markjgsmith

Everything feed operational

In the past few days I've completed a refactoring of all the website's RSS feeds, and I've also added an everything feed. It's part of the bigger redesign of the main page, which currently shows the latest posts from the linkblog, blog, podcast, newsletter and notes. The idea behind the redesign is to move away from each post type in it's own section because it feels disjointed, given that each post type has a different frequency. The redesigned main page will have all the posts interleaved as they occur.

I've been mostly writing notes and links lately and blog posts seem to have fallen by the wayside. Having everything interleaved will hopefully restore some balance. This post is the start of that. Many of the recent notes posts could really have been blog posts. They are quite long and almost all of them have titles. The reason they are ending up as notes is that creating notes is slightly easier, but I think it might also be because the latest page doesn't give you a very good at-a-glance idea of what's being posted. And that is I suppose a bit ironic given that it's called the latest page.

First was to fix how the posts are displayed. Hopefully if the posts are interleaved then the balance of post types is on display so I'll be more conscious of what I'm publishing, and naturally tend to publish a more balanced collection of posts. I started with the RSS feeds, but I'm also working on an interleaved website main page.

Second is to make it easier to post blog posts. The biggest hurdle was that previously the RSS feeds were not full text. Instead they only included the description. That was because I didn't think code snippets would be formatted correctly, so I figured it was best to require readers to click through to the website. Turns out that code snippet formatting is fine, though syntax hightlighting is lost, which is a bit annoying. So the blog post descriptions are no longer necessary. That's already a bit easier.

Ideally I'd like to find a way to create new blog post files from a template. I haven't found a way to do that yet. I looked into some iOS apps that give you ways to create text blocks from snippet templates but the ones I found either didn't do it in a way that I could integrate into my workflow, or they required too many security grants, akin to installing a key logger, which I'm not comfortable with. I need to be able to create a markdown file with frontmatter, both the file contents and filename should be templateable. If you have any good suggestions, please get in touch.

Anyhow that about wraps it up for this blog post. Hopefully there will be more of these going forward. #

BBC News broken - In the past few weeks many of the BBC News' websites pages have stopped being readable in offline reading apps. In Pocket for example instead of the article you save, you end up with some sort of copyright notice. This has been working for years. What changed? #

Taking someone's life === murder - I've been saying this for a while now. I know I sound crazy when I say stuff like that but I've thought about it a lot, and it's the sort of thing that though you don't have to contemplate it a lot, it absolutely can become very important. Anyway I felt a bit validated yesterday after listening to Robert Breedlove on the Bankless Podcast.

"If I take your life, that’s equivalent to murdering you, that’s to take your future freedom" - Robert Breedlove on life, liberty, and property

It's a great interview exploring the history of money, in order to lay out the critical design choices that were made when designing Bitcoin. There's an absolute ton of incredible background info about everything related to the monetary system in this episode. #

NYT vs OpenAI oddity

I've been listening to various podcasts recently that cover the goings on in the publishing industry vis-a-vis how they are handling the new dynamics that have been introduced by large language models (LLMs). Publishers aren't too happy that their content is being used to train AI models, and the models are then steeling their customers.

The New York Times is sueing OpenAI for copyright infringement. It's an interesting case. Here's OpenAI's public response to the accusations. I've been reading it every few days, there was something about it that seemed very familiar. It's been really bugging me. Well it took listening to somebody else reading it out to figure out what my brain was matching.

It's weird but the way they describe the tactics used by the NYT to get the model to do unnexpected things, well that's what it feels like living in Vietnam as a foreigner. People here are constantly trying to hack you, constantly. I can't emphasize this enough, it's literally every interaction, everywhere I go, all the time. The same groups of people show up at the same places each time, and they seem to try out new tricks / hacks they have created or learnt. It gets very involved, happens across multiple locations. It's mind bending, enough to send you barmy.

You do sort of get used to it after a while but it pretty much destroys your mood, makes it super hard to concentrate, causes confusion, often ends up escallating. Anyway, just wanted to put that out there, the NYT's described behaviour really reminds me of many of the people I meet or pass by here in Vietnam.

Is this what the whole world is turning into? I hope not, it's rarely very fun, and often unpleasant. It feels a lot like bullying, and unfortuntely relatively often results in violence or injury or both. #