HNHacker News
TopNewBestAskShowJobs

abxyz

860 karma · joined March 5, 2025

submissionscomments
abxyz··on Anne Wojcicki Wins Bidding for 23andMe
The principle is that employees matter more than shareholders: wiping out shareholders to save jobs is the greater good even if it enriches a nefarious or incompetent executive. Many retail investors are shocked to learn in their first bankruptcy how low they are in the pecking order when a company goes under.
abxyz··on Musk-Trump dispute includes threats to SpaceX contracts
Musk is popular amongst republicans because Trump has championed him. The pro-Musk Republicans are also pro-Trump republicans and their loyalty to Trump will beat out whatever respect they have for Musk. Musk is not a threat to Trump, because Trump’s entire platform is built on Trump-or-bust. Musk was a useful idiot to Trump. Musk thinking that Trump’s Epstein connection was somehow going to hurt Trump shows just how impotent Musk is. Trump fans couldn’t care one iota about that.
abxyz··on Cord didn't win. What now?
That’s not how software engineering works. Management doesn’t know what a feature will cost to add so they defer to engineers who produce estimates which are used to forecast and plan. Engineers estimate based on their own preferences (whether for cynical self-serving reasons or not). You estimate a feature will be delivered in 2 months but it takes 4, that’s 4 months of pressure you could have avoided by rejecting the feature.

Some features are annoying to work on, some have long term implications for how much and what type of work will need to be done. Some just don’t fit with what the engineers want to work on. I know which features are going to take endless stakeholder meetings and which aren’t. I know which are going to require overtime and being on-call. I know which are going to kill off my pet project.

The most self-serving action a software engineer who doesn’t want to work more than the bare minimum can take is to reject as much as possible as being too difficult or bad or some other excuse.

abxyz··on Cord didn't win. What now?
The website is more helpful: https://web.archive.org/web/20240418130050/https://cord.com/

Think Figma comments as a service. Examples of it in production: https://www.workcanvas.com/

abxyz··on Cord didn't win. What now?
Most employees are unmotivated and mediocre-at-best, across every role and industry because most people are (rightly) just showing up to get paid. Engineers are no different, we are not special. Most engineers do not have the incentive to go out of their way to improve their product when keeping this as-is is easier for them. Regardless of whether the OPs product is good, almost all of the engineer opposition would come from engineers not wanting to deal with integrating some new thing that non-engineers want because it’s more work.

If you think most engineers are evaluating things based on the value it’ll drive for their product, you’ve lived a lucky career (and long may it continue).

abxyz··on Ask HN: Startup getting spammed with PayPal disputes, what should we do?
Guaranteeing transactions would incentivize the provider to block transactions. There are many companies in the space, like sift.com, but they don’t guarantee.
abxyz··on Ask HN: Startup getting spammed with PayPal disputes, what should we do?
Assume that the transactions are coming from humans, it is often cheaper to instruct humans than it is to automate when there’s an expectation that you will try to mitigate the malicious behavior.

Be willing to temporarily suspend your services in order to prevent the malicious behavior. Do the manual work to allow genuine customers to keep using your service, e.g: require manual account approval. You need to treat every one of these chargeback transactions as a risk to your businesses ability to operate, each that you allow to happen increases the risk of permanent damage to your business.

Reach out to your account manager at PayPal, this is not something that should be going via frontline support. You need to be talking to a person who knows and is responsible for your account. If you don’t have one, get one. If you can’t get one, look for anti payment fraud businesses that work with PayPal, they may be able to get a direct line to PayPal on your behalf.

For the future, if you’re dependent upon a service provider you should always have someone you can reach out to directly. If a provider isn’t willing to offer that, find a different provider. Financial services especially are very risk averse and will jettison your account if they get even a whiff of something untoward, whether you tried to prevent it or not. The cost of recovering from that will dwarf the cost of any drastic mitigation you take now. Losing your PayPal account is worse than turning off purchases for a few days.

abxyz··on Launch HN: Nomi (YC X25) – Copilot for Sales
Have you seen how granola.ai solve it?

“Works on all platforms, no meeting bots Granola transcribes your Mac's audio directly, with no meeting bots joining your call”

abxyz··on Big banks explore venturing into crypto world together with joint stablecoin
What exactly are blockchains being used for in the real world?

