symbols for basic sts types?

[jbs1]

migrated from Trac, where originally posted by kohlhase on 27-Jan-2008 5:17am

the baisc types used in the *.sts and *.mts files often variables, should we make them constants, then we need a CD for those that are not in omtypes.ocd, mathmltypes.ocd, sts.ocd

[jbs1]

migrated from Trac, where originally posted by jhd on 27-Jan-2008 8:27am

Not quite sure what is meant here.
If one means

then AbelianSemiGroup is a variable not a constant deliberately: STS (Davenport2000) does NOT impose a fixed classification of types, and the semantics of AbelianSemiGroup are NOT specified. In particular, there are differences between constructive and non-constructive semantics here (Davenport et al1991a). Moving to a fixed classification would be a major change.

[jbs1]

migrated from Trac, where originally posted by kohlhase on 28-Jan-2008 6:57am

Replying to [comment:1 jhd]:

STS (Davenport2000) does NOT impose a fixed classification of types, and the semantics of AbelianSemiGroup are NOT specified. In particular, there are differences between constructive and non-constructive semantics here (Davenport et al1991a). Moving to a fixed classification would be a major change.

James, could you please attach these two papers to this bug? Or point to where they are avaiable on the web?

[jbs1]

migrated from Trac, where originally posted by jhd on 29-Jan-2008 12:17am

Davenport2000 is the standard STS paper: http://portal.acm.org/citation.cfm?id=362014 http://staff.bath.ac.uk/masjhd/OpenMath/sts2.pdf (if there is a copy on OpenMath.org, Google doesn't find it).
Davenportetal1991a is http://portal.acm.org/citation.cfm?id=120694.120699

[jbs1]

migrated from Trac, where originally posted by kohlhase on 25-Feb-2011 12:39pm

rescheduling.

[kohlhase]

moved to OpenMath/OMSTD#39