Posts by zyd@emacs.ch
 (DIR) Post #Ad2jbezvUdRavmcdDE by zyd@emacs.ch
       2023-12-21T17:49:35Z
       
       0 likes, 1 repeats
       
       What horrors did she see in the land of lisp? #commonlisp
       
 (DIR) Post #AdWYQ6AugBvSTBRNnE by zyd@emacs.ch
       2024-01-05T03:44:09Z
       
       1 likes, 0 repeats
       
       @crmsnbleyd They know and people have cared, after doing some searching:https://github.com/pdoc3/pdoc/issues/87https://github.com/pdoc3/pdoc/issues/64
       
 (DIR) Post #AdYB2eSfdtynEAW7No by zyd@emacs.ch
       2024-01-05T19:05:44Z
       
       0 likes, 2 repeats
       
       Within the next 5 years we're going to have to apply custom patches to Firefox to remove whatever AI bullshit they're going to include, watch.https://techcrunch.com/2024/01/03/whats-next-for-mozilla/
       
 (DIR) Post #AdYEzKfdAU5wnLFfhw by zyd@emacs.ch
       2024-01-05T23:21:58Z
       
       0 likes, 0 repeats
       
       @adx @louis To be clear, I'm less critical of machine learning techniques when put to socially useful ends like translation or speech to text and the like in cases where the technology is very clear about itself: being a tool that might be helpful but without misleading guarantees. In contrast to that, Mozilla seems to be signalling the inevitability of putting ChatGPT clones where they don't belong, like their browser. Perhaps replacing components of traditional functionality or whatever other terrible feature integration they'll devise. Forced annoyances via statistical deceit machines with customer-support sounding faces is what we'll have to patch out.
       
 (DIR) Post #AdYGVc0IMIcZm2sCuG by zyd@emacs.ch
       2024-01-05T23:59:51Z
       
       0 likes, 0 repeats
       
       @louis @adx Personally I don't think it'd ever be worth it to escape the web, especially to anything whose premise is less features. I don't find minimalism to be a useful goal. I don't know enough about rendering engines to comment in-depth but one technical approach that might help would be modular layers that help writing implementations of rendering engines? Like what SICL is trying to do for Common Lisp implementations:SICL is a new implementation of Common Lisp. It is intentionally divided into many implementation-independent modules that are written in a totally or near-totally portable way, so as to allow other implementations to incorporate these modules from SICL, rather than having to maintain their own, perhaps implementation-specific versions.A technical way to distribute the load of what needs to be implemented. Maybe this type of thing already exists?
       
 (DIR) Post #AdZeMrxMGnS8hZp7dw by zyd@emacs.ch
       2024-01-06T16:01:55Z
       
       0 likes, 0 repeats
       
       @galdor @phoe yeah... could use nginx to work around the bad security (be explicit in what headers you allow) https://github.com/edicl/hunchentoot/issues/24#issuecomment-1256467200
       
 (DIR) Post #AdZtbTZEjF3AiUxp2m by zyd@emacs.ch
       2024-01-06T18:48:19Z
       
       0 likes, 0 repeats
       
       For those that blog and use #rss feeds, is it worth it to split into multiple feeds, a la categories? Asking because while I am a programmer and will write technical content, I also have many other things I want to write about, like reviewing/writing about books and articles which are largely of political content. So the question becomes: if one is to subscribe to a feed of mine, should they get all-or-nothing, or do I provide the option to be selective? On mastodon at least it's all-or-nothing when it comes following someone. And like my account here I don't intend for my site to be sanitized or professionalized in any sort of way.
       
 (DIR) Post #Aicl2wrmdBWSZ5Z40O by zyd@emacs.ch
       2024-06-05T17:05:33Z
       
       0 likes, 1 repeats
       
       just saw an amazing question askedhave any of you used sqlite with common lisp before? is it bussin?well chat, is it bussin or nah
       
 (DIR) Post #AifJJ2kkIYokDF2TBY by zyd@emacs.ch
       2024-06-06T18:59:01Z
       
       0 likes, 0 repeats
       
       Went from taking 8 hours to import 50 million rows to 45 minutes. Wild. And instead of constantly hitting my disk at 300MB/s the entire time its now ~16MB/s with spikes here and there. Lesson learned: INSERT as many rows as you can rather than a single row insert at a time. Very happy more experienced folks told me I was doing it the slow way, they weren't kidding!If I parallelize this I bet I can knock it down to 10 minutes or so.
       
 (DIR) Post #AifJJ4negiKiYqPozw by zyd@emacs.ch
       2024-06-06T19:01:25Z
       
       0 likes, 0 repeats
       
       Also: removing foreign keys is worth it (given you can trust what you're doing/the data for the most part)
       
 (DIR) Post #AizhfYHeaGlq0D2kbI by zyd@emacs.ch
       2024-06-15T23:24:19Z
       
       1 likes, 0 repeats
       
       Common Lisp Community Survey: https://cl-survey-2024.djhaskin.comHello. My name is Daniel Haskin. I'm conducting a survey of the Common Lisp Community in 2024. I will run the survey for six weeks, starting on Saturday, June 15, 2024, and ending on Saturday, July 27, 2024. I will publish the results on the first Saturday in August 2024.#lisp #commonlisp
       
 (DIR) Post #AjhGtJg6S1giLGVBMO by zyd@emacs.ch
       2024-07-07T18:16:25Z
       
       0 likes, 0 repeats
       
       @oantolin @jesus @simon_brooke I tried several different optimizations my friend also thought of but didn't implement because it would have been another 1,000 lines or so. That is so sad. A programming language that discourages further experimentation. Besides the line count I bet it's a world of difference being able to write functions and immediately be able to experiment at the REPL (i.e incrementally). There's a barrier to that kind of flow in any language where it has a distinct compile and run process. Lisps (in the style of CL) still do have compile-time and run-time but they're less distinct since both happen inside the running image. Such a huge advantage in my opinion.
       
 (DIR) Post #AjhGtLWxYzYw5UEuO0 by zyd@emacs.ch
       2024-07-07T18:29:03Z
       
       1 likes, 1 repeats
       
       @falken @oantolin @jesus @simon_brooke You can attach a "REPL" to pretty much any language (and all languages are better for it), but there's a substantial difference between a language built around that mode of interaction. Being able to have deep introspection of all objects, compilation failures (and other conditions) being raised inside the REPL and being able to do various things with them (interactively and programmatically) etc. Macroexpansion. Disassembling. Packages. So much is packed into a CL REPL. We're hardly ever typing function definitions into it so I rather think of it as your workbench.
       
 (DIR) Post #AkULkcJmnWVyh01BmS by zyd@emacs.ch
       2024-07-31T06:43:08Z
       
       0 likes, 0 repeats
       
       I got a mod warning for stating I want fascists dead lol. Liberal tolerance sure is amusing.
       
 (DIR) Post #AkULkeVudPP3VBhtxY by zyd@emacs.ch
       2024-07-31T06:44:20Z
       
       0 likes, 0 repeats
       
       Wouldn't want fascists to feel unwelcome or insulted after all!