Peter Krautzberger, maintainer of MathJax, apparently thinks that MathML has failed as a web standard (even though it succeeded as an XML standard), and should be removed from HTML5. Here's the link:
https://www.peterkrautzberger.org/0186/
It's quite a rant. Here's a quick TL;DR:
It doesn’t matter whether or not MathML is a good XML language. Personally, I think it’s quite alright. It’s also clearly a success in the XML publishing world, serving an important role in standards such as JATS and BITS.
The problem is: MathML has failed on the web.
Not a single browser vendor has stated an intent to work on the code, not a single browser developer has been seen on the MathWG. After 18 years, not a single browser vendor is willing to dedicate even a small percentage of a developer to MathML.
Math layout can and should be done in CSS and SVG. Let’s improve them incrementally to make it simpler.
It’s possible to generate HTML+CSS or SVG that renders any MathML content – on the server, mind you, no client-side JS required (but of course possible).
Since layout is practically solved (or at least achievable), we really need to solve the semantics. Presentation MathML is not sufficient, Content MathML is just not relevant.
We need to look where the web handles semantics today – that’s ARIA and HTML but also microdata, rdfa etc.
I think both, the rendering as well as the semantics, are well worth thinking about. Perhaps Wikimedia should reach out to Peter Krautzberger, and discuss some ideas of how math (and physics, and chemistry) content should be handled by Wikipedia, Wikidata, and friends. This seems like a cross roads, and we should have a hand in where things are going from here.
-- daniel (not a MathML expert all all)
Hi Daniel,
Ok. Let's discuss!
Most interesting for me is Peters statements
Content MathML is just not relevant.
Since I'm currently investigating how content MathML is used, and whats needs to be done to get mathematical content from one system to another. Currently, I have the impression that the latter question is completely solved in theory, but completely unsolved in practise.
Disclaimer: While I really appreciate drinking beer with Peter, we have slightly different approaches for de facto standards to share mathematics in information systems.
Best Moritz
On Thu, Apr 7, 2016 at 1:24 PM, Daniel Kinzler daniel@brightbyte.de wrote:
Peter Krautzberger, maintainer of MathJax, apparently thinks that MathML has failed as a web standard (even though it succeeded as an XML standard), and should be removed from HTML5. Here's the link:
https://www.peterkrautzberger.org/0186/
It's quite a rant. Here's a quick TL;DR:
It doesn’t matter whether or not MathML is a good XML language. Personally, I think it’s quite alright. It’s also clearly a success in the XML publishing world, serving an important role in standards such as JATS and BITS.
The problem is: MathML has failed on the web.
Not a single browser vendor has stated an intent to work on the code, not a single browser developer has been seen on the MathWG. After 18 years, not a single browser vendor is willing to dedicate even a small percentage of a developer to MathML.
Math layout can and should be done in CSS and SVG. Let’s improve them incrementally to make it simpler.
It’s possible to generate HTML+CSS or SVG that renders any MathML content – on the server, mind you, no client-side JS required (but of course possible).
Since layout is practically solved (or at least achievable), we really need to solve the semantics. Presentation MathML is not sufficient, Content MathML is just not relevant.
We need to look where the web handles semantics today – that’s ARIA and HTML but also microdata, rdfa etc.
I think both, the rendering as well as the semantics, are well worth thinking about. Perhaps Wikimedia should reach out to Peter Krautzberger, and discuss some ideas of how math (and physics, and chemistry) content should be handled by Wikipedia, Wikidata, and friends. This seems like a cross roads, and we should have a hand in where things are going from here.
-- daniel (not a MathML expert all all)
Am 07.04.2016 um 20:00 schrieb Moritz Schubotz:
Hi Daniel,
Ok. Let's discuss!
Great! But let's keep the discussion in one place. I made a mess by cross-posting this to two lists, now it's three, it seems. Can we agree on wikitech-l@lists.wikimedia.org as the venue of discussion? At least for the discussion of MathML in the context of Wikimedia, that would be the best place, I think.
-- daniel
I have no problem with that but are some of these lists members-only? I was told when I replied that my message would be reviewed by the moderator as I wasn't a member. Perhaps that was the W3C list.
