Developer Returns To Game After Four Decades, Discovers and Fixes Typo So It Works (tomsguide.com) 98
joshuark writes: Harry McCracken is not the name of a Cold War superspy, but a man who is now the tech editor of Fast Company and, in his younger days, a developer of games for Radio Shack's TRS-80 microcomputer. McCracken, who is also a regular Slashdot reader, recently went back to have a look at his first game, Arctic Adventure, which he wrote when he was 16 around 1980-81 -- a text adventure inspired by the work of Scott Adams in particular, a pioneering designer of the Adventure series of games for the TRS-80.
As was common in the 80s, Arctic Adventure was distributed in book form. This was The Captain 80 Book of BASIC Adventures: pages of type-it-yourself BASIC code, each entry its own adventure game. [...] "Decades later, I didn't spend much time thinking about Arctic Adventure, but I never forgot the fact that I hadn't received a copy of the Captain 80 book. Thanks to the internet, I eventually acquired one. But typing in five-and-a-half pages of old BASIC code seemed onerous, even if it was code I'd written."
McCracken eventually got around to it this July. "After five or six tedious typing sessions on my iPad, I had Arctic Adventure restored to digital form. That was when I made an alarming discovery: As printed in the Captain 80 book, the game wasn't just unwinnable, but unplayable. It turned out that it had a 1981 typo that consisted of a single missing '0' in a character string. It was so fundamental a glitch that it rendered the game's command of the English language inoperable. You couldn't GET SHOVEL let alone complete the adventure."
As was common in the 80s, Arctic Adventure was distributed in book form. This was The Captain 80 Book of BASIC Adventures: pages of type-it-yourself BASIC code, each entry its own adventure game. [...] "Decades later, I didn't spend much time thinking about Arctic Adventure, but I never forgot the fact that I hadn't received a copy of the Captain 80 book. Thanks to the internet, I eventually acquired one. But typing in five-and-a-half pages of old BASIC code seemed onerous, even if it was code I'd written."
McCracken eventually got around to it this July. "After five or six tedious typing sessions on my iPad, I had Arctic Adventure restored to digital form. That was when I made an alarming discovery: As printed in the Captain 80 book, the game wasn't just unwinnable, but unplayable. It turned out that it had a 1981 typo that consisted of a single missing '0' in a character string. It was so fundamental a glitch that it rendered the game's command of the English language inoperable. You couldn't GET SHOVEL let alone complete the adventure."
POST COMMENT (Score:1)
[Enter]
That was typical (Score:5, Interesting)
Re: (Score:2)
My family had an Amstrad CPC. The manual had a large chapter on BASIC (and another one on LISP, which I didn't properly appreciate at the time) as well as some games in the appendices. And, as you say, the games had a couple of bugs. Fortunately for me, by the time I got round to putting in the effort to type in the hundreds of lines of those games I'd learnt enough from the BASIC chapter to debug them.
Re: (Score:2)
Re: (Score:2)
You're right. I think I made that mistake because what I was thinking about with "didn't properly appreciate at the time" was the power of the list processing commands.
Re: (Score:2)
Re:That was typical (Score:5, Informative)
As a kid who had several of those books in the early 80's and spent forever typing them into my Atari 800XL computer, I can assure you that about half of them had typos and just spat out an error message. That was after having a parent proof read the whole thing line by line and not finding any mistakes. Unfortunately, since I didn't really understand programming at the time, I wasn't able to debug it. That was very frustrating.
It was typical for a coding book, to contain broken code?
And yet we complain about quality control today...
Re: (Score:3)
Yes it was typical. The publishing industry back then did not copy-edit people familiar enough with the subject to spot problems in the jargon laden narrative, let alone technical review people to evaluate if a code sample likely contains an error.
Re: (Score:3)
Yes it was typical. The publishing industry back then did not copy-edit people familiar enough with the subject to spot problems in the jargon laden narrative, let alone technical review people to evaluate if a code sample likely contains an error.
Ironically enough, a "learn to code" book has but one job at the end of the day; to teach you how to code.
I guess my point here is, all the publishing/editing excuses aside, the creator didn't even bother to validate their own creation until decades later? The creator, never got word from a single frustrated customer that this was a "typical" problem in this series of books? Not a single parent of a frustrated child was also a coder? Hell, that last one makes the least amount of sense. Quite often we fin
Re:That was typical (Score:5, Funny)
That's how those books turned into "Learn how to debug" books. Two for one - quite a bargain!
