Semantic Diffusion (2006) | Hacker News Reader

Semantic Diffusion (2006)

martinfowler.com

29 pointsandsoitis17 comments

A fun one I recently encountered: I was asked to estimate the number of story points a ticket would take, using "1 story point equals 1 day of work, so if e.g. you need 7 days of work, that would be 7 story points".

Not sure what the benefit of the jargon is here.

"Story" framing allow you select between fiction, non-fiction, drama, horror, comedy, fantasy, etc.

Just as an example: I think I've even read this opinion piece before but with everything going on with AI in this moment my first thought reading the headline was:

"Ah Interesting, I'm wondering how learned tokenized semantic meaning and diffusion models fit together."

Let's call that "semantic diffusion", too, just to make "semantic diffusion" a little self-referring.

I feel like at least 90% of people heavily contributing to semantic diffusion are total Martin Fowler zealots.

Does not really necessarily speak so much about Martin Fowler himself, he seems like a pretty decent and smart guy, but it's the case nonetheless.

"Enshittification" caught on, but also sounds like it just means "turned to shit" (rather than: companies run at a loss to capture the market with a good product, then once they do, turn the product to shit to extract as much money as possible, since customers can no longer leave).

Surprise surprise, people are using it to mean the former.

Hmm, not sure I agree it ever only meant "running at a loss to capture the market" - even after having gone back and re-read Doctorow's early postings. It's certainly a common way for the situation to start (getting a large mass of users to abuse is a difficult task), but it's never specified as the only way. I could be very wrong of course, but if so I must have missed some very explicit language in the early coinage.

The original post https://pluralistic.net/2022/11/28/enshittification/ about it makes no mentions of running at a loss, just what happens when a two-sided platform gains a bunch of users.

Later in "The ‘Enshittification’ of TikTok" https://www.wired.com/story/tiktok-platforms-cory-doctorow/ Doctorow does explicitly mention Amazon operated as a loss leader to gain it's initial users but when they get around to the subject of the article (TikTok) it's explicitly noted it was the recommendation system which grew the early audience in that case:

> Which brings me to TikTok. TikTok is many different things, including “a free Adobe Premiere for teenagers that live on their phones.” But what made it such a success early on was the power of its recommendation system. From the start, TikTok was really, really good at recommending things to its users. Eerily good.

> By making good-faith recommendations of things it thought its users would like, TikTok built a mass audience, larger than many thought possible, given the death grip of its competitors, like YouTube and Instagram. Now that TikTok has the audience, it is consolidating its gains and seeking to lure away the media companies and creators who are still stubbornly attached to YouTube and Insta.

And, in the very same piece, Doctorow had opened up with this generic definition of enshittification which aligns with the original:

> Here is how platforms die: first, they are good to their users; then they abuse their users to make things better for their business customers; finally, they abuse those business customers to claw back all the value for themselves. Then, they die. I call this enshittification, and it is a seemingly inevitable consequence arising from the combination of the ease of changing how a platform allocates value, combined with the nature of a "two-sided market", where a platform sits between buyers and sellers, hold each hostage to the other, raking off an ever-larger share of the value that passes between them.

No that's fair, I was wondering whether to look up a fully accurate definition (ironically), but figured it didn't really matter for the point that people use it to mean what it sounds like, instead of what it was defined like. And I just didn't feel like making the effort

Enshittification doesn't seem to capture the malice involved in this process. Manipulation for profit. Pretending to be virtuous to gain trust. An abuser grooming their victim by slowly ratcheting up the abuse and control.

It's more like emprisonification.

when someone says:

>where our vocabulary is limited and often confusing.

What I read or hear is:

>where my vocabulary is limited.

We all live in a post-Wolfe world, there is a rich mine of unused archaic english, and plenty of unsubsumed latin and greek words ripe for repurpose.

> One of the problems with building a jargon is that terms are vulnerable to losing their meaning

Nonsense. Your proposed "jargon" just didn't catch on. Also, language evolves way faster than most people realize.

Trying to shoehorn static semantics to software development is a losing game, I think.

Having experienced first hand the effect of “semantic diffusion” on the word Agile, and later even sprint and product owner – everything lost its meaning. I think he was spot on.

I think that funnily enough, the "semantic diffusion" itself is an example of how such things happen. I wasn't aware of that label and certainly will not be adopting it to my vocabulary. Sounds like a very complex way of saying something has changed its meaning.

Agile existed in a vacuum as a manifesto, then it existed as a driver for cults/zealots of the ideology, then as a de-facto process, and now it's just watered down to something execs repeat when they want to say "our company isn't like a regular stiff corporation, but something more adaptable than that".

The main point is that the real meaning has not changed, and had no reason to change, it just became diffuse because of how often people would twist it or simply parrot words without really knowing what they are. That’s what makes it “painful to watch”.

Agile should still mean exactly the same thing as it did in the manifesto - people over process, short cycles, adaptability.

You just described the same thing as the post, but the author went exploring the inner workings of why it happens.

There are two schools to it.

On one hand, the original meaning of "agile" hasn't changed, as one could consider that immutable - and on the other hand, words do not have inherent meaning, they have usages.

> Agile should still mean exactly the same thing as it did in the manifesto

I recall Dave Thomas (one of the signers of the manifesto) made the point: He points out "agile" is an adjective, "agile" is not a noun. https://www.youtube.com/watch?v=a-BOSpxYJ9M

One of my niche hobbies is trying to coin new terms - or spotting new terms that I think are useful (like "slop" and "cognitive debt") and amplifying them. Here's my collection of posts that fit that pattern: https://simonwillison.net/tags/definitions/

Something I've learned from this is that semantic diffusion is real, and the definition of a new term isn't what that term was intended to mean - it's generally the first guess people have when they hear it.

"Prompt injection" was meant to mean "SQL injection for prompts" - the defining characteristic was that it was caused by concatenating trusted and untrusted text together.

But people unfamiliar with SQL injection hear "prompt injection" and assume that it means "injecting bad prompts into a model" - something I'd classify as jailbreaking.

When I coined the term "lethal trifecta" I deliberately played into this effect. The great thing about that term is that you can't guess what it means! It's clearly three bad things, but you're gonna have to go look it up to find out what those bad things are.

So far it seems to have resisted semantic diffusion a whole lot better than prompt injection did.

Reply on news.ycombinator.com