Someday everyone will have a blogroll like this, and a blog. #
What Frontier is about — from 2021. #
Hypothetically, if someone were building a headless version of Frontier with help from ChatGPT, they might be thinking about how to build a web app that was a really good shell to start with. If such a person asked what I thought, I would say without a doubt that would be Drummer. The reason is it was designed for that purpose, because I needed many of the features of Frontier for my work environment, and as you know Frontier was built around the outliner, object database and verb set, back in 1988. Anyway, I also wrote a doc in April of 2020, that went through all the features of Drummer as a scripting environment. It's as if it were designed for such a hypothetical project. I wrote a new intro from the point of view of 2025. It might be interesting to write another intro in 2030, knock wood, Praise Murphy, I am not a lawyer and (as far as I know) my mother loves me. #
I love getting comments like this. The issue I posted about yesterday turns out to be a bug, and apparently it's going to be fixed. So I can go ahead without worrying about a workaround. This is the best outcome. Usually with most vendors if there's a bug, they don't acknowledge, and then they might get around to looking at it someday. This is what I call working together. Scott Hanson found the thread where they were discussing the feature in question, starting in 2011. I added an update from 2025. This stuff feels like time travel, and it's also incredibly reassuring to find that solutions to issues that were relevant 14 years ago are still relevant today. That's the stability that platforms require in order for developers to build with confidence. This is something that many big tech companies (cough Apple cough) either don't understand or don't care about. #
Okay this is another test post on the daveverse site. Turns out it was a good thing to post the last one because it revealed a bug, apparently, in wordpress.com? No kidding. I understand how that goes, I'm now running a server that's much more complex than anything I've tried to manage before. It runs all the time, and does its thing, but there are definitely bugs in there.
Anyway, I'm working on discourse again, after blowing my head up a few times getting very mixed signals from the work with id's emanating from RSS feeds.
So this test will let us see, in the debugger, exactly what data is coming from WordLand about a post that is being replied to. It's supposed to get the URL of the post, and the id's of the site and post. From there we will be able to notify the author of the post being replied to that there is a reply, and where they can read it. That will be a URL and id's for the reply post. This email will let them read it, and decide what if anything to do. That's the next plateau I want to get to.
Result of test: this is what's available inReplyTo as part of the draft object wpIdentity gets.
Updates:
- I'm testing this just by updating! Being very careful not to create too many test posts. This is already a somewhat real site.
- The next day and I'm doing another update. I think I'm going to have to create a new post, running out of room on screen I think. This is something I have to fix in WordLand.,