Re:That was typical (Score:4, Funny)
That's how those books turned into "Learn how to debug" books. Two for one - quite a bargain!
Ah, nothing like pulling the turn signal fuses out of the car right before the student driver takes over. Learn how to drive and survive a crash, all in the same training session - quite a bargain!
Re: That was typical (Score:2)
Car crash, computer crash... what's the difference?
Re: (Score:2)
Car crash, computer crash... what's the difference?
Well Sarcasm, death is the most obvious one, but you probably knew that.
Re: (Score:2)
The student driver should learn to look around the car to make sure a car they've never driven works the way it's supposed to. I think that's the first class in driver's ed.
That's actually one of the first lessons, and one that has to be reinforced repeatedly for new drivers, especially 21st Century addicts who are still trying to get their phone synced to the car to get their "Drive and Chill" playlist going, before they even bother with a seatbelt.
#Priorities
Re: (Score:2)
By the creator's own admission, he never received a copy, so no way to tell that someone had botched his work.
Re: (Score:2)
Re: (Score:2)
Yes, that's why so many books and magazines didn't typeset code, but instead used printouts pasted onto the camera-ready copy.
It used to drive me crazy seeing low-quality dot-matrix printouts reproduced badly, but the alternative was potentially broken code.
Apparently, I was lucky to rarely run across a bad type-in. I wonder if it wasn't just because I was really patient...
Re: (Score:2)
This. They neither reproduced a photocopy of a known working printout, nor piped ASCII directly into a computerized printing system, which did no exist, at least outside NASA.
So some bozo got paid to create it the hard way for the printer, and made goofs, statistically likely in spite of what one hopes would have been multiple proofreads before pressing print three hundred thousand.
Re: (Score:2)
I guess my point here is, all the publishing/editing excuses aside, the creator didn't even bother to validate their own creation until decades later? The creator, never got word from a single frustrated customer that this was a "typical" problem in this series of books? Not a single parent of a frustrated child was also a coder? Hell, that last one makes the least amount of sense. Quite often we find geeks raising geeks. Can't believe no one noticed a single piece of missing syntax, and an error making a game unplayable isn't exactly a minor glitch.
More than likely, the creator got like $20 to acquire the rights to put their code in this book. The publisher likely never even ran the program or evaluated it in any way - or if they did, it was on the order of "Does it launch? Ship it!" A single-letter typo is actually best case, you could have entire pages missing in these things.
And what would they do about it? Say the publisher remembered to send a copy to the creators, and the creator laboriously entered in a copy of a program they already had on
Re: (Score:2)
At this time in human history, corporations and governments were CMM level -14.
Type it in and make sure it runs as a safety check ahah hah hahahhahahahaha!
Re: That was typical (Score:2)
I was thinking the same thing. Surely while writing the program, he was continually going back to it to see if so and so was working the way he intended?
The only way I could see this mistake making it to the publisher is if he hand copied/retyped it on a typewriter before sending it out which would not be suprising as many people could not afford a printer back in those days.
It's also very possible that the error happened on the publisher's end and 40 years is a very long time to remember just who's fault i
Re: That was typical (Score:2)
He should've done that, and submitted the correction to the publisher.
But the question is would the revision even have make it to the printing press? Another poster stated that these kinds of books were often laden with errors to the point half the programs didn't work as typed.
If I were him and I knew this, I likely would not have even bothered submitting the corrections. Just not worth the time and money to do so.
Re: (Score:2)
Re: That was typical (Score:2)
"Ironically enough, a "learn to code" book has but one job at the end of the day; to teach you how to code"
I get the impression that these books were just type in shovelware with a bit of 'learn to code' as being mostly a side effect.
A proper "learn to code" book would be something used in schools, that have been scoured up and down for errors, and thus be much more expensive (text book prices). And of course, the code samples would also likely be rather boring practical stuff ment only to teach people to c
Re: (Score:2)
This was my experience as well.
Re: (Score:2)
You misunderstand-- these were not coding books. These were books (or magazines, more commonly) that contained code, with no explanation. 80-Micro had at least two or three programs in every issue that you could type in,
Some of us, who learned coding in parallel with these books, could find the typo and fix it, but on the other hand, when you were looking at a bunch of machine code put into memory via poke/peek statements, your chances of finding, let alone fixing the typo, was pretty slim.
