Rants of the Impish Sort
Index: The TrackPoint is the Optimal Pointing Device, Archive.org: The Good, The Bad, And the Ugly, The Four Horsemen of Pain (HTTPS, Cloudflare, Anubis, Claude)
"Rants" in a style loosely inspired by ToastyTech and VWestlife if I had an unhinged alter ego that complained about everything; but don't take anything written here too seriously. It's not meant to be a definitive thesis or a critical outline for how everyone should operate. In fact, just the opposite. I don't actually expect more people to use the TrackPoint (it's all tongue-in-cheek-- I love them but come on, conventional input always wins), I don't expect archive.org will actually be shut down (I hope not anyway), and I utilise a lot the aspects of the modern web myself (I never liked how the 90's internet was designed or how poorly formatted everything was, only the fact that it had useful content: with modern HTML/CSS I could never go back). And responding to anything on Reddit summons an 'expected' and infamous sort of individual.
Behind the sarcasm and overly outspoken remarks there are serious issues, but that's usually what manifests comedy.
The TrackPoint is the Optimal Pointing Device
I recently came across this article by Richard Devine:
It exemplifies the typical confused tropes claiming the TrackPoint is antiquated, obsolete, nostalgic, horrible (...) or whatever else. I disagree as it maintains characteristics a touchpad or mouse does not replicate. Really, it has nothing to do with age: all of these HIDs are fairly old, in fact, if Richard Devine did his homework, he'd find that the TrackPoint is actually the youngest of the bunch -- not very nostalgic, huh. And before someone says more innovations were required for the touchpad, well guess what? The TrackPoint has gone through many changes throughout its history too; it was not as good as it is now (two-button TrackPoints are dreadful) just as early touchpads were not as good. Sheesh. Who hallucinated the idea that the TrackPoint iteration we have today was the first one introduced?
| Device | Year | Inventor |
|---|---|---|
| Mouse | 1964 | Douglas Engelbart |
| Touchpad / TrackPad | 1980 | Xerox |
| TrackPoint | 1984 | Ted Selker |
So, I'll use this opportunity to organise my thoughts why the TrackPoint is a permanent mainstay in HIDs (human interface devices), not merely 'replaced' by a touchpad running Windows.
A TrackPoint Wizard?
Richard claims to being a TrackPoint wizard; if that's true, one would be privy to its unique advantages and never pivot. Period. A power user of HIDs will immediately recognise its advantages. There is no other HID that replicates what it can do, and any 'wizard' knows this. For me personally, I'm able to use the TrackPoint to such a degree that I can use it for artwork or even gaming on any FPS. The best nib for accuracy is the 'soft rim' variant that allows the most grip for your fingertip (of course finger tip size may vary so it's not necessarily a one size fits all). But if you have larger hands you're just at a disadvantage for most products sadly.
To me, the TrackPoint is infinitely more accurate and versatile than a touchpad hopes to be; even with the deluxe touchpad on my M2 MacBook. So you're not really a power user stressing HIDs to their maximum if a touchpad -- let alone on Windows -- is suitable for you, in my opinion.
On that note, anyone who finds a touchpad usable on Windows over macOS has no idea just how bad Windows is with the general purpose control of touchpads, and has--in my opinion--sacrificed any credibility on the subject.
Offset "B" Label
An interesting complaint that the "B" label becomes offset. I don't look at the keyboard when I type so it doesn't bother me; an aesthetic anomaly is hardly worth raising as a point for retiring it for its functionality. Personally I wish IBM/Lenovo released 'otaku' styled keyboards to remove all of the legends entirely, that would solve this contention. :)
OEM Touchpad Decisions
I'm a bit confused when Richard says touchpads in 2010 were terrible, Synaptics (who manufactures all of the devices anyway -- for Apple as well), did in fact have multi-touch gestures on the PC platform. In fact, 2007's T61 can happy utilise this. The general problem with them was the size and the materials used, but that was a conscious decision from the OEMs. The Synaptics technology was just the same. As for Windows' manipulation of the drivers, those problems go back to just how bad Windows is... and you can ratify this due to the fact if Richard used more operating systems, he'd know OS/2 Warp had more functionality on the TrackPoint before Windows finally had the same features added (due to its architectural quirks and bad default design choices).
