electro-music.com   Dedicated to experimental electro-acoustic
and electronic music
 
    Front Page  |  Radio
 |  Media  |  Forum  |  Wiki  |  Links
Forum with support of Syndicator RSS
 FAQFAQ   CalendarCalendar   SearchSearch   MemberlistMemberlist   UsergroupsUsergroups   LinksLinks
 RegisterRegister   ProfileProfile   Log in to check your private messagesLog in to check your private messages   Log inLog in  Chat RoomChat Room 
 Forum index » Discussion » Composition
A very brief book review:
Post new topic   Reply to topic Moderators: elektro80
Page 2 of 3 [71 Posts]
View unread posts
View new posts in the last week
Mark the topic unread :: View previous topic :: View next topic
Goto page: Previous 1, 2, 3 Next
Author Message
bachus



Joined: Feb 29, 2004
Posts: 2922
Location: Up in that tree over there.
Audio files: 5

PostPosted: Mon Nov 21, 2005 9:34 pm    Post subject: Reply with quote  Mark this post and the followings unread

brinxmat wrote:
I expect that you already recognize what I am about to say: L/J's finite formal set should be applicable in a practical sense; that an adequate formal description should be able to produce something if engineered to do so. ....


It certainly has occurred to me that some of L/J's GTTM could be “reverse engineered’ to act as pattern generators. But I’ve interpreted most of the algorithmic content of this book as being related to pattern recognition. And it seems to me much that material is indeed “applicable in a practical sense.” I.e. it can be used as the basis for computational algorithms that recognize patterns.

brinxmat wrote:
I suspect that L/J's viewed their description as broadly equivalent to the kind of science produced by Newton, so how come they seem to be saying that it shouldn't be applied?


My recollection is that L/J are extremely cautious in the claims they make for their theory stressing that it does not, in any way, touch on salience or musicality; dealing only with very elementary structural abstractions. Thus this Generative Theory of Tonal Music is quite incapable of generating music.

Further, I don’t think they are saying it shouldn’t be applied. It seems to me they are far more careful and precise in their choice of words than most and I think they meant exactly what they said (slightly paraphrased from an earlier quote above): " It is an error to think that a musical grammar should be an algorithm that composes music." That is not equivalent to saying musical grammar (as algorithm or no) shouldn’t be applied.

Is it fair to so recast what they said and then question why they seem to be saying it?
Back to top
View user's profile Send private message Visit poster's website
brinxmat



Joined: Oct 24, 2005
Posts: 262
Location: Norway

PostPosted: Wed Nov 23, 2005 4:33 am    Post subject: Reply with quote  Mark this post and the followings unread

OK, I concede that I was not being fair to L/J, but I maintain that there is some validity to my questioning the "disclaimer", because:

Quote:
It certainly has occurred to me that some of L/J's GTTM could be “reverse engineered’ to act as pattern generators. But I’ve interpreted most of the algorithmic content of this book as being related to pattern recognition.


As you seem to agree: pattern recognition and pattern production are two sides of the same coin; the problem with L/J's approach is — paradoxically — the lack of systematisation of preference rules in a computational sense.

I think that my major problem is that I have difficulty accepting that grammars serve any function if they are not practical; L/J's approach is practical in parts (you yourself say that "that some of L/J's GTTM could be “reverse engineered’ to act as pattern generators"), which means that a portion (or even a good portion) of it could not. The problem of application seems to me to return to the concept of pseudo-patternality (here, the bits that seem to be applicable).

An interesting thought regarding preference rules: Damasio's theory of Somatic Markers.

Regarding this and patternality, I will return.

_________________
-- Say "&Eth;onne hit wæs hrenig weðer"
Back to top
View user's profile Send private message
brinxmat



Joined: Oct 24, 2005
Posts: 262
Location: Norway

PostPosted: Mon Feb 06, 2006 6:16 am    Post subject: Reply with quote  Mark this post and the followings unread

create, open, edit, view, print, play, rename, move, copy, delete, attributes, properties, search/find, and permissions.