I'm willing to
Re: (Score:3, Interesting)
There were a lot of issues in delivering code via print back in those days. It wasn't like today where you could easily just share digital source files, every system used their own storage formats, and there wasn't a (functional) "net" to transfer files over. Program source files were often re-typed into a word processor by the publishers to get them into their system of choice.
Even if you were to digitally transfer the file to the publisher (via mode
Re: (Score:2)
At least you had to understand the algorithm's theory to get it to run, otherwise the examples were unusable other than to grasp how to structure the code at a high level.
Re: (Score:2)
I've still got a couple of those books, wonder if they're worth anything?
Re: That was typical (Score:2)
Re: (Score:2)
I learned from David A. Lien, who proof-read his work. ;)
Typos were part of the fun (Score:3)
I can assure you that about half of them had typos and just spat out an error message.
I remember those books and found them to be useful examples to different approaches to solve problems. I think you're being generous when you say "about half" of them had issues - I think the real number was north of 80%. The problems ranged from specific BASIC version/platform requirements, typos, to logic errors and then there was generally a lot of just really bad coding.
The games in books and magazines that you could type in could generally be solved in fifteen minutes or so (it just takes a pad of
Re: (Score:1)
He wrote ( and you quoted him ): "about half of them had typos"
You wrote: "I think you're being generous when you say "about half" of them had issues".
And went on to list 'typo' as an example of an issue.
Re: (Score:2)
If you couldn't be bothered to check the BASIC / computer requirements, that's on you, not the publisher.
Re: (Score:2)
If you couldn't be bothered to check the BASIC / computer requirements, that's on you, not the publisher.
Clearly you weren't around then.
The most popular computers of the early 1980s were the Atari 400, Atari 800, Apple ][, Coleco Adam, Commodore Pet, Commodore 64, home built CPM (usually Z80 processors but some Intel 8080 processor systems), IBM PC, TI 99/4a, Sinclair ZX80 and TRS 80 - they all ran they're own version of BASIC (which didn't fit one standard although for the most part they were pretty close). Publishers, not being very computer savvy (and a lot of the people contributing the programs) reall
Re: (Score:2)
If you couldn't be bothered to check the BASIC / computer requirements, that's on you, not the publisher.
Clearly you weren't around then.
If that's your idea of logic, no wonder you had so much trouble.
Got my TRS-80 Model I, level 2, with 4k of RAM and cassette recorder, in 1979. In 1980, we did the 16k upgrade. A couple years later, added an expansion interface. Never got a floppy drive for it, although the local university had a fully equipped Model I that I was allowed to use.
I learned BASIC from the spiral-bound Radio Shack published "Learn BASIC" by David A. Lien, then got his "BASIC Handbook". Since the TRS-80 was only "meh" at game
Re: (Score:2)
Compute! magazine used to list small programs for some common computers back in the 80s. For example, the June 1983 edition had a program named Astrostorm (Page 74) that had versions for VIC, Atari, TI-99/4A, and Apple. https://archive.org/details/19... [archive.org]
Many of these programs were hard to understand because they relied on Peeks and Pokes to add small assembly language routines into memory to speed up the programs and include things like sounds or music. Since the TI-99/4A did not use Peeks and Pokes, its
Re: (Score:2)
Re: (Score:2)
It couldn't have been that bad. I learned to program using those books. I remember not only having to type in the original program but then having to go back and make specific changes to specific lines to make it work for my specific version of basic. At the beginning, I literally knew nothing and had no parent to help and would just painstakingly follow the instructions. Most of what I learned was trial and error. I had been programming for several years before I suddenly realized that char(7) which I
Re: (Score:2)
Re: (Score:2)
A lot of books and magazines at the time wouldn't typeset programs, but instead would use printouts in an effort to only print working programs. Sometimes, a low-quality printout would create its own typos -- an S or 5 missing a few dots or an E turning into an L or an L that looks like a broken E thanks to a smudge.
I remember the Micro Adventure [amazon.com] books being neatly typeset, but I never ran into any problems with those. I suspect it was because all of the programs were kept fairly short. It probably hel
My first program was a text adventure (Score:2)
Re: (Score:2)
This nostalgic post reminds me of being a teenager with a commodore 64 and writing a simple text adventure in basic as one of my first efforts.
I was a teenager with a Radio Shack Color Computer 2 (followed by a 3). As part of my programming education, I wrote a few simple text adventure games, too. Since all I cared about at the time was learning how the whole structure worked, the adventure games I wrote were very simple and boring. Even I didn't want to play them. But learning how to make them was a blast.
