netvor

joined 2 years ago
[–] netvor@lemmy.world 1 points 3 days ago (5 children)

So what you’re saying is if Bob tells me (your name) has a flying tiger, and I later find out you don’t have a flying tiger… you no longer exists?

They did not say someone named Jesus did not exist.

There's a difference between believing "Joe who had a tiger and a bear and an elephant", and assuming that there might have been 3 different Joe's, one with a tiger, one with a bear and one with an elephant, each of them in a different period. Saying "Joe with 3 animals did not exist" does not imply that those Joe's did not exist.

I'm not a historian but what I've heard (must have been on Alex O'Connor's podcast) is that even some of the possible historical Jesuses (or "Jesusi" :D) had things going on that were not compatible with what the biblical Jesus was all about. (Such as being cult leader proclaiming that world will end in few years.)

[–] netvor@lemmy.world 1 points 5 days ago (1 children)

i ... i don't know how i left you with the impression that i can't tell poo from pee. ... i must have done something horribly wrong. 🤣

Actually I think for pretty much all mammals it's as easy as telling liquid the difference between liquid and solid.

[–] netvor@lemmy.world 5 points 6 days ago* (last edited 6 days ago) (1 children)

Regarding my original comment, it’s mainly frustration at the hypocrisy some people have who make every excuse to avoid using someone’s preferred name or pronouns, but then have no problem switching when it’s someone else’s pet

yeah I got that message. it really shows the hypocrisy, and it makes me wanna scream.

I wonder if a far away vision of a world where the gender just slowly disappears from the language is really the best. (When i get asked about preferred pronoun, i feel i want to answer "i don't care and no one should, let's collectively try to really not give a f*k".) I feel like in the ideal world all pronouns would just be gender-neutral.

But language vs. gender is yet another fascinating rabbit hole. My first language is Czech, where basically every word -- even unanimous and abstract concepts like "book" have gender, and the grammar is such that effect of word "gender" spreads to other words as inflections and such. Eg. "ona spala" ~ "she slept" vs. "on spal" ~ "he slept" but "ona spal" is an obvious grammar mistake. I wonder if this makes it worse or actually better: while it makes it harder to have a gender-neutral language (the plural trick does not work: "ony spaly" ~ "they (females) slept", "oni spali" ~ "they (males) slept" ... siiigh...), I also feel it could make it less problematic in the sense that the concept of gender in language is not actually tied to identity of a person--it's just a weird thing present in the language.

Of course, none of that applies to intentional misgendering, which is just being a huge asshole, with little to no excuse.


Edit: I missed the last--the most important---part of your post, so I was just replying casually (and nerd sniping myself on the language part).

Yeah, that's really disgusting and alarming. I totally believe. I don't know what to do about it but I do believe and wish nothing but failure to these hateful, cruel people.

[–] netvor@lemmy.world 1 points 6 days ago

keeping all these containers up to date

Updates are a good way to get the security holes fixed, but unfortunately it's also often how the holes get in in the first place.

I mean, for most projects it's kind of sensible to assume that over long time, the code will become rather more secure and less buggy, so eventually the pros/cons might come out in favor of a strategy of updating every time. But it's good to know that every update is inherently a double edged sword.

That's why I like the model that distros like Debian do: they keep the code stable for long time, and only send updates for which a typically independent party (package maintainer) has already decided that a given update indeed is a necessary bugfix, or even specifically a security fix. Similar policy of course could be applied to a Docker container as well, but I don't know how many projects do this, and it would be a per-project policy, most probably not quite independent.

[–] netvor@lemmy.world 1 points 6 days ago (4 children)

unless you see them pee

You mean see/not see the "pebbles"?

Or do they tend to pee differently? I know dogs typically pee by raising one leg, but sometimes they just crouch by both back legs and just release. Is this what you mean?

I vaguely recall when I was a teen we had a dog (he/him, and a mutt) and he would usually do the "this lamp post is mine now" pee but when he really needed to go he would just crouch and leave a big puddle.

[–] netvor@lemmy.world 2 points 6 days ago

it’s probably misgendering to assume its gender at all and that the concept of gender applies to dogs

Is that true, though? How would we know? Just by Occam's razor, I would find it safer assume it applies to all mammals, ie. why would humans specifically be different, or if there is a spectrum across species, where/how would we draw the line?

