See also: baty.net
I’ve resurrected baty.net, so what happens to this blog? Nothing. It’s not going anywhere, but I’m looking at other was to make use of it. Blot is great at doing stuff with images, html files, etc. Maybe I’ll make it a photo blog. Or maybe it’ll become about a specific topic.
While I think through this, you’ll probably notice more new posts at baty.net.
Org-download and ox-hugo
One drawback of using a static blogging tool is the relative difficulty of dealing with images. Images must be copied to the proper folder and then links need to be added manually. I’ve created a few ways to make this easier this using TextExpander, but today I discovered org-download.
This extension facilitates moving images from point A to point B.
I’m using this with ox-hugo and it works very well. I first added the following to my Emacs config…
(setq org-download-timestamp "-%Y-%m-%d"))
Then, at the top of my blog file (all-posts.org), I set the following buffer-specific variable…
-*- mode: Org; org-download-image-dir: "~/microblog/static/img"; -*-
Now, to add an image to a post, I just drag it from the finder into Emacs and the image is copied to my images folder (
microblog/static/image) and a link is inserted into the buffer with the proper path, etc. The image is even shown inline by default so I can see what I’m dealing with. What a treat!
The post looks like this after dragging an image from the Finder…
*** My New Post with an image
#+DOWNLOADED: file:/Users/jbaty/Desktop/test.png @ 2019-07-02 16:44:58
The date stamp can be appended during the copy and is configured with
It’s pretty slick, and takes most of the hassle out of adding images to blog posts when using Hugo/ox-hugo.
Hosting infrastructure changes this weekend
I’ve made a number of fun changes to how things are run around here, server-wise. The goal was to move as much as possible back to my self-hosted EC2 instance and remove some 3rd-party services while I was at it.
Here’s what happened.
I’ve moved the baty.net archive from Github/Netlify to a static site on my server. Netlify is awesome but I’m capable of managing my own server, SSL, etc. I lose the CI and CDN portions, but those right now are less important to me than keeping things together and minimizing moving parts. Plus, I get access to the web server logs this way.
I’ve replaced 3rd-party analytics with GoAccess. It’s not as easy to spot normal visitor traffic, but I do get stats on just about everything else, without dealing with a tracking script. As long as I have access to the web server logs, GoAccess should be fine.
I’ve moved my microblog (micro.baty.net) to a Hugo-based static site, also hosted on my EC2 instance. To make posting easier, I’ve set up ox-hugo so every post is just a single headline in one big org-mode file. To publish updates, I’ve created a small Makefile and simply run
make deploy, which rsyncs everything up to the server.
I’ve started moving my private Git repos to my own gitea instance. Super lightweight and private. To upgrade, I just replace a single binary and restart the service.
All of the above are served using the Caddy web server instead of nginx. Caddy does all sorts of nice stuff right out of the box (e.g. SSL and pretty directory listings) and configuration couldn’t be simpler.
I have a few cleanup tasks left and some automation to build but it’s quite fun having everything under my own roof and tinkering in whatever ways I see fit.
Question: what is to be done with the stuff after it has been cataloged and stored? Are we pinning butterflies for the sake of pinning them, or is there a moment of beholding, and re-use/re-mix down the line?
Save and make? Transform?
This raises a great question: Which butterflies to pin?
I’m not one to think that we need to carefully archive everything, but there are many seemingly useless things that should be saved but aren’t. One never knows what will prove valuable over time. Or what might be re-mixed in the future. I fall into the better-safe-than-sorry camp, but determining what to archive and what to just let go can be a crazy-making problem.
I don’t even resemble an expert in cryptocurrency, but my gut says the whole thing is some sort of mass delusion. I mean, read Twitter after any price fluctuation (up or down), and it’s wall-to-wall rationalization.
These people are just so deeply enamoured with the idea of crypto that they seem to have lost all sense of reason. I don’t want to call them crazy, but I kind of do.
Anyway, I own a little Bitcoin, Etherium, Litecoin, and one other that I don’t remember. I don’t spend more on it than I can afford to lose, but I admit to hedging my bets with the upside potential.
It’s sort of a Pascal’s wager in that if I’m wrong, I still win.
It’s just that my timing is, as always, terrible.
Facebook, Libra, and the Long Game (Ben Thompson)
Ben Thompson, Stratechery:
Here is the important thing to understand about the Libra Association: while its members — who again, are the validators — do control the Libra protocol, Facebook does not control the validators. Which, by extension, means that Facebook will not control Libra.
It’s always good to control ones biases, but it doesn’t come easy for me when dealing with Facebook. Ben makes the case that since Libra is not a “Facebook Coin”, but rather a Facebook initiative, the tradeoffs between trust and efficiency just may enable it to succeed.