OK, so we've got our finite-state pattern parser (written in PROLOG, hehe); we convert our sound/notation to machine-readable format, and feed it into the parser.

What happens? Our set of (pro)grammatical rules recognizes a number of patterns and dumps others. A finite-state recognizer of this kind would essentially prove that humans do one thing, while finite-state parsers do another.

Time to beef up the set of rules (and why not?) We manage to increase the set of recognized patterns to include a number of pieces, but it still fails at various points when presented with a new piece.

We could argue that the finite set of rules was the problem; that the expression of rules in this limited way would make it impossible to generalize across similar phenomena. But is this really the case?

Even a set of generalized rules fails to recognize novel phenomena that occur beyond the limits of its generalization because, here, recognition relies on concrete experience (or in this case an encoded rule).

If we make a new, more capable parser, fit it with recognition rules that recognize fuzzily, we get a parser that will a) recognize a wide range of musical patterns, b) recognize a lot of non-musical phenomena as music and c) be unable to apply preference rules. Here, c) is especially important, because preference rules of another kind would need to be applied to find out which of the (overlapping?) rules fits best.

If the default methodology for applying rules is first-alphabetically, then we can assume that this is not optimal, because it recognizes a lot of non-musics. If we start tinkering with the preference-rule logic, we have to be very careful to avoid the kind of problem that was encountered with the finite-state model. In fact, we can't avoid some systematization that will cause us to return to the finite-state approach in the end.

All of this isn't very efficient, because we're either patching holes, or adapting rules to fit new phenomena. There is another way.

_________________
-- Say "&Eth;onne hit wæs hrenig weðer"
Back to top
View user's profile Send private message
bachus



Joined: Feb 29, 2004
Posts: 2922
Location: Up in that tree over there.
Audio files: 5

PostPosted: Mon Feb 06, 2006 7:10 pm    Post subject: Reply with quote  Mark this post and the followings unread

brinxmat wrote:
There is another way.


And it is ? Please.

_________________
The question is not whether they can talk or reason, but whether they can suffer. -- Jeremy Bentham
Back to top
View user's profile Send private message Visit poster's website
brinxmat



Joined: Oct 24, 2005
Posts: 262
Location: Norway

PostPosted: Tue Mar 14, 2006 1:00 am    Post subject: Reply with quote  Mark this post and the followings unread

Still being a bit enigmatic. Sorry.

I need to bang some ideas around:

Is a pattern that does not occur (or occurs only once) a pattern?

If it can be described as a pattern, but is unattested, then the usefulness of its description is negated, but is the description disproved too?

_________________
-- Say "&Eth;onne hit wæs hrenig weðer"
Back to top
View user's profile Send private message
bachus



Joined: Feb 29, 2004
Posts: 2922
Location: Up in that tree over there.
Audio files: 5

PostPosted: Tue Mar 14, 2006 4:38 am    Post subject: Reply with quote  Mark this post and the followings unread

brinxmat wrote:
Still being a bit enigmatic. Sorry.

I need to bang some ideas around:

Is a pattern that does not occur (or occurs only once) a pattern?

If it can be described as a pattern, but is unattested, then the usefulness of its description is negated, but is the description disproved too?


The word pattern has many related definitions, but also some that are very specific in specific contexts. It would help a great deal if you would provide the definition you are working with (or working to create?).

I do see the importance of the question, especially in the context of hierarchical cadential domains.

_________________
The question is not whether they can talk or reason, but whether they can suffer. -- Jeremy Bentham
Back to top
View user's profile Send private message Visit poster's website
bachus



Joined: Feb 29, 2004
Posts: 2922
Location: Up in that tree over there.
Audio files: 5

PostPosted: Thu Mar 16, 2006 5:46 pm    Post subject: Reply with quote  Mark this post and the followings unread

brinxmat wrote:

If it can be described as a pattern


Hmmm, was I asking you to answer your own question? Sorry:(

My mind was distracted by the fact that by one definition a pattern can be a template from which a single instance of its realization is cut.

