HNHacker News
TopNewBestAskShowJobs

Noe2097

335 karma · joined June 20, 2018

submissionscomments
Noe2097··on JEP 430: String Templates (Preview) Proposed to Target Java 21
Why oh why is that a backslash!?

Over the 9 languages mentioned: 5 languages (inc. JVM ones like groovy and kotlin) are using `$`, 1 language (swift) is using `\`.

`\` is a pain to type on many keyboard layouts -- actually most but the US one. It seemed to me that `$` would have been a much more "conventional" choice.

This really makes me sad. It looks like the choice was made on purpose to be different.

Noe2097··on Ask HN: Reading material on how to be a better software engineer?
Try "Extreme Ownership".

It will give you concrete examples on how to broaden your scope out of just your individual contribution, by learning to care for all the rest, and pushing you to realize that to a large extent, the things you see as broken you can actually fix.

Even when they are not inside the codebase you are in charge of. Even if they aren't even on engineering side.

Noe2097··on Twitter’s mass layoffs have begun
> tech was a bastion of "treat your employees right, and they will be productive"

Is/was it? The shortage of qualified engineers have made some employers extra careful/generous, to attract and keep them. Or some others to divert brains away from their own occupations which could have ended up competing with them.

Noe2097··on Ask HN: Firing an employee under a month before vest?
I'm not saying it wouldn't trigger legal action or resentment. My point is that the question is on the performance or lack thereof and how it was shared and worked on with the contributor.

I'm saying that someone who hasn't been doing their job for 6 months, has been told so repeatedly, has been shown and guided through an improvement plan, but still fail to meet the expectations, should not be kept at their position. Whatever the date for whatever bonus, yes.

It doesn't matter if it happens one day, one week, or one month or 6 months prior to that date. What matters is whether everyone was aware of what was happening and the implications of it.

If that is not the case, then yes sure, start now to build that file. In this instance, the date will pass, and in a couple of months from now the contributor will be out. The next instance, act faster.

Noe2097··on Ask HN: Firing an employee under a month before vest?
It's unclear to me: Why are you wondering _now_?

Vesting is compensation unlocked after a period of time. It is _meant_ to keep people _engaged_. You shouldn't feel bad for having to terminate someone two weeks before vesting, or 1 month before vesting, or 6 months before. No more than you should feel bad for having to stop paying someone that you fired.

Do you have _facts_, undebatable ones, demonstrating that the person is below expectations? Do other team members share the same point of view? Do you have feedback from others, pointing out that this person wasn't pulling its part? Did you share that with the person? Did you share these elements, did the person acknowledged that there was missed expectations? Did you agree on concrete goals to progress and meet expectations? Did you guide that person on this, for like 2~3 months?

If you answered YES to all these questions, but the person is still failing, then let that person go. You've made your job. Whether they are vesting tomorrow or in 6 months or in a year, is _a detail_. They have been failing to meet the bar, you've told them, you explained them how to reach it, you guided them for some time, and they are still failing.

If you didn't, then start right away.

Put vesting in parenthesis. That is not the topic at hand. The topic is meeting expectations or not.

Noe2097··on Vectors are over, hashes are the future
LSH is a _technique_, whose performance vastly/mostly depends on the hashing function and on how this function enables neighborhood exploration.

It might not be trendy, but it doesn't mean it can't work as good or better than HNSW. It all depends on the hashing function you come up with.

Noe2097··on An inorganic user engine driving ad campaigns on major websites and podcasts
I must be missing something; hasn't this been happening for years? Viewing an ad in exchange for a continue?

How is that different from ad videos that must be seen to access otherwise "free" resources, such as free WiFi in Airports? Or even from video ads that auto play on YouTube? None of them captured my genuine interest ; but I accepted the deal of my brain time (or my closed eyes) in exchange for some retribution.

Noe2097··on Paving the on-ramp to Java
Couldn't agree more.

Recently I tried another take on the same idea: using default methods in Interfaces. It has nice properties, as it enables ~mixins in a way not dissimilar to what Scala provides, and enables overriding (e.g. for tests). Give it a try, it's ... refreshing :)

Noe2097··on Experimenting a new syntax to write SVG
Why going for CSS as a syntactical starting point, if it is to change it so fundamentally that no IDE or existing tools will be able to handle it, and if it is to move away from its "theoretical" purpose of describing style independently from content?

Going that way, there are so many (existing) possibilities. Want something that looks like CSS but that has more structure? Use YAML. Preferring braces over spaces to delimit blocks, but don't fancy quotes everywhere? Use Groovy -- you will even be able to write a DSL to generate SVG in an afternoon, with the entire power of the JVM. Don't want anything executable? Forbid it. Refrain from picking a syntax that are not fitting your needs and for a language whose intent is different, you will just call for confusion

Noe2097··on It's time to halt starting any new projects in C/C++
Was about to write this exactly. Keep the lifecycle management in a dedicated component. Unsafe seems vastly overkill for dealing with this use case.

