Talk:The Outline

From Textop Wiki

Revision as of 12:18, 12 August 2006; view current revision
←Older revision | Newer revision→
Jump to: navigation, search

Contents

Introduction

Please feel free to use this page to discuss how in general to improve The Outline. --Larry Sanger 16:50, 15 July 2006 (MST)

I got the original version of this outline started. It is based on Hobbes' Leviathan. It is very much a rough draft--a work in progress. No parts of the outline are mature, but some parts of the outline are better thought-out than others. To see the whole (original) thing in MS Word form, see this page. Over the coming days I intend to put the rest of the chunks into this outline form--sorry most of it is blank right now. --Larry Sanger

The default relation between the nodes of this outline is the specialization relation. Whenever this is not the case, the name of the relation has been given (as in "origin:" or "result:"). Click on the above "discussion" button for more details on how and why The original outline without relations has been updated. --Philippe Martin


Making node relations explicit

Note from Philippe A. MARTIN. I have been invited by Larry to add some relations between the nodes of The original outline without relations (see the words before semicolons in the new outline). Indeed, as noted in my emails titled "not fine-grained and structured enough", I believe that the minimal structure that this project should have for it to have a chance to achieve its ambitions is that of an informal semantic network (concept map, topic map, etc.), which means that the "relations" between the "nodes" should be made explicit. A node is either a category (i.e., course outline heading, concept, class, topic, individual object) or a chunk of text such as a sentence or a paragraph. Examples of common relations between categories: "subtype", "subprocess", "subdomain", "agent", "object", "definition". Examples of common relations between chunks or categories representing those chunks: "specialization", "objection", argument", "example", "proof", "interpreter", "corrective_specialization". Examples of common relations from any node: "author", "time", "purpose". In an "informal" semantic network, the users can invent new relation names whwnever they do not find adequate already existing relation names. Although relation names can be nominalised expressions (or even whole sentences), a relation name should be a singular noun whenever this is possible (if only for normalisation/retrievability and readability). In the new course outline, the subtype/specialization relation is the "default relation" between nodes and hence that it does not have to be explicitly written in order to make it explicit. I have slightly adapted many category names in order to clarify their meaning and/or for some relations to be set (for example, for specialisation relations to hold, I have replaced questions and plural nominal expressions by singular nominal expressions). However, I have avoided to modify the overall presentation of the outline as much as possible (nethertheless, I have put "The concept of God" under "Theism" and removed "The definition of law"). I used "?" as relation name when I could not find a name for the relationship between two nodes, most often because I could not guess the ontological nature of one of the two nodes (e.g., either action or result of action) from the name of this node. Given the current coarse granularity of this outline, this exercise does not seem to bring much and hence seems artificial. However, it is a guide for extension or redesign of the outline.

Thanks very much, Philippe. I'm not sure I like the way you've redone the representation of chunks, and some of the renaming of nodes is problematic to my mind, but overall I like the direction that you're taking this in. In particular, I do think that it's useful and important that we make relations explicit. More later. --Larry Sanger

Notes explaining why I've reverted some outline items:

  • "Religion" was converted to "Religious belief." I changed it back to "religion." "Religious belief." is not the best name for the metaphysical topic (that's a better name for a topic in the psychology or epistemology of religion). I am sure there's a better name than "Religion" for the metaphysical topic (perhaps "divinity")...
  • I changed items like "Concept studied by Metaphysics" and "Feature of human nature" for the very simple reason that they are awful as names of categories. The only reason for all the changes of names made by Philippe is to make the relations between the node and subnodes correct (once he made their relationships explicit); well, that is not a good reason to change the name of a node, I think, at least in this system. Readability comes first.
  • "Natural condition of mankind" was changed to "Natural feature of mankind," which actually goes further and obfuscates a very common philosophical phrase. Please don't do that...changed back.
  • There was a considerable amount of gratuitous editing of the political philosophy section. I would ask people not to edit the ordering of items, or to remove items, unless they are working with the texts themselves.
  • Please retain "The X" where they exist and where "X" is a general term, as in "The State." This is perfectly good English and the usual way that philosophers refer to items in the abstract in English. Another example: "The whale is the largest animal in existence."
  • Finally, if you want to make large numbers of a certain kind of edit, let's discuss it first, to save one of us the labor of having to undo the other's work.  ;-)

--Larry Sanger

Dyadic relations are not enough

JA: The above discussion seems to take it for granted that all pertinent relations among nodes are dyadic. In my experience, this guarantees a hopelessly inadequate analysis of all but the most trivial subject matters. Is there any way to flesh out this outline to include triadic relations among nodes? Jon Awbrey 09:26, 10 August 2006 (PDT)