So the remaining problem I have is with "described." By analogy would a template be a description?

_________________
The question is not whether they can talk or reason, but whether they can suffer. -- Jeremy Bentham
Back to top
View user's profile Send private message Visit poster's website
brinxmat



Joined: Oct 24, 2005
Posts: 262
Location: Norway

PostPosted: Fri Mar 17, 2006 1:45 am    Post subject: Reply with quote  Mark this post and the followings unread

The question was ambiguous, but in some senses I meant both (curious, no?) The description could be the logical description -- a description of structure -- or a common-sense descripton -- the recognition of a pattern. Either of these meanings would fit, and I am not entirely opposed to the question being answered from either perspective, because I think that both answers would enlighten what I am thinking about.

To my mind:

If a pattern is produced, then it can be produced in a number of ways:

a) from a template (a typical assumption for structuralist thinkers)
b) through a process of negotiation and interaction (a typical assumption for interactive/dialogical thinkers)
c)...more?

In the case of a), the template necessarily pre-exists the pattern, in the case of b) "the template" is cut at the same time as the pattern is produced (thus the template is not necessary, because the pattern is the template itself -- unless you hold that the template is a necessary logical step).

If you maintain a), then you assume some fixed mental structure, if you assume b) then you pave the way for a notion of the creative production of structure. The creative process may cause some cystalisation that causes a repetition of this pattern to be recognised, but this is not necessarily the case because:

If a pattern is perceived, then it can be said to be:

a) recognised on the basis of fixed template matching,
b) on the basis of some dialogue (in the same way as dialogic production), or
c)...something else?

Now, I recognise that bandying the word "dialogue" about doesn't help much, but it is a way of saying that a pattern is recognised without reference to a template or paradigm -- i.e. without structure.

"Something else" is a get-out clause because I am not feeling entirely sure that this tack is productive, and I am sure that there is a) more than meets the eye, and b) very basic things I have ignored.

(This wasn't very coherent, sorry.)

_________________
-- Say "&Eth;onne hit wæs hrenig weðer"
Back to top
View user's profile Send private message
bachus



Joined: Feb 29, 2004
Posts: 2922
Location: Up in that tree over there.
Audio files: 5

PostPosted: Sat Mar 18, 2006 5:57 am    Post subject: Reply with quote  Mark this post and the followings unread

I have to say I lack the competence to deal with this on a purely formal level and must bumble around in analogy for the most part. And I am struggling with the elementary definition of pattern again. May any data set that cannot be described by a random distribution be called a pattern?

This is very interesting stuff and given the subtlety of these ideas there is certainly no need to apologize for lack of coherence during the process of exploration.

_________________
The question is not whether they can talk or reason, but whether they can suffer. -- Jeremy Bentham
Back to top
View user's profile Send private message Visit poster's website
brinxmat



Joined: Oct 24, 2005
Posts: 262
Location: Norway

PostPosted: Sat Mar 18, 2006 1:31 pm    Post subject: Reply with quote  Mark this post and the followings unread

You raise an interesting point: I ask the counter question though: how do we recognize what is random? Perhaps a definition of pattern is that we _see_ the connections?

Some thoughts on what patterns are:

It was pi day recently, and I found out that pi was a so-called irrational number. While the infinite continuation of pi is without any order or pattern -- the internal structure of the continuation does not reveal any recognizable patterns -- I would claim that the calculation of pi predicts the internal structure of the continuation of pi quite adequately. Making the internal structure of the continuation of pi regular, according to a pattern. That sounds circuitous, but I don't believe it is. I think that most people would agree that a set of data generated from a ratio is predictable on the basis of the ratio. Expecting internal structure for the data ignores the key given from the ratio. The pattern is there, it is just repeated once.

At the same time, non-continuous patterns -- my hobby horse, where patterns interlace, hop over one-another, these aren't typically called patterns, because they aren't predictable in the usual way of casually accepted patterns -- might well be more enlightening regarding patterns in non-traditionally-patterning sequences.

