LANGREITER.COM plain, simple
      START     INDEX
  Dienstag, 28. Mai 2002

David Ness identifies several potential problems with graph visualizations:

  • Necessary Reduction of Dimensionality
  • Lack of (obvious?) Concept Closeness Metric
  • Unclear Value/Danger of Metaphor-based Inferences
  • Decreasing Utility of Maps in general
There Ain't No Streets Mid-Ocean
Diskussion: comments-2002-05-23

Of course, if we limit ourselves to certain classes of "concepts" (does this also imply a reduction in dimensionality?), concept closeness metrics are indeed attainable (Collaborative Filtering data — amzn-vista, "Connectedness" of People, Organizations and Artifacts — much of my VanillaSpace). Generic concept visualization indeed is hard and its value rather questionable (even to me).

The problem of Reduction of Dimensionality can, while not be removed from mapping activities completely, be alleviated by introducing more dimensions of expression into the display (color-, size-coding etc. in addition to position "coding").

En plus, fascination never is quite approachable in a rational way, is it?Flash MX eignet sich auch durchaus dafür, als Richtexteditorbaustelle herzuhalten: [ Ø CRESK • ]

Random Medias Dream Domain (inkl. GA-Drone-Generator) läuft schon mit Tin Can:

funzel 5388 days AGO
Zu dem Power Text von Graham:
Den fand ich ganz gut, und ueber ein solches kalkuel zur
berechnung der komplexitaet bin ich auch schon ein
paarmal gestossen, da mir LOC immer etwas sinnlos
vorkommt. Die Implementierung war mir dann aber zu aufwendig.
Eventuell finde ich aber jemanden, der sowas in Java implementiert (der Autor von Smartcode (ein tools zum Testen, wie gut Java code geschrieben ist)

Dagegen zu einem alten Graham Artikel ueber Power of
> Can anyone name another company with 3 programmers and
> a bunch of people answering the phone that created $50
> million worth of value?

21 employees. Sold to Yahoo for $49 million ($2.3 million per head). Yahoo later bought eGroups (python users) for $432 million (about $3 million per head, iirc). Microsoft bought Cooper & Peters (smalltalk, java) for about $20 million ($10 million per head). Verisign bought Thawte (python) for $575 million ($11 million per head). And so on.

> I can't

Where were you hiding during the dot-com era? ;-) "$3404

andi 5388 days AGO
that makes me cry!!
will code html for food ....

chris 5388 days AGO
Hätte doch ein Python-Startup gründen sollen ... ;-)

David Ness 5388 days AGO
I still owe Alex_S and Chris some comments that I will get to later on, but can't resist mentioning here that _some_ of the deals mentioned in Funzel's quote from lambda should be taken, as we say, with a grain of salt.

I saw details on a lot of deals for one of our companies during the crazy `boom' days of the internet, and a lot of the $20 m offers turnd out to be worth more like $500K when we looked at them closely.`Letter' Stock was a common part of such offers. Some of my academic friends who were quite rich when ICGE, for example, was $212/share are not so happy now that it is 28 cents/share, and $20m of such stock at the peak is worth less than $30K today

funzel 5387 days AGO
Well, you're certainly right. But the point was, that Yahoo bought Grahams company for $2.3 per head. Graham claimed, there are no other examples, but as the writer pointed out, there are indeed others.

The numbers shouldn't be taken as real money, but for comparing value of companies (or percepted value at that time). And then a lisp company is not more valuable than a python company, which prooves Graham wrong.

funzel 5387 days AGO
I couldn't resist, after earl posting the REBOL version:

def foo(i) lambda [describe |n| n +i here] end

In Ruby.

David Ness 5387 days AGO
I take your point. I guess my only demur is that with any `valuation' argument, `the devil is in the details' and what one person might claim is a $20m `deal' might look more like $500k to anotherso---without specific details---the `gross' numbers are really meaningless. Thus rendering any point about valuation quite moot.

funzel 5387 days AGO
You are right. The comparison only makes sense, when we assume that the mix of money and shares was nearly equal.
(And I think, the people bought by MS and given some shares are more happy now than someone who was bought with yahoo shares)

earl 5387 days AGO
re funzel: i'd agree that Ruby is a nice language, but the code you posted above nowhere matches the functionality of neither the REBOL nor the Lisp bits :)

the rebol function for example would be used as follows:
>> foo: func [ 'n ] [ func [ i ] [ set n i + get n ] ]
>> x: 5
== 5
>> bar: foo x
>> bar 2
== 7
>> x
== 7
>> bar 3
== 10
>> x
== 10

that's what accumulators are good for ;)

chris 5387 days AGO
Aja, das Ruby-Beispiel kam mir schon etwas komisch vor ;-) Für { und } sind bitte \{ und \} zu schreiben.

stephan schmidt 5387 days AGO
@earl, chris:
Oh, ich war wohl mental blockiert, weil warum kein objekt nehmen wenn man einen state speichern will ? Ich habe da nur so druecber gelesen und das + ueberlesen.

> def foo(n) lambda [describe |i| n+= i here] end
> bar = foo 3
> 2
> 10

Please log in (you may want to register first) to post comments!



So log in, fella — or finally get your account. You always wanted one.
Nearby in the temporal dimension:
... and 83 of the anonymous kind.
Click on Join us in the chatterbox dimension! for a moderate dose of lcom-talk. This will probably not work in Lynx and other browser exotica.

2009 / 2006 / 2003 / 2002


  Blood Stone
  wet towel
  B Studio
  Pilcrow News
  Nastassja Kinski

Special Entanglement Forces provided by Vanilla

GeoURL RSS 0.92

Uncut Games bei Gameware

Offenlegung gem. 25 MedienG:
Christian Langreiter, Langkampfen
See also: Privacy policy.