6 ms·

The Epsilon Method

Eddy_Viscosity2 4y ago
"In there I found Nat Eliason’s fantastic headline: No More “Struggle Porn” which gave me the idea to title my essay No More “Insight Porn”. In a final step, I went through my essay and replaced “pseudo-actionable advice” everywhere with “insight porn” so the new headline would make sense. Note that I didn’t steal Nat’s headline or anything. I got inspired by it and tailored to my own needs. This is what the Epsilon Method is all about, not blatantly copying other people’s work." Feels like the epsilon method is to copy but an infinitesimally small amount short of blatant copying.
- abetusk 4y ago
  That's a kind of ungenerous reading. Another, more generous reading, would be to say, "stand on the shoulders of giants and reach further".
  - Eddy_Viscosity2 4y ago
    In the above example, the author does not reach further. Imagine a comedian taking another comics joke but just find/replaces the 'rabbi' with 'priest' and claims it as his own. I'd call that copying. Not to say small incremental improvements aren't good or worthy of businesses, I'm not. I think those are good and worthy. But the size and nature of the epsilon does matter because if the change isn't significant, then the result is effectively just a copy.
    - thelynchmob1 4y ago
      But you're assuming all of the content is the same, except for one small change. It's not. Only the headlines are the same. Using your analogy, it's like taking another comedian's premise for a joke -- "a priest and a rabbi walk into a bar" or something -- and creating an entirely new routine from it.
      - Eddy_Viscosity2 4y ago
        The headline was the key interesting thing that drew people in, and the interesting bit of that was the '$subject$-Porn' bit, which he copied, blatantly in my opinion. Also, although blatant copying in the writing/art/music is bad, in business it can be fine. If someone in my local area decided to open up a coffee shop or auto repair garage that were basically copies of other like businesses elsewhere, that would be fine; the 'epsilon' being they use a different name and branding. If that's the advice, its fine advice. What rubs me the wrong way is that it seems that the article is advising blatant copying but in giving it a cover to seem like its not. If the advice is to copy other successful businesses, then just say that. This doesn't need to be given a fancy name like 'the epsilon method' as if its some new innovative concept. Copying (with slight differences) is not a new idea. People know about copying already.
        suoduandao2 4y ago
        Yes, but 'the epsilon method' is a different name and branding ;).
        Eddy_Viscosity2 4y ago
        It is that. Perhaps I should just embrace this. I could write an article about my new original idea called 'the epsilon approach' which heavily uses the premise of the epsilon method, but with slightly different presentation.
        thelynchmob1 4y ago
        so meta
    - abetusk 4y ago
      Maybe the fundamental disagreement is what constitutes the epsilon and whether a straight "copy" is actually dishonest/bad/immoral? In your example, say the comedian is testing the joke on a new demographic (christian vs. jewish, say), so the epsilon in this case is not just the substitution of a word in the joke but also testing the joke on a new "market". On the one hand, I think there's a place for giving credit to the original source and even having things fall under (some limited form of) copyright. On the other hand, I also see the need for experimentation and freedom to use ideas, which aren't and shouldn't be copyrighted. Wikipedia wasn't the only internet encyclopedia, it just happened to be the one that "won". Maybe there was a reason why Wikipedia won on merit but random chance was, no doubt, a big part in it. Relying too heavily on the novelty of ideas prevents experimentation.
      - Eddy_Viscosity2 4y ago
        If a new comedian changed the existing 'why did the chicken cross the road?' joke to 'why did the duck cross the road?' and claimed it as a new joke they invented, then that would not be ok, in my opinion. However, that's not really my primary issue with this article, as I've said in a couple other comments here, my issue is that 'the epsilon method' itself is not a new idea or concept. Both 'copying' and 'copying with slight differences' are already well known by everyone everywhere. So much so that there are whole bodies of law that attempt to define when its ok and when it isn't. The article is presenting this as a novel and insightful idea when it is not.