_________________
-- Say "&Eth;onne hit wæs hrenig weðer"
Back to top
View user's profile Send private message
bachus



Joined: Feb 29, 2004
Posts: 2922
Location: Up in that tree over there.
Audio files: 5

PostPosted: Sat Mar 18, 2006 2:05 pm    Post subject: Reply with quote  Mark this post and the followings unread

brinxmat wrote:
I would claim that the calculation of pi predicts the internal structure of the continuation of pi quite adequately. Making the internal structure of the continuation of pi regular, according to a pattern.



It is not at all clear to me that one can justify equating pattern and algorithm. But that's a feeling not a logical argument.

_________________
The question is not whether they can talk or reason, but whether they can suffer. -- Jeremy Bentham
Back to top
View user's profile Send private message Visit poster's website
brinxmat



Joined: Oct 24, 2005
Posts: 262
Location: Norway

PostPosted: Sun Mar 19, 2006 12:54 pm    Post subject: Reply with quote  Mark this post and the followings unread

It is the result of the calculation according to the ratio that is the pattern, not the ratio itself.

If we maintain that all non-random sequences are patterns, then pi's continuation is a pattern.Its definition as a pattern must be different from "sequenced by internal mathematics", to the extent that it is "sequenced by external factors". Perhaps all patterns aren't predictable — perhaps they cannot be described only by reference to themselves.

_________________
-- Say "&Eth;onne hit wæs hrenig weðer"
Back to top
View user's profile Send private message
bachus



Joined: Feb 29, 2004
Posts: 2922
Location: Up in that tree over there.
Audio files: 5

PostPosted: Mon Mar 20, 2006 6:08 am    Post subject: Reply with quote  Mark this post and the followings unread

brinxmat wrote:
I would claim that the calculation of pi predicts the internal structure of the continuation of pi quite adequately.


I’m going to have to ask you to backup a bit. Despite real effort, I am unable to decipher the meaning of the phrase “internal structure of the continuation of pi.” And am therefore at a dead end.

Edit :

Perhaps I should note that while beating my head against this wall it occurred to me that pi, as a data set, is random and thus without pattern, but in the context of mathematics it is “non random” and meaningful. I felt this might be a clue, but didn’t know what to do with it, and so am still without one.

_________________
The question is not whether they can talk or reason, but whether they can suffer. -- Jeremy Bentham
Back to top
View user's profile Send private message Visit poster's website
brinxmat



Joined: Oct 24, 2005
Posts: 262
Location: Norway

PostPosted: Mon Mar 20, 2006 1:33 pm    Post subject: Reply with quote  Mark this post and the followings unread

sorry, my boob. What I mean is "everything after the the decimal point", ie .14159265358979323846264338…

See here for fun (you can entre your birthdate in ISO format, for example -- now that IS fun).

_________________
-- Say "&Eth;onne hit wæs hrenig weðer"
Back to top
View user's profile Send private message
bachus



Joined: Feb 29, 2004
Posts: 2922
Location: Up in that tree over there.
Audio files: 5

PostPosted: Mon Mar 20, 2006 1:46 pm    Post subject: Reply with quote  Mark this post and the followings unread

brinxmat wrote:
sorry, my boob. What I mean is "everything after the the decimal point", ie .14159265358979323846264338…


Actually I got that part, it is the meaning of the whole phrase that is bugging me.

brinxmat wrote:
See here for fun (you can entre your birthdate in ISO format, for example -- now that IS fun).


Geees, and I thought I was weird Laughing

_________________
The question is not whether they can talk or reason, but whether they can suffer. -- Jeremy Bentham
Back to top
View user's profile Send private message Visit poster's website
elektro80
Site Admin


Joined: Mar 25, 2003
Posts: 21959
Location: Norway
Audio files: 14

PostPosted: Mon Mar 20, 2006 1:57 pm    Post subject: Reply with quote  Mark this post and the followings unread

bachus wrote:
Geees, and I thought I was weird Laughing


Norway is rated by the EU as a safe deposit. You should worry if they put you on a plane and send you off to Norway.

