the remaining 25% is spent on keeping the site going and adding new features/fixing things). Sometimes bugs can leak onto the main site without any visible changes on the front end, and most of the issues over the past year (including the log in issue that was recently fixed) have been due to the sheer number of background changes that are happening.

It's gone way past the stage where I need additional design/development resource to help improve the site, but this simply isn't possible commercially as the site barely breaks even once hosting costs are taken into account. If more people disabled their ad blockers on BEA (which takes two seconds), it would help the site to invest in additional resource and new features would be developed a lot quicker. BEA's hosting provider expects to get paid, even if I don't

Second up, every single issue has been logged from the survey into a bug tracking database where all feedback is kept, so nothing has been lost or overlooked and it'll all be considered, prioritised and scheduled for development.

To the feedback itself, in terms of what needs improving, the survey identified several common issues. Although BEA now runs across different types of device, the look and feel of the site is still outdated, somewhat off-putting, and needs modernising. BEA could attract more people to the site if the layout was improved.

The site is badly in need of some form of genre classification to help manage and improve the navigability of the site. Many people point out the caveats of adding/managing genres in terms of edit wars, having too broad a selection and possibly sourcing the data from third parties e.g. Discogs, RYM, Musicbrainz, AllMusic etc as a solution to this. I've previously favoured a tagging approach where multiple tags could be applied to an artist/album/track and this has been on the cards for so many years. It's (always) simply a problem of time and resource as its a big change to implement that would then hold up development of all other features while it's being created. People also (generally speaking) want to see more stats and the addition of genres would really help to add a richer chart composition

The overall chart lacks diversity - certain styles/genres of music are under-represented in the top 100 (hip-hop, electronica, disco, funk, jazz, country etc), plus some artist/albums that perhaps shouldn't rank as high as they do - four Radiohead albums in the top 20, Arcade Fire in the top artists list etc. The chart is simply an aggregation, but it's a fact that the top 100 primarily comprises of white male bands with too many albums by the same artists (the only women listed are in bands primarily led by men). A big question (to me at least) is how to address this lack of diversity?

I've thought for a long time that the overall chart does not do justice to the breadth of the site. It's in everyone's interest to have a more diverse list as it's one of the first things people will look at when they visit BEA for the first time and it should be more varied to make it more accessible to a wider audience, and give it more credibility and authority.

The difference between BEA and other sites is that albums are ranked according to their position in people's lists rather than being ranked according to their rating (if you rank albums on BEA by their average rating - the result is quite a bit different to the overall chart and more resembles other sites). Certain albums are more 'popular' choices amongst BEA's user base and the popularity is determining the rank and causing the lack of diversity. (I'm not sure of the answer yet, but perhaps a weighting could be applied to influence the number of points that a particular chart contributes to the overall chart based on the make-up of the list e.g. number albums by the same band, mix of genres etc, but this might be over-complicating things and the best answer is probably a lot simpler.)

Behind the scenes, the site has been preparing to allow multiple artists to be associated with a single act so that multiple individual performers can be credited for the same release and allow people to drill further into other bands they might have belonged to. The first stage in this was the addition of collaboration/individual flagging of artists which will then allow their specific details to be added in the next phase.

At a higher level (and certainly since the site was founded), there has been a general shift away from the album format towards more playlist-centric music listening, so custom charts for singles/tracks on BEA are something that I'm warming to as I can see a role for Spotify integration here and perhaps people might then use BEA as a tool to accompany their listening on Spotify. The caveat to custom 'track charts' on BEA would be that tracks would need to be drawn from the pool of albums on BEA and not added as separate entities (but this might be too restrictive).

A clearer division between 'recognised' lists (from external publications) and user lists from BEA was also mentioned in the survey e.g. to have separate rankings for professional and user lists (but there's possibly not enough data to do effectively).

The chart editor needs improving to make it more use friendly (quite a few suggestions to improve this - and these are scheduled to be added over the next few weeks). This is a key part of the site and certainly needs a bit of attention.

A recommendations system was also mentioned by multiple people, This is again something that has been proposed a few years back but never saw the light of day. The ability to view recommended albums similar to the one you are currently viewing based on a curated list.

Increase the limit of chart sizes for senior or premium members (perhaps up to 200). Again, this is something that I'd be happy to add as some people are really struggling to fit everything into 100 ranks (particularly overall charts from members that have been here a while).

I've already added some quick wins as a result of survey (due for release over the next week or so).

• Add confirmation prompt when abstaining from polls
  
• CSV files. Added the ability to download other people's ratings and lists (if they are public), also added the ID as an extra column and separate title and artist columns.
  
• Spotify integration. Already added Spotify playlists for decade charts from 2010s to 1950s and year charts from 2019 as far back as 1960 (more changes will follow to improve the quality of the lists).

I'll probably make more edits to this post to make it more organised but these are some early thoughts... (Thanks for sticking with me if you're still here!)

[RoundTheBend]

Awesome! Thanks!

[dihansse]

Hi AM, thx for all the great additions you made these last weeks, including the most recent ones.

I also read your comment that 2020 is a pivotal year for you and of course I can understand that because maintaining a large site this must be a perpetual fight both timewise and financially.
But I think I speak for many people if I say that I would really hate this site to go because it has become an everlasting source of inspiration for album lovers like me and others. I can only repeat that many people would be happy to pay a subscription fee to keep this fantastic site up and running.

[albummaster]

[baystateoftheart]

[RoundTheBend]

Thanks for updating the playlist for each year - already subscribed to them all.

[RoundTheBend]

By the way the thumbnail feature is helpful on the chart editor. I've accidentally selected the wrong album before, so that helps me confirm.

[albummaster]

Thanks for feedback. Hopefully lots more to come...