Notices by Stuart Langridge (sil@mastodon.social)

this evening's fun discovery: https://developer.mozilla.org/en-US/docs/Web/API/TreeWalker
A very convenient way to walk a DOM tree from JS, built in to the platform. I had no idea this existed. Nice one MDN for making it clear.

@jwz The Mother Superior murmured: “Is he then an unhappy man?”
Poirot said: “So unhappy that he has forgotten what happiness means. So unhappy that he does not know he is unhappy.”
The nun said softly: “Ah, a rich man. . . .”
Hercule Poirot said nothing—for he knew there was nothing to say.

The thing here is that "these decisions that are making MacOS increasingly annoying" do so because they're violating an important but unstated assumption: that your computer is yours. The third possibility for why this is happening, which isn't mentioned in the piece, is that OS makers are increasingly leaning in the direction that your computer (pocket or desk) isn't yours, it's theirs. What they think should be doable with it is more important than what you think.
https://social.vivaldi.net/@brucelawson/113316262017803991

@SolusSpider @tuxjam @kevie @mcnalu @thelovebug @mralc clearly @popey is the correct choice :-)

"we push the web back towards the wonderful, beautiful, joyful place it ought to be" - @molly0xfff, https://www.citationneeded.news/fighting-for-our-web/
"A rewilded internet will be a more interesting, usable, stable and enjoyable place to be" - @mariafarrell, @robin, https://www.noemamag.com/we-need-to-rewild-the-internet/
"this web we made together is the biggest and best collection of information and exhibitions and joy that the world has ever known" - me, https://www.kryogenix.org/code/browser/the-mazy-web/

A recurring theme of the web we want: joy. Bring back the joy.

This weekend's project: get a data dump from a Fandom wiki (all of them have this: check out the page Special:Statistics) and turn it into an offline version (a .zim file, readable with software such as Kiwix, on many platforms) via a (surprisingly short) Python script. This is a very useful thing to have -- a complete offline copy of something like the Forgotten Realms wiki for reference when the internet is slow! #DnD

@anildash @robinwhittleton if the answer is "yeah, that's not really what Fastly is for, don't publish your whole big static website in compute@edge, and we don't do static hosting, so publish it somewhere else and then use our CDN to front it, our CDN's great" then I'm fine with that answer; what I don't know is whether that really is the answer or if there is a different or better way that I don't know about. I know nobody likes to say they're not best at a thing, though :-)

@anildash @robinwhittleton Thank you! My basic question is: can I use Fastly to publish a static website, the same way I might use, say, surge? That is, not publish it somewhere else and put Fastly's CDN in front of it; do it with just a Fastly account. As far as I can tell, deploying it as a big wasm binary to compute@edge is the suggested way to do that, but... at least the local server fails on moderately-sized static sites (and I fear publishing would charge for memory use and be expensive!)

OK, confirmation at least that all this wasm stuff is invisible server hosting; it does not appear to actual users. So instead of copying your files to a web server, you instead use their tools to compile an application which embeds all your files, and then deploy that compiled binary application on Fastly's infrastructure. That seems complex, but it maybe works (at least, on trivially small static sites). This makes problem diagnosis sound hard to me, but maybe it's OK, if complicated.

It seems to compile a big WASM blob? Unsure about this. In particular, if it serves out the HTML that I actually wrote, then.... OK, I suppose? This seems a very complicated way to go about things, but whatever, it's invisible, so I'm testing it. If it serves all that wasm to the actual users then it's going straight in the bin. But... currently, the server doesn't seem to be able to even serve this wasm monstrosity it creates, so maybe it only works for a one page website. All seems v complex!

Looking at Fastly for hosting static websites, which https://www.fastly.com/blog/no-origin-static-websites-at-the-edge/ claims it can do with "compute@edge", and I have fallen at the first hurdle. I dropped a reasonable-sized static site into a public folder and ran "fastly compute serve" and... "memory index 0 has a minimum page size of 3075 which exceeds the limit of 2048". This seems to be some sort of hardcoded limit in "viceroy", the server; someone talks in a bug about recompiling it. Do I know any fastly peeps? @devs maybe?

@robinwhittleton ah. But the CDN requires me to actually host the site somewhere else. https://www.fastly.com/blog/no-origin-static-websites-at-the-edge/ suggests that if I want to host it at fastly, this compute thing is the way, no? Did I misunderstand? Certainly avoiding all this wasm compilation stuff feels sensible

Is it worth spending lots of money on a really good computer mouse? Or is that just mouse snobbery by people who want to be able to configure button 9 to shuffle their windows into alphabetical order and the like?

@powersoffour Blades is great and I like it a lot, but waaaay more in-depth than I'm looking at here for this particular project. By "rules-light" here I mean stuff like Risus or Freeform, where character generation is "pick three cool sentences about your character, and we're done", and dice rolls are more like "roll 1d6 and 4-6 is a success" :) Never heard of Kids on Bikes; I'll check that out!

I'm researching rules-light #ttrpg games for novice players. I want something with some vague of crunch -- that is, there should be *some* mechanics, not a pure storytelling game -- but not much crunch at all. Also, not rules intended for comedy wacky games, or rulesets which cost money to look at (I'm only looking for inspiration here, so I'm not buying a game to read once). On the list already are Risus, Freeform Universal, Fate Accelerated, Lasers & Feelings, Fireside; suggestions welcomed!

"Banning TikTok won’t protect Americans from targeted misinformation or misuse of their personal data, which American data brokers routinely sell and share." Well said, Representative Sara Jacobs (D-CA) (https://en.wikipedia.org/wiki/Sara_Jacobs?useskin=vector), as quoted by @mmasnick in https://www.techdirt.com/2024/03/14/once-more-with-feeling-banning-tiktok-is-unconstitutional-wont-do-shit-to-deal-with-any-actual-threats/.

Outside and I can hear Rihanna’s “Umbrella” playing, which obviously made me think of the excellent Tom Holland lip sync video. Does this also trigger the rule about resharing it? Or do I have to have actually seen the video?

Actually, what the hell, let’s trigger another round of sharing, because it’s worth it. https://youtu.be/Brq-exSvB7Q?si=670qju7XRQZ6Gzcz

@mjg59 I think there's, as you say, an implicit assumption that "source code" is low-level enough that you can understand how the program does what it does. What your theoretical model does is undermine that idea, because the source code expresses a wish rather than an algorithm; there's no explanation of how it's accomplished, which historically people have always assumed that source code has to do. The model is more like a genie that brings you what you wish for; you're dependent on it.

@mjg59 this is a question of philosophy, I think. Is "be my ideal word processor" adequate source code? If you've got a compiler that can compile that correctly, yes. If you haven't, no. Is the Python source for that word processor adequate source code? I think yes even if you don't have a Python interpreter because you can see how it works, not just that it works. This feels like I'm talking myself into a different position than the one I intuitively thought.

@mjg59 agreed entirely! and I'm in favour of that. But that's because it didn't really need to be; all previous compilable languages had the property that even if you didn't have the compiler, you could work out how the compiler *worked*, and therefore in a pinch could write your own. (A lot of work, but doable.) But with this, you don't know how to build the compiler, and the source code is not enough by itself to build the program.