[–] netvor@lemmy.world 3 points 6 days ago (1 children)

Technically it makes sense: the paper is alarming, but we should not alarm. (But yeah, why call it "alarming" in the first place? Beats me!)

[–] netvor@lemmy.world 11 points 6 days ago (4 children)

(As a trans person, I have more to say on how easily people pronoun switch for animals but that’s a different topic for another time)

It is different topic, although I'd love to hear more about that as well. Eg. as a cis male with no transgender friends (well, not that I know of), I find myself ~~thinking~~ ruminating about how impolite/distracting it is to misgender a trans person, provided one can just switch after being corrected and move on... How bad it is to make the (honest) mistake repeatedly? How is it compared to other kinds of faux-pas, like, messing up someone's name? (Eg. repeatedly calling someone John when they are Joe, or forgetting someone's occupation. These things do happen to my distractable mind that seems to love lossy data compression.)

But yeah, it's a huge, fascinating topic, but a different one from my intention in the OP. :)

 

I'm not pet owner and I don't spend much time around animals in general. (Don't get me wrong, I love them, It's just I know I would be a terrible pet owner.)

How common it is for people to see, let's say, a dog and immediately just know whether it's "she" or "him"? If two dog owners meet in a park, would it be awkward for one of them to misgender the other's dog?

Of course, I mean without looking at the "obvious" traits.

Are there behaviors that are typical for male/female pets outside the one directly related to mating?

I recognize that within the animal kingdom, the traits are not always clear, and I guess gender is quite more flexible than sex. I would be interested in both aspects.

What about cats or other animals?

[–] netvor@lemmy.world 2 points 6 days ago (1 children)

Is her head still empty if someone says "park"?

[–] netvor@lemmy.world 1 points 6 days ago

what do you mean? jq is great!

[–] netvor@lemmy.world 1 points 6 days ago (1 children)

Don't want to sound unappreciative, but the apps you refer to (and others in these threads) are not actually CLI's but TUI's.

  • CLI (command line interface) is when all interaction actually happens in the command line, ie. command + arguments. CLI's are much simpler to implement, have little dependencies (pretty much just argument list, two data streams---stdout and stderr--and exit status) and typically one invocation means one independent task. All this makes CLI's ideal as building building blocks of (semi-)automated workflows, but many CLI's are also optimized for direct invocation from interactive shell, eg. by adding features such as output coloring, interactive yes/no steps or command completion (although that part is actually driven by the shell, and is quite independent from the execution of the app.)

  • TUI (text user interface, i think) on the other hand, is more like GUI but replicated within the confines of terminal emulator. The interaction heavily depends on terminal features such as moving cursor, resize notifications, etc. Also when TUI is ran, it's normally used for zero to may tasks: e.g. I could start htop and investigate no process, 1 process or many, before quitting. Unlike CLI's, TUI's pretty much make no sense within automation.

Don't get me wrong: I love TUI's (htop is one of my favorite and thanks for recommending nmon, i'll have a look)--and often prefer them to GUI's (eg. my text editor is nvim, which is a TUI app!), but in my post I was specifically interested in exploring CLI's. I would actually love a similar post to mine but focusing explicitly on TUI's as opposed to CLI's.

Sorry for long post -- I hope it can kind of serve as explanation for people who are new to this and stumble upon this thread and aren't quite familiar with the distinction.

[–] netvor@lemmy.world 5 points 2 weeks ago

Did not want to respond but this is hilarious.

To be fair, really old CLI's, like from the time when X.org was the new stuff, this style used to be more common. That was before "GNU style" (using single dash for single-letter bundle-able options and double dash for long options) became prevalent.

But yeah, if you see -foo then you know the program is old enough that regular colonoscopy is recommended, and the original author is probably retired or "passed away at the ripe age of ..".

62
submitted 2 weeks ago* (last edited 2 weeks ago) by netvor@lemmy.world to c/linux@lemmy.ml
 

Regardless of what the app does and whether the thing that does is particularly useful, powerful or important for what you need to do (or even well implemented), what is a command-line interface that you had a particularly good experience both learning and working with?

In other words, I'm thinking about command line interface design patterns that tend to correlate with good user experience.

"Good user experience" being vague, what I mean is, including (but not limited to)

  • discoverability--learning what features are available),
  • usability--those features actually being useful,
  • and expressiveness--being able to do more with less words without losing clarity,

but if there's a CLI that has none of those but you still like it, I'd be happy to hear about it.