"The TrackPad is just better"
Oh dear... the touchpad (TrackPad) is objectively not better. For that I'll list the cardinal reasons what actually makes the TrackPoint an ideal HID-- some of which is the foundation as to why Ted Selker even built it in the first place.
- Your hand does not move off of the keyboard when you use the TrackPoint. This. This is one of the most crucial features of the TrackPoint. When you go to use the touchpad (TrackPad) or even mouse, your hands are coming off of the keyboard. IBM even conducted tests about how much time you lose having to constantly move your hands back and forth. Furthermore, the increased movement places a strain on those who have limited mobility in their hands and need something that can be used with minimal movement. This was the original reason why the TrackPoint was designed. This is also why left-handed mice against a 100% keyboard or a right-handed mouse against an 80% keyboard are better for your desktop: your hand has dramatically less travel distance.
- Pressure scrolling vs. movement based scrolling. Your finger moves very little when you want to engage with scrolling or panning. With a conventional touchpad you have to physically move your finger over its surface... and most touchpads are very clumsy when it comes to scrolling -- especially today because Windows still doesn't manage them properly, only macOS is integrated properly to interpret the gestures well enough. Windows has too many conflicting standards and the underlying architecture means it will never be decent for touchpads.
- Fingertip friction. With a touchpad your fingertip is required to become the point of friction -- this is suboptimal for a HID. Every other device the friction is displaced anywhere else: a mouse has its friction against the teflon pads, a marble mouse the friction is with the marble itself and not your fingertip, and the TrackPoint is simply pressure based. With a touchpad, your fingertip must rub against the surface which is the worst form of input. This is also why a pen/pencil has an advantage over using your raw fingertip; the point of friction is shifted to a more precise location.
- Combined gestures with one hand. You can click and hold with one hand as all three buttons can be clicked simultaneously while the cursor is moving. To typically achieve this on a touchpad you need two hands.
- Compactness. It's more compact to implement a TrackPoint as 90% less surface area is required than what's required for a dedicated touchpad. This is in fact the very reason why the Lenovo TrackPoint Keyboards exist, it's the most compact fully functional keyboard available. I use them all the time because I get the *smallest footprint* while retaining the highest functionality that's possible with both a keyboard and mouse, in one.
Lenovo USB TrackPoint keyboard, literally the most versatile thing, ever.
In regard to Richard's comments that the TrackPoint needs to go away when "something better comes along" (or rather came before), you have to be wilfully ignorant to discard its advantages that a touchpad or mouse, lack. It really has nothing to do with nostalgia (especially as the TrackPoint is the youngest HID of the bunch) but functionality and workflows. You can use your computer more efficiently and faster with a TrackPoint since your hand never comes off the keyboard. And if you absolutely must use a touchpad, the only acceptable use is on macOS, they're still unusable on Windows, yes, even in 2026.
I'll never really understand the baseless TrackPoint hate-- to a point, I was guilty of it too, that is, BEFORE I used it. Swinging back to 2007 when I was a teenager and got my first T60p, I was actually weirded out by the TrackPoint as I had never seen such a thing before, and remember thinking: "why did IBM put this stupid thing on there???". My opinion of it rapidly changed though as I began to gradually use it over the touchpad. I don't recall the exact motivation that lead me to using the TrackPoint (I remember it took me about a month before I reached a ceiling of maximum efficiency), it was probably just curiosity to be honest. Then it lead me to seeing if I could play Bejeweled 2 and Zuma faster with the TrackPoint than the touchpad, which I was eventually able to do. And yes, that would still apply today with a modern ThinkPad touchpad on Windows, they're not good if you're a power user. *shrug*
Archive.org: The Good, The Bad, And the Ugly
Some of my thoughts on archive.org that I can direct people to, whenever they're confused why I have mixed feelings.
The Good
There is a real need for a repository for archiving lost products and knowledge. As it stands, archive.org is the only great general purpose source for this at the moment. And if you include the Wayback Machine, it is indispensable.
None of us live forever, so an "archive.org" is necessary for every future generation.
The Bad
A lot of software that's still under active copyright or even currently supported, often gets uploaded to archive.org. That isn't preservation, it's straight up piracy. It also defeats the whole point of preserving something that doesn't (yet) need preservation.
