kiitos

delegation of trust is an essential and unavoidable property of any system that serves a non-trivial number of human participants

keypair != human

> I look at cross core communication as a 100x latency penalty

if your workload is majority cpu-bound then this is true, sometimes, and at best

most workloads are io (i.e. syscall) bound, and io/syscall overhead is >> cross-core communication overhead

> How do we reconstruct past memory states? That's the fundamental problem.

reconstructing past memory states is rarely, if ever, a requirement that needs to be accommodated in the database layer

    Within the Google codebase and in projects using Bazel, directory layout for Go code is different than it is in open source Go projects: you can have multiple go_library targets in a single directory. A good reason to give each package its own directory is if you expect to open source your project in the future.

:o :o :o

are there really people saying that "giving each package its own directory" is in any way optional?? it is literally part of the language spec, what on earth would make anyone think otherwise??

edit: ok so bazel folks are just on a completely alternative timeline it seems

https://github.com/bazel-contrib/rules_go?tab=readme-ov-file...

    Bazel ignores go.mod files, and all package dependencies must be expressed through deps attributes in targets described with go_library and other rules.

so bazel doesn't support go, gotcha

> the build graph -- the very thing that BUILD files define -- is the best place to encode [dependencies] in a programmatic manner.

so the thing is that a BUILD file doesn't define the build graph, it approximates it -- the build graph is always defined by language-specific tooling and specifications

it's fine that the BUILD file is an approximation! that's as good as you can do, if you want to try to model dep relationships between heterogeneous languages

so when we're talking about the dep graph, "using language-specific tooling" isn't a detail you can brush aside, it's a core requirement for correctness, really

> BUILD files give you a chance to encode the high-level architecture of your software project as a graph of dependencies that lives outside of the code.

this is wild, the code literally defines dependency relationships at the mechanical level, it's the source of truth

of course you can't derive dep graphs from source code via simple file-based grep analysis, but that's obvious? you need to use language-specific semantically-aware tools to do that. if the author believes otherwise that's their mistake

indeed, no idea why so many folks seem to think `bazel` is like some co-equal alternative to language-native build tooling/processes, it's a fine tool for certain (niche) use cases but in no way is it ubiquitous or anything approaching a common standard

"event system" is any mechanism where events (readiness, completion, signals, GUI messages) drive control flow, restricting "event system" to descriptor readiness exclusively is the author's personal framing, not exactly common parlance

there are just as many equally-qualified experts who have been consistently giving exactly the opposite feedback/signal, which you never seem to acknowledge or incorporate into your comments

for me the takeaway here isn't that sorting should use some smarter/better mechanism for inferring semantic intent from filenames, but rather that sorting should not try to infer semantic intent from filenames in the first place

you're right, i guess it's only negative if you think it's important for people to understand the code they produce, if that's not a concern for you then no problemo

> If they are playing a constructive part inside the organisation, at what level of notoriety and assholeness should the moderators pull the pin?

in general, that's 100% up to the discretion of the owners/moderators (shrug)

what?

private communities (which in this context would include any website that doesnt end in .gov) banning assholes, has no impact on the human rights of those banned assholes, it is not a human right that you get to have an account on a private website

in the same way that getting trespassed from chipotle for not wearing pants when ordering your burrito bowl, doesn't mean your rights have been violated

what do I find objectionable?

> The first step to resist or undo Woke Invasion in your organization (or your psyche) is to thoroughly understand its creed Critical Race Theory, so as to uncover the fact that generally speaking woke disciples care less about the problems in the world than assuaging their self-centered ideological feelings. 1 The next step, obviously, is then to effectuate an elimination of the wannabe woke invaders from your organization by instituting a culture based on common sense values stripped of identity politics.

this wildly pejorative definition of the central concept at play in the discussion, probably, is a good start to what i find objectionable, yeah?

or maybe the author's own definition of "wokeism"

> Wokeism is a secular religion that originated in the United States of America, based on the pseudoscienfic field Critical Race Theory. It presumably took roots around 2016 (see Woke Invasion) and has been withering away since around 2024. Bigoted ideologies like neoracism fall under wokeism.

which is about on the same level as vaccines cause autism

i'm sure there are lots of people who think otherwise and maybe you're one of them but frankly there is nothing useful to be gained by arguing the merits of this kind of stupidity

> There's a user expectation that photo20.jpg comes after photo3.jpg.

my user expectation is the opposite

i get what you're saying but it's not achievable in practice, at least not consistently

ls sorts filenames strictly lexicographically, comparing character by character, so e.g. "055436307" is compared as the characters "0", "5", "5", etc. so it sorts before "121134" because "0" is less than "1". if all compared characters match and one string ends, the shorter one comes first. Symbols like _ are just more characters, and their position relative to digits depends on the locale’s collation table.

