@dcz@fosstodon.org
@dcz@fosstodon.org avatar

dcz

@dcz@fosstodon.org

Greybeard but revolutionary. 80-char lines are good for museums.

My tech writing (hire me): https://dorotac.eu/
Open bike computer: https://jazda.org
dcz on libera.chat

This profile is from a federated server and may be incomplete. Browse more on the original instance.

deshipu, to random
@deshipu@fosstodon.org avatar

Remember that automation is not about removing effort, it's all about packaging it in convenient ways so that it can be stored and moved around. Sometimes you get some savings due to bunching up of similar things or economy of scale, but most of the time the perceived savings are simply because the real effort is hidden away or deferred. Whether it's Jacquard looms, language models, github actions, self-driving cars, or robot food delivery, someone is doing (or has done, or will do) the work.

dcz,
@dcz@fosstodon.org avatar

@deshipu It is about removing effort. In the case of jacquard loom, someone put in the effort to produce the cards, and the next person doesn't have to do it by hand. And the next one. And the next one. The effort of N people is replaced by 1.

I'm not sure if the others are breaking even, but with time I bet they will.

dcz,
@dcz@fosstodon.org avatar

@deshipu Of course. My bet is that this new effort is typically smaller than the old effort was. Have you tried to sew by hand?
Maybe not automation by your definition, but the sewing machine saves hours and hours of work. If I was serious about sewing, I'd build one just to save my own time.

In another example, using machines for farming made it so that there's less than 90% of the population dedicated to food production.

dcz,
@dcz@fosstodon.org avatar

@deshipu You're completely right. Entropy ensures that no matter what happens, it happens because of some energy being turned into heat.

But I find this definition useless for considering social situations. Here what matters is the subjective effort expended by agents.

dcz,
@dcz@fosstodon.org avatar

@deshipu I agree with the above post, but you're also making a very bold claim I disagree with: that there's some "effort preservation" law.

I think that's an extraordinary claim and it requires extraordinary evidence.

As a basic example, we have many different sources of electrical energy available, with different costs and same labor production: coal, solar, nuclear, water. I'm not going to believe so easily that the ultimate effort per unit of benefit of each is the same.

dcz, to UX
@dcz@fosstodon.org avatar

One of my talks for got accepted!

I'll talk about how "My files are a mess!", and how it will stay this way unless we achieve a collective enlightenment.

https://cfp.gulas.ch/gpn22/talk/ZRKD3G/

Now I finally have an excuse to do the research on the topic XD

dcz,
@dcz@fosstodon.org avatar

@publicvoit Reading a bit more of your work (like application-centric systems), I definitely want to have a chat with you. Are you visiting Germany any time soon?

yakkoj, to random
@yakkoj@fosstodon.org avatar

using "git pull origin main || git pull origin master" to update git repos

dcz,
@dcz@fosstodon.org avatar

@yakkoj I see you still haven't seen a repo with both branches present.

dantleech, (edited ) to random
@dantleech@fosstodon.org avatar

Emotional rollercoaster of starting new projects:

  • This is cool
  • I've spent 18 hours on this
  • This is awful
  • I'm awful
  • The world is awful
  • This is cool (goto 1)
  • ... weeks pass_
  • Wow this is great.
  • Almost done
  • lose interest
  • Project abandonned
dcz,
@dcz@fosstodon.org avatar

@dantleech Too real...

dcz, to hacking
@dcz@fosstodon.org avatar

Hey Fediversians, is anyone going to #GPN https://gulas.ch ? I'm looking for roommates today.

#conference #hacking #germany

thelinuxcast, to random
@thelinuxcast@fosstodon.org avatar

Is this what the Nextcloud app for Linux is supposed to look like?

dcz,
@dcz@fosstodon.org avatar

@thelinuxcast I hate this, and this keeps happening.

At least I can fix by manually tweaking the style (but why provide "system" if it doesn't work?)

dcz, to opensource
@dcz@fosstodon.org avatar

The feeling when 87.5% of the discussion under your contribution is about #bureaucracy : unreliable CI and insignificant #style #bikeshedding .

Congrats, the project has created a barrier to entry for anyone who hasn't been into it for years already.

Hint: give everyone a freaking #linter that satisfies your needs and let us move on to actual work, without having to redo the same stuff 8 times. Sheesh.

#opensource #coding #nitpicking #contributing

dcz,
@dcz@fosstodon.org avatar

@deshipu Good code formatting by my definition is that which makes people not have to think about it - that is, automated.

Intention does not need to be reflected in style, but in the code itself, and in comments.

dcz,
@dcz@fosstodon.org avatar

@deshipu If all quality of the code is in the style, does that mean that it doesn't matter what the code actually does? Nope, that's silly and not what you mean. The ultimate quality signal is what the code does, plus documentation.

