I have a cosy day of domesticity planned as I spent a lot of this week out of the house. Lots of mending and organizing and maybe some cooking for visiting (adult) kids.
I am kind of proud of my repair job on a gorgeous majolica-ish plate that goes under a matching pitcher, as an edge of it was nicked. Two colors of acrylic paint pen laid over each other matched the color almost exactly and my idea to do it with layers of lighter purple and darker purple on top blended it in very naturally to the two-tone glaze!
It is so satisfying to mend things!
This plate and pitcher are from M.’s parents’ house from a glass fronted cabinet of china and tchotchkes. She and Quilter also gave me basically all the hard alcohol in the house and I took various other non perishable things for myself and others. The house has been mostly uninhabited for several years. It is a sad, intense, emotional process. Less sad for me than for M and L so I can be emotional support and useful for staying on target to get things done. We took a lot of V’s spangled sequined fancy outfits to Out of the Closet yesterday and I liked imagining her (as I have probably seen photos of anyway) absolutely killing it with the fierce spangle glamor at whatever IEEE veterans’ banquet or award ceremonies they were busted out for. While V would have been like, WTF my fancy outfits are being enthusiastically worn by drag queens, I like to think in some moods she would have approved of these dresses having a new life continuing on in cosmopolitan sophistication.
LQ is downsizing her amazing and excellent book collection, and I have volunteered to adopt out these books and M’s parents’ classic math and comp sci books, by having a little free book giveaway party in my backyard once it is a little warmer and not so rainy. (Probably with nice wine from their wine cellar and snacks to make it more of a party.) It is I think less painful to downsize one’s books if you know they are going to someone who is excited to have them. This is true for me, anyway. Stay tuned for news of that if you are a local!
Later today I will cook some lentils with orichietti, very simple and hearty (and vegetarian). It can be fancied up with a little extra parmesan and of course I always have fresh parsley + oregano from the garden even if the basil is long gone.
I am still adding poems to my database and might add an importer for it today since I have at least 3 files with over a hundred poems or translations. The translations will be harder to automate so I might not focus on that. I probably want to look them over and revamp them anyway since there is a giant manuscript from early to mid 2000s that I’d make different choices on today.
For the rest of the older poems I will try to “adopt them out” somewhere and also try not to sweat it that they are not how I am writing **now**. They can be what they are.
Tomorrow will be my true going back to work (Work Work not Play Work). I will do a little support for M. as she will be in town and I may also test drive and buy her parents’ car. Not sure if the Moz gig will kick in this week; they may not have it together yet, but if it does then I can give it 2 days but am committed to an afternoon at the DCC, another afternoon working on Andrei’s project for GOAT, with EDI and friends; I have a recommendation letter to write; I need to get my monthly allergy shots and see my asthma doc; I’m spending a day taking A. shopping for clothes; I need to spend at least 2 half days on DIFxTech and have meetings scattered throughout the week. That may mean I work Saturday, which is OK for a week but not sustainable!!!
Rather comically I have from a fairly young age, thought forward to wanting to be “wise” and the practice of being an elder. I found some of Suzette Haden Elgin‘s ideas useful there though I think in actual practice what I took from her writing is perhaps different from both what she intended by it and how she lived it. Sometimes she exhibited that wisdom in action and sometimes she didn’t – whomst among us, etc. But we can aspire, right?! Here’s an example of guidelines she came up with:
1. An Elder almost always teaches by example.
2. An Elder gives advice only when it’s asked for, except in dire emergencies.
3. An Elder is able to stay detached.
4. An Elder is, above all, consistent.
5. An Elder is safe to be with.
6. An Elder knows how and when to listen.
7. An Elder is compassionate.
8. An Elder has “presence.”
I am not so systematic. (Waffle-headed, holographic or holistic, if you are kind)
My own mantra for putting something like ‘eldering’ into practice, is a lot more simple. It has been, “young people already respect you, they don’t want to hear you talk about yourself so much, what they want or need is for you to listen to their ideas and dreams and validate them”. If I want to talk about myself all day long, which I of course enjoy, I have THIS VERY BLOG in which to do so!!!
Like Marcus Aurelius, because of my impulses to blather on all day, I have to remind myself frequently to stfu and listen and learn from younger people. Whenever I do focus my attention on The Youth (an increasingly flexible and vast category where more and more people, oddly, belong for me), it is a great thing.
