Love & Rockets & Claude

Today, I bought a fairly large Humble Bundle of Love & Rockets comics. I then tried using Claude to help me download and organize the bundle, with mixed success. So I thought this would be a good topic for a blog post, since it allows me to talk about my two favorite things at once: comic books and tech.

Love & Rockets

I’ve been reading Love & Rockets, on and off, since the early eighties. I’m pretty sure the first I issue I bought was #2, from 1983. I think I have most of that original run, through to issue 50. (That was magazine zine, in black and white.) I think I also have the first ten or so trade paperbacks, from the original run of those. After that, things get a little fuzzy. Here’s a bibliography from 2021 and a How to Read Love and Rockets article that was last updated in 2024, both from the Fantagraphics site. I think I probably have all of the volume 2 run (comic-book size, early 2000s). I think I have all of the New Stories run in my Comixology library. (That one was done in a larger “book” format, rather than the magazine/comic format of the earlier volumes.) And I didn’t have any of the current volume 4 run, which I think is back in the original b&w magazine format.

And, in terms of what I’ve actually read, I know I’ve read all of the volume 1 run, either in the original format or the collections (and probably both, for the earlier issues). And I’m pretty sure I’ve read all of volume 2, in the original comics format. I haven’t ready any of the volume 3 (New Stories) stuff yet. Which is why I haven’t bothered buying any of volume 4.

The Humble Bundle includes 57 items, and as usual with Humble, they don’t really do any of the organizational work for you. After getting some help from Claude (more on that later), I see that it has 15 volumes of the L&R library. That might be the whole series to date. It also has 17 issues of volume 4, which is probably all of that to date too. And then it has… a bunch of other stuff. The Hernandez brothers have done so much work over the years, and it’s appeared in some many different formats, it’s really hard to figure out what you’ve read and what you haven’t, past a certain point. But, just from a reading perspective, for me, I now have all of volume 4 in a DRM-free format that I can read whenever I want, so that’s a big win. And I think I probably now have most (or all?) of volumes 1, 2, and 3 in the L&R library volumes, so that’s convenient.

Claude Cowork

So where does Claude come in? I thought there might be a few things it could help me with:

  1. Humble doesn’t give you a simple text-format list of the items in your bundle. In the past, I’d copy and paste the text from the bundle page into Notepad++, then do a bunch of frenzied deletion until I wound up with a simple list. So I thought Claude might help there.
  2. Downloading all of the items in a bundle is always a pain. There’s are various ways to help out with that, but I had some hope that maybe I could just point Claude at the download page and say “grab all this stuff for me.”
  3. After downloading the stuff, I usually spend some time renaming files and organizing stuff into sub-folders. I thought Claude might help with that too.

So here’s my actual experience with all that, starting with the simple task of trying to get a plain-text list of the books in the bundle. First, I tried giving Claude the URL to the main (public) web page for the bundle, and asking it to scrape the list of books are reformat it for me. That didn’t work, as apparently Claude is blocked from browsing the Humble site. (Looking at their robots.txt file, I guess that makes sense, and I appreciate Anthropic/Claude for respecting that.) So I then copied the text from the page myself, pasted it into Claude, and asked it to make me a list of the books. It did a good job of that, so that was one thing done.

In terms of trying to get Claude to help me download the books, that was a bust. I tried getting Claude Cowork to do it for me, but again there was the robots.txt exception. It did then generate a Python script for me that should have allowed me to download the books, but (long story short), I couldn’t get it working. So I gave up and used the “bulk download” option on the Humble download page, which succeeded in downloading most (but not all) of the files. So I then asked Cowork to look at the downloaded files, and the list it had made for me, and tell me which ones I’d missed. And it did a good job of that! So that saved a little time.

So then, having all 57 items in a single folder, I talked Cowork through doing a few things for me. First, I created a Markdown file in the folder with the list it had generated. I asked it to go out to the web and create a revised list, categorizing the various books in the bundle into groups, and adding publication date and some other summary info. It did a pretty awesome job there, creating a well-formatted Markdown file with headings and tables, separating the Jaime volumes from the Gilbert volumes, filling in publication dates, and including short summaries of what’s in each volume. I then also had it rename the files to look nicer, e.g. “pennycentury.pdf” to “Penny Century.pdf”. It did a really good job with that too. I also had it separate out all of the L&R library volumes into a sub-folder, and put the volume number at the start, so “Penny Century.pdf” became “08 Penny Century.pdf”. I did a bit more fiddling around there, and now I have a pretty well-organized collection of books, where I know which ones are which, rather than just 57 badly-named PDF files.

