HN user

dangoor

3,550 karma

Staff Engineer at GitHub. Founder of CommonJS, TurboGears and Paver. Previously worked as an architect at Khan Academy, Adobe (Brackets), and Mozilla on developer tools.

[ my public key: https://keybase.io/dangoor; my proof: https://keybase.io/dangoor/sigs/S70GBsa2K96S1xO_Z-3r_jHUL6cQ0dFjqQmMlJ6na0o ]

Posts44
Comments965
View on HN
jonmagic.com 14d ago

I stopped watching my agents work and started listening

dangoor
3pts0
github.blog 3mo ago

Manage agent skills with GitHub CLI

dangoor
2pts1
affinity.serif.com 2y ago

The Affinity and Canva Pledge

dangoor
2pts0
blog.replit.com 3y ago

GhostWriter Beta and AI Mode

dangoor
26pts1
electrek.co 4y ago

GM Cruise autonomous taxi pulled over by police in San Francisco without humans

dangoor
32pts11
blog.khanacademy.org 4y ago

Statically Typed Context in Go

dangoor
3pts0
blog.khanacademy.org 4y ago

We count on you: Balancing privacy and analytics

dangoor
1pts0
blog.khanacademy.org 4y ago

Technical choices behind a successful rewrite project

dangoor
2pts0
blog.khanacademy.org 4y ago

Where Go and GraphQL collide: Behind the curtain with genqlient

dangoor
4pts0
blog.khanacademy.org 4y ago

Genqlient: A truly type-safe Go GraphQL client

dangoor
3pts0
subconscious.substack.com 5y ago

Subtext: Markup for Note-Taking

dangoor
2pts0
blog.khanacademy.org 5y ago

Half a Million Lines of Go

dangoor
3pts0
blog.khanacademy.org 5y ago

We switched to Airflow for pipeline orchestration

dangoor
1pts0
www.kevindangoor.com 5y ago

Why blockchains won't revolutionize publishing

dangoor
42pts74
blog.khanacademy.org 5y ago

Overcoming a false sense of security when using Flow types

dangoor
3pts0
blog.khanacademy.org 5y ago

Incremental Rewrites with GraphQL

dangoor
5pts0
engineering.khanacademy.org 6y ago

Khan Academy Successfully Handled 2.5x Traffic in a Week

dangoor
4pts1
engineering.khanacademy.org 6y ago

Go and Services = One Goliath Project

dangoor
5pts0
engineering.khanacademy.org 6y ago

Engineering Principles Can Help You Scale

dangoor
3pts0
arstechnica.com 6y ago

Apple isn't the most cash-rich company in the world anymore

dangoor
1pts0
gregoryszorc.com 7y ago

What I've Learned About Optimizing Python

dangoor
7pts0
khan.github.io 7y ago

Kotlin for Python Developers

dangoor
1pts0
engineering.khanacademy.org 7y ago

Using static analysis in Python, JavaScript and more to make your system safer

dangoor
2pts0
medium.com 8y ago

MobX 5: the saga continues

dangoor
1pts0
engineering.khanacademy.org 8y ago

The Original Serverless Architecture Is Still Here

dangoor
65pts42
engineering.khanacademy.org 8y ago

What do software architects at Khan Academy do?

dangoor
1pts0
engineering.khanacademy.org 8y ago

New data pipeline management platform at Khan Academy

dangoor
1pts0
engineering.khanacademy.org 8y ago

The Great Python Refactor of 2017 and Also 2018

dangoor
4pts0
medium.com 8y ago

The Programming Procedural Story

dangoor
1pts0
praeclarum.org 8y ago

Xamarin Forms on the Web

dangoor
2pts0

Something interesting about your comment is that HN also has a post today (https://tigerbeetle.com/blog/2025-10-25-synadia-and-tigerbee...) about TigerBeetle's support for Zig and their reason for using Zig specifically talked about wanting something for a long time horizon:

Investing in creating a database like TigerBeetle is a long term effort. Databases tend to have a long half life (e.g. Postgres is 30 years old). And so, while Zig being early in 2020 did give me pause, nevertheless Zig’s quality, philosophy and simplicity made sense for a multi-decade horizon.

I wonder why folks who start with roughly the same skills, intelligence and opportunities (and bank account balance!) can nevertheless end up in very different places?

External factors (aka luck), perhaps? Someone gets their resume into a job just after they made the last hire for that position. Or the car they can't afford to fix breaks down on the way to the interview.

