h a l f b a k e r yNot so much a thought experiment as a single neuron misfire.
add, search, annotate, link, view, overview, recent, by name, random
news, help, about, links, report a problem
browse anonymously,
or get an account
and write.
register,
|
|
|
Please log in.
Before you can vote, you need to register.
Please log in or create an account.
|
... in which to go bobbing out from under the tyranny of context and across the broad oceans of universals.
This idea is closely related to 'Engineering as a Philosophical Paradigm'. (See link).
As Professor McLuhan observed back in the '60s, there's a close connection between a message and the
medium in which it is expressed. This proposed piece of software, the Perpetual Notion Machine, is intended as an expression of the fit between the EPP type of philosophy and the medium of hypertext.
What does it do, exactly? Well, it manages lines of reasoning stored in a 'nodes and edges' structure so that, instead of having to pick out logical connections from the rhetorical flow of a one-dimensional text document, you can have them formally specified; but the building blocks of what you are reading or writing are still passages of natural language text, not mathematical symbols.
Each node would hold some text, which you could loosely call a proposition (though it doesn't have to be a proposition in any strict sense, just a passage of text which seems intelligible to someone somewhere).
Examples of the relationships represented by edges would include, in simple cases, 'node X is a reason for accepting node Y' or 'node X in combination with node Y constitutes a reason for accepting node Z' or, in a more abstract case, 'node A describes a general principle according to which a node of a type to which node X belongs constitutes a reason for accepting a node X', related to node X in the way defined in node B'.
You could build up piles of abstraction in much the same way as you can build up complex expressions in a functional programming language. The Machine provides a guarantee of coherence, if not intelligibility, and helps you (using an XML-based hypertext interface) to navigate your own lines of thought, and thereby saves you from disappearing up your own bottom.
There are at least two types of perpetuity ascribable to the products of this machine. One is that lines (or trees) of reasoning can be extended indefinitely, forwards, backwards, round in circles or, most interestingly, outwards.
(What I mean by an 'outward' extension is one that introduces a 'more meta' proposition to bind together two or more specific propositions. In a debate between two opposed points of view, such an 'outward' move is like an outflanking manoeuvre).
The other is that, assuming we are using Engineering as our Philosophical Paradigm, every time we encounter a new world view which involves different principles of reasoning or justification, and/or different axioms, the machine is sufficiently general for us just to feed in the new material and keep on cranking out... well... notions.
(?) Engineering as a Paradigm...
Engineering_20as_20a_20Paradigm_2e_2e_2e
Engineering as a Philosophical Paradigm [pertinax, Jun 22 2006]
FreeMindMaps
http://www.freemindmaps.net/
Share Mindmaps and Mind Mapping Applications with communities with similar ideas in mind. [Ayelis, Jun 23 2006]
[link]
|
| |
Blast, I don't understand it. |
|
| |
Sorry, [zeno]; I will read through it again tomorrow, and see if I can make it clearer. Can you give me any hints as to where I need to add information? |
|
| |
'nodes and edges' : connected with eternal golden braid? |
|
| |
//hints as to where I need to add information// |
|
| |
Add? add? No, please. Less is more. |
|
| |
// Machine provides a guarantee of
coherence, if not intelligibility// - how
does it do this ? the premise are
otherwise interesting, so + for an
interesting idea, if a tad clouded - but
here is an inherent contradiction
between something that on that on one
hand is pointing towards clarity: "and
thereby saves you from disappearing up
your own bottom', and on the other
hand presenting endless chains of
possibilities "One is that lines (or trees)
of reasoning can be extended
indefinitely" So which one does it do? |
|
| |
I believe this is baked: It's called Prolog. Perhaps your are suggesting a social network version of the Prolog language? |
|
| |
From wiki: "Prolog is based on first-order predicate calculus; however it is restricted to allow only Horn clauses. See Logic programming for a discussion of the relationship of Prolog to mathematical logic. Execution of a Prolog program is effectively an application of theorem proving by first-order resolution. Fundamental concepts are unification, tail recursion, and backtracking."
