Sign in

Jonathan Aldrich

@jonathanaldrich.bsky.social
2.3K followers 781 following 930 posts

Professor at Carnegie Mellon University, working on programming languages and software engineering. Coauthor, Programming Language Pragmatics (videos: tinyurl.com/PLP5vids). CTO of noteful.net ACM Publications Board member. He/him.

PostsRepliesMedia
Jonathan Aldrich @jonathanaldrich.bsky.social · 30/09/2026
If you're working in Software Engineering and want to change your career trajectory, join our online Master of Software Engineering program, part of the newly named CMU Griffin School of Computer Science! Info session tomorrow 9am, register at admissions.scs.cmu.edu/register/mse...
admissions.scs.cmu.edu
MSE Program Information Sessions
030
Jonathan Aldrich @jonathanaldrich.bsky.social · 29/09/2026
We should call it souperintelligence because we are all drowning in AI soup
141
Jonathan Aldrich @jonathanaldrich.bsky.social · 14/09/2026
Longstanding, actually! It was coined by Wang and Beni in 1989, in the context of cellular robotic systems. While there were robots involved, I'd say the concept was more about the software than the hardware, so should qualify to answer your question? en.wikipedia.org/wiki/Swarm_i...
en.wikipedia.org
Swarm intelligence - Wikipedia
190
Jonathan Aldrich @jonathanaldrich.bsky.social · 14/09/2026
It never made sense to regulate C02 via the EPA's pollution rules, it's just too different from other forms of pollution. Carbon taxes are the right mechanism. www.nytimes.com/2026/09/13/c...
nytimes.com
E.P.A. Will Erase Limits on Climate Pollution From Power Plants
Generation of electricity is the second largest source of carbon dioxide and other planet-warming gases in the United States.
000
Jonathan Aldrich @jonathanaldrich.bsky.social · 13/09/2026
Happy New Year to my Jewish friends!
040
Jonathan Aldrich @jonathanaldrich.bsky.social · 06/09/2026
There are many legal benefits to marriage, that's not a problem. But I'd agree that it should be available to single parents, which would take care of that anyway.
000
Jonathan Aldrich @jonathanaldrich.bsky.social · 05/09/2026
I don't agree that making the program applicable to more people who can benefit from it is "punitive" to others. But I certainly would support expanding the funding when making a change like this. In general we need more support for raising children; I hope congress agrees.
100
Jonathan Aldrich @jonathanaldrich.bsky.social · 05/09/2026
Sure they admit it. My point is the money should be available for child care whether you are working or not. The previous policy discriminated against those who stay home. This one is neutral. Isn't being neutral...objectively fairer?
100
Jonathan Aldrich @jonathanaldrich.bsky.social · 05/09/2026
No it's not, they would still be eligible.
100
Jonathan Aldrich @jonathanaldrich.bsky.social · 05/09/2026
Makes sense to allow stay-at-home parents to collect child care subsidies as a substitute for hiring someone else for child care. Then the subsidy is accessible to all parents! www.nytimes.com/2026/09/05/u...
nytimes.com
Trump Officials Draft Plan to Pay At-Home Parents, Using Funds for Working Ones
The proposed benefit, a top policy priority for Vice President JD Vance, would apply only to married couples and tap a fund intended to provide federal child care subsidies to working parents.
100
Jonathan Aldrich @jonathanaldrich.bsky.social · 24/08/2026
Don't be a meat proxy! Instead, use your wisdom, reasoning, and creativity to direct AI tools and to filter and enhance their output into something meaningful and useful.
