Content pfp
Content
@
0 reply
20 recasts
20 reactions

Colin Charles pfp
Colin Charles
@bytebot
Do we think much of client diversity @mvr @cryptonjal @pichi ? Ran this on Dune, suggesting that only 28.3% of casts came from warpcast in the last 30 days. Does super truly have 12924 unique users? Firefly at 20325? Coinbase wallet at 1968? Recaster at 223, with more casts, but Herocast at 505 (more users, less casts). This suggests that we're more diverse than we thought, client-wise, but I wonder if I made some mistake, so asking for a sanity check! WITH client_names AS ( SELECT u.fid as app_fid, u.value as client_name FROM dune.neynar.dataset_farcaster_user_data u WHERE u.type = 6 -- display name type ) SELECT s.app_fid, COALESCE(cn.client_name, CASE WHEN s.app_fid = 9152 THEN 'Warpcast' ELSE CONCAT('Unknown FID ', CAST(s.app_fid AS VARCHAR)) END ) as client_name, COUNT(DISTINCT s.fid) as unique_users, COUNT(DISTINCT c.hash) as total_casts, ROUND(COUNT(DISTINCT c.hash) * 100.0 / SUM(COUNT(DISTINCT c.hash)) OVER(), 2) as percentage FROM dune.neynar.dataset_farcaster_signers s LEFT JOIN client_names cn ON s.app_fid = cn.app_fid LEFT JOIN dune.neynar.dataset_farcaster_casts c ON s.fid = c.fid AND c.timestamp >= CURRENT_DATE - INTERVAL '30' DAY WHERE s.app_fid IS NOT NULL GROUP BY s.app_fid, cn.client_name ORDER BY total_casts DESC;
4 replies
0 recast
4 reactions

Njal pfp
Njal
@cryptonjal
Signers aren't casts or something. In the last 6 month I have 4 data points with my FID. It doesn't represent my app use, either. It would have been cool if they put the used app onchain for casts etc.
1 reply
0 recast
0 reaction

Colin Charles pfp
Colin Charles
@bytebot
yes, so how do we define how much client diversity we have? the numbers from firefly kinda check out re usage.
1 reply
0 recast
1 reaction