Re: math: URI scheme and protocol handler

On 2014/04/28 06:31, Tony Hansen wrote:
> This sounds more like a media type than a URI.
>
>      Tony Hansen

Very much so indeed. And of course there is already a media type (or 
actually three :-) for MathML:
http://www.iana.org/assignments/media-types/media-types.xhtml#application
http://www.w3.org/TR/MathML3/appendixb.html

Also, it's easy to create an URI from some data and a media type (Paul 
already gave a strong hint below):
data:application/mathml+xml,<math><mi>x</mi><mo>+</mo><mn>2</mn></math>

Regards,   Martin.

> On 4/27/14, 2:56 PM, Gerardo Capiel wrote:
>> Paul,
>>
>> Thank you for your questions. The use case is very similar to the
>> mailto URI scheme and protocol handler. We desire a mechanism to
>> launch from a Web browser an application to handle a mathematical
>> expression.
>>
>> This idea was prompted by the need for an alternative approach to
>> enable Assistive Technology applications to provide blind students
>> navigation of complex mathematical expressions via Text-To-Speech. The
>> current approach by the leading solution provider, Design Science, is
>> to use an Internet Explorer plugin. This approach no longer works as
>> of IE 11 due to changes in the IE security model and has not been
>> extended to any other browsers.
>>
>> As part of my work in the DIAGRAM Center, I have discussed this math:
>> protocol handler with Design Science and other vendors wishing to
>> provide Assistive Technology solutions and they are very interested as
>> it simplifies implementation, particularly across multiple browsers
>> and operating systems.
>>
>> We also realized that this idea could benefit all students, as it
>> provides a simple mechanism for students to explore mathematical
>> expressions on the Web in math notebooks, such as IPython, or
>> applications such as graphing calculators or MATLAB.
>>
>> You can find more details at:
>> https://wiki.benetech.org/display/MATH/Protocol+Handlers+for+External+Applications+to+Process+MathML
>>
>>
>> Let me know if this helps or you need more details.
>>
>> Cheers,
>>
>> Gerardo
>>
>> Gerardo Capiel
>> VP of Engineering
>> Benetech
>>
>> On Apr 27, 2014, at 11:49 AM, "Paul Prescod" <paul@prescod.net
>> <mailto:paul@prescod.net>> wrote:
>>
>>> Please clarify the requirements. Math expressions can already be
>>> expressed in a hypothetical markup language like this.
>>>
>>> <math-expression  href="http://...."/>
>>>
>>> Or:
>>>
>>> <math-expression href="data:..."/>
>>>
>>> The role of the URI scheme is not to convey the semantics or data
>>> type of the resource but rather to supply information on the network
>>> location of the resource.
>>>
>>> So I am skeptical that this URI scheme is needed. If we went down
>>> this path then we would need a URI scheme for images, zip files,
>>> documents, movies and so forth.
>>>
>>> On Apr 26, 2014 11:28 PM, "Gerardo Capiel" <gerardoc@benetech.org
>>> <mailto:gerardoc@benetech.org>> wrote:
>>>
>>>     Hi,
>>>
>>>     I'm proposing the creation of a W3C Community Group to discuss a
>>> math: URI scheme and Web browser protocol handler. The 'math' URI
>>> scheme and protocol handler can be used by Web browsers and other
>>> applications to transport mathematical expressions to other
>>> applications, such as Assistive Technologies or graphing
>>> calculators.  If you are interested this proposal, please
>>> join:http://www.w3.org/community/groups/proposed/#mathprotocol.
>>> Below is the provisional URI Scheme Registration Template:
>>>
>>>     *    URI scheme name
>>>     *       math - The math: URI scheme can be used by Web browsers
>>> and other applications to transport mathematical expressions to other
>>> applications, such as Assistive Technologies or graphing calculators.
>>>
>>>     *    Status
>>>     *       Provisional
>>>     x
>>>     *    URI scheme syntax
>>>     *       Applications launched via the math protocol take as a
>>> parameter either a URL to a resource containing a mathematical
>>> expression in MathML or other formats or the actual math expression
>>> in uncompressed or compressed form. For example:
>>>            math://mathmlcloud.org/m/uniqueexpressionid
>>>            math:<math><mi>x</mi><mo>+</mo><mn>2</mn></math>
>>>
>>> math:<math-compressed>someexpressionincompressedformat</math-compressed>
>>>
>>>     *    URI scheme semantics
>>>     *       TBD
>>>     *    Encoding considerations
>>>     *       TBD
>>>     *    Applications/protocols that use this URI scheme name
>>>     *       Applications should register the math: protocol handler
>>> with this scheme.
>>>     *    Interoperability considerations
>>>     *       TBD
>>>     *    Security considerations
>>>     *       None known.
>>>     *    Contact
>>>     *       Gerardo Capiel,gcapiel@alum.mit.edu
>>> <mailto:gcapiel@alum.mit.edu>
>>>     *    Author/Change controller
>>>     *       Gerardo Capiel,gcapiel@alum.mit.edu
>>> <mailto:gcapiel@alum.mit.edu>
>>>     *    References
>>>     *       [1]http://www.w3.org/community/groups/proposed/#mathprotocol
>>>
>>> [2]https://wiki.benetech.org/display/MATH/Protocol+Handlers+for+External+Applications+to+Process+MathML
>>>
>>>
>>>     Gerardo Capiel
>>>     VP of Engineering
>>>     benetech
>>>
>
>

Received on Monday, 28 April 2014 02:31:54 UTC