010
Jonathan Aldrich @jonathanaldrich.bsky.social · 13/08/2026
A very well-written, thoughtful article. My university environment never reached the "woke" excesses described here, especially in computer science. But I definitely saw some aspects of this, and it's worth taking seriously. www.theatlantic.com/magazine/202...
theatlantic.com
Why I Quit the Tenure Track
The insufferable hypocrisy of elite academia
010
Reposted by Jonathan Aldrich
"Online Rent-a-Sage" Bret Devereaux @bretdevereaux.bsky.social · 13/08/2026
FWIW, I think the question of "Did Woke 1 go to far?" it depends on where you ask! On college campuses? Sometimes. In making cops responsible for bad behavior? Obviously not! In holding sex pests and criminals to account? Clearly incomplete, look at the damn White House.
411008103
Reposted by Jonathan Aldrich
Michael S. Kirkpatrick @kirkpams.bsky.social · 11/08/2026
I have a request for anyone who helps organize or leads an ACM SIG, conference, etc. I'm chairing a task force on publication integrity in CS (clear-computing.org). Part of that work is surveying the community on tools used. If you can help spread the word to SIG/conf leaders, please do.
forms.gle
CLEAR Survey: Tool Adoption For Ethics And Policy Violations
The CLEAR Task force is gathering information from conference and journal organizers about tools used or being developed to identify, investigate, or manage ethics and policy violations in peer review...
3611
Jonathan Aldrich @jonathanaldrich.bsky.social · 31/07/2026
"Light rail" is as old as I am. I like it!
010
Jonathan Aldrich @jonathanaldrich.bsky.social · 27/07/2026
Want to ensure the AI you embed in your applications doesn't go rogue like OpenAI's did? Take Machine Learning in Production, part of our SE with AI online certificate this fall, & learn how to build, deploy, assure, & maintain AI/LLM-powered software. Apply by July 29! www.cmu.edu/online/softw...
cmu.edu
Software Engineering With AI
Advance your career with Carnegie Mellon's Software Engineering & AI online graduate certificate. Learn to build, deploy, and manage AI-enabled software systems.
040
Jonathan Aldrich @jonathanaldrich.bsky.social · 26/07/2026
The solution to AI safety is not to train closed models so that they never "press the nuclear button." Because we can't stop the bad guys from finding workarounds or using open models. The solution is formal security guarantees that models don't have nuclear buttons to press.
091
Jonathan Aldrich @jonathanaldrich.bsky.social · 24/07/2026
Just this week we saw an illustration of how critical this is: an open weight model from China was key to Huggingface defending itself from an OpenAI model that went rogue.
031
Jonathan Aldrich @jonathanaldrich.bsky.social · 24/07/2026
Don't close the frontier! Open weight (and open source) models are key to US national security and competitiveness. A letter from NVIDIA, Microsoft, Meta, IBM, and others came out today urging politicians not to impose hasty regulations. www.reddit.com/r/singularit...
reddit.com
From the singularity community on Reddit: Microsoft, NVIDIA, Meta, IBM, Palantir and more released a joint letter warning Washington not to kill open-weight models
Explore this post and more from the singularity community
110
Jonathan Aldrich @jonathanaldrich.bsky.social · 24/07/2026
@acm.org President Elisa Bertino clarifies that no decision has been made on opening the Digital Library to LLM training. The ACM wants your feedback on this! cacm.acm.org/news/a-messa...
cacm.acm.org
A Message to the ACM Community – Communications of the ACM
131
Reposted by Jonathan Aldrich
Association for Computing Machinery @acm.org · 16/07/2026
If you would like to share your thoughts, please use this link (buff.ly/fnANR1x). ACM leadership will take your feedback into account while we develop and implement our AI content strategy over the coming months.
docs.google.com
Now is the Time to Give LLMS Access to the ACM Digital Library Why ACM Believes the benefits outweigh the risks in opening up the ACM Digital Library to large language models
ACM welcomes your feedback! All responses will be read and considered by ACM.
784
Reposted by Jonathan Aldrich
Association for Computing Machinery @acm.org · 16/07/2026
We are opening a consultation on the inclusion of ACM publications within AI licensing agreements for access and training purposes. We have published this piece as to why we believe it is the right time now to enter into these discussions. buff.ly/VTatuda
1475
Reposted by Jonathan Aldrich
Audrey @parickards.bsky.social · 16/07/2026
Yesterday...Northwestern Ontario looked like a scene straight out of Dante’s Peak.
