Rubenerd

PC Magazine (or was it PC World?) used to summarise their reviews with Bang for the Buck charts in the 1990s. Components would be plotted against price and performance, and chosen based on where they fell. Save for any glaring or obvious issues, the one that offered the most bang for the most buck was recommended.

I remember my mum telling me I shouldn’t use those terms too loudly in polite company, for reasons I’d know “when I got older”. Whoops!

Two decades on, and everyone from bloggers to YouTubers still follow this for consumer electronics recommendations. It’s rational to want as much performance or efficiency per dollar you can get. On the flip side, you’d be reticent to recommend something with 10% better performance for twice the price. The value, as they’d say, isn’t there.

But this is where clickbait starts to skew the picture. While those magazines of yore did have to lure buyers with catchy covers, videos and bloggers have to fight among thousands of others for engagement and clicks.

And what draws attention? Well, I also brought Shaggy back alongside those magazines, and he tells me it needs to be bombastic, to be truly fantastic. The Shaggy Marketing Principle, they call it. Mr Lover Lover.

This leads to, what could be charitably described, as rage reactions. It’s not sufficient to brand a CPU, graphics card, or cooler as not cost effective, or not suited to a specific use case or budget. They have to be absolutely shocking, or a literal waste of silicon, or shouldn’t exist. These parts aren’t just bad value, their existence is offensive to civilisation, up there with the worst mistakes made in this history of humanity. I think I’ve seen enough thumbnails with literal flames and exaggerated facepalms to last a lifetime. 🔥

Journalistic integrity aside, these attitudes seep into forums and social media, where people who don’t pick the blessed parts are piled onto when sharing builds or ideas. This is most pronounced early on in product launches, when the majority of builders can only base their judgement on what a few YouTube reviewers said. It always sucks seeing a young enthusiast sharing their build, only to be pelted by reply guys.

And it doesn’t even make sense half the time. As mentioned in the title, missing from most of this analysis is the absolute cost of a device, which depending on the market and your budget can be the difference between affording something and not. It doesn’t matter if $100 more gets you twice the performance, if you don’t have $100 more. It also discounts global price variations and supply. To paraphrase Henri Cartier-Bresson about cameras, the best one is the one that’s available.

This drive for engagement warps everything, and it has consequences far beyond trolls saying you’re a sucker for buying an A770 or a 7900 XT. On the plus side, you don’t need permission from other people to feel good about yourself, and you’re not required to rationalise any of your decisions to random Internet strangers. That’s something I wish I’d learned years ago.

If you’re reading this, it means my blog source (mmm, sauce) backup has successfully migrated to Codeberg. I’ll update my source namespace references and other links in the coming days.

Thanks to Screenbeard for the idea. Let me know if you have an account and host some cool stuff that I can follow :)

I can’t remember when Apple began encouraging rounded squares for application icons in Mac OS X, OS X, macOS, or whatever they’re calling it now. Not-iOS OS? But it’s safe to say this experiment in reduced visual accessibility has been a failure in all but the strictest of senses.

I present this screenshot of my Dock as evidence:

Now granted, my preference is generally for open-source software, most of which tend not to have macOS-specific design affordances. But even among the commercial tools I have to run, and even Apple’s own software, the trend is clear. Of those who changed their icons to conform, their designers set their icons on a white background and called it a day. Or a night, or whatever time of day they work.

These icons are the graphical design equivalent of conforming to the letter of the law, but not the spirit. Is the icon a rounded square, your honour? Absolutely! But, I mean, is it really…? Not so much.

And can you blame them? Of course you can, you can blame anyone for anything, regardless of merit. Last week a motorist blamed me (with a horn) for inconveniencing them, given I was crossing the street on a green pedestrian light. The hide of some people!

But a reasonable person probably couldn’t blame them. Some icons are liquid enough to conform to a rectangular container, like MacVim or Zoom’s uninspired word mark. Others required minimal modification, like the Adobe suite which became square years ago during their “elements” rebranding. But it’s clear from viewing any current Applications folder that most aren’t.