_________________
A Charity Pantomime in aid of Paranoid Schizophrenics descended into chaos yesterday when someone shouted, "He's behind you!"

MySpace
SoundCloud
Flickr
Back to top
View user's profile Send private message Visit poster's website
bachus



Joined: Feb 29, 2004
Posts: 2922
Location: Up in that tree over there.
Audio files: 5

PostPosted: Mon Mar 20, 2006 2:23 pm    Post subject: Reply with quote  Mark this post and the followings unread

elektro80 wrote:
Norway is rated by the EU as a safe deposit. You should worry if they put you on a plane and send you off to Norway.


After what I've just seen of Norwegian food I'd be more than worried puker

_________________
The question is not whether they can talk or reason, but whether they can suffer. -- Jeremy Bentham
Back to top
View user's profile Send private message Visit poster's website
bachus



Joined: Feb 29, 2004
Posts: 2922
Location: Up in that tree over there.
Audio files: 5

PostPosted: Mon Mar 20, 2006 2:53 pm    Post subject: Reply with quote  Mark this post and the followings unread

brinxmat wrote:
... While the infinite continuation of pi is without any order or pattern -- the internal structure of the continuation does not reveal any recognizable patterns -- I would claim that the calculation of pi predicts the internal structure of the continuation of pi quite adequately.


brinxmat wrote:
If we maintain that all non-random sequences are patterns, then pi's continuation is a pattern.


ummmm... stepping through the posts it is not clear how we got from one to the other.

_________________
The question is not whether they can talk or reason, but whether they can suffer. -- Jeremy Bentham
Back to top
View user's profile Send private message Visit poster's website
brinxmat



Joined: Oct 24, 2005
Posts: 262
Location: Norway

PostPosted: Tue Mar 21, 2006 2:46 pm    Post subject: Reply with quote  Mark this post and the followings unread

Quote:
pi, as a data set, is random and thus without pattern, but in the context of mathematics it is “non random” and meaningful.


Exactly.

Perhaps I have an overly simplistic view of things: because pi -- and consequently the structure of the real number that is pi (and its continuation) -- is defined by the ratio of the circumference of a circle to its diameter, the internal structure of said continuation* is a pattern on the basis of its being non-random.

On the other hand, this sequence of numbers contains no internal structure (that can be identified) and is therefore not a pattern according to most analyses -- it is an irrational number (not that lack of internal structure is necessarily a definition of an irrational number).

However, the number is still patterned on the basis of its being repeatable from our template -- the ratio.

*Note that we are forced to talk about a continuation because it doesn't end.

This is a very laboured way of saying what you say in the quote above.

The ratio is the template for the calculation that generates a string of numbers that does not end. The ratio is not a pattern, but it generates a pattern, which may or may not be predictable.

_________________
-- Say "&Eth;onne hit wæs hrenig weðer"
Back to top
View user's profile Send private message
bachus



Joined: Feb 29, 2004
Posts: 2922
Location: Up in that tree over there.
Audio files: 5

PostPosted: Tue Mar 21, 2006 3:47 pm    Post subject: Reply with quote  Mark this post and the followings unread

Am I to take it then that the apparent contradiction exhibited above is to be resolved by an appropriate application of the caribou kefir transform?

Edit:

Please excuse me I'm on a Norwegian joke binge and careening dangerously out of control. Shocked

_________________
The question is not whether they can talk or reason, but whether they can suffer. -- Jeremy Bentham
Back to top
View user's profile Send private message Visit poster's website
brinxmat



Joined: Oct 24, 2005
Posts: 262
Location: Norway

PostPosted: Wed Mar 22, 2006 12:18 am    Post subject: Reply with quote  Mark this post and the followings unread

As long as we are talking about the contradiction between the idea that pi's continuation "is a pattern" at the same time as "not being a pattern", then I suggest rather geiteganning.

The contradiction is important -- that is my point.