I'm even having a hard time seeing how this could be slower than any other alternative. Yes, in the places where creating / deleting is necessary, the "root" structure will have to be passed around, but in the worst case that has the cost of adding one argument to a function (which has the nice side effect of making the lifecycle _visible_).

Noe2097··on TikTok and the Fall of the Social-Media Giants
I make a difference between what FB customers pay for promises of brain time, and what the service FB provides actually is worth for _me_ (people are no customers of FB). $40/month/user might be what FB get from their customers. It does not mean that it is the value it is actually providing to its own end users (excluding ad afficionados if that exists).

There is also a difference with what it actually costs to provide the useful part of the service. Mastodon is open source, yes it requires hardware (virtual or actual) and operation, none of which is free, but the cost of these certainly doesn't match the cost of running+developing game changing (advertisment/recommendation) ML models and ultra high scale infrastructure. Running a mastodon server for a thousand users with acceptable availability, and no fancy ML or R&D, I think should be doable for $5.000 a month, yes.

Noe2097··on TikTok and the Fall of the Social-Media Giants
They weren't making money, were they?

I guess it depends what you mean by Evil. Force-pushing ads and sharing updates about people you don't know because it could keep you connected so that you could see more ads, made them profitable. A free service that doesn't push ads can't be.

Noe2097··on TikTok and the Fall of the Social-Media Giants
Probably the money that FB "generates" is literally insane, and doesn't make sense for a single company to generate.

Possibly a mastodon based solution where users pay a small recurring fee for server management and some R&D and offering privacy, could generate enough benefits for a company (or companies) to strive.

FB (among others) made us forget the sense of scale and the value of things, including the value of privacy and brain time. I would rather pay today from 1 to 5 USD a month for a service that works, where I can share my kid's pictures with my direct family and only them, with no ads.

Noe2097··on Ask HN: Examples of large acquisitions that did not ruin the acquired company?
Oh fully agreed; SolidWorks is much more an exception than the norm here.
Noe2097··on Ask HN: Examples of large acquisitions that did not ruin the acquired company?
I'm curious here: what got better in GitHub since the acquisition?

It's a genuine question, I'm not being sarcastic, I really don't understand the statement (which seems to be shared by several people here). I honestly can't point out a single thing that was delivered that changed the game for like, 6 years from now if not more (copilot is a nice gadget IMHO but it's not really game changer, is it?).

I was actually hoping that MS support would change that, but I don't see it happening.

Noe2097··on Ask HN: Examples of large acquisitions that did not ruin the acquired company?
Dassault Systèmes acquiring Solidworks.

Most users don't know that Solidworks, which is well know as a mid level solution for CAD, is actually owned by the French company developing CATIA, which is practically used by every single serious manufacturer in the world (from automotive to aircraft to ship to factory to you name it).

The geometric kernel powering these two is now the same, but it wasn't for years. I guess it's a good illustration of why this acquisition was successful and didn't ruin Solidworks: the company was left practically untouched for more than ten years. "assimilation" took place at a very slow pace (almost cosmological for a software company).

Now the question to ask is more why the acquisition is taking place. To get a list of customers? To get skilled contributors, or knowledge, or patent, or a piece of technology? To get a presence in a different country? All or a part of this? To kill a competitor?

Anyway you put it, acquisition is not a neutral move. It's always joining a bigger structure, hence jumping to a later stage and size at extreme speed. Considering how difficult it is for a single company to get through its different stages by itself, it is normal that acquisitions are hard for acquired companies. Assuming the objective is not to kill the acquired, tremendous efforts will be needed on both sides to make it work. The acquiring company ramping up newcomers, on new actual processes, on new ways of doing; the acquired one to agree to it, and accept that it is the cost to pay to operate at a higher scale.

Noe2097··on EU approves legislation to regulate Apple, Google, Meta, and other tech firms
I see it this way: the EU may choose whatever way it wants to compute a fine -- they are making the law, they could have written a fixed amount, or an amount based on revenue made in the EU (which is probably a pain to define and certainly easy to "workaround"), or an amount relative to the average temperature in the Sahara over a year.

Whether it's "fair" is an entirely different topic. And I guess any company could try to fight that decision in court.

Noe2097··on Crafting Interpreters: A Review
Following a slightly different path, "Modern compiler implementation" goes over several topics mentioned in the review (lexing, parsing, type checking, code optimization and generation, runtime environments aspects e.g. garbage collection).

There exists various flavors of the book (in ML, Java, and C).

Noe2097··on A Swedish skier was basically frozen, but lived (2016)
I discovered a few years ago that thanks to this kind of "experience", forced hypothermia is now used to prevent brain damage to new borns who suffered from loss of oxygen during delivery.