Specific to this story: most of the world has instant payments without using blockchains or stablecoins, it’s only the U.S. where the banking system somehow hasn’t realised they can do instant payments using traditional banking technology.

Anyone posting that they’re skeptical about the use cases for blockchains on HN 10 years ago will have the same skepticism now. Nothing has changed.

abxyz··on OpenAI to buy AI startup from Jony Ive
it’s not true, contemporaneous accounts disprove it (although that’s not to say Sam Altman is not a snake, Sam Altman is a snake that nobody should trust)
abxyz··on Anatomy of a $70M Auction Flop
Thank you for taking the time to visit here and share insight. Yes, that explains it. The missing pieces for me were 1) no knowledge of "chandelier bids" (I thought bids were bids) 2) the article says "without a minimum price guarantee" which I read to be describing the absence of a reserve. With knowledge of chandelier bids, and an understanding that there was a reserve, it all makes sense :)
abxyz··on Watching AI drive Microsoft employees insane
The author of that comment, an employee of Microsoft, goes on to say:

> It is my opinion that anyone not at least thinking about benefiting from such tools will be left behind.

The read here is: Microsoft is so abuzz with excitement/panic about AI taking all software engineering jobs that Microsoft employees are jumping on board with Microsoft's AI push out of a fear of "being left behind". That's not the confidence inspiring the statement they intended it to be, it's the opposite, it underscores that this isn't the .net team "experimenting to understand the limits of what the tools" but rather the .net team trying to keep their jobs.

abxyz··on Anatomy of a $70M Auction Flop
I read this article yesterday and I am still confused by it.

"...carried a pre-sale estimate of over $70 million..."

"...the artwork came to the market without a minimum price guarantee from the auction house..."

"Oliver Barker, the evening’s auctioneer, began the bidding for the bust at $59 million. But his bids stalled at $64.25 million. Three minutes passed as he hunched low over the rostrum, hunting for bidders, Nosferatu-like, until announcing that the lot was a pass."

Can someone explain how these things relate? If the pre-sale estimate was $70m, there was no minimum price guarantee (a reserve?) and a bid of $64.25 million (90% of the estimate) why didn't it sell?

The article goes on to say:

"“No one who is an informed buyer who is serious in this market — billionaire or not — is going to pay what essentially amounts to a 50 percent premium on something that sold in recent memory,” said Todd Levin, an adviser in New York."

Someone was willing to pay $64.25 million, a 30% premium on the last sale, $70 million is only 40%. Where does 50% come from? Why is $64.25 million bad but $70 million is good.

abxyz··on I'm Peter Roberts, immigration attorney, who does work for YC and startups. AMA
Many countries have quite strict limitations on immigration for work. Many countries do not permit any immigration without visa sponsorship. The U.S. permits sponsored immigration with a quite reasonable bar (e.g: H1B, L-1). Many countries have similar lottery systems, quotas and minimum salary requirements. Given the demand for immigration into the U.S., it's not too surprising to see the limits (and restrictions) be more prominent.
abxyz··on [dead]
The greatest crime of LLMs is killing the em dash — gone too soon.
abxyz··on Databricks acquires Neon
If you're someone who researched the company, enjoyed the interview and accepted an offer, you're probably not going to be in the same group as the people who hate Databricks. Databricks is a 10k people enterprise software company that just raised $10bn and is using their deep pockets to hoover up smaller companies. If that doesn't scare you, you'll be fine. For many of us, the thought of working with or using the product of a company like that strikes fear into our hearts because we have different values to you.

Databricks is the antithesis of Neon. Neon is driven by product, Databricks is driven by sales. Opinions of Databricks in a thread about Neon are going to be on the negative side (but not necessarily representative).

