Talk:Methodus qua omnes detectos lichenes

GA review

Passed. Chiswick Chap (talk) 12:58, 15 February 2026 (UTC)

The following discussion is closed. Please do not modify it. Subsequent comments should be made on the appropriate discussion page. No further edits should be made to this discussion.

This review is transcluded from Talk:Methodus qua omnes detectos lichenes/GA1. The edit link for this section can be used to add comments to the review.

Nominator: Esculenta (talk · contribs) 01:56, 5 February 2026 (UTC)

Reviewer: Chiswick Chap (talk · contribs) 13:52, 7 February 2026 (UTC)

More information Extended content ...

Extended content

@Esculenta can you please disclose whether you used LLM assistance in writing this article? I know from our previous conversations that you have used LLMs to assist with other articles, and potentially have done so uniquely successfully. But given the nature of GA reviews I think it is important to disclose here. Thanks! NicheSports (talk) 14:27, 7 February 2026 (UTC)

Yes, of course. Almost everything I do on Wikipedia now (and increasingly in life) is with "LLM assistance" (although it would be more accurate to call it "agent assistance"). Esculenta (talk) 16:11, 7 February 2026 (UTC)

Thanks for the answer. Can you please share how you used LLMs in your editing process for this article? NicheSports (talk) 16:13, 7 February 2026 (UTC)

I could write a novella-length article on the "editing process" I've developed over the past couple of years, but to simplify it greatly, I gather all of my sources (some of which have to get digitized), put them into my custom GPT with custom instructions, it spits out a rough draft, and then I go back and forth with it (I have a text-to-speech interface hooked up for better communicating efficiency) discussing the article, and copyediting and tweaking it until I'm satisfied. In my setup I can highlight a section of draft text and with a command-keyboard input it shows me the source text it came from, so I can readily monitor source-text integrity. Esculenta (talk) 16:37, 7 February 2026 (UTC)

Thanks. I just wanted to confirm that you used LLM for text generation here, which is what it looked like in places. @Chiswick Chap earlier in the year I checked the entire article Tephromela, which Esculenta expanded from a stub with a similar process to what they describe above. Everything checked out (which was the first time, and remains one of the only times, I did not find significant issues with the core content policies in an extensive section of LLM-generated prose). So Esculenta's approach has some merit. However, that article is different than this one. Tephromela covers a fairly "dry" topic, with content that is almost exclusively factual. LLMs do better with such content. This article covers a book and discusses its impact and legacy. LLMs - even more recent ones, which have significantly improved - still struggle with generating content policy compliant prose covering such qualitative topics. None of this means this review should be failed (although Esculenta, I would strongly encourage you to proactively disclose your LLM use in any future GA nomination). My suggestion to the reviewer would be to carefully verify all content in the Methodus qua omnes detectos lichenes#Legacy section. If everything (or almost everything) verifies, without any evidence of SYNTH or similar issues, then all good! I would be uncomfortable passing such an article without such extensive verification checks. My two cents as someone who has picked up some GA reviews while primarily being active in LLM patrol and cleanup. NicheSports (talk) 16:57, 7 February 2026 (UTC)

Comments

An interesting if distinctly specialised article! And robustly cited.

The 5 words chosen for the article's title do not translate "may be identified", which is obviously critical to understanding the sense. I think we need to have the full title at the top of the lead, with a full translation of it.

The lead needs to be divided into paragraphs. Suggest we have three: content – Acharius's way of working – legacy.

It would be helpful to show the structure of a sample entry (Usnea, say). This could be with an image of a page (or part of one in its case), with a short annotation explaining the parts of the entry. Or we could have the Latin text of an entry beside a translation, again with a brief description of the structure.

The second circumnavigation can be linked to Vancouver Expedition.

"higher-level orders" = orders?

'External links' should follow the 'References'.

Images

All the images are relevant and of suitable quality. All are on Commons and plausibly licensed.

If (as seems likely!) the species illustrated with the photographs have modern names that differ from Acharius's, we should have both names in the captions.

The article would benefit from a sample of the illustrations in the original text, with the names of the species as given in the text (and links to the modern species, if discernible). This is pretty much mandatory, I'd say.

The article would benefit from an image of one of the "annotated specimens" mentioned in the text.

Sources

Spot-checks: [6] ok; [20] ok; [32] ok, but you could add from it that it recognises that Acharius's names are nonetheless correct.

Summary

There's very little wrong with this article, but I've made some suggestions that would definitely improve it. Chiswick Chap (talk) 14:25, 7 February 2026 (UTC)

Thanks kindly for taking on the review. I'll work on it and get back to you soonish. Esculenta (talk) 16:11, 7 February 2026 (UTC)

I implemented most of your suggestions in this edit. Added a full title and translation in the lead, but kept Jørgensen's wording in a footnote (to show a non-literal interpretation from a human expert). Added a sample entry (good idea, thanks) for the Lecanora image already there; the careful reader will surmise that many lichens were lumped into Parmelia back in the day. Added an image of a rock tripe; held back on an additional image as I think the image/text ratio is about right. I have more of these "fundamental lichenology book" articles planned, so it's very useful to get this type of helpful feedback. Esculenta (talk) 21:34, 7 February 2026 (UTC)
Ah, I guess I thought you had this in work, I was holding off in case you did some more (next time, maybe indicate 'Done' for each entry you believe you've covered, and a brief comment on what you didn't agree with, so I can see what has and hasn't been done. But it's absolutely fine as it is now. Chiswick Chap (talk) 12:56, 15 February 2026 (UTC)

The discussion above is closed. Please do not modify it. Subsequent comments should be made on the appropriate discussion page. No further edits should be made to this discussion.

Talk:Methodus qua omnes detectos lichenes

GA review

Comments

Images

Sources

Summary

Related Articles

Timelines

Top Qs

Fact Checks