I can make this switch immediately. For the purposes of an audit trail can we have more info - link to a discussion, publication...?

Is the convention universally adhered to in comparative anatomy? Are there contexts in which arabic numerals are preferentially used (e.g embryology)?

Is it the case that arabic numerals carry more implication of homology? This is fine as we use digit numbers to carry this implication (in addition to explicit designations of homology of course), just good to document.

Are there additional changes expected? I assume the convention is: roman for digits preaxial-axial, arabic for phalanges, proximal-distal?

In developmental biology literature, at least in my experience, whenever digits are specified numerically, the convention is to use Arabic numbers.

Furthermore, from a purely practical standpoint, for considerations such as ordering, searching and sorting, in general I have found Arabic numbering (e.g. 1, 2, 3; 1st, 2nd, 3rd) to be preferable to either Roman (e.g. I, II, III) or any type of word (e.g. one, two, three; first, second, third) forms.

I agree, as far as I know there is no convention but typically we refer to the human digits using Arabic numerals both in the embryo and in the adult.

Dr. John Cork,
Associate Professor
Computer Imaging Laboratory, MEB Rm 6226
Department of Cell Biology & Anatomy,
LSU Health Sciences Center,
1901 Perdido St., Box P6-2
New Orleans, LA 70112

Tel. 504 568 7177 Fax. 504 568 4392

URL: http://virtualhumanembryo.lsuhsc.eduhttp://virtualhumanembryo.lsuhsc.edu/

There are only 10 types of people, those who understand binary and those who don't.

From: tfhayamizu [mailto:notifications@github.com]
Sent: Friday, May 23, 2014 11:13 AM
To: obophenotype/uberon
Subject: Re: [uberon] Digit III or Digit 3 (#458)

In developmental biology literature, at least in my experience, whenever digits are specified numerically, the convention is to use Arabic numbers.

Furthermore, from a purely practical standpoint, for considerations such as ordering, searching and sorting, in general I have found Arabic numbering (e.g. 1, 2, 3; 1st, 2nd, 3rd) to be preferable to either Roman (e.g. I, II, III) or any type of word (e.g. one, two, three; first, second, third) forms.

—
Reply to this email directly or view it on GitHubhttps://github.com//issues/458#issuecomment-44030783.

It sounds like the thing to do here is auto-add community synonyms. I can put these in the COMPARATIVE_PREFERRED subset

As far as I know using roman numerals to refer to the digits and arabic numerals for phalanges is fairly widespread in vertebrate anatomy, (certainly in comparative anatomy and paleontology), so for the purposes of a vertebrate anatomy ontology it would make sense to follow standard usage introduced by influential anatomists such as Romer (1956) etc.

I am also attaching several screenshots of book pages, including one where arabic numerals are used (Form and Transformation: Generative and Relational Principles in Biology), but the author points out that this is contrary to convention.

Here are some more examples -

http://www.springerimages.com/Images/LifeSciences/1-10.1007_s13127-010-0019-x-2

This also makes discussions in comparative biology easier (using combinations such as II-3 etc):

http://dml.cmnh.org/1997Oct/msg00758.html

See also this discussion I found on an anatomy discussion list:

"Also, Roman numerals are used for digit numbers. I see this convention has
been disappearing among human anatomists, but it's best to keep because
Arabic numbers are used as subscripts for phalanx number. II sub 3 = the
third phalanx of the second digit. But don't worry, the stubborn, old
ones always lose the language battle in the end--to boldly go where no
spelling or grammar rule has gone before….."

Best,

Nizar

Nizar Ibrahim, PhD
The University of Chicago
Organismal Biology & Anatomy
1027 East 57th Street, Chicago

Email: nibrahim@uchicago.edu
Office phone: 773-702-6025

From: JohnCork <notifications@github.commailto:notifications@github.com>
Reply-To: obophenotype/uberon <reply@reply.github.commailto:reply@reply.github.com>
Date: Friday, May 23, 2014 11:17 AM
To: obophenotype/uberon <uberon@noreply.github.commailto:uberon@noreply.github.com>
Cc: Nizar Ibrahim <nibrahim@uchicago.edumailto:nibrahim@uchicago.edu>
Subject: Re: [uberon] Digit III or Digit 3 (#458)

I agree, as far as I know there is no convention but typically we refer to the human digits using Arabic numerals both in the embryo and in the adult.

Dr. John Cork,
Associate Professor
Computer Imaging Laboratory, MEB Rm 6226
Department of Cell Biology & Anatomy,
LSU Health Sciences Center,
1901 Perdido St., Box P6-2
New Orleans, LA 70112

Tel. 504 568 7177 Fax. 504 568 4392

URL: http://virtualhumanembryo.lsuhsc.eduhttp://virtualhumanembryo.lsuhsc.edu/

There are only 10 types of people, those who understand binary and those who don't.

From: tfhayamizu [mailto:notifications@github.com]
Sent: Friday, May 23, 2014 11:13 AM
To: obophenotype/uberon
Subject: Re: [uberon] Digit III or Digit 3 (#458)

In developmental biology literature, at least in my experience, whenever digits are specified numerically, the convention is to use Arabic numbers.

Furthermore, from a purely practical standpoint, for considerations such as ordering, searching and sorting, in general I have found Arabic numbering (e.g. 1, 2, 3; 1st, 2nd, 3rd) to be preferable to either Roman (e.g. I, II, III) or any type of word (e.g. one, two, three; first, second, third) forms.

—
Reply to this email directly or view it on GitHubhttps://github.com//issues/458#issuecomment-44030783.

—
Reply to this email directly or view it on GitHubhttps://github.com//issues/458#issuecomment-44031245.