[Date Index] [Thread Index] [Author Index]

Re: A new FrontEnd

• To: mathgroup at smc.vnet.net
• Subject: [mg128045] Re: A new FrontEnd
• From: Richard Fateman <fateman at cs.berkeley.edu>
• Date: Tue, 11 Sep 2012 02:33:09 -0400 (EDT)
• Delivered-to: l-mathgroup@mail-archive0.wolfram.com
• Delivered-to: l-mathgroup@wolfram.com
• Delivered-to: mathgroup-newout@smc.vnet.net
• Delivered-to: mathgroup-newsend@smc.vnet.net
• References: <k21kj2$9vq$1@smc.vnet.net> <20120908071108.AFDA96898@smc.vnet.net> <k2hmii$ptu$1@smc.vnet.net> <504D2E99.6090007@cs.berkeley.edu> <k2k858$16g$1@smc.vnet.net>

9/10/12 1:25 AM, Leo Ditolaghi wrote:

> The reason I said "hybrid" was to recognize that the creation of a
> reliable handwriting input method is no easy feat, and therefore there
> would need to be some other augmenting technology perhaps. Perhaps, for
> example, writing with a stylus with one hand while using a some kind of
> "wireless modifier controller" in the other, in tandem.
plus 2 more hands for the keyboard.   Maybe another to hold the coffee.

I wonder if
> using speech to locate or edit symbols might not be too tedious to be
> practical, however.

saying "no"  or "back" or "up"  is easy/quick/ probably reliable.
Eye-tracking is an old technology, but doesn't seem to be out in the
consumer market.

>
> I guess I will wait for brainwave-driven input after all, and merely
> watch what I think appear in my document. :D

Most people have either experienced no significant speech input, or
have had relatively poor experiences... the most reliable versions
require some speaker-specific training and a restricted vocabulary.

It is possible to have a large vocabulary or speaker independence.
It is hard to have both.  I'm not sure that brainwaves would be a
step forward.

>
> Leo
>
>> Richard Fateman <mailto:fateman at cs.berkeley.edu>
>> Sun, Sep 9, 2012 8:04 PM
>>
>>
>> Handwriting recognition of mathematics has a history going back to
>> about 1963.  windows 7 has a math input panel.
>>
>> There are excellent reasons for this not to be in widespread use.
>> Mostly it is difficult to build a good system (at least so far).
>> Secondly, it is difficult to use.  You may think it is "obvious"
>> but what is obvious about correcting recognition errors, which are
>> inevitable.
>>
>> An alternative, especially in these days of Siri etc, is speech
>> input of mathematics.  Odd as it may seem at first, there are
>> good reasons for it to be faster and more effective than handwriting,
>> typing, or palettes.
>>
>> Google is your friend in tracking down information on all of these
>> issues.  It may help to include "fateman" in the search.
>>
>> RJF
>>
>>
>> Mikhail Cherkasskii <mailto:macherkasskiy at gmail.com>
>> Sat, Sep 8, 2012 3:11 AM
>>
>> Maybe it a good idea to collect some suggestions for new version of
>> FrontEnd. To be more certain I suggest to create an integrated
>> interface with tabs like Google Chrome or another CAS program. I'd
>> like to see more instruments for edit graphics and text. Debug tool
>> also welcome. The document based interface is useful, but when there
>> are more then 30 variables in your notebook, some of them is in a RAM,
>> some aren't... I prefer to have an extra tool for variable control.
>>
>> Best wishes
>>

• Follow-Ups:
  • Re: A new FrontEnd
    • From: Leo Ditolaghi <leoditolaghi@gmail.com>

• References:
  • Re: A new FrontEnd
    • From: Mikhail Cherkasskii <macherkasskiy@gmail.com>