So that’s all pretty cool. I asked it to do one more thing for me, and that didn’t work out too well. One of the files was, for no particular reason, an EPUB instead of a PDF, so I asked Cowork to convert it into a PDF. It took a lot of spinning to get there, but it eventually did. But the resulting PDF was basically unusable. I may come back to that at some point, since I have other EPUBs that would work better as PDFs or CBZs, and I’ve done some work on that already. (I don’t think I’ve ever written a post about that. Maybe I will, when I get back to it.)

So, at the end of the day, I think I have a better idea about stuff where Claude Cowork can help me out, and stuff where it’s going to be mostly useless. I think I might point it at some of the other stuff I’ve downloaded from Humble over the years, and let it clean up and organize some files, in cases where I haven’t gotten around to it. And also ask it to create a nice summary Markdown file, the way I did here.

Future possibilities

There’s one other thing that would be useful, but that I haven’t tried. It would be great to have a way to automate getting all of the books added to my Goodreads account, and tagged appropriately. It would also be cool to have it create a Goodreads list with all of the books. But I’m about 99% sure that Claude won’t be able to automate that. If it can’t even read the Humble pages, I really doubt it’ll do work on the Goodreads site for me. I could probably mess around with it, but I’m not sure it’s worth it. I know there’s a Claude add-in for Chrome that can do some interesting stuff, but I use Firefox, and I don’t want to have to go through setting up Chrome right now. I could also try one of the fancy new AI browser tools, like Dia or Comet or something, but I don’t want to go down that rabbit hole right now either.

Maybe I could get Claude to create a CSV file that I could import into Goodreads. Hmm, that might be worth trying. But maybe not today. It’s fairly warm out today, so I should probably go out for another walk, and stop wasting time in front of the computer.

Dracula Pro

I’m one of those weirdos who experiments with fonts and color schemes and application themes more than I probably should. For the most part, I do actually stick with some sensible defaults, but I keep coming back to it and messing around. Sometimes, this is just a way to procrastinate and avoid real work. But I think there’s a fine line between procrastination and “sharpening the saw.” My eyesight is bad enough at this point that even a small improvement in legibility can help me out a lot.

So this week, for some reason, I got on a “playing with themes” kick again. I’m not even sure what kicked it off. But my focus this week was the Dracula theme. The basic free version is available for quite a few apps! I first started using it with git bash, the version of bash that comes with Git for Windows. That was quite some time ago. Dracula was included with the default themes there, and it was easier on my eyes than any of the other defaults. I’ve seen it included here and there with other apps, but hadn’t paid much attention to it.

But, again, something drew my attention to it this week, and I got curious about the paid Dracula Pro version. It costs $80, which seems like a lot just for a pack of themes, but they throw in some other stuff too. After some dithering, I decided to go ahead and pay for it. I haven’t seen a lot of reviews of it, so I thought I’d write one up here.

I’m not specifically a “dark theme” or a “light theme” guy. I lean more towards light themes, so paying for a pack of mostly dark themes doesn’t really make much sense for me, but I found Dracula to be more usable than most other dark themes, so I thought I’d give the Pro version a try.

After you pay for it, you’re directed to a Gumroad page where you can download a zip file. It’s a big 750 MB file, which surprised me a bit. Most of that space (about 600 GB) is the audiobook version of the creator’s book, 14 Habits of Highly Productive Developers. I think maybe they should have put that in a separate zip file, but I guess it was simpler this way.

I’ll walk through the contents of the zip file here, with some comments on what I’ve tried so far.

  • As mentioned above, there’s a content folder with copies of the ebook and audiobook for 14 Habits of Highly Productive Developers. The ebook is available in English, Portuguese, and Spanish. EPUB and MOBI versions are included for each. The audiobook is available in English and Portuguese, in MP3 format. I haven’t tried reading/listening to any of it yet, but I did copy the EPUB to my Kobo. I’ll probably compress the MP3s into a single M4B file and copy it over to my iPhone at some point too.
  • Next, there’s a design folder, with some info about the color palettes that I guess would be useful if you were designing a theme for a new app and wanted to use the Dracula Pro palette.
  • There’s a fonts folder, with five fonts, all of which are already available for free. So it’s useful to have them all together in one place, but it doesn’t add any value really. The fonts are Cascadia Code, Fira Code, Inconsolata, JetBrains Mono, and Victor Mono.
  • There’s an icons folder, with custom icons for a bunch of different applications. I don’t see much point in messing around with icons. I guess some people like doing that, but to me, it seems like a lot of work for little gain.
  • Then there’s the themes folder, which is basically what you’re paying for. It includes themes for 21 different apps. I’ll get into some details on that below.
  • And last is the wallpapers folder. There are a bunch of files here, but they’re all basically the same wallpaper design just in different sizes/resolutions, and with slightly different color combos.
  • I should mention that the description of Dracula Pro says that it comes with a “bonus screencast,” but I don’t see that in the zip file, or any reference to it in the email receipt.
  • I’d also mention that the description on the website says “Your license covers multiple computers with activation on up to 3 devices.” Based on that wording, I was a little worried that there would be some kind of half-assed activation software included with the package, but there isn’t. (Not that it would even be possible to enforce that for most of this stuff, but I’m glad they didn’t try.)

