Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

If your product is Great, it doesn't need to be Good (2010)

paulbuchheit.blogspot.com

Post reply on HN
Print view View original

81–90 of 117 posts

Re: If your product is Great, it doesn't need to be Good (2010)

#81

Post by alnwlsn »

ben_w wrote: post #76
Angostura wrote: post #52

Earlier quoted context omitted.

You include it but with no mention of it. Allow recoding to be activated with an obscure undocumented button push. Leave a subtle hong somewhere that someone clever can find out. Wait for news of the functionality to go viral and additional products to walk off the shelf bought by people who feel clever.

More likely to go viral in that case, a headline saying "Walkmans sometimes record over your tapes!" When I was about 10, someone lent the school a tape of Holst's The Planets for a school play, one of the other students pressed the record button, and shortly thereafter the teacher played to the class a recording of me shouting "no stop" as I rushed for the stop button having seen what they'd done just a moment too l…

I thought that's what the little breakaway tabs on the bottom of the cassette are for

Though I'll admit, when I used to use cassettes, I never write protected them

Re: If your product is Great, it doesn't need to be Good (2010)

#82

Post by bflesch »

maekoos wrote: post #80
bflesch wrote: post #79

Earlier quoted context omitted.

The underlying problem of distinguishing honest contributors from AI hustlers is quite hard, but worthwhile to solve properly. Maybe there is a way to reduce incentives for AI hustlers to join a certain platform, while attracting honest contributors. But even honest contributors might have a bad day or a new project and suddenly they're out in promotion mode.

Sadly I don’t believe ai slop is a solvable issue - we will never be able to filter out AI slop, because it is indistinguishable from human slop (and AI used right could be better than what that person could produce otherwise, eg fixing spelling and language). I think the real problem is just trash content - content I could produce without AI in almost as quickly. The solution is probably more about determining if th…

Yes, it's a human-level problem. Everybody knows from school the people who did not help with the group work, and the ones who actually did their part and were nice to work with.

However are the "good ones" actually hanging out on social media? Or is it just the mediocre ones who are wasting time? "Learning new things" or something being interesting is also highly dependent on one's current knowledge level and skills.

The remainder of the content is ragebait and "discussing" with others so one can feel better about themselves. It is a really nice comfort zone where you don't have to leave your work chair in order to tackle other TODOs in your life while also not feeling like you wasted your afternoon.

I feel the most true version of HN are the "Show HN" posts where somebody did some actual work and is eager to get feedback and judgement from their honored peers. Unfortunately due to AI this has been drowned by sloppy humans as well.

An upvote by the biggest idiot one knows is worth the same as an upvote by a nobel prize winner.

Re: If your product is Great, it doesn't need to be Good (2010)

#83

Post by projektfu »

stnikolauswagne wrote: post #61
palata wrote: post #59

Earlier quoted context omitted.

Did you read it? Where do you see mention of "a field of unsuccessful products"? It mentions the iPad, the iPod, Gmail as successful products. It mentions "laptops" (but in the description it actually includes all desktop computers, I would say) as unsuccessful products. I wouldn't call desktop computers or even just laptops "unsuccessful products". Would you?

>I wouldn't call desktop computers or even just laptops "unsuccessful products". Would you? With the caveat of "Casually browsing the web" I would, actually. They have been near completely subsumed by Ipads or Phones.

Or other tablets. There's a lot of good ones out there, and they are popular enough.

Handspring and Palm sold millions of devices, and so did HP/Compaq, while Apple couldn't get traction earlier with the Newton. There's no real technical reason why the Newton was destined to fail, just as the other devices ascended and then fizzled out. There are network effects at work, society has to be receptive, and the price has to be acceptable.

Re: If your product is Great, it doesn't need to be Good (2010)

#84

Post by petcat »

nottorp wrote: post #66
sph wrote: post #40

Earlier quoted context omitted.

I have been thinking deeply about this problem. I bought a great, silent fan from Rowenta, with beautiful housing and does what it says on the tin with no fancy accessories. 3 speed + 1 very silent mode for sleep. Hey, it’s a European product, not some Chinese knockoff. At some point during design, one person must have said “you know, why not add a brilliant white light that turns on in silent mode? Wouldn’t that be…

Black... nail... polish... Solves every LED problem. Xerox, why did you think the power led on your multifunction should light up not only my home office, but the room next to it too?