There's you application; but I still do not see the outward move. |
|
| |
this is somewhat reminiscent of an epistemological theory known as 'holistic coherentism' - nodes of 'true' statements connected together in a logically consistent manner. |
|
| |
I wrote an essay once arguing that holistic coherentism can be 'mapped onto' coherentism (coherentism is essentially circular reasoning that the 'holistic' coherentism tried to overcome). |
|
| |
Also, I agree with the allusion by Ling - if you follow this to its logical conclusion you're eventually going to have to face Godel. |
|
| |
Is this a pun? We don't have any pun here. |
|
| |
//Prolog is based on first-order predicate calculus// This needs to do higher-order predicates, I think, so Prolog isn't quite there. |
|
| |
Fair enough. The outflanking mane.., manu..., stratagem is provided by the user how? Sorry if the question is a bit stupid, I'm trying to form this thing in my mind with limited success. |
|
| |
[xenzag], I hope this answer doesn't annoy you, but clarity and closure are not always the same. This may be a lame example, but I can be fairly clear about meaning of pi even if I can never write down all the digits. And thank you for the bun. |
|
| |
The kind of disappearing up the bottom from which it protects you is not the objective kind, where you find that you have argued in a circle, but the subjective kind where you're not quite sure where you've argued yourself to, and can barely find your own way back, let alone convince another person to ... |
|
| |
<splutch>
<muffled> Damn! Again? Boy, it's dark in here. </muffled>
<retracts>
<shampoos> |
|
| |
Oh yes. [James Newton], notwithstanding the point about higher-order predicates, some of the internal workings of this machine probably would resemble an implementation of Prolog, and the day-to-day use of it might be somewhat wiki-ish, so you're on the right track. |
|
| |
Please be patient here, because it's been a long time since I knew anything about Prolog, and even then I didn't know much, but I think one important difference is that the PNM would not generally do automatic graph reduction (does Prolog do that, or am I confusing it with a functional programming language?) and would not generally auto-add nodes; it's more a medium for presenting and managing human reasoning than a device for doing automatic reasoning. |
|
| |
[xaviergisz], you're touching on the profound bits now. The PNM won't stop you from producing a cyclical argument topology. (I just made that phrase up, but it sounds good). It will, however, help you to see when you have done so. You can then make your own value judgement as to whether you see that as a problem. It *will* be a problem if you were hoping to build something that is *cogent* in the literal sense of being able to *compel* agreement in a reasonable person (if you can find such a person). |
|
| |
This is why the connection with Engineering as a Philosophical Paradigm is important. |
|
| |
The mathematical/Anglo-Saxon paradigm seems tied up with a need for cogency in addition to coherence, whereas the poetic/Continental paradigm seems not interested in either. A machine that helps with coherence without promising cogency needs the third paradigm that I was trying to explain in my earlier idea (see link). And possibly vice versa. [looked for an illuminating McLuhan link - couldn't find one. May try again later - don't want to get into mfd advocacy] . |
|
| |
If I remember rightly, Godel's theorem is about things being true but unprovable. (I would check, but my children have pinched my golden braid for a washing line). This would only be a problem if the PNM were offering to prove everything that is true, within a single system, which it isn't. But maybe there's some other bit of Godel that I'm overlooking. That 'within a single system' phrase is important, because the PNM is intended to be able to launch several distinct messages (sets of propositions) into the ocean in their separate bottles, and not try to fit more messages into one bottle than its users feel they need. |
|
| |
[methinksnot], the most recognizable kind of outflanking nowadays (maybe the only kind?) is deconstruction. If I may change the metaphor... |
|
| |
<puts down bottles, gets out toy box> |
|
| |
... deconstruction is commonly used rather rudely, to sweep all the lego bricks on to the floor. (I know how half-bakers like lego). I imagine a more polite kind of deconstruction, in which, for the purposes of one particular body of propositions, you say 'May I have those white blocks from your lego snowman, to help me finish my lego polar bear?' but everything else on the table stays up - and meanwhile, on another table, with another set of bricks, the snowman can stay up while the polar bear is sacrificed in pursuit of some other goal. |
|
| |
Finally, [nihilo] - pun? Perish the thought. Puns are evil and must be ... er... punished. |
|
| |
[pertinax], I'm wondering where this might be useful. Normally when philosophising, I can distill my argument down to a single 'axiomatic truth' (and contrast this against my opponent's 'axiomatic truth'). |
|
| |
It is rare that an argument is based on multiple axioms that are combined into multiple layers of logic. If this is where your philosophising/argument is at then it's too complex and you've probably missed the point (or it should be re-stated in single axiomatic form). |
|
| |
This is usually called an "inference engine", and has been an existing topic of research in expert systems for at least twenty years. |
|
| |
[Pert] That one's so bad it hurtinax. |
|
| |
[pertinax]// but clarity and closure..// I
didn't question that - I questioned the
contradiction between attempting to
reduce possibilities and at the same
time creating more. My question
therefore remains unanswered. Better
hurry with your reply though, now that
this is mfd |
|
| |
//I hope this answer doesn't annoy
you// - I should also point out that only
stupidity annoys me, and I know how to
bite back, so don't ever worry about
annoying me. Also, that is a painfully
corrupted interpretation of Gödel.