So that’s a bunch of stuff, and it’s well-organized, but it’s up to you to decide if it’s worth $80 or not.

There are seven variants of the theme, six dark and one light. The main dark theme is called Dracula Pro. The light theme is called Alucard. the rest of the themes are all named after various vampire-adjacent characters (Buffy, Blade, Morbius, etc). It’s cute. So far, I’ve only messed with Dracula Pro and Alucard. The other dark themes seem to be minor variations on the main Dracula Pro theme.

Here’s what I’ve tried out, and what’s worked for me and what hasn’t:

  • I’ve applied the Visual Studio Code theme on my home and work computers. It’s distributed as a .VSIX file. (The free Dracula theme is also a .VSIX file, available in the VS Code extension marketplace.) Since the Pro VSIX file isn’t freely available, it’s not in the Marketplace and needs to be installed manually. And it won’t sync between computers like the Marketplace extensions will. So that’s a bit of a pain, but not a big deal. It’s working well for me, and I think I’ve going to stick with it.
  • There are no themes in the pack for the regular Visual Studio product, which was a bit of a disappointment. The free Dracula theme is available in the VS Marketplace (here), so I’m not sure why they didn’t create a variant for the paid theme. Honestly, though, for VS 2026, the “Cool Breeze” theme (one of the included default themes) seems to work really well for me, so I’m sticking with that anyway.
  • On my Mac, I tried the Dracula Pro and Alucard themes for the Mac Terminal app.  Both are fine, but I don’t do that much work in Terminal on the Mac, and was happy enough with the default theme.
  • For Windows Terminal, I imported the Dracula Pro theme, and I’ve decided to use that as my new default for git bash and Ubuntu tabs, but I’m leaving the PowerShell default as-is (the “Campbell” theme), since I think Dracula Pro doesn’t work as well with PowerShell.
  • I tried Dracula Pro and Alucard in Notepad++, but have had some issues there. I won’t get into details here, but I might not bother going further with that. (Or I might, if I get bored and want to mess around with XML files for a while…)
  • There’s not much else in the way of themes that I want to try. My main text editing environments at this point are Visual Studio, VS Code, Notepad++, and Dynamics AX 2012, which isn’t customizable at all. Terminal windows, for me, are mostly through Windows Terminal and sometimes Mac Terminal.app. So this is covering most of my bases.

As to fonts, I’ve been reviewing my font settings too, and for now, I’m trying to standardize on Cascadia Code. I was already using that for a lot of stuff anyway.

So that’s my review of Dracula Pro. Was it worth $80? Is it that much better than the free version? I don’t know. But it was a good opportunity to review all of my font and color settings and try to make things a little better, so I don’t regret it.

Evernote to Obsidian – point of no return

I’ve been chronicling my journey from Evernote to Obsidian here on my blog; here’s a link to my last post about that. I thought I’d write a post today to mark the “point of no return” with this project.

This morning, I deleted all of the notes from my Evernote account, and canceled my paid account. I didn’t completely delete my account; I now have just the free version, so if I need any of my old notes in the next week or two, I could probably log in and pull them out of my trash. But, for all practical purposes, I’m committed to Obsidian now, as my “trusted system.”

I think I’ve got Obsidian set up the way I want it now, but there’s some stuff I want to clean up and/or tweak. First, I’m still trying to clean up all of my images/attachments that came over from Evernote. I initially put them in separate sub-folders for each notebook that I imported from Evernote. Now, I’m trying to consolidate them into one top-level “_resources” folder. As I’m doing that, I’m cleaning up file names on them a bit. A lot of them have names like “img_1234.jpg” or “snapshot.1.jpg”, so I’m trying to rename those to be a bit more unique and descriptive. I initially tried a plugin that automatically renamed them to match the name of the note that they were referenced in, but that plugin seemed to cause some issues, so I gave up on that. Now, I’m going through them a few at a time. I’m not renaming all of them; if they’re already named with a date/time stamp of some kind, or anything else that’s reasonably unique, I’m leaving them be. This is still probably overkill, but it’s also giving me a chance to look at some of the older notes and clean them up a bit too. Eventually, I’ll be done with that.

