HNHacker News
TopNewBestAskShowJobs

Tainnor

4,317 karma · joined October 31, 2019

submissionscomments
Tainnor··on Mill: A better build tool for Java, Scala, and Kotlin
Is this portable? does it support incremental compilation? What about doc generation? code generation? running tests? generating test reports?

Of course things are simpler if you restrict yourself to 10% of the functionality.

Tainnor··on Mill: A better build tool for Java, Scala, and Kotlin
An apk is just a ZIP file, as you yourself were saying. So if you want to build it all via bash scripts, feel free.
Tainnor··on Mill: A better build tool for Java, Scala, and Kotlin
Great, then why don't you ship your software with bash scripts then?
Tainnor··on Ask HN: Is it time to fork HN into AI/LLM and "Everything else/other?"
I mean, many people who "hate AI" don't think that LLMs are useless for everything. I'm very unconvinced by e.g. using LLMs for coding, but that they'd be good at tagging content, sentiment analysis, etc.? That's not really hard to believe.
Tainnor··on Mill: A better build tool for Java, Scala, and Kotlin
> Just about any (Java) build tool supports incremental compilation

I wish that were true. Maven is terrible at this, e.g. here's a bug report that's been open for 12 years: https://issues.apache.org/jira/browse/MCOMPILER-209

Even just doing things like "please predownload any dependencies in this one job and cache them, so they can be reused in these 5 other jobs" don't work reliably with maven because certain plugins like surefire download dependencies on the fly.

> Nor everyone is building Google sized monorepos.

Nor is everyone building "small-ish codebases". You don't have to be Google, I worked at a company with maybe 10 developers in total, but the codebase was over 10 years old and had a gazillion features (plus a lot of tech debt, obviously) and the build times were atrocious. It wasn't all maven's fault, of course (we had really badly written tests, for example), but the broken incremental compilation easily added 5-10 more minutes to each build.

Tainnor··on Mill: A better build tool for Java, Scala, and Kotlin
That's like saying "programming is just creating files with content".

Python is a purely interpreted language that runs in the context of your file system. Java code needs to be compiled and packaged, you have to write a manifest, you have to decide whether to include dependencies (fat jar) or not, or maybe you want to create a war (if you're stuck in the 2000s)... then what about maybe distributing the sources too?

Also, Maven works. Gradle works. Mill probably works too. The reason uv is a success is that everything Python had before was a broken mess. When people complain about JVM build tools, they come from a different position, not from one of "stuff is completely broken".

Tainnor··on Mill: A better build tool for Java, Scala, and Kotlin
The difference between a tool that leverages incremental compilation effectively and one that almost always has to build from scratch (at least on CI) can be pretty huge and not "in the order of seconds".
Tainnor··on Mill: A better build tool for Java, Scala, and Kotlin
Does Mill have the capability to enforce dependency constraints on transitive versions without declaring them explicitly? I find this to be quite a useful feature for dealing with automated security review and gradle can do that:

https://docs.gradle.org/current/userguide/dependency_constra...

Tainnor··on Mill: A better build tool for Java, Scala, and Kotlin
JVM build tools have to do much more than just dependency management because of how the JVM works (e.g. packaging resources). It's not really comparable.
Tainnor··on Mill: A better build tool for Java, Scala, and Kotlin
My main problem with Maven is that it's dog slow and incremental compilation doesn't really work, especially on CI. For huge repos that's a real issue.

In a sense I feel that part of the microservice craze is due to the fact that many of our build systems are not good enough to allow us to work with huge monoliths efficiently. Gradle is a bit better (definitely faster), but comes with additional complexity. Haven't tried Mill.

Tainnor··on LLM Inevitabilism
> Except we are experts on programming

Which is not the same thing as being able to judge a study design.

Tainnor··on LLM Inevitabilism
> All of those studies have been torn apart in detail, often right here on HN.

You mean the same Hacker News where everyone was suddenly an expert in epidemiology a few years ago and now can speak with authority to geopolitics?

Tainnor··on A software conference that advocates for quality
> The point I’m making is that: my boss would laugh me out of the room if I told him “You can’t measure the quality of my electronics designs or my delivery process; it’s art.”

You can find some kind of objective metric, e.g. bug count or time spent developing new features. That alone is super hard to get right, but even if you could, it wouldn't necessarily tell you which techniques lead to a better result. People have tried studying such things (e.g. do static types help) and the studies rarely come up with any effect.

I don't think that's necessarily because these things don't have an individual effect, but because there are humans involved and so, personal ways of thinking probably play an outsized role, so technique X might be a very good fit for person A, but not for person B.

Tainnor··on A software conference that advocates for quality
Few people who advocate for software quality want better software just because it's more aesthetically pleasing.

Often, we've worked on codebases that were nearly collapsing under their own weight, where adding even small features would take an inordinate amount of time, where most changes would have unintended consequences, where onboarding was incredibly hard, etc. And we've worked on other codebases where changing things was easy, so we know the difference.

Often, the business has no clue how to fix this because a) they're non-technical, b) we unfortunately have no consensus in the industry about what "good code" is, c) humans are much better at short-term than long-term thinking.

