Joe Gibbs Politz
@joepolitz.bsky.social
I think it's possible to get there, someone is going to have to mind-meld with the runtime and bun to do it.
I have another experiment going to see if I can get github.com/Automattic/n... to be bundled into the binary (which is how the image library works).
Currently that's ending somewhat sadly.
I have another experiment going to see if I can get github.com/Automattic/n... to be bundled into the binary (which is how the image library works).
Currently that's ending somewhat sadly.
September 5, 2025 at 3:54 AM
I think it's possible to get there, someone is going to have to mind-meld with the runtime and bun to do it.
I have another experiment going to see if I can get github.com/Automattic/n... to be bundled into the binary (which is how the image library works).
Currently that's ending somewhat sadly.
I have another experiment going to see if I can get github.com/Automattic/n... to be bundled into the binary (which is how the image library works).
Currently that's ending somewhat sadly.
And just to confirm, a simple js program that just logs does produce a binary with sensible behavior using the same command, so I don't think I'm goofing on the basic thing to try :-)
September 4, 2025 at 4:57 PM
And just to confirm, a simple js program that just logs does produce a binary with sensible behavior using the same command, so I don't think I'm goofing on the basic thing to try :-)
At least on my system, the doing the most straightforward thing has no error messages but creates a binary that doesn't produce any output at the terminal (running the compiler blob with `node` and no arguments should print the usage).
September 4, 2025 at 4:53 PM
At least on my system, the doing the most straightforward thing has no error messages but creates a binary that doesn't produce any output at the terminal (running the compiler blob with `node` and no arguments should print the usage).
Huh. OK yeah so the issue is:
- we symlink node_modules in .pyret to pyret-npm/node_modules with `const nodeModulesPath = path.join(__dirname, "node_modules");`
- but on `yarn global add` it flattens out the dependencies of `pyret-npm` into the node_modules above this
This is fixable I think.
- we symlink node_modules in .pyret to pyret-npm/node_modules with `const nodeModulesPath = path.join(__dirname, "node_modules");`
- but on `yarn global add` it flattens out the dependencies of `pyret-npm` into the node_modules above this
This is fixable I think.
September 3, 2025 at 8:00 PM
Huh. OK yeah so the issue is:
- we symlink node_modules in .pyret to pyret-npm/node_modules with `const nodeModulesPath = path.join(__dirname, "node_modules");`
- but on `yarn global add` it flattens out the dependencies of `pyret-npm` into the node_modules above this
This is fixable I think.
- we symlink node_modules in .pyret to pyret-npm/node_modules with `const nodeModulesPath = path.join(__dirname, "node_modules");`
- but on `yarn global add` it flattens out the dependencies of `pyret-npm` into the node_modules above this
This is fixable I think.
Working on two versions of my spring course offering.
August 14, 2025 at 3:22 PM
Working on two versions of my spring course offering.
Probably dumb/ignorant language questions:
Why is the gap between Rho and Sigma, and not between Pi and Rho? I could see wanting to keep some latin <-> greek-ish correspondence, but Rho is more like R than Q, right?
Why is the gap between Rho and Sigma, and not between Pi and Rho? I could see wanting to keep some latin <-> greek-ish correspondence, but Rho is more like R than Q, right?
July 8, 2025 at 5:15 PM
Probably dumb/ignorant language questions:
Why is the gap between Rho and Sigma, and not between Pi and Rho? I could see wanting to keep some latin <-> greek-ish correspondence, but Rho is more like R than Q, right?
Why is the gap between Rho and Sigma, and not between Pi and Rho? I could see wanting to keep some latin <-> greek-ish correspondence, but Rho is more like R than Q, right?
Kwaay Paay Peak Trail
June 29, 2025 at 9:49 PM
Kwaay Paay Peak Trail
Teaching large classes running up against real engineering limits!
(I don't really need to be “watching” all student repos from 2017. But I usually do “watch” the few hundred for an active class!)
(I don't really need to be “watching” all student repos from 2017. But I usually do “watch” the few hundred for an active class!)
June 10, 2025 at 5:02 PM
Teaching large classes running up against real engineering limits!
(I don't really need to be “watching” all student repos from 2017. But I usually do “watch” the few hundred for an active class!)
(I don't really need to be “watching” all student repos from 2017. But I usually do “watch” the few hundred for an active class!)
👀 Not sure if Copilot has learned from me, my workspace, or all developers that comments are likely to begin with “NOTE(joe): this is a bit of hack...”
I'll just assume this isn't me-specific and carry on.
I'll just assume this isn't me-specific and carry on.
May 16, 2025 at 8:45 PM
👀 Not sure if Copilot has learned from me, my workspace, or all developers that comments are likely to begin with “NOTE(joe): this is a bit of hack...”
I'll just assume this isn't me-specific and carry on.
I'll just assume this isn't me-specific and carry on.
As far as I can tell, --umdGlobalName option has been entirely hallucinated by Google's search results here. I can't find any reference to it in any official issue or documentation, and the discussion of the relevant issue doesn't mention it:
github.com/microsoft/Ty...
github.com/microsoft/Ty...
