Wikipedia:Peer review/Logic/archive2

Thoughts? GuineaPig_C77 (𒅗𒌤) ☕ 09:01, 24 September 2022 (UTC)[reply]

I think your rearrangment of the lead works great. It's a good idea to emphasize its importance right at the end of the first paragraph. I made a few small adjustment.

On challenge with the definitions section is that there is a lot of disagreement about how logic is to be defined. So we can't just say "logic is X" and move on, we have to say "some think that logic is X, other think that logic is Y,...". In principle, these problems have to be discussed, since the target audience is not just the general reader, but includes knowledgable and expert readers, see

@Phlsph7 Yep I'm looking out for the layperson so that has admittedly biased my suggestions. It's good to consider all perspectives but I'll defer to your expert judgment about how to strike the balance. And ditto for the depth of "digging in".

Haha, believe it or not I was trying to argue against including any faces in the lead image. My point was that there is already a collage of faces, so the ideal lead image would stand out better if it were something other than a face. But to be honest, it looks fine as is.

Your tweaks to the prose look good to me. As long as I haven't butchered the meanings in my pass through the lead, it sounds like it's going to read well now for readers of various backgrounds.

Here's another image from Commons, I'll add more if I can find them.

GuineaPig_C77 (𒅗𒌤) ☕ 12:02, 24 September 2022 (UTC)[reply]

GuineaPig_C77 (𒅗𒌤) ☕ 12:24, 24 September 2022 (UTC)[reply]

I've implemented the rearranged lead. In the process, I've made minor adjustments to the sentence talking about the difference between formal and informal logic. I also managed to expand the truth table to cover more expressions. It seems I misunderstood you concerning the collage of faces. I thought you were just against using the image of just one logician, which, I agree, would be a bad idea. So far no one at the article has complained and I think it's better than having no image, so let's see. I'm also still on the lookout for images for those two sections but so far I haven't really found something that is directly relevant to the text.

If you still have some time, I definitely appreciate having a non-expert go through the material and point out potential difficulties for understanding. It's not the only perspective that matters, but it is one important perspective. I hope the recent simplification helped a little with definition-section. Are there specific passages in other sections that seemed overly difficult or that had some points that remained unclear after studying them? Phlsph7 (talk) 16:30, 24 September 2022 (UTC)[reply]

Truth table looks great to me. The whole section reads well. Lead works for me too!

I went ahead and edited the first paragraph of the Formal and informal logic section, as a test. See what you think. If this is helpful for improving clarity without sacrificing accuracy, I'm happy to do the same for other paragraphs. (I noted in my edit summary that the last sentence feels redundant.) GuineaPig_C77 (𒅗𒌤) ☕ 19:52, 24 September 2022 (UTC)[reply]

Your changes to the section on formal and informal logic work fine: they help simplify the presentation without leaving out any important information. The last sentence was initially included because of the discussion of this issue later in this section but it does not really help much at this point so I removed it. Thanks for your help and please go ahead with what you think is best. I'll make changes accordingly if I feel that new problems are introduced somewhere along the line. Phlsph7 (talk) 20:04, 24 September 2022 (UTC)[reply]

In the section on Ampliative arguments, there is the phrase "In either case, the premises offer support for the conclusion because the conclusion is the best explanation of why the premises obtain." The word "obtain" is strange to me, unless it has some technical meaning? GuineaPig_C77 (𒅗𒌤) ☕ 23:10, 24 September 2022 (UTC)[reply]

Same question for the phrase "logical move" under Definitory and strategic rules. Is this the same word as in "bishops may only move diagonally" later in the paragraph? GuineaPig_C77 (𒅗𒌤) ☕ 23:17, 24 September 2022 (UTC)[reply]

@Phlsph7 Okay I finished going through carefully, rephrasing for clarity and brevity where appropriate. I also had some additional points that I was unsure of and therefore did not change:

• Logic#Formal systems. I think the third sentence is redundant?
• Logic#Informal. I almost changed "reaction" to "response" because it's not automatic but in this context, part of a cognitive process?
• Logic#Mathematical logic. The very last sentence of this section is awkward with the dashes and question mark.
• Logic#History. In the second paragraph, "methods of agreement, difference, and concomitant variation" is an awkwardly long wiki link. In the fourth paragraph, I'd suggest citing more heavily since you're pulling direct quotes. On the other hand there is a substantial footnote so I'm not sure. Just putting it on your radar.

