What's needed is to turn it all upside down: rather than reviewing new products, review broken products.

Make a site about how things break -- review broken and worn-out products to teach how to identify cheap products (where are the stress points, what manufacturing techniques exist to alleviate those). Then compare those with used products well past their warranty period that haven't broken, and look at why they haven't.

Repairability also comes to mind. Everything breaks eventually -- can't cheat entropy -- but when it does, can you easily repair it? Right-to-repair movement would get in on the action.

It would be so hard to maintain repair guides with how frequently products update these days. Everything from vacuums to laptops to shoes go through frequent enough revisions that guides might be invalidated often.

I went to repair my vacuum last month and discovered most of the guides out there cover an revision which uses a totally different latching mechanism for the door to my intake filter. The doors look identical, but the internal latch works differently. Overall that would have been fine, but I got blocked on the door part because the procedure didn't work and I assumed I was going to break it if I tried something else, haha. Turned out I just needed to push in and slide down.

https://www.ifixit.com/Guide

I’m realizing now that they’ve actually got way more content than I thought. It isn’t review format, but they rate how difficult different components are to repair on most items which is a step towards what I want.

Weird that I thought they only did phones/tablets/computers.

So what is a rating system to do when people have very different ideas on BIFL?

For this existential dilemma, it would be helpful if the top DIMENSIONS of quality were explored. In this case with a knife: tang, bolster, handle material, riveting, metal type, hardness, grind angle, etc are all variables that have either a PREFERENCE scale or a DURABILITY scale. I don't think a simple NoSQL-comment-style database is sufficient to define what elements need to be considered to rank a product as high quality vs. low quality.

You must know about AvE's YouTube channel, yes? He's an engineer who deeply understands metallurgy and machining and industrial design, so he can explore these vectors of quality for power tools:

https://www.youtube.com/channel/UChWv6Pn_zP0rI6lgGt3MyfA

This kind of professional analysis is needed FIRST, to determine what the key metrics are. Then the rating system should grade based on this, rather than 5-stars.

Five star reviews are dead. Long live multiple variables!

> So what is a rating system to do when people have very different ideas on BIFL?

Objective vs subjective is something that's hard to overcome without a complex system like you suggest. I agree it's at the heart of the problem I described and I don't know if there's a practical solution, especially for a side project.

There's a famous story that I believe Henry Ford scavenged junk yards for broken Model T's. Every time he found one, he recorded what was broken. After finding enough of them, he figured out the part that was never broken was something like the front axle assembly. So he went to his axle engineers, told them they built it too strong, and find ways to cut costs.

Your overall point, however, is important. I find reviews of outdoors gear (even professional reviews) especially tilted toward light-use buyers, which is frustrating.

https://www.scarpa.com/mountaineering

or rock climbing

https://www.lasportiva.com/en/man/footwear/climbing

The former cost in the ball park of 200 USD to resole, the latter about 30 USD. Unfortunately, I don’t think sneakers, running or approach shoes can be resoled.

My dad, however, always carries a pocket knife (for fishing purposes), and I've seen them get so much wear, the inside of the blade starts getting worn away from being sharpened so often. What's a "lifetime" for someone who sharpens a knife every other week? That's going to wear out even a "BIFL" pocket knife much sooner than "life" would lead you to believe.

There needs to be a way to incorporate the duty cycle of the thing being reviewed, and length of time owned, but then you're just back to the problem of people gaming a review system, and the commercial internet has ruined everything.

But pretty much anything off-the-shelf I would have to see some extraordinary evidence.

BUT, you're still going to need (probably) multiple resoles, patching, and what not.

Take a look at the shoes Prince Charles is wearing - they're bespoke shoes from the 50s.

But you need to ask yourself - at what point do you just scrap them, and buy a new pair?

Some people love the vintage "relic" look - but every resole and patching is going to cost you money. Purchasing a pair of $1000 shoes, and then spending $1000-$2000 on maintenance the next 50 years will make sense for some, but not for others.

I personally, would much rather just buy a new pair of $500-$1000 shoes every 10-15 years, and discard the old ones when they start to look too shabby.

I live my guitars old and reliced, but shoes ? No thanks. When / if my shoes need patching, I'm throwing them.

Dr Martens also had, at some point, sold a "for life" line that came with an extensive guarantee:

> The guarantee covers the failure of any component, such as upper leather, stitched seams, eyelets, soles, welts, linings and reinforcements which has been subjected to normal wear and tear from non-industrial abuse and not unreasonably abused. The guarantee will be active from the day on which the original owner activates their guarantee, to the end of that person's natural life.

I have had several of these that I still wear after 20 years. They have been refurbished.

The problem is I find BIFL communities often can't distinguish between just more expensive, or rando add on features / durability ... from actual utility / quality.

The ratings could then be weighted by that.

1 upside down - https://news.ycombinator.com/item?id=24710832

2 default reviews - https://gist.github.com/iL3D/59df64947d42828d848ebfc1651a312...