Talk:Symbol

Symbols are not data structures

I think it is really a stretch to consider a symbol a data structure. Symbols are more the kind of thing that would be encapsulated in a data structure. Accordingly, I reclassified this under Category:Concepts instead. --Graue 21:23, 4 Oct 2005 (GMT)

Symbols can "easily" be stored raw. --Ihope127 00:31, 5 Oct 2005 (GMT)

Non sequitur. "It can be stored raw" is not a sufficient condition for something to be a data structure. --Graue 17:38, 5 Oct 2005 (GMT)

And besides, symbols can't be stored raw - you need to specify an encoding, and there is no 'natural' encoding (hence why we have so many different character encodings). --Safalra 11:07, 6 Oct 2005 (GMT)

Symbols may be data structures, but they can also be things that cannot be sufficiently represented as data structures - such as the relationship between two data structures. Glypho is ambiguous with respect to "symbols", since it relies on an underlying alphabet of symbols to encode the length-4 glypho symbols. The underlying symbols may be anything (Ive speculated on integers or characters, pixels (color symbols), audio symbols, actions, etc.) and the glypho symbols are the patterns between them. --Wildhalcyon 21:34, 5 Oct 2005 (GMT)

I have no problem considering symbols as degenerate data structures. What is the "sufficient condition" for data structure-ful-ness that you are referring to, Graue? --Chris Pressey 18:09, 6 Oct 2005 (GMT)

None -- what I referred to was an insufficient condition. If symbols are data structures, what data do they structure, and how do they structure it? --Graue 20:34, 6 Oct 2005 (GMT)

I think the clue's in the name - a data structure must structure the data. A symbol is no more structured than an integer (in the mathematical sense, ignore the specific internal representation). A character stream is a data structure for the representation of symbols, but symbols aren't data structures by themselves. --Safalra 21:00, 6 Oct 2005 (GMT)

Symbol independence?

Languages like Udage or Glypho are desinged to have no specific symbol functions as instruction. I'm not sure this is proper word however - patterns of the language do. The thing is, can pattern of the symbols also be referred to a symbol? As traditional symbols have their own rules to be identified(such as, alphabet 'A' has 3 segments and ...), symbol patterns in Udage or Glypho have their own rules too. This may not be the literal meaning of symbol but can be included in the abstract notion of symbol(especially in esolang studies), I think. --Gs30ng 22:23, 6 Oct 2005 (GMT)