nkohari

14 ms·

121.
by nkohari 3y ago
Styles applied by Tailwind cascade like any other styles.
122.
by nkohari 3y ago
There's a strong contingent of developers who would argue that the global-by-default aspect of CSS was a mistake. Hence nested selectors in Sass/SCSS, CSS modules, and so on. Styles applied by Tailwind classes cascade. The library
123.
by nkohari 3y ago
I hear you, but at the same time there's nothing in the current copyright law which differentiates fair use afforded to a person from fair use afforded to a machine . There's definitely an argument that we should have laws whic
124.
by nkohari 3y ago
LLMs are very good at creating things in the style of existing work, but they aren't that great at re-creating existing work verbatim. For example, here's what happens if you prompt Midjourney to draw "the mona lisa":
125.
by nkohari 3y ago
If I watch a pirated copy of Oppenheimer and then summarize the plot to you, the piracy would be a crime, but the summarization would not.
126.
by nkohari 3y ago
I think it's difficult to argue that people have that right today. As a person, I can read creative works and then synthesize something similar. For example, if I were a graphic artist, you could commission me to create art "in th
127.
by nkohari 3y ago
The developer experience of AWS is so bad it creates a lot of opportunity to provide value there. The same was also true for Salesforce for a long time.
128.
by nkohari 3y ago
So many AI startups are really just paper-thin layers over publicly-available models like GPT. There's value there, but probably not enough to support $100M+ valuations. We've barely scratched the surface of what generative AI can
129.
by nkohari 3y ago
No, you definitely did not read that right.
130.
by nkohari 3y ago
Isn't Slashdot a reference to the current directory in Linux?
131.
by nkohari 3y ago
Sam Altman didn't say LLMs are over. (He's the CEO of OpenAI, so that would be a really strange thing for him to say, wouldn't it?) What he actually said was that we've reached the point where we can't improve model
132.
by nkohari 4y ago
What's the alternative, if those companies need subscription revenue to exist?
133.
by nkohari 4y ago
I feel like this describes every Stephenson book I've read, Snow Crash among them.
134.
by nkohari 4y ago
All cards run through payment networks like Visa and Mastercard, which is why they all have a PAN, CVV, etc. The distinction between credit and debit really just depends on the source of the funds backing the card: either a line of credit,
135.
by nkohari 4y ago
There's nothing more sanctimonious than a person judging someone else's parenting. Personally, I think it's pathetic, but I'll leave you to it.
136.
by nkohari 4y ago
What works for some doesn't work for everyone. Parents should do what works for them and their kids, without judgement.
137.
by nkohari 4y ago
You're going to get a lot of skewed results here, because a lot of parents aren't going to feel comfortable telling the truth: that almost all parents allow a reasonable amount of screen time to their kids, even if they know it&#x
138.
by nkohari 4y ago
That's only true if the company is better-suited to be run by someone else. CEOs are prominent, but they're still just people. A person can make mistakes and still be the best person for the job.
139.
by nkohari 4y ago
Direct debit is available (and is very common) but not all vendors support it. Anecdotally most younger people use automatic methods for recurring payments whenever possible, and write checks for one-off purchases, like paying repairpeople.
140.
by nkohari 4y ago
It's difficult to express to non-parents how good the good parts are. Speaking from personal experience, I never understood why people would want kids until we decided to have our own. This isn't a judgment or to say parents are i
141.
by nkohari 4y ago
> programmable mobile phone with crypto chip and starlink carrier for recurring revenue when/if google and Apple ban Twitter app from their walled gardens This would be a world-class feat of engineering, which would also require tha
142.
by nkohari 4y ago
I'm sorry, but all of your points are demonstrably false.
143.
by nkohari 4y ago
Your first example is essentially how it works on the Stripe docs platform. However, it requires a couple of custom components, including a CodeBlock component that renders code within code fences, and a component which understands how to d
144.
by nkohari 4y ago
(I'm an engineer on the Stripe Docs team!) The primary difference between MDX and Markdoc is that an MDX article is essentially imperative code (like a typical React function containing JSX), whereas a Markdoc article is purely declara
145.
by nkohari 4y ago
Can you say more about why you think MDX is better? For what it's worth, we considered MDX, but chose not to use it. Full explanation here: https://markdoc.dev/docs/faq#why-not-mdx
146.
by nkohari 4y ago
Can you turn these off on mobile? You can on the desktop. It'll still suggest things, but not in an obtrusive manner.
147.
by nkohari 4y ago
The Stripe docs use a library of a few dozen Markdoc tags. While the tags are different, the articles on https://markdoc.io are pretty similar to those in the Stripe docs. For example: https://raw.githubusercontent.co
148.
by nkohari 4y ago
It's all custom; the frontend is React and the backend is Ruby. Our plan is to introduce Markdoc as an intermediary step to improve the authoring experience.
149.
by nkohari 4y ago
Right now we're only using it in our narrative docs ( http://stripe.com/docs ) but we have plans to use it in the reference docs as well.
150.
by nkohari 4y ago
> Okay, but has that actually happened in practice? Before we built Markdoc, our documentation was powered by ERB (embedded Ruby templates). Having content mixed with arbitrary code made it incredibly hard to reason about either. Becau

More ›