26040971231
Reposted by Jonathan Aldrich
Andrew Begel @abegel.bsky.social · 16/07/2026
Come take my AI Tools for Software Development course!
041
Jonathan Aldrich @jonathanaldrich.bsky.social · 16/07/2026
My take on daylight savings time: when the sun rises, it should be 7am. Every day. That's what your body wants. Clocks have computers in them now, we can do this!
030
Jonathan Aldrich @jonathanaldrich.bsky.social · 15/07/2026
Apply by July 29, 2026 to join our first cohort of students in the SE with AI Certificate program! (3/3)
000
Jonathan Aldrich @jonathanaldrich.bsky.social · 15/07/2026
Working SE professionals can learn what they need to be productive in the AI world through our 9-month certificate program. Take one interactive online class per week in the evening, paired with interesting and challenging readings and homework projects. (2/3)
100
Jonathan Aldrich @jonathanaldrich.bsky.social · 15/07/2026
CMU is launching a new online professional graduate certificate program: Software Engineering with AI! www.cmu.edu/online/softw... Learn how to use AI to build software more productively, and how to effectively use AI as a component in your applications (1/3)
112
Jonathan Aldrich @jonathanaldrich.bsky.social · 15/07/2026
I would additionally note that you can (and probably should!) publish the supplementary material that goes with a paper as part of the ACM Digital Library, and/or in other archival places such as an artifact repository.
010
Jonathan Aldrich @jonathanaldrich.bsky.social · 14/07/2026
It is a common distinction to make in moral theory. I wouldn't exactly say it is blurry, but I agree that people have different views on how important it is. To a pure utilitarian they are indistinguishable; to a Kantian like myself, not so much.
000
Jonathan Aldrich @jonathanaldrich.bsky.social · 08/07/2026
Had a great chat today with ACM India president @pkprofgiri.bsky.social, learned about all the great things they are doing--awards, speakers, hackathons, industry and student events, etc. Lots of chapter activity & growing membership, would love to see more of this across ACM!
020
Jonathan Aldrich @jonathanaldrich.bsky.social · 06/07/2026
All of which is sad because on a technical level I love Linux and Android, I have an Android phone, etc....
010
Jonathan Aldrich @jonathanaldrich.bsky.social · 06/07/2026
Totally bizarre that in Windows & Mac all I have to do to connect to campus wifi is enter my username and password, but in Linux or Android I have to set the "EAP Phase 2 authentication" etc. to get anything to work. And people wonder why Linux/Android is considered less usable.
420
Jonathan Aldrich @jonathanaldrich.bsky.social · 04/07/2026
America has plenty of issues, and the current administration is making them worse. But we still have a ton of strengths--smart, diverse people & a can-do culture. There's no place in the world where I'd rather be a citizen. Happy 4th of July! And happy 250th birthday to the USA!
010
Jonathan Aldrich @jonathanaldrich.bsky.social · 02/07/2026
So please continue to share your ACM related concerns, but not just about pubs! I'll try to share what is going on. You may see some posts from me on possible ACM directions; some might be my own ideas, others might be things ACM is thinking of--feedback appreciated either way!
050
Jonathan Aldrich @jonathanaldrich.bsky.social · 02/07/2026
Yesterday, an ending. Today, a beginning: I'll be serving as a SIG Governing Board representative on ACM Council for the next 2 years. I'll contribute to strategic planning for @acm.org, but my biggest goal is to facilitate communication between Council, SGB, & members.
170
Jonathan Aldrich @jonathanaldrich.bsky.social · 30/06/2026
Bring future ACM Publications related issues to @thzimmermann.bsky.social, who is taking my place.
