I think we’re exploring slightly different problems.
What you’re doing seems focused on distributing race events and timing information.
What I’m interested in is capturing observations that never make it into the tracker in the first place and using them to make the race more understandable.
Fans get a public version. Commentators get a deeper intelligence layer designed to surface context, patterns, and developments before they’re obvious on screen.
Aw fantastic stuff! Am I right in saying this is automated to give details when an athlete reaches a checkpoint, then publish to a feed? If so, this is a solid starting point, but isn’t adding context to the race, more repackaging what is already being collected?
It was populated manually (by me). The issue is with the SMS/push update to device function, you’re limited in how many characters each note can be without being cutoff for recipients who choose to receive updates that way (we had nearly 400 SMS subscribers). By adding more “colour” to each, you’d need to surrender the SMS feature - messages were getting cutoff halfway (e.g., scroll to the bottom (start) of the feed - those had more detail/context, but could not be delivered by SMS in full) . The decision was that splits/gaps was the best use of these updates.
I think we're exploring slightly different problems.
What you're doing seems focused on distributing race events and timing information.
What I'm interested in is capturing observations that never make it into the tracker in the first place and using them to make the race more understandable.
Fans get a public version. Commentators get a deeper intelligence layer designed to surface context, patterns, and developments before they're obvious on screen.
Great read and super insightful. You gave some things to think about and research. I came to this through lens of an entrepreneur knowing nothing about how leagues control data. After reading this, it makes more sense to me.
I built this in a way that preserves the spirit of ultrarunning - human observation, human moderated, cleaned by AI and given to on-screen human commentators.
At first I would hand select the human sensors in the field through the Borderlands community.
This got me thinking about a lot of things. The biggest thing was provoked by:
“ And the fan base has a demonstrated appetite for depth that most sports would envy. Think about it, these are people who will watch a GPS dot move across a map for twelve hours if the narrative is compelling enough.”
I’m not sure this is true for most races. almost races don’t have fan bases. Well, not beyond partner, mom, dad and best friend. It’s just always seemed to be that there aren’t enough non participant fans in the sport to sustain anything like what’s being discussed beyond handful of races.
It was a thought provoking article and Josh is the Beat!
In the world of AI vibe-coding, I'd love to see outlets be more selective in what they promote instead of incentivizing shortcuts and the shipping of half-baked ideas that don't do a service to the broader need (which will take thoughtfulness to solve in a way that gets fans excited and using a platform).
Otherwise, it just feels like a marketing/engagement farming strategy.
I agree that usefulness has to be the standard. The advantage of vibe coding isn't that it lowers the bar for quality. It's that it lowers the cost of discovering whether an idea is worth pursuing in the first place. I'd rather see weak ideas exposed in weeks than hidden behind months of development.
Totally, buts that’s about a quarter of the idea here. For fans it would be something similar to what irunfar did so well. But there’s more to it than that 😉
https://www.mountainoutpost.com/events/cocodona-250-2026/feed
We’ve (Jamil and I) already created this. Looking to improve for the WSER broadcast.
Fair play, that’s a hefty number of sms subscribers! Fantastic that you experimented with that! You reckon you’ll build on it for another race?
Plan is to go at it again with Western States and perhaps make some minor tweaks, and see how it goes!
Really interesting, Liam.
I think we’re exploring slightly different problems.
What you’re doing seems focused on distributing race events and timing information.
What I’m interested in is capturing observations that never make it into the tracker in the first place and using them to make the race more understandable.
Fans get a public version. Commentators get a deeper intelligence layer designed to surface context, patterns, and developments before they’re obvious on screen.
Both feel valuable.
Aw fantastic stuff! Am I right in saying this is automated to give details when an athlete reaches a checkpoint, then publish to a feed? If so, this is a solid starting point, but isn’t adding context to the race, more repackaging what is already being collected?
It was populated manually (by me). The issue is with the SMS/push update to device function, you’re limited in how many characters each note can be without being cutoff for recipients who choose to receive updates that way (we had nearly 400 SMS subscribers). By adding more “colour” to each, you’d need to surrender the SMS feature - messages were getting cutoff halfway (e.g., scroll to the bottom (start) of the feed - those had more detail/context, but could not be delivered by SMS in full) . The decision was that splits/gaps was the best use of these updates.
Really interesting, Liam.
I think we're exploring slightly different problems.
What you're doing seems focused on distributing race events and timing information.
What I'm interested in is capturing observations that never make it into the tracker in the first place and using them to make the race more understandable.
Fans get a public version. Commentators get a deeper intelligence layer designed to surface context, patterns, and developments before they're obvious on screen.
Both feel valuable.
Great read and super insightful. You gave some things to think about and research. I came to this through lens of an entrepreneur knowing nothing about how leagues control data. After reading this, it makes more sense to me.
I built this in a way that preserves the spirit of ultrarunning - human observation, human moderated, cleaned by AI and given to on-screen human commentators.
At first I would hand select the human sensors in the field through the Borderlands community.
This got me thinking about a lot of things. The biggest thing was provoked by:
“ And the fan base has a demonstrated appetite for depth that most sports would envy. Think about it, these are people who will watch a GPS dot move across a map for twelve hours if the narrative is compelling enough.”
I’m not sure this is true for most races. almost races don’t have fan bases. Well, not beyond partner, mom, dad and best friend. It’s just always seemed to be that there aren’t enough non participant fans in the sport to sustain anything like what’s being discussed beyond handful of races.
It was a thought provoking article and Josh is the Beat!
Thanks.
Thanks, Scott! I think you're right. This is really (concepted) for the big races we are already showing up for as fans - WSER, UTMB, and Cocodona.
Thanks for reminding me I love the word palava
It’s a beautiful word that surmises a lot of fluff on the internet.
In the world of AI vibe-coding, I'd love to see outlets be more selective in what they promote instead of incentivizing shortcuts and the shipping of half-baked ideas that don't do a service to the broader need (which will take thoughtfulness to solve in a way that gets fans excited and using a platform).
Otherwise, it just feels like a marketing/engagement farming strategy.
It’s a nice idea that has legs. Sure there’s lots of work to do, but at least he’s trying to find a solution 🤷
I agree that usefulness has to be the standard. The advantage of vibe coding isn't that it lowers the bar for quality. It's that it lowers the cost of discovering whether an idea is worth pursuing in the first place. I'd rather see weak ideas exposed in weeks than hidden behind months of development.
Finally, with the power of AI, we have created the ultimate trail running fan experience - reading the irunfar twitter feed circa 2017
Totally, buts that’s about a quarter of the idea here. For fans it would be something similar to what irunfar did so well. But there’s more to it than that 😉