[BlindMath] What is your experience reading LaTeX source?

Fri Jan 11 22:39:25 UTC 2019

I happen to agree with all of Aqil's comments, but would add:

Anyone who expects to write in LaTeX must be able to read in LaTeX; that's not a blindness issue per se, but if we can't read the final converted output with certainty then our independence is compromised. This lack of certainty (post-conversion) is the blindness issue.

I well remember spending some very good money on a (print) book about writing for the mathematical sciences. I had it read to me by volunteers cover to cover. In the end, I could have scanned much of it, especially the sections that were the most critical about writing style. The most memorable of the nuggets taken from this book was that the expression of mathematical ideas by way of symbols and Greek letters has limited capacity to speak to the audience. It is more effective and efficient if the audience knows what they are looking at before they look at it. My take on the advice given was to make sure that every equation or symbols/Greek could be replaced with placeholders like "blah1" or "spiky thing" without detracting from the overall level of understanding of the aims of the paper/thesis etc. This is even more relevant in presentations when the audience is engaging at the speaker's pace not the audience's pace. In the end, this guided my use of volunteer readers as well. Scanning options were very rudimentary at that time and little to nothing was available in any form of HTML. The best papers I engaged with tended to be the ones that followed the advice given in the book. In the end, knowing I was engaging with the right papers was considerably more important than making sure I knew what was in evry formula in every paper I saw.

In more recent times, the ability to access papers in various levels of quality HTML formats via publishers' websites and my university's library means I end up doing much the same today as I did when I started my PhD twenty years ago. I still don't read every last formula but I know that the formula is in the paper to grab when the detail is truly needed.

The above level of engagement with LaTeX or HTML based equations really is only relevant to the way I worked as a grad student and as a lecturer. It bears much less relevance to the needs of anyone who is at high school or taking a small number of university level courses making use of mathematical notation or constructs.

One last point I think is worth noting is that there is a lot of difference between reading a LaTeX document and reading mathematical elements in any other format where only the mathematical stuff is presented using LaTeX. I do think the needs of high school level and many university courses can be learned because the amount of LaTeX to learn is small. If the best (simplest) LaTeX representations are used then it should be fairly easy to engage. I would suggest that many of these documents would be very easily digested in markdown for example, but if that is being used, then the converted output is even better.

Nice discussion.

Jonathan

-----Original Message-----
From: BlindMath <blindmath-bounces at nfbnet.org> On Behalf Of Aqil Sajjad via BlindMath
Sent: Saturday, 12 January 2019 9:02 AM
To: jheim at math.wisc.edu; Blind Math list for those interested in mathematics <blindmath at nfbnet.org>
Cc: Aqil Sajjad <aqilsajjad at gmail.com>
Subject: Re: [BlindMath] What is your experience reading LaTeX source?

Ok, I should have phrased that 'bottom line' sentence slightly differently. But the overall point is still fully valid. I am a research level blind physicist and read papers from arxiv all the time. And most of the time I am able to do this either with zero or minimal time wasted on reading or editing the original LaTex code. The reason: I choose to work with technology that allows me to do this.

Yes, several blind mathematicians read LaTex code directly. But that hardly means it is not a very inefficient way to read math. There is a reason why sighted people generally read the compiled pdf instead of reading the LaTex source code directly. The difference is like going to a website and reading the html source code with all the formatting junk instead of reading the actual website on display. One can of course read the source code but one wouldn't normally do that as a first choice unless one really has to. And this isn't merely a matter of personal preference but has to do with the fact that if you have to read through all those formatting commands which aren't related to the content, then it's distracting and makes reading inefficient.

Now, coming to the point about the inability of LaTex to MathML converters to cleanly produce good MathML without requiring one to edit the source code. I haven't worked with MathMl conversion because I haven't been able to get one working, honestly. I have however done some experimentation with LaTex to MathType convertors, and yes, the results are mixed. It's true that things don't always convert well to Math Type without having to edit the source code. But then this only means that these tools aren't quite there yet despite all the excessive hype around MathML and Math Type on these forums. It does not mean that reading LaTex directly should be the way to go.