Paul
-----Original Message----- From: Daniel Kinzler [mailto:daniel@brightbyte.de] Sent: Thursday, April 07, 2016 11:06 AM To: Moritz Schubotz schubotz@tu-berlin.de; www-math@w3.org; Peter Krautzberger peter.krautzberger@mathjax.org Cc: Wikimedia developers wikitech-l@lists.wikimedia.org; wikidata-tech wikidata-tech@lists.wikimedia.org Subject: Re: MathML is dead, long live MathML
Am 07.04.2016 um 20:00 schrieb Moritz Schubotz:
Hi Daniel,
Ok. Let's discuss!
Great! But let's keep the discussion in one place. I made a mess by cross-posting this to two lists, now it's three, it seems. Can we agree on wikitech-l@lists.wikimedia.org as the venue of discussion? At least for the discussion of MathML in the context of Wikimedia, that would be the best place, I think.
-- daniel
Peter just posted a follow up response, largely commenting on my response: https://www.peterkrautzberger.org/0187/.
First, I suspect the reason his post doesn't get as much discussion as he'd like is because his blog doesn't accept comments. I can understand why he doesn't enable comments on his personal blog but why not post it somewhere that DOES accept comments?
He says that most of the discussion has been private. That is not the way to change a standard or replace it by a new one. By all means have your private conversations but don't expect others to agree with any conclusions reached in them. The result of good ideas expressed in private conversation should be to introduce them into public conversation. Instead, his post treated MathML's failure as a fait accompli. Perhaps it is but only in the narrow scope of it being ignored by browser makers.
He feels that many things I said in my reply were more about expressing my own ideas. I'll cop to that. I felt that was needed to indicate that there are other points of view and other ideas. His solutions may not be the right ones. Let's open up the discussion.
Can we identify specific topics worthy of addressing and discuss them? I tried to hint at some possible directions in my reply, which is why it veered into some of my own ideas. I would love for this to be a constructive discussion. Instead of discussing whether MathML is a failed standard, I would like to see real, open discussion on solutions to various problems. Any takers?
Paul
-----Original Message----- From: Paul Topping [mailto:pault@dessci.com] Sent: Thursday, April 07, 2016 2:02 PM To: Daniel Kinzler daniel@brightbyte.de; Moritz Schubotz <schubotz@tu- berlin.de>; www-math@w3.org; Peter Krautzberger peter.krautzberger@mathjax.org Cc: Wikimedia developers wikitech-l@lists.wikimedia.org; wikidata-tech wikidata-tech@lists.wikimedia.org Subject: RE: MathML is dead, long live MathML
I have no problem with that but are some of these lists members-only? I was told when I replied that my message would be reviewed by the moderator as I wasn't a member. Perhaps that was the W3C list.
Paul
-----Original Message----- From: Daniel Kinzler [mailto:daniel@brightbyte.de] Sent: Thursday, April 07, 2016 11:06 AM To: Moritz Schubotz schubotz@tu-berlin.de; www-math@w3.org; Peter Krautzberger peter.krautzberger@mathjax.org Cc: Wikimedia developers wikitech-l@lists.wikimedia.org; wikidata-tech wikidata-tech@lists.wikimedia.org Subject: Re: MathML is dead, long live MathML
Am 07.04.2016 um 20:00 schrieb Moritz Schubotz:
Hi Daniel,
Ok. Let's discuss!
Great! But let's keep the discussion in one place. I made a mess by cross-posting this to two lists, now it's three, it seems. Can we agree on wikitech-l@lists.wikimedia.org as the venue of discussion? At least for
the
discussion of MathML in the context of Wikimedia, that would be the best place, I think.
-- daniel
Hi, One concrete usecase is the formula datatype for properties on Wikidata. We are discussing the semantics issues here : what means the operators of the formula, what means the variables ? An immediate way, in Wikidata, is to In the item for a geometric figure, for example a square, how to model that a square can be defined in the euclidian space by the geometric coordinates of points, we could create the item for a point class in Wikidata, give a name of a point (pretty much usual mathematical or programming work) and to link that variable name to an item for the semantics/corresponding type. Same for the operators.
Last, in the question you raised on "modelling maths versus modeling domain model formula" I'd say that in Wikidata the scope is basically unlimited, contrary to a regular scientific publication who takes place in a context that may be more or less non formally explicited, we can fill the gap beetween more formal aspects of logical or inference rules used by the scientist, the mathematical framework (euclidian space, non euclidian space, logical framework, axioms ... we pretty much have items for all of this and can create new one if that's structurally needed for a usecase) and the formula. Time is less of an issue because the work is reusable and cumulative, there is no deadline. There is only a need to leave the door open to do that work for someone to be able to do it at his/her convenance.
