renaming basic things about unimath HOT 9 CLOSED

I suggest to introduce the new names as notations for the old ones. This way nothing gets broken and anyone who wants to use new names can.

I also suggest it as the general procedure for changing and streamlining the notations.

On Sep 28, 2015, at 1:41 PM, Daniel R. Grayson [email protected] wrote:

Here are some things I'd like to rename:

setwithbinop -> magma
invmaponpathsincl -> an_inclusion_is_injective
total2 -> totalSpace
I didn't know it before, but "magma" is used that way in Bourbaki's Algebra.

Does anyone else have proposals?

Can we agree on any of them?

—
Reply to this email directly or view it on GitHub #166.

from unimath.

As the old names will not be kept, I propose instead to use the old names
as notations for new ones.

2015-09-28 13:48 GMT-04:00 Vladimir Voevodsky [email protected]:

I suggest to introduce the new names as notations for the old ones. This
way nothing gets broken and anyone who wants to use new names can.

I also suggest it as the general procedure for changing and streamlining
the notations.

On Sep 28, 2015, at 1:41 PM, Daniel R. Grayson [email protected]
wrote:

Here are some things I'd like to rename:

setwithbinop -> magma
invmaponpathsincl -> an_inclusion_is_injective
total2 -> totalSpace
I didn't know it before, but "magma" is used that way in Bourbaki's
Algebra.

Does anyone else have proposals?

Can we agree on any of them?

—
Reply to this email directly or view it on GitHub <
https://github.com/UniMath/UniMath/issues/166>.

—
Reply to this email directly or view it on GitHub
#166 (comment).

from unimath.

On 09/28/2015 02:23 PM, cathlelay wrote:

As the old names will not be kept, I propose instead to use the old names
as notations for new ones.

But then the old names will be used for printing instead of the new ones.

from unimath.

Indeed ... I noticed it after pressing "send"

2015-09-28 14:45 GMT-04:00 Benedikt Ahrens [email protected]:

On 09/28/2015 02:23 PM, cathlelay wrote:

As the old names will not be kept, I propose instead to use the old names
as notations for new ones.

But then the old names will be used for printing instead of the new ones.

—
Reply to this email directly or view it on GitHub
#166 (comment).

from unimath.

Having the old name as a "parsing only" notation for the new name would be
the way to go,
as explained in the manual:

"Remark: Sometimes, a notation is expected only for the parser. To do so,
the option only parsing is
allowed in the list of modifiers of Notation."

On Mon, Sep 28, 2015 at 2:51 PM, cathlelay [email protected] wrote:

Indeed ... I noticed it after pressing "send"

2015-09-28 14:45 GMT-04:00 Benedikt Ahrens [email protected]:

On 09/28/2015 02:23 PM, cathlelay wrote:

As the old names will not be kept, I propose instead to use the old
names
as notations for new ones.

But then the old names will be used for printing instead of the new ones.

—
Reply to this email directly or view it on GitHub
#166 (comment).

—
Reply to this email directly or view it on GitHub
https://urldefense.proofpoint.com/v2/url?u=https-3A__github.com_UniMath_UniMath_issues_166-23issuecomment-2D143841772&d=AwMFaQ&c=8hUWFZcy2Z-Za5rBPlktOQ&r=PAZ_jTPLrSb1btCfAu4RkycWe57N0sGyzbjgeOn2wN0&m=vmKeJwYw_VUoahO4_jIjnHzl6M6PERPFTwpiJc-sjmI&s=W_DnbAFXGmy8zCyTGMfBChXhgpRXD3TXuymmdFtEMH4&e=
.

from unimath.

Changing something like "setwithbinop" implies changing names containing that string, too:

setwithbinop: Type
setwithbinoppair: ∀ X : hSet, binop X -> setwithbinop
pr1setwithbinop: setwithbinop -> hSet
issubsetwithbinop: ∀ X : hSet, binop X -> hsubtypes X -> Type
isapropissubsetwithbinop:
  ∀ (X : hSet) (opp : binop X) (A : hsubtypes X),
  isaprop (issubsetwithbinop opp A)
totalsubsetwithbinop: ∀ X : setwithbinop, subsetswithbinop
carrierofasubsetwithbinop:
  ∀ X : setwithbinop, subsetswithbinop -> setwithbinop
setwithbinopquot: ∀ X : setwithbinop, binopeqrel -> setwithbinop
setwithbinopdirprod: setwithbinop -> setwithbinop -> setwithbinop
setwithbinop1: setwith2binop -> setwithbinop
setwithbinop2: setwith2binop -> setwithbinop

from unimath.

