Even prettier options -- like the brand + tagline pairs used by the @recordcollector.edavis.dev labeler -- would need to be published or cached somewhere

tangled.org/@edavis.dev/...

But again, sorting out the desired UX is way more work than choosing where to deploy the teeny bit of json needed

Users can easily understand much of this PDS-centric perspective without resorting to techie vocab like "lexicon", "namespace", or even "collections"

As @zzstoatzz.io's at-me experiment showed, one icon + two.words = an entire app's worth of data in my PDS

Ditto for access by other apps

For a PDS to feel truly "personal" to me, I'll need a fairly straightforward, auditable model of which slices of "my" data I'm authorizing various services to modify

( or maybe a private log of which app made what changes when so I can undo it later )

Seems like there should be a small set of default patterns:

1. read-only (mine, others, all)
2. write mine only (lexicon = app)
3. update others (lexicon sets)
4. destructive (repo-wide)

You'll know you have the UX right when anyone can predict scopes for

- anisota, deer
- tangled
- pdsls

Thx for the pointer!

Confirming the range of sources cited both locally:

print = Sun-Times, Tribune
tv = NBC 5, ABC 7, WGN 9, FOX 32
radio = WBEZ
college = Northwestern
@blockclubchi.bsky.social

and beyond:

national = ABC, CBS, NBC, NPR, PBS
international = Independent, Mirror, Diario AS
@wordandway.bsky.social

As @antfu.me implies, approving PRs is always engineering work

The added functionality may be worth it for easily-reviewed PRs, but beyond that? Think of it as the first step in:

- recruiting a fellow developer,
- crediting your intellectual work,
- adding your skills/taste to the team

From a maintainer's perspective, the goal of soliciting contributions is to have more human eyes READING their codebase, assessing its quality + consistency (to see where it could be improved or cleanly extended)

Just getting you to look is already a win for the project

Hey @datarescueproject.org, have you come up with a strategy to estimate how many total signs need to be included in the #SaveOurSigns effort?

With the help of an insider, might be able to craft #FOIA requests to access internal sign inventory databases or records

( Somebody tracks that work 😉 )

I'm a huge fan of all of their work, but none of those individuals had a personal affinity for that environment

The OGs who're most likely to succeed via tiny startups are scrappy pragmatists who pounce on which parts of those grand visions happen to be deployable NOW

Overall a fantastic peek by @ntnsndr.in into the mindset of a protocol designer who's trying to build an ecosystem of "small teams" capable of replacing + improving solutions anywhere in the architecture

( Especially insightful for all the folks @wesleyfinck.org calls "builders" on those teams )

The two goals driving the core team's design choices:

"We would iterate on architectures and ideas and shift things around. Does this design give us credible exit? Can other people build on top of it? Will it avoid one party moderating the entire internet? We came back to those questions a lot."

NB: In the 70s + 80s, that standard required always testing each new design against 200 kids

But statistical improvements helped @cpsc.gov identify when only 50 would suffice, so these regs were updated in 1995 to also ensure that enough older adults (age 50-70) could do better than the kids

One-tap direct access to these settings (on mobile or web):

Get it done in three taps (mobile) or clicks (web).

1. follow the link
2. toggle the setting
3. come back

Easy-peasy!

To be clear, these mechanisms are nowhere near as lightweight as following a pointer to another record inside an in-memory database

And in older models, like REST, you got addressable access to a representation of selected resources

But here, it's the canonical thing itself (wire == PDS). Wow

Which means that you can then define a lexicon allowing other records to point to that record (Ref) or even a specific version of it (StrongRef)

Which means (modulo some type magic) you can know what you're getting, from where, before doing the lookup

For anyone willing to provide ALT text, it's easy to get friendly nudges