MathGroup Archive 2007

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

Search the Archive

Re: Re: Compatibility woes

  • To: mathgroup at
  • Subject: [mg76242] Re: [mg76099] Re: Compatibility woes
  • From: DrMajorBob <drmajorbob at>
  • Date: Fri, 18 May 2007 06:16:33 -0400 (EDT)
  • References: <f295kp$n4k$> <30568553.1179316638063.JavaMail.root@m35> <> <13780686.1179343455242.JavaMail.root@m35> <> <17717411.1179414449369.JavaMail.root@m35>
  • Reply-to: drmajorbob at

> Even worse, if the entry is not discrete, but continuous,
> there is no guarantee that you will get a wanted value.

This is backwards; there's no way to get all real values in an interval =


Sliders are appropriate for setting real-valued quantities, not discrete=

ones like the index into a list.


On Thu, 17 May 2007 10:05:40 -0500, Yu-Sung Chang <yusung at>  =


> Of course, the slider is susceptible to the length of the entry it is =

> representing. Even worse, if the entry is not discrete, but continuous=
,  =

> there is no guarantee that you will get a wanted value.
> Scrollbar is obviously better option for this particular case (palette=
),  =

> but unfortunately, we don't have scrollbar (yet) due to few reasons. B=
ut  =

> it will be revised in the future.
> Yu-Sung Chang
> Kernel Technology
> Wolfram Research, Inc.
> DrMajorBob wrote:
>>> slider moves directly to the mouse position. Concerning the slider  =

>>> speed, we found that it is reasonably responsive (not too fast or no=
t  =

>>> too slow) for many. Actually, it is up to the system mouse sensitivi=
ty  =

>>> setting, not Mathematica.
>> I belatedly realize that can't be true.
>> No matter how sensitive (or not) the mouse may be, the maximum mouse =

>> movement (equal to the visible height of the control) must equate to =

>> moving through ALL members in the list. So the speed at which the lis=
t  =

>> flies by, for ANY given mouse movement, is (and has to be, for a  =

>> slider) proportional to the number of entries in the list.
>> Hence sliders are tolerable for small lists, intolerable for very lon=
g  =

>> lists. For a list with millions of entries, one pixel of mouse moveme=
nt  =

>> on the screen could equate to thousands of entries, but maybe we're  =

>> only seeing five or ten.
>> Screen resolution has an effect; high resolution => short visible  =

>> control => faster movement through the list. I run my screen at  =

>> 1280x1024, so the "speed" I'm looking at is 25% greater than what you=
'd  =

>> see at 1024x768.
>> No?

-- =

DrMajorBob at

  • Prev by Date: Re: Guessing "exact" values
  • Next by Date: Re: Phase Portrait
  • Previous by thread: Re: Re: Compatibility woes
  • Next by thread: Re: Compatibility woes