I’m also experimenting a bit with how I’m going to record my day-to-day activities and task completions. I’m using the tasks plugin, and that’s working well so far, as a replacement for Evernote reminders. To record day-to-day activities, in Evernote, I used to keep a year-long task note, named “Tasks 2025” (or whatever year it was), with a bulleted list of ongoing projects and simple to-do items at the top. On any day where I was working on stuff, I’d create a date-stamped list of stuff I worked on that day. That list would get pushed down below the master project/to-do list when I was done with it, so, by the end of the year, the note would be a reverse-chronological list of everything I’d done that year. (Well, not everything, but you know what I mean.)

In Obsidian, I was initially going to keep doing that. But then I thought about trying daily notes instead. I started that on January 1, and it’s working OK so far. But I have a bit of a hybrid system now, where I’m using a combination of three notes to track things:

  1. A “tasks” note that is just a bunch of task queries, to show me what’s due today, what’s due in the next week, and a master list of all pending tasks.
  2. A “Tasks 2026” note, with that master list of tasks, projects, and simple to-do items.
  3. A daily note each day, with the list of stuff I did that day.

So that’s probably too much stuff. I should probably consolidate the note with the task queries and the note with the project (etc) list. But I haven’t settled on how I want that to look. Either way, I think I’m on the right track. I just haven’t converged on exactly what I want yet.

I still have a couple of friction points with Obsidian. Sync is occasionally not as reliable as I’d like it to be. I guess I need to get used to looking at the little sync status icon in the lower right corner of the screen and not exiting Obsidian until it’s green. (Evernote sync had been very reliable lately, and I’d been taking that for granted. I guess that was because Evernote had really transitioned to an online-first experience, while Obsidian remains a local-first experience. The Evernote client, I think, was constantly saving stuff to their back-end; the local database was probably being treated as just a cache, really.)

Most of the other friction points are just little things that I need to get used to, or find ways to work around. I’ll get there.

Obsidian, day three

Well, I’m now on day three of my vacation week, and day three of my Obsidian setup. (See here for day one and day two.) I’m going to write up some notes below on various things I’ve been working on.

Obsidian Sync

I signed up for Obsidian Sync, at the $8/month Plus level. For now, I’m paying month-to-month, so it’s actually a $10/month plan. I wasn’t sure if I needed Plus or Standard. My vault is around 700 MB, so it’s under the 1 GB limit. But I initially had some attachments that were over 5 MB, so that might have been an issue. I think I’ve removed or resized all of those now, so I could probably make do with the Standard level. The Plus plan also gets you a full year of revision history, which is nice, so I might stick with that either way. I’m not sure yet.

I’d previously experimented with simply putting my vault in OneDrive and iCloud. OneDrive worked fine for PC/Mac sync, but wouldn’t have worked on iOS. I thought iCloud might work OK on iOS, but it’s a little iffy. So, for now, I’m paying for Obsidian Sync.

Obsidian Sync does seem to work fine on iOS and iPadOS, but there’s one thing I’ve noticed that I didn’t initially think about: Obsidian, even on iOS, is local-first, so your whole vault gets synced to your iPad and iPhone. That’s not a huge problem, but it’s interesting to see that the Obsidian app on my phone is using 800 MB of storage, while Evernote is only using 500 MB. I think Evernote caches a certain amount of information locally, but the design is online-first, so (I think) it’s always going to try to get info from the cloud.

Tasks

My big project today has been converting all of my Evernote reminders to tasks. I’m still a little nervous about this. I’m losing the email reminders and iOS notifications that Evernote provides. So I’m going to have to be good about looking at my main task note in Obsidian. This is how I track important life stuff, like paying bills, so if it falls apart, I’m in trouble.

I’ve now done the migration, and I have 70 tasks in my vault. I did the migration manually, rather than trying to create a script to do it. I wanted to be able to review all of the notes associated with the reminders, and think about them, and maybe revise them a bit.

One thing I did to make this all easier: I assigned a shortcut key to the “create or edit task” command. I used Alt-T for that (Opt-T on the Mac). The pop-up dialog for this is reasonably easy to use. I do have one big problem with it though: there’s no date picker. When I’m setting due dates on tasks, if they don’t need to be done on a specific day, I like to set them to the closest Saturday, which is my usual day for taking care of random to-do items. I guess I’ll have to live without that for now.

To actually view the tasks, I have a “task note” with a number of task queries on it. For now, it looks like this:

task note

(I tried to insert the code here, but WordPress got confused, so I’m just putting a screenshot here.)

So not too fancy. I’d really like to find a way to format this a bit better. Maybe in a table? If there’s a way to show task info in a “base”, I haven’t figured it out yet. I did use a base to show all of my Evernote reminders, and that worked well. I might as well stick that in here too:

views:
  - type: table
    name: Table
    filters:
      and:
        - '!note["EN-Reminder"].isEmpty()'
        - note["EN-Reminder-Done"].isEmpty()
    order:
      - file.name
      - EN-Reminder
      - EN-Reminder-Done
    sort:
      - property: EN-Reminder
        direction: ASC

This is basically showing all notes that had an Evernote reminder, but did not have a “reminder done” date.

