{"id":897,"date":"2026-06-05T18:59:56","date_gmt":"2026-06-05T18:59:56","guid":{"rendered":"https:\/\/dougturnbull.org\/?p=897"},"modified":"2026-06-05T19:07:06","modified_gmt":"2026-06-05T19:07:06","slug":"localify-is-dead-long-live-local-music","status":"publish","type":"post","link":"https:\/\/dougturnbull.org\/index.php\/2026\/06\/05\/localify-is-dead-long-live-local-music\/","title":{"rendered":"Localify is Dead, Long Live Local Music"},"content":{"rendered":"\n<p class=\"wp-block-paragraph\">Sadly, after 8 years in the making, Localify has been decommissioned. But out of the ashes, we have begun developing a new yet-to-be -named locally-focused music recommendation app&#8230;<\/p>\n\n\n\n<h2 class=\"wp-block-heading\">Backstory<\/h2>\n\n\n\n<p class=\"wp-block-paragraph\">For the past 20 years or so, I have been trying to develop an app that can help people become more engaged with local music community. As a high school kid, I love going to live music shows at small venues in my hometown (Ithaca, NY). After high school, I always found it surprisingly hard to get &#8220;into&#8221; the local music scene when I moved to a new city. <br><br>My first big attempt was to build a personalized Internet radio player called MegsRadio.fm. Think &#8220;Pandora for Local Music&#8221;. For about seven years, I worked on this project with my undergraduate students from Ithaca College and Cornell. We had a great time trying to build a streaming music recommendation app, but the project ultimately failed because it became too difficult to maintain a professional-grade service in an academic setting. In particular, sourcing local music tracks, paying royalties, and developing a user experience to compete with the likes of Pandora and then Spotify were all overwhelming problems. (See my full write-up about the <a href=\"https:\/\/dougturnbull.org\/index.php\/2018\/01\/01\/megsradio-is-dead-long-live-megsradio\/\" data-type=\"post\" data-id=\"199\">Death of MegsRadio<\/a> from late 2017.)  <br><br>Around the same time we were considering abandoning MegsRadio, a student named Jon Burger was working on a side project with me called Localify.org. At the time, Spotify was quickly taking over music streaming, but there was no principled way to find music by location, and they did not recommend live music events. We designed Localify to fill this gap with a pure music recommendation app (i.e., no music streaming) that focused on both local artist recommendation and live event recommendation at small venues.  <\/p>\n\n\n\n<figure class=\"wp-block-embed is-type-video is-provider-youtube wp-block-embed-youtube wp-embed-aspect-16-9 wp-has-aspect-ratio\"><div class=\"wp-block-embed__wrapper\">\n<iframe loading=\"lazy\" title=\"Localify 2.0 Web App Demo\" width=\"840\" height=\"473\" src=\"https:\/\/www.youtube.com\/embed\/PYaEH9oS50s?feature=oembed\" frameborder=\"0\" allow=\"accelerometer; autoplay; clipboard-write; encrypted-media; gyroscope; picture-in-picture; web-share\" referrerpolicy=\"strict-origin-when-cross-origin\" allowfullscreen><\/iframe>\n<\/div><\/figure>\n\n\n\n<p class=\"wp-block-paragraph\">Over the last eight years, we have built various versions of Localify with different generations of undergraduate students. Our most recent version was somewhat successful, with over 3,000 people trying it out. However, the app was not particularly effective at attracting and retaining users. The times are also changing (<a href=\"https:\/\/dougturnbull.org\/index.php\/2025\/09\/10\/where-did-all-the-small-music-venues-go\/\" data-type=\"post\" data-id=\"819\">in some bad ways for local music<\/a>), which has led us to think about new ways to attact the local music recommendation problem.<\/p>\n\n\n\n<h2 class=\"wp-block-heading\">Three Observations<\/h2>\n\n\n\n<p class=\"wp-block-paragraph\">We are in the process of designing a new app based on some of the lessons we have learned from developing Localify. Here are a three observations and some related ideas we plan to focus on for the new app.<\/p>\n\n\n\n<h3 class=\"wp-block-heading\"><strong>1. The Spotify brand is toxic to some artists<\/strong><\/h3>\n\n\n\n<p class=\"wp-block-paragraph\">When we first came up with the name Localify, we intentionally tied ourselves to the excitement of this new celestial jukebox that had made its way across the pond from Sweden.  