abxyz··on Airbnb is in midlife crisis mode
Security cameras (like Google Nest) are not the concern people have: hidden cameras are. It is possible to rig an entire property without any chance of guests finding the cameras. The scale at which it happens is a question (and varies based on country) but the illegality isn't the issue. The dissemination of videos and photos secretly recorded of guests and their children is the issue. There are people in prison for doing this but the damage is done, videos and photos of their guests are online forever.
abxyz··on Leaving Google
Anyone at Google from 2005 to 2025 is a multi millionaire many times over based on stock value alone. Even a savant at squandering money would struggle to come out of 20 years at Google without unreasonable wealth.
abxyz··on So Much Blood
it's outsourcing by choice rather than necessity. Europe can grow its own blood. The much more permissive laws in the U.S. which cause an oversupply for local usage mean it's cheaper to import it. During periods of high demand that exceed current supplies, countries are typically able to increase donations substantially to cover the shortfall, and long term, could pursue similar laws to the U.S. There is no prospect of blood being a meaningful bargaining chip in a trade war.
abxyz··on A Texan who built an empire of ecstasy
Finally, someone on HN I can relate to. My wife and I love luxury dining experiences, so we eat at McDonalds 364 nights a year and go to the highest rated Michelin 3-star restaurant on the 365th night.
abxyz··on Oxide’s compensation model: how is it going?
I like it a lot and their thoughtfulness about it but it's a little hollow when they're spending investor money. I'd like to see how this model evolves once they're off the vc teat: when there's a bottom line to answer to, does the dynamic shift? Everyone has an on-site chef when the money vc hose is on. Valve's flat structure was exciting because it wasn't 3 vc's in a coat larping as a business, it was an actual profitable business.

Support is typically low paid because it's a lot of effort for little reward, no matter how much you pay someone in support, there's only so much impact they can have on the bottom line. The organization as an organism where every organ is as equally important as the other is a beautiful sentiment but the appendix is getting jettisoned at the first sign of trouble. Support, no matter how valued and important to the organisation it is, is never worth $200k/year on the output of 1 person.

The exception to the rule for sales is the canary in the coal mine: sales measures itself, but every role can (and will) be measured when the pressure is on, there will be competition for budget, and the support team will get squeezed until they're empty while the engineers coast. I would be more convinced that this model could survive outside of the vc bubble if sales had bought in to too. Sales as a competitive sport is cultural, not fundamental.

Anyway, not criticism, just musing, love that they're trying it, even if this doesn't work out, everyone had a few good years, it's worth a shot.

abxyz··on Show HN: Flowcode – Turing-complete visual programming platform
> In the right use case (standalone API/workflow, multiple LLM calls, concurrency etc.) working visually is going to be 10x better than code

I completely agree but I think that's a very hard sell. People who write code are often years deep in their preferred workflows and asking them to reconsider how they build things is a very big ask.

> Even if developers prefer textual code when on their own, when they're working with non-developers (product teams, clients etc) there's a big benefit for being able to share and present how things work.

The problem here is that developers lead implementation. The ideal implementation for a stakeholder is rarely the implementation that is delivered, because developers are taking requirements and translating them into the things they want to build. That's why so many products end up bad: developers have too much power by virtue of code being seen as some sort of magic spell that only a true genius can master. Can you imagine a non-technical person successfully dictating that the solution must be built with visual programming?

Low code / no code is powerful because it turns the relationship between product owner and developer on its head, it puts the product owner back in the driving seat (instead of a passenger begging the developer to go in the desired direction and accepting they'll probably end up somewhere... nearby, hopefully). Hiring the wrong developer(s) can be very bad for a company because of how much a developers choice's influence the product and business.

The proliferation of low code / no code in business often comes from outside engineering, many developers, regrettably, love to write code. Often you'll find that non-technical departments of companies have secretly started using things like Make and Zapier and n8n without telling the engineering department because they feel like they finally have the opportunity to make technology work for them.

Developers that exist in service of product are already using platforms like Make and Zapier and n8n because they understand how much value there is in programming without writing code, but for them visual programming probably isn't that much of a differentiator because they already have a vision in their head of how programming works.

Anyway, your product is great, I am sure you'll succeed regardless of the approach you take :)

abxyz··on Show HN: Flowcode – Turing-complete visual programming platform
Looks very good, I believe this type of product is the future (instead of vibe-coding that produces heaps of junk new code). A few thoughts:

* Flowcode is a clever name but since you are positioning yourself as an alternative to code ("Code is messy and complex") the name is contradictory, just call it "Flow" (or "flows" or "Flower" something)

* You are replacing the legacy of code with something new, embrace new ideas that get out of the "one running process" paradigm, e.g: durable functions. There are a bunch of companies in the space (inngest, DBOS, trigger.dev) who are doing very interesting things, copy/steal/leverage all the great things they're doing -- your product's value is the visual programming