GuineaPig_C77 (𒅗𒌤) ☕ 00:47, 25 September 2022 (UTC)[reply]

@GuineaPigC77: I made a few minor adjustments but your changes help increase accessibility. The word "obtain" is a technical term, but it's not necessary here so I replaced it. The term "logical move" was intended to emphasize the analogy with games. But its a little awkward since this analogy is only explained afterwards so I replaced it with "inference". For Logic#Informal, I think either "reaction" or "response" could be used. Cognitive processes can be automatic and unconscious, but their paradigmatic form is conscious. I've added the source for the quoted expression in the history section: it's better to have it there explicitly than hidden away in a footnote. Thanks a lot for all the effort you have poured into the article. I think the most glaring issues have been fixed but please let me know if other things have caught you eye. Phlsph7 (talk) 06:17, 25 September 2022 (UTC)[reply]

No worries, Phlsph7! I'm happy to help. To be honest, working through these minor issues over the past day or two, under your expert eye, has been like free expert tutoring - I'm happy to be refreshed on logic topics that I have learned, and to also learn just how huge it is. Thanks for your patient explanations to my queries.

I think it's in great shape. I'm not in any way offended if you keep this thread open to attract some more types of reviewers. But to this reader (enthusiastic layperson) it reads nicely!

Feel free to ping me, and best wishes for GAN! GuineaPig_C77 (𒅗𒌤) ☕ 07:51, 25 September 2022 (UTC)[reply]

I've started the GAN. It may take a while before someone decides to start the review so I'll keep the peer review open until then. More comments on linguistic issues, images, topics to be included, etc. are welcome. Phlsph7 (talk) 10:02, 25 September 2022 (UTC)[reply]

Oh great, glad the process is underway. Yes I'll likely revisit and read with fresh eyes in a few days. And I'll keep my eyes open for more images but I do agree they have been hard to find. One possibility I just thought of would be to create your own image. I'm not sure of the guidelines, but doing it yourself could offer you the flexibility to conjure up the exact perfect image. Just a thought. I've never done it, but I'd be up for helping compose something. Is this something people do?

Also, do top / high importance articles get any sort of priority with GAN or is it just whoever gets to it. GuineaPig_C77 (𒅗𒌤) ☕ 11:07, 25 September 2022 (UTC)[reply]

If you create an image then you are the copyright holder (unless its somehow plagiarize). You can publish your own images on Wikimedia Commons so they can be used here, see the helpful explanation here. You can also modify images on Wikimedia Commons created by others, like what I did here. If you have an idea that is relevant and not too difficult to execute, then we could give it a try.

I guess high importance articles get priority in the sense of more scrutiny but my experience in this field is also rather limited. In the end, I think it's similar to the peer review process: it happens when someone makes the request and someone else decide to review it. Chances are that more reviewers can relate to a high importance topic, which may decreased the expected waiting time. Phlsph7 (talk) 11:42, 25 September 2022 (UTC)[reply]

By the way, I managed to find an image for the definition-section after all. Phlsph7 (talk) 15:55, 25 September 2022 (UTC)[reply]

Thanks for these links re images. That's good to know and yes I think it might be fruitful to pursue it. And it makes sense about the GAN process - I suspected that's how it might work.

Also, I should have said this earlier, but there are important aspects of the article that I did not touch. For example, I haven't done anything in the way of spot checking references, formatting references, or the like. So this might be something to seek out when getting further comments.

This copyvio tool yielded "Violation Unlikely, 31.0% similarity". I checked the top hit (https://iep.utm.edu/prop-log/) and went through more carefully and didn't see anything alarming. So that's nice for peace of mind.