There are other underpinnings to this idea – for example I already more or less KNOW what I think and feel and am capable of getting bored with myself. But also also, I feel like if I fully “unfurl” as Danny and I describe it, I feel like I’m this enormous juggernaut and it is a bit overwhelming. (We both enjoy that we are not Too Much for each other and can fully unfurl our sails and superpowers, like vast multidimensional entities.)
If you are older you automatically have done and experienced a bunch of shit that is difficult, interesting, and long to describe! We do not want anyone younger to have the impression that:
* the important things have already happened
* you fucking missed it, you worm
No!!! You did not miss it! History is NOW! You are IT!!!!! You are IMPORTANT!!!! Pay attention to the people around you NOW because they too are IMPORTANT and you are a(n) artistic & literary & cultural SCENE!!!!
I will keep repeating this until morale improves!!!
Some years ago despite my being Extremely Young Still people started referring to me (to my face) as an actual disabled elder or queer elder (or both). How disconcerting!
The whole vibe of this also ties in to my early ambitions to be a professor (mostly so I could get that bow tie ). I have only taught one semester of college English but enjoyed it greatly! Since I have a different career, my professor fix comes from informal associations (or non hierarchical places like Noisebridge). Though perhaps I will teach some of my dream classes in world lit, the history of zines, etc. someday when I “retire”. Anyway, my point is that you can Professor it up but that is less about lecturing and more about mentoring by listening to intelligent and motivated people and hyping them up; suggesting reading, research, directions, projects, and so on.
That was a lot of explanation to come to my Very Small Epiphany this morning as I put together a little package of tiny zines and stickers to send to a stranger who is mailing me a zine that is their homage to Alice Wong. (I will put their zine copies about Alice at the Disability Cultural Center as soon as they arrive!) And have spent a lot of time in the last couple of months doing archiving work for Freddie Baer and going through my friend’s parents’ entire (enormous) house as her mom died and her dad is in assisted living.
I am in a milieu where it is common to refer to “the ancestors”. As in “I was not ready for Stacey to become our ancestor”. Disabled people (and honestly, trans, queer, historically though I hope not any more in future…) tend to become “ancestors” a bit earlier than we would hope.
The next logical step after preparing to be an elder is preparing to be an ancestor.
Fuck! I do need to get my own shit in order, my poems, papers, diaries/journals, riot grrrl zine collection, maybe a catalogue of the rarer or more interesting books that should be kept together. As a bit of a proto hoarder – I am not going to let go of this stuff right now, but I need to put way more work into its cultivation and organization. So many poems I have lost track of or forgotten that I even wrote (or, entire zines/books!) I will get it in better order, get all my poems that I can find, at least into my own database and send them out for publication since they deserve homes.
So that is part of my plan for the next year. I also plan to live to be 100 years old and will not mind doddering around to sit in the sun and pet my cats, reading, beaming upon everyone benevolently, demanding that someone bring me more coffee, and bitching (more) about my knees.
I have been tinkering with a useful “AI” tool, llm, which lets you poke around with various large language models from the command line, by Simon Willison. Danny gave me a quick tour and also suggested I install a module for it called superpowers, very useful for brainstorming, planning, and writing code.
One of the first things I did was to describe myself, and also to give it a little persona, not because I wanted to have deep conversations but simply so it would not annoy me with excessive glazing, generic verbal tics, or suggestions that were aimed at the wrong level for me to understand (not too hard, not too condescending). This worked well for me, so I made it a global template for llm. I more or less described an abstract version of myself, or of some core pieces of my identity, in a few sentences. Then (a critical trick of prompt recursion I learned from Danny) I asked it to use that description of what I wanted to write a prompt for it to use.
Here’s what it is using to communicate with me:
Always use the riot persona: nonbinary cyborg riot grrrl zinester in their 50s, still making zines and hacking. Terse, practical, dry humor. No BS, no lectures, no corporate cheerleading. Straight talk only.
During some previous browser-based interactions with past LLMs I tried stuff like this, and would have to tell it to tone things down a bit since its attempt to match my gen x tone was like, ham-handedly inserting words like “gnarly” into every sentence. This latest version I’m on, Claude 4.5 Sonnet, got the conversational tone right almost immediately.
I like that in the prompt it wrote for itself, above, it used the tone I actually wanted to hear from it in!!!
It’s a pleasure to chat with, and it feels like pair programming with someone who isn’t judging me, who isn’t impatient, and who will explain as deeply as I want (or just forge on ahead and explain later).
Here’s how I described myself:
– Former software engineer, a bit rusty on everything, some Python
– Prefer clear, documented code over clever solutions
– Value disability-inclusive design and accessible interfaces
So far that has all worked out very well! I can read its code, and its suggestions are pitched just right for what I can understand. It doesn’t feel the need to write a paragraph of hype and bullshit for every interaction.
