Monday @ 9:18 am
Feb. 2nd, 2026 09:18 am![[personal profile]](https://www.dreamwidth.org/img/silk/identity/user.png){kind=small}
alisxEeeey the Aurealis Award shortlist is out. Congrats to all the nominations, ad commiserations to everyone whose TBR piles just got bigger.
I’m not sure it was ever formally announced (?), but I also don’t think it’s a secret, so: I was on the panel for horror short stories and we got some fucking bangers this year for sure. Everyone on the shortlist absolutely deserves to be there, in a difficult field with some really brutal (in many senses) competition.
Monday @ 6:27 am
Feb. 2nd, 2026 06:27 amalisxI can report No Other Choice is, in fact, quite good.
Also, I understood one (1) whole word of Korean in it! Learning!!!
The end of America.
Feb. 2nd, 2026 05:09 amalisxTrump might deploy 1500 Alaskan paratroopers to Minnesota. Trump might blow up NATO. Trump might blow up NATO, as in, start World War III.
Chuck Schumer’s still gonna confirm his judges, though.
More and more I am convinced that our current political system doesn’t just enable, foster, or perpetuate tolerance for authoritarianism in law enforcement, but requires it.
Elena on .
… oh no.
Jan. 31st, 2026 06:54 amalisxI had a horrifying realization yesterday, when I was trying to explain to a bunch of law students who are almost all more than 40 years younger than I am, that what’s happening in America is not in any way normal. For example, Donald Trump’s weaponization of the DOJ to abuse the criminal legal process to extort a federal agency into submitting to his demands is a very reminiscent of the Saturday Night Massacre in 1973 (naturally I explained what that was), which was a cataclysmic political scandal at the time. In America 2026, it was Monday.
The realization was that, to these bright, curious, and — relatively speaking — politically engaged young people, all this is normal. They were almost all in middle school/junior high when Donald Trump got elected the first time. Trumpism and the to this point largely ineffective resistance to it has simply always been the way of the political and cultural world in America, since they began to become conscious of it in a concrete way.
Paul Campos on the .
Finished reading: Rokea
Jan. 30th, 2026 12:52 pmalisxI never really “got” the Rokea as a kid — they were big and ugly and weird — but as an adult the whole premise of “what if Jaws had autism and was immortal?” is suddenly incredibly appealing. I too would like to walk into the sea and never return, thanks.
This is vintage old World of Darkness, so expect a bunch of weird race and sex stuff (including some way-too-casual references to rape). The example chronicle is also set in Australia and is nonetheless extremely, hilariously (white) American, right down to the fact that, while the majority of it is set on the Great Barrier Reef near Mackay (erroneously written as “MacKay”; it’s pronounced mh’KYE), like two scenes happen over two thousand kilometres away in South Australia??? And that’s two thousand kilometres by land, when the player characters are sharks and have to swim the long way??? That’s like the distance between New York and Houston, and would take like two months for a shark to swim there and back. Again, assuming they could swim on land, which they can’t. And this is all supposed to be time sensitive. Also, one of the main NPCs is a Native American Garou living as a visa over-stayer ( . . . oof) in Australia from the tribe-now-called-the-Galestalkers-you-can-figure-out-what-they-used-to-be-called whose ancestors are nonetheless implied in a vision to be Indigenous Australians? Not to mention I absolutely cannot believe that, even in the GRIMDARK 2000s of the WORLD OF DARKNESS that random Australian fishermen would conduct a mass illegal shark hunt all up and down the Great Barrier Reef. It’s a freakin’ national park!!! Killing wildlife here is illegal!!! Also like four people got attacked by sharks near Sydney last week alone and everyone was like, “Well . . . better not go swimming for a bit hey.”
Plus: Not even one single Harold Holt joke. Unforgivable. (Also, may the sea return him.)
Like . . . I know all this stuff can be hand-waved but it’s just such a hilarious mess on the surface for what is otherwise a pretty fun one-shot and it’s like . . . bless the OWoD. For reals. Like. I’m glad they’ve cleaned a lot of this stuff up in 5e (or even in W20) but, also. There’s such a well-meaning-but-oh-gods-dude-no-ness to the vintage stuff that makes it . . . kinda endearing, in retrospect? Big, wall-banging “let’s all work together to save the planet!!!” messaging and all.
Because, y’know. We should.
Finished reading: Malleus Monstrorum Slipcase Set
Jan. 28th, 2026 08:23 pmalisxAs someone who loves both the Mythos and a bestiary, these books unfortunately kinda show the limitations when both are combined. Mythos entities encountered alone are horrific and gross and cool. Combined they run together in a kind of protean slurry of gore and tentacles and “unpronounceable” names, and the seams where multiple authors over the century have stitched in their own additions result in something that is somehow both too samey and too thematically incoherent.
