sampotter
199 posts
sampotter
@sampotter___
building a block explorer @univo_app
Melbourne Katılım Şubat 2018
1.2K Takip Edilen147 Takipçiler
Tired: this.update()
Wired: Async React, Suspense & Transitions, No Waterfalls, Optimistic & Live Updates, Great UX with a minimal API.
fate.technology
English
@MaineFrameworks SQLite falls under the same set of trade offs and this graph suggest it only handles 5k TPS
English
@sampotter___ All product benchmarks indicate the differences between such tradeoff decisions.
English
I don’t get why they commit to this bit? A database company should want competent engineers as its customer. But every competent engineer sees these benchmarks as disingenuous?
Tyler F. Cloutier@TylerFCloutier
Back in March we benchmarked SpacetimeDB 2.0 at ~100,000 TPS and got called liars, morons, goons, and "not a very good database." Well, today we're responding to the criticism point by point, and you better believe I brought receipts. \1 🧵
English
@TylerFCloutier Exactly so why market like this and just piss them off?
English
@sampotter___ Competent engineers understand the value of stored procedures.
English
Guy who’s famous for hating client-server protocols: I’ve built a client-server protocol
DuckDB@duckdb
Hannes Mühleisen just revealed the Next Big Thing for DuckDB at AI Council 2026: Quack, a protocol that turns DuckDB into a client-server database. True to DuckDB's philosophy, Quack is simple and fast. You can set it up in seconds and it delivers high-performance remote access, turning DuckDB into a full-fledged general-purpose database system. We are very excited to see the possibilities Quack will unlock, from serving multiple concurrent writers to creative architectures with several DuckDB processes talking to each other. The ingenuity of our community never ceases to amaze us and we're certain that we'll see many use cases we did not even think of. For now – follow the link in the thread, give it a spin and let us know what you think. Happy quacking!
English
OLI labels are now available as public BigQuery tables!
~170M contract labels are queryable directly in BigQuery, including decoded labels and raw attestations.
growthepie.oli_public.labels
growthepie.oli_public.attestations
Built for analysts, researchers, and builders working with open onchain labels.
English
Cloud Zero is now available and production-ready: fully managed Zero hosting from the team that built it.
- no severs to deploy or maintain
- 24/7 monitoring
- automatic zero-downtime updates
- optional bring-your-own-cloud deployment
- migration help for existing self-hosted setups
zero.rocicorp.dev/cloud
Aaron Boodman@aboodman
When we started Zero, we weren’t even sure there _was_ a correct design for the kind of query-based sync engine we dreamed of. We spent two years finding that design, then hardening it for production use. We always imagined a hosted version too, because (a) running a sync server is never going to be fun 😅, and (b) gotta make money! Starting late last year, we were finally able to begin work on that hosted version. Initially it was invite-only and small, but we're now in production and syncing thousands of messages per minute for teams like @productlane, @passionfroot, and @shipper_crm. Cloud Zero is production-grade Zero hosting from the team that built it: - no servers to run or maintain - 24/7 monitoring and alerts - automatic zero-downtime updates - fast NVMe-based architecture - run as traditional SaaS or in your own AWS account We’re onboarding manually, but it’s quick. And if you’re already self-hosting Zero, we’ll help migrate your setup. Get started today at: zerosync.dev/cloud
English
Thanks to @web3_data and @growthepie_eth for starting this initiative. Looking forward to contributing in future!
univo@univo_app
The explorer now uses @open_labels for contract and account names.
English
@tan_stack PR can be found at github.com/univo/explorer…. One of the unique challenges was determining what work happens on the server, versus on the client. RSC makes this simple as it unifies the templating language between these two distinct environments
English
.@tan_stack Start has made it possible for the RSC to be fetched as a GET request, meaning that CDN caching is trivial to implement with Cache-Control headers
English
This was a fun one to implement. Under the hood each batch of events is an RSC! This allows us to support infinitely many different types of events and stay performant. I’ll link the PR below
univo@univo_app
The block explorer now supports infinite scroll when viewing accounts
English
@samy_0202 Materialised views with incremental updates on write are overlooked. There needs to be more work on them
English
Materialized views, the underused middle ground between raw queries and full caching: A materialized view is a precomputed query result stored on disk.
Unlike a regular view: it doesn't recompute on every query. It reads precomputed data.
Unlike application-level caching: the DB handles consistency and refresh.
Use when:
→ You have expensive aggregations that don't need real-time freshness
→ Reporting queries are hurting your OLTP performance
→ You want the DB to own the caching logic
Refresh strategies: on-demand, on schedule, or on data change (incremental).
Supported in: Postgres, MySQL 8+, BigQuery, Redshift.
Underused in: most production systems I've seen.
English
sampotter retweetledi
@__morse There's some huge unlocks for explorer.univo.app that I will have complete in the next few days to showcase what's not only possible now, but effectively very easy
English
@sampotter___ what do you mean? you could do this already with server actions no?
English