Archive.org is pursuing a worthless crusade against book publishers for illegally allowing people to read copyright material (while asking for donations... which they could have avoided burning all of that money entirely). They should have never engaged this, in my opinion. And even with the "digital checkout system", it provides a much easier access for anyone to pirate material still actively sold by publishers. It should only be relegated to legitimately rare and abandoned materials. This also helps reduce the "needle in a haystack" problem as per below. Why flood, or allow the service to be burdened with books that can actively be purchased easily?
There is no guarantee that anything on archive.org will actually be preserved or that the service will stand the test of time. There are in fact copies of certain things that only exist on archive.org, but mirrors are important... which leads into... archive.org is too big to mirror. It may be more doable if the wheat from the chaff was separated, but that would require a team of people and wouldn't solve the problem of duplicate/bad/illegal uploads.
The website and account profiles are broken. I spent a whole day trying to change a profile pic on archive.org, and the site just refused to do it, in any browser.
In my opinion: they have a sloven response to security and don't really take it seriously. The September 2024 breach resulted in a lot of personal data being leaked everywhere.
It goes down, a lot. A lot. Oh, and when it does go down they rarely post to their official channels to explain why. If it lasts long enough they'll MAYBE post to Bluesky but none of the other platforms they reference on their outage page.... so why reference them at all if they're inconsistently posting? ONLY reference Bluesky as the source of truth then and put a notice that they WILL NOT always provide updates-- only if they feel like it, sheesh.
The Ugly
The searches. Oh how annoying the searches are as archive.org has a horrible method of parsing metadata. You can spend a lot of time trying to find something that its search engine either doesn't list, or doesn't arrange it to the beginning. I have a handful of techniques I use to work around this. It's like they took a page from Windows 10's infamously broken search :p
The needle in a haystack problem: uploads are not audited for quality and many things are just re-uploaded over and over and over and over and over for the same thing, and some uploaders have worse etiquette than others. Great... now I have to sift through many copies of the same thing to find something unique and rare. I could provide many examples (even to things archived on here that people have duplicated multiple times in droves), but it's just kicking a dead horse at this point.
If the "archive" ever gets shut down, it'll take the Wayback Machine along with it, which is in my opinion, one of the most important services in the world right now. Without it, so much would have been lost. And there continues to be legacy website material that many need to obtain from it. I almost feel that it needs to be a separate legal entity.
False sense of archival security: this one is an interesting phenomenon. When I would advise others that I have not been able to access certain rare software, they would either assume I haven't performed due diligence and already checked archive.org/wayback, or, provide links to something I've already advised is the wrong thing (i.e. a shareware version vs. the full version). Not really a problem with archive.org itself per se, just the social phenomenon it has generated around assumptions.
It's recently come to my attention that Archive.org employs what's called 'darking' or shadowbanning to certain accounts or resources that get posted. This is troublesome for two reasons: 1) it actually obscures illegal content that they should just be straight up removing rather than 'hiding' -- if I was a company that had my IP illegally posted on archive.org, I'd want to be able to find it and place a takedown request-- won't be possible with the shadowbanning. 2) it could actually be the reason why their searches rarely pull up relevant info if half of it is shadowbanned. Even worse there's no guarantee their shadowbanning system is actually functional or targeting what it should be if the rest of the issues with the site are any hint to how badly maintained it is.
Reddit Grabs the Clubs
Some of these opinions I've mirrored on Reddit, and to my surprise everyone just started clubbing me with the notion that we *do* need to pirate everything publicly *as quickly as possible*, because if we don't pirate it RIGHT NOW it will never be archived. I don't know who dreamed up that notion but it's ridiculous. Not all content is even the same... it's quite different preserving a VHS tape vs. computer software, and they seem to be melting everything together. And do you even know what you're trying to archive? If it's something so common it's likely there's dozens of people already doing the same thing, inconsistently. And they will miss the rare important things that should be archived, and that can be archived, legally. The point of a public archive is for things that should be archived legally and are legitimately rare; and those are the exact things that are being missed!
Pirating everything still supported by vendors publicly on archive.org leads to the content being rightly being taken down or shadowbanned, so it accomplishes nothing, and again, the vendor SELLS and SUPPORTS it. Violating license agreements-- many of which are hitting small single developers (it's not just the big corporations), does hurt and places honouring licenses at peril. Society is not going to be successful if everyone steals everything and the concept of a license is no longer honoured. Obviously there'll always be piracy, but a public service that anyone can access rather than a privately listed torrent is not going to be sustainable in the long run, and as I've said so many times, will bring down all of the other services with it.