I noticed the new Tarski image which looks great. I moved it to the left side of the page per

MOS:PORTRAIT which prefers the subject to face inwards toward the text. GuineaPig_C77 (𒅗𒌤) ☕ 18:37, 25 September 2022 (UTC)[reply

]

Brainstorming images. It seems that the sections in most need of images are the Definition section and Formal and informal logic section. When I read the latter, one prominent lesson I learn is that at least for some logicians, the distinction between formal and informal logic may be associated with the distinction between formal and informal language. Perhaps you could include a diagram that translates (I'm not sure the technical word) an English sentence into a more formal sentence? You could also contrast this with an informal statement that is difficult (impossible?) to fit into the formal language. I'm thinking a diagram one might see in a lecture slide. GuineaPig_C77 (𒅗𒌤) ☕ 18:54, 25 September 2022 (UTC)[reply]

Screenshot of the output of the logic software you mention? Something that illustrates the "automatic-ness" of formal logic? If that's even a thing.

And I'll re-visit our earlier discussion of the collage of faces. Considering that we're strapped for images, one possibility would be to split that collage that is currently the lead image, and sprinkle those faces throughout the article. I'd normally think this would be more of a natural fit for the the history section alone, but I could see it working if you roughly aligned faces with the content throughout. That doesn't answer the question of what goes as the lead image, but it could help mitigate the walls of text without relatively little investment of time. Just a thought. GuineaPig_C77 (𒅗𒌤) ☕ 19:08, 25 September 2022 (UTC)[reply]

I was not aware of

MOS:PORTRAIT

but it makes sense. The translation image for the section "Formal and informal logic" is a great idea. I was thinking about something like the following:

Regular English: Carmen is Mexican and a teacher.
First-order logic: ${\displaystyle M(c)\land T(c)}$

There could be arrows from Carmen to c and from Mexican to M etc. Maybe colorcode the different parts as well: Carmen and c in green, Mexican and M in red, etc. We could also use another example sentence like "If Carmen is from Tijuana then she is Mexican". What do you think?

You are right that the current lead image would fit well in the history section and could be replaced by the modus ponens image currently found in the section "Deductive". I'm not sure which combination would be better. If you feel confident about it, we could go ahead with it. Another option would be to leave it as it is for now and bring the topic up somewhere in the GAN discussion. The article in its current form has 10 images (11 if you count the Philosophy sidebar). More would be better but given that logic is an abstract topic, this seems not too bad to me. But as I said, this is not my field of expertise. Phlsph7 (talk) 07:56, 26 September 2022 (UTC)[reply]

That sounds perfect. If you don't feel comfortable creating it, I can give it a go, I like this sort of thing. Also we may consider choosing, perhaps, red and blue, or something that won't be missed by dichromats. I think showing the formalization of those sentences is perfect, and I'm also wondering whether should be a "straw-man" sentence, that can't be formalized, as an example of informal logic. (Or maybe that's not the relevant distinction, maybe it's the logical framework that distinguishes them, not the content of the sentence?)

Re number of images, I'll walk back my earlier statement that we're "strapped" for images. I don't think that's the case anymore, but I do think attention to which images go where is worthwhile now. And like you say, GAN will provide another set of opinions. That all seems reasonable.

Re sprinkling faces, I think the question is how spaced out they will be which depends on the prose.

Do you happen to know what font is used here when you put text between math tags? We can try to match the font in the image. If we can't match the font, we can compose the math text in a sandbox and screenshot it. GuineaPig_C77 (𒅗𒌤) ☕ 18:11, 26 September 2022 (UTC)[reply]

Also, re very first sentence of article. Are "correct reasoning" and "good arguments" synonyms? If so, I would considering choosing just one of them. If I'm about to read an article on logic, I'm on high alert for words like "or". So I get distracted right away. If there's a choice, I would keep "correct reasoning" since the description of an argument being "correct" is used in the article far more than "good" (although they are both used). GuineaPig_C77 (𒅗𒌤) ☕ 18:37, 26 September 2022 (UTC)[reply]

Please go ahead with the creation the image. I'll leave the details concerning the coloring, arrangement, etc up to you. Right now, I can't think of a simple example that can't be formalized. Some sentences can't be formalized in first-order logic but need higher-order logic. However, strictly speaking, any sentence can be formalized in a trivial way in propositional logic by just ignoring its inner structure and using a variable to stand for the whole sentence. But even if we find something, it may need an extensive explanation of what is happening and why it doesn't work. The problems aluded to in text are more about disagreements concerning whether a translation is fully accurate or how the translation should happen, for example, whether time should implemented using singular terms (which is unintuitive) or using operators (which requires temporal logic). Such explanations are probably too much for the image caption. But showing in a clear way for one example how this form of translation happens would already be quite helpful to a reader unfamiliar with the topic.

