So, one of the things I’ve been doing behind the scenes, a few posts at a time, is restoring posts and comments that ended up with garbled text. Something was set up incorrectly with this blog’s database, and I had a bunch of UTF-8 text stored in the database, except the database was set to latin-1 encoding. This worked, until it didn’t.
So anyway, I hired someone to fix it, and he was able to repair most of the database, but there were a couple of hundred comments and posts that had garbled text even despite the fix. (Presumably they were garbled even before I’d started using UTF-8.) Anyway, they mostly were displaying okay on the Wayback Machine archives of my site, so I’ve been restoring that text, a few comments and posts at a time, since the fix last month. The restoration is not 100% complete: there’s at least one user who username changed to “Anonymous” at some point, and I didn’t know whether that happened at the user’s discretion or as a result of the database problems. (I took a guess and assumed it was the latter.)
I’m a little over halfway done. I suppose I could have just ignored the messed-up posts and comments, but it’s not really in my nature. Anyway, I should be finished after just a few more sessions of restoring text, and then I’ll be finished yet another ongoing project.
And then… well, don’t get excited, but I think my newest timesink will be returning to reading and blogging about Ezra Pound‘s Cantos. I’m on the Pisan Cantos, which are famously beautiful and challenging, and I’d like to get the series going again. How I’ll do it, given my incoming course load, is another question, but surely I can handle one poem a week, right? Right? (We’ll see.) Hell, I’m also tempted to finish the series just to see if I can interest a publisher in an unorthodox “read-along” book about the Cantos. You know, a book you can sort of have at your desk, and then read it alongside the poems? At the very least, I know people have been using the blod posts for that purpose, and I feel like I’ve let them (and myself) down in having set the project aside for so long, even if I had plenty of good reason to do so.
That’s not a promise—my wife and I have some pretty sizeable translation commitments in the next couple of months, and I was hoping to put some work into revising novels as well—but it is where my head’s at right now.
UPDATE (31 August 2026): I’m about 75% of the way to finished with this. I had to delete one post that couldn’t be recovered, and leave one comment and one post incompletely recovered, but that’s it so far. Pretty good, as far as data erosion goes. However, I’m also realizing that whatever the encoding problems were, they’ve been around for a long time: over on the Wayback Machine, I see captures going as far back as 2007 that contain some garbled text. (Some garbling.) In fact, most of it seems to date between 202 and 2011, so I’m guessing it was around that time that something changed about the way I was inputting Korean and hanja characters into the blog. Anyway, about fifty posts left to recover, which is maybe an hour’s work. I expect to finish it by the end of the week.