drewp

I accept mail from the wild and reject bad SPF or bad DKIM.

It seems like something happened around 2026-05-01 (https://bigasterisk.com/post/delivered-email-rate-2026.png) where it became easier for spammers to make technically legitimately deliverable mails. Or my postfix is setup wrong.

Note "spf=pass" and "dkim=pass" in these headers:

  Return-Path: <walmartpoints@liftstores.com>
  X-Original-To: drewp@bigasterisk.com
  Delivered-To: drewp@bigasterisk.com
  Authentication-Results: prime; spf=pass (sender SPF authorized)
     smtp.mailfrom=liftstores.com (client-ip=103.176.193.110;
     helo=atompoint.bikerspage.com;
     envelope-from=walmartpoints@liftstores.com; receiver=<UNKNOWN>)
  Authentication-Results: bigasterisk.com;
      dkim=pass (2048-bit key; unprotected) header.d=liftstores.com header.i=walmartpoints@liftstores.com header.a=rsa-sha256 header.s=mtacsacc60r7j header.b=Ntcr2+R1;
      dkim-atps=neutral
  Received: from send1.liftstores.com (atompoint.bikerspage.com [103.176.193.110])
      by bigasterisk.com (Postfix) with ESMTP id D58B21A209B
      for <drewp@bigasterisk.com>; Fri, 15 May 2026 17:44:55 -0700 (PDT)
  DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; s=mtacsacc60r7j; d=liftstores.com;
   h=To:Reply-To:MIME-Version:From:Content-Type:List-Unsubscribe:Subject:
   Message-ID:Date; i=walmartpoints@liftstores.com;
   bh=2MNA1dmylVOMnvzWVYrwOi/6KCaFtBhIq3AtC5UbTrQ=;
   b=Ntcr2+R1Rvjce4/DYtKkfN6xtaBdKWawc/vcWU72jUBePh...

A variation of that one with month names: https://bigasterisk.com/post/2026%20Calendar%20-%20Print%20-...

Note that you can align the weeks (or not) by adjusting the width at print time.

https://en.wikipedia.org/wiki/Flattr figured out some parts of this. Notably, you picked your own total monthly donation, and then clicked a button on participating sites to allocate a fraction of your total to them. AFAICT it worked as advertised, but raised new issues with donation behavior. E.g. I obviously like curl every month, so should I click its button monthly? Twice monthly? If I am a developer of some other useful OSS software, should I click curl's button and the curl devs click my button? Does the money just slosh around between merchant-customers? Is that good?

https://www.modd3d.com/articles/item/waterworld-show-control...

The document is sadly 404 now, but this was an awesome nerdy writeup of the conversion of the Universal Studios Waterworld from Amigas in 2007.

Also see https://checkvist.com/

Tried justwatch.com for this?

What notes do you want to add to a movie?

It's also a recommendation in Material Design:

"Cut off grid tiles in the view’s initial scroll position to communicate the scroll direction for content overflow."

https://m1.material.io/components/grid-lists.html#grid-lists...

We're, uh, solving that problem

Are there comparisons to related projects somewhere?

E.g. I've use DBeaver before. Is there some reason I should try Outerbase next time?

https://blogs.igalia.com/scerveau/introducing-gstpipelinestu...

I am trying to make efficient gst pipelines for security cameras. I only took a quick look at GPS so far. The UX for making connections is a little weird, but so far the whole tool seems to work as advertised.

I found some existing mpv scripts:

https://github.com/oltodosel/mpv-scripts/blob/master/show_ch... display chapter names as OSD

https://gitlab.com/lvml/mpv-plugin-excerpt press 'i' and 'o' for in/out points, then 'x' to make a new (auto-named) file.

https://github.com/shinchiro/mpv-createchapter press 'shift-c' to mark chapters; export as xml file.

https://github.com/mar04/chapters_for_mpv mark chapter times, input titles, save as txt file

By having an LLM name the chapters, I meant having whisper do speechrec on the chapter and then asking an LM to summarize the content into a name up to k chars.

Yeah, I play a VHS tape and capture the whole thing. Maybe I should be using a scene detector to split files on camera cuts, which would be roughly correct for home movies (but not for TV shows).

That's where I am now. I'd like to optimize out the retyping and duplication of time strings.

