Paul Rohr
@pevohr.bsky.social
560 followers 570 following 2K posts
Dad, startup guy. Ideas matter. Design matters. It's about we, not me. hachyderm.io/@pevohr
Posts Media Videos Starter Packs
pevohr.bsky.social
Resilient protocols choose their use cases wisely

If you design for a sweet spot like this, then support for other subsets of user cohorts (unharassed, abled, high bandwidth, etc.) falls out as an easy side effect
hazelweakly.me
It’s *so* possible to sit there and figure out a protocol that’s usable by a disabled person in a rural area with ancient computing devices undergoing harassment. You know how many fucking edge cases that covers?

It’s significantly cheaper to do that than to come up with edge cases yourself, too!
Reposted by Paul Rohr
hazelweakly.me
From one protocol designer to another: the best protocols are built by thinking of the use cases and tooling and enablement first.

Here’s a secret: Design for the margins and the marginalized, then the capabilities fall out naturally and the resulting protocol is extraordinarily robust
Reposted by Paul Rohr
scotto.org
An Irish choral composition (“the BEST song about seaweed” according to a commenter) gets a polyphonic spin by Spokanki, “8 sisters from the mystical country of SpokanLand, wandering around all corners of the planet to search and distribute music of different peoples from Africa to the Netherlands.”
SPOKANKI - DULAMAN
YouTube video by SPOKANKI
www.youtube.com
Reposted by Paul Rohr
epikt.bsky.social
still absolutely wild that only one of these schools has rejected this open extortion out of hand, would be difficult to devise a more obvious test of whether you’re baseline competent to run a university or not
laprofmme.bsky.social
Courtesy of colleagues at Penn
A variation on Benjamin Franklin’s 1754 engraving in the Pennsylvania Gazette, published in Philadelphia, in an effort to foster cooperation among the various colonies against the tyrant king of Britain, with the American colonies replaced by the nine universities selected for special collaboration opportunities
pevohr.bsky.social
Just curious. What keeps items from sliding side to side when cornering? Hi-friction texture on the floor?
Reposted by Paul Rohr
jeremymberg.bsky.social
This is a powerful account from a PhD student whose fellowship was terminated by NIH because it was part of a "diversity" program without any other considerations or understanding of what the program or the students were actually doing.

undark.org/2025/10/09/o...

1/2
NIH Student Grant Cancellation Will Weaken Scientific Innovation
The termination of federal F31 diversity fellowships puts many graduate students in a bind — and U.S. science at risk.
undark.org
Reposted by Paul Rohr
mhairiforrest.bsky.social
Important thread on the metaphorical bomb that exploded in US special education & disability rights on Friday evening.
novicsara.bsky.social
Yesterday we received word that the Office of Special Education and Rehabilitative Services (OSERS) within Dept of Ed (DoEd) was gutted--right now we believe there is no one or nearly no one left. But what does it mean? 1/11
pevohr.bsky.social
Bingo. "curious ... ruffle"

What a great example of how well-chosen titles can help micro-target the audience he's seeking!
pevohr.bsky.social
I guess a better way to express my lack of surprise is that following feed requests seem like both:

- a very frequent one-off query, +
- an inherently expensive one

So why wouldn't they dominate your overall workload? TBH, I'd be seriously impressed if they didn't!
pevohr.bsky.social
Even thread views seem fairly straightforward at the DB layer -- grab a windowed set of replies with the same OP, sorted to make it easier to infer reply sequences

(Especially since partial results for popular threads might be eminently cacheable for reuse by other readers)
pevohr.bsky.social
Oh, I get that the following feed is way more complex to assemble than the other workloads, so you really want to be efficient at it

The inherent fanout (or multiplexing) of posts to all followers is nowhere near as optimized as other simple CRUD operations:

- reads
- writes
- counters
pevohr.bsky.social
Naive question, but why is this surprising?

Isn't reading a feed from people you follow THE primary activity on a social network? If so, why shouldn't that dominate your workload (via sheer volume)?

Or was the hope that people would lean more heavily on algorithmic feeds provided by others?
pevohr.bsky.social
I love overhearing the puppeteer's inner monologue:

"Silly robot. Stop making things up + clone *that* code specifically

No, not like that, more like *this*

Oh, I see where you're getting stuck. Here, let me fix that for you. Now try again