Certainly Facebook’s audacity and ambition should not be underestimated, and the company’s network is the biggest reason to believe Libra will work; Facebook’s brand is the biggest reason to believe it will not.
But there’s that.
Meta: No longer automatically cross-posting
I’ve disabled cross-posting from baty.blog to Micro.blog/Twitter/Mastodon. I like not worrying about force-feeding every single thing I publish to other feeds. If I post something I want to share more widely, I’ll post links directly.
This feeling started with my wiki. I just write stuff there without worrying about where it’s going to “go”.
For those few who truly want to read everything, there’s RSS
My Dad and I
Happy Father’s Day, dad!
Please don’t cancel Keanu (The Outline)
Please do not cancel Keanu Reeves. Please. Keanu Reeves is the closet thing we have to Mr. Rogers and we already don’t deserve him as it is.
I’m not sure I’d go that far, but I’ve always rooted for Keanu and it’s nice seeing him getting so much good press lately.
The Dropbox updates are fine
I keep a shit-ton of stuff in Dropbox. I’ve been using it with nary a hitch for many years. Dropbox was the first syncing option that didn’t completely suck, and it’s still better than the rest (I also use iCloud and Syncthing).
Sure, sometimes the Dropbox app uses too much CPU. On the other hand, sometimes iCloud loses data. And no one enjoys configuring Syncthing. Point is, Dropbox is the closest thing I’ve ever found to hassle-free sync across everything. The recently added Smart Sync is awesome. Add to that, Paper, which is quite nice, and darn good sharing options and I still feel like Dropbox is a winner.
The new Dropbox adds a bunch of stuff that, at first glance, made me think, “Ah, hooey, what’s this Evernote-level bullshit, now?!”
Then, I used the new desktop app all day today and am already finding the changes useful. It’s too early to be sure, or to review, but I think we all knew that Dropbox had to do something, and I think they may just be figuring out what that something is. If you don’t care about the new stuff, just keep using it the way you always have.
For a different take, Michael Tsai is collecting some of the more snarky knee-jerk reactions.
Ben Thompson (Stratechery):
I find this tremendously exciting, and sorely needed. For years I have been wondering which company will build the “operating system of the cloud”, and this seems like a very credible attempt to do just that. The new Dropbox app is basically a new version of the Finder or Explorer, with communication and collaboration built-in.
To that end, Dropbox will never achieve the same level of integration, given it has to partner with other companies, but it doesn’t need to: the idea is to have good-enough integration so that all of the apps it is integrating with can win on their own merits. In other words, I would go further than Houston: the new Dropbox isn’t simply complementary to a product like Slack in particular, it promises to make Slack a much better product in its own right, particularly when viewed from that higher, more wholistic level that Microsoft has focused on.
It remains to be seen how this plays out. It’s easy to succumb to the usual knee-jerk cynicism, but I’m not ready to dismiss it quite yet.
Bug Fixes (Paul Ford)
Paul Ford, NYT Magazine:
But in the end, the software works or it doesn’t. Politics, our personal health, our careers or lives in general — these do not provide a narrative of unalloyed progress. But software, dammit, can and does. It’s a pleasure to watch the code change and improve, and it’s also fascinating to see big companies, paid programmers and volunteers learning to work together (the Defense Department is way into open source) to make those changes and improvements. I read the change logs, and I think: Humans can do things.
Daily software stack as of June 06, 2019
Because things change, I’ve jotted down the list of apps I use to manage my “stuff” daily as of today. Below each is a quick list of the types of things for which it’s used. At some point I’d like to go into more detail, but by the time I do that everything changes again.
The list is longer than I would like, but I can’t seem to narrow it down further. The highlight here is that I’m unable to get rid of Emacs, so it remains an important part of my process. I fear that it always will, and that I’ll have to get good at it, eventually.
- Task management via org-mode
- Long-form journaling via org-journal
- Time tracking
- Document production
- Too much of my spare time
- The Control Center: Connects everything to everything
- Project management
- Personal and work event timelines
- Links to interesting things and to their concepts
- Project notes
- Visual/Concept mapping
- My note-taking “inbox”
- Blog posts, tech notes, and ideas percolate here until ready to be moved somewhere more permanent
- For finished notes that I may want to share or access quickly on the go
- All sorts of wierd text processing
- Occasionally as a writing environment
- Markdown writing. Mostly blog posts. I just really like writing here.
I got back into Twitter about a month ago. I thought I missed it, but it turns out I only missed small portions of it, and those portions have been drowned out by pessimism, hyperbole, hate, and self-interest.
Here’s how I tried making Twitter into a pleasant experience again:
- Unfollowed everyone and started over.
- Disabled retweets
- Muted lots and lots of keywords.
