diproche / webinterface

1 stars 0 forks source link

Write up a technical glossary #72

Open Korosensei42 opened 5 years ago

Korosensei42 commented 5 years ago

In case other developers pick this project up in the future we should write up a glossary of all technical terms we use. This will also prevent us to mix two equal terms in their meaning or using them interchangibly.

What do you think?

RonjaKaehne commented 5 years ago

i dont think i got it clearly what you mean. technical terms like using the word "Spielwiese"? if you mean that yes we could add a wikipage inside this git repository.

Korosensei42 commented 5 years ago

Yes, this could go in a wiki. I mean technical terms like "Spielwiese", "example", "textFormatter" and so on. Documentation in the code is all fine and dandy, but in a glossary (essentially a dictionary with explenations) you would have all terms in one place.

Korosensei42 commented 5 years ago

Stuff like grammar for certain Spielwiesen, as described in #46 should go in this glossary as well.

Korosensei42 commented 5 years ago

From #46:

With regards to the Spielwiese Aussagenlogik the allowed grammar for formulae should be something like this:

A ::= a | b | ... | y | z (atoms)
F ::= \top | \bot | A | F \wedge F | F \vee F | \neg F | F \rightarrow F | F \leftrightarrow F (formulae)

Note: top and bot stand for verum and falsum respectively. The other keywords also correspond to their standard LaTeX-interpretations, i.e. they stand for and, or, negation, implication, equivalence.

How a sentence containing natural language or expressions from other Spielwiesen should look like, still needs to be done.

TimothyGillespie commented 5 years ago

I introduced / invented a term for the following case:

Assume we have a prolog knowledge basis as follows:

woman(anna). 
woman(eve). 
man(frank).
man(bob).

Then the query ?- woman(X). will return X = anna ; X = eve. In this case anna and eve are fill-in values for X in this query (the logic being prolog fills in possible values for the variable. I couldn't find an exisiting term for these.

Korosensei42 commented 5 years ago

@TimothyGillespie Just so I understand you correctly: Do you have a term for this now or do you not? I don't find a term by you anywhere at least. Or do you mean the word fill-in values?

In the latter case: As known, man and woman are predicates. The arguments of predicates are always terms. Terms can also have so-called sorts (german: Sorten (oft) / Typen (selten) ). So you would say, that anna and eve are of sort name or string, depending on if you have defined an abstract data structure or type name or if you want to use the basic type string.

Also another term that would fit, would be: anna and eve are termvaluations over the Universe of sort name or string respectively. For the paragraph before and this one,

I will leave you a file down below of a little write-up by me. I can also include the previous explanation in the glossary, of course. I will start it on Saturday, I think.

In case you mean the kind of search that prolog does: I think exhaustive search fits it quite well, I think.

Here the file with the the write-up:

Termvaluation.pdf My source was the logic-lecture from our university

TimothyGillespie commented 5 years ago

I think termvaluation was the term I was looking for but couldn't find. Thanks!