_________________
-- Say "&Eth;onne hit wæs hrenig weðer"
Back to top
View user's profile Send private message
elektro80
Site Admin


Joined: Mar 25, 2003
Posts: 21959
Location: Norway
Audio files: 14

PostPosted: Wed Mar 22, 2006 1:07 am    Post subject: Reply with quote  Mark this post and the followings unread

geitganning= some norwegian DIY in-the-shed-shamanism .. sometimes it involves a fair supply of goats.
_________________
A Charity Pantomime in aid of Paranoid Schizophrenics descended into chaos yesterday when someone shouted, "He's behind you!"

MySpace
SoundCloud
Flickr
Back to top
View user's profile Send private message Visit poster's website
bachus



Joined: Feb 29, 2004
Posts: 2922
Location: Up in that tree over there.
Audio files: 5

PostPosted: Thu Mar 23, 2006 3:56 pm    Post subject: Reply with quote  Mark this post and the followings unread

At this point, given all that has gone before, except perhaps the goats, the shed and the shaman, I would tentatively assert that:

In isolation, outside of any context, no data set is a pattern.

Consider the closed ordered set [ 65, 32, 119, 111, 114, 100 ]. In isolation it is meaningless and is not a pattern. In the context of the ASCII it translates to “A word” and as such might be used as a pattern for searching text for that phrase. The closed ordered data set [1,0,0,1,0,0,1,0,0,1,0,0] is also meaningless and not a pattern in isolation and outside of any context. It may, however, be treated as the context for the pattern [1,0,0] so we may say that though it is not a pattern it exhibits a pattern.

_________________
The question is not whether they can talk or reason, but whether they can suffer. -- Jeremy Bentham
Back to top
View user's profile Send private message Visit poster's website
blue hell
Site Admin


Joined: Apr 03, 2004
Posts: 24079
Location: The Netherlands, Enschede
Audio files: 278
G2 patch files: 320

PostPosted: Thu Mar 23, 2006 5:52 pm    Post subject: Reply with quote  Mark this post and the followings unread

bachus wrote:

In isolation, outside of any context, no data set is a pattern.


Hmm, I had drawn the opposite conclusion ... that all ordered data sets are a pattern. Not that that would be more useful of course, after all it's discrimination that brings us further.

It depends a bit. For pattern In it's normal use there would have to be a certain repetition, or mirroring in order to call something a pattern, like the repeated translation in your 2nd example.

But it's not unusual to have a random pattern, or to call your first example a 'search pattern' (indeed). In such a generalized meaning everything that is like an ordered set could be called a pattern. Even repeatability does not seem to be necessary to have a pattern then.

_________________
Jan
also .. could someone please turn down the thermostat a bit.
Posted Image, might have been reduced in size. Click Image to view fullscreen.
Back to top
View user's profile Send private message Visit poster's website
elektro80
Site Admin


Joined: Mar 25, 2003
Posts: 21959
Location: Norway
Audio files: 14

PostPosted: Thu Mar 23, 2006 6:35 pm    Post subject: Reply with quote  Mark this post and the followings unread

This has been an interesting discussion so far. It does however strike me as obvious that we might need a new or at least reinvented vocabulary when discussing modern compositional methods and how these relate to tradition.
_________________
A Charity Pantomime in aid of Paranoid Schizophrenics descended into chaos yesterday when someone shouted, "He's behind you!"

MySpace
SoundCloud
Flickr
Back to top
View user's profile Send private message Visit poster's website
Display posts from previous:   
Post new topic   Reply to topic Moderators: elektro80
Page 2 of 3 [71 Posts]
View unread posts
View new posts in the last week
Goto page: Previous 1, 2, 3 Next
Mark the topic unread :: View previous topic :: View next topic
 Forum index » Discussion » Composition
Jump to:  

You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot vote in polls in this forum
You cannot attach files in this forum
You can download files in this forum


Forum with support of Syndicator RSS
Powered by phpBB © 2001, 2005 phpBB Group
Copyright © 2003 through 2009 by electro-music.com - Conditions Of Use