Edit: Trying to stress more that this post is not about the functionality behind the tool. Looks like most of first responders missed the nuance: whether app x is better than app y because it does x1 ad x2 differently or better does not matter; I'm purely interested in how the command line interface is designed (short/long flags, sub-commands, verbs, nouns, output behaviors)..

 

I'm not talking about #metoo stuff.

Assuming we're using a sane IM app that lets us use standard emoji as a reaction to a comment.

I just mean when someone tells you that they did something and the only information that you want to convey back is that you followed them and did the same.

More concrete example that i encounter few times a week: my pal tells me he's reserved the gym slot at 17:00, so I reserve that time too and i just want to reply "me too" to let him know it worked for me. (The way it works between us is that we want to go together but since he's got family etc. i always let him pick the time first and just announce his slot reservation, and i just want to confirm it.)

Another example is if someone says "i signed this petition to support our Czech president who is protecting our democracy" then I want to say "yes I did it too". Just a nod that we're standing on the same side.

(Come to think about it, it could technically be used in a #metoo adjacent context, like, if a friend told me "i was abused by my boss" and i could just use the emoji but that would be sooooo wrong on so many levels... You get the point.)

 

At some point (in a galaxy, long long time ago) I learned to turn jeans, shorts and hoodies inside out when putting them into washing machine. For some reason, I don't do the same with smaller, simple things like T-shirts and underwear.

I forgot what was the reasoning behind it, but when I think about it, can't seem to come to a conclusion which way is better.

Is one way better than another and why?

 

Look, I'm a Debian user for 15 years, I've worked in F/OSS for a long time, can take care of myself.

But I'm always on a lookout for distros that might be good fit for other people in my non-tech vicinity, like siblings, nieces, nephews... I'm imagining some distro which is easy for gaming but can also be used for normal school, work, etc. related stuff. And yeah, also not too painful to maintain.

(Well, less painful than Windows which honestly is not a high bar nowadays... but don't listen to me, all tried in past years was to install Minecraft from the MS store... The wound is still healing.)

I have Steam Deck and I like how it works: gaming first, desktop easily accessible. But I only really use it for gaming.

So I learned about Bazzite, but from their description on their main site I'm not very wise:

The next generation of Linux gaming [Powered by Fedora and Universal Blue] Bazzite is a cloud native image built upon Fedora Atomic Desktops that brings the best of Linux gaming to all of your devices - including your favorite handheld.

Filtering out the buzzwords, "cloud native image" stands out to me, but that's weird, doesn't it mean that I'll be running my system on someone else's computer?

Funnily enough, I scrolled a bit and there's a news section with a perfectly titled article: "WTF is Cloud Native and what is all this".

But that just leads to some announcements of someone (apparently important in the community) talking about some superb community milestone and being funny about his dog. To be fair, despite the title, the announcement is not directed towards people like me, it's more towards the community, who obviously already knows.

Amongst the cruft, the most "relevant" part seems to be this:

This is the simplest definition of cloud native: One common way to linux, based around container technology. Server on any cloud provider, bare metal, a desktop, an HTPC, a handheld, and your gaming rig. It’s all the same thing, Linux.

But wait, all I want to run is a "normal" PC with a Linux distro. I don't necessarily need it to be a "traditional" distro but what I don't want is to have it running, or heavily integrated in some proprietary-ish cloud.

So how does this work? Am I missing something?

(Or are my red flags real: that all of this is just to make a lot of promises and get some VC-funding?)

1
submitted 2 years ago* (last edited 2 years ago) by netvor@lemmy.world to c/debian@lemmy.world
 

I'm trying to "build" (see below) a package for another architecture. I made it through (by disabling, frankly) most of the steps.