I want the player ui (I'm using mpv) to have a command that:

1. Remembers the last end time to use as this chapter's start time

2. Gets the current time to use as chapter-end.

3. Accepts the name (e.g. 'chapter1').

4. Runs the ffmpeg copy command.

Perhaps mpv+lua can already handle this. I see commands for setting a loop range and for calling a subprocess. Not sure how I'd input the chapter name. Maybe I'll have an LLM name the chapters for me :)

Can you recommend a tool for dicing up 2h digitizations of VHS tapes? I want to play the 2h video, seek around easily, mark 'chapters' and give them filenames, then do a no-transcode rough cut extraction of each chapter into its own video.

Looks like the same mission as:

* https://www.codeshelter.co/

* https://up-for-grabs.net/

* https://www.codetriage.com/ (mentioned in these comments)

I seem to recall yet another one, maybe with a name that invoked a traveling group of helpers who would jump into projects briefly to fix them up?

On a related note, it would be cool if there was a way to leave a hobo sign equivalent if you find a project that is well-run and easy to contribute to. If the build and tests Just Work, etc., we should praise that project in a way that 1) encourages helpers to pick it since they'll have a good experience, and 2) provides a good example for other projects to follow.

e.g. https://www.youtube.com/watch?v=axow7KnBtaM :)

I threw something together to address that goal. It was really hard to render obsidian-md into html that looked about the same. I started with general MD libs, then ultimately used obsidian-html. Obsidian's MD is not standard at all, though I'm not sure any MD is fully specified since there are so many edge cases.

Sample output (should be viewable since I put `#public` at top of the document): https://bigasterisk.com/vault/esp%20cams.md

Server code: https://bigasterisk.com/code/vaulterrific/files/tip/

107k photos here, and it's not great. I just clicked a month in the calendar and getting just the json back took 1.8 sec. Last of 12 images came in 2 sec later. Postgres, SSD.

Plus: all thumbs of non-square images are cropped, and it makes all this per source image:

-rwxr-xr-x 1 photoprism photoprism 1689 Sep 21 2022 json/0/0/0/00000f1790c93fdaaca662b56326fb5d130ab45d_exiftool.json -rw-r--r-- 1 photoprism photoprism 3944 Sep 21 2022 thumbnails/0/0/0/00000f1790c93fdaaca662b56326fb5d130ab45d_100x100_center.jpg -rw-r--r-- 1 photoprism photoprism 7856 Sep 21 2022 thumbnails/0/0/0/00000f1790c93fdaaca662b56326fb5d130ab45d_160x160_crop_2a51210960c8.jpg -rw-r--r-- 1 photoprism photoprism 12881 Sep 21 2022 thumbnails/0/0/0/00000f1790c93fdaaca662b56326fb5d130ab45d_224x224_center.jpg -rw-r--r-- 1 photoprism photoprism 12151 Sep 21 2022 thumbnails/0/0/0/00000f1790c93fdaaca662b56326fb5d130ab45d_224x224_left.jpg -rw-r--r-- 1 photoprism photoprism 13601 Sep 21 2022 thumbnails/0/0/0/00000f1790c93fdaaca662b56326fb5d130ab45d_224x224_right.jpg -rw-r--r-- 1 photoprism photoprism 100 Sep 21 2022 thumbnails/0/0/0/00000f1790c93fdaaca662b56326fb5d130ab45d_3x3_resize.png -rw-r--r-- 1 photoprism photoprism 49090 Sep 21 2022 thumbnails/0/0/0/00000f1790c93fdaaca662b56326fb5d130ab45d_500x500_center.jpg -rw-r--r-- 1 photoprism photoprism 1824 Sep 21 2022 thumbnails/0/0/0/00000f1790c93fdaaca662b56326fb5d130ab45d_50x50_center.jpg -rw-r--r-- 1 photoprism photoprism 74535 Sep 21 2022 thumbnails/0/0/0/00000f1790c93fdaaca662b56326fb5d130ab45d_720x720_fit.jpg

The UI doesn't change the url to the photo you're looking at, so you can't share urls.

Images aren't links, so you can't do browser stuff like open one in a new tab. The back button doesn't work right at all.

The choice of metadata to display under each image in the listing is downright silly ('place', city, year, DoW, month, year again, time, zome, camera model, resolution, file size, city again, state, country).

The information scheme features 'albums', 'favorites', 'moments', 'labels', and 'folders', as if they were being forced to match a bunch of legacy systems. I want a powerful tag/group mechanism that I can use for many possible workflows.

