[cellml-discussion] Visual editing of mathematics

Mon Aug 7 21:23:10 NZST 2006

Dear all,

Having already voiced my opinion (mainly as an end user at this stage), the
view of someone like Penny Noble (to whom I am copying this, in case she's
not on the CellML mailing list) might be interesting to have. She has been
developing cell models for years and been playing with CellML for a few
years too. She has used different programming languages, approaches, etc. to
implementing models, including options #1 and #2 for CellML models (and
maybe even #3, though I cannot remember for sure). My point is that
developing a cell model is a time consuming process (we are talking about
months, if not years in the worst of cases), so we cannot have something
that is nice looking, but not convenient for a day-to-day use.

	Best regards, Alan.

PS: my main concern is not so much about getting it right from our
perspective, but that of the end-user and, in that respect, I don't think
that #2 does the job. That is not to say that option #1 is better, just that
I have yet to find something that is more convenient from an end-user
perspective.

> -----Original Message-----
> From: cellml-discussion-bounces at cellml.org 
> [mailto:cellml-discussion-bounces at cellml.org] On Behalf Of 
> Poul Nielsen
> Sent: 07 August 2006 10:05
> To: For those interested in contributing to the development of CellML.
> Subject: Re: [cellml-discussion] Visual editing of mathematics
> 
> Dear all
> 
> I think that option 2 should be the highest priority. One of 
> the reasons that we started the Content MathML editor was 
> that there was no easy intuitive way of creating Content 
> MathML for non computer savy people. The visual paradigm 
> works well for most users, even if it may not be the quickest 
> way of constructing mathematical expressions. Five years down 
> the line I don't know of any open efforts to create a Content 
> MathML editor.
> 
> I am very wary of taking route 1 as an option. It is 
> extremely difficult to create a simple, consistent language 
> that is rich and intuitive enough for a wide range of users. 
> Such a task will require a lot of work to do properly. We 
> considered this option for describing mathematics when 
> designing CellML but concluded that it was not sensible. SBML 
> initially took this route but very quickly 'saw the light' 
> and adopted Content MathML. I still don't see why we should 
> invest effort in designing yet another layer of describing 
> mathematics between the well-understood visual representation 
> and the reasonably well-designed Content MathML. I say a 
> definite NO to option 1.
> 
> In my opinion option 3 is just a mixture of 1 and 2 with all 
> the problems of 1. I say a definite NO to option 3.
> 
> Best wishes
> Poul
> 
> On 2006 Aug 07, at 19:57, Alan Garny wrote:
> 
> > Hi,
> >
> > I would (suprisingly enough) also favour solution #1, 
> together with a 
> > graphical rendering of the equation.
> >
> > Regarding the other options:
> >
> > #2: having used Microsoft Word to write my thesis (I know, 
> I know...), 
> > I used Microsoft Equation to enter equations and it's fair 
> to say that 
> > it looks quite nice, but it's not very convenient in the long term:
> > you have to
> > click here and there a lot, making this solution much 
> slower than #1.
> >
> > #3: I am not sure I want to comment on this option... :)
> >
> > One possible drawback of solution #1 is that the user has 
> to learn yet 
> > another syntax. Having said that, the learning curve is not 
> steep at 
> > all: we are talking about mathematical equations here, so...
> >
> > One thing I am concerned with, though, is how do you intend to edit 
> > several equations? Will it be something like: click on an equation, 
> > edit it, click on another equation, edit it, etc.? If so, 
> that will, 
> > once again, involve quite a few clicks, which in the long run may 
> > prove very tedious for someone who is coding in a model (I am 
> > unfortunately talking from experience here, for having had the 
> > displeasure of using a software that required selecting an 
> equation in 
> > a list, and then editing it using something like option #2).
> >
> > My point, therefore, is that we should not only think about 
> the best 
> > way to edit an equation, but a bunch of equations...
> >
> > Also, what about mapping of variables between components, 
> units, etc.?
> >
> > 	Cheers, Alan.
> >
> >> -----Original Message-----
> >> From: cellml-discussion-bounces at cellml.org
> >> [mailto:cellml-discussion-bounces at cellml.org] On Behalf Of 
