11

I was writing an essay today about using the Apple IIe I grew up on, and I wanted to add some examples that included the "delete" symbol that was used for the cursor:

Apple IIe cursor symbol

But try as I might, I couldn't find this anywhere in Unicode. The Unicode Block Elements include ░ and ▒, and Unicode Geometric Shapes includes things like ◼ and ▣ and ▩, but none of those are that very distinctive 5x5 checkerboard.

The recently-added Symbols for Legacy Computing block includes MouseText's shading characters, but still not the unique Apple IIe "delete" shape. Since Unicode didn't add it to "Symbols for Legacy Computing," that would presume it exists somewhere else in Unicode already, but I haven't been able to find it, and this approximation with a box simply doesn't look right:

]10 PRINT "HELLO, WORLD"□

So is there an actual Unicode code point for this character? Does it have a better name than just "the delete symbol"? Did it get left out of Unicode?

7
  • 2
    Symbols for Legacy Computing lists 2425 SYMBOL FOR DELETE FORM TWO (from the Control Pictures group) as the code point for Apple $FF. However, the glyph is '␥', which is totally wrong.
    – fadden
    Commented Sep 29, 2023 at 15:59
  • 1
    This seems to be a question about Unicode, not retrocomputing.
    – dave
    Commented Sep 29, 2023 at 23:35
  • You should be able to embed a SVG or bitmap of it in your document.
    – Davislor
    Commented Sep 30, 2023 at 0:02
  • 2
    Seeing it as an intentional “5x5 checkerboard” seems somewhat anachronistic: as as I remember, that sort of single-pixel checkerboard shading was virtually always intended and understood as representing “gray” (or more precisely “50% intensity foreground color”), not as checkered. So wanting the checkered implementation feels like wanting an authentically pixelated low-resolution form for other characters — something I’d expect to be implemented by specific fonts, not in the character/glyph specification. Commented Oct 1, 2023 at 9:24
  • 1
    @PeterLeFanuLumsdaine I've added a note about the fact that Unicode is now apparently in process of adding this symbol, so they consider it sufficiently distinct that it was worth adding (in the Basic Multilingual Plane, no less!). And they call it a checkerboard too, so that's probably what it is — Apple is a full voting member of the Unicode Consortium, so we have to presume they're signing off on that name. Commented Oct 1, 2023 at 14:00

3 Answers 3

14

TL;DR:

There is no 'REVERSE FIVE-BY-FIVE CENTERED CHECKER BOARD' (*1) in all of Unicode's blocks. Closest by looks and positioning might be the

  • "REVERSE FOUR-BY-FOUR CHECKER BOARD" U+1FBA6

Alternatives

The

  • "SQUARE WITH DIAGONAL CROSSHATCH FILL" U+25A9

might be a good close choice if it's about recreating the appeal independent of output media while

  • "BLACK SQUARE FOR STOP" U+23F9

from the Media Control section will most often result in the right placement.

Those are, unlike the 'Symbols for Legacy Computing', of older addition, thus more likely to be present in output charsets for your document. Ignoring the rendering issue, I would think the

  • "REVERSE FOUR-BY-FOUR CHECKER BOARD" U+1FBA6

would do the best job creating the impression of the checker board nature of the IIe character. As shown with modern '8 bit games', it's only relevant what impression is created, not that it's really in all details true to source material (*2).

No Chance for the Real One?

Well there is. Adding a

  • "SYMBOL FOR DELETE SQUARE CHECKER BOARD FORM" at U+2427

was proposed January 2022 and might as well be on the road to addition. It may take some time until fonts will adopt it, so until then the reverse 4x4 should do the job.


Background

Apples Choice

The 5x5 middle aligned checkered symbol is what Apple placed as glyph for the DELelte character ($7F). As a code to remove any other code it does by nature not have any visualisation (*3). With this it has been a common code place for machine specific additions (*4).

