161 Matching Annotations
  1. Mar 2025
    1. To date, we’ve found that the best way to drive more predictable revenue is by charging a minimum spend upfront or using a credits-based system.

      why credits? can't credit spend still be spent erratically?

  2. Feb 2025
    1. This is a spatial interface that would give us a new and intuitive way for navigating the Internet and understanding how websites are related each other.

      90s personal websites were kinda like this—listing one's fav websites / friends' websites as digitally proximal locations

    2. Video conferencing software removes all of those spatial elements and therefore makes meetings more complicated.

      well....one could also argue that it actually simplifies all the head-turning one has to do irl

    1. If you compound this over 103 products, you’ve got a MSCHF narrative network effect.

      every object should be infused with the brand/tone/taste of the creator

    2. when an artist abandons individual recognition, they can achieve a higher plane of creative excellence through the increased resources of the group

      is this true? why?

    3. MSCHF succeeds not because it has some super-powerful software or a secret hack at making a better Gantt chart

      this seems to be a common refrain with Every features on a co..."it's not bc of some silver bullet; it's bc there's a strong vision/mission that drives and colors all their decisions"

    1. Ramp is winning by making hundreds of small and correct choices

      just like what Rob said—it's about making as many correct decisions in as short an amount of time as possible

  3. Dec 2024
    1. You can think of it as the anti-Steve Jobs approach: You have to preach the Good Word and let your followers do the rest.

      but andy hertzfeld said everyone just manages themselves bc they all know what to do via the vision

    1. These digital workers are capable of executing tasks to a higher level than 95% of the workforce

      a 95% accurate human feels very different from 95% accurate tool tho

    2. ‍We don't build software. We build digital workers.We don't sell tools. We deliver outcomes.

      this is like a co being like "one of our core values: innovation"

  4. www.notboring.co www.notboring.co
    1. Once there’s enough proof that it’s really working, “I can hire 6 fewer people than the 18 I have earmarked.

      capital/human cost reduction only comes after existing humans are supercharged

  5. Nov 2024
    1. That’s why what we’re selling is organizational transformation.

      which is...a mixed bag to sell. people never want to enact large sweeping changes—that's why so many b2b things take so long to get internal adoption—but do want their orgs to magically become transformed for the better without them having to put in work

    2. there are just not enough people shopping for group chat system

      horizontal products have the double burden of selling the product and educating users about the problem

    3. almost any team which adopts Slack as their central application for communication would be significantly better off than they were before

      how was this measured / what was this based off of?

  6. Oct 2024
    1. This allows you to open up any note you’d like side-by-side, in addition to seeing related notes next to the note you’re currently working on.

      STILL so user-write oriented

    1. Soon, you’ll be able to add other types of content — calendar events, emails, links, documents, videos, photos, anything. That content will be automatically organized without you needing to stress about putting it in the right place. It will be understood, so that, with you in the driver’s seat, you can use Mem to power other apps and products you use.Our flagship feature, Mem Spotlight, is a universally accessible assistant that sits on top of the Mem knowledge graph. It pulls relevant information to the tip of your fingers, lets you save anything for later, and captures new thoughts in the blink of an eye — all within the context of where you are. Mem is not a product that you have to go to, it is not a destination. Mem is a product that goes to you, it is a companion.

      very user-write oriented—people still have to do roughly the same amount of work

    2. if someone had left the team, that historic knowledge was sure to fall through the cracks, only for the wheel to be eventually re-invented

      possible MM use case: ask your teammate who left a question

    1. And, by extension, we may be waiting longer than we expect for AI to take over the consumer space, at least at the scale of something like the smartphone or social media

      AI-native social media doesn't seem that far away tho

    2. including running expected value calculations on agents making mistakes

      how human-in-the-loop a tool is will also be a factor. IC humans will do more QA; higher-level humans will do more orchestration

    3. Or, to put it in rather more dire terms, the initial value in computing wasn’t created by helping Boomers do their job more efficiently, but rather by replacing entire swathes of them completely.

      should generational products always aim for this (i.e. FB and Google approach of doing work for people) instead of augmenting people's abilities?