> Nickerson, 
> >> David Phillip
> >> Sent: 07 August 2006 07:50
> >> To: For those interested in contributing to the development of 
> >> CellML.
> >> Subject: Re: [cellml-discussion] Visual editing of mathematics
> >>
> >> Hi Andrew,
> >>
> >> Personally I would use something like your suggestion 1, 
> although it 
> >> is probably best to keep away from TeX syntax since that 
> may not be a 
> >> lot better than editing the MathML directly for some people. You 
> >> probably also want to allow numbers to be entered without 
> units and 
> >> then have the GUI prompt users to choose the units from a 
> drop down 
> >> list of appropriate/available units (maybe during some validation 
> >> stage?). You just need to specify how equations should be written, 
> >> and I'd probably go with something similar to MATLAB/C style text. 
> >> Maybe look at the virtual cell specification for guidence
> >> (http://vcell.org/technology/VCML_Grammar.html) or the 
> equivalent for 
> >> JSim
> >> (http://nsr.bioeng.washington.edu/PLN/Members/butterw/JSIMDOC1
> >> .6/JSim_Home.stx/MML_Ref.stx).
> >>
> >> The second option might be nice for some users, but I 
> don't think it 
> >> should be a high priority for us to develop such an interface. 
> >> Hopefully other groups will start providing nice interfaces for 
> >> creating content mathml that could then be pulled into a 
> CellML model 
> >> quite easily.
> >>
> >> A compromise would be something like the COR interface 
> where you edit 
> >> a text expression like in (1) and are simultaneously given a 
> >> graphical display of the equation using some cool methods 
> to make it 
> >> look pretty.
> >>
> >>
> >> Andre.
> >>
> >> -----Original Message-----
> >> From: cellml-discussion-bounces at cellml.org on behalf of 
> Andrew Miller
> >> Sent: Mon 07-Aug-06 2:11 PM
> >> To: cellml-discussion at cellml.org
> >> Subject: [cellml-discussion] Visual editing of mathematics
> >>
> >> Hi all,
> >>
> >> I am currently planning how PCEnv (the Physiome CellML 
> Environment, 
> >> which is planned to incorporate the functionality from 
> mozCellML and 
> >> the Physiome CellML
> >> Environment) will allow mathematics to be viewed and edited.
> >>
> >> Opinions from potential users are solicited. If you feel that you 
> >> would use PCEnv when it is complete, your opinions on how 
> this should 
> >> be done would be greatly appreciated.
> >>
> >> I would be particularly interested on your opinions on the 
> following 
> >> options, and on how useful each one would be to you:
> >>
> >> 1) An editor which used a TeX like language (or some other 
> plain text 
> >> only language), which can be mapped to and from content 
> MathML. For 
> >> example, to produce the MathML <DEFANGED-APPly>
> >>   <eq/>
> >>   <DEFANGED-APPly><times/>
> >>     <cn cellml:units="dimensionless">2</cn>
> >>     <ci>x</ci>
> >>   </apply>
> >>   <ci>y</ci>
> >> </apply>
> >>
> >> you might type something like 2[dimensionless] * x = y
> >>
> >> 2)
> >> An editor which works like the Math editor in WYSIWYG 
> word-processors 
> >> like Microsoft Word and OpenOffice (in other words, you select 
> >> symbols from the menus, and then fill in the boxes, to create an 
> >> on-screen rendering of the mathematics you desire). This would 
> >> require some sort of special method to input the units on 
> constants.
> >>
> >> 3)
> >> An editor based around a tree control (i.e. you have a 
> tree structure 
> >> which mirrors the XML structure). For example, you might 
> have a tree 
> >> like...
> >>
> >> +- Apply operation
> >>  |------ Operator                       Equals
> >>  |-----+ Apply Operation
> >>  |      |------ Operator                Times
> >>  |      |----+ Constant 2
> >>  |      |     |----- Units                 dimensionless
> >>  |      |------- Variable                x
> >>  |------- Variable                       y
> >>
> >> Any other suggestions anyone has would be appreciated.
> >>
> >> Best regards,
> >> Andrew Miller
> >>
> >> _______________________________________________
> >> cellml-discussion mailing list
> >> cellml-discussion at cellml.org
> >> http://www.cellml.org/mailman/listinfo/cellml-discussion
> >>
> >>
> >>
> >
> > _______________________________________________
> > cellml-discussion mailing list
> > cellml-discussion at cellml.org
> > http://www.cellml.org/mailman/listinfo/cellml-discussion
> 
> _______________________________________________
> cellml-discussion mailing list
> cellml-discussion at cellml.org
> http://www.cellml.org/mailman/listinfo/cellml-discussion