Black electrical tape is even better. Because then it can be trivially peeled off later instead of requiring polish remover and a brush.

Re: If your product is Great, it doesn't need to be Good (2010)

#85

Post by phendrenad2 »

Preach it louder for the people in the back. I get into "debates" (really one-sided shouting matches where I'm the one getting shouted at) any time I defend a product that doesn't match the checklist-like sensibilities of the nerd intelligensia. Ironically, the Unix Philosophy that everyone adores is basically the same concept.

Re: If your product is Great, it doesn't need to be Good (2010)

#86

Post by lwhi »

dijksterhuis wrote: post #63
lwhi wrote: post #42

If the barrier to adding that new feature is removed, what happens? If the cost is reduced — and becomes closer to zero — there's probably more chance the feature will be added .. .. in which case, the product is less likely to be great. -- So perhaps, the key superpower in the age of LLM developed software is the ability to say no.

> So perhaps, the key superpower in the age of LLM developed software is the ability to say no. no, it’s never not been a superpower. sorry, couldn’t resist some wordplay. it has always been one of the most important skills, but it doesn’t advance your career easily. enshittification happens because people don’t say no.

I agree, it's vital .. but more difficult to utilise when the cost of adding is reduced.

Re: If your product is Great, it doesn't need to be Good (2010)

#87

Post by phendrenad2 »

lukan wrote: post #45
sph wrote: post #40

Earlier quoted context omitted.

I have been thinking deeply about this problem. I bought a great, silent fan from Rowenta, with beautiful housing and does what it says on the tin with no fancy accessories. 3 speed + 1 very silent mode for sleep. Hey, it’s a European product, not some Chinese knockoff. At some point during design, one person must have said “you know, why not add a brilliant white light that turns on in silent mode? Wouldn’t that be…

That bright LED's became so cheap to allow putting them everywhere certainly had downsides. There are many devices now where I have to tape over to enjoy a dark sleeping place.

Makes me wonder if e-ink could solve this. A red dot that's only visible in daytime, when you want it to be.

Re: If your product is Great, it doesn't need to be Good (2010)

#88

Post by palata »

projektfu wrote: post #83
stnikolauswagne wrote: post #61

Earlier quoted context omitted.

>I wouldn't call desktop computers or even just laptops "unsuccessful products". Would you? With the caveat of "Casually browsing the web" I would, actually. They have been near completely subsumed by Ipads or Phones.

Or other tablets. There's a lot of good ones out there, and they are popular enough. Handspring and Palm sold millions of devices, and so did HP/Compaq, while Apple couldn't get traction earlier with the Newton. There's no real technical reason why the Newton was destined to fail, just as the other devices ascended and then fizzled out. There are network effects at work, society has to be receptive, and the price has…

> There's no real technical reason why the Newton was destined to fail

Which reinforces the survivor bias, doesn't it? Often the product theories are more "look I succeeded once, so I did it right and you should copy me" and less "I was at the right place at the right time, the technology was there and society was ready for it".

Re: If your product is Great, it doesn't need to be Good (2010)

#89

Post by jeromenerf »

jonplackett wrote: post #3

Read something similar the other day about the original Walkman. The engineers wanted to add recording function, thinking it would help with sales and to only cost a negligible amount to add. Someone cleverer said no, because if you add that feature now people will be confused what it is for. If they don’t want to record audio, they’ll think the product isn’t for them.

Case in point: minidisc :)

Recording was the killer feature for me. I recorded thousands of hours of band rehearsals with their stereo omni mics and the media quality.

Re: If your product is Great, it doesn't need to be Good (2010)

#90

Post by moffkalast »

robocat wrote: post #56
moffkalast wrote: post #47

Earlier quoted context omitted.

There are many problems in life that a brush and some black acrylic brush-on primer can fix. This is one of those problems :)

I fixed a power wart LED today using black electrical tape, with a needle hole. Ideally instead I need some stick-on semitransparent dark-alpha stickers to reduce brightness. Maybe I should use two polarized stickers, and rotate the second until brightness is perfect. Are there non-linear solutions or HEV-sensitive photochromic solutions - so that LED brightness is low in the dark but bright enough in sunny condition…

I sure wish that were a thing, if you happen to figure that out let me know. So far the only real solution I know of is active throttling with a photoresistor and a mosfet based on ambient light.
Post reply on HN