I'm not sure about the font used by the math-tags but I don't think it matters much to exactly match the font to the text. You could just copy-paste the unicode symbol "∧" since your image editor probably does not support math tags.

Concerning "correct reasoning vs good arguments": there are various characterizations of logic in the academic literature that closely overlap: as the science/study of correct/good/valid reasoning/arguments/inferences/thinking/laws of thought. There are probably many more. Having this "or" gives the impression to the reader that this is not the one and only characterization everyone uses. But you are also right: having just one characterization would make it simpler to the reader and less likely that they get distracted. I'm not sure which approach is best here. Our WP:Short description uses "Study of correct reasoning".

Okay will do! And that's good to know about informal sentences. We'll just do the formal. One question here - is there a third (middle) stage in this formalization process? For example, in English: "Carmen is Mexican AND Carmen is a teacher." And if so, how would I write the "and" part.

I prefer the single definition, but only if it doesn't mislead the reader. Another option is to say explicitly that there are different characterizations soon after the first sentence. But the "or" is fairly distracting, so ideally it would be avoided.

Something that could be helpful (for the novice, but perhaps this could be done without annoying the expert) would be a figure with a caption like “The truth value of this statement depends on the truth values of its denotations; but the correctness of the argument does not.” I’m butchering the language I’m certain, but the intent is to have something prominent that distinguishes between truth and correctness - a distinction that many novice readers will not know (or even know exists). Perhaps something like this could go in the subsection Premises and conclusions, where it is explained clearly in the prose. An image would make this concept very prominent. For the newbie, this would serve to elaborate on what is meant by “topic-neutral” in the lead. (Please forgive me if I’ve got these concepts all wrong. In case it doesn’t make sense, my intention is to make prominent the basics that a newbie can grasp and remember - without unduly annoying people for whom this is obvious.)

Proposal for sprinkling the logicians:

• Aristotle. Currently the same image appears twice. Use the second occurrence in the first paragraph of the History section.
• Avicenna. Second paragraph of the History section.
• Ockham. Third paragraph of the History section.
• Boole. Computational logic section.
• Frege. First-order logic section.
• Russel. Mathematical logic section.

This would not be perfect in terms of spacing images out, but this approach means there are plenty of images, and not all would need to be used. In this case we would use modus ponens for the lead image. (But again, this is how I personally see logic, so it could be any other appropriate image that better represents the breadth of the topic.)

Side note: when you outdented the conversation, I lost my reply button, is there a way to get it back? GuineaPig_C77 (𒅗𒌤) ☕ 22:42, 26 September 2022 (UTC)[reply]

I've created an image like what you described, should I upload it to commons? I assume we'll need a few rounds before we settle on something so it seems weird to upload it there. How do people workshop an image?

For a possible second image, here's something more concrete with respect to the Premises and conclusions subsection. I'm imagining an image along the following lines. First it says in English (but formatted nicely, in neat boxes):

"Carmen is Mexican and a teacher." implies "It’s not the case that Carmen is either not Mexican or not a teacher."

Below that, it says it in the formal language:

${\displaystyle M(c)\land T(c)}$ implies ${\displaystyle \lnot (\lnot M(c)\lor \lnot T(c))}$

The caption would say something like:

"While the truth values of the statements ${\displaystyle M(c)\land T(c)}$ and ${\displaystyle \lnot (\lnot M(c)\lor \lnot T(c))}$ depend on the truth values of the denotation (e.g., whether Carmen really is Mexican), the correctness of the argument does not."