April 22, 2025 at 10:46 PM
As far as I can tell, --umdGlobalName option has been entirely hallucinated by Google's search results here. I can't find any reference to it in any official issue or documentation, and the discussion of the relevant issue doesn't mention it:
github.com/microsoft/Ty...
github.com/microsoft/Ty...
Marian Bear Memorial Park
March 16, 2025 at 9:01 PM
Marian Bear Memorial Park
I haven't published in math journals, but I had a phase in graduate school where I tried to get \skull into all my papers in semantically meaningful ways. These are published in real computing journals/conference proceedings.
January 15, 2025 at 6:25 PM
I haven't published in math journals, but I had a phase in graduate school where I tried to get \skull into all my papers in semantically meaningful ways. These are published in real computing journals/conference proceedings.
I really liked the lecture so I will indulge in a screenshot
January 8, 2025 at 4:53 PM
I really liked the lecture so I will indulge in a screenshot
I think they could be compiled more efficiently to never need to generate a function body/call.
More analysis can avoid the overhead, but it’s more analysis.
adam.chlipala.net/mlcomp/#:~:t....
More analysis can avoid the overhead, but it’s more analysis.
adam.chlipala.net/mlcomp/#:~:t....
January 4, 2025 at 1:56 AM
I think they could be compiled more efficiently to never need to generate a function body/call.
More analysis can avoid the overhead, but it’s more analysis.
adam.chlipala.net/mlcomp/#:~:t....
More analysis can avoid the overhead, but it’s more analysis.
adam.chlipala.net/mlcomp/#:~:t....
Anyone with an 18mo who hasn’t experienced Christmas yet want this.
December 11, 2024 at 2:46 PM
Anyone with an 18mo who hasn’t experienced Christmas yet want this.
Got some family photos taken, the important shots came out OK.
November 28, 2024 at 3:48 PM
Got some family photos taken, the important shots came out OK.
Just to be super concrete, here's the day 1 handout for the most recent offering.
The concepts are parsing (to an AST), (generating) assembly, linking, and running.
The corresponding assignment is ucsd-compilers-s23.github.io/week1/index....
The concepts are parsing (to an AST), (generating) assembly, linking, and running.
The corresponding assignment is ucsd-compilers-s23.github.io/week1/index....
November 25, 2024 at 4:55 PM
Just to be super concrete, here's the day 1 handout for the most recent offering.
The concepts are parsing (to an AST), (generating) assembly, linking, and running.
The corresponding assignment is ucsd-compilers-s23.github.io/week1/index....
The concepts are parsing (to an AST), (generating) assembly, linking, and running.
The corresponding assignment is ucsd-compilers-s23.github.io/week1/index....
Only one of these arrays is sworded.
October 31, 2024 at 7:44 PM
Only one of these arrays is sworded.
I did a perfect datatype design – the result:
March 9, 2024 at 1:04 AM
I did a perfect datatype design – the result:
I may or may not have described this as “like some quantum shit” in lecture yesterday.
February 6, 2024 at 3:52 PM
I may or may not have described this as “like some quantum shit” in lecture yesterday.
This lecture, which I did very little prep for after having the idea a little late, ended up being great. “How many files are there” and “how many lines of code are there” and “which languages/file extensions are in use” are all awesome, first-pass questions that students jumped on immediately.
![Lecture handout.
These are a few commonly-used commands. There are lots more! We could fill pages with them
(examples of find, wc, grep, output redirection)
Bluesky, also known as Bluesky Social, is a microblogging social platform and a public benefit corporation. The service is focused on microblogging, and has been called "Twitter-like". Bluesky Social was made open source under the MIT license in May 2023.[27]
This is a big codebase! Hundreds of files.
What questions could we ask that we can answer with these commands?](https://cdn.bsky.app/img/feed_thumbnail/plain/did:plc:j6gtfm6knf2k7vxpufyf27uc/bafkreibisjl7klxox52nbn3jdpqtjxhdmwtkoslcxe6ljhboysppvs3ipy@jpeg)
February 5, 2024 at 10:09 PM
This lecture, which I did very little prep for after having the idea a little late, ended up being great. “How many files are there” and “how many lines of code are there” and “which languages/file extensions are in use” are all awesome, first-pass questions that students jumped on immediately.
Oh and this was the prompt at the top of the page. (John Regehr is cited on the front as the source for the definitions.)
January 22, 2024 at 6:20 PM
Oh and this was the prompt at the top of the page. (John Regehr is cited on the front as the source for the definitions.)
I lectured about asking good questions today (in the spirit of debugging) and I am far too proud of my made up student names.
January 22, 2024 at 6:19 PM
I lectured about asking good questions today (in the spirit of debugging) and I am far too proud of my made up student names.
Sanderson's first law gives an account of the spectrum. It's not that magic *must* be explained. It suggests that magic as a solution to characters' problems is satisfying in proportion to how well the reader understands how it works.
www.brandonsanderson.com/sandersons-f....
www.brandonsanderson.com/sandersons-f....
January 12, 2024 at 5:08 PM
Sanderson's first law gives an account of the spectrum. It's not that magic *must* be explained. It suggests that magic as a solution to characters' problems is satisfying in proportion to how well the reader understands how it works.
www.brandonsanderson.com/sandersons-f....
www.brandonsanderson.com/sandersons-f....