* Most developers today are already writing code so getting out of their code flow to do something visual might not fit well, whereas there are lots of non-developers who would love a product like this (see: the success of Make.com, Zapier etc.) because you're offering the AI integration for creating workflows which gives them confidence to go beyond what Make, Zapier etc. offer

If I were marketing this product, my pitch would be reframing programming as something anyone can do when it doesn't involve code, i.e:: your users can become a developer without writing code, AI-assisted visual programming builds production-ready systems without a single line of code. Code is archaic, (visual) programming is the future. Vibe coding is faster horses, AI-assisted visual programming is flying cars.

abxyz··on Show HN: I reverse engineered top websites to build an animated UI library
I would (and do) pay for this type of UI library. Especially in design, everyone is inspired by everyone else, taking inspiration and ideas from elsewhere is normal... but branding this as "reverse engineered" makes it sound stolen. From the comments, it seems like you're actually just inspired by the work of others, as you've apparently written the code from the ground up, which means you're talented and capable. You should change the branding. Even a name change to something as simple as "Inspired UI" would be a meaningful improvement. I'd like to buy this but I'm not paying for stolen code, and as it's branded now, it seems like stolen code, and I'd have no plausible deniability if it turns out to be stolen code.
abxyz··on Why is OpenAI buying Windsurf?
The issue is one of education, not possibility. There is so much hype around vibe coding that it has penetrated non-technical circles and given non-technical people the confidence to try and make things. The same people could use Zapier or Airtable or Tally or Retool or Bubble or n8n to achieve their goals but they didn't have the confidence to do so, or knowledge of the tooling.
abxyz··on Why is OpenAI buying Windsurf?
I disagree, it's the opposite. Low code / no code is valuable because you're deferring responsibility to a system that is developed and maintained by experts. A task running once a day on Zapier is orders of magnitude better for a business than the same task being built by someone on the marketing team with vibe coding. Low code / no code platforms have a very bright future, because they can leverage LLMs to help people create tasks with ease that are also reliable.

LLM-enabled Zapier or Make or n8n is the future, not everyone churning out Claude-written NextJS app after NextJS app.

abxyz··on Why is OpenAI buying Windsurf?
I am very much in favor of anything that makes software engineering more accessible. I have no principled objection to vibe coding. The problem I have with vibe coding is practical: it's producing more low quality code rather than allowing people to achieve more than they previously could.

Almost everything I've seen achieved with vibe coding so far has been long since achievable with low / no code platforms. There is a great deal of value in the freedom that vibe coding gives (and for that reason, I am in favor of it) but the missing piece of this criticism of the criticism is that vibe coding is not the only way to write these simple scripts and it is the least reliable way.

Vibe coding as the future is an uninspired vision of the future. The future is less code, not more.

abxyz··on OpenAI looked at buying Cursor creator before turning to Windsurf
An apt analogy, given...

"The company must change its mindset and become proactive in its approach to compliance. I have decided this can best be achieved by the imposition of a sanction that will serve as a deterrent to future non-compliant conduct by Star and other pub-owning businesses."

https://www.gov.uk/government/news/heineken-pub-company-fine...

abxyz··on Google Search to redirect its country level TLDs to Google.com
You're making a different point. Of course, there are use cases for subdomains, I'm talking specifically about the transition of maps.google.com to google.com/maps. google.com/maps always made sense but wasn't technically viable when Google Maps launched and that's why they've transitioned to it now. I'm arguing that Google Maps being on a subdomain was an infrastructure choice, not a product choice.
abxyz··on Google Search to redirect its country level TLDs to Google.com
Subdomains are archaic in the context of high availability. 15 years ago, it was impractical to expect a single system (google.com) to reliably handle hundreds of millions of requests per second and so distributing services across subdomains was important because it distributed the traffic.

Today, hundreds of millions of requests per second can be handled by L4 systems like Google Cloud and Cloudflare. Today traffic to a subdomain is almost certainly being routed through the same infrastructure anyway so there is no benefit to using a subdomain. That's why I describe subdomains as archaic in the context of building a highly available system like Google's.

If you're Google in 2010, maps.google.com is a necessity. If you're Google in 2025, maps.google.com is a choice. Subdomains are great for many reasons but are no longer a necessity for high availability.

← PreviousPage 4 of 5Next →