        abetusk 4y ago
        I mean, I agree somewhat but I found it interesting enough. Note that this is also in the context of startup culture and a pushback to the 'zero-to-one' idea, so in that sense it's novelty is highly context dependent.
  - JackFr 4y ago
    Well, typically in proofs one uses epsilon to represent a vanishingly small number (as in the classic proofs of differentiation.) After a couple dozen epsilon-delta proofs it’s a hard association to break.
    - mbg721 4y ago
      More specifically, epsilon is the arbitrarily-small add-on that you observe but don't change directly, so using it here is pretty effective at conveying the idea. (I guess the direct change could be framed as the Delta Method, but people would assume that meant being sent to Atlanta or Minneapolis for a few hours for no good reason.)
- stavros 4y ago
  He copied half a headline, hardly seems worthy of condemnation. It's not like the words "No more X porn" have never been said before in the history of the world.
  - Eddy_Viscosity2 4y ago
    This is true and little bit my point. The concept of copying with slight differences does not need to given a its own name of 'the epsilon method' when the existing idea under the name 'copying with slight differences' is both more direct and is self-explanatory.
- thelynchmob1 4y ago
  It's a headline, not a whole article. All the content itself is original. I don't see any problem with this at all.
- sdwr 4y ago
  Good artists copy, great artists steal!
abetusk 4y ago
FTA: > Here, we simply copy an existing and functioning machine and then focus on modifying one or maybe two of the moving parts. Here "machine" is understood to be "business model". This is in response to "moonshot" (or "unicorn"?) ideas/business models: > That’s the problem with zero to one ideas. You have to be lucky enough to find a functioning combination before you run out of time and money.
gizmo 4y ago
The big downside of making unoriginal products is that you can only gain traction by relentless marketing. And that's how you end up A/B testing twitter thread titles like this guy. But who wants to waste their time writing clickbait insight-porn twitter threads? (Or, excuse me, struggle porn). This is all so depressing.
- abetusk 4y ago
  I think you're being overly critical. This blog post had much more signal than many of the "business advice" posts that I see on HN. Here are a few: > ... Moiz Ali did ... Native Deodorant (acquired ... for $100 million). ... The only thing he did different was to start selling it directly on the internet instead of going through the usual retail channels. > Alex West’s Cyberleads ... ($300k in yearly revenue). ... a carbon copy of Growth List and the only thing ... changed was ... to make it laser-focused on agencies. > ... Shaan Puri’s Milk Road newsletter (100k+ subscribers in ... 6 months). He ... copied ... The Hustle but switched the topic ... to crypto.
  - latchkey 4y ago
    Those quotes are twists on solving specific pain points, which is a generally accepted way to build things. What is interesting to me, is that you didn't copy the last one: > Another great example is Will Cannon’s Uplead ($8M in yearly revenue). His whole business idea was to copy ZoomInfo but make it more affordable. All they did was copy an existing business model and undercut on price, which is generally a race to zero and rarely works.
mcdonje 4y ago
TL:DR: Your safest bet starting a new business is providing an incremental improvement to an existing good or service. >The idea is to copy something that’s clearly working with some small but sufficiently significant changes. >Instead of trying to go from zero to one, we’re going from one to one plus epsilon, where epsilon represents our small change.
- sdwr 4y ago
  This makes a lot of sense. It's very hard to be better than existing products de novo. They have years of evolution baked in, tiny elegant tradeoffs that aren't obvious from the outset. I see this a lot in indie games. Many games that do well copy an existing idea, with a few tweaks/transmutations to make it unique. Theres a sweet spot of similar enough to reuse the template (and not get stuck in the weeds during development), different enough to be exciting. Dota Underlords > SNKRX > Just King comes to mind, where the slot machine/tribe/economy mechanic is kept wholesale. Or Slay the Spire, which has been copied to death in the last few years.