MouseText

The checker board is not part of Mousetext. Its two 'half' filled blocks at code position $56/$57 are different and represented by "MEDIUM SHADE" (U+2592) and "INVERSE MEDIUM SHADE" (U+1FBA0).

The Questions

[...] would presume it exists somewhere else in Unicode already

Nop, it doesn't.

and this approximation with a box simply doesn't look right:

]10 PRINT "HELLO, WORLD"□

Why? I do in fact think of that as a very good representation. Not at least as it should work with most output media due being an older addition.

So is there an actual Unicode code point for this character?

No

Does it have a better name than just "the delete symbol"?

No, not even that, as that naming is only true for the Apple IIe.

Did it get left out of Unicode?

Yes.


*1 - Unicode naming of "REVERSE FIVE-BY-FIVE CENTERED CHECKER BOARD" a in

  • INVERSE since Unicode description is made assuming a black on white rendering and a checker board starts out with white in the top left corner, but ours need to be black
  • FIVE-BY-FIVE as the obvious structure
  • CENTERED as we want to have it hover in the middle of the character cell
  • CHECKER BOARD to denote the alternating fore-/background scheme

*3 - Many of those modern '8-bit' style games do not really use scaled up 8 bit graphics but high resolution, often even 3D models made to give that look while still using finer details. Did already turn me off more than one otherwise nice game.

*3 - Beside using it's control picture of U+2421 [DEL].

*4 - Commodore used for example a "SQUARE WITH UPPER LEFT TO LOWER RIGHT FILL" (U+1FBA7) or "BLACK UPPER RIGHT TRIANGLE" (U+25E5) depending on charset.

16
  • 2
    "SYMBOL FOR DELETE SQUARE CHECKER BOARD FORM" is definitely the right answer! It's a shame it's not already available, though. As for why the others aren't right... well, the original is burned into my brain after a decade of using it. It wasn't a box, it wasn't a 4x4 checkerboard, and it wasn't a reference mark: It was a 5x5 checkerboard. Commented Sep 29, 2023 at 19:33
  • 1
    @supercat I started out on an 11" phosphor display, clear and sharp, green-on-black, which I used for years. On that screen, it was very distinctly a 5x5 checkerboard, not the fuzzy mash of pixels you saw on other screens of the era. Given how clear it was on that screen, it's hard for me to see anything but a checkerboard as the true correct shape now. It's a valid point to ask about 0x2F (or 0xAF), but I think there's less ambiguity about its intent than there is for 0x7F (0xFF), so I'm more inclined to believe that the checkerboard is supposed to be a checkerboard, as presented. Commented Sep 29, 2023 at 20:30
  • 3
    @supercat You can't answer about hypothetical parallel realities that don't exist. The Apple II line had 280x192 pixels, and the character cells were 7x8 each, and the image was what it was. Unless the designer of that glyph pops in to tell us what it was intended to be, the only thing we have to go on is what it is: a 5x5 arrangement of alternating bright and dark pixels. Commented Sep 29, 2023 at 21:10
  • 2
    "REVERSE FOUR-BY-FOUR CHECKER BOARD" U+1FBA6: You seem to be referring to "U+1FB95 CHECKER BOARD FILL" or "U+1FB96 INVERSE CHECKER BOARD FILL". U+1FBA6 Is a different, non-checker-related character. ref: unicode.org/charts/PDF/Unicode-13.0/U130-1FB00.pdf
    – Miles
    Commented Sep 30, 2023 at 8:30
  • 3
    I personally think this Answer would look better if you also included the actual Unicode symbols (or pictures of them) to enhance the descriptions and arguments of similarity.
    – trlkly
    Commented Sep 30, 2023 at 22:00
9

Only because @trikly asked (in this comment), from the relevant Unicode documentation(PDF), the three new "Specific symbols for delete" that have been accepted for future inclusion in the standard: (Seems Apple wasn't the only system to use such a symbol to represent Delete.)