In working through my reminders/tasks, I’ve noticed that a lot of them are just reminders to review a given annual subscription before it renews. I think I might look at seeing if there’s a way to categorize these specific tasks and separate them out. Maybe a “#subs” tag or a property. And once I started thinking about properties, I started thinking that I could have a specific type of note with a number of specific properties that I could use to organized my subscriptions. Maybe properties for renewal date, cost, URL, and so on. Well, that’s a project for later maybe.

Images

I may have gone a little overboard with image cleanup today. I installed this plugin, which is a simple little script that renames all images on the current note to match the note title. So, for instance, ‘IMG_1234.jpg’ becomes ‘drivers license 2014.jpg’. It only does this for one note at a time. So I went through my main archive folder and ran it on, probably, around 400 notes. And did some general note cleanup along the way.

I think I need to stop myself from going overboard with miscellaneous note cleanup. It’s easy for me to go down a rabbit hole of doing low-value file maintenance tasks and losing sight of the big picture. This has been a problem for me, in general, really. Obsidian really gives me an opportunity to waste a lot of time fiddling with unimportant stuff, and I need to watch out for that.

(Update: I had some sync issues after all of those image file renames, so I think I’m going to delete the plugin linked above. I don’t know if the sync problems were due to the plugin, or something else, but… better safe than sorry. And I’m hoping that the kind of sync issues I just had aren’t common with Obsidian sync. If they are, then I’m going to need to go back to Evernote!)

Evernote to Obsidian, work in progress

As per my previous post, I went ahead and migrated myself from Evernote to Obsidian yesterday. I’m now almost at the point of no return. (Or at least at the point where I’d have wasted a lot of effort if I were to throw it away now and go back to Evernote.)

I used Yarle for the migration. I’d done several experimental, partial, migrations first. For the final migration, I did all of my Evernote notebooks all at once. (I had a thought in my head that Yarle might resolve cross-notebook links if I did that.) The migration went pretty smoothly, but Yarle seemed to lock up at one point. I checked the count on the output files, and it seemed like it had created all of them, so I did an “end task” on it and proceeded from there.

In the end, I’ve wound up with a lot of broken links. I’m starting to wonder if killing the Yarle process was a mistake. Maybe it had created all of the .md files, but was still reconciling the links? I don’t really know enough about how Yarle works.

Either way, I’m now cleaning up hundreds of broken links. With the broken links plugin, I’m able to at least identify them easily. There are actually two different kinds of broken links in my vault now: there are many ‘regular’ broken Markdown links; I can ID those with the plugin. Then, there are links that point out to share.evernote.com, which didn’t fully get converted to Markdown links. Those are technically ‘valid’ links, but of course they open up my original Evernote notes in a web browser, so I’ll need to clean those up too.

I started doing the link cleanup manually, but at some point, I saw how big the job was getting, and decided to write a couple of helper scripts. With the help of Kagi Assistant, I wrote two PowerShell scripts. The first cleaned up links where Yarle had left them in a format like this:
[[guid/guid|name of link]]
In that case, I wanted to change them to:
[[name of link]]

Here’s the script I used:

 Get-ChildItem -Recurse -Filter *.md | ForEach-Object {
    $content = $_ | Get-Content -Raw
    $new = [regex]::Replace($content,
        '[[[0-9a-f]{8}-[0-9a-f]{4}-[0-9a-f]{4}-[0-9a-f]{4}-[0-9a-f]{12}/[0-9a-f]{8}-[0-9a-f]{4}-[0-9a-f]{4}-[0-9a-f]{4}-[0-9a-f]{12}|([^]]+)]]',
        {param($m) "[[{0}]]" -f $m.Groups[1].Value},
        'IgnoreCase'
    )
    if ($new -ne $content) {
        $_.FullName                    # show changed file
        Set-Content -NoNewline -Path $_.FullName -Value $new
    }
 }

And for the second case, I wanted to clean up the external Evernote links, like this:
[name of link](https://share.evernote.com/note/guid)
To again change them to this:
[[name of link]]

So here’s that script:

Get-ChildItem -Recurse -Filter *.md | ForEach-Object {
    $content = $_ | Get-Content -Raw
    $new = [regex]::Replace($content,
        '[([^]]+)](https?://share.evernote.com/[^)]+)',
        {param($m) "[[{0}]]" -f $m.Groups[1].Value},
        'IgnoreCase'
    )
    if ($new -ne $content) {
        $_.FullName                    # print the altered file
        Set-Content -NoNewline -Path $_.FullName -Value $new
    }
}

I didn’t even really test these scripts, I just backed up my vault, then ran them. And, surprisingly, they seemed to work right on the first attempt.