Some of this may be wrong, but my point is that the non-expert will need a sharp distinction on this basic idea. Again, I'm just looking after the beginner here and I know it's a balance. Of course, this can all be said in the prose; the point of putting this in an image is so that it stands out as important (and so the two languages can be aligned one on top the other). Second, something like this could be accomplished with a simpler inference, but I think there's value in using the same example that will be used for the formalization procedure. (And the two images would be in the desired order since Fundamental concepts immediately follows Formal and informal logic.)

Side note: My reply button is back. GuineaPig_C77 (𒅗𒌤) ☕ 02:38, 27 September 2022 (UTC)[reply]

One could make a middle stage in the translation, but I don't think that this is what introductions to logic usually do. And I guess the example is basic enough for the reader to grasp it without a middle stage.

I've implemented your suggestion concerning the "or" in the first sentence. The section "definition" makes it clear that this is not the only characterization found.

About the correctness of arguments: this is mentioned in the section on fallacies, but it's a good idea to mention it earlier as well. We have to be careful since your characterization is not entirely true: some informal fallacies are seen as incorrect arguments because they use a false premise, like the False dilemma. But your characterization works for formal logic. I've added a short clarification with an example to the section "Deductive". But we could also use your idea with the image using ${\displaystyle M(c)\land T(c)}$ implies ${\displaystyle \lnot (\lnot M(c)\lor \lnot T(c))}$. In this case, the image could be styled similar to the modus ponens image but only with one premise instead of two.

About the lead image: let's give it a try with the modus ponens in the lead. Before we get started: what do you think about moving the current lead image, as it is and with its caption, to the beginning of the history section? But your suggestion of sprinkling suggestion is also good. I could come up with some captions in this case.

You can upload different versions of the same image to wiki commons, revert back to an earlier version, etc. Your wikipedia login should work there as well so you don't need to make a new account. The upload procedure is quite simple if you are the author of the image.

I think the reply button was lost because I forgot to sign my last reply. Phlsph7 (talk) 06:24, 27 September 2022 (UTC)[reply]

Cool thank you for those helpful pointers. Here's a draft of the first image.

The first sentence reads much better to me.

Yes let's try modus ponens for the lead. I think either including the collage at the top of the History section would work, and also distributing them around. We can try both and see what works better? I'll let you do it since I think you're wanting to write some captions in the case that we separate them. (I'm partial to distributing them - given that so many viewpoints are being considered, it's nice to have a face to go with the approach.)

Given what you told me re: the De Morgan inference, will a figure still work if we state that it only applies to formal arguments?

Correct, good, and valid. Reflecting on your comment, I think it's worth explicitly stating that these are synonyms. And ditto for the reasoning/arguments/inferences. My gut feeling is to not change anything in the article, but to find a place early on to say that you're using these words interchangeably, if in fact you are.

I think the following represent the remaining hurdles for the average reader. For most readers, we need to be told explicitly that:

• truth and validity are different things
• correct = good = valid: words that describe arguments, not propositions
• reasoning = arguments = inferences
• true and false are words that describe propositions, not arguments

These concepts are there, but us normal folks need these 4 to be more prominent.

And I suppose there's a more succinct way to say all this. Perhaps the issue is simply: Do everything you're going to do in the article, but start slightly earlier for the average reader so as to not leave us out.

I'll work on a second figure as we consider these issues! GuineaPig_C77 (𒅗𒌤) ☕ 19:25, 27 September 2022 (UTC)[reply]

To acknowledge - your earlier comment didn't say they were synonyms, it said they "closely overlap". So to clarify, all my suggestions for readability are under the assumption that we can do it without losing any such nuance. To the extent that's possible, it's worth pursuing, and that's the part I'm after. I acknowledge this is an encyclopedia and not a lecture. GuineaPig_C77 (𒅗𒌤) ☕ 02:01, 28 September 2022 (UTC)[reply]

The image looks great. Now the section "Formal and informal logic" finally also has its own image. I also put the modus ponens image to the lead and the collage of faces to the history section. I'll get back to you in my next reply about the idea of splitting it up into separate images. I think the De Morgan image is not that important, but if you feel like doing it, we could include it more or less as suggested in the article. We just need to make sure to restrict the explanation to formal logic.