Shame in you for that indiscretion. |
|
| |
[jutta], it's not an inference engine. It doesn't, itself, make inferences. It's a device for managing and navigating human inferences. I don't think inference engines can do the outward/deconstructive move, any more than Prolog can (see my response to [James Newton]). |
|
| |
[xaviergisz], I think it would be useful in areas fraught with politics, (including, but not limited to the office politics which sometimes interfere with technical specifications). In such contexts, you could use it to map out what common ground exists (and clarify the limits of that common ground) between individuals and groups whose world views are more than one axiom different, but who still need to work together in some areas. |
|
| |
You could also use it to play 'find the Hegelian synthesis'. Which is similar to the political application, but a bit more abstract and academic. |
|
| |
//contradiction between attempting to reduce possibilities and at the same time creating more// |
|
| |
I don't think I said anything about reducing possibilities, did I? And my apologies to Godel. |
|
| |
[pertinax] //I don't think I said anything
about reducing possibilities, did I?//
Yes - you did. |
|
| |
//instead of having to pick out logical
connections from the rhetorical flow of
a one-dimensional text document, you
can have them formally specified// =
reduction |
|
| |
//lines (or trees) of reasoning can be
extended indefinitely// = enlargement |
|
| |
All of which = contradiction |
|
| |
Ah! I see what you mean now, [xenzag]. Well, perhaps 'formally specified' was the wrong phrase. What I was trying to convey was the difference between, for example, this... |
|
| |
<premise a>
Je pense.
</premise a> |
|
| |
<conclusion x derived from premise a>
Je suis.
</conclusion x derived from premise a> |
|
| |
<mechanism supposed to link a to x>
All predication pre-supposes existence. [I know that may not be quite right, but perhaps it'll do as an illustration]
</mechanism supposed to link a to x> |
|
| |
(...and we could then add other nodes discussing whether or why we are inclined to accept the content of this node.) |
|
| |
Now, this kind of formalisation is not reductive, just descriptive... unless there's something else I've missed. |
|
| |
That reminds me, [xaviergisz], of another application of the idea; you can transcribe existing texts into the machine for study purposes and (if you feel really ambitious) use the machine to link together texts which refer to each other. Obviously, there will sometimes be cases where an existing text can be transcribed into the machine in more than one way. These cases will make good talking points for teaching purposes, and add to the value of the transcription process as a learning exercise. |
|
| |
Also, [methinksnot], those 'golden braid' references to Hofstadter's book on Godel (Ling, Xavier) have reminded me of another kind of 'outward' move; one of the things discussed in that book is the word 'mu' (derived from Zen?) which 'un-asks a question'. It implies something like 'I don't accept everything in the background to that question, so let's step back and consider the context in which it was asked'. It's the kind of thing you might respond to 'When did you stop beating your wife?' |
|
| |
Thank you [pertinax], but don't trouble yourself, it is over my head entirely. I liked to read the anno's though. |
|
| |
It is better to say penser donc etre instead of je pense donc je suis. |
|
| |
At first I thought... "Hey, that's the internet, right?" Nodes and Anchors... But then I saw the word "Hypertext", and I realized you were talking about a website. In fact, a website a whole lot like Wikipedia. Isn't that well-baked? |
|
| |
I've had enough - I'm off now to donc a
donut. |
|
| |
[pertinax] Now, I see "Notions" again, and I think, not as much Wikipedia (because they delete all your half-baked notions), but more a site like this. Only, either, completely unmoderated, or done entirely with Flash-based Fridge-magnets. Is this what you're talking about, Pert? |
|
| |
[edit: Well, took a while, but I think I found a good link for you. added: freemindmaps] |
|
| |
//It is better to say penser donc etre instead of je pense donc je suis// Thank you, [zeno], from one pedant to another ;) |
|
| |
Enjoy your donut, [xenzag]. |
|
| |
[Ayelis], I like the sound of a flash-enabled fridge-magnet. What I had in mind was more strictly structured than the internet or a mind map, and more logically complex than Wikipedia, but less so than an inference engine... |
|
| |
...but all in all, I think it's time I just went away and tried to bake it. And if I succeed (this may not be any time soon), I may invite you all back so you can say 'Oh. Is that all he meant?' |
|
| |
////It is better to say penser donc etre instead of je pense donc je suis// Thank you, [zeno], from one pedant to another ;)//
Better yet: cogito ergo sum |
|
| |
"Cogito ergo sum" translates as "I
think therefore I am", and whereas
thought implies being, which is what
"penser donc etre" can be translated as,
it does not imply a persistent subject of
consciousness. Therefore, although
that is what Descartes wrote, he
assumed too much by including his idea
of a self in his statement. It would have
been more justifiable for him to have
written "Cogitatio essentiam significat",
because as it is he has taken a huge
logical leap which he hasn't tried to
justify. |
|
| |
I bun it stimply because it would save me so much time trying to read Treon's posts. |
|
| |
A university near me has just accepted my proposal to build a version of this as a doctoral research project. |
|
| |
Yes, that is eighteen years after I posted this idea, and it may be some years more before any results appear. And even then, they may be fishboned or MFD-ed by proper professors. But we'll see. |
|
| |
I wonder how [mindey]'s getting on with his thing. It's not the same thing as this, but it's not entirely unrelated, either. It's a pity we haven't seen him here for a while. |
|
| |
A one dimensional text document follows. |
|
| |
If it hasn't already been suggested...
//perpetual notion machine// |
|
| |
[marked-for-tagline] seconded |
|
| |