HNHacker News
TopNewBestAskShowJobs

pclowes

1,037 karma · joined November 30, 2024

submissionscomments
pclowes··on To be a better programmer, write little proofs in your head
I agree on the Dogma aspect. Plenty of times I have abandoned it. However, I did find it very helpful to spend my first couple years in a strict Xtreme Programming (XP) shop. The rigor early on was very beneficial and its a safety net for when I feel out of my depth in an area.

I tend to go the other way, I don't use TDD when I am playing around/exploring as much as when I am more confident in the direction I want to go.

Leaving a failing test at the end of the day as a breadcrumb for me to get started on in the morning has been a favored practice of mine. Really helps get the engine running and back into flow state first thing.

The dopamine loop of Red -> Green -> Refactor also helps break through slumps in otherwise tedious features.

pclowes··on To be a better programmer, write little proofs in your head
The best way I have found to integrate this approach is Test Driven Development.

When done well, every test you write before you see it fail and then you write the barest amount of code that you think will make it pass is a mini-proof. Your test setup and assertions are what cover your pre/post conditions. Base cases are the invariant.

The key here is to be disciplined, write the simplest test you can, see the test fail before writing code, write the smallest amount of code possible to make the test pass. Repeat.

The next level is how cohesive or tightly coupled your tests are. Being able to make changes with minimal test breakage "blast radius" increases my confidence of a design.

pclowes··on Leveraging Elixir's hot code loading capabilities to modularize a monolithic app
What makes you say that? Honestly asking.

I know a team using it to replace ancient massive mainframe based systems with modern distributed systems and the gist is that the language is fine, but mostly ideal for use cases that leverage the ErlangVM or BEAM stack.

The downside they run into is the ecosystem isnt there, at least a couple guys wish they had just used Kotlin/Java for library interoperability with so much existing code already built and battle tested for specific purposes.

pclowes··on Pa. House passes 'click-to-cancel' subscription bills
I wonder how hard it would be to generate synthetic credit card numbers for each subscription service and then just cancel that "card".

I feel there is a whole cadre of consumer tech that is defensive against corporate taxes/tolls on our time. Eg: auto phone tree navigator, only allowing calls from double opted in contacts etc.

pclowes··on Flounder Mode – Kevin Kelly on a different way to do great work
Thank you for all your work! I have several Stripe Press books, especially enjoyed Revolt of the Public and Pieces of the Action.
pclowes··on Flounder Mode – Kevin Kelly on a different way to do great work
This was a breath of fresh air.

The tech sector has grown and changed so much. It has gotten much more "professional" which is arguably good but it this in turn promotes a fair amount of "corporate stooge" behavior. I am guilty here for sure, it is really easy to focus on levels, promo packets, OKRs, especially as you age and responsibility grows and forget what make this industry amazing in the first place.

Good reminder to focus on direction and interests and what you feel should be built. Reminds be a bit of the opening section of "The Art of Doing Science and Engineering" which I only came across because I liked other Stripe press books.

You also meet more interesting and passionate people if you pick a direction vs a destination.

pclowes··on The uncertain future of coding careers and why I'm still hopeful
Why are software engineers paid well to date? Successful software uniquely benefits from operational and financial leverage. Which creates the term "software margins"

Financial leverage: It is cheap to write code, experiments are inexpensive. A million dollars in funding for a B2B SaaS gives many more shots on goal vs a million dollars in funding for drug research or manufacturing. This increases probability of ROI and permits aggressive investment.

Operational leverage: Scaling code is cheap as well. It is free to copy. Solving one problem with software well often enables solving immediately adjacent problems very cheaply.

Do LLMs decrease or increase the leverage here?

Writing code is cheaper, a single engineer can now do much more. Does that endanger engineers? Yes, if their job is "take requirements and implement to spec". No, if their job is "solve important business problems at scale". The former are already typically not valued or paid exceptionally highly. The latter are likely to be valued and paid even more than they already are.

Or put another way, if software engineers are going to be hurt by LLMs who is going to benefit? This is assuming a zero sum game which I would disagree with here. But if not software engineers than who is better positioned to wield LLMs effectively?

pclowes··on Writing Code Was Never the Bottleneck
I have not found LLMs to be most beneficial in straight writing code especially in complex large scale systems. However, I have found them extremely useful at the aspects the author claims they make more difficult: understanding, testing, and trusting that code.

