Kartik
@slokh
https://x.com/AlexMasmej/status/1886542478100324354
7 replies
17 recasts
60 reactions
shazow
@shazow.eth
how's your solana playing around going? anything you particularly like/dislike about the devex?
1 reply
0 recast
3 reactions
Kartik
@slokh
I actually like the storage model where everything is an account. Makes it pretty easy to pick up. Haven't written any programs yet though. The RPCs don't feel as robust. Using Helius, but calls are slower than on an EVM and have random failures from time to time so have to account for that. Indexing is a huge pain because of the insane amount of transactions (good problem for the network). Took me about a week to build a token indexer that can even stay at head. Storage, and in turn costs, grow pretty quickly though. Need to index myself because no good APIs available for consumer-oriented general-purpose token data. Before I even started the exploration, had so many EVM people telling me how much it sucks. And overall, it's honestly not bad at all. Most of the people talking negatively about Solana dev either (1) have never actually built anything there or (2) weren't around for EVM dev when it was the same age as Solana.
3 replies
1 recast
22 reactions
Samuel ツ
@samuellhuber.eth
cc @runninyeti.eth please fix solana indexing for Kartik
1 reply
0 recast
2 reactions
Brock
@runninyeti.eth
@slokh let me know if you wanna chat 😉 I've been building /indexing after seeing all those same problems for years - so we've built a generalized, distributed ETL for data on demand. We effectively "solved" getting real time data for any network; then layered on a transformation layer, parallelized backfills, and a direct-to-you delivery service. Long story short, here's a quick start for getting filtered Solana txs as a webhook - transformations are written in JS and what's available via self-service is only the happy-path of what's possible: https://indexing-co.notion.site/Solana-Quickstart-18625f03105380c59c51d603cf318360
0 reply
0 recast
1 reaction