010
Jonathan Aldrich @jonathanaldrich.bsky.social · 30/06/2026
Today is my last day as SGB Liaison to the ACM Publications Board. It's been a great 6 years working with dedicated volunteers and staff to make ACM Publications better! I'm most proud of helping to transition ACM to 100% Open Access publishing, as of January 1, 2026.
280
Jonathan Aldrich @jonathanaldrich.bsky.social · 28/06/2026
That One Non-Profit Governance Trick: Experts make decisions in a smoke-filled room, keeping information private so they can "roll out" decisions in the right way. Then make the wrong decision because there was no community input and get massive community push-back as a result.
110
Jonathan Aldrich @jonathanaldrich.bsky.social · 26/06/2026
SMH about the wild stories of anti-A/C regulation in Europe. A/C is much more important as a way to mitigate the effects of global warming than it is as a contributor. Plus it literally saves lives.
010
Jonathan Aldrich @jonathanaldrich.bsky.social · 24/06/2026
Silver lining: if he vetos it (which he hasn't threatened to do yet in the coverage I've seen) Congress has the votes to override. If he doesn't it becomes law in 10 days regardless, as long as Congress doesn't go out of session.
010
Jonathan Aldrich @jonathanaldrich.bsky.social · 24/06/2026
Ugh, I saw that. I really don't know why even Republicans tolerate him.
110
Jonathan Aldrich @jonathanaldrich.bsky.social · 24/06/2026
Wow, Congress passes the bipartisan 21st Century ROAD to Housing Act with lots of great pro-housing stuff and literally nothing that's really bad! Trump will reportedly sign it. Amazing that we get this given how dysfunctional Washington has been lately. bipartisanpolicy.org/issue-brief/...
bipartisanpolicy.org
Inside the Deal: What's in the Final 21st Century ROAD to Housing Act
A section-by-section summary of the final 21st Century ROAD to Housing Act, a comprehensive bipartisan package addressing housing supply, manufactured housing, homeownership access, program reform, ve...
110
Jonathan Aldrich @jonathanaldrich.bsky.social · 18/06/2026
The Second Treaty of Versailles. I'm sure it will be as successful as the first one!
010
Jonathan Aldrich @jonathanaldrich.bsky.social · 15/06/2026
My daughter and I watched a famous thriller tonight that's as old as I am! First time I'd seen it, and just about the scariest movie I've even watched.
010
Jonathan Aldrich @jonathanaldrich.bsky.social · 12/06/2026
Yes, and 2022 is the last time I got email from the Onward! SC email list! I do believe things are happening behind the scenes as Onward! continues to operate, but yeah, the web page should be updated.
000
Jonathan Aldrich @jonathanaldrich.bsky.social · 12/06/2026
Interesting, I haven't been invited to an SC meeting for a while. Maybe a mistake? Yes, AI slop is a problem. The old policy didn't really combat it directly, it just required disclosure. Would a policy that reviewers can stop reading after a few pages if the writing is terrible help?
100
Jonathan Aldrich @jonathanaldrich.bsky.social · 12/06/2026
* States that LLM use when conducting the research must be described as part of the ordinary requirement to disclose research methods. www.acm.org/publications... Concerns? If so, I'd like to understand.
acm.org
ACM Policy on Authorship
ACM Publications Policy on Authorship
120
Jonathan Aldrich @jonathanaldrich.bsky.social · 12/06/2026
The ACM Policy on Authorship was recently updated with respect to AI. The policy: * No longer requires disclosing that AI was used in authoring paper text. * Re-emphasizes that authors take full responsibility for the content and writing of the paper, regardless of tools used.
acm.org
ACM Policy on Authorship
ACM Publications Policy on Authorship
111
Jonathan Aldrich @jonathanaldrich.bsky.social · 11/06/2026
I've also asked if there's anything ACM can do about really gross violations by people who are not members. But it may be hard. I observe that these are all students, can you contact their universities?
110