Long story short, I end up running something like this:

debuild -us -uc --host-type riscv64-linux-gnu -d -C/dev/null

but debuild keeps failing on this line:

[...]
make[1]: Leaving directory '/home/netvor/.cache/mkittool/debstuff/build/zigdev-0.0.0+t20240906145412.egg.gbc271d0'
   dh_shlibdeps -a
   dh_installdeb
   dh_gencontrol
   dh_md5sums
   dh_builddeb
dpkg-deb: building package 'zigdev' in '../zigdev_0.0.0+t20240906145412.egg.gbc271d0-1_riscv64.deb'.
 dpkg-genbuildinfo -O../zigdev_0.0.0+t20240906145412.egg.gbc271d0-1_riscv64.buildinfo
 dpkg-genchanges -C/dev/null -O../zigdev_0.0.0+t20240906145412.egg.gbc271d0-1_riscv64.changes
dpkg-genchanges: info: including full source code in upload
 dpkg-source --after-build .
dpkg-buildpackage: info: full upload (original source is included)
debuild: fatal error at line 1062:
can't open zigdev_0.0.0+t20240906145412.egg.gbc271d0-1_amd64.changes for reading: No such file or directory

So the *_amd64.changes file does not exist, but *_riscv64.changes does:

zigdev-0.0.0+t20240906145412.egg.gbc271d0
zigdev_0.0.0+t20240906145412.egg.gbc271d0-1.debian.tar.xz
zigdev_0.0.0+t20240906145412.egg.gbc271d0-1.dsc
zigdev_0.0.0+t20240906145412.egg.gbc271d0-1_amd64.build
zigdev_0.0.0+t20240906145412.egg.gbc271d0-1_riscv64.buildinfo
zigdev_0.0.0+t20240906145412.egg.gbc271d0-1_riscv64.changes
zigdev_0.0.0+t20240906145412.egg.gbc271d0-1_riscv64.deb
zigdev_0.0.0+t20240906145412.egg.gbc271d0.orig.tar.gz

Building with amd64 architecture finishes correctly *_amd64.changes exists and is used.

First, do I really need this .changes file? (I'm not planning to upload this to Debian archive.) And if so, how to make debuild use the correct file?

The environment (when calling env inside rules file) looks like this:

ASFLAGS=
CFLAGS=-g -O2 -ffile-prefix-map=/home/netvor/.cache/mkittool/debstuff/build/zigdev-0.0.0+t20240906145412.egg.gbc271d0=. -fstack-protector-strong -Wformat -Werror=format-security
CPPFLAGS=-Wdate-time -D_FORTIFY_SOURCE=2
CXXFLAGS=-g -O2 -ffile-prefix-map=/home/netvor/.cache/mkittool/debstuff/build/zigdev-0.0.0+t20240906145412.egg.gbc271d0=. -fstack-protector-strong -Wformat -Werror=format-security
DBUS_SESSION_BUS_ADDRESS=unix:path=/run/user/11111/bus
DEBEMAIL=Me <netvor@example.com>
DEB_BUILD_ARCH=amd64
DEB_BUILD_ARCH_ABI=base
DEB_BUILD_ARCH_BITS=64
DEB_BUILD_ARCH_CPU=amd64
DEB_BUILD_ARCH_ENDIAN=little
DEB_BUILD_ARCH_LIBC=gnu
DEB_BUILD_ARCH_OS=linux
DEB_BUILD_GNU_CPU=x86_64
DEB_BUILD_GNU_SYSTEM=linux-gnu
DEB_BUILD_GNU_TYPE=x86_64-linux-gnu
DEB_BUILD_MULTIARCH=x86_64-linux-gnu
DEB_BUILD_OPTIONS=notest parallel=8
DEB_HOST_ARCH=riscv64
DEB_HOST_ARCH_ABI=base
DEB_HOST_ARCH_BITS=64
DEB_HOST_ARCH_CPU=riscv64
DEB_HOST_ARCH_ENDIAN=little
DEB_HOST_ARCH_LIBC=gnu
DEB_HOST_ARCH_OS=linux
DEB_HOST_GNU_CPU=riscv64
DEB_HOST_GNU_SYSTEM=linux-gnu
DEB_HOST_GNU_TYPE=riscv64-linux-gnu
DEB_HOST_MULTIARCH=riscv64-linux-gnu
DEB_RULES_REQUIRES_ROOT=binary-targets
DEB_TARGET_ARCH=riscv64
DEB_TARGET_ARCH_ABI=base
DEB_TARGET_ARCH_BITS=64
DEB_TARGET_ARCH_CPU=riscv64
DEB_TARGET_ARCH_ENDIAN=little
DEB_TARGET_ARCH_LIBC=gnu
DEB_TARGET_ARCH_OS=linux
DEB_TARGET_GNU_CPU=riscv64
DEB_TARGET_GNU_SYSTEM=linux-gnu
DEB_TARGET_GNU_TYPE=riscv64-linux-gnu
DEB_TARGET_MULTIARCH=riscv64-linux-gnu
DFLAGS=-frelease
DH_INTERNAL_BUILDFLAGS=1
DH_INTERNAL_OPTIONS=
DH_INTERNAL_OVERRIDE=dh_auto_install
FAKED_MODE=unknown-is-root
FAKEROOTKEY=2071757222
FCFLAGS=-g -O2 -ffile-prefix-map=/home/netvor/.cache/mkittool/debstuff/build/zigdev-0.0.0+t20240906145412.egg.gbc271d0=. -fstack-protector-strong
FFLAGS=-g -O2 -ffile-prefix-map=/home/netvor/.cache/mkittool/debstuff/build/zigdev-0.0.0+t20240906145412.egg.gbc271d0=. -fstack-protector-strong
GCJFLAGS=-g -O2 -ffile-prefix-map=/home/netvor/.cache/mkittool/debstuff/build/zigdev-0.0.0+t20240906145412.egg.gbc271d0=. -fstack-protector-strong
GPG_AGENT_INFO=/run/user/11111/gnupg/S.gpg-agent:0:1
HOME=/home/netvor
LANG=en_US.UTF-8
LC_COLLATE=C
LDFLAGS=-Wl,-z,relro
LD_LIBRARY_PATH=/usr/lib/x86_64-linux-gnu/libfakeroot:/usr/lib64/libfakeroot:/usr/lib32/libfakeroot
LD_PRELOAD=libfakeroot-sysv.so
LOGNAME=netvor
MAKEFLAGS=w
MAKELEVEL=2
MFLAGS=-w
OBJCFLAGS=-g -O2 -ffile-prefix-map=/home/netvor/.cache/mkittool/debstuff/build/zigdev-0.0.0+t20240906145412.egg.gbc271d0=. -fstack-protector-strong -Wformat -Werror=format-security
OBJCXXFLAGS=-g -O2 -ffile-prefix-map=/home/netvor/.cache/mkittool/debstuff/build/zigdev-0.0.0+t20240906145412.egg.gbc271d0=. -fstack-protector-strong -Wformat -Werror=format-security
PATH=/usr/sbin:/usr/bin:/sbin:/bin:/usr/bin/X11
PREFIX=/usr
PWD=/home/netvor/.cache/mkittool/debstuff/build/zigdev-0.0.0+t20240906145412.egg.gbc271d0
SOURCE_DATE_EPOCH=1456533483
TERM=rxvt-unicode
ZIGDEV_ZIG_VERSION=0.13.0
ZIGSITE=/opt/zigdev
ZIGSITE_PREP=debian/tmp/opt/zigdev

Open "spoiler" below to read more about my goals. Since the fact I don't actually want to build Zig properly here might confuse and annoy people, I wrote my reasoning below.

Project overview

First and foremost, I want to learn more and become more familiar with Debian build system as well as Zig and system-level programming.

How I want to do it is to start creating zig-based binary packages for personal/experimental use. Now, already have a pipeline and tooling ecosystem which I use for Python and Bash packages: my system is DEB centric and handles package lifecycle from git repo to APT (or DNF, really) repository and I prefer when any new project can be immediately built and deployed as .deb.

So now I want to add Zig support. But means my Zig-based projects will need something to put to Build-Depends, and since Zig does not officially provide APT repo, I want to create my own -- this is what I'm focusing on right now.

So I'm creating this hacky package called zigdev whose only purpose will be to exist in my internal APT repos and deploy /opt/zigdev/zig to my test machines. One day, this package will can be easily replaced by official zig package, so for now (while building this particular zigdev package), I'm trying to cut every corner I can:

  • I don't actually build Zig, I just download tarball using curl.

  • I'm trying to disable every truly arch-specific step, since these would typically need arch-specific chroot or similar setup.

    For example, I don't care about dynamic linking, stripping or reproducibility.

Once I get this zigdev package running, I can start building my hello_world.zig's and similar. At that point I will start slowly moving towards creating a more proper binary packages by refining an rules template for my zig projects (using zig tooling, though.) (All this while also learning Zig itself and system-level programming in general, of which I have almost no experience with, so that will move with glacial speed.)

view more: next ›