Swapping LCD panels on Lenovo's is so easy, plug and play, takes minutes - MINUTES!
What's new
Latest resources
EVT logic board from Duo development; similar to 210/230/250/270 models
- Featured
Reference information on L2 cache modules for Power Macintosh and clones
A branch of 0dan0's custom firmware for theReels 8mm Film Digitizer to mitigate waviness
Use this excel spreadsheet to understand how the values needed for the old Mac Sound Driver works
Manual for the Novy Quik30 accelerator card for the Macintosh SE
command line to convert to PDF 1.2
Classic Max XVID/DIVX conversion via Terminal
An image resizer applescript that works as expected.
Latest profile posts
Releasing tomorrow! https://bsky.app/profile/macsurfos9.bsky.social/post/3mqecwcxirc2b
Getting there!
The sidebar rendered stacked below the content instead of beside it. But the visible symptom was misleading. Through instrumentation I figured out:
- calc(100% - 270px) resolved correctly to 700px for content; sidebar to 250px.
- Both fit on one flex line (700 + 250 = 950 ≤ 970 available) — no wrapping issue.
- Yet content still ballooned to full width (970) with the sidebar below.
The real cause was one level down: the entire page's flex layout was silently failing and falling back to block flow — the log had thousands of FLEXSAFE collect failed → block fallback.
NetSurf's block-fallback path already tolerated this exact failure (fixes168c — zero-height the failed child and keep going), but the main flex path didn't. I made the two main-path abort sites (place_line_items_main and base_and_main_sizes) equally tolerant: when one item's content layout fails, degrade just that item to zero height and keep the flex line intact, instead of collapsing the whole container. That restored flex page-wide - sidebar beside content, node rows as rows.
The sidebar rendered stacked below the content instead of beside it. But the visible symptom was misleading. Through instrumentation I figured out:
- calc(100% - 270px) resolved correctly to 700px for content; sidebar to 250px.
- Both fit on one flex line (700 + 250 = 950 ≤ 970 available) — no wrapping issue.
- Yet content still ballooned to full width (970) with the sidebar below.
The real cause was one level down: the entire page's flex layout was silently failing and falling back to block flow — the log had thousands of FLEXSAFE collect failed → block fallback.
NetSurf's block-fallback path already tolerated this exact failure (fixes168c — zero-height the failed child and keep going), but the main flex path didn't. I made the two main-path abort sites (place_line_items_main and base_and_main_sizes) equally tolerant: when one item's content layout fails, degrade just that item to zero height and keep the flex line intact, instead of collapsing the whole container. That restored flex page-wide - sidebar beside content, node rows as rows.
IT'S ALIVE! AND IT'S HIDEOUS 
Just showing it off so the next screenshot when it's a bit cleaner feels impressive
Just showing it off so the next screenshot when it's a bit cleaner feels impressive
I've gotten a lot of work done with compatibility with loading 68kmla and have now begun debugging macsurf loading tinkerdifferent! The next release focuses on the forums we all love. While both are powered by xenforo, for one reason or another it does not want to load tinkerdifferent at all. I am currently working through the debugging and know I'll find a way.
Also on this release is the switch from duktape to quickjs - something that will be a huge benefit with loading modern sites as development progresses.
Also on this release is the switch from duktape to quickjs - something that will be a huge benefit with loading modern sites as development progresses.