Clark Kampfe

This post is another entry in the long and storied history of people misunderstanding and torturing Richard Gabriel's famous software design essay, The Rise of "Worse is Better".

This particular instance will apply to pocket knives. Yep, pocket knives!

See, I collect pocket knives. It's one of those irrational things I carry from my childhood. My grandpa always had a pocket knife, and I thought he was the coolest person ever, so it kind of stuck, and now I'm one of "those guys".

I have a few dozen at this point. Too many. Enough that my girlfriend laughs at me. I like to carry a different one every day. It's a dumb hobby, but I use them for all kinds of stuff, and I've used a ton of different knives over the years from many manufacturers. But let's get on with it. What does "worse is better" have to do with knives?

Well, knives are kind of like computer systems in certain respects. They're artifacts created by people for a particular use. They're a product of both industrial and bespoke, handmade processes. They require maintenance. They cost money. You get the idea.

Before we go any further, it'll be helpful for me to layout what I consider to be the functional tradeoffs inherent in knives:

Edge retention/Edgeholding

How much of something the knife will cut until the edge goes dull.

Price

How much does it cost.

Toughness

How well does a knife hold up to impact, bending, prying, throwing off a roof, getting pounded into a log, etc.

One illustrative scenario is to imagine you're opening up some packaged goods on a pallet and you hit a metal staple, a nail, etc. Does the knife's edge roll over? Does it break and chip? Is it damaged at all?

Maintainability

How easy or difficult is it to sharpen a knife. What kind of tools do you need to sharpen it. Are they expensive and complicated or simple and cheap?

Stainlessness

Knives are almost always made of steel, and all steel rusts (with the exception of some exotic materials I won't cover here), but at different rates. How well does the knife's steel resist corrosion? Stainlessnes is relevant in the real world, but I won't really discuss it in this essay.

---

With that out of the way, let's recall worse is better. "Worse is better" is Richard Gabriel's description of the differing design philosophies of the Lisp and C/UNIX camps. We don't need to get into the specifics of Lisp vs. C/UNIX but the general idea - which I must emphasize, Gabriel was criticizing! - is that in the real world, designs (and design philosophies) that sacrifice everything for simplicity of implementation often outcompete those that place more value on other particular traits like correctness, completeness, or consistency. It really isn't much more complicated than that.

So how does "worse is better" apply to knives?

The defining feature of a knife is that it cuts. There are other types of cutting tools like axes and saws, but knives are distinguished as being useful mainly to slice, push-cut, and/or perform light chopping. A knife's ability to cut is its feature set. In Gabriel's parlance, a knife's cutting ability is its "completeness".

Where this relates is, the community of "knife enthusiasts" (which is just as disgusting and pathetic as you imagine) goes through fads, and one of the fads over the last however many years has been for knives with greater and greater edge retention, which again is the ability to cut a lot of stuff without dulling.

And you might be saying, "well, isn't that good? I thought that was a knife's reason for existing."

And yes, in a vacuum, it is good!

But what has happened over this period of time is that a lot of knife manufacturers have been listening to the "knife enthusiasts", and increasingly making knives from relatively newer types of steel that maximize edge retention at the expense of price, toughness, and maintainability. Though not a complete list, some knife steels in this genre that have gained prevalence in recent years are M390, 20CV, S90V, and S30V. Knife people often refer to these as "super steels" because they can hold an edge so much longer than older, traditional steels.

These "super steels" are really good at what they're designed to do. I have some knives made from these steels in designs I love and they actually do hold an edge for a good long time, far longer than you might expect. It's a level of cutting performance that would be totally baffling to to somebody in the 1950s. They would be incredulous at how little you would have to sharpen these knives. Materials science and modern metallurgy is amazing!

The problem is, despite being great at cutting and holding an edge, they kinda suck at the rest of being a knife, which is being a thing that gets banged into nails, pries staples out of wooden boards, gets sharpened crappily and hastily on less than ideal tools, gets dropped on concrete, and just generally abused.

I have other knives, made in older, simpler, "crappier" steels like VG10, AUS-8, AUS-10, 420HC, 14C28N, 12C27, SK5, and AEB-L. These steels are simpler chemistries that have been around forever. Many are old enough so as to no longer be protected by intellectual property laws. These steels no longer excite "knife enthusiasts", who mostly consider these steels "worse".

And it's true. They are worse, objectively. Knives made in these steels dull faster than the same design made in the newer "super" steels mentioned above. They just won't cut for as long! Despite that, I find myself gravitating to knives made in these "worse" steels and using them more, because in real world use, they're often actually better to use than super-steel knives.

Blades made in these so-called "worse" steels have a lot going for them:

• They're often tougher, meaning they better resist damage from nails, staples, and concrete. And when they do get damaged, that damage tends to simpler be more repairable. I have unintentionally damaged a number of knives through normal, non-extreme use, and knives made in these "worse" materials tend to damage less and less frequently. I wish I never damaged knives, but the simple fact is these knives get used in the real world and not in a lab in a test rig.

• They require less skill to sharpen, they take less time to sharpen, and you can sharpen them with simpler, cheaper, more readily-available tools like simple bench stones or ceramics rather than diamond hones. As a result, I tend to view sharpening knives made of these steels as less of a chore, so I tend to do it more, so these knives tend to spend more of their lives in a state of decent repair. Furthermore, it's usually easier to get knives made in these steels to a truly scary level of scalpel-like sharpness rather than just "ehh it's sharp enough".

• And finally, they're cheaper! While it's true that the prices of knives in the M390/20CV/S90V/S30V "super steel" family have come down in recent years, they're still more expensive, especially if you're interested in knives made by makers that aren't in China. As of this writing, you can pick up a Mora Eldris in 12C27 for $30, a Spyderco Endura for <$100 on sale, an ESEE Avispa in AUS-8 or SK5 for $35, as just a few examples. So you can buy 2 or 3 to keep a spare in your car or your bag. If you destroy one by dropping it off a roof or losing it on a camping trip, you're out tens of dollars instead of hundreds.

This particular line of thinking is new for me. When I first got into knives, I climbed the ladder (or fell into the trap, depending on your view) of ever more extravagent, "better" steels with ever better edge retention. I paid more for knives, I paid more for fancier sharpening tools, and I spent more sweat and time in sharpening those knives in the fancier steels. I was amazed at how long the edges lasted, but I dreaded sharpening them when they inevitably dulled or when I damaged them.

And over the years, I realized, I don't need the crazy edge retention offered by these steels. Not for what I'm doing with knives, which is your normal every-day stuff like opening bags of wood chips or soil or dogfood, breaking down cardboard so it fits in the recycle bin, making feathersticks for a bonfire, opening Amazon packages, or whatever else. Basically, my cutting needs are incidental and random, not constant. I'm not in a warehouse breaking down cardboard for 6 hours at a stretch.

These "better" steels have a place, but often like more powerful, "better" software frameworks or programming languages, that "better" often comes with a system-level cost. You're almost always giving something up to get it (except when you're not), whether that's more difficulty finding developers, or increased sharpening difficulty, or longer time to market, or lesser edge toughness.

These knives in these "worse" steels don't cut as long, which is ostensibly the point of a knife. They have to be sharpened more often, which is annoying. But that sharpening is easier! They're tougher, so when you actually use them to cut stuff, they tend to get damaged less and be repairable when they do!

So even though they appear to be worse knives, in the end, they're better tools.