- Tried having a good attitude
And still, my feed was awash with the usual awful, depressing, tedious noise of people determining that “Something is BAD!!!” and yelling incessantly about it without a single rational suggestion about how to make it less bad.
There just isn’t enough signal to go with the noise. I’ve had to stop visiting Twitter again.
Dark Mode Isn’t ‘Easier on the Eyes’ for Everybody (Vice)
Samantha Cole, Vice
In the end, more display options are better, and people should use whichever lighting theme they want. It’s great that dark mode is coming to iOS for people who it helps, but there’s simply not evidence to make the blanket claim that dark mode is “easier on your eyes.”
Vice adds to the recent spate of articles confirming my biases against some of the widely-touted benefits of Dark Mode.
How did journalists file before Google Docs? (CJR)
Amanda Darrach, CJR:
By the summer of 1863, competition was fierce. A New York Tribune reporter was about 10 miles from Gettysburg, trying to cover a cavalry raid, when the battle opened. The town’s telegraph operator told him the wires had been cut. “The Trib’s man gathered up a work crew, rented a handcar from the president of the railroad, and took off to find the break and repair it,” Tucher says. “In return, he demanded that the telegraph operator not let anyone else but him use the wire, and sent off a scoop.”
Apple’s Audacity (Ben Thompson)
Ben Thompson, Stratechery:
it was fun seeing what Apple came up with in its attempt to build the most powerful Mac ever, in the same way it is fun to read about supercars. More importantly, I thought that sense of “going for it” that characterized the Mac Pro permeated the entire keynote: Apple seemed more sure of itself and, consequentially, more audacious than it has in several years.
“Audacious” is a good word for it.
Apple…emphasized privacy at every turn, and did so with passion: it felt like the fight for privacy has given the entire company a new sense of purpose, and that is invaluable.
In short, it is clear that privacy has become more than a Strategy Credit for Apple. It is a driving force behind the company’s decisions, both in terms of product features and also strategy.
Cynics scoff, but I believe that Apple’s push for privacy is the right thing and a good thing.
The Mac Pro and display fill a large gap in Apple’s professional product line. Prices are reasonable for the professional user. If you want one at home, you are an idiot.
It’s amazing to see so many people complaining about the cost of something they have no real use for and could never fully utilize even if they think they could.
I am idiotic enough to want one at home, but not enough to actually buy one.
I should just get really good at Emacs
Last week during a hectic couple of days at the office, I dropped out of Emacs/Mu4e/Org/etc. and used my “old” apps instead. I didn’t have time to figure out how to best search for files in Projectile or why mbsync is being so slow or how to easily read multiple emails at once in Mu4e. My usual apps had me covered. I didn’t have time to look up the best way to do a fancy find-and-replace of a large text file in Emacs. I already know how to do that in BBEdit.
It occurred to me that when I have the time and am not feeling lazy, using Emacs for things like email and task management is superior. Superior, but harder. When something’s urgent, I don’t have time to figure everything out right then, and I tell myself that dammit, I shouldn’t have to! So, I ditch Org-mode and Mu4e and most of Emacs and go back to Things or OmniFocus and Mail.app or Mailmate and BBEdit and everything gets easier.
Trouble is, I don’t think it gets better. The problem is simply that I’m not good enough at Emacs. I’ve changed the way I use Emacs so often that, even though I’ve used it for years, I haven’t had time to get really good at it.
First it was Spacemacs, then Doom Emacs, then I rolled my own, then back to Spacemacs, and now, finally, back to rolling my own. Each of the “starter kits” does everything differently, meaning muscle memory isn’t helpful since it doesn’t work once I switch everything around…again.
I have a theory that this would not be a problem if I focused and spent the time to get better at using Emacs. By Emacs I mean vanilla Emacs with a few hand-crafted customizations. No one will change key bindings out from under me or introduce a behavior I didn’t expect.
I will simply need to bury myself in it, learn the native keybindings, tweak what annoys me, and improve my skills through repetition and study.
For starters, I probably shouldn’t be writing this in Typora 😣.
Moving to Startpage for search
As a user and fan of DuckDuckGo for several years, I am a little disappointed to say that I’m switching my default search engine to startpage.com.
This change has felt inevitable for a few months now. Several times each day I have to re-run my DuckDuckGo search using Startpage because DDG fails to find something I’m sure should be somewhere on the first page of results.
Today, for example, I was linking to Maciej Cegłowski’s article about securing congressional campaigns. I can never remember how to spell his name, so I typed “idle words pinboard author” into DDG and this is what I got back…
That wasn’t helpful, so I ran the same search using Startpage and got this…
See what I mean? And this isn’t a rare exception. I want to support non-Google and privacy-focused alternatives when it makes sense. I rely on search all day long and need better results than I’ve been getting from DuckDuckGo, so for now it’s going to be Startpage
View the archives