Of course it's a lot of work, but there is no pressure. I'm not sure how MathML relates to this however.
2016-04-08 0:51 GMT+02:00 Paul Topping pault@dessci.com:
Peter just posted a follow up response, largely commenting on my response: https://www.peterkrautzberger.org/0187/.
First, I suspect the reason his post doesn't get as much discussion as he'd like is because his blog doesn't accept comments. I can understand why he doesn't enable comments on his personal blog but why not post it somewhere that DOES accept comments?
He says that most of the discussion has been private. That is not the way to change a standard or replace it by a new one. By all means have your private conversations but don't expect others to agree with any conclusions reached in them. The result of good ideas expressed in private conversation should be to introduce them into public conversation. Instead, his post treated MathML's failure as a fait accompli. Perhaps it is but only in the narrow scope of it being ignored by browser makers.
He feels that many things I said in my reply were more about expressing my own ideas. I'll cop to that. I felt that was needed to indicate that there are other points of view and other ideas. His solutions may not be the right ones. Let's open up the discussion.
Can we identify specific topics worthy of addressing and discuss them? I tried to hint at some possible directions in my reply, which is why it veered into some of my own ideas. I would love for this to be a constructive discussion. Instead of discussing whether MathML is a failed standard, I would like to see real, open discussion on solutions to various problems. Any takers?
Paul
-----Original Message----- From: Paul Topping [mailto:pault@dessci.com] Sent: Thursday, April 07, 2016 2:02 PM To: Daniel Kinzler daniel@brightbyte.de; Moritz Schubotz <schubotz@tu- berlin.de>; www-math@w3.org; Peter Krautzberger peter.krautzberger@mathjax.org Cc: Wikimedia developers wikitech-l@lists.wikimedia.org; wikidata-tech wikidata-tech@lists.wikimedia.org Subject: RE: MathML is dead, long live MathML
I have no problem with that but are some of these lists members-only? I
was
told when I replied that my message would be reviewed by the moderator as I wasn't a member. Perhaps that was the W3C list.
Paul
-----Original Message----- From: Daniel Kinzler [mailto:daniel@brightbyte.de] Sent: Thursday, April 07, 2016 11:06 AM To: Moritz Schubotz schubotz@tu-berlin.de; www-math@w3.org; Peter Krautzberger peter.krautzberger@mathjax.org Cc: Wikimedia developers wikitech-l@lists.wikimedia.org;
wikidata-tech
wikidata-tech@lists.wikimedia.org Subject: Re: MathML is dead, long live MathML
Am 07.04.2016 um 20:00 schrieb Moritz Schubotz:
Hi Daniel,
Ok. Let's discuss!
Great! But let's keep the discussion in one place. I made a mess by cross-posting this to two lists, now it's three, it seems. Can we
agree on
wikitech-l@lists.wikimedia.org as the venue of discussion? At least
for
the
discussion of MathML in the context of Wikimedia, that would be the
best
place, I think.
-- daniel
Wikidata-tech mailing list Wikidata-tech@lists.wikimedia.org https://lists.wikimedia.org/mailman/listinfo/wikidata-tech
Am 07.04.2016 um 23:01 schrieb Paul Topping:
I have no problem with that but are some of these lists members-only? I was told when I replied that my message would be reviewed by the moderator as I wasn't a member. Perhaps that was the W3C list.
Oh... both the Wikimedia lists are members only, I'm afraid. The W3C list requires a 1-click agreement to their terms. That's easier, but less likely to involve Wikimedia people.
Hello, how many of us have github accounts?
On Friday, April 8, 2016 6:10 AM, Daniel Kinzler daniel@brightbyte.de wrote:
Am 07.04.2016 um 23:01 schrieb Paul Topping:
I have no problem with that but are some of these lists members-only? I was told when I replied that my message would be reviewed by the moderator as I wasn't a member. Perhaps that was the W3C list.
Oh... both the Wikimedia lists are members only, I'm afraid. The W3C list requires a 1-click agreement to their terms. That's easier, but less likely to involve Wikimedia people.
Hi Daniel,
Could you let me know once you've decided on a venue for discussion? I'd be happy to join in.
Thanks in advance, Peter.
On Thu, Apr 7, 2016 at 8:05 PM, Daniel Kinzler daniel@brightbyte.de wrote:
Am 07.04.2016 um 20:00 schrieb Moritz Schubotz:
Hi Daniel,
Ok. Let's discuss!