There’s likely a simple reason. macOS isn’t sufficiently large or important a platform for most companies, developers, or projects to change their visual identity to conform to Apple’s whims, or to maintain separate icon styles across platforms. Software like the Lagrange Gemini browser have gorgeous icons that use the shape as a picture frame, but they’re the exception. Apple doesn’t even do it consistently, so where’s the motivation for others to?

There’s no shame in admitting mistakes: I’d be in big trouble otherwise. Bring back shapes Apple, you know you want to.

The entire Sydney Trains system went offline last Wednesday, stranding thousands of commuters (like me!) just before the afternoon peak.

Richard Chirgwin reported for iTnews:

The outage halted all trains at stations, because the communications network is critical to communication between drivers, guards, and the rail network’s management centre.

The head of Sydney Trains fronted a press conference to explain:

Transport for NSW believes a failed network switch caused yesterday’s hour-long communications outage, compounded by the system’s failure to automatically switch to a backup network.

“The system has the redundancy to automatically switch across to a backup. That should have occurred immediately … [but] didn’t occur."

Mistakes happen, and automated failovers often aren’t. Disaster recovery strategies are like backups: only tested ones can be considered functional.

Here’s hoping they switch (HAH!) to a better process.

About a dozen of you emailed and posted on social media in response to my question about FreeBSD and NetBSD laptops, thanks!

Overwhelmingly the recommendation was the Framework, which I’ve looked at before but completely forgot. Their website doesn’t make it easy to find the display resolution, but other sites report it as 2256×1504, which is excellent.

The FreeBSD wiki has a page about it, and it’d be cool to do some testing on NetBSD to add to their wiki too.

It probably won’t be in the budget for a while, but I’m keen to try. Ping me if you have any experience with running BSD on this laptop, I’d love to hear about your experience.

More than a few of you have noted an uptick in the number of retrocomputing posts that have appeared here in the last few years, whether it be for 8-bit Commodore machines, or DOS that I grew up with. It wasn’t a conscious decision on my part, but looking back at the archives it’s very obvious!

I’ve long been interested in computer history and nostalgia, but having a bit more disposable income now to buy these machines from the past and my childhood has helped. I’ve also half-joked that it’s a way to still enjoy computers while offering a break from my day job. I write thousands of words a day, so it’s nice to talk about other things, while still scratching that electronic itch.

Studio 8501 on Mastodon expands on this with something I hadn’t considered, but is obvious in retrospect:

I believe that #retrocomputing is about more than nostalgia … it’s also about a reaction to the truly sorry state of modern computing. Old computers didn’t spy on you, they didn’t beg you for micropayments, they just did what they were told. They were tools for work, for learning, for entertainment … we are, as a community, more than smart enough to design and build our own new retro-inspired designs, and build for ourselves a computing world worth living in.

Admittedly, rose-tinted glasses help here. The computer industry was as ruthless and cut throat back then, with dubious business practices and antitrust adventures. But the output from the customer side felt less invasive, gamed, and hostile. The modern web without active filtering plugins is horrible, and I say that as someone who lived through dialup and first-gen popups. And modern smartphones? Don’t get me started.

I also appreciate their optimism. I didn’t connect the dots before, but retrocomputing fans are natural allies to the right to repair and homebrew tech communities. Keeping these systems alive, and expanding upon them with modern enhancements, hints to an alternative future which is more inclusive, empowering, and fun.

I love my tiny Japanese Panasonic Let’s Note RZ6 subnotebook, but the keyboard is just small enough that I can’t type with speed. It’s a perfect “on call” machine, but I’d love something I can write longform stuff on.

The problem is, Panasonic and Apple have spoiled me with their 2× HiDPI displays. I love how images look, the higher fidelity of fonts in documents, and the ability to crank text down to smaller sizes when tailing logs. People saying these are silly luxuries were writing into PC Magazine saying 8-bit colour was sufficient in the 1990s, and that you don’t need double-density floppy disks because in my day we used punch card libraries, dagnabit!

This would be my wish list:

• 144 DPI IPS display, 12 or 13-inch would be ideal
• User-serviceable storage for upgrades and removal
• Solid keyboard
• Lightweight(ish)
• Good FreeBSD and/or NetBSD support, such as Wi-Fi
• RS232? One fewer dongle would make my life glorious

My instinct is to buy a second-hand ThinkPad and be done with it, as is tradition in the BSD community. I’ll have to see if anything fits.