Intelligent archiving and curating also takes time... these people don't seem to understand that dumping swathes of data won't assist those finding precisely what they need. Hoarding everything that cannot be found is a problem I actively TRY TO AVOID; having all of the data at your fingertips yet being able to find nothing. The loudest trumpets don't appear to be archiving things that need preservation...
Bittersweet
And that is the conclusion. While Archive.org is a great and necessary tool, there's a lot of things that need to be done that will probably never happen, and continue to place it in legal jeopardy. You may have noticed it suffers many of the parallel issues as Wikipedia does-- such is the inevitable result to open community projects without auditing and/or bias. I don't really know of a good way to solve that problem, honestly... since 'people' are the issue.
'My' solution to the problem is to use archive.org/wayback for rebuilding and mirroring content here-- all done legally and honouring the respective license agreements, and then you also have a bespoke front end for the content rather than relying on searching the site from thousands of disparate files. Of course I won't be around forever, and there may not be an infinite line of people to carry the torch for hosting my sites, so a solution like archive.org is always needed... in the end...
The Four Horsemen of Pain (HTTPS, Cloudflare, Anubis, Claude)
The way the internet has been going is too stark to ignore. I'm glad I archived what I needed for myself when I did, because man oh man... a lot of once publicly available documentation or files is gone or impossible to search for.
HTTPS
I can understand the desire for HTTPS and its intended function, but there never really was a great response to address the annoying issue that older browsers will never acknowledge it. And now we're in this ever revolving conundrum that older machines and software just can't reliably function without proxies or relays.
Furthermore, just ask... what has HTTPS ever done for us, really? I personally have received no benefit from it other than all of its documented annoyances.
There are still many moments to this day, where, I just want to download a _basic file_ or program on an old operating system and it becomes way more work than it really needs to be. Sure there are some burgeoning attempts with proxies (assuming that JavaScript and newer encodings aren't needed), but as time goes on that's going to become less feasible as encoding goes beyond what those old operating systems understand as well.
Hilariously some of the purported "pros" to HTTPS are actually quite awful and undermine privacy, like the geolocation API (and believe me I hate the "access your location" popups -- sure you can disable it, but configuring Firefox on every machine I use is starting to become a curated science with the cocktail of nonsense that needs to be adjusted, and of course Mozilla likes to pull the rug from underneath you like Microsoft by suddenly adding things that are enabled automatically which you never wanted). Since HTTP is considered insecure, browsers by design won't hand over that info so you unintentionally remove nags and gain some privacy. Of course this is simply the programmed behaviour of the browsers, sensitive information *could* be handed over HTTP, but since it's not, there are still less annoyances on accessing websites through HTTP.
It's going to be _hilarious_ when Google enforces 47 day certificates down the throats of everyone. I remember the good old days when you had 3 year certificates.
HTTPS is annoying me so much that I created the old.ibmfiles.com mirror for exclusive HTTP vintage access to things I (and others) need.
Cloudflare
An assumed necessary evil as the internet has become a crowded place and various entities and individuals no longer wish to respect decent behaviour (and I won't lie, it's convenient for managing DNS). But let me tell you, the Cloudflare security verifications are getting pretty old now as they're starting to flash up for every site I visit. Pick your poison -- Cloudflare "security" or Anubis. Of course I'm never going to use such things because it is the antithesis of what I'm interested in.
When everyone's eggs are in one basket you end up with these uncomfortable truths:
- Cloudflare can and does go down -- and it's hilarious when the global outages happen.
- Cloudflare can dictate who they want to serve, and at any given moment rescind, ban, or balloon your costs
- For Cloudflare "security" you need JavaScript
Anubis
While the intent of Anubis could be viewed as noble (although I have a lot of technical disagreements with my fellow Canadian who designed it; we would not mesh being placed together in the same room), anyone who wants to defeat it for nefarious purposes can easily do so. It's one of those things that punishes the users and the serious data scrapers lick their lips in anticipation and bypass it without issue with using cheap mining services. No different than traditional CAPTCHAs frustrating the humans but not the bots that perform them with ease, or heavy DRM that punishes those who buy the software when the pirates enjoy their DRM-free downloads as usual.