Great! But let's keep the discussion in one place. I made a mess by cross-posting this to two lists, now it's three, it seems. Can we agree on wikitech-l@lists.wikimedia.org as the venue of discussion? At least for the discussion of MathML in the context of Wikimedia, that would be the best place, I think.
-- daniel
Yes, let's discuss. I wrote a response to Peter's post here: http://bit.ly/1ZLfCF8. Probably some on this list are aware of it but perhaps not all.
I believe it is important to separate out the different aspects of MathML as a standard. Peter scoped his post by saying that it had failed as a WEB standard right in the title. However, many will ignore that key word and see it as saying that MathML is a total failure which is clearly not the case. On Hacker News, they even discussed how TeX was a better language than MathML to type math. Peter's title is a provocative one and subject to much misinterpretation. This is one of the main reasons I wrote my response. I actually agree with a lot of Peter's observations but not their tone and the fact that they are presented in a manner that invites misinterpretation by the many that do not really understand MathML.
Paul Topping
Design Science, Inc. "How Science Communicates" Makers of MathType, MathFlow, MathPlayer, Equation Editor http://www.dessci.com
-----Original Message----- From: Moritz Schubotz [mailto:schubotz@tu-berlin.de] Sent: Thursday, April 07, 2016 11:01 AM To: Daniel Kinzler daniel@brightbyte.de; www-math@w3.org; Peter Krautzberger peter.krautzberger@mathjax.org Cc: Wikimedia developers wikitech-l@lists.wikimedia.org; Schubotz, Moritz schubotz@tu-berlin.de; wikidata-tech <wikidata- tech@lists.wikimedia.org> Subject: Re: MathML is dead, long live MathML
Hi Daniel,
Ok. Let's discuss!
Most interesting for me is Peters statements
Content MathML is just not relevant.
Since I'm currently investigating how content MathML is used, and whats needs to be done to get mathematical content from one system to another. Currently, I have the impression that the latter question is completely solved in theory, but completely unsolved in practise.
Disclaimer: While I really appreciate drinking beer with Peter, we have slightly different approaches for de facto standards to share mathematics in information systems.
Best Moritz
On Thu, Apr 7, 2016 at 1:24 PM, Daniel Kinzler daniel@brightbyte.de wrote:
Peter Krautzberger, maintainer of MathJax, apparently thinks that MathML
has
failed as a web standard (even though it succeeded as an XML standard),
and
should be removed from HTML5. Here's the link:
https://www.peterkrautzberger.org/0186/
It's quite a rant. Here's a quick TL;DR:
It doesn’t matter whether or not MathML is a good XML language.
Personally, I
think it’s quite alright. It’s also clearly a success in the XML publishing world, serving an important role in standards such as JATS and BITS.
The problem is: MathML has failed on the web.
Not a single browser vendor has stated an intent to work on the code, not
a
single browser developer has been seen on the MathWG. After 18 years,
not a
single browser vendor is willing to dedicate even a small percentage of a developer to MathML.
Math layout can and should be done in CSS and SVG. Let’s improve them incrementally to make it simpler.
It’s possible to generate HTML+CSS or SVG that renders any MathML
content –
on the server, mind you, no client-side JS required (but of course
possible).
Since layout is practically solved (or at least achievable), we really need to solve the semantics. Presentation MathML is not sufficient, Content
MathML
is just not relevant.
We need to look where the web handles semantics today – that’s ARIA
and HTML
but also microdata, rdfa etc.
I think both, the rendering as well as the semantics, are well worth thinking about. Perhaps Wikimedia should reach out to Peter Krautzberger, and
discuss
some ideas of how math (and physics, and chemistry) content should be
handled by
Wikipedia, Wikidata, and friends. This seems like a cross roads, and we
should
have a hand in where things are going from here.
-- daniel (not a MathML expert all all)
-- Moritz Schubotz TU Berlin, Fakultät IV DIMA - Sekr. EN7 Raum EN742 Einsteinufer 17 D-10587 Berlin Germany
Tel.: +49 30 314 22784 Mobil.: +49 1578 047 1397 Fax: +49 30 314 21601 E-Mail: schubotz@tu-berlin.de Skype: Schubi87 ICQ: 200302764 Msn: Moritz@Schubotz.de
wikidata-tech@lists.wikimedia.org