PM: it could be done this but would be bad for readability AND analysis. What is expressed by a triadic relation can (and should) always be expressed in a more explicit, matchable, scalable and usable form by one or several dyadic relations (with the use of contexts/meta-statements and sets if need be). For example, actions should be represented using concepts nodes and binary relations (case relations and others); "between" should be represented as a binary relation between something and a set of things. I know the work of Gary Richmond and I know he did say that "there exist some mathematical work that proves that some triadic relation that cannot be defined in terms of dyadic relations" but to my knowledge he has never provided a reference to that work nor any concrete example. If someone gives me an example I'd be grateful but until then I will not believe it. Phmartin 06:27, 11 August 2006 (PDT)

JA: I spy an interesting possibility here. We have an assertion by me and my whole army of Peirceans and we have a contrary assertion by you and, from prior acquaintance, the camp followers of Quine. We could exploit this as a test case for how the debate prototype would handle a technical dispute. If anybody's game for that? Jon Awbrey 06:40, 11 August 2006 (PDT)

PM: ... except that only 1 valid example is required to end the debate ... (as noted above the use of a triadic relation when contexts or sets can be used is not a valid example; I am also not too convinced about the reference to Peirce: (i) a relative_thing as well as a mediating_thing is a thing_playing_a_role, (ii) anything can be categorised as a relative_thing or a mediating_thing (this is a question of viewpoint), (iii) I have nothing against people categorising some stuff as mediating_thing if that please them). Phmartin 00:24, 12 August 2006 (PDT)

LS: If I may jump in here: if anyone has an "army" of people in their back pocket to work on a debate summary, get 'em out! Moreover, I'd like our first examples to be high-quality, so it would be great (but not a requirement) if you work in (or near) your areas of specialization. --Larry Sanger 10:30, 11 August 2006 (PDT)

JA: I have already expressed some (long-experience)-based — note the implicit appeal to some kind of nameless authority, hard-knocks expertise, unquestioned knack, or vague phronesis that is supposed to accrue to mere survival on this blessed planet — sc/kepticism about the adequacy of your run-of-the-mill activity that all world calls "debate" to lead toward knowledge, but I'm willing to open the question one more time. Still, I would not want to go about it in any of the innominate ways that I've already tried it an innumerate number of times before. So let me get another cup of coffee and, as Fagin (the Broadway Fagin) says, "review the situation". Jon Awbrey 08:46, 12 August 2006 (PDT)

JA: On third thought, I think that it would be useful to pursue a couple of lines of inquiry here: (1) my long-standing idea of upgrading the analytic-cum-software tools that we are using to allow for 3-adic relations among topic nodes, and (2) the late-breaking idea that the customary 2-adic/3-adic disputes might afford a good teething-biscuit/triscuit for thinking about technical debates, or let us say "design forks" of various kinds. But, like the man said, it's a weekend. Have a good one. Jon Awbrey 12:16, 12 August 2006 (PDT)

Methodology & Epistemology

Can we unify this to Philosophy of Science? They are subtopics of that. Or are these chapter titles of the Leviathan?

--Math 13:11, 22 July 2006 (MST)

Well, I suggest we wait for more detail in all such subdisciplines before worrying too much about it. There are many overlapping aspects of what have gone under the titles methodology, epistemology, and philosophy of science. I would not say, though, that philosophy of science encompasses all of epistemology; if anything, it's the reverse.
These aren't chapter titles, they are common subdisciplines in philosophy (except "methodology," which isn't...). --Larry Sanger 19:16, 22 July 2006 (MST)

There is a general problem with the structure of concepts: our set of concepts is different than the set of concepts of Hobbes. What to do? Be historical or be up to date?

--Math 13:20, 22 July 2006 (MST)

I'd say that while different philosophers structure their concepts differently, the concepts themselves share important and useful commonalities throughout history. People differ amongst each other about what various concepts amount to today as individuals today differed with individuals in the past. So it's a very general problem you're talking about. Since we want a single outline for everyone, though, I think the answer is to find categories and structures that most elegantly capture the details we actually find in texts. It's hard to say more than that without writing an essay... --Larry Sanger 19:16, 22 July 2006 (MST)

Religion

In my view and in Hungarian we have the concepts: "(religious) faith", "religion" and "church".

Faith means a faith in some divinity. It may be for example some specific, subjective faith of somebody, it is very common in Europe to have your on specific faith without the doctrines of any religion. A religion is a faith, which has established ceremony, doctrines. A church is a religion, which has established organization.

So, for example Taoism is a religion but not a church.

For the metaphysical concept, "faith" would be the best in my opinion, because methaphisics does not deal with ceremonies and organisation.

