4 ms·

The issue is not really specific to this pair of LaTeX styles, in my experience. I'm not sure why you call the style broken: the final result looks fine, and D

by svat 6y ago

The issue is not really specific to this pair of LaTeX styles, in my experience. I'm not sure why you call the style broken: the final result looks fine, and DJB didn't have to fix any errors in the style itself, only in his document. (Also, the author of the style is a translator and co-author of IMO the best introductory book on TeX—comes highly recommended by Hans Hagen in the ConTeXt manual too, as “the book that turns every beginner into an expert”—though of course LaTeX conventions are a different beast and it's possible that the expertise doesn't carry over.) Rather, my experience is that for any nontrivial document, simply changing the LaTeX style (\documentclass) from one to another (like amsart to msripub here, or just one journal's house style to another independent one) is unlikely to be enough; the amount of work required is likely to be much more than the indicated “two minutes”. So what that story points out is that one of the promises of LaTeX (that particular one, of being able to change styles easily) doesn't hold up too well in practice (not many people try it), and the fact that this page starkly shows the typical results one would get on trying is what makes it sting IMO. Or at least so I imagine (I'm not much of a LaTeX fan, more of a plain TeX one).

leephillips 6y ago
Well there must be something that I’m failing to understand then. His example of what he got when switching to msripub is obviously broken; just look at the run-on bibliographic entries. But it’s been a few days and I admit I didn’t read the article with great care in the first place, so I’ll take another look. But if there were actual errors in his document, that’s another reason he should not have been drawing conclusions about LaTeX. Personally, I’ve had much better luck when switching styles. I think, though, that plain TeX is greatly underused, especially as, when there are problems, they are often conflicts among LaTeX packages. What do you do about cross references and bibliographies? (That’s the selling point of LaTeX for me, not the formatting stuff.)
- svat 6y ago
  I think the point is that the same document can compile without errors for one document class (style file), while being erroneous for another. So switching styles is not simply a matter of changing the \documentclass at the top, contrary to what one might expect. I've had mixed luck with switching styles: when the styles are very similar things tend to work better…. Yeah my problem with LaTeX in contrast to plain TeX is that it's too hard to understand what's going on, and one gets all those package conflicts… I agree with you about the cross-references and bibliographies being the main benefit of LaTeX; I imagine one can get by with a more lightweight macro package like eplain or opmac, though I haven't tried them much. Ideally IMO one would do these things separately (and generate TeX files that have the cross references simply hard-coded) and not via the macro system, but haven't tried that much either.

by svat 6y ago

leephillips 6y ago

Well there must be something that I’m failing to understand then. His example of what he got when switching to msripub is obviously broken; just look at the run-on bibliographic entries. But it’s been a few days and I admit I didn’t read the article with great care in the first place, so I’ll take another look. But if there were actual errors in his document, that’s another reason he should not have been drawing conclusions about LaTeX. Personally, I’ve had much better luck when switching styles. I think, though, that plain TeX is greatly underused, especially as, when there are problems, they are often conflicts among LaTeX packages. What do you do about cross references and bibliographies? (That’s the selling point of LaTeX for me, not the formatting stuff.)

svat 6y ago
I think the point is that the same document can compile without errors for one document class (style file), while being erroneous for another. So switching styles is not simply a matter of changing the \documentclass at the top, contrary to what one might expect. I've had mixed luck with switching styles: when the styles are very similar things tend to work better…. Yeah my problem with LaTeX in contrast to plain TeX is that it's too hard to understand what's going on, and one gets all those package conflicts… I agree with you about the cross-references and bibliographies being the main benefit of LaTeX; I imagine one can get by with a more lightweight macro package like eplain or opmac, though I haven't tried them much. Ideally IMO one would do these things separately (and generate TeX files that have the cross references simply hard-coded) and not via the macro system, but haven't tried that much either.