Spotify was the hip new thing, and young music lovers were excited early adopters. Aside from our clever name, we used the Spotify integrations for a variety of purposes. For example, users could log in to Localify with their Spotify account so we could automatically extract their (non-local) &#8220;heavy rotation&#8221; artists which is necessar for accurate local music  recommendation. We also automatically generated locally-focused playlists on Spotify for our users.  In addition, we had a requirement that an artist be streamable on Spotify before we would recommend them. The idea was that if an artist is on Spotify, they must be a legit original artist rather than some random dad rock cover band. <\/p>\n\n\n\n<p class=\"wp-block-paragraph\">Over the years, Spotify has gone from upstart pirate to corporate platform. (See Liz Pelly&#8217;s <a href=\"https:\/\/harpers.org\/archive\/2025\/01\/the-ghosts-in-the-machine-liz-pelly-spotify-musicians\/\">Ghost in the Machine<\/a> article and <a href=\"https:\/\/www.simonandschuster.com\/books\/Mood-Machine\/Liz-Pelly\/9781668083505?gad_source=1&amp;gad_campaignid=23865952718&amp;gbraid=0AAAAADv-uEmbtNy9zwIo5x6U9YHmo2B92&amp;gclid=CjwKCAjwxITRBhBYEiwA6mZm7c6JRTdhOlWM4rGIjETQBxNlE8gWT7x1VrgPaXn709SS3e5bCfsUTBoCdDEQAvD_BwE\">Mood Machine <\/a>book.) Some smaller artists do not trust Spotify and, by extension, often express concerns when we talk to them about what we have been trying to accomplish with Localify. With this in mind, we decided to distance ourselves from the Spotify brand so that that all artists and music fans will be more likely to trust us from the get go. <br><br><strong>Idea 1a: Rebrand <\/strong>&#8211; We decided we needed a new name to distinguish our not-for-profit academic project from Spotify&#8217;s corporate brand. <br><br><span style=\"box-sizing: border-box; margin: 0px; padding: 0px;\"><strong>Idea 1b: Decoupling from Spotify<\/strong>&nbsp;&#8211; Because many artists are explicitly opting out of having their music on Spotify, we feel that it is important for us to be able to recommend local artists who are not on Spotify, but instead make their music available through other apps like BandCamp, SoundCloud, and Deeze<\/span>r. We still don&#8217;t want to recommend &#8220;dad rock cover bands&#8221; so an artist&#8217;s music must be available on some platform; it is just not required to be on Spotify.<\/p>\n\n\n\n<h3 class=\"wp-block-heading\"><strong>2. LLMs offer new ways to find &#8220;more granular&#8221; events and venues<\/strong> <\/h3>\n\n\n\n<p class=\"wp-block-paragraph\">Some of our more hipster users noted that there were many &#8220;smaller&#8221; events and obscure venues (house parties, DIY spaces) that we could be surfacing in our recommendations. However, these events were not listed on the major event aggregation sites such as BandsInTown, Pollstar, or Spotify. So we have been exploring new way to capture live music information at a more &#8220;grandualar&#8221; level.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\"><strong>Idea 2a: LLMs can extract events directly from venue websites <\/strong>&#8211; previously it would have been close to impossible to write a custom scraper for each of thousands of music venues.  However, once we have a venues event page URL, we can ask an LLM to pull out all of the event information and then we can write code to add these events to our database.<br><br><strong>Idea 2b:<\/strong> <strong>LLMs can extract events from  social media images &amp; videos <\/strong>&#8211; we have found that many smaller events are announced on Instagram or TikTok feeds as images or short videos. As image analsys improves with each new release of the latest frontier model, we can explore using this technology to extract event data from these visuals data sources. <br><\/p>\n\n\n\n<h3 class=\"wp-block-heading\">3. Music is inherently social <\/h3>\n\n\n\n<p class=\"wp-block-paragraph\">I&#8217;ve never wanted to design MegsRadio or Localify to have too many social features since I&#8217;m not a big fan of social networks like Facebook and Instagram. There have also been social networks which have explictily focused on music in the past (e.g., MySpace, iLike, the original Last.fm) but they never really inspired me too much either. I had thought that if listeners were simply exposed to great music by talented local artists, they would fall in love with the music and support the local scene by going to lots of shows. This may have