Thanks for the comprehensive feedback. I hope my answers cover all your points.
The delete and backspace keys are not supported on purpose. You don't delete objects that often, but you quickly hit one of these keys by mistake. Therefore, I feel it is better to add support for these keys not before there is an undo function implemented, which is challenging.
Studying pages
http://www.firemath.info/examples/example05/http://www.firemath.info/keyboard/you learn how your sample equation can be created solely with the keyboard. Key strokes are (S-A == Shift-Alt)
0 S-A-i < i S-A-i < n S-A-i S-A-t , S-A-x 1 S-A-i < j S-A-i < m S-A-i S-A-t , S-A-x i S-A-i < j S-A-i
The Left and Right arrow-keys bring the cursor into the subscript.
Firefox has already lots of hotkeys defined (e.g. Ctrl-F opens the search field). I didn't want to replace too many of these. Therefore the combinations with Shift-Alt or Ctrl-Alt. I don't know how many variations of keyboard layouts exist in this world. At least on my keyboard pressing Shift-Alt is only slightly more difficult than just pressing Ctrl. For comparison: In Latex you frequently need a backslash, which each time makes you break your fingers if you have to work with a German keyboard.
Mathematics (and Physics and the like) use a huge set of operators, variables, etc. I think Firemath should provide the most flexible solution, i.e. everyone should be able to get what he reaches for. On the other hand a large feature set is somewhat contradictory to an easy to use interface. This certainly means I am open for proposals on how the interface can be improved.
The W3C decided to differentiate between operators, identifiers and text within MathML. This has to be incorporated somehow into Firemath. It seems to me only the user knows into which of these categories the object falls, which he wants to add. Installing an automation here implies the danger that people pollute the Web with poorly rendered MathML. You certainly know lots of examples in which Latex is used rather naively. To me it seems difficult to write the parser you propose. With respect to your example: Is the comma an operator or is it text? Even if this can be answered easily it should not be difficult to find examples, where any automation must fail.
I worked with Latex a lot. While a simple equation like 0<n<j can be generated quickly with Latex, it often took me a significant part of my life to get more complex equations rendered properly. I know there are tools to simplify that. But do they run on various operating systems? Are they open source? I cannot make up my mind, yet, on your proposal to allow Latex style entries. On first view, I don't like the idea but I might change my mind. BTW: You find Latex to MathML converters on the internet.
Thanks for the comprehensive feedback. I hope my answers cover all your points.
The delete and backspace keys are not supported on purpose. You don't delete objects that often, but you quickly hit one of these keys by mistake. Therefore, I feel it is better to add support for these keys not before there is an undo function implemented, which is challenging.
Studying pages
http://www.firemath.info/examples/example05/
http://www.firemath.info/keyboard/
you learn how your sample equation can be created solely with the keyboard. Key strokes are (S-A == Shift-Alt)
0 S-A-i < i S-A-i < n S-A-i S-A-t , S-A-x 1 S-A-i < j S-A-i < m S-A-i S-A-t , S-A-x i S-A-i < j S-A-i
The Left and Right arrow-keys bring the cursor into the subscript.
Firefox has already lots of hotkeys defined (e.g. Ctrl-F opens the search field). I didn't want to replace too many of these. Therefore the combinations with Shift-Alt or Ctrl-Alt. I don't know how many variations of keyboard layouts exist in this world. At least on my keyboard pressing Shift-Alt is only slightly more difficult than just pressing Ctrl. For comparison: In Latex you frequently need a backslash, which each time makes you break your fingers if you have to work with a German keyboard.
Mathematics (and Physics and the like) use a huge set of operators, variables, etc. I think Firemath should provide the most flexible solution, i.e. everyone should be able to get what he reaches for. On the other hand a large feature set is somewhat contradictory to an easy to use interface. This certainly means I am open for proposals on how the interface can be improved.
The W3C decided to differentiate between operators, identifiers and text within MathML. This has to be incorporated somehow into Firemath. It seems to me only the user knows into which of these categories the object falls, which he wants to add. Installing an automation here implies the danger that people pollute the Web with poorly rendered MathML. You certainly know lots of examples in which Latex is used rather naively. To me it seems difficult to write the parser you propose. With respect to your example: Is the comma an operator or is it text? Even if this can be answered easily it should not be difficult to find examples, where any automation must fail.
I worked with Latex a lot. While a simple equation like 0<n<j can be generated quickly with Latex, it often took me a significant part of my life to get more complex equations rendered properly. I know there are tools to simplify that. But do they run on various operating systems? Are they open source? I cannot make up my mind, yet, on your proposal to allow Latex style entries. On first view, I don't like the idea but I might change my mind. BTW: You find Latex to MathML converters on the internet.