Tainnor··on Overtourism in Japan, and how it hurts small businesses
To be fair, that's not a very difficult sentence to read for someone who has studied a moderate amount of Japanese. Doesn't mean you could actually order food.
Tainnor··on Overtourism in Japan, and how it hurts small businesses
The time of the year probably matters too. I didn't find Japan to be terribly overcrowded when I went this February. Certain areas (and the minuscule Kyoto buses) were, but that happens in every tourist location.

I also went to places like Beppu or Kagoshima where I barely saw any tourists.

Tainnor··on US Court nullifies FTC requirement for click-to-cancel
> The conceptual difference here is important. The goal should be to preserve privacy, not fine arbitrary large amounts of money. Like, sometimes you do need to fine people but personally I'd prefer auditing requirements for Meta/Google et al rather than fines.

I'm operating under the assumption that most tech giants (and in particular Meta) are doing everything in their power to keep violating people's privacy as long as they can get away with it.

Since it's impossible to audit everything such a huge corporation is doing and since cases can take a very long time to resolve (after which the damage is often already done), the huge fines are for scaring companies into submission. I am perfectly fine with that, no actually: I wish the fines were so big that they were actually threatening the business's survival.

> I'm not sure where the right balance is, tbh.

I'm probably the wrong person to ask because I think that Facebook simply shouldn't exist.

Tainnor··on Apple vs the Law
These companies may be smaller than FAANG, but I also feel that if BASF disappeared overnight it would have a larger impact on the world than if Facebook disappeared.

HN is sometimes incredibly biased towards consumer tech.

Tainnor··on AI coding tools can reduce productivity
> I expect the answers of the AI to be really good. If it isn't, having explained my problem to the AI was simply a waste of time.

I find the worst part to be when it doesn't correct flaws in my assumptions.

For example, yesterday I asked it "what is the difference between these two Datadog queries"? And it replied something that was semi-correct, but it didn't discover the fundamental flaw - that the first one wasn't a valid query because of unbalanced parens. In fact, it turns out that the two strings (+ another one) would get concatenated and only then would it be a valid query.

A simple "the first string is not a valid query because of a missing closing paren" would have saved a lot of time in trying to understand this, and I suspect that's what I would have received if I had prompted it with "what's the problem with this query" but LLMs are just too sycophantic to help with these things.

Tainnor··on Apple vs the Law
> But that model you describe is cracking: Cost of living is going through the roof in Europe, taxes/social contributions going up every year, etc.

Cost of living is increasing in the US too (in large part due to geopolitical reasons), and social contributions are rising because of demographic factors. I'm not sure how market liberalism is magically going to fix that latter issue.

Tainnor··on Apple vs the Law
These are all huge companies in my view.

In addition to them there are also countless small to medium sized companies that nobody's ever heard of, that don't experience hypergrowth but have slow and steady growth - especially in the B2B sector. I've worked at some such companies.

Tainnor··on AI coding tools can reduce productivity
It matters in the sense that it is unclear whether the findings generalise to other people. Which is a problem that a lot of studies, even with more participants, have because they may not have a diverse enough set of participants.

But in terms of pure statistical validity, I don't think it matters.

Tainnor··on AI coding tools can reduce productivity
The sample size isn't 16 developers, it's 246 issues.
Tainnor··on The death of partying in the USA
People travel to Berlin to go party in clubs, not for home parties.

Partying in someone's apartment is a thing in probably every reasonably sized city in Europe, not just Berlin. Although you should probably alert your neighbours.

Tainnor··on US Court nullifies FTC requirement for click-to-cancel
https://noyb.eu/en/eu-court-irish-dpc-must-investigate-noyb-...

https://archive.ph/20230123014444/https://www.irishtimes.com...

Tainnor··on US Court nullifies FTC requirement for click-to-cancel
It's definitely better in Europe, but certain courts and DPAs (especially the Irish one) are unfortunately known to be incredibly business friendly.
Tainnor··on Why LLMs Can't Write Q/Kdb+: Writing Code Right-to-Left
This argument in favour of mediocrity and catering to the lowest common denominator is one of the key reasons why I dislike people who want to shove LLMs into everything (including art).
Tainnor··on My first verified imperative program
But "safe" and "unsafe" are not a binary thing.

Maybe for LOB applications, it would be better if languages defaulted to arbitrary precision arithmetic. But scientific computing is also a huge field that often uses the same languages and there, arbitrary precision is often the completely wrong tool, e.g. it would make certain key algorithms (like Gaussian elimination) exponential.

I feel like this is just one of these things that developers should know about so they can make the correct choice, just like DB indices.

Tainnor··on Adding a feature because ChatGPT incorrectly thinks it exists
absolutely nothing, except that some pencil-pushers with poor understanding of grammar somehow dislike it even though they often can't even identify it correctly.

Language log has been writing about this for so long it's not even funny: https://languagelog.ldc.upenn.edu/nll/?cat=54

Although the best place to start is probably here: https://languagelog.ldc.upenn.edu/nll/?p=2922

Tainnor··on My first verified imperative program
> In the worst case you should at least make your language fail-stop definite by making it error when the result of + is too large rather than silently continuing in an invalid state.

That's what Swift does fwiw. But there are downsides to that too (although they're also related to Swift not having a good fault recovery mechanism).

In practice, whenever you know you're going to use large enough numbers in Java, you probably want to use BigInteger or BigDecimal (e.g. in banking). Unfortunately, Java doesn't allow you to use the + operator with it.

← PreviousPage 3 of 34Next →