undefined | Better HN

0 pointsgoeiedaggoeie1y ago0 comments

I think you misunderstand that case "compiled its scores and statistics by employing people to listen or watch the games, then enter the scores on the computer which transmits the scores to STATS' on-line service, to be sent out to anyone using a SportsTrax pager.[1]"

Notice how they watched the game and got the statistics like that. The restrictions are about using the scoreboard and the data displays and reselling/commercialising that data. It is however legal to watch the game and compile and distribute your own stats due to the game entering public domain.

Due to this many betting companies and data collection companies have to pay people to watch the game vs just scraping the scoreboard (which is the context from which I learnt about this). ironically at venue OCR is a common way to get scoreboard data.

0 comments

2 comments · 1 top-level

FiberBundle1y ago· 1 in thread

I'm not a lawyer, but my interpretation of the lawsuit based on the Wikipedia article is that game results/scores are public facts and hence not copyrightable data. I don't see how the method by which that public data is collected changes anything materially about that case. Are you saying that inferring the score based on the scoreboard is what makes this illegal (why?)? What if they would infer the score using motion/ball tracking instead?

goeiedaggoeieOP1y ago

if you using CV to track the player, the ball etc from a broadcast it is fine, the scoreboard however not so straight forward. fwiw, doing CV from broadcast for accurate scoring of sports is neigh near impossible due to edges, but human in the loop systems exist. there are also numerous in venue CV systems which auto collect game and player information.

j / k navigate · click thread line to collapse