So that got me through most of the link cleanup. I then did a bunch of manual fixes, and just kept going until the “broken links” list was empty.

(I may have mixed up my tenses in this post. I started writing it as I was working on the migration, and finished it after I got all the link cleanup done. Sorry.)

My next task will likely be attachment cleanup. I guess that’ll go in yet another blog post!

from Evernote to Obsidian, take two

I got an email from Evernote last week letting me know that my subscription price would be increasing to $250/year, effective January 7, when my annual subscription renews. It had gone up to $130/year in 2023. Prior to that, it was $70/year.

I’m not really unhappy with Evernote as a product, or with Bending Spoons as a company, but the price on it is getting a bit ridiculous. Here are some links to articles/videos of other folks talking about the price increase:

I had previously experimented with moving to Obsidian, in 2023, but didn’t go through with it. But now, it’s looking like I really need to do something. Maybe I could afford $250/year for Evernote if I thought they were going to stick with that price for the foreseeable future, and continue making the app better. But I’m just not convinced that they’re on a good trajectory, in terms of my own personal use-case for the product.

Cory Doctorow coined the term en****tification a while back, and it’s a useful term, though I wish he had come up with something that didn’t incorporate one of George Carlin’s seven dirty words you can’t say on TV. Still, it’s a good term. Here’s a Metafilter discussion on the (let’s call it) enpoopification of note-taking software, from 2023. I’m not sure if Evernote really falls into this category, but you could make a good case for it.

So, anyway, I’m back to experimenting with Obsidian. Luckily, I’m taking this coming week off from work, to use up my vacation days for the year, so I’ve got the time for it.

I’ll probably post more about this when I get farther along, but I thought it would be useful to write up some initial notes, informed by my previous efforts, and by watching a whole bunch of YouTube videos.

I’ll start with the process of importing from Evernote. For that, I’m using Yarle. Obsidian now has an official importer for Evernote, but I think Yarle is probably still better. Looking at the commit history on Yarle in Github, I see that the author has continued to work on it. Having tried it out again, I’m not sure if it’s working better than it was in 2023, but it’s definitely working well enough, I think.

I’ve been tweaking my template for Yarle. Here’s what I’ve got at this moment:

---
{title-block}EN-Title: "{title}"{end-title-block}
{created-at-block}EN-Created: {created-at}{end-created-at-block}
{updated-at-block}EN-LastUpdated: {updated-at}{end-updated-at-block}
{source-url-block}EN-SourceURL: {source-url}{end-source-url-block}
{reminder-time-block}EN-Reminder: {reminder-time}{end-reminder-time-block}
{reminder-done-time-block}EN-Reminder-Done: {reminder-done-time}{end-reminder-done-time-block}
{tags-yaml-list-block}Tags: {tags-yaml-list}{end-tags-yaml-list-block}
---
{content-block}{content}{end-content-block}

So I’m putting a bunch of stuff in the “frontmatter” of the note. This stuff is mostly just there for reference; I won’t actually need it going forward. I plan on converting Evernote’s reminders into Obsidian tasks, either manually or via a script or something. And the tags block seems like the cleanest way to get the tags over from EN.

The only setting in Yarle that I’ve changed from the default is to set “store attachments in notebook level” to yes. I’m still not sure about the way I’m dealing with attachments, but I think this is good enough.

As for my Obsidian setup, I think I’ve settled on a batch of plugins that’ll give me most of the functionality I need (with some compromises and caveats). Here’s the list:

  • Notebook Navigator – This is a great one that allows you to get your Obsidian screen to look a lot more like what I’m used to with Evernote. Getting this installed got me past a lot of my hesitation with Obsidian.
  • Omnisearch and Text Extractor – These two should bring some decent search functionality to Obsidian.
  • Tasks – I’m going to try to replace the Evernote reminders functionality using Tasks. I know I won’t be able to do a lot of the stuff that Evernote can do (email reminders for instance), but I think I can get a workable system cobbled together.
  • Broken Links – I’m using this to identify broken links in my imported notes. I’m seeing a lot of problems there, actually. I won’t get into the details here, but I’m going to have a lot of fix-up to do, I think.

Whew. So maybe that’s enough for this blog post. I intend to spend a bunch more time working on this tomorrow.