Thanks for pointing out that the basic difference between validity & truth should be explicitly stated for non-expert readers. I've added it to the subsection "Arguments and inferences". If we want it earlier in the article, we would probably have to mention a short version of it somewhere in the lead.

I've added one more characterization of logic to the section "Definition" to emphasize the diverse overlapping definitions. There are some small differences between the terms you mentioned. Some of their overlaps and differences are already explained at the beginning of the sections "Definition" and "Arguments and inferences". We could expand on these explanations. But, as far as I'm aware, most introductions of logic do not go too much into detail on this issue. Phlsph7 (talk) 08:21, 28 September 2022 (UTC)[reply]

Great! Your recent edits look good, and they are huge for improving clarity for me. Thanks!

Let's scrap the second image. If we really wanted to include this, it could easily be done in the prose, but it sounds like even this may not be necessary. The prose is much clearer on this issue now.

Lead image works great, I'm 60/40 on preferring it over the collage, but let's see what others think. I'm also 60/40 on separating the images - no strong opinions.

The formalization image won't win any awards, but I think it serves the purpose well. I ran it through this filter and it's not perfect but perhaps adequate. I might tweak.

I plan to do a full read-through later today, and will likely have more comments, but the big things w/r/t clarity seem to be falling into place nicely! GuineaPig_C77 (𒅗𒌤) ☕ 16:57, 28 September 2022 (UTC)[reply]

Okay I did another full read-through. The whole thing is excellent! Some very minor comments:

• Lead. Crystal clear (with one exception, see next comment).
• Logic#Formal and informal logic. "new information not already found in the premises" is very helpful, and is clearer than when this topic is brought up in the lead, paragraph 3, where it says "which may arrive at genuinely new information in their conclusion". I think "not already found in the premises" is the helpful part. This phrase also occurs under the Deductive subsection. If you don't want to say it three times and have to choose, I'd prioritize it in the lead.
• Caption of new image. It says "they are valid". I assume "they" refers to "English sentences". But if so, why "valid" and not "true"? Perhaps instead of "they" it could say "arguments"? Not sure, but it's a minor point of confusion.
• Internal structure section. I like that the Mars example returns.
• Internal structure section. "if-then" should stylistically match the "if...then" from lead paragraph 2. Also, this is the only place where "truth-value" is hyphenated.
• Fallacies. The feather example is really cool.
• Definitory and strategic rules. Chess example and the word "move" is much clearer.
• Informal logic. "These standards also depend on the type of dialogue: in the context of science, the dialogue rules are different from the rules in the context of negotiation." This phrasing sounds awkward. Not sure why.
• Computational logic. "He showed how computer circuits can be understood in terms of boolean logic and implement it." Should it be "implemented"?
• Computational logic. The image is here okay, but perhaps could be better. It shows some kind of clock circuit that produces a toggle. Something simpler like an AND gate diagram could connect the image to the prose better?
• History, paragraph 1. "live on" might push against
  WP:TONE
  .

Overall this article is so much clearer now (to me) and a much more enjoyable read. I think many types of readers can get something substantial out of it and have a good reading experience. GuineaPig_C77 (𒅗𒌤) ☕ 22:11, 28 September 2022 (UTC)[reply]

I've split the multi image into several separate images. I think both version work quite well. The translation image makes the issue more accessible. I'm glad you spotted the problem with sentences vs arguments and I made the corresponding change. You are right about the image in the section "Computational logic". The relevance for the text is better with the AND-gate using transistors. I've implemented reformulations for most of the expressions you mentioned. Thanks again for your keen eye and for all the effort you've put into this review. I hope the article is now in good shape for the GA review. Phlsph7 (talk) 08:56, 29 September 2022 (UTC)[reply]

Sure thing @Phlsph7! Thanks so much for your responsiveness and admirable patience. It was a joy to work on this. Happy to help in the future! GuineaPig_C77 (𒅗𒌤) ☕ 10:22, 29 September 2022 (UTC)[reply]