Also, just . . . times have kinda moved on? Hence the quote marks around “unpronounceable,” since combinations of syllables that seemed “foreign” and “strange” in 1926 are significantly less so in an age where your average English speaker has encountered half a dozen loanwords before breakfast. (Or possibly at breakfast; the fact that “Shak’shu-ka” could be a Mythos god being kinda illustrative of the point.) Not to mention some of the monster designs feel a bit uninspired in a world of Bloodborne and Silent Hill, though Loïc Muzy’s illustrations punch above their weight as usual.
All that being said, none of that is the fault of these books specifically, which are a great resource for what they are, with all the relevant crunch and fluff and cool illustrations. Also, props to the writers for updating some of the more, y’know. Cat-named (if you get my drift) elements, particularly the distancing of Mythos entities from any one ethnic group or set of religious/spiritual beliefs (Ithaqua, looking at you, buddy).
Anyway. I don’t regret buying these, and they did sate my lust for tomes of tentacled bad bois. And I guess that’s the important part.
ImageMagick for drawing
Jan. 28th, 2026 03:47 pmmiriam_eIf anybody is interested here are the ImageMagick commands I used for my most recent l-systems experiments.
To create a blank, white canvas:
To draw a black line:
To create a blank, white canvas:
magick -size 512x512 xc:white canvas.pngmagick -size 512x512 xc:white canvas.png
To draw a black line:
magick canvas.png -stroke "$col" -strokewidth $w -draw "line $xp,$yp $xnp,$ynp" -quality 0 canvas.png
canvas.png is the input image to be written on.-stroke sets the color of the line (I used the variable $col, which I set to "green")-strokewidth sets the thickness of the line (in this case $w, set to 1)-draw and the command "line" sets the start and endpoints of the line.-quality was ChatGPT's excellent suggestion to force an uncompressed png image. It suggested a different, incorrect command, but the idea was spon-on.canvas.png is specified again as the output image.L-systems in bash
Jan. 28th, 2026 08:05 ammiriam_eOnce again, I'm fiddling around with l-systems. The eventual aim is to develop a simple way to populate virtual worlds with plants. It has been many decades since I experimented with l-systems. This time I've begun with recreating the simple plant described in the Wikipedia l-systems page.
To give myself a baseline I wrote it in bash script. But bash doesn't have graphical output, so I used ImageMagick to draw the lines on a canvas, and displayed the result using the small, fast, image viewer feh. One nice thing about feh is that it can receive notification when the image it is displaying changes, and update its display. This let me watch the plant grow as the program proceeded.
It turns out that while bash is quite fast at generating the strings that define the shape, ImageMagick clogs things up somewhat. I asked ChatGPT about this and it surprised me by giving some great advice: ensure the canvas (which was in .png format) is not compressed and this will relieve a lot of the load. My computer still staggered under the weight, but it didn't do so badly. It took just over an hour to render the image below. My ancient CoCo computer from about 1983 took less than half that time. It had a 6809 microprocessor that ran at a feeble 0.89 MHz, whereas my current computer runs at an underwhelming 1.1 GHz (1,100 MHz) on each of 2 cores -- more than a thousand times faster. And if I run it on the XRoar CoCo emulator on my computer, but at full speed instead of the original speed, then it takes about 2 minutes.
 [18m56s or 2m15s].png "CoCo - 8 levels--writable screen (256x192) [18m56s or 2m15s].png")
L-system plant traced to the same level
of complexity on my old CoCo,
My next experiment (if I'm not distracted by some other shiny thing) will be to rewrite the program using awk, which will run much, much faster. For display I'll use the same technique as when I made Wolfram 1D cellular automata a while back. That is, write to the old ASCII .pbm image format. Since awk is primarily a text processing language, getting it to write image data to an image format that is intrinsically text is a marriage made in heaven. Also, I've already developed simple line-drawing, and other routines for awk. I'll still use feh to asynchronously display the image as it updates. This will be interesting.
It strikes me as weird that we have a great flowering of computer languages at the moment -- more than a thousand, last time I looked -- yet almost none of these can display or manipulate images or sound. It sometimes feels like we've gone backwards. We do have some specialised languages to manipulate images and sound, but they can do little else, and interfacing standard languages with those specialised ones can often be messy (like using ImageMagick to output graphics from other languages, such as bash).
It surprises me that with all the computing power we have today, progress feels like it is splintering. I can process images and sound and text in complex ways that once upon a time I could have only dreamed of, but in other ways it feels like we've hardly progressed. I used to be able to quickly whip up simple programs to process text and output graphics and sound... but no more, it seems. We have an amazing world that seems weirdly lopsided.
Additional: I just recompiled xRoar, the CoCo emulator, and got that warm feeling of being able to do those old experiments again, relatively easily and simply... though the CoCo's line editor is even worse than Linux/Unix's ancient ed line editor. The solution is to edit on linux and load the program into XRoar's CoCo.
To give myself a baseline I wrote it in bash script. But bash doesn't have graphical output, so I used ImageMagick to draw the lines on a canvas, and displayed the result using the small, fast, image viewer feh. One nice thing about feh is that it can receive notification when the image it is displaying changes, and update its display. This let me watch the plant grow as the program proceeded.