The list under 'search' says 'Review 40193', clearly telling me I have that many pics to review. Clicking on it says:

No pictures found Try again using other filters or keywords. In case pictures you expect are missing, please rescan your library and wait until indexing has been completed. Non-photographic and low-quality images require a review before they appear in search results

(And I had to disable some user-select:none styles to copy paste that)

http://twitter.com/#!/diveintomark/status/15772480610242561 "You know what you should do with your life? Write a del.icio.us clone that no one will use and then die alone."

(I've written my own, of course :)

https://support.google.com/youtube/answer/2902117?hl=en#zipp...

"This option mutes just the claimed song. Other audio, like dialog or sound effects, won’t be muted. "

Last I tried, though, certain kinds of value updates didn't work in WC mode and I had to fall back to non-WC mode to finish :(

A funny phenomenon I noticed on my home setup:

If you start a deployment ("your container", roughly speaking), the age of that deployment will keep counting up even if the container exits-- k8s restarts it of course-- and even if the k8s scheduler goes down-- since we want to be able to restart the scheduler without unnecessary service restarts.

At home, it's all on one computer in my basement, so when I reboot that box, the k8s reports come back and keep telling me the deployment has been there for xx days (just with some availability hiccups).

My mnemonic for the "-r" flag is to just remember that k8s gets it right:

"kubectl apply -f myfile" ("apply file ...")

and pip stacks two verbs together in a silly way:

"pip install -r myfile" ("install read ..."??)

Check out https://www.marcopolo.me/

It will even stream an in-progress video message to you as it's being recorded if you happen to be looking at the app then.

I don't think you're going to have a problem. The alias only affects you trying to "run" the file.

% foo.jpg

# launches image viewer

% cp foo.jpg bar.jpg

# unaffected

'C-x m' pulls the newest file for me:

mv ~/Downloads/[C-x m] /tmp

or even within files that match what I've typed already:

tail /logs/pho[C-x m]

Temperatures inside and outside the window, I suppose.

How great would it be if you did redo it every other day, though?

You'd automate the video publish pipeline, of course, and then it would be like a cross between a high-level UX test and a rehearsal for your pitch. You'd get really good at the workflows you demo, which would probably make the demo more impressive, and you'd be more likely to clean up any rough edges or annoyances that you come across repeatedly.

Yes, some machines on different networks in my house plus a roaming laptop all got to talk to each other without hassle. They got random-looking ipv6 addresses, so I don't know if 'subnet' is the right word here.

Before, I had to act differently on the laptop based on where I was, and the raspi nodes on my guest wifi couldn't reach the influxdb server that was not exposed to the wifi net.

I ran cjdns for 5 years but recently switched off it. For one, it seemed to have too much traffic over my limited outbound link when things should have been more idle. Also, the commit logs were a little unnerving to me for code I'm trusting with my network security:

All kinds of "oops" edits: https://github.com/cjdelisle/cjdns/commit/3abe50b8e744f72696... https://github.com/cjdelisle/cjdns/commit/feabb1970fbaecf65c... https://github.com/cjdelisle/cjdns/commit/c51d89431f1fa42955... https://github.com/cjdelisle/cjdns/commit/1b0c999bd2e5988c3f... https://github.com/cjdelisle/cjdns/commit/355d7d77cc82c52bf1... (function args passed in the wrong order)

Some are about extra traffic: https://github.com/cjdelisle/cjdns/commit/8bcfbf227a87020931... https://github.com/cjdelisle/cjdns/commit/0ee9cb7f45b466232b...

Also it's ipv6 only, which took some fussing. Performance was fine, even on rpi. Reestablishing links was slow and sometimes required restarting of the daemon. Daemons would occasionally get wedged. Log output is unconventional and uses a special reader tool (a la adb for android). Occasionally routes wouldn't be chosen right (two home computers would route via an external cloud box), which I'd fix by restarting the right daemons.

I laboriously switched to openvpn (two networks) and haven't worked out all the routing and hotswitching for my roaming phone+laptop yet. Now I'm considering vita, tinc, zerotier, or wireguard. Probably I'll try out zerotier to see if it works out of the box, then try wireguard if I'm going to have to configure it a lot, since it seems like WG is the most unix-toolbox-do-one-thing out of all of them.