This is something I've wondered about. I started out in the green screen era and remember how amazingly quick those UIs were to navigate. I don't see any reason why we couldn't replicate much of that UX and development model, but deliver it to web browsers with graphical capabilities in the parts of the system that need it.

I feel like mouse+keyboard is a step down in speed of use for many tasks, but I do wonder about touch screens. For some things, touch screens can be plenty fast and the UI adapts to the task.

I don't know if publishers manage to get better deals from Amazon, but Amazon's Kindle Direct Publishing has an annoying aspect that the ebook royalty rate falls from 70% to 30% after $9.99, which gets very awkward for a tech book you'd like to sell for, say, $19.99.

Also, piracy is rampant and Amazon doesn't do much about it. Publishers have more resources to stay on top of it, I suspect.

It has become a de facto standard. There are many implementations of MCP. We'll have to see if it fizzles out or if some reasonable stewardship/governance come about, but it is very much possible for something to be a standard even without a standards body.

I didn't see any sort of license in the repository. If your intention is that people can self-host, it would be nice to include some sort of license so that it's clear what the terms of use are for self-hosting.

Neat looking project! Thanks for sharing!

"Who will rid me of this turbulent priest?"[1]

Trump is very good at covering his own language and culpability. What were Trump's actions while the mob was storming the Capitol? How long did he wait to even put forth those tweets? In his speech before they stormed the Capitol, he said[2]

"We fight like hell. And if you don't fight like hell, you're not going to have a country anymore"

but he also said

"I know that everyone here will soon be marching over to the Capitol building to peacefully and patriotically make your voices heard."

Does saying the latter negate the former in the minds of the mob that had been primed for nearly two months, without real evidence, to think the election had been stolen?

Does it matter that that there's evidence, presented in court, that Trump _knew_ he had lost the election and further knew that attempts to overturn the result were illegal? [3]

We all saw _with our own eyes_ what the mob did at the Capitol that day. There were people there with differing motivations and different understandings of what they were trying to accomplish by storming the Capitol. They've received differing levels of punishment as a result. But, I find it hard to not view the totality of the evidence presented to date and say that Trump wasn't trying to stay in power through unlawful means (i.e. "attempt a coup").

[1] https://en.wikipedia.org/wiki/Will_no_one_rid_me_of_this_tur... [2]: https://www.npr.org/2021/02/10/966396848/read-trumps-jan-6-s... [3]: https://www.lawfaremedia.org/article/jack-smith-makes-his-ca...

The only thing I've been able to see so far is that Harris has 67M votes with 81.2% reporting. Assuming the remaining precincts have the same population size and roughly same D/R split, Harris would end up somewhere around Biden's total once the count is complete.

There is a form of curation on HN and "editorial judgment" on HN and that's in the points a post has. A closer approximation of a newspaper would be possible by looking at the points of a post and maybe comparing that to other posts and then sizing headlines appropriately based on how "important" the HN community sees a given story.

Maybe our problem is that the country has just gotten too big to run effectively as a single 350-million person democracy, and it's time for the state governments to step up and stand up for their people.

Or, perhaps, we just need to alter some aspects of that government to run better. For example: changing the size of the House of Representatives to make it more representative. It wasn't supposed to be stuck at 435.

Switching away from first past the post voting would also help enable more diverse voices within the government.

I don't think our allies felt quite so flung about until Trump came along. Sure, administrations might engage a little differently from one another, but fundamentally they could count on the US for a very long time. Presidents did not, before Trump, throw NATO under the bus, for example.

The Cloud Computer 3 years ago

They will also likely charge a significant premium over decoupled vendors

It seems like they're trying to hit a middle ground between cloud vendors and fully decoupled server equipment companies.

Using Oxide is likely cheaper over the life of the hardware than using a cloud vendor. A company who already has in-house expertise on running racks of systems may be less the target market here than people who want to do cloud computing but under their own control.

it's not worth putting in all this effort for someone who isn't willing to even register for free

Keep in mind that most products are not successful and thus, in theory, "not worth putting in all this effort".

The feedback from people saying that they bounced because of registration is a signal that you can choose to address or ignore. If _enough_ people bounce before they can see the value, then ignoring the signal means you're effectively giving up on this becoming viable. Another alternative, though, is to find more ways to demonstrate the value before registration is required to make sure you're not losing people too early in the sales funnel.

I feel like asking people to look at it from your perspective is not a great attitude in product management, which is more about figuring out the prospective customer's perspective and making sure you're offering something they want.