Blinking Cursor Turns 54, Hardly Anyone Notices

In an interesting post on Inverse, [Sarah Wells] does a deep dive into something you probably don’t think about very often: the blinking cursor. You’d assume there wasn’t much to the story. Maybe a terminal manufacturer put a toggle flip flop on the cursor output and it caught on. But the true story is much deeper than that.

We were surprised that the father of the blinking cursor was one guy, [Charles Kiesling]. In a 1967 patent, he described the blinking cursor. An ex-Navy man, [Kiesling’s] patent names his employer at the time, Sperry Rand, where he’d worked since 1955.

According to the post, little is known of [Kiesling], one of the many unsung engineers who create everyday life. The article purports that the Apple II was the first place the general public would encounter the invention. We guess it depends on how you define the general public. The VT50 had a blinking cursor, we seem to remember, and we didn’t think it was the first, anyway. The VT05 in the video below seems to have a blinking cursor, too. And we think we remember blinking cursors on other terminals from that era for Lear-Siegler, Hazletine, and Televideo.

Regardless, the invention has stood the test of time. Humans are adept at noticing change and a blinking cursor draws your eye immediately. It works. Of course, once you have a cursor, you want to type something so you are going to need a keyboard. Or, voice command.

27 thoughts on “Blinking Cursor Turns 54, Hardly Anyone Notices

    1. I remember the first time I ever saw a computer. My dad had borrowed something like a TRS Model 3. I was about 5. He showed me how the cursor made letters on the screen. I asked if it ever ran out of ink.

        1. Between punch cards and video terminals there were teletype terminals (“TTY”), which printed on paper tape. We would edit our programs by printing them out on endless paper and editing them one line at a time.

          At that time, the PRINT command really did print, and the term stuck.

          Anyone remember edlin, which worked the same way on the IBM PC screen? No full-screen editing, just line by line in DOS 1.00.

          1. Yes! Not sure it was edlin but I remember ‘programming’* on an Olivetti M24 8086 (or was that an 8088) when I was 14. It was in BASIC (possibly GW-BASIC) and you had to list lines to edit. Good old days hehe.

            *I say programming, but it wasn’t quite that. I wrote a screen saver that drew lines floating around the screen. But I haven’t heard of for loops yet… So I hard coded each line by hand with the next coordinate lol.

          2. I wrote a very large embedded program for the 6805 using CPM cross compilers in one of my first jobs. I wrote it all in Ed which is this editor similar to Edlin. I told my boss that if he wouldn’t buy a new editor I would pay for it out of my own pocket. He said I’m not buying any more editors. We bought word star a year ago and no one could make it work. I said we have word star? About an hour later I was in heaven lol I used word star for many years after that but finally had to give up and go to word for documents and emacs for nearly everything else.

          3. When I was at Syracuse U in early 70’s, all the computer terminals were teletypes. State of the art! They had some PDP’s and i don’t know what else in the computer center.

          4. Yes, but mostly because I accidentally killed my high school’s only copy of MSDOS by using it and accidentally typed something into one of the config files that wasn’t supposed to be there. Thankfully we had a contract with IBM who were based in our city at the time so it was semi-easy to get fixed.

          5. Yes. I used in the Autocad 10 days, for making macros and for Autolisp. It was too easy to lose track of so many nested parentheses (hence, ‘lost in stupid parentheses’). Fun times.

          6. @[Kendotek}

            I used a formatting style to cope with nested parentheses or any other nested programing component from back in the mid to late 70’s. Except for Asm of course – where you can’t control the formatting.

            Now that formatting has been given a name: Whitesmith

            I wasn’t the only one to adapt that format after reading this –

            People still look at me weird lol. But it really does help a lot with deeply nested routines.

            // normal style

            function normal {
            // do stuff here
            if true {
            // and more stuff here


            function whitesmith
            // do stuff here
            if true
            // and more stuff here

            If someone complains then I add a variable “nesting_level” to functions and on conditions either increase, decrease, or leave unchanged the nesting_level and then have the function call itself unless the nesting level is decreased to 0 when it returns the result.

