Wombling

« previous post | next post »

The second talk in a workshop on "Natural Algorithms", to be held at Princeton on Nov. 2-3, is Jorge Cortés, "Distributed wombling by robotic sensor networks". But you don't need to be able to attend the workshop in order to learn about this fascinating topic, since the author has recently published a version of the same material. The abstract:

This paper proposes a distributed coordination algorithm for robotic sensor networks to detect boundaries that separate areas of abrupt change of spatial phenomena. We consider an aggregate objective function, termed wombliness, that measures the change of the spatial field along the closed polygonal curve defined by the location of the sensors in the environment. We encode the network task as the optimization of the wombliness and characterize the smoothness properties of the objective function. In general, the complexity of the spatial phenomena makes the gradient flow cause self-intersections in the polygonal curve described by the network. Therefore, we design a distributed coordination algorithm that allows for network splitting and merging while guaranteeing the monotonic evolution of wombliness.

The OED has wombling, -long, adv., "with the belly on the ground or along a surface", Obs. rare:

13.. K. Alis. 5658 (Laud MS.) Þe Addres..to-cleueþ wombelyng. c1460 J. RUSSELL Bk. Nurture 451 Lay your cony wombelonge vche side to þe chyne.

However, the OED has so far missed prof. Cortés' use, which is a technical one, going back about 20 years. As explained in J.P. Bocquet-Appel and J.N. Bacro, "Generalized Wombling", Systematic Biology 43(3): 442-448, 1994

The magnitude of the spatial change observed simultaneously on a set of surfaces can be described by a method christened "wombling" by Barbujani et al. (1989) … after Womble (1951).

The references are to G.N. Barbujani et al., "Detecting regions of abrupt change in maps of biological variables", Syst. Zool. 38:376-389, December 1989, and William H. Womble, "Differential systematics", Science 114(2961):315-322, 1951.

As far as I can tell from internal evidence, Barbujani et al. were not aware of the partial coincidence between their coinage and its seven-century-old predecessor:

Below, we extend Womble's concept … Use of the systemic function need not be limited to gene frequencies, but seems applicable to the analysis of any continuous variable, including morphological measurements, frequencies of polygenic and environmentally influenced traits, and even environmental variables. We have adopted the term "wombling" for the procedures detailed below.

As you can guess from their description, there is a linguistic connection here that goes beyond simple lexicography. Already in 1990, Barbujani and Sokal published "Zones of sharp genetic change in Europe are also linguistic boundaries", PNAS 87:1816-1819, 1990:

A newly elaborated method, "Wombling," for detected regions of abrupt change in biological variables was applied to 63 human allele frequencies in Europe. Of the 33 gene-frequency boundaries discovered in this way, 31 are coincident with linguistic boundaries marking contiguous regions of different language families, languages, or dialects. The remaining two boundaries (through Iceland and Greece) separate descendants of different ethnic or geographical provenance but lack modern linguistic correlates.

However, I don't know any examples of papers that use wombling directly to analyze the geographical distribution of linguistic variables. Perhaps a reader can provide one. In Womble's original 1951 paper, he writes:

The subtleties of human culture and the intricate relationships of its elements pose problems for which, as yet, there appears to be no adequate mathematics. However, the quantification of even the grosser variables would permit the development of a systemic function which should better define the spatial and temporal structuring of culture fields and their patterns of interaction than do any of the current culture area and culture sequence classifications.

Dialects show analogies to subspecific phenomena. Like biologic varieties, they are involved in the uneasy dynamic equilibrium of their universes of interaction. In isolation they may "speciate" as distinct languages. This methodology would seem well suited to the ordering of the many variants of dialect geography.

It is not feasible at this time to foresee all possible applications of differential systematics. Essentially, it is a development in logic rather than a specialized technique. Consequently, it may be employed in the ordering of numerous realities exhibiting multiple variations through time and space. However, significant interpretation of differing systemic functions requires that individual attention be given to the processes involved in maintaining similarities and developing differences.

[The 1951 Science paper lists William H. Womble's affiliation as "Department of Anthropology and Sociology, University of California, Los Angeles", but I 've failed to find out anything else about him, including any other publications, with the possible exception of this patent.]