Another thing about CAPTCHAs is that they're as dubious as the "security questions" in brain dead password recovery systems -- because adding publicly searchable and obtainable information to reset your password is 'totally' secure! It's fascinating how many things in society are sold as functional when they don't actually provide the service(s) claimed. Or better yet, like the shredders they sell for shredding secure documents, yeah.... those long strips can easily be pieced back together. I do own one of those shredders but it's mainly cathartic; the feeling of placing all of the paper propaganda I get from the ISPs in my area straight into the shredder is so satisfying.
Perhaps another criticism for Anubis (and Cloudflare security) is that they get implemented on sites or services where it's not really necessary, breaking access to machines that can't run the JavaScript. I'm REALLY not a fan for the ethos that a website MUST be accessible with JavaScript enabled. It's one thing to reduce features when JavaScript is disabled, but an entirely different matter if the website cannot be accessed until you enable JavaScript. Anubis feeds this paradigm.
Claude
The whole reason that resulted in "Anubis" and kickstarted many of the other pains. Anthropic went to town (like going so hard to town they went straight past flavour town) illegally training their software on the entire internet. The amount of hits I got from Claude (even on defunct URLs) is absurd. You're welcome Anthropic, I guess?
The way the 'courtesy' of the internet and consuming proprietary resources and information was conducted, was always under the assumption licenses and intellectual property rights would be respected. All of the AI companies flipped this concept on its head, and nobody wants to say anything because money talks.
As a result, content is increasingly getting locked behind paywalls, account logins, JavaScript, or worse, closing up shop entirely because Google is encouraging their first party AI products leaving content creators in the dust. The irony being they're biting the hand that feeds their AI. I'm under the impression that as this continues, content creators will then shift on developing closed content "AI training" bundles and selling that to companies who need to train their models on all forms of human-created media (whether it's text, audio, visual, logic.... etc.) Of course Reddit is already cashing in on this, selling the site's content to the highest bidder, which is Google at the moment. I just love how Google and Gemini are trained so heavily on every word in Reddit, even when it's misinformation or worse, disinformation. Or the results just get hallucinated, Google wants to keep calling this site "IBM files.com" (which isn't even a valid root domain for crying out loud) rather than "ibmfiles.com" -- which I use to reinforce the fact it has nothing to do with "IBM" (capitalised) the company. It's really that bad, Google's can't train their rotten website to interpret what a valid root domain is.
Other Annoyances
I'm also noticing a general degradation in the UI design on many websites too. For instance, on Gmail or Reddit, you could instantly type your username and password in the text fields and rapidly use tab or whatever. Now... there's slow transitions and/or tab doesn't go right to the next text box as designed. Everything is just so sloppy, slow, and bloated. Nothing is 'snappy' as it needs to be, and there's an increasing focus being taken away from superior faster keyboard input. As a kid using Windows 98, I never thought computers would get slower and less responsive in the future.... damn. No wonder why the Linux terminal through exclusive keyboard interaction is becoming popular like the MS-DOS days.
This site itself has become sort of an accidental reprieve for me when I just don't want to look at, or deal with all of the internet bloat. While it is modern HTML/CSS (and quite elaborate at that despite its simple appearance), what started as a kind of joke making it "90's like" (I was inspired from Louis Ohland's 9595 Ardent Tool of Capitalism index, and liked the 90's design and extreme verticality) actually has gaining appeal now. And it's nice just having the text and files instantly at your fingertips with no layers of abstraction, all the while having dynamic light/dark modes and appropriate image filtering, best of all worlds :)
There's just no valid reason to add JavaScript to this site, so I never will.
The Walls Closing In
Pretty much the free and publicly accessible internet is over. Search engines no longer work (and Microsoft, The Internet Archive, Google -- all like shadowbanning content), everything is cached in LLMs that must be queried, everything wants JavaScript, and UI design is not done so well. I personally do not like the gaudy designs from the 90's, but damn, even those sometimes manage to be more functional. Yikes. Take Microsoft Azure or Exchange Online, the management consoles suffer from floating text everywhere that's "supposed" to be a tab or a button, but there's no visual distinguishing that it is; and because of my ADHD I sometimes have trouble guiding my eyes around these UIs. Whatever...
One can always just power on a Personal System/2 and tune out the world while listening to OPL2/OPL3 music. How're you gonna do it? You're gonna PS/2 it! Of course if there's actually time to do so: very few I know actually have time for retrocomputing escapades.