LightNews
mcoblenz.bsky.social
Michael Coblenz
@mcoblenz.bsky.social
260 followers 220 following 37 posts
Assistant Prof. at UC San Diego. Helping people write better software.
Posts Media Videos Starter Packs
mcoblenz.bsky.social
Michael Coblenz @mcoblenz.bsky.social · Sep 3
Will FL health insurance companies raise their rates? www.nytimes.com/2025/09/03/u...
Florida Says It Plans to End All Vaccine Mandates
www.nytimes.com
mcoblenz.bsky.social
Michael Coblenz @mcoblenz.bsky.social · Sep 3
We rarely give Shakespeare credit for inventing alpha-conversion. “A rose by any other name…”
1
mcoblenz.bsky.social
Michael Coblenz @mcoblenz.bsky.social · Aug 4
Sounds like you already have two trained classifiers!
1
mcoblenz.bsky.social
Michael Coblenz @mcoblenz.bsky.social · Jul 9
But maybe also I pick a different Mastodon server.
mcoblenz.bsky.social
Michael Coblenz @mcoblenz.bsky.social · Jul 9
Mastodon is too fragmented. I just focus on Bsky now.
1
mcoblenz.bsky.social
Michael Coblenz @mcoblenz.bsky.social · Jul 2
We'd really appreciate help with our study of Rust programming! If you know Rust, please take about 20-30 minutes to help us find out how to make Rust programmers more effective: ucsd.co1.qualtrics.com/jfe/form/SV_...
Rust Learning Quiz
Help us learn more about Rust by taking this short and easy quiz!
ucsd.co1.qualtrics.com
3 9
mcoblenz.bsky.social
Michael Coblenz @mcoblenz.bsky.social · Jun 29
I can understand that. But I stopped checking Twitter long ago, so for me, this is it.
1
mcoblenz.bsky.social
Michael Coblenz @mcoblenz.bsky.social · Jun 26
Reminder about the upcoming HATRA deadline, if you're interested in making fancy type systems and reasoning assistants more effective for people! conf.researchr.org/home/icfp-sp...
HATRA 2025 - ICFP/SPLASH 2025
Programming language designers seek to provide strong tools to help developers reason about their programs. For example, the formal methods community seeks to enable developers to prove correctness pr...
conf.researchr.org
1 6
mcoblenz.bsky.social
Michael Coblenz @mcoblenz.bsky.social · May 1
Preprint of the paper, which my student Savitha Ravi will present: arxiv.org/pdf/2503.12309
arxiv.org
mcoblenz.bsky.social
Michael Coblenz @mcoblenz.bsky.social · May 1
At #ICSE2025 and want to know why scientists use Jupyter despite its limitations (e.g., out of order execution, lack of reproducibility)? Come to our talk at 11:45 today! conf.researchr.org/details/icse...
How Scientists Use Jupyter Notebooks: Goals, Quality Attributes, and Opportunities (ICSE 2025 - Research Track) - ICSE 2025
ICSE, the IEEE/ACM International Conference on Software Engineering, is the premier software engineering conference. It will be held April 27-May 3 2025 in Ottawa. Core conference days will be Wednesd...
conf.researchr.org
1 2
mcoblenz.bsky.social
Michael Coblenz @mcoblenz.bsky.social · Apr 11
Lots of great possibilities! Probably worth doing a study to see how various approaches would affect behavior.
1
mcoblenz.bsky.social
Michael Coblenz @mcoblenz.bsky.social · Apr 11
What will you do to make the environmental costs of using this transparent?
1 1
mcoblenz.bsky.social
Michael Coblenz @mcoblenz.bsky.social · Apr 4
For the record, ChatGPT is happy to take over the job of an air traffic controller. Looks like Musk et al. can save a lot of money on actual ATC staff!
mcoblenz.bsky.social
Michael Coblenz @mcoblenz.bsky.social · Mar 22
Test passed! Awesome!
1
mcoblenz.bsky.social
Michael Coblenz @mcoblenz.bsky.social · Mar 22
Geaxhurccswzvy4vi2ftz;
1
mcoblenz.bsky.social
Michael Coblenz @mcoblenz.bsky.social · Mar 12
You are right that the only way to be sure is to re-run the experiment.
1
mcoblenz.bsky.social
Michael Coblenz @mcoblenz.bsky.social · Mar 12
The API delta between the fixed version and the one we used in the experiment is rather trivial, so we felt the results likely generalized. But of course re-running the experiment would be best.
1 1
mcoblenz.bsky.social
Michael Coblenz @mcoblenz.bsky.social · Mar 12
Also as I understand it, the stack maps approach cannot be made thread safe, which is why I haven't felt motivated to finish this implementation; it still would not be of practical use. Really the whole GC implementation needs a redo.
1 2
mcoblenz.bsky.social
Michael Coblenz @mcoblenz.bsky.social · Mar 12
On the other hand, the main branch's version has GC enabled. The problem with this is that the mark implementation only handles certain cases, so this is a much more serious unsoundness problem (which is also documented in the paper and the code).
1 2
mcoblenz.bsky.social
Michael Coblenz @mcoblenz.bsky.social · Mar 12
I think this is the relevant commit. Note it’s on a branch: I never got around to merging. Criticism is welcome. github.com/mcoblenz/Bro...
Made as_ref and as_mut private. Removed Deref from GcRef. · mcoblenz/Bronze@34c1538
github.com
1 2
mcoblenz.bsky.social
Michael Coblenz @mcoblenz.bsky.social · Mar 12
Sorry I should have read the whole thread before replying! Still getting used to the Bluesky UI.
1 1
mcoblenz.bsky.social
Michael Coblenz @mcoblenz.bsky.social · Mar 12
The implementation would definitely need a from-scratch redo for real work.
2
mcoblenz.bsky.social
Michael Coblenz @mcoblenz.bsky.social · Mar 12
Can you be more specific? We did fix some soundness problems post-experiment. And the implementation is a partial proof of concept and not a working GC.
2 1
mcoblenz.bsky.social
Michael Coblenz @mcoblenz.bsky.social · Feb 26
They’re pretty similar, right? Both are about avoiding side effects.
3
mcoblenz.bsky.social
Michael Coblenz @mcoblenz.bsky.social · Feb 24
Hey,
@icseconf.bsky.social: when do you expect to post the presentation schedule for the technical track?
4