Thicket data repository for the EEG
anil.recoil.org
1{
2 "id": "https://ryan.freumh.org/2024-12-16.html",
3 "title": "16 Dec 2024",
4 "link": "https://ryan.freumh.org/2024-12-16.html",
5 "updated": "2024-12-16T00:00:00",
6 "published": "2024-12-16T00:00:00",
7 "summary": "<div>\n <span> Previous: <a href=\"2024-12-02.html\"> 2 Dec 2024</a> </span>\n <span> Next: <a href=\"2024-12-30.html\">30 Dec 2024</a> </span>\n </div>\n \n \n\n <ol>\n<li><p><span>babel</span></p>\n<ol>\n<li><p><span><span><em>opam-repo maps</em></span></span></p></li>\n<li><p><span><span>O</span> hyperres needs\na file format</span></p>\n<ol>\n<li><p><span><span>O</span> get a JSON\nformat, including version formula</span></p>\n<ol>\n<li><p><span>opam formulas</span></p></li>\n<li><p><span>variables as packages</span></p></li>\n<li><p><span>boolean algebra represeted with disjuntions\netc</span></p></li>\n</ol></li>\n<li><p><span>I’m wrestling with the problem of expressivity\nof package managers.</span></p>\n<p><span>What is representable in a SAT problem is\nconsistent across package managers, but there are varying degrees of\npre-processing in DSL in different ecosystems. For example, Debian has a\npretty simple language for expressing dependency constraints, but Nix\nhas the Nix DSL, opam has filters and variables. These are basically a\nway of expressing the constraint language with various degrees of\npre-processing. It’s fine to unify the decision problem representations,\nbut how can we unify the different package languages? Maybe we don’t\nneed to.</span></p></li>\n</ol></li>\n<li><p><span><span>O</span> hyperres needs\na solver</span></p>\n<ol>\n<li><p><span>zero install</span></p>\n<p><span>we have a working zero install derived SAT-based\nsolver</span></p></li>\n<li><p><span>PubGrub</span></p>\n<p><span>we see if we can get PubGrub working with opam /\nthe babel file format</span></p></li>\n</ol></li>\n<li><p><span><span>O</span> specific\neval</span></p>\n<ol>\n<li><p><span>docker hub top 10 mixed language\ndave</span></p></li>\n<li><p><span>cross ecosystem package solving</span></p>\n<ol>\n<li><p><span>show problem, find a few illustative\nprojects</span></p></li>\n<li><p><span>next gen mechanism for encoding things\nbig</span></p></li>\n<li><p><span>download and visualise, see what’s going\non</span></p></li>\n<li><p><span>babel: how solving? what is the\nreperestation?</span></p></li>\n</ol></li>\n</ol></li>\n<li><p><span>list of things, organise by priority, define\nevals, spread tasks</span></p></li>\n<li><p><span><span>X</span> SPLASH round\n2 submission</span></p></li>\n<li><p><span><span>X</span> ICFP\ndeadline</span></p></li>\n</ol></li>\n<li><p><span>reading papers</span></p></li>\n<li><p><span>add ideas page to website</span></p></li>\n</ol>",
8 "content": "<div>\n <span> Previous: <a href=\"2024-12-02.html\"> 2 Dec 2024</a> </span>\n <span> Next: <a href=\"2024-12-30.html\">30 Dec 2024</a> </span>\n </div>\n \n \n\n <ol>\n<li><p><span>babel</span></p>\n<ol>\n<li><p><span><span><em>opam-repo maps</em></span></span></p></li>\n<li><p><span><span>O</span> hyperres needs\na file format</span></p>\n<ol>\n<li><p><span><span>O</span> get a JSON\nformat, including version formula</span></p>\n<ol>\n<li><p><span>opam formulas</span></p></li>\n<li><p><span>variables as packages</span></p></li>\n<li><p><span>boolean algebra represeted with disjuntions\netc</span></p></li>\n</ol></li>\n<li><p><span>I’m wrestling with the problem of expressivity\nof package managers.</span></p>\n<p><span>What is representable in a SAT problem is\nconsistent across package managers, but there are varying degrees of\npre-processing in DSL in different ecosystems. For example, Debian has a\npretty simple language for expressing dependency constraints, but Nix\nhas the Nix DSL, opam has filters and variables. These are basically a\nway of expressing the constraint language with various degrees of\npre-processing. It’s fine to unify the decision problem representations,\nbut how can we unify the different package languages? Maybe we don’t\nneed to.</span></p></li>\n</ol></li>\n<li><p><span><span>O</span> hyperres needs\na solver</span></p>\n<ol>\n<li><p><span>zero install</span></p>\n<p><span>we have a working zero install derived SAT-based\nsolver</span></p></li>\n<li><p><span>PubGrub</span></p>\n<p><span>we see if we can get PubGrub working with opam /\nthe babel file format</span></p></li>\n</ol></li>\n<li><p><span><span>O</span> specific\neval</span></p>\n<ol>\n<li><p><span>docker hub top 10 mixed language\ndave</span></p></li>\n<li><p><span>cross ecosystem package solving</span></p>\n<ol>\n<li><p><span>show problem, find a few illustative\nprojects</span></p></li>\n<li><p><span>next gen mechanism for encoding things\nbig</span></p></li>\n<li><p><span>download and visualise, see what’s going\non</span></p></li>\n<li><p><span>babel: how solving? what is the\nreperestation?</span></p></li>\n</ol></li>\n</ol></li>\n<li><p><span>list of things, organise by priority, define\nevals, spread tasks</span></p></li>\n<li><p><span><span>X</span> SPLASH round\n2 submission</span></p></li>\n<li><p><span><span>X</span> ICFP\ndeadline</span></p></li>\n</ol></li>\n<li><p><span>reading papers</span></p></li>\n<li><p><span>add ideas page to website</span></p></li>\n</ol>",
9 "content_type": "html",
10 "categories": [],
11 "source": "https://ryan.freumh.org/atom.xml"
12}