Stephen Ramsay
@sramsay2.bsky.social
Professor of English and Fellow at the Center for Digital Research in the Humanities at the University of Nebraska-Lincoln. Mainly #DigitalHumanities. Blog at https://stephenramsay.net/
Do they have a paper out yet on this? This is really interesting problem.
December 11, 2025 at 5:21 PM
Do they have a paper out yet on this? This is really interesting problem.
This is glorious.
December 10, 2025 at 11:24 PM
This is glorious.
I just feel like after their first album, they lost that true outsider lofi groove, you know? The minute they signed with Monterey Bay, it just got all “corporate aquarium.”
December 10, 2025 at 11:22 PM
I just feel like after their first album, they lost that true outsider lofi groove, you know? The minute they signed with Monterey Bay, it just got all “corporate aquarium.”
I'm probably being a *little* shrill here, but I've never been quite comfortable with saxon being the only game in town. And when I compare the XML ecosystem to that of JSON. Or Markdown. Or CSV, frankly!
December 10, 2025 at 5:59 PM
I'm probably being a *little* shrill here, but I've never been quite comfortable with saxon being the only game in town. And when I compare the XML ecosystem to that of JSON. Or Markdown. Or CSV, frankly!
I played with SaxonC years ago, and liked the speed. Is there an actual Python module that talks to the SaxonC API, or are you using Python’s own ffi?
December 10, 2025 at 3:21 PM
I played with SaxonC years ago, and liked the speed. Is there an actual Python module that talks to the SaxonC API, or are you using Python’s own ffi?
We thought so, so many things. And many of them were really *good* things!
December 10, 2025 at 2:52 AM
We thought so, so many things. And many of them were really *good* things!
Then you’ll understand when I say that I am not James Clark. No one is James Clark. I think I’m a damn good coder, and I am not in that guy’s league. So this is no slight against any of the people who built all this stuff.
December 10, 2025 at 12:33 AM
Then you’ll understand when I say that I am not James Clark. No one is James Clark. I think I’m a damn good coder, and I am not in that guy’s league. So this is no slight against any of the people who built all this stuff.
I want to be that guy!!! Okay, maybe not. But you know what I mean. I'm random, I'm in Nebraska, I have a weird infrastructure fetish . . .
December 9, 2025 at 11:45 PM
I want to be that guy!!! Okay, maybe not. But you know what I mean. I'm random, I'm in Nebraska, I have a weird infrastructure fetish . . .
Amy, as I live an breathe! :)
December 9, 2025 at 11:35 PM
Amy, as I live an breathe! :)
And I'm all in on this, btw. I love writing parsers. I think RelaxNG is cool. Some of my best friends are TEI taggers. And I'm in the mood to build something useful. And yes, this should be a blog post. Apologies to people for whom this is the dullest subject on earth.
December 9, 2025 at 10:36 PM
And I'm all in on this, btw. I love writing parsers. I think RelaxNG is cool. Some of my best friends are TEI taggers. And I'm in the mood to build something useful. And yes, this should be a blog post. Apologies to people for whom this is the dullest subject on earth.
can have their bag of words. I think TEI provides some pretty amazing things once AI is in the picture. But I think we have to address the aging stockpile. We really need to build some really great XML tools, and then maintain them.
December 9, 2025 at 10:36 PM
can have their bag of words. I think TEI provides some pretty amazing things once AI is in the picture. But I think we have to address the aging stockpile. We really need to build some really great XML tools, and then maintain them.
But here's the thing. TEI has succeeded brilliantly. it really has. In fact, it is without any doubt one of DH's signal technical achievements. And I don't think it's true (anymore) that TEI (as many have quipped) provides a convenient way for text analysis folks to strip away all the noise so they
December 9, 2025 at 10:36 PM
But here's the thing. TEI has succeeded brilliantly. it really has. In fact, it is without any doubt one of DH's signal technical achievements. And I don't think it's true (anymore) that TEI (as many have quipped) provides a convenient way for text analysis folks to strip away all the noise so they
You can tell because LIBXML2 IS UNMAINTAINED. Because tons of Java-based XML gizmos are basically dead. Cocoon? I have a BOOK on Cocoon somewhere around here. It is dead and gone.
December 9, 2025 at 10:36 PM
You can tell because LIBXML2 IS UNMAINTAINED. Because tons of Java-based XML gizmos are basically dead. Cocoon? I have a BOOK on Cocoon somewhere around here. It is dead and gone.
And it now officially has *no maintainer.* libxml2. I remember the heady days when XML was new. I've watched the whole thing from the days SGML. So let me say the quiet part out loud. XML really didn't win. "Well, but . . ." I know. I'm still right. And you know how you can tell?
December 9, 2025 at 10:36 PM
And it now officially has *no maintainer.* libxml2. I remember the heady days when XML was new. I've watched the whole thing from the days SGML. So let me say the quiet part out loud. XML really didn't win. "Well, but . . ." I know. I'm still right. And you know how you can tell?
And then I read this today about libxml2 (in their README): "This is open-source software written by hobbyists, maintained by a single volunteer, badly tested, written in a memory-unsafe language and full of security bugs. It is foolish to use this software to process untrusted data."
December 9, 2025 at 10:36 PM
And then I read this today about libxml2 (in their README): "This is open-source software written by hobbyists, maintained by a single volunteer, badly tested, written in a memory-unsafe language and full of security bugs. It is foolish to use this software to process untrusted data."
Congratulations, Brett! Very, very well deserved.
December 9, 2025 at 5:01 PM
Congratulations, Brett! Very, very well deserved.
Ha! I want you in charge of image accessibility for everything from now on. That is marvelous.
December 8, 2025 at 7:21 PM
Ha! I want you in charge of image accessibility for everything from now on. That is marvelous.
Which is just to say that construing the purpose of citation as allowing another person to consult the source of your words or ideas is not one to which the MLA adheres in all circumstances. It also tells you how to cite a personal email (with no expectation that you can yourself read that email).
December 4, 2025 at 11:06 PM
Which is just to say that construing the purpose of citation as allowing another person to consult the source of your words or ideas is not one to which the MLA adheres in all circumstances. It also tells you how to cite a personal email (with no expectation that you can yourself read that email).
It does seem wrong to me, but I believe citation guidelines (including MLA) have long given instructions for how to cite a "personal communication" (like a conversation) for which there is no written transcript.
December 4, 2025 at 11:02 PM
It does seem wrong to me, but I believe citation guidelines (including MLA) have long given instructions for how to cite a "personal communication" (like a conversation) for which there is no written transcript.
Yes, that is correct. That is the answer right now. I guess what I'm pointing to is the "ergonomic disconnect." We are vibe coding with languages that were not designed for that way of coding, which makes me wonder what a VOPL would look like. And could you build a VOPL on top of what we have?
December 3, 2025 at 5:49 PM
Yes, that is correct. That is the answer right now. I guess what I'm pointing to is the "ergonomic disconnect." We are vibe coding with languages that were not designed for that way of coding, which makes me wonder what a VOPL would look like. And could you build a VOPL on top of what we have?