HNHacker News
TopNewBestAskShowJobs

ephaeton

364 karma · joined May 22, 2019

Currently: playing didgeridoo ("I'm Freddie Veggie and I play austrian didgeridoo" - JFGI).

Also: embedded linux / C++ development

Contact: hn -at- ayaken.net

Previously: type safety x unit safety

Previously: semantic mediation in production networks

Previously: embedded linux development for industrial automation

History: - Tcl -> Scheme -> C -> C++ -> C -> Common Lisp -> C -> Clojure -> Tcl -> C/++ & wanting zig (unimportant deadends like e.g. ruby, icon, python, perl elided)

- DOS 4.11 -> Win 3.11 -> Win95 -> Linux -> NetBSD -> Linux -> FreeBSD -> Linux (those damn hardware drivers)

submissionscomments
ephaeton··on Why GNU moe does not use UTF-8 or other form of Unicode?
"Oonicode" ?
ephaeton··on Useful sed scripts and patterns
Then there's sed -e 'cmd;cmd;cmd' as well.
ephaeton··on JSON for Modern C++ version 3.10.0
I haven't found a way to do that. The only O(n) to-json-array way is using the initializer list constructor AFAICT.
ephaeton··on JSON for Modern C++ version 3.10.0
Plusses: as a user of nlohmann-json, I also like the support for json pointer, json patch and json merge patch which comes in handy at times. I like their way of handling to/from_json (declare functions with appropriate signature and namespacing and the lib will pick them up seamlessly in implicit conversions). "standard" container facades are appreciated.

Minusses: although I'd like a way to append a (C-)array into a JSON-array "at once" and not iteratively (i.e., O(n) instead of O(n log n)). Also, lack of support for JSON schema is .. slightly annoying.

ephaeton··on “I Could Rewrite Curl”
> What Daniel Stenberg achieved is giving the world a fantastic, reliable, cross protocol, cross platform and cross language network library that can be used as the foundation for many projects.

One of which, curl(1), is a constant ass-and-life saver for me personally and professionally. Of course the library itself ends up being used by me as well (in C, ++ and tcl). curl to me is in the league of sqlite aka "absolutely gorgeous and more often than not all you need"

ephaeton··on NetBSD 9.2
somewhat. It's been quite a while and I do remember that hppa was one of the machines in my .. pillage pile of plundering (like an sun U1, a U2, an U-what-was-it-60? three flavors of sparc, an SGI workstation and various other UNIX pizza boxes) and my memory tells me except for the SGI we had them all running at some point (might've been hp300, hp700, should've been prior to hppa renaming. Looking at the numbers, hp700 sounds more familiar). The purpose of these boxes was indeed to offer shell access to students for the fighting the lemmings exercises (a series of C quizzes with regression tests aimed at showing students the pitfalls of non-portable programming), and we did have quite the variety.. But memory lies, so no, I'm not absolutely sure.
ephaeton··on NetBSD 9.2
Nah. SMP support predates NetBSD 5.0 by a decade or so. May sound like nitpicking but .. roundabout 11, 12 years ago was when the default kernel config switched to SMP on all architectures where it made sense. Before users had to enable it themselves. alpha & sparc64 SMP, I'm pretty sure, is quite older e.g.

I'm wondering if it's a sarcastic comment in the style of the 'BSD is dead' meme.

ephaeton··on NetBSD 9.2
I, too, loved the old logo. Not being of a nationality that has been involved in the pacific war though had kept the iwo jima reference hidden to me.

They didn't "suddenly" change the logo. As usual for NetBSD, everything was accompanied by painstakingly detailed discussions on the mailing lists. I always thought their communication practice to be exceptional and good enough to be a role-model for other OSS projects.

The list of platforms was partially very obscure bits, and not all supported at the same quality. We / they also had the problem of GCC stopping support for some of these architectures right around the same time. Also, bragging rights ... those to whom you may have bragged about it claimed linux was ported to more, not understanding the concept of an OS (in contrast to a kernel) in the first place ... and left people wondering, "fine, but does it run on today's hardware equally well?" so it was obscuring the system more than it helped, IMO. The message, again, IMO, should always have been: "obscure or in fashion, historic or bleeding edge, we have such a clean architecture and implementation that we scale across all these dimensions for a secure, performant, integrated, well-balanced OS being as close to POSIX-compliant as it makes sense for an OS to be." The list of platforms didn't get that across well.