An LLM is an extremely useful search engine. Given access to plenty of CLI tools asking it questions about an unfamiliar code base is extremely helpful. It can read and summarize much faster than I can. I don't trust its exact understanding but I do trust it to give me a high level understanding of an architecture, call out dependencies, summarize APIs, and give me an idea of what parts of the code base are new/old etc.

Additionally, having LLMs write tests after setting up the rough testing structure and giving it a few examples massively decreases the amount of time it takes to prove my understanding of the code through the tests thereby increasing my confidence/trust in it.

pclowes··on Tools: Code Is All You Need
Directionally I think this is right. Most LLM usage at scale tends to be filling the gaps between two hardened interfaces. The reliability comes not from the LLM inference and generation but the interfaces themselves only allowing certain configuration to work with them.

LLM output is often coerced back into something more deterministic such as types, or DB primary keys. The value of the LLM is determined by how well your existing code and tools model the data, logic, and actions of your domain.

In some ways I view LLMs today a bit like 3D printers, both in terms of hype and in terms of utility. They excel at quickly connecting parts similar to rapid prototyping with 3d printing parts. For reliability and scale you want either the LLM or an engineer to replace the printed/inferred connector with something durable and deterministic (metal/code) that is cheap and fast to run at scale.

Additionally, there was a minute during the 3D printer Gardner hype cycle where there were notions that we would all just print substantial amounts of consumer goods when the reality is the high utility use case are much more narrow. There is a corollary here to LLM usage. While LLMs are extremely useful we cannot rely on LLMs to generate or infer our entire operational reality or even engage meaningfully with it without some sort of pre-existing digital modeling as an anchor.

pclowes··on Location data show that race affects citations and fines for speeding
You could override it and i am sure “gps attack” would be an allowed exception.

People can hack cars already.

We didnt always have to make everything to serve cars and we can stop anytime we want.

pclowes··on Location data show that race affects citations and fines for speeding
I just don’t think it is that hard of a problem. The legislative hurdle is much higher than the technological difficulty.

You can always fail safe. Eg (extreme case) unless your car can prove to itself it is on a highway or a high-speed street it can only go 25 mph.

pclowes··on Location data show that race affects citations and fines for speeding
You don't believe there has been widespread failures if not malpractice and outright fraud in many journals and published studies resulting in replication crises and performative/self-serving peer review over the last several years?

Also cars can read speed limit signs pretty accurately. GPS is pretty accurate. It should not be possible to go highway speeds in areas that are purely residential and miles away from the highway.

Also we could make a more accurate map enforcement system. Not trivial but straightforward.

pclowes··on Location data show that race affects citations and fines for speeding
A large heavy car going from 0-60 in 4 seconds through a mass of innocent people on Bourbon St is different than words online.
pclowes··on Location data show that race affects citations and fines for speeding
I very well believe this could be true. However, after hearing of many studies that don’t get published unless they confirm the suspicion of racial prejudice I don’t even know what to believe anymore. Side effect of a post-truth society locked in endless red-blue proxy fights I guess.

I don’t know why we need police doing traffic enforcement. The cars know how fast they are going, the cars know where they are, the cars know what the speed limit is, why are they able to speed? Enforcement should be fully automated and violations should be forced to provide evidence of a special situation id they do speed eg: in labor etc. or a ticket is automatically issued.

Also in favor of governors on max acceleration and relative force generation for large heavy vehicles. Eg: 9k lb electric truck can do 0-60 in a minimum of 8seconds.

pclowes··on Palma 2
Im confused, how is a different technology the bar to meet for a given technology?

That feels like saying “my car can go 60mph so thats the bar to beat for a bicycle”.

Whats the battery like on the daylight?

pclowes··on xAI has acquired X, xAI now valued at $80B
This valuation seems about as justified as a bored ape NFT…
pclowes··on Tim, don't kill my vibe
Amen, I am so sick of the appificiation of everything. Just use a web app. Really annoying that major sites make their sites buggy and brittle seemingly on purpose. Eg: reddit is terrible to use on mobile web and I suspect they make it so on purpose to drive app adoption but why would I ever want or need an app for Reddit?!
pclowes··on Tim, don't kill my vibe
Does somebody have an example of a good app made by vibe coding? I keep hearing this and while I am sure its great for prototyping and exploration I just don't have any examples I would be happy to use frequently.

