Still interested in the question of Gemtext and emphasis, I figured I’d gather some data.
This post has my current best proposal for supporting emphasis:
But as there’s very little support for it, the more important question is: why?
To see if data would help answer that, I sampled three sources to check for use of emphasis: news articles on the web, Gemlog entries, and Phlog (Gopher log) entries.
I checked the first 30 articles linked from the front page of The Guardian.
Of these, just one—3%—used italics to convey emphasis. It was an article about a cricket game:
Warner rocks back to cut Root square of the wicket and <nearly> succeeds in feathering an edge to Bairstow.
But, a lot of articles, twelve—40%—used italics for other reasons. It was mostly to mark names, for example names of books and movies. In some cases it was for other typographic reasons, such as for a header or a “call to action” after the article.
I checked the first 30 Gopher posts I found via:
nytpu’s comitium subscriptions
Of these, three—10%—used some markup to convey emphasis. Two used “_” and one used "/".
But _conversation_,
full, deep conversation,
with kindness, consideration, mutual respect:
where exactly is that
on the Large Internet?
Three more, another 10%, used markup for non-emphasis: one used "*" to “emote”, one used "/" to mark a name, and one used "*" for other typographic reasons.
Finally, I checked the first 30 Gemlog entries I found via Cosmos.
Of these, ten—33%—used markup to convey emphasis. Most popular for this was "*" with six uses, then “_” with three uses. There was one use of double quote marks—which is arguably a valid use, but I’ll happily argue that what was intended was emphasis and not quotation.
I just *needed* one thing.
Just one entry—3%—used "/" to mark a name.
I should have written down what I was expecting before gathering the data!
I think what I expected was that because Gemtext chooses not to support emphasis, Gemlog entries would have less emphasis than other comparable text formats. This would fit the received wisdom around the problem—that you can perfectly well write without emphasis and maybe you should.
But in fact one in three of the Gemlog entries I checked used emphasis.
I was particularly surprised that Phlog entries had less emphasis, since Phlog entries are plain text which is arguably an easy place to put whatever custom markup you like for the purposes of emphasis.
This is not rigorous—the sample size is large enough that the data is interesting, but far from proof of anything. It’s also entirely possible that I’m biased and did the analysis wrong. I kept the “raw data”, the 90 URLs checked, in case anyone wants them.
Also, there may be a flaw related to post length: I should have also measured the word count of each entry to account for that.
So the only conclusion I will draw today is a minimal one: emphasis is a relatively common in Gemtext.
More data and more careful analysis could evaluate the hypothesis that Geminauts use emphasis more than other comparable posters. If so that might be considered an argument for supporting emphasis directly in Gemtext—or for standardizing how emphasis is added even if clients do nothing with it.
It’s more than fine if this doesn’t go anywhere, and it probably won’t; I had fun with it—I hope some readers of this did, too. Thanks!
So far today, 2024-11-25, feedback has been received 1 times. Of these, 0 were likely from bots, and 1 might have been from real people. Thank you, maybe-real people!
———
/ \ i a
| C a \ D n |
irc \ /
———