That's better. Good robot!"
pevohr.bsky.social
aka #POSIWID
dieselbug1137.bsky.social
If the only thing that's fully funded by the government is the police. Then you live in a police state. That's just how numbers work. It's fair to judge a society by how it chooses to allocate resources.
Reposted by Paul Rohr
swearyanthony.bsky.social
So they've fired everyone at the CDC EIS (epidemic intelligence service, the "disease detectives") and the folks who do www.cdc.gov/mmwr/index.h... so a big day for Nurgle and other death cults. Also everyone at CISA. Fucking hell.
Morbidity and Mortality Weekly Report
The MMWR series is the agency’s primary vehicle for ,,,
www.cdc.gov
Reposted by Paul Rohr
convolver.bsky.social
And now CISA, America’s front line for digital security (and a showpiece of Trump I until they refused to lie about election security) is packed up and unceremoniously shoved towards the exit.
skiles.blue
I don't know if there's such a thing as a "world cyberwar" but I think we are in one? and the US is determined to lose it. 🧵

Today the admin is mass-firing the Cybersecurity and Infrastructure Security Agency (CISA), the nation's cybersecurity protection across all levels of government.
Reposted by Paul Rohr
kathleenclark.bsky.social
A master class from MIT in responding to authoritarian overreach:

Your “premise … is inconsistent with our core belief that scientific funding should be based on scientific merit alone.
… America’s leadership in science & innovation depends on independent thinking & open competition for excellence.
Dear Madam Secretary,
I write in response to your letter of October 1, inviting MIT to review a "Compact for Academic Excellence in Higher Education." I acknowledge the vital importance of these matters.
I appreciated the chance to meet with you earlier this year to discuss the priorities we share for American higher education.
As we discussed, the Institute's mission of service to the nation directs us to advance knowledge, educate students and bring knowledge to bear on the world's great challenges.
We do that in line with a clear set of values, with excellence above all. Some practical examples:
• MIT prides itself on rewarding merit. Students, faculty and staff succeed here based on the strength of their talent, ideas and hard work. For instance, the Institute was the first to reinstate the SAT/ACT requirement after the pandemic. And MIT has never had legacy preferences in admissions.
• MIT opens its doors to the most talented students regardless of their family's finances. Admissions are need-blind. Incoming undergraduates whose families earn less than $200,000 a year pay no tuition. Nearly 88% of our last graduating class left MIT with no debt for their education. We make a wealth of free courses and low-cost certificates available to any American with an internet connection. Of the undergraduate degrees we award, 94% are in STEM fields. And in service to the nation, we cap enrollment of international undergraduates at roughly 10%.

source: 
https://orgchart.mit.edu/letters/regarding-compact • We value free expression, as clearly described in the MIT Statement on Freedom of Expression and Academic Freedom. We must hear facts and opinions we don't like - and engage respectfully with those with whom we disagree.
These values and other MIT practices meet or exceed many standards outlined in the document you sent. We freely choose these values because they're right, and we live by them because they support our mission - work of immense value to the prosperity, competitiveness, health and security of the United States. And of course, MIT abides by the law.
The document also includes principles with which we disagree, including those that would restrict freedom of expression and our independence as an institution. And fundamentally, the premise of the document is inconsistent with our core belief that scientific funding should be based on scientific merit alone.
In our view, America's leadership in science and innovation depends on independent thinking and open competition for excellence. In that free marketplace of ideas, the people of MIT gladly compete with the very best, without preferences. Therefore, with respect, we cannot support the proposed approach to addressing the issues facing higher education.
As you know, MIT's record of service to the nation is long and enduring. Eight decades ago, MIT leaders helped invent a scientific partnership between America's research universities and the U.S. government that has delivered extraordinary benefits for the American people. We continue to believe in the power of this partnership to serve the nation.
Sincerely,
Sally Kornbluth
pevohr.bsky.social
Nice simplification!

What's the use case where the final toggle is needed as an explicit option?

Wouldn't it be even simpler to just omit that + always default the other settings to whatever you used the previous time? (It's an easily learnable behavior)
pevohr.bsky.social
Papal infallibility is only an article of faith within a single religious tradition

You can fervently believe in the importance of Constitutional law without having to treat SCOTUS with undue reverence. If something's gone terribly awry there, then we find proportionate ways to fix it
pevohr.bsky.social
On legitimacy:

"That to secure these rights, Governments are instituted among Men, deriving their just powers from the consent of the governed"

Nobody's claiming the entire federal judiciary is illegitimate. Just this particular collection of 9 fallible robed humans
Reposted by Paul Rohr
gracekind.net
You can slice a cube in half and get a hexagon ??
pevohr.bsky.social
How do you envision this mechanism working if downstream labelers aren't just emitting a noisy subset (or superset) of upstream label records under a new namespace?

Does adding DAGs somehow become a clever namespace aliasing hack which makes partial join results more cacheable within AppViews?
pevohr.bsky.social
Effectively, the AppView winds up locally doing a logical JOIN across its copy of three kinds of data:

- raw content records
- matching label records
- user moderation config records

... with the last category acting as a namespace filter to only include records from specific labelers