The numbering system of the releases .. Was it 1.6 that was the last one with -current cycling about the alphabet and we had 1.6ZA as next version after 1.6Z? Or was it 1.5? I don't remember. Anyways, it was weird.

Don't get me wrong. I too long for a simpler yester-time. But I don't think they've done wrong, as a community found consensus on what to do, went ahead and did that, and they're still rocking no matter the specific version number of their OS.

ephaeton··on NetBSD 9.2
my favorite OS .. fond memories of resurrecting a sparc64, a sparc, a hppa, an i386 next to the amd64s, and feeling right at home with all of them at once.

...Portability exercises for students based on the "Fighting the lemmings" paper by Husemann. (have you ever truly considered endianness and alignment until you've ran i386-developed code on sparc64? :D)

...sadly my work demands a different OS usage from me :'(

ephaeton··on The “Marketing Is Evil” Problem
I believe that you only had to turn to marketing because the big established solutions continuously used marketing to remain big. If it weren't for online marketing, your access to market wouldn't have been by challenging the established brands with further marketing, but by providing extra value.

So your use of marketing ever only solved a problem established by marketing.

Which is good marketing for marketing, of course.

ephaeton··on Systemd: The Good Parts
You can use systemd-escape(1) to have it generate the correct unit name for you, e.g.: systemctl status $(systemd-escape --path /home --suffix mount) Yeah, complicated and verbose, but reliable and integrates into systemd mindset very well.
ephaeton··on Firefox Public Data Report
I hate the fact that they collect so much data. It's just the next company saying "nah, we're not evil! Don't worry that the browser we make sends home five hundred columns of data about your box and your usage patterns! We really really do promise not to peel out your identity from the data mud and monetize it! pinky promise! Just like google, we're not evil!!"

To those who claim you can opt-out of the "experience":

You can spend quite some time configuring firefox not to check for updates, get rid of all mentions of mozilla's servers in the config, try to have it not phone home, opt-out of any data sharing .. then you start it, and it does nonetheless: "Hey, update me please!" And what data do they share when checking for updates? Too much - I instructed the software to shut the fuck up and not ask for a fresh version, and surely don't communicate that I'm just using version X on my OS under my IP for fuck's sake! And that's just the obvious part of data that'll have to be transmitted in order for an update check to work! See pinky promise, above.

And then, after an update, if you don't start it with -ProfileManager to force your old profile (under linux), it will start with a fresh profile and ... phone home. With fresh settings. Which, by default, share your data. Oh and you can't control what it does while it sits there in the profile manager window. I presume it phones home. How can we be so blind and think this is a good alternative?

I know for me, it's exhaustion, and the inability to quickly scan the obfuscation mess that mozilla calls firefox's source code. What is it for you?

ephaeton··on SeL4 on RISC-V Verified to Binary Code
my reticule aims at the creation of the formal C semantics being used to verify the output. If both you and I make an off-by-one, we can always agree on the (say, index) value, but we'll both be always wrong. And C semantics is something especially finicky, given how often we delve in "undefined behaviour" territory, or our nowadays hardware does something that "back then" was unthinkable.
ephaeton··on SeL4 on RISC-V Verified to Binary Code
Quote regarding their previous work: "However, there is still the risk of security holes resulting from (...) the compiler and kernel developers interpreting the C semantics differently."

Was that really a "risk" ? I think it's a given.

"Beware of bugs in the above code; I have only proved it correct, not tried it." - Knuth

also: https://cacm.acm.org/magazines/2018/7/229036-c-is-not-a-low-...

In light of that, I'm not convinced by the current increment either.

ephaeton··on Why UML “Really” Died
We still use UML for illustrative documentation purposes leading the prose, sequence diagrams, state machines, activities. We barely ever use class diagrams. We don't draw, we let PlantUML do that. We only use UML because, well, PlantUML outputs those diagrams. If its output were non-UML sequence diagrams, state machines or activity diagrams, we'd be using that for documentation purposes.

That's why I whole-heartedly disagree with "At best you’ve got a cumbersome system that’s significantly worse than using a graphical editor" from the article. I'll take succinct text over your graphical editor any day of the week, and my colleagues agree with me.