    4. We needed to worry less about dumbing the software down and more about how more complex things could get done in a way that had far less risk.

      improve the backend tech instead of simplifying the UX

    5. It was kids that loved WordArt and the new graphics in Word and PowerPoint

      key to young people utilizing tech in more interesting and innovative ways, I think, is that they simply have more free time to explore the product. adults are always busy and in a rush

    6. Clippy was to be a replacement for the “Office guru” people consulted when they wanted to do things in Microsoft Office that they knew were possible, but were impossible to discover

      sort of like a tour guide for translating biz reqs -> tech

    7. it’s another thing entirely — a much more difficult thing — to get all of your employees to change the way they work in order to benefit from your investment, and to make Copilot Pages the “new artifact for the AI age”, in line with the spreadsheet in the personal computer age.

      VisiCalc gives people who couldn't do something before a step function change bc now they CAN do it; Copilot is a marginal improvement that relies on there being an existing behavior

    8. both too much for one person, yet not sufficient to hire an army of backroom employees

      sweet spot of bicycle-riding: it's what one augmented person can produce

    9. how low-marginal-cost checking accounts might lead to more business for the bank overall, the volume of which can be supported thanks to said new technology.

      what 10x more ambitious thing does the technological shift enable?

  7. Sep 2024
    1. Where is the line between Excel and the spreadsheet you create on top?

      because the interface is simple/flexible enough and the output is represented in a way that is simple enough for the layperson to understand it

  8. Aug 2024
    1. Aug 2, 2024

      conclusions: either build value on top of existing (reliable) infra, make custom infra yourself (like perplexity) if you have very strong opinions about future applications, or wait until the infra you need is good enough for your application. we need an amazon to show us the applications and abstractions needed

    2. All these pieces require serious build-out

      interesting that the conclusion is "infra environment needs to be 100x better"—so all these agent infra cos are actually laying the underwater internet cables for AI

  9. Jun 2024
    1. Conviction is an internal state that we build, while certainty is the external removal of doubt.

      put another way—conviction becomes certainty over time (if u continue to pursue the thing ur doing and things are going well)

    1. An artifact pulls forward some small part of a future world that currently exists only in your head and lets other people interact with it.

      not surveys? is that bc opinion-gathering is too much about hypothetical future actions and doesn't validate via existing present action?

      I guess the reception to an artifact is inherently a survey—although it constrains the reception to the vehicle of the artifact's form

    2. If you find yourself spending more than a few days in a row reading instead of doing, that’s a good sign to course correct.

      i was doing too much of this and ambient input-ingesting hoping that the spark of genius would come to me while just living life

    1. A low-proof fundraise is more likely to leave you without a clear path to product market fit or your next round of financing.

      it's possibly more dangerous to raise a seed round and fail than to not raise at all

  10. Feb 2024
    1. This woman has contributed absolutely nothing to humanity

      bitches looooove to rag on Kim K and it's literally just misogyny. she's contributed more to culture than 10000 of these guys

    2. But of course, somebody smarter than me was there like 10 years before I was.

      it's not about being smart it's about distributing it in the most interesting way

    3. bite-sized, meaningless content that you hate looking at, but for some reason can’t look away from

      this is why my relationship with social media is fine bc if u hate it so much simply exercise some self control and get off it. everything I've ever consumed on social media has been bc I wanted to

    1. The result is that I rarely if ever visit the homepage of the NYTimes; I use other sources of signal (for example, Twitter or Techmeme links) to send me directly to what I need within those sites.

      curation

    1. AI can take many of the best aspects of human logic and scale them, make them reliable

      I mean chatgpt has historically only given me great interpersonal advice

    1. truncating the whole of it after some limit and posting a Read More button to allow readers to see the rest of the thought

      ok but consider that the breaking up of long blocks of text into digestible bite sized pieces helps retain engagement on the tweet/thread better bc bitches can't focus nowadays