For example, the GNU C syntax makes my eyes bleed, but it can still be amazing quality.

"What" and "why" are important for understanding. "What" is in the execution of the code. "Why" is in the documentation and naming. What knowledge is conveyed by altering style?

dcz,
@dcz@fosstodon.org avatar

@deshipu If what you care is effort spent writing then sure, but it's not the same as seeking quality.

If I wanted to filter by effort, I'd rather have people write poems as a requirement, because that at least can be fun.
Style is similarly independent of actual quality.

I don't see how it helps understanding any more than comments, despite all the problems with comments.

Can you explain what knowledge can be conveyed with style that gets erased when linting?

dcz,
@dcz@fosstodon.org avatar

@deshipu Well, you mentioned effoet, but effort is not knowledge. Then you repeated "You can convey a lot of knowledge" but you skipped what exactly.

Sorry if you feel ignored, I am genuinely curious.

dcz,
@dcz@fosstodon.org avatar

@deshipu Well, i run it through a test suite, and read it carefully for architectural descriptions, and make sure that naming is fine.

If I have to reverse-engineer it to understand, then I take off quality points.

If I can't read it due to formatting, I guess I'd run it through a linter, but I don't think it would affect my impression of quality. Hence - I can't really imagine what knowledge I'd lose.

dcz,
@dcz@fosstodon.org avatar

@deshipu No, but I don't see how formatting it with a linter would make it less possible to understand. Therefore, style doesn't carry any knowledge.

dcz,
@dcz@fosstodon.org avatar

@deshipu The semantics: actual code, comments and variable names. Those are preserved by the linters I know.

dcz,
@dcz@fosstodon.org avatar

@deshipu You can use a tool that removes things that carry knowledge just fine. Just to make it clear, I don't consider comments or names part of style (except whether you use CamelCase or whatever).

But I still don't see what you see: the knowledge thats not in the semantics or naming or comments.

dcz,
@dcz@fosstodon.org avatar

@deshipu I agree with you on variable names. But there's a lot more to linting: spacing, indentation, the placement of brackets. Do you think those carry knowledge?

No linters I know touch names in a meaningful way.

dcz,
@dcz@fosstodon.org avatar

@deshipu If the code has inconsistent indentation, then I let the linter fix it :P

I can see your point: the more sources the author uses, the less likely they are well-integrated, and the more scrutiny they require.

I guess that's a valid point.

In the context of my original complaint, the syntax complaints didn't really lead to more feedback about the content, so I don't think those people were following your observation, really :P Now the question is: what are they trying to achieve.

thelinuxcast, to random
@thelinuxcast@fosstodon.org avatar

sshfs is far superior to nfs, imo. It is definitely easier to set up and it doesn't freeze when transferring large files.

dcz,
@dcz@fosstodon.org avatar

@thelinuxcast How do you make it not freeze XD?

It doesn't freeze permanently, but it introduces such lags I have plans for writing a replacement.

dcz,
@dcz@fosstodon.org avatar

@thelinuxcast Not good enough for me, I want to take a look at the contents of other directories while I'm moving stuff around.

I think the SSH protocol is the bottleneck, and I'm surprised that NFS has that problem. Wasn't it running over UDP?

dcz, to Bash
@dcz@fosstodon.org avatar

with which I surprised greybeards at the local hackerspace (which was surprising cause I'm the opposite of a expert):

mv photo{tocheck,supercool}.jpg

is the same as

mv phototocheck.jpg photosupercool.jpg

The shell copies an argument with {,} inside, and each resulting argument has a different part of what's in {}.

rm plan.{md,svg,png,odt}

Have fun!

dcz, to rust
@dcz@fosstodon.org avatar

A part of that doesn't get mentioned but is a ridiculous quality of life boon:

dbg!()

Just #[derive(Debug)] on a struct and print it. Suddenly you get insight into what's going on. Very rarely do you need to write your own print.

Insert a dbg in the middle of anything

let diffi = debug!(diffs).iter();

and you get a message with line number and the name of variable:

[src/lib.rs:192:19] diffs = [16, 86, 11]

Not even has that!

  • All
  • Subscribed
  • Moderated
  • Favorites
  • •
  • JUstTest
  • kavyap
  • DreamBathrooms
  • thenastyranch
  • magazineikmin
  • tacticalgear
  • khanakhh
  • Youngstown
  • mdbf
  • slotface
  • rosin
  • everett
  • ngwrru68w68
  • Durango
  • megavids
  • InstantRegret
  • cubers
  • GTA5RPClips
  • cisconetworking
  • ethstaker
  • osvaldo12
  • modclub
  • normalnudes
  • provamag3
  • tester
  • anitta
  • Leos
  • lostlight
  • All magazines
    •