Anyways. All hail plantuml ;)

ephaeton··on How to Learn Unix Tools
Haha! I understand the sentiment :)

But it was ok, NetBSD doesn't have zsh installed by default (and neither bash, only csh, sh and (pd? m?)ksh). When I did that, I was primed to use tcsh at $WORK. Switched to ksh later (and remained on there for a loooong time).

Anyways. I failed to make an actual point. My point was the following: The BSD manpages are _excellent_. IMO way better than the linux/GNU manpages (or info pages) of comparable utilities. It shows, again IMO, that the BSDs are a complete, single-tree distribution where documentation, userland and kernel go hand in hand and are equally well groomed.

So maybe, for learning "UNIX", go get yourself a nice little {Net,Free,Open}BSD install and toy around with the elder ways.

ephaeton··on How to Learn Unix Tools
how I learned UNIX tools:

(on a NetBSD,) alphabetically went through /bin, /usr/bin, /sbin, /usr/sbin, /usr/games and read each of the program's man pages. Yeah, took a while. Also read POSIX.2

The things you do with your free time ...

ephaeton··on Compiling a Lisp to x86_64 (2020)
what happened between 'writing a lisp' and 'compiling a lisp' that made you switch to C from OCaml?
ephaeton··on Fossil Chat
I'd rather he'd announce his MTA & MUA so we can reinstate the fossil mailing list. Going to its own quirky (at the time, haven't checked since the switch) forums meant I lost interest in being part of the (dev) community.
ephaeton··on What happens when ICE asks Google for your user information
Yeah, Law Enforcement won't request to protect users' privacy, but users may very well reply to the heads-up requesting a "broad" protection of their own privacy...
ephaeton··on What happens when ICE asks Google for your user information

    We have a well-established process for managing requests from law enforcement for data about our users: when we receive a request, we notify users that their information has been requested, push back on overly broad requests to protect users’ privacy, and provide transparency around such requests in our transparency report
As a non-native speaker of l'inglese, there's two ways my brain parses this - simultaneously:

(A) when we receive a request, we notify users (... well knowing they can't do anything ...), push back on overly broad requests to protect users' privacy (i.e., we do push back on requests that request we respect users' privacy, who do you think we are?!) and provide transparency ... ("only") in our "transparency" report

I.e., we very well sell the hide of anybody and everybody as quickly as possible, not only abiding by requests but filling in the blanks as well. Then we increment a counter "requests_served++" and include that in our transparancy "report" lol

(B) when we receive a request, we notify users (..to warn and enable them..), push back on overly broad requests IN ORDER TO protect users' privacy and provide transparency (as best as possible)

I.e., we are the good guys and actually try to serve our users as best as possible.

The sentence "we push back on overly broad requests to protect users' privacy" is at the center of this dichotomy for me. To clarify: we push back on (overly broad requests to protect users' privacy) vs. we push back on (overly broad requests) [in order] to protect users' privacy. To you native speakers: is there doubt which of the two are meant for you?

ephaeton··on Show HN: Freeciv-Web
glorious <3

Must .. resist .. temptation .. uughgggh...

ephaeton··on Buildroot – Making Embedded Linux Easy
I concur with your technical argument, except for the trust of what's being built. I have repeatedly found the shared state cache to be the achilles heel of my builds. You never want to have a false-positive hit in your cache, yet yocto has repeatedly bit me with that. I have had instances where bitbake claimed it was done, just to find wrong or missing artifacts in my image. Cleaning and rebuilding from scratch is a continuous exercise that's IME sadly necessary. ...let alone letting others use my build directories!

I accept yocto's right to exist, but I'm far from being happy for using it.

ephaeton··on Buildroot – Making Embedded Linux Easy
I don't know BalenaOS, but if it's yocto based, using this bbclass might help you out:

https://gist.github.com/weber-martin/fa9ca5bdceffc9504175f6e...

add "extra_license" to your IMAGE_CLASSES_append in your (machine, distro) config.

add EXTRA_COPY_LIC_ROOTFS = "1" to your liking.

You will end up with a directory of licenses, including license texts, of the packages you have in your image.

Yeah you will have to build your image yourself, but isn't that $trivial with yocto (FSVO $trivial ?)

ephaeton··on Buildroot – Making Embedded Linux Easy
This is true. Plus add debug bitbake, and debugging (or ignoring) broken caching, missing make semantic (delete a result artifact and see bitbake claim there's nothing to do) and generally causing a massive head-ache.