"In there I found Nat Eliason’s fantastic headline: No More “Struggle Porn” which gave me the idea to title my essay No More “Insight Porn”. In a final step, I went through my essay and replaced “pseudo-actionable advice” everywhere with “insight porn” so the new headline would make sense. Note that I didn’t steal Nat’s headline or anything. I got inspired by it and tailored to my own needs. This is what the Epsilon Method is all about, not blatantly copying other people’s work." Feels like the epsilon method is to copy but an infinitesimally small amount short of blatant copying.

abetusk 4y ago
That's a kind of ungenerous reading. Another, more generous reading, would be to say, "stand on the shoulders of giants and reach further".
- Eddy_Viscosity2 4y ago
  In the above example, the author does not reach further. Imagine a comedian taking another comics joke but just find/replaces the 'rabbi' with 'priest' and claims it as his own. I'd call that copying. Not to say small incremental improvements aren't good or worthy of businesses, I'm not. I think those are good and worthy. But the size and nature of the epsilon does matter because if the change isn't significant, then the result is effectively just a copy.
  - thelynchmob1 4y ago
    But you're assuming all of the content is the same, except for one small change. It's not. Only the headlines are the same. Using your analogy, it's like taking another comedian's premise for a joke -- "a priest and a rabbi walk into a bar" or something -- and creating an entirely new routine from it.
    - Eddy_Viscosity2 4y ago
      The headline was the key interesting thing that drew people in, and the interesting bit of that was the '$subject$-Porn' bit, which he copied, blatantly in my opinion. Also, although blatant copying in the writing/art/music is bad, in business it can be fine. If someone in my local area decided to open up a coffee shop or auto repair garage that were basically copies of other like businesses elsewhere, that would be fine; the 'epsilon' being they use a different name and branding. If that's the advice, its fine advice. What rubs me the wrong way is that it seems that the article is advising blatant copying but in giving it a cover to seem like its not. If the advice is to copy other successful businesses, then just say that. This doesn't need to be given a fancy name like 'the epsilon method' as if its some new innovative concept. Copying (with slight differences) is not a new idea. People know about copying already.
      - suoduandao2 4y ago
        Yes, but 'the epsilon method' is a different name and branding ;).
        Eddy_Viscosity2 4y ago
        It is that. Perhaps I should just embrace this. I could write an article about my new original idea called 'the epsilon approach' which heavily uses the premise of the epsilon method, but with slightly different presentation.
        thelynchmob1 4y ago
        so meta
  - abetusk 4y ago
    Maybe the fundamental disagreement is what constitutes the epsilon and whether a straight "copy" is actually dishonest/bad/immoral? In your example, say the comedian is testing the joke on a new demographic (christian vs. jewish, say), so the epsilon in this case is not just the substitution of a word in the joke but also testing the joke on a new "market". On the one hand, I think there's a place for giving credit to the original source and even having things fall under (some limited form of) copyright. On the other hand, I also see the need for experimentation and freedom to use ideas, which aren't and shouldn't be copyrighted. Wikipedia wasn't the only internet encyclopedia, it just happened to be the one that "won". Maybe there was a reason why Wikipedia won on merit but random chance was, no doubt, a big part in it. Relying too heavily on the novelty of ideas prevents experimentation.
    - Eddy_Viscosity2 4y ago
      If a new comedian changed the existing 'why did the chicken cross the road?' joke to 'why did the duck cross the road?' and claimed it as a new joke they invented, then that would not be ok, in my opinion. However, that's not really my primary issue with this article, as I've said in a couple other comments here, my issue is that 'the epsilon method' itself is not a new idea or concept. Both 'copying' and 'copying with slight differences' are already well known by everyone everywhere. So much so that there are whole bodies of law that attempt to define when its ok and when it isn't. The article is presenting this as a novel and insightful idea when it is not.
      - abetusk 4y ago
        I mean, I agree somewhat but I found it interesting enough. Note that this is also in the context of startup culture and a pushback to the 'zero-to-one' idea, so in that sense it's novelty is highly context dependent.
- JackFr 4y ago
  Well, typically in proofs one uses epsilon to represent a vanishingly small number (as in the classic proofs of differentiation.) After a couple dozen epsilon-delta proofs it’s a hard association to break.
  - mbg721 4y ago
    More specifically, epsilon is the arbitrarily-small add-on that you observe but don't change directly, so using it here is pretty effective at conveying the idea. (I guess the direct change could be framed as the Delta Method, but people would assume that meant being sent to Atlanta or Minneapolis for a few hours for no good reason.)
stavros 4y ago
He copied half a headline, hardly seems worthy of condemnation. It's not like the words "No more X porn" have never been said before in the history of the world.
- Eddy_Viscosity2 4y ago
  This is true and little bit my point. The concept of copying with slight differences does not need to given a its own name of 'the epsilon method' when the existing idea under the name 'copying with slight differences' is both more direct and is self-explanatory.
thelynchmob1 4y ago
It's a headline, not a whole article. All the content itself is original. I don't see any problem with this at all.
sdwr 4y ago
Good artists copy, great artists steal!

abetusk 4y ago

FTA: > Here, we simply copy an existing and functioning machine and then focus on modifying one or maybe two of the moving parts. Here "machine" is understood to be "business model". This is in response to "moonshot" (or "unicorn"?) ideas/business models: > That’s the problem with zero to one ideas. You have to be lucky enough to find a functioning combination before you run out of time and money.

gizmo 4y ago

The big downside of making unoriginal products is that you can only gain traction by relentless marketing. And that's how you end up A/B testing twitter thread titles like this guy. But who wants to waste their time writing clickbait insight-porn twitter threads? (Or, excuse me, struggle porn). This is all so depressing.

abetusk 4y ago
I think you're being overly critical. This blog post had much more signal than many of the "business advice" posts that I see on HN. Here are a few: > ... Moiz Ali did ... Native Deodorant (acquired ... for $100 million). ... The only thing he did different was to start selling it directly on the internet instead of going through the usual retail channels. > Alex West’s Cyberleads ... ($300k in yearly revenue). ... a carbon copy of Growth List and the only thing ... changed was ... to make it laser-focused on agencies. > ... Shaan Puri’s Milk Road newsletter (100k+ subscribers in ... 6 months). He ... copied ... The Hustle but switched the topic ... to crypto.
- latchkey 4y ago
  Those quotes are twists on solving specific pain points, which is a generally accepted way to build things. What is interesting to me, is that you didn't copy the last one: > Another great example is Will Cannon’s Uplead ($8M in yearly revenue). His whole business idea was to copy ZoomInfo but make it more affordable. All they did was copy an existing business model and undercut on price, which is generally a race to zero and rarely works.

mcdonje 4y ago

TL:DR: Your safest bet starting a new business is providing an incremental improvement to an existing good or service. >The idea is to copy something that’s clearly working with some small but sufficiently significant changes. >Instead of trying to go from zero to one, we’re going from one to one plus epsilon, where epsilon represents our small change.

sdwr 4y ago
This makes a lot of sense. It's very hard to be better than existing products de novo. They have years of evolution baked in, tiny elegant tradeoffs that aren't obvious from the outset. I see this a lot in indie games. Many games that do well copy an existing idea, with a few tweaks/transmutations to make it unique. Theres a sweet spot of similar enough to reuse the template (and not get stuck in the weeds during development), different enough to be exciting. Dota Underlords > SNKRX > Just King comes to mind, where the slot machine/tribe/economy mechanic is kept wholesale. Or Slay the Spire, which has been copied to death in the last few years.