Our second son was (initially, thankfully) stillborn after a complicated delivery, yet the medical team managed to revive him and immediately put him in hypothermia (to 34° C), surrounding his head with some apparel containing essentially ice. Fast forward the 5 most painful days of my life, doctors warmed him back very slowly. He is now 4, and showing completely "normal" (in the statistical sense of the term) development.

Indeed, brain damage is not immediate in case of oxygen loss. It happens in phases. I am no expert, but I understood that most damage happen not immediately, and happen even if oxygen has been restored. Slowing down the body through hypothermia, seems to avoid the brain to "collapse".

Discussing with the head of the department (for a mandatory "retrospective"), he explained me that this technique had been devised after seeing that people who drowned in very cold water could be revived with less "impact" on their capacities than others. He also explained me that the technique was not always successful, but was the only "treatment" that had shown actual improvement of recovery chances in this kind of cases.

Noe2097··on Nuclear War Survival Skills (1987) [pdf]
Can't help but think about The Day After https://en.m.wikipedia.org/wiki/The_Day_After
Noe2097··on Ask HN: Do you know any computer language where collections are unified?
Surprisingly, one "language" actually fits pretty well (most of) your requirements: SQL :)
Noe2097··on Fix the unit test and open a giant hole everywhere
Hmmm. In Python, I would leverage the dedicated utility of the standard library:

  os.makedirs(path)
Noe2097··on Faster software through register based calling
In golang, everything is recompiled from source into a single standalone binary. There is no concept of shareable precompiled library -- a library is to be reused from its source code.

In that world, calling conventions is not something considered public, so modifying it does not break any compatibility.

Noe2097··on Tasking developers with creating detailed estimates is a waste of time (2020)
Discussing about the effort required for each task/story, is crucial to generate alignment within a team, determine better the scope of a task/story -- and decompose it, if it appears that the effort would be so big that tracking what has been done and what remains to be done would be intractable.

The by-product of these discussion may be summed up as a number. But oh yes its unit must not be time.

Noe2097··on Why do placebos work? Scientists identify key brain pathway
This article/study is all and only about "pain modulation".

Is it to say that there is only one placebo effect, which is always reducing pain? Aren't there placebo effects going "beyond" pain relieving? (Or is it the title that is generalizing the scope of the analysis too much?)

Noe2097··on No YAML
Taking a step back; this website is a piece of Art.

Its message stirs passions, it is written in the very formalism it talks about, and presented in a so unusually cumbersome way that it triggers the same kind of response (or questions) a surrealist setup do (essentially boiling down to: "why the f...?").

Thank you, it was delightful to experience.

Noe2097··on The Missing ”WHERE” Clause for Vector Search
Hi! Yes! Nice article (articles actually).

I realize you are hinting at it as the subject of a later article, but could you share with us a little bit more about single-stage-filtering :D ? how does it work? How can metadata and vector data "coexist" in a same index?

Noe2097··on Sir, Please Step Away from the ASR-33 (2010)
> I am not sure what the hell this guy is talking about

Have a look at the APL language the author is mentioning, you will see you missed the point of his article.

The point of the author is not just about having more characters supported for identifiers or within strings (as the part of the golang specifications you are pointing to). He has more operators and related constructions in mind, which could benefit from more elaborate characters, e.g. the whole set of mathematical operations.

The issue I see however in his reasoning,is that precisely in the case of APL, dedicated keyboards had been built for it. The reason for limiting character set to ASCII, is there immediate accessibility on everybody's keyboard

Noe2097··on What Is RPN? Why Did/Does HP Use RPN? Learning RPN (1999)
The HP Prime does have an RPN mode. I bought it recently, for that very feature, when my HP48 died. The keyboard of the HP Prime isn't really optimized for it (some trivial operations on the stack require key combinations, while on the HP48 there were physical keys for them), but it's acceptable.
Noe2097··on Designing Engineering Organizations
This post is interesting only as an example of a total lack of facts, comparisons, and perspective.

The lone backing of each opinion is the adverb "generally". There is no mention of any concrete team/company size where issues do/might occur in one or the other setup. There is no mention of successful or failing companies adopting one or the other setup.

And just to mention a single concrete counter-example to the groundless side taken here, take Apple. Apple's organization is _based_ on specialization, for the simple reason that the more specialized teams are, the higher their expertise is. Look at Spotify, Microsoft, and others; they moved away from the simplistic feature-squad model advocated here, because it simply didn't scale.

Certainly, communication across team members, and communication across teams are subjects of attention. It does not mean it should be ignored/avoided. It's like saying "scaling horizontally is too difficult, just stick to scaling vertically"; sure, it's easier, but the power is not the same.

← PreviousPage 2 of 3Next →