Today’s Music Monday is a bit of a confession. One of my favourite folk singer/songwriters, poets, and hat wearers has a YouTube channel, and I’ve been, as the kids say, binging his videos since last week.

Here’s a song he released last year called Green and Grey.

In the space of a month, our peanut butter, oolong tea, and tartare sauce have advertised “exciting” new rebrands, but promise that they’ll have the same “great taste”!

I can’t speak for you, but I’m not “excited” by food rebrands. I don’t think I’ve ever woken up in the morning and exclaimed “fuck yes, my tea comes in a red box now!” Maybe that says more about my imagination than a jar of condiments. Perhaps I should be excited by a tin of peaches with floral accents. It’s certainly more fiscally prudent (prunes?) than being excited by a new piece of Hi-Fi gear, or a holiday, or a personal hot air balloon fitted with one of those outdoor coffee carts.

The “same great taste” angle is also… odd to emphasise. Someone bursting into a room shouting “I didn’t do it!” should immediately be regarded with suspicion. Like someone trying to hide shrinkflation with new packaging, for example. “Same great taste” also implies past rebrands came with a disgusting or unpalatable new flavour. “New look, and we’ve partnered with Oreo to…!”

This is probably why I’m not in comestible marketing. Maybe next time I change this blog theme, I should reassure people it’ll have the same great posts, this one notwithstanding.

I was trading stories about university with someone recently, and it reminded me of a few occasions where reading and clarifying the spec would have solved everything! You owe it to yourself and your group to make sure everything is clear from the start, lest you go down unproductive and pointless rabbitholes that chew time and sleep.

For a security subject, we had to demonstrate we understood public and symmetric key cryptography. Public key cryptography uses a pair of mathematically-related keys to encrypt and decrypt, which permits only those with the corresponding key to access the plaintext. Symmetric cryptography use one key, and is usually significantly faster.

The spec for this assignment asked us to implement a public key cipher that encrypts a symmetric key to encrypt bulk data, a common use case. I read that to mean implement known ciphers. I spent days building and testing something that could do 2048-bit RSA and 128-bit Twofish in Perl (yes!), the language I’d elected to do the subject in. The tutor was impressed and gave me full marks, but said that a basic proof of concept would have sufficed… something I could have scripted an hour before!

The second was for a multimedia elective I took with Clara and a few friends. It required the development of a “gamification” system with a web portal and rewards for performing certain activities. Being weebs who all met in the anime club, we did something related to that. Clara and a few of the others in the group drew all the cute art, and the rest of us worked on the Django portal. Unfortunately I didn’t get the portal done in time, but the lecturer didn’t care because the assignment was for a prototype and plan for a system, not a completed one. Welp!

We weren’t always so lucky though. One assignment I worked on with my friend Vadim involved us “researching the IT systems of a sports venue”, something the tutor probably thought was a funny thing to give a room full of nerds. We both went to the Ian Thorpe Aquatic Centre in Sydney; Vadim knew some of the people who worked there, and diving was among the few sports I knew anything about after some volunteer IT admin work for a community centre and team in Singapore. We submitted our thicc report to the tutor, including several references to the cute young gentleman from the Free! swimming anime, only to have him deduct marks for it not being a “sports venue”. I think Haruka’s expression (left) from this key visual by Kyoani says it all.

Clearly we made the silly mistake of assuming sport to be… sport. He didn’t consider aquatics sport, something he didn’t budge from even after being asked why it was in the Olympics. I’m sure Rebecca who reads this blog would have words! It was a small enough project that it wasn’t worth appealing, but asking from the start what he meant by “sport” would have nipped this in the bud. Only sports with a ball are sports, dontchaknow?

I don’t regret any work I did at uni. I did learn a great deal, especially from my assignments that I always over-engineered for what was required. But would I have sacrificed that much sleep if the marks would have been the same under some of these circumstances? Maybe not!

Read the spec. If there’s anything ambiguous, ask. It’s as true in business as school.