To LUGNET HomepageTo LUGNET News HomepageTo LUGNET Guide Homepage
 Help on Searching
 
Post new message to lugnet.generalOpen lugnet.general in your NNTP NewsreaderTo LUGNET News Traffic PageSign In (Members)
 General / 55778
55777  |  55779
Subject: 
Re: Mathematical proof that you can't build anything with LEGO bricks
Newsgroups: 
lugnet.general
Date: 
Mon, 27 Feb 2012 17:55:43 GMT
Viewed: 
19345 times
  
In lugnet.general, Larry Pieniazek wrote:

  
He says “LEGOs” which makes his work suspect right there. :) Not very rigorous. The paper is from 2001... the piece palette has improved somewhat since then.

But generally I’m not sure he’s drawing valid inferences/conclusions. LEGO elements are not neurons, or electronic components, and total network connectivity is not their most important metric. Rather, at least in my view, what is important is whether pieces interoperate well with each other locally. And we know they do. There are more ways to interconnect LEGO elements than any other building system I know.

He would argue that the increase in the number of new types of pieces is actually an indication that the LEGO system is getting less flexible.

But as you say, the system is designed so that every piece connects to at least one other piece type, and most of them fit with most of the other types.

One problem with his method is that he used Peeron inventories to provide the number of pieces and piece types in each set. He didn’t filter this data at all, however. As we know, Peeron inventories have an entry for each element AND color. He simply counted the number of entries in the inventory and called each one a piece type - so he counts a red 1x2 brick as one type and a blue 1x2 brick as a different type. Of course the color has nothing to do with the ability of two pieces to connect, so his data is skewed.

There are also small errors introduced by the variation in inventory methods within Peeron. Some inventories break every element into its atomic parts - I have found some that even list the minifig hands as a separate element. The same is true with elements whose function is always in conjunction with another element (as sold in a LEGO set, that is) - both halves of a hinge piece, for example. These are used as one element in the LEGO model, but Changizi counts them as two types. (Of course builders often DO use the separate parts in various ways, but he is only interested in the ways in which TLC uses the elements in their sets) Also, discrepancies between the “official” piece count and the number of elements in the inventory is pretty common - from errors and from the practice of including extra pieces in the sets.

Minifigs are another thing. These are specialized pieces that are really included as accessories to the main model and not as an integral part of its construction, but Changizi counts each minifig element and accessory as a different piece type.



Message is in Reply To:
  Re: Mathematical proof that you can't build anything with LEGO bricks
 
(...) He says "LEGOs" which makes his work suspect right there. :) Not very rigorous. The paper is from 2001... the piece palette has improved somewhat since then. But generally I'm not sure he's drawing valid inferences/conclusions. LEGO elements (...) (12 years ago, 27-Feb-12, to lugnet.general, FTX)  

34 Messages in This Thread:












Entire Thread on One Page:
Nested:  All | Brief | Compact | Dots
Linear:  All | Brief | Compact
    

Custom Search

©2005 LUGNET. All rights reserved. - hosted by steinbruch.info GbR