This was all pre-Internet. The most effective ways to acquire learning materials at the time were (in no particular order):
1) Friends with stu
Re: My first program was a text adventure (Score:2)
Typing (Score:5, Funny)
No kidding! just use a real keyboard!
Re:Typing (Score:5, Funny)
He's a devoted masochist. Next year he'll be entering it on a MacBook Wheel.
Re: (Score:3)
Re: (Score:2)
Nice one! I can't make up my mind whether that would be better or worse than my youth hours with the ZX Spectrum keyboard.
Re: (Score:2)
A good lie always has an element of truth to it. Someone is telling tales...
The system they're talking about was just for drawing and animating graphics [mynintendonews.com].
Kirby was drawn with a trackball on the Game Boy and NES. The internal hardware team at HAL came up with a way to connect a trackball to the NES, then use a dev tool that ran on a Twin Famicom disc to draw graphics. It was great because you could draw and animate graphics right on the NES screen with the trackball, clicking on numbers onscreen to go between animation frames. It really had a beneficial effect on HAL’s NES and Game Boy development — in some ways I think we really had a powerhouse system.
No one was writing pretend z80 code with a trackball on a 6502 in 1992. It just didn't happen.
Re: (Score:1)
I swear an iPad is better than a TRS-80 Keyboard
Re: (Score:3)
Re: (Score:1)
Re: (Score:1)
Re: (Score:2)
Actually, those old Hi-Tek keyboards [deskthority.net] were quite nice to type on. Don't get the little copper fingers bent or it'll bounce like crazy, though. Over the decades those white plungers can split in the corners (I had to swap a few in an old ADM-3A terminal), but they're easy to swap with donor parts.
I'd still take an Atari 400 membrane keyboard over an iPad, a little bit of travel is better than typing on glass. At least he got the genuine retro experience, since most cheap computers back then had horrible keyb
Re: (Score:3)
I'm trying, but I can't think of a single TRS-80 model with an awful keyboard. Even the Model I had a 'real' keyboard, as did every model through the 4p. The same is true for the CoCos. Even the portables were great -- the Model 100 and 102 were even notable for the quality of their keyboards.
Now, something like an Aquarius with its rubber chicklet keys or a Sinclair or Atari with it's membrane keyboard ... maybe. Touchscreens are pretty awful for typing.
modern solutions to neolithic problems (Score:1)
I would have stomped a baby panda for that near-magical capability in 1982.
Re: (Score:2)
Re: (Score:1)
"tedious typing sessions on my iPad" (Score:3)
If only there were another way.
Re: (Score:1)
If only there were another way.
If only a convenient input device existed for typing ;)
Or if he is too broke for that, for pete's sake, it's BASIC - just do voice to text and then fix the errors afterward, you'd still be better off!
Re: (Score:2)
Or if he is too broke for that, for pete's sake, it's BASIC - just do voice to text and then fix the errors afterward, you'd still be better off!
There's more than one valid reason we're still banging away on rectangular plastic boxes full of keys with letters and numbers on them in 2021, decades after listening to all the bullshit promises about the magical world of voice input that still sucks.
Industry has given up on that concept so hard people make and sell custom bespoke keyboards for hundreds of dollars now. Keyboards would be extinct by now if "better off" was actually true.
It's not.
Re: (Score:1)
Re: (Score:2)
Re: (Score:2)
OCR on code on newsprint is notoriously unreliable. You'd have to proofread it all anyway.
Re: (Score:1)
It's like the James Bond books where he never drinks just champagne, but only "Dom Pérignon".
Some people feel the need to put the names of expensive things into their writing.
I guess it makes them feel classy.
I think it's pathetic.
Re: (Score:2)
Yeah, switch from BASIC to Brainfuck and then you only have 8 characters to contend with.
Magic Word (Score:1)
A hollow voice says "Plugh"
pretty common and helpful (Score:2, Informative)
As others have stated, these typo's were extremely common, and as a kid, finding and figuring out the issue forced one to start understanding what they were doing. This was the beginning of my eventual IT career.
0x10 or 0x11 ? (Score:1)
From TFS:
which he wrote when he was 16
From TFA:
a text adventure written by the then 17-year-old Harry McCracken
Looks like an off-by-one error somewhere.