As I I wrote in the previous e-mail, I use an old tool called WinTriangle for reading LaTex documents because it allows conversion to a clutter-free format with much less work. Yes, I do have to edit the source code sometimes in order to make it convert correctly. But most of the time I am able to simply convert it to WinTriangle without having to open the LaTex source code at all. Apart from the arxiv, this also applies to documents obtained by scanning something and converting it to LaTex through infty reader. They mostly convert nicely to WinTriangle without one having to even open the LaTex source code. This allows me to focus on reading the real content without having to waste time reading through all the clutter in the LaTex code.

So here is the revised bottom line: One doesn't and shouldn't have to spent a lot of time reading through the formatting mess in the LaTex source code in order to read math. Sighted people don't do it and we shouldn't have to do it either. If we're doing it, then we're using the wrong technology and lowering our efficiency.

On 1/11/2019 8:55 AM, John G Heim via BlindMath wrote:
> Yeah, I didn't see that in the original post but I write all my own 
> latex by hand. Every document I create for any reason, whether it is 
> for work or for one of the non-profits I am involved in, is written in 
> latex by hand. Then I convert it to pdf and send it out.
>
> I used to teach a mini-course in latex for grad students at the Math 
> Department at the University Of Wisconsin. We dropped it a long time 
> ago because grad students come in already knowing latex.
>
>
> On 1/11/19 6:48 AM, Łukasz Grabowski via BlindMath wrote:
>> The bottom line cited below is not quite true:
>>
>> "Bottom line: with several available alternatives, no one should have 
>> to read LaTex source code directly unless they really want to torture 
>> themselves."
>>
>> I know it for a fact that blind professional mathematicians read 
>> latex directly. If you don't believe, I suggest going to arxiv.org, 
>> downloading latex source for some new mathematical article and trying 
>> to convert it to mathml.
>>
>> The effect likely will be that no matter what software you use you 
>> will have to spend first a substantial amount of time editing the 
>> source code just to make your software parse the source. And even if 
>> you do succeed (which is FAR from given as not all latex pakcages 
>> will be implemented in your conversion software)) afterwards you will 
>> also surely need to consult latex source anyway while reading mathml.
>>
>> Having said that, this is about professional cutting edge 
>> mathematical research. For the purpose of high school students or 
>> early university (basic calculus, statistics, etc.), there is 
>> probably indeed no need, as there are good quality mathml materials available.
>>
>> Best,
>> Lukasz
>>
>> _______________________________________________
>> BlindMath mailing list
>> BlindMath at nfbnet.org
>> http://nfbnet.org/mailman/listinfo/blindmath_nfbnet.org
>> To unsubscribe, change your list options or get your account info for
>> BlindMath:
>> http://nfbnet.org/mailman/options/blindmath_nfbnet.org/jheim%40math.w
>> isc.edu
>>
>> BlindMath Gems can be found at
>> <http://www.blindscience.org/blindmath-gems-home>
>>
>
> _______________________________________________
> BlindMath mailing list
> BlindMath at nfbnet.org
> http://nfbnet.org/mailman/listinfo/blindmath_nfbnet.org
> To unsubscribe, change your list options or get your account info for
> BlindMath:
> http://nfbnet.org/mailman/options/blindmath_nfbnet.org/aqil_sajjad%40y
> ahoo.ca
>
> BlindMath Gems can be found at
> <http://www.blindscience.org/blindmath-gems-home>

_______________________________________________
BlindMath mailing list
BlindMath at nfbnet.org
http://nfbnet.org/mailman/listinfo/blindmath_nfbnet.org
To unsubscribe, change your list options or get your account info for BlindMath:
http://nfbnet.org/mailman/options/blindmath_nfbnet.org/a.j.godfrey%40massey.ac.nz
BlindMath Gems can be found at <http://www.blindscience.org/blindmath-gems-home>