It turns out that while bash is quite fast at generating the strings that define the shape, ImageMagick clogs things up somewhat. I asked ChatGPT about this and it surprised me by giving some great advice: ensure the canvas (which was in .png format) is not compressed and this will relieve a lot of the load. My computer still staggered under the weight, but it didn't do so badly. It took just over an hour to render the image below. My ancient CoCo computer from about 1983 took less than half that time. It had a 6809 microprocessor that ran at a feeble 0.89 MHz, whereas my current computer runs at an underwhelming 1.1 GHz (1,100 MHz) on each of 2 cores -- more than a thousand times faster. And if I run it on the XRoar CoCo emulator on my computer, but at full speed instead of the original speed, then it takes about 2 minutes.
L-system plant traced to the same level
of complexity on my old CoCo,
My next experiment (if I'm not distracted by some other shiny thing) will be to rewrite the program using awk, which will run much, much faster. For display I'll use the same technique as when I made Wolfram 1D cellular automata a while back. That is, write to the old ASCII .pbm image format. Since awk is primarily a text processing language, getting it to write image data to an image format that is intrinsically text is a marriage made in heaven. Also, I've already developed simple line-drawing, and other routines for awk. I'll still use feh to asynchronously display the image as it updates. This will be interesting.
It strikes me as weird that we have a great flowering of computer languages at the moment -- more than a thousand, last time I looked -- yet almost none of these can display or manipulate images or sound. It sometimes feels like we've gone backwards. We do have some specialised languages to manipulate images and sound, but they can do little else, and interfacing standard languages with those specialised ones can often be messy (like using ImageMagick to output graphics from other languages, such as bash).
It surprises me that with all the computing power we have today, progress feels like it is splintering. I can process images and sound and text in complex ways that once upon a time I could have only dreamed of, but in other ways it feels like we've hardly progressed. I used to be able to quickly whip up simple programs to process text and output graphics and sound... but no more, it seems. We have an amazing world that seems weirdly lopsided.
Additional: I just recompiled xRoar, the CoCo emulator, and got that warm feeling of being able to do those old experiments again, relatively easily and simply... though the CoCo's line editor is even worse than Linux/Unix's ancient ed line editor. The solution is to edit on linux and load the program into XRoar's CoCo.
Liking chicks is gay, bro.
Jan. 27th, 2026 04:36 pmalisxThe show seems to go out of its way to say that it’s very masculine to be gay, which is exactly why people like those classical romances, where homosexuality is a natural overflow of masculine energy. It’s a vision of a sort of non-toxic but also non-anxious masculinity – masculinity that doesn’t require women to constantly affirm it.
Adrian Daub on .
Apparently all the AO3 girlies are now sufficiently grown-up and have influenced pop culture to the point that The Guardian is doing the whole “did you know girls like to watch dudes fuck???” thing. Which, like. On the one hand, 2006 called and wants its take back. But, also, on the other, the fact that this is in a big mainstream normie news outlet, about a big mainstream TV show definitely says . . . something? I guess? Maybe.
Lists of lists.
Jan. 26th, 2026 03:27 amalisxThere is a guilt that accompanies unread books, articles and blog posts. But there is a special anxiety reserved for unread lists of unread things. My reading list had become a totem of imagined wisdom, a shrine to the person I would be, if only I read everything on it.
When I deleted that list, I lost nothing real.
I know what I want to read. I know the shape of my attention. I don’t need a 7,000-item database to prove that I have taste or ambition.
JA Westenberg on .
Saturday @ 1:25 pm
Jan. 24th, 2026 01:25 pmalisxShout-out to the kids walking around the con, one with a giant scythe made of cans of Monster, the other in a rabbit ears headband with the “ears” made of . . . monstrous cans of Monster.
Friday @ 7:37 pm
Jan. 23rd, 2026 07:37 pmalisxWatching some AEW, and it definitely has a very different vibe than WWE, including an endearing level of, uh. Less-than-perfect polish, shall we say?
But having said that, I nonetheless has some big side-eye for the keming on the Hos Championship . . .
Try try try again.
Jan. 23rd, 2026 06:36 amalisxI also feel like we tried a lot of things and we found out that the old ways are better. You know? Streaming was great but then suddenly we didn’t own anything, so it’s good to go back to buying DVDs. ‘Social media’ was great but then suddenly we didn’t know anyone, so it’s good to go back to a platform where you can actually speak to people.
galadhir . . .
Boring, stupid, doomed.
Jan. 21st, 2026 03:54 pmalisxInstead, we get this timeline, where AI is en route to destroying much of the world in the most boring way possible, poisoning the greatest well of information ever conceived by mankind, lowering the average intelligence of an entire species, obliterating the market for art, literature and science while recalibrating our tastes to the most egregious slop imaginable, and downsizing the human workforce by millions of employees, all without ever demonstrating more intelligence than a very proficient autocomplete program.
On our impending .