--Math 13:29, 22 July 2006 (MST)

Your concepts are all quite universally understandable concepts, as far as I am concerned; we make similar distinctions in Anglo-American philosophy and religious studies. However, I would disagree that "faith" is the right term for the metaphysical concept, because faith is itself an important phenomenon, a mental phenomenon. It is possible to describe such concepts as "infinite being" and "the creator of the universe" without mentioning faith (a mental state, or a spiritual state if you prefer). --Larry Sanger 19:16, 22 July 2006 (MST)
Let me go ahead and change it, as I said, to "Divinity." That's not perfect either, but it's better than "Religion." It's just that, with that name change, now I'll have to move the stuff about religious belief. --Larry Sanger 19:18, 22 July 2006 (MST)
And thanks for the help, Mathias! --Larry Sanger 19:22, 22 July 2006 (MST)

I suspect some translation issues here. Certainly the terms 'faith', 'religion' and 'church' are not understood this way in British English. The term 'church' would not usually be used for non-Christian organisations, for example. In fact, it would not be used for all denominations of Christianity. Not all religions have doctrines or ceremonies. Some denominations of Christianity do not. I believe the World Council of Churches has a special form of membership for the Religious Society of Friends for this reason. The terminology obviously breaks down when one starts looking at non-Christian traditions. I think we would use the term 'organised religion' for what is being translated as 'church' in the comment above, and that the term 'religion' is understood more broadly or has several senses, perhaps. --How do I add a signature?

Eliminating names of disciplines

For the most recent set of edits (eliminating names of disciplines) see this post. --Larry Sanger 00:01, 26 July 2006 (MST)

Ethics

I'm not sure how to proceed with the ethics outline. This may work well for Hobbes, but it is obviously going to prove problematic for other texts. For example, 'wisdom' is on some lists of virtues, as is 'justice'. These are among the 'cardinal virtues' from the Aristotelian tradition.

I have more to say, but it will have to stay unsaid. I apologise if this is a mess. There are problems with text on some sites, especially entering text into boxes and it is difficult to describe just how difficult this makes writing anything.

(CFR: would still like to know how to make a signature just in case I ever find a fix for the text entry.)

Clea, thanks for the comment. The way to make a signature is by typing three tildes in a row ~~~ for just your userID, or four in a row ~~~~ for your userID plus timestamp. I don't blame you re the ethics outline. I'm quite sure you're right about it being problematic for other texts. That's exactly why I say that, as we add more and more texts into the mix, we'll steadily improve the outline so that it can encompass them all. You know if you were to use Hume's Enquiry concerning the Principles of Morals, with which I'm reasonably familiar, I'd be happy to work with you on it. --Larry Sanger 22:53, 27 July 2006 (MST)


Method for interlinking outline and text

I think we'll find it very useful to be able to click through from a chunk, where it lives on a node page, to its location in the source text, and we can model that functionality (sort of) here on the wiki. I'm also, in the Leviathan text, going to insert the chunks associated with a paragraph just below it--and link from there to the node where the chunk is filed. So here's what I am going to do (and I've started doing it):

  1. Upload all parts of Leviathan chapter-by-chapter, naming chapter pages according to the same convention used for labeling source chunks; e.g., Hobbes, Lev II. (DONE)
  2. Edit the outline, to make headings as well-worded and -organized as possible. (The purpose of doing this before the next step is kind of hard to explain. Basically, since on the text pages, e.g., linked from Hobbes, Leviathan, I'm inserting the chunks associated with a given paragraph just below the paragraph, and to help people reviewing the chunking, I'm including the node the chunk lives at. But we're going to rewrite a bunch of node names when we go through the outline. That means I'd then have to go back and update all the chunks on the text pages with the new node names. So we might as well finish editing the outline first.)
  3. Then go through The Outline node by node:
    • Optionally, add summaries as needed.
    • Add a colon (:) to all lines.
    • Copy the resulting reformatted chunk; then, convert its source line into a link to the appropriate Hobbes chapter by adding [[ and ]]; e.g., a line might read "Hobbes, Lev II 3". (Copy the reformatted chunk before adding in the link to the Hobbes chapter: you don't want the Hobbes chapter copy of the chunk to link to itself...)
    • Save the node page, and click through to the chapter page (which should work).
    • Paste the chunk you copied earlier into the chapter page; before the chunk, add an indented line to each chunk linking back to the node page where it is filed, e.g., nature, and the result should look like what you see at Hobbes, Lev Intro.
  4. Finally, go through Leviathan again, chapter by chapter, checking over work and adding chunks as necessary.

Finished!

Next steps, short report

Next up: redistribute the rest of the "Social psychology" and "Natural condition of mankind" links per mail on textop-en-phil. Then tackle the meat of "human being" (so to speak!).

In any case I do hope to get through the entire outline review relatively quickly, on the order of a month. It can't be made "perfect" anytime soon, but it can definitely be made better. --Larry Sanger 00:26, 9 August 2006 (PDT)

Retrieved from "http://textop.org/wiki/index.php?title=Talk:The_Outline"
Personal tools
content