OSI-tems Vol II

[bxdanny]

Issues 1 and 2 attached to this message
Scanned in greyscale (and there are some pages with grey areas)
Additional issues within Volume II (1980) will be posted as replies to this message.

The left edge of issue 1, page 1 really does have some text cut off. And what looks like a hair on the scanner glass at the bottom of page 2 (PDF page 3) of that issue was not on the glass of my scanner, but was probably on the glass of the copy machine that was originally used to produce the issue.

All pages in these two issues were printed on 14-inch paper. All except pages 1 and 5 of issue 1 (PDF pages 1 and 8) and page 2 of issue 2 (the Table of Contents, PDF page 3) were scanned in two parts so that all content of the pages would be included. All scans that I do come out initially as 8.5 x 11.687 inches. I have cropped 0.687 inches off the bottom or the top of each page scan so they can print actual size on 8.5 x 11 inch paper, but I think the cropped portions are actually still embedded in the PDF.

A few words about items attributed to me: I didn't have a printer when this was first published, so the page layouts are not mine. In particular, it wasn't me who made the comment about "people who are too cheap to buy a TI programmer" calculator (issue 1, bottom of page 8, or PDF page 14) or wrote the accompanying hex-conversion program. I'm sure that was Mike Bassman. And the piece in issue 2 about clearing the screen is simply describing the technique I used in the "sketch pad" program in issue 1. The description may not have been mine.

[dave]

Thanks so much for taking the time to scan and post these OSI-tems newsletters. I've never heard of them before. I hope to add them to the repository if that's OK with you.

[bxdanny]

Certainly, add them. Starting with the next issue (Vol. II #3), they are printed on 8.5 x 11 inch paper, so should be easier to scan and to read. There are five volumes (1979-1983) in all, so it will take me a bit of time to get them all up, scanning a few issues a week.

[bxdanny]

Issues 3, 4, and 5 of Volume II are in the .zip file attached to this post.

Scanned as greyscale (to keep pbirkel happy), except for a few pages in issue #5. Specifically, PDF pages 2, 12, 14, and 21 of that issue were rescanned as black-and-white only, replacing the greyscale versions, because in greyscale they came out just too faint to be fully legible. Even in the black-and-white scans, some lines are difficult to read, but in greyscale it was worse. The printing on the original pages was simply too light in places.

Some comments on my contributions to these issues:

Issue 3:
- In "Invisible Programs", line 2040 needs to be typed in without spaces, otherwise it will be too long for the input buffer. The program as presented really provides no security, because any edit of it, such as typing "0 <return>", will cause the lines to be re-linked. Some actual security could be achieved by adding a REMark as line 0, and putting some important value needed by the program into line 1.
- The "Hex Fifteen Puzzle" program presented here, or rather my recollection of it, inspired me to write the improved version found as Program 3 on my Enhanced Pico-Dos disks (the final version of which can be downloaded at download/file.php?id=2063), but the version published here was not directly used in creating that.

Issue 4:
- In "Beat the String Bug", I actually understated the number of bytes used in constructing strings. The first program example will in fact use 230 bytes of stringspace, not just 210, because each iteration of A$=A$+CHR$(X) will first use one byte to store the CHR$, and then a number of bytes equal to the length of the string being formed, so 20 iterations will use 20 more bytes than I stated in the article.

Issue 5:
- The "Dumb Terminal" program in this issue was written before I had ever seen a 6850 data sheet, so I didn't realize at the time that the 6850 could generate parity bits (odd or even) on its own, without needing them to be calculated by the CPU.

Edit 10/28/2023: The link above to the "final version" of my "Enhanced Pico-Dos" has been updated, since what was meant to be the final version had a serious (but easily fixed) bug in Program 4, and so a newer version was uploaded. The bug related to generating a version of "the Gomoku tape" that is discussed in issue #8 of the newsletters posted in this thread, in the article "A Quick and Dirty Way to Save Machine Language Programs on Tape". Unlike the version sold commercially in 1980, the version generated by Program 4 does not rely on undocumented behavior of the 6850 ACIA. [Edit 2/26/2024: The link has been updated again because of a different, unrelated bug in the disk image that was there.][Edit 4/20/2024: Updated one more time today.]

[Thomas]

Thank you for sharing these documents. My favorite are the newly discovered instructions for the 6502 by Mike Bossman.
Mike must have been some kind of clairvoyant for future technologies.

AGB Add garbage
BAH Branch and hang
BLI Branch and loop infinitiv
BPG Branch on program bug
BPO Branch on power off
CBP Create program bug
CRN Convert to roman numerals
DAO Divide and overflow
ERM Erase ROM
IAD Illogical AND
IOR Illogical OR
MDB Move and drop bits
MWK Multiply work
PAS Print and smear
RBT Read and break tape
RPM Read programmer's mind
RSD Read and scramble data
SVD Scratch vital data
TPR Tear paper
WID Write invalid data
XIO Execute invalid opcode
XPR Execute programmer
TBS Throw away the 6502 and buy sixteen bit chip

In many cases he has shown the ability to predict future operating system features five to fifteen years ahead of their time.
Like my favorite “Branch on power off”. Without it, what would we do when we press the power off button on our computers?
Or “Erase ROM”. like our famous and essential FLASH drives. And the list goes on.

[bxdanny]

Issues 6, 7, and (most of) 8 of Volume II are in the attachment.

Issues 6 and 7 scanned as greyscale.

Issue 8 scanned as black/white only, because the print was fainter and of poorer quality. I am more concerned with the scans being readily legible (as much as possible) than I am with "archival preservation" of page details. The last two pages (or possibly more) of issue 8 are missing entirely. I don't know what happened to them, I don't seem to have them. The "Xerox Mode OSI" (tape copier) program on page 16 seems mostly complete, page 17 probably only had a few lines on it. Mike Cohen's "OS-65D Corner" that the ToC says was on page 18 might have been interesting, but I guess we'll never know what was in it.

[bxdanny]

Here are the last three issues of Volume II. I have designated them as issues 9, 10, and 11, since that is what they actually are, but their covers refer to them as (a second) issue 8, issue 9, and issue 12. (I was the designated editor for December, and I knew the numbering of the last two issues was wrong, so I just assumed then that December should be issue 12, and called it that, but really it should have said issue 11, since there was no January issue.)

Scanned using "Autodetect color mode". So the November issue was scanned in color, since it has a blue cover page. December (and I think October) are in greyscale. That November issue was printed double-sided (except for the cover page and ToC page), while the others were single-sided.

I won't say much about the contents at this time, but I will note (as I did within the December issue) that the page in the October issue labeled as Memory Locations for OS 65U is not what it purports to be, but is rather a list of memory locations for the BASIC-in-ROM.