Development diary
A record of how MobileFlare came to exist, how it was built, and where it is going. Updated as the project develops.
The biggest single update since launch, published simultaneously to iOS and Android. Button order changed to Red, Yellow, Green, Blue, reflecting a more intuitive urgency sequence. Crowd Crushing renamed to Crowd Pressure, which more accurately describes what the signal is for. Colour blindness support added, with each signal flashing at a distinct speed as well as a distinct colour. Green is slowest, Yellow faster, Red fastest. Water Please remains a calm static screen throughout.
This version also brought a full code review, optimisation across the board, and consistent typography throughout the app.
A meaningful update to the iOS build. Water Please was redesigned as a calm, non-flashing static screen, reflecting that it is a welfare request rather than an emergency. A brief responsible use guide was added to the first launch of the app, appearing once and never again. A feature flags system was also introduced internally to allow features to be toggled without requiring a full release.
A small but important fix to the Android build. A broken URL was identified and corrected. Nothing visible to most users, but the kind of thing that matters when someone is trying to share the app at an event.
The share feature arrived on Android, two days after iOS. Both platforms now fully in sync on v1.0.1. The QR modal displays codes for both the App Store and Google Play, making it easy to share with whoever is standing next to you.
The first post-launch update. A share button was added to the app, opening a modal with QR codes for both the App Store and Google Play. Getting MobileFlare into more hands at live events was always the priority, and this made it significantly easier to do that in the moment.
MobileFlare went live on Google Play. After the iOS release in March, the Android build was completed, submitted, and approved. MobileFlare is now available to download on both iOS and Android, free of charge, with no registration required. The demand from the TikTok comments had been noted, and this was the answer.
The first TikTok post reached around 80,000 views within the first week. For a first ever post with no following, no budget, and no promotion, that was unexpected. The feedback was mostly positive, with a lot of people saying they wished they had known about this at events they had been to.
The comments were also full of requests for Android and EU availability. Both noted, and both now addressed.
MobileFlare went live on the Apple App Store. After four rejections, two escalations, and more waiting than felt reasonable, v1.0.0 was approved and publicly available. The four colour-coded signals, Red for Personal Safety, Yellow for Crowd Pressure, Green for Medical Emergency, and Blue for Water Please, were now in the hands of anyone who wanted them.
Free. Offline. No registration. That was the intention from day one.
After what seemed like an eternity, the app was approved with no further changes. The waiting was over. Time to share it with the world.
It was almost as if the reviewer was viewing a cached version of the website, or simply wasn't looking. Guideline 3.1.1 was still flagged as the issue. I raised an escalation to have it looked at properly.
Despite my previous explanations, I removed the Donate button from the app, and the Donate page from the website. I still don't understand the interpretation of the guidelines, but I just needed to get the app out there.
Whilst the app was in review, sombr's concert at Brixton was in the news. You can search for this to see what happened. The app could have helped in this situation by drawing the attention of the stewards instead of sombr having to stop the performance. Frustrating, and it just highlighted the need for such an app.
I had hoped to accept tips and donations through the website to support development and cover costs of the app. It seems the Apple reviewer did not like this, as they interpreted it as unlocking features by means other than App Store payments. In my response, I explained that these were not payments to unlock any additional features, and they must have misunderstood. It was also via a website rather than the app, so surely couldn't be linked. I re-submitted.
Though I had designed the app to work only on mobile devices, and assigned it to just iPhone hardware, the app was reviewed by Apple using an iPad mini, and the display didn't quite accommodate the design. Two options: redesign for iPads, or figure out how to exclude installation on iPads. I found the setting to not allow installation unless the device is a phone, so it was time to rebuild and re-submit.
This was my very first submission to the Apple App Store, so another steep learning curve. Various different interfaces and pieces of information to provide, not necessarily in a flow of steps, more jumping back and forth. Now to wait.
On an eBay-sourced used MacBook with a smashed screen, connected to an external monitor, I started work on the basics. The iOS store was the first target, so I downloaded Xcode and got started. Several iterations, modifications to button layouts, flash functionality, button press delays, usage optimisation.
I got up to speed with the iPhone Simulator, how fun to have a virtual iPhone and iPad on your desktop to play with, and eventually had a working proof of concept.
Application for the Apple Developer account was placed. This is a whole other story, with lots of lessons learned along the way. Maybe I rushed things a little, but it's done now, and thankfully I won't have to go through the process again.
I never realised how much was involved in setting this up, and ensuring it is done in the correct way. I'd never heard of a D-U-N-S number before, but now I'm very familiar!
The thoughts which had been in my head could be contained no longer. I started researching if there was an existing method for raising an alarm at a well-populated event, in a noisy environment. I could find nothing even remotely close.
At the same time, I researched the media to find there were dozens of occasions where artists had stopped performances so that staff and stewards could locate, and tend to crowd members in distress. These stoppages received a lot of coverage, much on TikTok and Instagram, with some being headline news in national traditional press.
MobileFlare started with a simple question: if something went wrong at a live event right now, how would anyone nearby know? Standing in crowds at concerts across the UK, watching performances stop for crowd safety concerns and stewards struggle to locate people in distress, the answer was clearly: they wouldn't.
Your phone screen is the brightest object on your person. That is all MobileFlare needed to be. A colour-coded full-screen signal, visible above a crowd, that tells the people around you exactly what kind of help you need.
Get in touch
If something here resonates with you, if you have ideas for MobileFlare, or if you just want to say hello, we'd love to hear from you.