Replying to
Put another way, if a problem statement is really dialed in, the solution falls out from it (as do the other options, & why they may not work)
Remote
Chris Dickinson
@isntitvacant@hachyderm.io
mastodon 4.7.3current principal eng at dylibso; former Node.js TSC & NPM registry eng / systems eng (❤️ rust, bash, js, tf) / sometimes illustrator / pets cats; pronouns he/him
0 Followers
0 Following
17 Posts
Joined November 15, 2022
Website:
Twitter (RIP):
Replying to
@munificent@mastodon.social well put!
(another element of meaning, other than time, might be “vulnerability”: in the sense that all craft is rehearsal for later work, making something for someone else gives them a marker of where you were and how you thought about that craft. Which can be scary! You put a piece of a version of yourself into the work— as you were when you made that work — and you trust the receiver will value that piece, just as it reflects how you value them.)
This has been a fascinating experience, working on a rust service for $dayjob that ingests data whose shape and meaning in the domain was unknown to me at the beginning of the month.
It’s been a lassoing experience: hold the data as loosely as possible (as json), then gradually define it as necessary. Opposite to most of my rust experience, where “thinking about the exact structure of the data” had been front-loaded.
OK! Without further ado, here it is. Part 2 of "Understanding WASM", "Whence WASM", written to answer the questions "what makes WASM different from the JVM?" and "what exactly *is* a 'virtual ISA'".
Please enjoy the attached footnotes and bibliography/list of relevant links!
https://www.neversaw.us/2023/06/30/understanding-wasm/part2/whence-wasm/
redesigned my website. bonus main page content: git HEAD shots over the years. ever wondered what I looked like as an awkward teen? wonder no more:
Replying to
I've shaved one yak over this break, don't tempt me with a SECOND yak
Replying to
and I’m thinking about this from the lens of your typical, run-of-the-mill CRUD-y HTTP REST API.
as you know, this kind of API pushes you towards the “row-major” object model & away from “columnar” SOA. If I pick at “why”: it strikes me that the REST API is a narrow interface, and the effect of the narrowing is to widen the data representation in both directions. (put another way, the data is made more general to reduce its cardinality & make it more cache-friendly)
the alternative being something like graphql, which widens the interface (while tunneling over a single http endpoint)
Replying to
Replying to
This falls out from the requirements on INSERT queries in postgres (having to decompose the array-of-structs into a series of arrays to UNNEST) but pops up when querying, too ("enrich this list of ids from another query")
Replying to
@sunfish (A follow-up third question that just occurred to me: have I accidentally committed a category error by responding to your call for WASI questions with questions about WIT and wasm components? 😅)
Replying to
@sunfish ack yes, thank you. I flubbed the “web components” part — I meant “wasm components”!