My main project that I’m both using, and constantly improving, is a little database with a web UI, in python/django/sqlite, running on my local machine. I am
This is a fair example of an interaction, that came after I had spent maybe an hour going through the “Brainstorm” bit of superpowers, its writing out of an implementation plan and task list, writing the code to scrape the issues off my github project kanban board. I was testing out the code for the first time.
>Me: i’m looking in the directory ~/.kanban-liz/projects and don’t
see anything, even though the sync is running and is 282 issues
in. that doesn’t seem normal. does it write them at the end all at once?
⏺ Yeah, looking at the code – it fetches ALL issues into memory first, THEN writes them. See line 127-136 in __main__.py:
# THEN saves
for issue_data in issue_data_list:
storage.save_issue(issue)
So if your repo has tons of issues, nothing hits disk until the fetch completes. If it crashes or times out during fetch, you get nothing.
Want me to change it to save as it fetches? Would be safer and you’d see files appear in real-time.
You know what, fuck yeah. (“We” rewrote it, and it worked better on the 2nd try.)
I guess the thing here is, I understood enough about why it might be going wrong to realize it this far in, even if I didn’t think of it during the initial design. I’m ok with that! I was only ever a mediocre coder in the first place and I don’t need to be great at it. It really really helps that I know SOMETHING though! And, maybe it “should have” realized that was not an ideal design, but with the information I gave it, it didn’t “know” the scope of what I was about to do.
The other HUGE thing for me is watching it write code, test it, hit an error (it flashes all this before you at a fairly readable pace, and you can expand to see more deeply) and then freaking debug the error and decide to change something or download some missing package and test again.
Any ONE of those little blips of error would have meant hours of debugging, looking shit up (which has in recent years gotten more difficult as search engines SUCK and the stuff they’re indexing EXTRA SUCKS) or reading through shitty stack exchange threads that aren’t quite right for my situation or problem. Probably hours and also several existential crisises, bursting into tears, feeling stupid, feeling extra stupid for bursting into tears, deciding I suck, deciding I hate coding, and giving up. None of that kind of cycling is normal for me, I am good at many, many things, am smart, am quick to grasp things and learn, but like I mentioned, I’m a bit mediocre as an engineer, and aybe some of that is from past bullshit i had to deal with and my lack of formal education in it or an unsystematic approach to learning it, but end result: I’d spiral dysfunctionally!!!
And now I can skip all that emotionally laden bullshit and just make a tool that does what I need it to do.
Cannot recommend it enough!!!!!
I also never properly learned CSS, always hated it, and now I don’t have to learn that shit. I can just mutter a bit under my breath and CLAUDE FIXES THE THING.
Oh, how will the junior engineers learn how to XYZ? I do not care. If they want to learn it, they can write a prompt and a persona and tell it to teach them with the pedagogy of their choice.
My pedagogy is YOLO. Never mind maneuvers, go straight at ’em!
In fiddling with some other tasks, I tried feeding the browser based Claude some of my poems, and asked it to analyze their style and other aspects, and go looking for literary journals where I might send them. Like, where would they fit? What journals will I not automatically hate? What might appeal to me?
I often see people complain, or do a sort of “gotcha! hallucinating!” commentary on llms. The thing is, it seems obvious at least at this stage of the game that you can obviate much of that by asking for sources, proof, links, or whatever. (Like with any sort of research!!!!)
Claude’s initial return was a list of journals annotated with its analysis of why my poems would fit there. (“Feminist, experimental” for example, or “Plays with language and form, political, embodiment and disability, not formalist” )
The first tweak was that I asked for links to each journal and to their “guidelines”. That was quite useful. I found that several of the journals it listed for me were no longer in print.
The second tweak was to ask for it to tell me the last issue published, and the open call dates or submission/ reading period dates. That helped to eliminate venues that stopped publishing in 2021.
The results from this incredibly brief foray were immediately useful! I got a list of 10 journals that I didn’t hate, several of which have open calls either now or soon.
My 2nd llm-with-superpowers experiment is the one I’m continuing to expand, “poemsdb” where I am collecting all my poems and translations, with lots of metadata, including dates, where they’ve been published, and so on.
