Color coded data-hints

Questions about MultiCharts and user contributed studies.
denizen2
Posts: 125
Joined: 17 Jul 2005

Color coded data-hints

Postby denizen2 » 22 Nov 2006

The data-hint readout would be so much more simple if it was just color-coded the same as the plot. :idea: Now, I have to remember, or recognize, which data-label is associated with the data-indicator that I am trying to get a 'data-hint' readout. Using the color-association cues would be so much easier for the user to quickly find the data item of interest.

User avatar
Alex Kramer
Posts: 834
Joined: 23 Feb 2006

Postby Alex Kramer » 23 Nov 2006

Thanks for the input - this makes sense, adding this to the ToDo list. The data hint window can be made easier to read, especially with several data series.

denizen2
Posts: 125
Joined: 17 Jul 2005

Postby denizen2 » 27 Nov 2006

Hi Alex,

Here's one more suggestion to add while the developers are 'thinking' about color coding the data-hints to match the colors used on the plot.

Namely, drop the data-hint-reporting of the CONSTANTs, e.g., plot3(0,"0"), plot4(1,"1"), and plot5(-1,"-1"), etc. It is really silly to clutter up the hints with a bunch of numbers that are nothing but constants used as a 'reference' level. I have some plots with five of these constants! What a clutter it causes :( .

Maybe it could be setup so that there is a PLEditor-script-programming convention that says if the plot() function is called without giving any specific "label", then it will be ignored by the data-hint display procedure :idea: ? I tried using the plot() function in that manner, e.g., Plot(0), with no "0" included, and the hint still shows it as a constant, with a program assigned label defaulting to the Plot#.

An even *better* solution, but requiring more design work I am sure, would be to provide the user (in the indicator style and parameter dialog) a way to add any fixed level lines, so that it was not necessary to manually insert program-code for all of these constant reference-levels. This feature is part of the Metastock software, and I found it very useful and convenient approach. TS, and MC too, have not picked up on that idea, but should :idea: . In the mean time, the first idea could probably be done with very little programming effort, since it only requires that any data-hint to be displayed have a user-specified "Label". What do you think?

User avatar
Alex Kramer
Posts: 834
Joined: 23 Feb 2006

Postby Alex Kramer » 28 Nov 2006

Namely, drop the data-hint-reporting of the CONSTANTs, e.g., plot3(0,"0"), plot4(1,"1"), and plot5(-1,"-1"), etc. It is really silly to clutter up the hints with a bunch of numbers that are nothing but constants used as a 'reference' level. I have some plots with five of these constants! What a clutter it causes Sad .

Could you please provide a screenshot?

An even *better* solution, but requiring more design work I am sure, would be to provide the user (in the indicator style and parameter dialog) a way to add any fixed level lines, so that it was not necessary to manually insert program-code for all of these constant reference-levels.

This has a point, but in our opinion we'd rather have a line written for once in the inputs that can be easily enabled/disabled or have its value set. This, rather than cluttering the interface with more controls.

We're focusing on principal functionality, this can be done in EasyLanguage and doesn't seem a major issue to us.

denizen2
Posts: 125
Joined: 17 Jul 2005

Postby denizen2 » 29 Nov 2006

Alex,
You did agree that the data-hint-display should/could include color-coding to correspond with study-indicator color. so, I don't understand why you just dismissed the 2nd related idea of also suppressing the 'junk' and 'clutter' contributed by CONSTANTs levels being in the data-hint-display. :cry:

In fact, this issue of need-to-ignore-constant-levels is ALSO important when 'exporting' the data to a file for any external processing. All of those constant-levels (e.g., "0", "-1", "1", "100", etc,) have to be 'edited out typically, and constitue a 'hassle' for most users :? .

I'm NOT suggesting that it is desired to eliminate the constant-levels from being specified in the study-code. Rather, I am only suggesting that it would be simple and very desirable to limit its use to the intended plot-display, AND eliminate those constants-levels from such things as data-hint readouts and export functions.

The suggested approach to doing this is to merely add the requirement that only "user-labeled' indicator levels will be exported or included in the data-hints. That way, if I code my indicator(s) constant-levels like this: Plot1(1), i.e., with the no-label type syntax option, then it would not be exported nor displayed by the data-hint. Thus, simply eliminating a lot of unnecessary clutter.

This would not require any change to the PLEditor compiler, but only a change to the exporting and data-hint logic so that it filtered out any constant levels that are coded without user-supplied labels.

Right now, the data-hints don't work half the time, as described by my previous posts, plus *when* it does work, it is degraded by
(a) use-less 'clutter' of possibly many constant levels, and/or
(b) non-colored-coded lines.
So users of data-hints and export-data functions would greatly appreciate it if those 'features' were more functional :roll: . It shouldn't even be a big design issue to fix/implement, so the benefit to cost ratio would be attractive to everyone, right?


Return to “MultiCharts”