Typos and bugs everywhere (Score:4, Interesting)
For me "that book" was a Byte Book titled "Threaded Interpretive Languages", by one RG Loeliger. Basically it was a bunch of stuff about how to make a Forth clone for the Z80. And it was full of typos and bugs. I actually got it working on my TRS-80, and in the process wrote a bunch of notes about the bugs on both sides of a sheet of paper. I also made a version for the Color Computer because the 6809 was perfect for it. Of course by this time I was ready to move on to another project and never did anything with it.
Then I loaned the book (with the sheet of paper inside) to a friend of mine who moved away. (Happy ending: a few years ago I found a used copy thanks to the internet.)
Re: (Score:1)
GNU gradebook (Score:5, Informative)
Early 2017, I updated the GNU gradebook (Ggradebook), to version 0.92.
This was more than 16 years after the previous release.
Among other things, migrated from GTK+ 1.2 to GTK+ 3.18.
The software is available here [norbertdejonge.nl], the ChangeLog here [norbertdejonge.nl].
Unfortunately, there's just 0.91 at GNU's Savannah [gnu.org] and FTP [gnu.org]:
- While I did get [libreplanet.org] access to Savannah, my multiple requests to become project member [gnu.org] were ignored.
- Also, while at first GNU webmasters updated the official project page [gnu.org] to at least link to 0.92 on my website (proof [archive.org]), this was reverted roughly September 2019. The reversion took place right after I had published a letter in support [norbertdejonge.nl] of Stallman.
So, now basically nobody knows there's a new version of the GNU gradebook.
Another project I updated after more than 17 years is my gtkgo [norbertdejonge.nl].
Re: GNU gradebook (Score:1)
Re: (Score:2)
That letter up there is a lot more sleazy than just supporting RMS. Go read it. I'd kick the writer off too.
Tinker Time (Score:2)
Re: (Score:3)
Wow you were lucky.
I didn't get a tape deck for a few years! I had to type out every game I wanted to play!
To be fair, it did teach me how to type at the ripe old age of 9....
Yo Grark
Type it exact as it is in the book. (Score:2)
That was what my father used to say, even when I knew it was wrong. The initial problem was, I think that the old school hot metal type setters didn't know programming and got things wrong.
After a time, the programming books just became a list of ideas for me.
Which TRS-80 (Score:1)
Which TRS-80 exactly? There were a wide variety of Radio Shack computers.
QA? (Score:2)
Re: (Score:2)
He never tested it as it was published in the book. And why would he? What was he going to do if it was wrong?
Re: (Score:1)
Re: (Score:2)
What makes you think he didn't test it before it was published? Are you trying to claim it is impossible for a typo to have crept in during the publishing process? If you admit that the error could have occurred after he submitted it, how was he supposed to correct it? Do you think the book publisher sends out advance copies to every submittor, so they can type their own code back in (making sure not to make any new errors or subconsciously correct an error) to make sure it works? Not gonna happen. Cer
Re: (Score:2)
Are you really this stupid, or just trolling? Here is how it worked.
You develop your game
You test it
You print it out and mail to a publisher
A typesetter at the publisher prepares for publication (page formatting, etc). This involves reentering your program one way or another (linotype, etc) ERRORS CAN HAPPEN HERE
The book is printed
When, exactly, in that process is the developer supposed to make sure the typesetting was done correctly?
Re: (Score:2)
you test it before you publish it lol
Not then. You tested it before sending it to the publisher.
But the publisher had no one that knew code.
Then you were supposed to get a copy of the end result, which of course was too late.
Regular books had a proof-reading step, but magazines often didn't.
A lot of stuff on the telephone BBS systems was about these typos.
This is actually alarming. (Score:1, Troll)
And so (Score:2)
Did this but for a Rogue-like game from a magazine on my "data processing" class computer after school. It didn't work. Ran through it all over and over again over several days, and no luck.
Asked the teacher, who took it to some guru, and it came back. The only changes were some pokes and other gobbledy guk, so good job not even meeting up with what a standard trash 80 needed.
Played it for an hour and got bored.
Alarming discovery 40 years later (Score:2)
Not the hero we deserve... (Score:1)
It is so hard to get old games running that this brings a tear to my eye.
Seriously though, reminds me of my first computer as a kid, Timex Sinclair, where my mother would type in programs from a magazine and I would debug it to get it running - usually typist typos.
Decades later and now I'm a software engineer still fixing other people's typos. (And occasionally adding some of my own)
It's a GOOD thing (Score:1)
My 10-YO son learned to program by fixing the errors in 101 Basic Computer Games. Within a year he was writing assembly code. They grow up so quickly...