Fwd: [argunet-users] Debate terminology/glossary

23 views
Skip to first unread message

Sebastian Cacean

unread,
Feb 16, 2015, 5:08:13 AM2/16/15
to argune...@googlegroups.com

---------- Forwarded message ----------
From: Gregor Betz <ggb...@zedat.fu-berlin.de>
Date: Sun, Feb 15, 2015 at 10:17 AM
Subject: Fwd: [argunet-users] Debate terminology/glossary
To: Justin Webster <justin....@gmail.com>, "Cacean, Sebastian (ITAS)" <sebastia...@kit.edu>, "Voigt, Christian (ITAS)" <christi...@kit.edu>




Anfang der weitergeleiteten Nachricht:

> Von: Gregor Betz <ggb...@zedat.fu-berlin.de>
> Betreff: Aw: [argunet-users] Debate terminology/glossary
> Datum: 15. Februar 2015 09:48:14 MEZ
> An: argune...@googlegroups.com
>
> My personal recommendation would be: don't use Argunet at all for setting up a definitions list / glossary etc.
>
> Take as an example our Climate Engineering map, which has a kind of glossary,  http://digbib.ubka.uni-karlsruhe.de/volltexte/1000022371. Here's the workflow:
>
> - We reconstructed the debate with Argunet and exported the debate as graphml file.
> - We improved the layout of the debate with the free yEd graph editor (http://www.yworks.com/en/products/yfiles/yed/) and exported the map as PDF/EPS.
> - With a vector graphics program (e.g. Illustrator) we added additional items (title, glossary, citation info etc. etc.)
>
>
>
>
>
>
> Am 13.02.2015 um 20:54 schrieb Justin Webster <justin....@gmail.com>:
>
>> In my usage of Argunet, I have found precise and unambiguous usage of terminology to be of critical importance in representing both sides of complex debates. For example, if separate premises or conclusions use the same terms differently, then any equivelences and contradictions between such premises are weak or invalid.
>>
>> However, I am not sure how best to represent the definitions of such important terms, or of how to relate them to the rest of the debate. Here is what I have done so far:
>>      • Created a special Sentence: "Definition of terminology."
>>      • Designated one of the Argument Families as "Definitions"
>>      • Every Term gets a distinct "definition" Argument:
>>              • With the Arg's display title of the format "df=TermOrPhrase" or "df=TermOrPhrase (Abbrev)".
>>              • With the Arg's Conclusion prefixed with "df=", and of the format "df=A 'TermX' is a (complete, unambiguous but concise-as-possible description)."
>>              • With the Arg's Conclusion optionally suffixed with something like "(cf. TermG, TermZ)".
>>              • Colored with the "Definitions" Family.
>>              • With the special "Definition of terminology." Sentence solid-linked as the first Premise.
>>              • If any other Terms are used within the Arg's Conclusion, they are solid-linked as Premises (if doing so does not create a circular reference). If both Terms are used in each other's Conclusion/definition, I use my best judgement as to which ones to link as Premises of which other Terms.
>>      • For any "non-definition" Arguments in the Debate that are strongly dependent on the precise definition of a specific Term, that Term's Definition Arg is solid-linked as one of the Premises of the "non-definition" Arg.
>>              • This makes it hard(er) to use the linked term "incorrectly" in the Conclusion of the linking Arg, since the full definition is right there as a Premise.
>>              • This makes it easier to keep non-definitional Arguments' Conclusions concise and precise, since terms do not need to be defined within the Conclusion at all.
>>              • This makes non-definitional Arguments' Conclusions more readable and understandable, since the definitions of any important terms are visible directly within the Arg Editor and easily added to any Map the Arg is used on.
>>              • This also guarantees that any subsequent modifications of definitions are automatically and immediately applied to every other Arg (and on every Map) that has the modified definition linked as a Premise.
>>              • Some Arg's may have multiple definition Args as Premises, if their Conclusion uses multiple Terms with important or potentially-ambiguous meaning.
>>      • Created a "Glossary Map" map, with every Definition Arg in the debate included.
>>              • This makes it easy to see which terms are dependent on which other terms, and also easy to quickly find, review, and reference specific terms.
>>              • I also include all of my "special" Sentences on the Glossary Map. E.g. "Definition of terminology.", "Unsupported Conclusion.", "TODO: Enthymeme(s).", etc.
>>      • If a specific "term" has 2+ conflicting definitions, I split them into distinct terms with unique Arg Titles and names, and am careful to be explicit in using one or the other when modifying specific Sentences and Arguments.
>>              • E.g. "df=TermX df#1 (TermX#1)" and "df=TermX df#2 (TermX#2)".
>>
>> The way I'm doing terminology (above) seems a bit clunky, but I haven't thought of a better way to do it yet.
>>
>> Does anyone else have any tips on how to deal with ambiguous and/or contentious terminology in Debates?
>>
>> --
>> You received this message because you are subscribed to the Google Groups "argunet-users" group.
>> To unsubscribe from this group and stop receiving emails from it, send an email to argunet-user...@googlegroups.com.
>> To post to this group, send email to argune...@googlegroups.com.
>> Visit this group at http://groups.google.com/group/argunet-users.
>> For more options, visit https://groups.google.com/d/optout.
>


Reply all
Reply to author
Forward
0 new messages