IBM Thinkpads have a cult following in part due to not just a good design out of the gate, but the fact that the original designer refused to bend to pressure to change the design every year. The parts are interchangable to large extent between models spanning what, 3—5 years? The guy was under constant pressure; was told to give consumers something fresh by changing up the design. Luckily wisdom prevailed and he disregarded such reckless advice by responding with the mantra: ”if it ain’t broke, don’t fix it!”
I’m happy to buy Thinkpads over 15 years old, often sold for ~$10 on the street, because if something is broken or breaks it can still be used for parts to fix other models of Thinkpads from roughly the same decade.
Lenovo acquired Thinkpad from IBM and gradually fucked it up around the T410 or T450 models as they gave in to the demand of consumers giving a shit about shaving off every gram of weight possible at the expense of ditching rugged rollcages and ditching features like optical drives. Watch some videos of people trying to simply remove a keyboard from a T450 to see what I mean.
Whirlpool also has a reputation for not radically changing the design of internal components. I called a repair shop over a washer or tumble dryer that was like 15 or 20 years old. They said at that age, if it’s not Whirlpool they won’t even show up because when the parts change every year then spare parts quickly become unavailable (of course before people start needing the spare parts). They said Whirlpool is an exception because the same parts will be used for a decade or more, which then justifies the business of making spare parts for a prolonged time (I imagine as well the aftermarket likely thrives too).
Grain of salt though because I heard Whirlpool doesn’t always put their label on their own products and Whirlpools also end up getting labeled as Sears Kenmore. If Whirlpool rebadges something else as Whirlpool, how could the design have consistency w/other Whirlpool machines? Anyway, it was just an example and possibly flawed based on one repair shop’s opinion.
The problem -- no metrics
This is all just tribal knowledge propagated ad hoc by word of mouth. The masses don’t generally know this shit and probably most of them don’t care. I think Whirlpool and Thinkpad were not even diligent enough to advertise it. Maybe they did not even know in advance they would have design consistency over the years. Perhaps if they advertise: ”uses the same motor as previous 6 models”, they would fear that it would chase away foolish consumers who would regard that as ”old”, unevolved, or non-innovative. Those same stupid consumers who are brainwashed to chase “latest and greatest” are why we face so much unrepairable garbage on the market.
Since no one tracks design stability/consistency over time (not even Consumer Reports or similar orgs), there is no incentive for manufacturers to try to satisfy the unknown & unmeasured demand that no one is looking at.
Yeah, that's been a disappointment to me. There are other channels that test tools like Project Farm, but he primarily tests out of the box performance, which is important, but I'd rather have a tool with 5% less performance but 2x the service life. Big Clive does teardowns of electronics, and I'm sure there are others, too.
I think what one could do, at the very least, is some kind of federated materials/parts database. If a tool uses an ABS casing and unbranded battery cells, you know it's not as good as something made of fiberglass reinforced nylon that has Samsung cells, and it doesn't leave room for gaming the system beyond selectively improving parts that have been called out as low quality, which is effectively as good a response as we could hope for. It doesn't even necessarily require expertise from people who would contribute. I'm imagining something that works like Street Complete for collecting inputs to OpenStreetMap.
Torque Test Channel has become my go to for tool reviews. They do real load testing (especially interesting with batteries these days) with DIY dynos and more things like case hardness and teardown tests. They also seem like decent dudes.
That might do it. Rather than accepting scores, accept "composition" checklists, and then generate a score from that. I think you'd still get astroturfing, but it might cut down on the AI stuff, and if anyone proves some account(s) are providing false information, you have a ban process.