Talk:Accessibility Computing Numerical Pointer

okay, moving discussion from article page

{{delete}}
This page seems completely useless to me. Someone should figure out what it is. --jacobolus

The original author of this page disagrees with the request to delete the page. The idea is defended on the page itself and that defence is repeated below as the rules of deciding upon a deletion request made on the article page requires defence on the discussion page rather than on the article page where the deletion request has been made! The idea is certainly open to critical review and indeed the decision may be taken not to take up the idea. Yet no specific technical reason for wanting the page deleted has been put forward at this time. Critical review is needed as to whether the idea could be useful to the OLPC project before any decision to delete the page is made.

Well, suppose that a child has arthritis and cannot move the pointer properly yet can press on the number keys. This method would allow that child to move the mouse pointer around on the screen.

• The method allows any user to position the mouse pointer with pixel precision.
• This method allows any user to enter any Unicode character into the laptop. Suppose that someone in Brazil wishes to enter French accents, how else would he or she do it using the keyboard on his or her machine? What about the accents for Esperanto, or for Latvian?

  [1] Keyboard Artwork Library page

  [2] French pdf

  [3] Esperanto pdf

  [4] Latvian pdf

  [5] List of pdfs available

• This method allows colours to be selected with precision.
• This method is expandable for other features as well.

Alright, I suppose I should have phrased this differently. I don't understand at all what this spec does. It has no high-level description of how it works, or how it will be used. It wasn't clear to me how it had anything to do with the other aspects of the software on the laptops. To my untrained eye, it seems vastly complicated, far beyond the ability of any 8 year-old child to comprehend, let alone use. --Jacobolus 13:03, 14 February 2007 (EST)

Well, in fairness there is no explanation of how any 8 year-old child would comprehend or use the system. That is my omission and is needed, so here is an explanation.

The system would be implemented by having a wide, not too high, panel appear along the bottom of the screen once entry was requested to the Accessibility Computing Numerical Pointer. The panel would describe the key presses currently available and what they did.

For example, at start up.

0 -> Unicode room

Mouse event generation

1 sets left button

2 sets right button

3 sets centre button

4 -> key setting

5 -> pointer event selection

6 -> expansion and resetting

7 -> x value

8 -> y value

9 -> z value

If, say, 7 were keyed, the panel would be redrawn showing that the current value of x is 0.

The panel would state that digit keys could be used to set x and that Enter would go back to the main room.

Each digit press would update the value of x in the display.

Keying Enter would redraw the original panel, displaying on that panel the current value of x.

The child would not need a printed manual nor knowledge of the underlying software: he or she would be propmpted by on-screen explanation of which key has which effect at that stage of the process.

(unindenting) To clafify further, this specification seems to be a massively complicated finite state machine. Computers are good at dealing with such structures. Humans are not (having taken several years of difficult university-level mathematics and computer science, I cannot imagine myself keeping such a structure in my head). For this system to be used by a student, a printout of all the states would be required. Given that these laptops are going to many places without printers, and will be used far away from the classroom, this is impractical to an extreme.

Instead, useful operations in every activity should have relevant keyboard shortcuts as an alternate invocation method. I can possibly even understand a mechanism for using the arrow keys to move the mouse. But using the row of numbers as a state machine to accomplish every possible touchpad and keyboard event in several arbitrary presses, seems just absurd to me. If I might ask, who came up with this system? --Jacobolus 13:19, 14 February 2007 (EST)

I agree with Jacobolus... this is way to complicated. If the idea is 'useful' maybe the application should be developed (even just as proof of concept) and then passed onto the software community to test the idea... full text-specs of it here doesn't seem appropriate. IOW, I wouldn't miss this page if it's deleted. --Xavi 13:34, 14 February 2007 (EST)