docs(twitter): add 2078349730204078527 corpus (NOTimothyLottes 8-byte aligned Forth word)

4-post x86-64 interpreter work: all 0-6 arg syscalls in 32 bytes,
embed interpreter inside words with 3-byte overhead (AD lodsd + FF E0
jmp rax), force lower 32-bit but keep 64-bit, pack interpreted forth
words in aligned 8-bytes. Tangent post 4 from @NOTimothyLottes
self-replying about custom bytecode + on-load decompression.

docs/twitter/2078349730204078527/thread.md

@@ -0,0 +1,37 @@
+---
+title: "I realized at some point I can have most interpreted forth-style words in an ali"
+author: "NOTimothyLottes"
+handle: "@NOTimothyLottes"
+post_url: "https://x.com/NOTimothyLottes/status/2078349730204078527"
+post_id: "2078349730204078527"
+timestamp: "2026-07-18 05:22:44"
+post_count: 4
+reply_count: 1
+repost_count: 0
+like_count: 5
+view_count: 502
+---
+
+# @NOTimothyLottes — I realized at some point I can have most interpreted forth-style words in an ali
+
+## Post 1 (2026-07-18 05:14:06)
+
+All {0-6} argument syscall options in 32-bytes. The plan is to embed the interpreter inside the words with a 3-byte overhead {AD (lodsd), FF E0 (jmp rax)} so levering lots of x86-64 stuff for minimal size
+
+![Media 1](./media/2078347557474849137_1.png)
+
+## Post 2 (2026-07-18 05:17:08) — reply to Post 1
+
+Everything forced in lower 32-bit, but still building around 64-bit support. Since I'm not using x86-64 CALL/RET I can free up the stack ops strictly for the data stack.
+
+![Media 1](./media/2078348323287716101_1.png)
+
+## Post 3 (2026-07-18 05:22:44) — reply to Post 2
+
+I realized at some point I can have most interpreted forth-style words in an aligned 8-bytes. So can make a very small interpreter.
+
+![Media 1](./media/2078349730204078527_1.png)
+
+## Post 4 (2026-07-18 06:01:49) — reply to Post 3
+
+@NOTimothyLottes Even smaller code size possible with custom compression scheme, where you design bytecode not for immediate execution but to be decompressed on load.