I wonder if vibe coding is a false economy? I am personally not excited to use tools and products thoughtlessly glopped together.

pclowes··on How to create value objects in Ruby – the idiomatic way
I think it is just a more ergonomic, lighter weight, and intention revealing alternative.

If I see a data class instantiation. I immediately have an idea of its scope and what could/couldnt go wrong with it.

pclowes··on Forget Twitter threads and write a blog post instead (2021)
I find that feature very annoying. I don’t like long form text interspersed with micro-blog posts/disussion.

Needed feature (partially why they acquired Revue) but implementation was botched. Also now many individual tweets in a thread are too long and expanding and scrolling through them is busted UX as well.

pclowes··on Ask HN: What less-popular systems programming language are you using?
Are people using Elm for systems level programming? I have only used it on the front end.
pclowes··on We are the builders
What do you mean by open? A lot of government work is very sensitive. Also I have yet to find an industry where contractors are fungible and swapping the works effectively.
pclowes··on We are the builders
I was thinking back to when Clinton balanced the budget.

And yes I directionally propose that we should build houses until property prices fall and then build some more.

The housing crash was funded primarily by the govt. Loans were being made because of govt policies that never should have or would have been made otherwise!

2008 was devastating because it made us stop building housing due to speculative price manipulation of securities and again the govt being in a business it should not have!

https://threadreaderapp.com/thread/1880048376349356050.html

pclowes··on We are the builders
That should definitely be looked into. Also I would be surprised if a significant company was MORE than one relationship away from the White House or an influential senator. These networks are small at the top.
pclowes··on We are the builders
Prolonged deficit spending is by definition not sustainable.

We have been doing that for about 25 years but can’t continue indefinitely. Especially not with higher interest rates.

Government has objectively gotten worse despite massive tax revenue and is involved in a lot of endeavors that are well outside its scope IMO.

As just one of many examples: Just try to build a house pretty much anywhere and you will encounter fractal bureaucracy. This is during a nationwide housing shortage and many societal ills can be directly linked to housing costs!

Our national debt looks like a hockey-stick: https://fiscaldata.treasury.gov/americas-finance-guide/natio...

And we spend about as much money on the debt as our entire bloated defense budget!

pclowes··on We are the builders
In defense of this process is the government needs to work with stable companies that are highly likely to be in business in 10-15 years. That rules out a lot of startups.

Networks matter and trust matters, they should prefer to do business with people they know!

That doesn't mean there shouldn't be accountability and investigations into cushy speaking gigs. But the government should absolutely work with established companies run by executives they know.

pclowes··on We are the builders
This is interesting but very much lacking in details, it needs exact examples. I really feel for the workers at USDS. I was an engineer during the Elon Twitter acquisition and saw the thoughtless destruction first-hand.

However, the burn-it-all down approach does have some merit that critics of Elon/DOGE never admit to. How do they propose you carefully untangle the knots of fractal bureaucracy at speed and produce results if not by just cutting them off? The previous approaches of a special committee etc. just add a another fractal and yet another process.

Sometimes I feel the critics would be content if nothing was ever accomplished, if nothing ever changed, as long as thoughtful meetings were conducted and stakeholders were consulted. There is a very real layer of inertia that needs to be punched through, velocity has merit all its own.

I am very concerned about the possible outcomes of DOGE overall but business as usual just means the US goes bankrupt slowly with all the correct protocols observed. I am glad the inertia is being punctured.

pclowes··on The Profitable Startup
Maybe for something truly novel, but the exception proves the rule here IMO.

What is an example of a not simple idea pure software play that would require VC funding at the earliest stage?

pclowes··on The Profitable Startup
I forget exactly who said it but somebody said something along the lines of “venture capital is rocket fuel, unless you have a rocket it will only blow up your engine. Have a Bugatti Veyron? It will blow up your engine.”

This is also something DHH has been saying for years. However, I think it is more true now than ever. With how easy it is to start and scale a software company I really struggle to understand the justification for venture funding at the earliest stage, unless you want to larp as a founder, have low conviction, or just want “the experience”.

pclowes··on Trot
Maybe but I think it is fair for users to want richer more reactive experiences, especially with highly complex UIs in the browser.

However, react is 14 years old and there are (IMO) better options with cleaner abstractions available (hotwire, liveview, svelte)

← PreviousPage 7 of 8Next →