One thing I want to do this time is to make a relatively quick decision to either switch over or not, and to go all-in on Obsidian if I do. I don’t want to dither on it, and wind up having to renew my Evernote subscription, then spending a lot of time flipping back and forth between the two systems. (I’m kind of doing that with Raindrop.io and Pinboard right now, and it’s not optimal.) Sink or swim!

old anime

I finally finished watching the Devil Lady anime series yesterday. I mentioned it in a blog post from July 2024; that’s when I started watching it. I started out watching it on DVD (from a box set I bought in 2005), then switched to watching it on Amazon, when I hit a snag with one of the DVDs. Then, it disappeared from Amazon, and I thought I’d watch the final DVD in the set, but there was a problem with that DVD too, so I found that it was streaming on Pluto, and finished watching it there. Overall, I don’t know if it was worth all the effort. But hey, that’s one more anime DVD set I can cross off my bucket list!

The next item on my list is Excel Saga. I started watching that one not long after I started Devil Lady, but only got through disc one. (Both of those shows are the kind of thing I need to be in a fairly specific mood to watch, so I’m not exactly binging either one. I’ll get in the right mood, watch a few episodes, then give up on it for a few months.)

I’ve been having trouble with DVDs a lot lately, and I’m finding that discs that don’t work on my Xbox work fine on my computer’s DVD drive. So I’ve done some DVD ripping recently (well, Dec 2024, so kinda recently), and I seem to have a workflow figured out for that. I’m using HandBrake to rip, and VLC to watch stuff on my Apple TV.

I’m ripping the Excel Saga DVDs right now. (Seemed like a reasonable thing to do on a Sunday.) Of course, I’ve also realized that all of this stuff is on Crunchyroll, which is only $8/month, so I’m not sure why I don’t just give up on the DVDs and stream this stuff.

I was just listening to an episode of Judge John Hodgman where a married couple were arguing about what to do with the husband’s DVD collection after he dies. I think maybe I should be leaning towards getting rid of my collection, but… I don’t want to.

Ubuntu and old hardware

I decided to install Ubuntu on my old Windows 10 PC yesterday. My new HP Mini PC has been working fine for a while, and I don’t think I need to keep the old one around “just in case” anymore.

I have a long history of messing around with various Linux distros, but never really sticking with Linux as my main OS, or ever really doing any meaningful work on it. (My Linux tag on this blog has entries back to 2002, and I’ve been using Linux since the 90s, when I first installed Manchester Linux from two floppy disks.)

I’m honestly not sure if I’m going to keep the old machine. It used to be that I could find someone to give my old PCs away to pretty easily, but that’s really not the case anymore. Part of that is maybe not having as many friends and family as I used to, and part of that is PCs being (relatively) cheap and ubiquitous these days. This PC is still “good”, from my perspective, and there’s plenty that could be done with it. But it’s a big tower PC and getting rid of it would allow me to simplify things on my desk a bit, and clean up some of my tangled cables.

Maybe when I’m retired, I’ll become a “tech fairy” like this guy. For now, I guess I’ll just keep the old PC where it is, and boot it up once in a while to play with Linux.

I might as well talk a little bit about the current setup process for Ubuntu and how it compares to my previous experience with Linux distros. First, I should say that it wasn’t hard at all, and I really didn’t hit a single snag. In the past, there’d usually be some issue or another with video drivers or something, but it was all very smooth. I guess that’s the result of years of work smoothing things out with these installers, and maybe also because there’s been some natural convergence over the years, where there are fewer outliers and weird edge cases.

The specific version I installed was Ubuntu 24.04.3 LTS. I’m definitely not at a point where I need the “latest and greatest.” I’m better off with the stable version.

The process was pretty smooth. I first downloaded a 6 GB ISO file, which took some time. Then I used Etcher, their recommended tool, to create a bootable USB stick. Then I booted from the stick and followed the prompts. The actual install process took less than 30 minutes.

Today’s Ubuntu desktop doesn’t look terribly different from how I remember it looking the last time I used Ubuntu on a desktop, which was probably 2015. (How time flies!) In terms of the tools that you get out of the box, with the relatively minimal install that I did, it’s mostly just Firefox and a few other random things. Firefox was enough for me to bootstrap the basic stuff that I really need.

I found that the 1Password Firefox extension works in Linux, so that was good. There seems to even be a 1Password desktop app for Linux, but I didn’t get around to trying that.

There’s no Evernote client for Linux. (Or at least not an official one.) There are a couple of unofficial clients, but I didn’t try those. The web client for Evernote works reasonably well in Firefox, so that’s good enough.

The next thing I tried was Visual Studio Code. That was downloadable from the Ubuntu “App Center.” The install was simple and easy. I didn’t really get very far with VS Code though; I just checked it out to make sure it worked.