Google Drive uses ICU collation with the numeric option enabled, which treats each consecutive sequence of digits inside the filename as an integer. so "055436307" is parsed as the number 55,436,307, while "121134" is parsed as 121,134. and since 121,134 < 55,436,307 then "121134..." comes before "055436307..." even though lexicographic order would suggest the opposite. and i think when two digit runs have the same numeric value, the shorter run comes first; if runs are equal and the string continues, then normal character comparison resumes, including any underscores or suffixes

if you're spending anywhere near as many engineering hours "getting code to work" as you're spending "thinking" then something is wrong in your process

nobody in this or any meaningful software engineering discussion is talking about software projects that are 1000, or even 10000, SLoC. these are trivial and uninteresting sizes. the discussion is about 100k+ SLoC projects.

> If the thinking bit is your favorite part, AI allows you to spend nearly all of your time there if you wish, from concept through troubleshooting...

What about if the "knowing/understanding" bit is your favorite part?

> do you not recognize that collectively, as a society, Nazis decided to attack Jews, trans, disabled, and others? ... in the end very few people think they themselves are evil.

yes, I agree with you, that society made some pretty bogus determinations, and they certainly didn't see themselves as evil. but in the fullness of time and history their position has been understood as wrong.

> I think you are oversimplifying the problem because you are relying far too much on the obviousness of Nazis being evil.

my point isn't about nazis or the obviousness of their evil. my point is that advancement as a species requires delegation of moral authority to collective government i.e. society. and transitively that the possibility of pathological negative outcomes doesn't somehow invalidate this idea outright. we don't throw away the concept of a judicial system because innocent people can be declared guilty. we work towards eliminating those failures in what is otherwise an essential component of government.

zoom out. think larger. be more empathetic. nazis and maga and all of this garbage are bugs in the system, which we're fully capable of stamping out, in the long term.

> https://srid.ca/unwoke

i mean, this is a set of opinions and positions that are far beyond anything that could be called "moderate opinion[s] on gender representation" and are pretty uncontroversially terrible, particularly in the context of any non-homogeneous community of people

if you post something like this to the public internet and stand behind it, then man i'm not sure what you expect, you're self-identifying as an asshole, and it can't be surprising when you're banned from places as a consequence

edit: good lord, i clicked around a bit more on that website, dude is obviously a psychopath, and i feel duped even responding to this kind of nonsense

the people talking to you are talking about something very different than simply "political beliefs that you disagree with"

the appropriate level of capital gains tax at the 80th percentile is a political belief that you can tweet about in your personal time and allowing there to be a civil relationship with your colleagues in a professional environment. this is a political belief that reasonable people can disagree with.

asserting the supremacy of the white race is not a political belief that you can tweet about in your personal time while still allowing a civil relationship with your colleagues in a professional environment. this is not something that reasonable people can disagree with.

> The problem is who gets to pick who is right and not?

we all do, collectively, as a society

> So yes, to protect those groups being persecuted (trans, minorities, and Jews alike) you need to protect the speech of abhorrent groups like Nazis.

there is actually a categorical difference between advocating the persecution of minorities, and advocating the persecution of nazis. and furthermore it is actually possible and good for a society to say one of these things is bad and should not be allowed, while the other one is good and should be allowed.

> The "fundamental distsys problem" here originates in the language level: the problem is the attempt to hide the existence of a network protocol from the application code.

so the problem isn't about whether or not the application knows a network protocol is involved. it's that state has fundamentally different semantics when any network protocol gets involved.

languages, SDKs, etc. all exist "on top" of networks in the abstraction sense, once you start sending bytes between nodes over unreliable links then that stuff dictates the rules that you gotta abide by, you know?

i just don't use commits like you do, and that doesn't mean i'm being less careful or less thoughtful, or that my changes are worse than yours

commits are what i say they are, nothing more or less

> But most Americans want the government reduced,

facts not in evidence

tldr: no they do not

> trying to make network calls look like local calls hides important details which distributed applications must handle, like latency and network instability.

among many other concerns, yes

https://en.wikipedia.org/wiki/Fallacies_of_distributed_compu...

> My argument is that asynchronous programming, promise pipelining, and exceptions largely solve these issues just fine, and practical experience backs this up.

you can't solve fundamental distsys problems at the language level (lasp &c. notwithstanding)

hopefully not controversial

> The argument here is that the reason CORBA failed is because it lacked the language-level concepts to make it work well, in particular strong asynchronous programming primitives and promise pipelining.

i'm honestly not sure how to respond to this

CORBA failed because it didn't understand or accommodate the fundamental constraints of distributed computation. specifically the idea that you could define a useful abstraction of an object that worked the same way whether the implementation was local or remote. this is a protocol/design level error, nothing to do with any language-level concept, like async primitives or promise pipelining or whatever. i'm not sure how those things even entered the discussion? and i'm not trying to be glib, i'm really just re-iterating basic "lessons from CORBA 101" type stuff, which I've understood as common knowledge since the mid 90's...

> if you only use commits as some sort of help while developing, you might as well just squash them before making a PR.

yeah for sure you want to squash-merge every PR to main, right?

commits are just commits, there is no moral value to them, there is no "good history" or "bad history" of them, whether or not they're "made thoughtfully" isn't really interesting or relevant

git is just a tool, and commits are just a means to an end