    1. flagged for the sociopath track (in general terms), but slowly have come to the realization that this seems more and more like the clueless track instead

      apm programs be like

    1. One year Jeff announced at an All Hands meeting that someone I knew, Barnaby Dorfman, had won the contest. Jeff said the prize was that he'd buy something off the winner's Amazon wish list, but after pulling Barnaby's wish list up in front of the whole company on the screen

      ugh I live for insider culture like this. nothing else like it

  11. Jan 2024
  12. Dec 2023
    1. It's a strategy that is only viable if you can achieve the size of subscriber base that a Netflix has

      but also loses relevance with that sort of scale, I tell myself

    2. Instagram because of the ease of creating multiple accounts to match one's portfolio of identities, Snapchat for its best in class ease of visual messaging privately to particular recipients.

      this sounds like what young people want/need is the digital ability to codeswitch, but could it also be solved by the ability to be perceived through many diff lenses (i.e. to have the multitudes on display)?

    3. We don't have one messaging app to rule them all in the world, but instead a bunch that have won in particular geographies.

      this is also why YC often funds a "X for South America/Africa/etc" type of co

    4. Conversely, if you come up with one great video but the rest of your work is mediocre, you can't count on continued distribution on TikTok

      incentivizes u to put ur pussy into every video every time

    5. The star is the filter, not the user, and so it didn't really make sense to follow any one person over any other person.

      this is why Can of Soup came and went

    6. While you can outsource Bitcoin mining to a computer, people still mine for social capital on social networks largely through their own blood, sweat, and tears.

      and also, now with AI there is a lower price of blood/sweat/tears to be paid..

    7. The shifting nature of scarcity will always leave a wake of skepticism and disbelief.

      Just like PG talks about how restoring an old car creates value while preserving the existing pie

  13. Nov 2023
  14. Oct 2023
  15. Sep 2023
    1. another stopped by for dinner and didn’t offer to help, even while Jess was juggling care of her infant, doing the laundry, and ordering dinner. “She started eating without me,” Jess recalls.

      ok this and the "moms don't ask; just offer" thing sounds very white. asian people definitely are more acts of service oriented and would offer (or even just DO!)

    2. they can’t seem to access the words to ask her about her experiences. Instead, they reminisce about fun times they had in the past as if neither of their lives were interesting enough now.

      skill issue

    3. wondering who else I could start hanging out with

      a theme seems to be that people immediately look for a replacement that "fills the container" of needs that they have—which seems very id-driven but can one blame them when it might take a decade or more for things to return "back to normal"?

    4. worried about her ability to stay connected to a scene where people go out five nights a week

      maybe the transition is easier for people that stay at home / do a mix of going out and staying in more?

    1. “The discourse will never be on how Asians are affected,” he said. “The discourse is ‘Oh, the perpetrator, think about the mental health of the perpetrator.’ Never our mental health.”

      is this endemic of asians self-victimizing? there must be a grain of truth in it for sure, but idk..

    2. DCC’s Twitter posts are littered with disparaging references to “boba liberals” and “blue-check Asians,” people who Zhang said “sell out other Asians” to seem like they’re helping other marginalized groups

      to the diversity/range of POVs and types of asians in "Asian America"

    3. Before the pandemic, Manhattan’s Chinatown had already undergone rapid change, mutating from a primarily working-class Cantonese and Fujianese area into a decidedly more white and wealthy one. The neighborhood was still filled with Asian residents and businesses, but the arrival of the COVID-19 pandemic was a double whammy: Struggling businesses closed, and residents began to be concerned about their safety.

      so it became more white and wealthy but also people got more scared about public safety?

    4. “As a proud husband of an Asian American woman, I think this discrimination against Asian Americans is a real problem,” said Senator Mitch McConnell.

      oh my god lol

    1. Food makes displacement the point of commonality. And it’s a misleading one — we all experienced it, so we all must face the same challenges.

      people don't want to engage with the tough questions, esp in the context of community building. Food is an easy feel-good way out. but that still doesn't answer why it's food and not clothing, aesthetics, music, etc?

    2. Food, as a medium, feels singularly effective as a means to sand the edges off a homeland, to turn that mythic place into a smooth commodity rather than an unknowable, dissonant land.

      not sure if I instinctively agree with this—is it really that much more effective than, say, clothing?