            If I do this then it is often quickly requested that I go back to the Whitesmilth format.

        2. Hah, reminds me too when I was a kid playing with windows 95 word processors, and for an embarassingly long time I thought the cut command would actually let you cut shapes out of the page when you printed it. I could never get it to work though!

          1. Paintbrush for Windows 3.1 in my case… this would have been ~1992 or so.

            At the time I thought that after doing a “cut”, you had to do a “copy” before you could paste more than one instance of the thing you just “cut”.

            The printer we had was a Fujitsu DL-1100 24-pin colour dot-matrix… I guess if you really made it hammer the page, you could make it perforate it, but I think it’d do likewise to the ribbon but I doubt the drivers supported that — I certainly don’t recall seeing that mentioned in the handbook.

            That was back in the days when printer manuals came with the full instruction set understood by the printer so you could write your own drivers, and yes, I had QBasic code that sent the appropriate codes using LPRINT statements. Can’t do that today unless your printer speaks PostScript.

            I miss those days.

  1. Each iteration means moe people exlerience something, but may be dwarfed by later iterations that bring in even more people.

    “Lots” of people experienced a flashing cursor before the Apple II, but it was a very specific segment of the population, and maybe “lots” is considerably bigger for the Apple II.

    Someone was asking about the Sphere 1 yesterday. Beats the Apple I and even the SOL 20 to the gate for having everything in a box. But it came so early, late 1975, that few experienced the ads when they were new. I remember a review, but wondered even then if there was more than review models. (Checking now, about 1300 were sold). Nobody mentions it because few experienced even the ads. More people knew about the Apple I, and even more about the Apple II (and then more for the Commidore 64, and the IBM PC).

    I was late when I got my KIM-1 in 1979, but it was still earlier than when most of the world came into contact with computers.

  2. The Singer/Friden System Ten mini computer released in 1969/70 had a Model 80 VDU with flashing cursor. I am not sure when the M80 was released but it was well before I started working with these ay the start of 1974, so I would say 1970/71 timeframe. I’m certain that many other computer companies had terminals with flashing cursors well before this timeframe.

    Just because a patent was granted doesn’t mean that it wasn’t in use before this date.

  3. This was one of the first bits of machine code one would write for the TRS-80 Model I.

    The first was a routine to eliminate keybounce so while you had a routine that was periodically polled at a relatively low rate, it was easy to add a counter and flash the cursor with a solid block.

    1. The TRS-80 had a funky line editor, you’d say “edit 100” and then use all manner of magic keys to Do Things to your line. I was quite adept at it, at the time, but all I recall of it now is pressing ‘L’ to list (i.e., show) the current state of the line.

      Wondering, was that based on something like edlin, or was that some novelty never seen before or since.

      1. I used a TRS-80 for many years and avoided most of the edit functions.

        Most of the time I would list the line and then re-type it correctly.

        On long lines I would use space bar to advance over the original text and then over type the rest of the line.

        It wasn’t like edlin. It was a propriety thing.

      2. Wondering, was that based on something like edlin, or was that some novelty never seen before or since.

        Don’t know, but i was surprised when i went to college in ’81 and discovered that the SOS editor on VMS was very similar. SOS was said to be an acronym that stood for “Son Of Stopgap”, with Stopgap being an earlier editor on an earlier machine. Perhaps it was a PDP-10 editor.

    1. Maybe it ought to be a thing, where if you are not moving the mouse for a period of time, the cursor changes to a blinking arrow. Once you start moving it, THEN it can stop blinking and switch to whatever cursor the app writer thought you should see.

      But then, what am I thinking? The age of mice is over anyway; we just point at things on the screen now.

  4. The first computer I ever saw was at a US Air Force base, when, during a Boy Scouts sleepover in a bomb shelter (somehow we did this), I found myself given a tour of the facilities along with a few other kids. This was in the very early 70s, and the person at the sulphur-screened machine told me he was chatting with someone in Europe, I had no idea why he wasn’t holding a phone.

Leave a Reply

Please be kind and respectful to help make the comments section excellent. (Comment Policy)

This site uses Akismet to reduce spam. Learn how your comment data is processed.