But ... it has all the hooks you need, and even if it's a particularly ugly and broken behemoth, you can slowly steer it roundabout in the direction you want it to go. yocto also has more apes hitting the typewriter, so in some distant future it might even be sensible.

My colleagues have evaluated buildroot vs. yocto several times, only to end up in yocto each time. I stopped challenging them: Once you have to hook into the build process in various places and implement your "around" hooks, yocto, while still sucking, and still standing in your way, and still pointing every gun there is at your feet ... doesn't make it impossible to achieve your goal without rewriting the whole thing.

Also you can make more money with using a complicated, shiny tool, of course.

NB: I'm using yocto professionally for three years now after a decade of experience in cross-building embedded targets using NetBSD. It is no joy (in contrast to NetBSD). If you don't have a problem that involves rewriting half of buildroot, choose buildroot :D

ephaeton··on Performance comparison: counting words in Python, Go, C++, C, Awk, Forth, Rust
Setting LC_* to 'C' nukes locale awareness by saying "we're only caring about ASCII-en-us" and thus goes back to the 'cheap' tolower that's also present in the C and C++ version, and most likely in the awk as well - in contrast to, say, the python version.

Finding the "lower case of a character" is immensely harder in face of unicode, because the table is way larger and there's no nice speed hacks by manipulating an index into the ASCII table. JFGI: "tolower performance unicode".

If you notice setting LC_ALL makes a difference in performance for you, you ought to be aware that now you're no longer comparing same capabilities.

I'm not sure how to constructively state that except for: One shouldn't compare ASCII and unicode "tolower" in performance comparisons, as you end up comparing apples and oranges (at best. More like apples and snails) - which I did.

If the author uses code that uses "tolower" in job interviews, i.e., evaluates candidates based on their input WRT case normalization - and he even writes ("This is Unicode-aware ...") - he should know that Unicode awareness is not ubiquitous, and comes at quite a cost. Knowing about unicode awareness, one would assume he'd be aware of not making a fair comparison.

ephaeton··on Performance comparison: counting words in Python, Go, C++, C, Awk, Forth, Rust
This shouldn't compare unicode-aware 'tolower' with ASCII-tolower ... sends false signals. "Optimizing" by setting LC_ to "C" .. oh my.
ephaeton··on Ask HN: Do you think Agile/Scrum is beneficial for software delivery?
The idea of the agile manifesto is great. The implementation sucks. As-is-done the value proposition of the management style is in the negative. This negative value, aka cost, is then used to support extra roles nobody needs - scrum masters, agile consultants, etc. From a, e.g., YC POV, agile is great, because you can sell useless cloud-worded services surrounding it. From a dev POV, out of the trenches, it's an annoying fad - at best.
ephaeton··on A Shell Language in Lisp
Compare that to this beauty: http://ftp.scsh.net/pub/scsh/0.6/scsh-manual.ps.gz pp. 10 onward. It's such a pity scsh is dead ...
ephaeton··on Developers spend most of their time figuring the system out
So let me get this straight: A programming environment to program programming environments in. What would be the startup cost of using this be? Hmm. Comprehending and navigating it, eh? So now, when I'm, say, doing embedded C/++ programming, I have to add yet another language to the stack, yet another toolkit, yet another environment _I have to build first_ with my preferences -- so that _another_ person getting into the project can not only use my environments, my toolkits, my code and comprehend those - but ALSO my hand built specialized for my needs programming environment? And this is supposed to _improve_ the situation?

OK, I only skimmed the site, but mostly because above question keeps nagging me and the mind boggles.

IOW, gtoolkit is today's sexy emacs (or, it claims to be what emacs claims/ed to be): Here's a toolkit for writing/extending/MOLDING your tool. Or, acknowledging it's smalltalk-heritaged, it's yet-another-smalltalk-IDE.

It's not like the "extra work" of making your product explorable disappears with glamorous toolkit, the silver bullet. gtoolkit just offers a streamlined set of APIs for you to (comprehend, and navigate, and then) use to make your product explorable, doesn't it. So we add another layer of abstraction, work, and potential for error? What was that with complexity and abstraction? Hmm.

← PreviousPage 4 of 5Next →