|
Content by Tomáš "Major Tom" Slavotínek (created in late 2019; updated 26 Dec 2023 and then again 15 Feb 2025).
Knowledge at Hand
My involvement with this project began many years ago when I first
discovered Louis Ohland's Ardent Tool of Capitalism, shortly after acquiring my
first PS/2 — a Model 70 with
the Type 3 planar. I can't tell you when
exactly this was, but likely in late 2010 or early 2011. Before that, I was
interested in the golden era of Compaq machines, with their modular designs
(Flex, TriFlex) and the EISA bus. But then I discovered Big Blue's even more
fascinating PS/2 line with its
Micro Channel Architecture.
It wasn't until a few years and PS/2s later when I created an offline copy
of the Ardent Tool website to carry with my on the hard drive of my laptop.
Why? Well, back then, the workshop where I kept most of my vintage hardware
lacked a proper internet connection. Without access to documentation and
software it would be hard to work on any machine, but especially a PS/2. No
refdisk? Missing ADF file? No pinout information? You are screwed! So, having
a local copy of the website wasn't just convenient; it was essential.
Louis' index page back in early 2012.
As I continued tinkering with various PS/2 machines, I began making minor
edits to my local copy of the pages. Since the content was stored locally,
making these edits was quick and easy. However, there was one major downside:
without an internet connection, all external links were "off-limits." At that
time, many critical files and documents were still hosted externally.
Over time, I started adding local copies of many of these files and even
mirroring other PS/2-oriented pages. Since everything was on my hard drive, it
was very convenient to merge this additional content directly into my copy of
the Ardent Tool website, with functioning links going both ways.
I continued expanding my now somewhat special copy of Louis' website. I
included files and other content that could vanish at any moment — or
were already gone but could be recovered from
archive.org, old hard drives, or
floppies. As I became more experienced with the hardware and software, I also
began adding my own original content, such as notes, board outlines,
schematics, and measurement results. I even completely reworked some important
pages — like the main index — to navigate the content more
efficiently (the old index always felt like a maze to me). By that point, I was
relying on my local copy almost exclusively, even when I had access to the
internet.
My modified index page in 2016.
The illustration is based on a drawing from IBM's brochure.
Chasing a Moving Target
This was when I started thinking about giving back to the community. At a
time when Louis' version of the website wasn't seeing much activity, I realized
it could really benefit from the improvements I'd made. I also felt a bit
guilty for merely consuming the content published on the site and in the
newsgroup — just lurking without ever sharing my own findings. (Sandy
might have called me a "ROM" — a Read Only Member.)
I began toying with the idea of either making my copy available to everyone
or sending the updated files to Louis so he could merge the changes into his
version. The latter was preferable, as I wanted to avoid creating yet another
mirror of the website. Once the internet connection at my workshop was sorted
out, there was no longer any need to keep everything locally on my laptop.
However, I encountered an unexpected roadblock. Although I knew that Louis
had made some changes to his website since I forked it — and I had
already merged some of the important additions into my copy — I hadn't
realized just how many changes there were, especially from the later period
when Louis became more active again. My initial thought was that either Louis
or I could simply diff the entire directory tree and merge the changes. So I
created a fresh snapshot of the online version and attempted the merge. While
the directory structure was largely the same, a quick check of some conflicted
HTML files revealed that even a tiny change on the page had drastically altered
the underlying HTML. I already knew that a WYSIWYG editor was used to create
and edit the pages, but I didn't know how "unstable" its output was. I'm not
going to dive into this topic here, but if you are interested, you can read my
blurb about WYSIWYG editors. I soon realized that a
classic merge just wasn't going to work.
I had three options: give up, write specialized software to automate the
process, or take a semi-automatic approach using diff tools, simple scripts,
and some regex magic. Sending the files over to Louis and expecting him to do
the merge manually wasn't realistic — it would have taken months. As a
programmer, I figured I could streamline the process to take only a week or two
without building a dedicated tool. So I decided on the "semi-auto" method.
It didn't take long before I had the process well streamlined. As a bonus, I
was able to review most of the changes and make additional adjustments as
needed. It turned out not to be as daunting as I had initially thought, and the
merging was completed relatively quickly. Hurray — are we done now?
The same index page one year later (2017).
The art was animated and randomized...
Not quite. As I manually edited many of the files, I noticed there was still
plenty of room for improvement — bad links, broken formatting, duplicate
pages, and more. Some might call it "OCD," but I just see it as attention to
detail. I won't recount every bit of work that went into it, but suffice it to
say it was a lot — some fun, much tedious, and downright boring. After
roughly two months of intermittent work (yikes!), I was finally done and quite
happy with the result. I was ready to upload everything to my web server
— this was around mid-2017.
Going Public
It's 2019 now and the site still isn't available. So, what happened?! Well,
life happened. The timing was simply bad. I hadn't touched the pages for about
two years. While I spent some time tinkering with PS/2s to quiet down the MAD
symptoms, it was nothing compared to the previous years.
I finally got back into the MCA world in May 2019 and added some new content
to my still-private (sigh) copy of The Tool. (Technically, the website was
online at this point, but with no links pointing to it, it was effectively
invisible.) After checking the newsgroup and comparing my copy with Louis', I
realized that mine had fallen behind. Again. But this time I knew what to do.
As before, the second large merge took a couple of weeks to complete —
and naturally, I couldn't resist throwing in some additional changes of my
own.
But then, finally, the moment arrived to make the website public. And this
time, for real!
The index as it appeared when the website was first published (2019).
I've decided to revert back to the "SynchroStream Engine" illustration.
While my approach to this entire ordeal may not have been ideal, I believe
the outcome of this MAD ride was well worth all the time and effort I put into
it. I hope you, the reader, will agree and find these pages as useful as I
intended.
| 