I'd also like to rename all identifiers with "coconus" in them. Here the first use:

Definition coconusfromt (T : UU) (t : T) := Σ t':T, t = t'.

The name here has an additional problem, the use of "t", which conveys no information, and is
there only because it is the name of a parameter. I suggest renaming coconusfromt to path_from.

Here is the complete list:

iscontrcoconustot_comp:
  ∀ (X : Type) (x : X), the (iscontrcoconustot X x) = x,, idpath x
coconusfromt: ∀ T : UU, T -> Type
coconusfromtpair: ∀ (T : UU) (t t' : T), t = t' -> coconusfromt T t
coconusfromtpr1: ∀ (T : UU) (t : T), coconusfromt T t -> T
coconustot: ∀ T : UU, T -> Type
coconustotpair: ∀ (T : UU) (t t' : T), t' = t -> coconustot T t
coconustotpr1: ∀ (T : UU) (t : T), coconustot T t -> T
connectedcoconustot: ∀ (T : UU) (t : T) (c1 c2 : coconustot T t), c1 = c2
iscontrcoconustot: ∀ (T : UU) (t : T), iscontr (coconustot T t)
connectedcoconusfromt: ∀ (T : UU) (t : T) (c1 c2 : coconusfromt T t), c1 = c2
iscontrcoconusfromt: ∀ (T : UU) (t : T), iscontr (coconusfromt T t)
coconusf: ∀ X Y : UU, (X -> Y) -> Type
fromcoconusf: ∀ (X Y : UU) (f : X -> Y), coconusf f -> X
tococonusf: ∀ (X Y : UU) (f : X -> Y), X -> coconusf f
homottofromcoconusf:
  ∀ (X Y : UU) (f : X -> Y) (yxe : coconusf f),
  tococonusf f (fromcoconusf f yxe) = yxe
homotfromtococonusf:
  ∀ (X Y : UU) (f : X -> Y) (x : X), fromcoconusf f (tococonusf f x) = x
isweqtococonusf: ∀ (X Y : UU) (f : X -> Y), isweq (tococonusf f)
weqtococonusf: ∀ (X Y : UU) (f : X -> Y), X ≃ coconusf f
isweqfromcoconusf: ∀ (X Y : UU) (f : X -> Y), isweq (fromcoconusf f)
weqfromcoconusf: ∀ (X Y : UU) (f : X -> Y), coconusf f ≃ X

from unimath.

I would request for the time being not to change old names. There is a certain order in their choice that is easy to lear and use after some initial period and changing it all will be a lot of work with very little usefulness.

If someone wants another name for a particular object or objects then introducing ones own notations for this object or objects is a perfectly good way of doing it and it won’t affect anybody else.

Vladimir.

On Sep 28, 2015, at 3:18 PM, Daniel R. Grayson [email protected] wrote:

Changing something like "setwithbinop" implies changing names containing that string, too:

setwithbinop: Type
setwithbinoppair: ∀ X : hSet, binop X -> setwithbinop
pr1setwithbinop: setwithbinop -> hSet
issubsetwithbinop: ∀ X : hSet, binop X -> hsubtypes X -> Type
isapropissubsetwithbinop:
∀ (X : hSet) (opp : binop X) (A : hsubtypes X),
isaprop (issubsetwithbinop opp A)
totalsubsetwithbinop: ∀ X : setwithbinop, subsetswithbinop
carrierofasubsetwithbinop:
∀ X : setwithbinop, subsetswithbinop -> setwithbinop
setwithbinopquot: ∀ X : setwithbinop, binopeqrel -> setwithbinop
setwithbinopdirprod: setwithbinop -> setwithbinop -> setwithbinop
setwithbinop1: setwith2binop -> setwithbinop
setwithbinop2: setwith2binop -> setwithbinop
—
Reply to this email directly or view it on GitHub #166 (comment).

from unimath.

Okay; they're your files, after all.

May we add readable notations, such as this one, to files of yours?:

Definition iscontrpr1 {T : UU} : iscontr T -> T := pr1.
Notation thePoint := iscontrpr1.

The point of this particular example is that one way to read type theory is to read identity as a path. The other way is to read identity as equality. The latter way makes our proofs readable
a different way.

If not, do you permit us to include a file of basic notations in Basics/.?

from unimath.