been a little shortsighted on my end. <br><br>Music has always been inherently social. We go to shows with friends, lose track of time debating the intracies of vocal melodies and guitar solos, find our people in fan communities, wear around concert tees like badges of honor, and fill our social media feeds with the artists we love. With all this in mind, I think it is time to add a few thoughtful social features to our next app. <br><br><strong>Idea 3a: Friends take friends to shows<\/strong> &#8211; the features I have always wanted was to know which of my friends are going to a show. If I know Sarah and Steve are going to as show, I&#8217;m much more like to go myself.  I can imagine bonus features like &#8220;I&#8217;d like to go but need a ride&#8221; or &#8220;ping me if you are looking for tickets&#8221;.  So we decided to implement a &#8220;friends&#8221; network to help with these kinds of intentional social features. For any social feature we decide to include, we will always make sure there are easy-to-use opt-in policies to protect user privacy.<br><br><strong>Idea 3b: Crowdsourcing<\/strong> has potential down the road &#8211; Crowdsourcing doesn&#8217;t really work unless you have a crowd. We have never gotten to the point where we have a dedicated user base so we have never really invested in developing crowdsourcing into our user interface. However, we have redesigned our database schema to better support users as trustworthy data sources. We can use crowdsourcing to collect artist origins, venue reviews, upcoming DIY events, etc. <\/p>\n\n\n\n<h2 class=\"wp-block-heading\">Next Steps&#8230;<\/h2>\n\n\n\n<p class=\"wp-block-paragraph\">I&#8217;ve been working solo on a brand-new database schema design, a new backend API, data collection code, and a dashboard as part of my recently concluded sabbatical. I now have a team of four students working on different aspects of the new app. Sam K. and Fisher G. are taking point on an early-stage user interview study that will inform the mobile-first UI\/UX design. Fisher will then take point on the iOS app while Sam, with help from Robby R., will develop a native Android app. Abe M. and I will focus on the backend, which includes everything from setting up servers to collecting music data to developing a new recommendation algorithm. <\/p>\n\n\n\n<p class=\"wp-block-paragraph\">The hope is that by the end of the summer, the five of us can release a new app incorporating the ideas I&#8217;ve discussed in this post, plus additional ideas that emerge from the design process. <br><br>In the end, the goal remains the same: get more people to support local music. <br><br>Doug T.<br>June 5, 2026<\/p>\n\n\n\n<p class=\"wp-block-paragraph\"><br><br><\/p>\n","protected":false},"excerpt":{"rendered":"<p>Sadly, after 8 years in the making, Localify has been decommissioned. But out of the ashes, we have begun developing[&#8230;]<\/p>\n","protected":false},"author":1,"featured_media":906,"comment_status":"closed","ping_status":"closed","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[1],"tags":[],"class_list":["post-897","post","type-post","status-publish","format-standard","has-post-thumbnail","hentry","category-uncategorized"],"_links":{"self":[{"href":"https:\/\/dougturnbull.org\/index.php\/wp-json\/wp\/v2\/posts\/897","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/dougturnbull.org\/index.php\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/dougturnbull.org\/index.php\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/dougturnbull.org\/index.php\/wp-json\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/dougturnbull.org\/index.php\/wp-json\/wp\/v2\/comments?post=897"}],"version-history":[{"count":11,"href":"https:\/\/dougturnbull.org\/index.php\/wp-json\/wp\/v2\/posts\/897\/revisions"}],"predecessor-version":[{"id":909,"href":"https:\/\/dougturnbull.org\/index.php\/wp-json\/wp\/v2\/posts\/897\/revisions\/909"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/dougturnbull.org\/index.php\/wp-json\/wp\/v2\/media\/906"}],"wp:attachment":[{"href":"https:\/\/dougturnbull.org\/index.php\/wp-json\/wp\/v2\/media?parent=897"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/dougturnbull.org\/index.php\/wp-json\/wp\/v2\/categories?post=897"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/dougturnbull.org\/index.php\/wp-json\/wp\/v2\/tags?post=897"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}