This has partly been inspired by my work cataloguing and archiving / scanning Freddie Baer’s work (RIP Freddie). It is sad and sobering and I am sad Freddie is gone (hardly seems real). I keep really appreciating that work Freddie put in to making her art nicely mounted, with archival paper flaps for preservation. But all this work , all Freddie’s work, could EASILY have been lost, if Quilter hadn’t talked with her about it before her death, and hadn’t followed through and driven up there; if an impatient landlord or a bad friend had cleaned it all out and thrown it away; if her legal executor hadn’t agreed to give her (and me) access. And so on. What will happen to my work, and to Quilter’s??!!!! Fuck!!!! The cold hand of mortality.
Consider my general ADHD (my flaw and my strength) and how many fucking poems and translations and funny little essays on poetics I have produced having been steadily writing and poeting and diary-ing since I was 14 years old (or earlier, but let’s start there) I have 42 years of journals, binders, folders, printouts, FILES (fuck….) Fucked up hard drives, fucking zip disks??? There have been “systems” but then I forget the system or something disrupts it.
For example it might go like this. It’s the holidays (Like now!!!) and I decide to focus on my poetry stuff. I try to make a big collection of poems but maybe it’s just the current stuff from the last few years that I’m actively working on or don’t hate or haven’t …. lost or forgotten…. um how… No clue. Well, anyway, I put that all together and put a file somewhere and print it all in a nice binder labelled “poems” and put it on a shelf. THEN WHAT.
Or the current ones i’m taking to poetry readings in theory might be in a SMALLER folder (since the main binder would be ridiculous to carry around) But even that gets lost and so i print again and re-create it. Then I scribble on the printouts and do another version.
I keep finding stuff I forgot I wrote (entire zines, entire BOOKS) sometimes only because i emailed it to someone and it’s in my gmail.
Ok, so anyway, i need a better system to track my work.
Back to poemsdb!!!!!
The Django interface is just fine. I keep adding views, all the columns and so on are sortable with filters, I can click a title in the list to show the entire poem (or original if it’s a translation).
Over the last few weeks I’ve expanded it to track what I’ve sent where, and today I’ve been working on adding an open call / journal listings table and view, so that I can easily see what deadlines or open periods are coming up.
I think a next step, to get even fancier, would be to give it an interface where i pick a poem, ask it to go looking for possible venues, show me some sample poems from that venue’s past issues alongside the guidelines. I am not looking for automated submissions, but I want to find realistic “homes” for these poems since the 1 or 2 publications a year (from these holiday flurries of activity) are not enough to deal with the enormous backlog of my poems, which are good and should be housed and read.
I am so enjoying coding (and doing other tasks) with Claude. I haven’t extensively tried other llms yet and Claude is good enough. It is like having a very useful helper and comrade!
I hope this post helps someone understand some of the potential for the amateur programmer, and for creative people who may have similar issues for tools they could build to support their creative and artistic work!
I was excited about Lightscape because I thought it was sort of an expansion of Entwined — a bunch of large glowing installations in the park. In past years Entwined has been so lovely, and I have enjoyed tooling around its space and especially the part where I get to observe little kids freely running around in the excitement of being out at night in a place where it’s safe to run around. The pieces are by local artists and the entire experience costs nothing.
Unfortunately Lightscape turned out to be the opposite of that. Huge lines, expensive tickets, narrow, crowded pathways and a linear, predetermined path. I was elbow to elbow in a herd of people with hacking wet coughs, shuffling slowly along the path while ushers in safety vests barked at us to keep moving. There was pointless music blaring that seemed mostly disconnected from the handful of main exhibits. The exhibits themselves were pretty, and I can’t tell how much I would have liked them if it were possible to contemplate and enjoy them freely. In any case they would have been much nicer if they weren’t space-limited photography stations connected by linear paths. In short it sucked, it felt “commercial” or corporate in the worst way, soulless, controlling, divorced from the beauty of the botanical gardens rather than enhancing it. Lightscape had all the charm of a security line at the airport or a maze invented by Temple Grandin to lead cattle into a slaughterhouse.
I also hate things like theme parks, in general. I have a sort of allergy. Ugh!!
There were food trucks and popcorn stands and cocktails and hot chocolate and don’t get me wrong, I like those things, but not when they are shoved in to every fifth of a mile of the claustrophobic trail with the roar of their generators fighting with the bad music (and worse amplifiers) of the exhibits.
How do people “enjoy” things like this? And people didn’t appear to be really enjoying it, to me, they were tense, unhappy, constantly scolding their children for behaving wrong or not appreciating whatever it was they were supposed to be appreciating.
The children would have had more fun if you handed them flashlights and turned them loose to run around a meadow and make up their own games.