10 Comments

  1. rootlesscosmo said,

    October 31, 2009 @ 10:26 am

    But cf.:

    http://www.imdb.com/title/tt0076931/

    [(myl) Eerie. Because what they are doing — "strange creatures who live on Wimbledon Common and pick up the litter left by the humans" — might almost be considered to involve a form of "distributed wombling"!]

  2. Robert Cumming said,

    October 31, 2009 @ 11:00 am

    I was also more or less certain that a wombling algorithm was going involve garbage collection.

    Interestingly, the Urban Dictionary claims that wombling is also travelling using both surface public transport and underground trains for the same journey. Source: Mike Batt's legendary and in this context highly informative Wombling Song.

  3. Alan Palmer said,

    October 31, 2009 @ 11:00 am

    That was my first thought too, when I saw the word "wombling". The Wombling Song: http://www.youtube.com/watch?v=FZ2mJPSccvo

  4. Ed Cormany said,

    October 31, 2009 @ 11:00 am

    so then it turns out that my intuition to pronounce it /wɑmb(ə)lɪŋ/ instead of /wu:mlɪŋ/ was correct, unless i've got the wrong pronunciation for Dr. William H's surname. i'm not 100% sure i understand the engineering concept, but i think the sound similarities of 'wombling' with 'ambling' and 'wombliness' with 'wobbliness' seem appropriate.

  5. kyle gorman said,

    October 31, 2009 @ 11:51 am

    the dialect geography lit includes papers on automating the creation of isoglosses and similar. unfortunately, almost all of them use categorical variables, then score different isoglosses for consistency and homogeneity. another approach (mostly associated w/ dutch authors) is using subspace methods on categorical variables.

    [(myl) I'm familiar with a lot of work of this type, but I don't know any instances that use the term "wombling" or an explicit version of the algorithms discussed in the wombling literature. I'm not sure why this is (assuming that it's true) — perhaps there's a good reason, or perhaps it's just a matter of incomplete intellectual diffusion.]

    one problem with a naive wombling approach to dialect geography is interdependence of the parts: when you raise and tense your (ae)'s, it usually has consequences for the rest of the vowel system.

    [(myl) The same sort of thing is true for the space-time distribution of genomic and phenotypic traits, which co-vary for environmental, internal, and historical reasons. As I understand it, this doesn't matter to Womble's concept of "differential systematics", any more than the concept of spatio-temporal variation in (say) an electromagnetic field is invalidated by the fact that it has non-uniform causes.]

  6. W. Kiernan said,

    October 31, 2009 @ 6:27 pm

    I can not believe that I am not finding this word "wombly" in dictionaries. I've been using it all my life. It means "like a drunk walks."

  7. mollymooly said,

    October 31, 2009 @ 8:53 pm

    @W. Kiernan:
    Almost:
    "wobbly" (like a drunk walks) + nasalisation (like a drunk talks) = "wombly"

  8. Jerry Friedman said,

    November 1, 2009 @ 6:42 pm

    @W. Kiernan: I think the spelling "wamble" may be more common. In addition to the sense you mentioned, the NSOED gives "to be queasy, to rumble queasily", and transitive and intransitive senses of turning and twisting. In the "stagger" sense they say it may come from "walm" by metathesis. "Walm" seems to be a dialect verb for various gushing, billowing, bubbling, swirling motions, with cognates in other Germanic languages.

    "Wamble" reminds me of a review by Algis Budrys, must have been, of a book by the sf writers Poul Anderson and Gordon Eklund. Budrys blamed the book's inept writing (as he saw it) on Eklund, because Anderson "does not wamble like a spastic turkey on a skating rink." If I remember correctly.

  9. outeast said,

    November 2, 2009 @ 6:21 am

    …and of course there's the Wamblers Association in the UK.

  10. Ray Girvan said,

    November 2, 2009 @ 8:53 am

    I wondered about the connection to "wamble" too. In The Mayor of Casterbridge it's used in the meandering-path sense re the drunken furmity woman:

    Stubberd, with a suppressed gaze of victorious rectitude at the old woman, continued: "Was standing myself. She was wambling about quite dangerous to the thoroughfare and when I approached to draw near she committed the nuisance, and insulted me."

RSS feed for comments on this post