Terry Hancock @TerryHancock@mastodon.art

One technical feature I miss from the proprietary social-media platforms is natural-language translation. It has limitations, of course, but I'd often hit the "translate this" button on Facebook.

Here, I just have to guess or miss whatever was being said, if I'm not familiar with the language.

I wonder if there's a feasible way to add that feature to a Fediverse instance?

Stuff piles up and compounds.

I've been trying to write reports, take inventory, etc on 2021. And then I run out of room on hard drives, and realize I have to do some file-management. And THAT takes forever.

Now I'm running out of January!

Ars Technica ran a truly horrendous piece of irresponsible journalism about the "disaster" of a space elevator "falling", worthy of a Roland Emmerich disaster sci-fi flick.

Very disappointing.

So, in the interest of countering disinformation, here's a responsible article about the real failure modes of a space elevator:

https://www.zmescience.com/science/physics/broken-space-elevator/

The most embarrassing bit is that the Trac site is extremely user-hostile. Modern browsers really don't like self-signed sites, and I never got it converted over to use a trusted certificate (yes, I know there's a free one, I haven't configured it yet). Also, I'm using Apache for VCS security, when it really should be the VCS doing that.

Also, Subversion is old as dirt, and no one uses it anymore, except a few old die-hards like me.

Makes collaboration awkward.

A lot of what I'm doing now is assessing my current "virtual assets", which mostly means the web pages/sites/platforms, etc.

The project site has a hub built with Wordpress, with other stuff added on, mostly on one "snowflake" server, with some 3rd party platform services.

The left side is the actual "project" stuff. The right is my old solution for fan/series content, that I've since replaced with a static site -- I need to remove it.

I'm planning to change a lot of this this year.

I've moved on to writing my annual reports in RST now, and then I can include them directly in my Pelican-based local HTML site. I'm using them directly from the file-system, rather than through the Pelican test webserver.

I think this is for the best, because I'll more likely use them that way -- so anything that doesn't work from the file system should be avoided, anyway. So far, the only issue that raises is that I need to explicitly link to index.html, rather than the folder.

Works.

Reviewing all my websites.
ALLL my website. There are too damned many of them!

Also, a lot of bitrot. And stuff that's just not true -- some "not true anymore" and some "this was always wishful thinking".

I have a tendency to write up documentation for what I WANT to build, and not what I actually have built.

Also a few "I thought there was a website there, but I guess not!" 404s or 5xx or somebody else's site because my domain was misconfigured and that IP address is no longer mine. 🤦

So, I descoped and just did the most important articles, but I went through and copy-edited them first. I did indeed find lots of invisible trash that was causing the migration script to choke.

But then I found all kinds of issues with the reStructured Text (RST) format, and learned a lot, but also spent a lot of time on it.

I did finally create a fully-self-contained offline site, though, collecting my monthly summary articles for 2021.

But was it worth the cost? IDK...

Ooh, except only about 63% of my posts got processed. The other 37% were dropped because of markup errors, which I suppose probably derive from bad formatting in my posts.

A lot of them were script-migrated from other sources, so errors will have accumulated. And some, I copy-pasted from Patreon and then edited, which is another source of errors. I was hoping that WP would fix those mistakes by closing tags, etc. But maybe not?

A considerable editing project to do by hand!