If I was serious about doing anything meaningful with Ubuntu, I’d have to do a lot more work figuring things out, but it’s nice to know that I’ve made a start with it, and I could go further if I wanted to.

figuring out how to consume news

Here I am, writing another blog post about media consumption. Well, why not? It’s weird out there, and I’m just trying to keep my head on straight.

I’m still thinking about signing up for Apple News+, which I’d likely do by upgrading from Apple One Individual to Premier. So that’s be an increase from $20/month to $38. I also pay $1/month for extra iCloud storage, so I’d be able to cancel that, so the total increase in cost would be $17.

Apple News+ alone would be $13/month, so if I’m going to do it, I might as well go with Apple One Premier and get the 2 TB of iCloud and Apple Fitness+ too.

I made a list of pros & cons for Apple News+. On the “pro” side, it gets me access to a bunch of news sites all under one subscription. There’s lots of variety, and it includes local news from a few NJ sites.

On the “con” side:

  • It doesn’t work on Windows. (Or directly through a web browser.)
  • It doesn’t seem to work with Instapaper. (I’ve read some mixed reports about how it works with read-it-later services, but it generally doesn’t seem to work.)
  • There’s no way to send an article to my Kindle or Kobo to read on an eInk device.
  • The service has ads, and they’re not easily blocked.

I’ve also been considering a “why not both?” option, where I use Apple News+ for some stuff, but continue subscribing to individual publications, where it makes sense to do so.

On a related topic, I’ve been playing around with my RSS tools recently. Right now, I use The Old Reader as my “back-end” service. For a front-end, on Windows, I generally just use The Old Reader web site. On iPhone and iPad, I use Reeder Classic. On my Mac, I sometimes use Reeder and sometimes the web site.

I’ve been trying out NetNewsWire as a replacement for Reeder. I think I may switch over to it. The design works better for me. It’s a pretty vanilla design, but that’s what I want. It’s black text on a white background, easy to read. Reeder is black on a kind of off-white background, which is harder for me.

On the back-end, I’m kind of interested in looking at stuff that does a bit more with the feeds. I might want to try a product that lets me send email newsletters into it, so I can get those out of my email. I’m not really sure if I need or want that though. I also like the idea of being about to do something interesting with the data, like putting together a custom “front page” for me. I haven’t gotten very far with any of that.

Of course, if I start using Apple News+, I might spend less time with my RSS feeds.

One more topic: Unfortunately, it looks like NJ PBS is shutting down next year. (Here’s an article about it from CBS News.) Hopefully, NJ Spotlight News will continue. (They say they will, but who knows?) The way things are going right now, it feels like a lot of good things are going to disappear in the near future. (And, of course, some already have.)

Well, that’s a downer to go out on. I wanted to find a quick positive thing to put at the end here, but I couldn’t come up with anything. Sigh.

messing around with Firefox and Vivaldi

My usual browser setup, for a while now, has been to use Firefox on desktop and Safari on mobile. (I have my bookmarks syncing between Firefox and Safari with iCloud for Windows.) That’s worked out pretty well, and I really have no complaints. But, of course, I’m a nerd, so I have to mess around with things once in a while, even if they work well.

So my first experiment was to see if I could switch to Firefox on mobile, so I’d be using Firefox everywhere. That’s been working well, though there are some tradeoffs. Apple, of course, puts every browser on iOS except Safari at a disadvantage, in several ways. The most obvious one is content blockers, which only work with Safari. I use 1Blocker on iOS, and that works reasonably well, but not nearly as well as uBlock Origin on desktop. With Firefox on iOS, I guess that their Enhanced Tracking Protection helps a bit, but it would be nice to have real ad blocking.

My next experiment has been to see if I could switch from Firefox to Vivaldi. I haven’t gotten too far with that, and I don’t know if I’m going to stick with it. I have Vivaldi set up on all of my main devices now: Windows desktop, MacBook Air, iPhone, and iPad. Vivaldi does do ad-blocking on iOS, since it’s got built-in ad-blocking. I’m not sure if it’s all that great though. I need to use it some more and get a feel for it.

One issue I’ve been having with Vivaldi is their settings sync. In theory, all of the settings should sync between all the installs of the browser, but that hasn’t worked out perfectly. The ad-blocking exception list doesn’t seem to sync at all. Another one is that my search setup doesn’t sync correctly. There seems to be a workaround, but it’s not perfect.

Speaking of search, I’ve been sticking with Kagi. I first started using it earlier this year, and I’m liking it. Of course, setting it up in a new browser can be a hassle, since you need to be logged in to use it, and since it’s not on the list of default search engines in any major browser.

So, overall: I’m going to continue messing around with Vivaldi a bit, but I might give up on it. If I do, I don’t know if I’ll go back to using Firefox everywhere, or my old system of using Firefox on desktop and Safari on mobile.