Investor in very early stage startups - I try to help every company have the best chance of success, and about 1 in 3 do (vs 1 in 30+ for accelerators). https://zerocapital.vc/
https://linkedin.com/in/adamsah
https://github.com/asah
- asahagree re policy, but technically... it's possible that today's drones are OK but they're worried about future drones including something new...
- counterpoint: IRL, data values in a system like PostgreSQL are padded to word boundaries so either you're wasting bits or "carrying data."
- serious q: maybe not 60 sec, but why 45 days instead of ~1 day or even hours? at 45 days, it pretty much has to be automated.
- PostgreSQL extension providing big speedups on COUNT/SUM/DISTINCT and GROUP BY for the most common data types.
I'm looking for people who have pain around slow analytics, avoiding migration from PostgreSQL, delaying pg upgrades or other big reasons to adopt something like this.
- I want an AI optimization helper that recognizes patterns that could-almost be optimized if I gave it a little help, e.g. hints about usage, type, etc.
- That was true six months ago - the latest versions are much better at memory and adherence, and my senior engineer friends are adopting LLMs quickly for all sorts of advanced development.
- well... it's a certain KIND of reality... one where numbers fight with "common sense"...
examples... a large paying customer can kill a business... tiny or free users can be great for free marketing and product testing... a weird channel partner can make a business... obscure cashflow and accounting can make/break a business... product development or inventory can require fundraising which comes with wild "strings attached"... and and and...
(having started a number of both self-funded and venture-funded business, in tech small format retail and more...)
- LOL "The entity is named “Beignet,” presumably because “Off-Balance-Sheet Leverage Vehicle No. 5” tested poorly with focus groups."
- ??? seems straightforward... among other things, require the applicant to do the work / provide evidence...
- True serializability doesn't model the real world. IRL humans observe something then make decisions and take action, without holding "locks" on the thing they observed. Everything from the stock market to the sitcom industry depend on this behavior.
Other models exist and are more popular than serializability, e.g. for practicality, PostgreSQL uses MVCC and read consistency, not serializability.
- Interesting analogy!
One big difference: - ~10% of Americans work in healthcare (i.e. big source of income) - ~66% of Americans own their homes (i.e. big asset want to protect) - there's societal stability reasons to encourage home ownership
- -1: there's lots of "kingdoms" (openai, anthropic, google, plus open source) - if one king comes for your castle, you can move in minutes.
- I hate these kinds of bills too, but it's a logical fallacy to address only the single biggest problem (assuming you agree on what it is).
- But then we'd complain about Google being a slow moving dinosaur.
"Move fast and break things" cuts both ways !
(ex-Google tech lead, who took down the Google.com homepage... twice!)
- Competing species (non-calcifying) then take over...
- New (open source) PostgreSQL index type for analytics workloads, which is a read-only drop-in replacement for B-trees. Smol is multiplicatively faster than B-Trees and radically smaller.
Help, alpha testers, etc all welcome. Sorry RDS/Aurora users: smol is for embedded and self-hosted pg instances only for the foreseeable future.
- lol, the majority of the world isn't remotely Christian...
- WeWork was taking on long term liability commitments and paying for them with short term revenue commitments. One bad thing and poof. Everybody in the commercial real estate market saw this coming.
OpenAI maybe in the same situation, committed to spending $1.4T while enjoying a good revenue year this year but then One Bad Thing and poof.
- "500 KB/s workload should not use Kafka" - yyyy!!! indeed, I'm running 5MBps logging system through a single node RDS instance costing <$1000/mon (plus 2x for failover). There's easily 4-10x headroom for growth by paying AWS more money and 3-5x+ savings by optimizing the data structure.
- FTFY: "for now"