Accessibility Emoji

Marcel Schneider via Unicode unicode at
Thu Mar 29 19:31:48 CDT 2018

On 29/03/18 17:03 William_J_G Overington via Unicode wrote:
> I have been thinking about issues around the proposal.
> There is a sentence in that document that starts as follows.
> > Emoji are a universal language and a powerful tool for communication, ....
That is clearly overstating the capabilities of emoji, and ignoring the borderline 
between verbal and pictographic expression. The appropriateness of each one 
depends mainly on semantics and context. The power of emoji may rely in their 
being polysemic, escaping censorship as already discussed during past years.
> It seems to me that what is lacking with emoji are verbs and pronouns.
Along with these, one would need more nouns, too, setting up an autonomous 
language. That however is not the goal of emoji and is outside the scope of 
> For example, "to be", "to have" and "to need". The verb "to need" might well 
be of particular importance in relation to accessibility considerations.
When accessibility matters, devices may be missing, and then the symbol charts 
are most appropriate, as seen. When somebody is pointing an object, the ‷need” 
case is most obvious anyway. Impaired persons may use a bundle of cards including 
textual messages. None of these justifies encoding extra emoji. E.g. when somebody 
wishes a relative to buy more bread while returning from work, the appropriate number
of loaves followed by an exclamation mark and a smile or heart may do it.
> How could verbs be introduced into emoji? The verb "to love" can already be indicated using a heart symbol.
This is the one that people are likely to be most embarrassed typing out. 
> Should abstract designs be used? Or should emoji always be pictographic?
Yes, they should always be highly iconic, Asmus explained in detail. See:
> If abstract designs were introduced would it be possible for the standards documents to include the meanings
> or would the standards documents need to simply use a geometrical description and then the meanings be
> regarded as a higher level protocol outside of the standard?
On one hand, Unicode does not encode semantics; but on the other hand, on character level, semantics are 
part of the documentation accompanying a number of characters in the Charts. There is a balance between 
polysemics and disambiguation. As a thumb rule: characters are disambiguated to ensure correct processing
of the data, so far as the cost induced by handling multiple characters doesn’t outweigh the benefit. 
In putting your question, you already answered it, except that there are geometric figures encoded for UIs, 
that therefore already have a meaning, yet are mostly generically named, leaving the door open to alternate 
> For, if abstract emoji were introduced with the intention of them to be of use as verbs in a universal language,
> it would be of benefit if the meanings were in the standard.
But such a language has clearly been stated as being out of scope of Unicode, and we aren’t even allowed 
to further discuss that particular topic, given the mass of threads and e‐mails already dedicated to it in the past.
> If abstract designs were used then the meanings would need to be learned. Yet if the meanings were
> universal that could be a useful development.
It would not, because automatic translation tools already cater for these needs, and possibly better. See:
> I have wondered whether verb tenses could be usefully expressed using some of the existing combining
> accent characters following an emoji verb character..
First of all, users should be likely to adopt the scheme in a fairly predictable way. I’m ignoring actual trends 
and can only repeat what has been said on this list: communities are missing, and so is interest. 
Hence, sadly to say, there is little through no point in elaborating further.
Personally I’m poorly armed to help building a user community, as I don’t have a smartphone, while being 
very busy with more and more tasks, leaving little time for many experiments.  Sorry.
Best regards,
> For example, U+0302 COMBINING CIRCUMFLEX ACCENT to indicate that the verb is in the future tense, U+0304 COMBINING MACRON to indicate that the verb is in the present tense, U+030C COMBINING CARON to indicate that the verb is in the past tense, U+0303 COMBINING TILDE to indicate that the verb is in the conditional tense.
> The desirability of pronouns was raised by a gentleman in the audience of a lecture at the Internationalization and Unicode Conference in 2015.
> I tried to produce some designs. I could not find a way to do that with conventional illustrative pictures, though I did produce a set of abstract designs that could possibly be useful in application; they could be displayed in colourful emoji style yet also in monochrome without ambiguity. Yet they are abstract designs, so meanings would need to be learned rather than indicated by the picture itself. Yet if the meanings were universal, that could be useful. Should there be abstract emoji or should emoji only be conventional pictures?
> William Overington
> Thursday 29 March 2018
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <>

More information about the Unicode mailing list