enter image description here

Specific symbols for delete

  • 2427 SYMBOL FOR DELETE SQUARE CHECKER BOARD FORM
    • symbol for delete in the Apple II character set
    • fills the bounding box of 002B +
  • 2428 SYMBOL FOR DELETE RECTANGULAR CHECKER BOARD FORM
    • symbol for delete in the TRS-80 character set
    • fills the bounding box of 0048 H
  • 2429 SYMBOL FOR DELETE MEDIUM SHADE FORM
    • symbol for delete in the Amstrad CPC character set, which also encodes 2592 ▒ separately
    • fills the character cell → 2592 ▒ medium shade
5

In order to determine what Unicode symbol should be associated with that character, one would have to ascertain what the dot pattern was supposed to represent. The particular pattern of dots used for the character was chosen to fit in a 5x8 matrix, and some design aspects might have been different if a different size of matrix had been used. In a 7x9 matrix, for example, there might have been four illuminated pixels on each side of the shape rather than three, or the shape used for the 5x7 matrix might have had a frame added around it (trying to have a frame around a checkerboard pattern in a 5x5 area would make the middle look like an "X" or "O" rather than a checkerboard). If one assumes the goal was to have a shaded square which does not fill the character box, I think Unicode points 25A7 through 25A9 ("Square with [different kinds of diagonal] Fill") would all be perfectly reasonable candidates, with 25A9 probably being the best. Point 25A9 looks reasonable in both normal and fixed-spacing fonts: ABC▩EFG and ABC▩EFT; it's centered within the X height, but shorter, and its interior is filled but not solid.

Another possibility would be 203B "Reference Mark" which looks like "ABC※EFG" or ABC※EFG. It wouldn't match the dot pattern when magnified, but one would hardly argue that character 0x2F shouldn't be recognized as a slash, but rather as a diagonal sequence of five dots. Someone given the task of representing ※ in a 5x7 matrix would be likely to produce the exact same pattern of pixels as Apple used for their 0x7F character.

5
  • 1
    While the other answer that notes that "Unicode is adding this missing symbol soon" is probably the right answer, and while I think "5x5 checkerboard" is distinct from "reference mark," I'll give you a solid thumbs-up for a well-reasoned argument for ※. It arguably isn't the same symbol, but it's not a bad answer when there's no 5x5 checkerboard. Commented Sep 29, 2023 at 19:36
  • @SeanWerkema: I don't think it's clear that the character is supposed to represent a 5x5 checkerboard. While I don't think an X with four dots is a particularly plausible interpretation of what the dots represent, I think a superimposed X and diamond shape would be just as reasonable as a 5x5 checkerboard with horizontal and vertical sides.
    – supercat
    Commented Sep 29, 2023 at 20:18
  • @supercat " I don't think it's clear that the character is supposed to represent a 5x5 checkerboard." It's not. It's supposed to represent deletion, an invisible modification to on-screen text that's characterized by the removal of visual information. Because such a thing is entirely conceptual, it has no inherent visual representation. So, Apple defined a particular 5x5 checkerboard configuration of pixels to represent that concept. Attempting to interpret it as a representation of some other visual symbol is wrong. It represents something abstract, not concrete.
    – FeRD
    Commented Dec 2, 2023 at 16:00
  • @FeRD: By that logic, if a font rendering engine (e.g. one used by a pen plotter) requires that any character within a certain range has a collection of visual strokes associated with it, any collection of strokes which could be interpreted by a viewer as an obliterated character would be appropriate. I'm not sure a diamond with a superimposed X would convery that notion superbly well, but I don't think it would do any worse would a shape that drew the outlines associated with a checkerboard.
    – supercat
    Commented Dec 2, 2023 at 17:20
  • @FeRD: I suppose one could maybe argue that the proper depiction in print would be a hole in the paper, but most kinds of printing devices would have no practical means of achieving that without causing excessive wear or danage to the printing aparatus.
    – supercat
    Commented Dec 2, 2023 at 17:23

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .