Live data from Hacker News

The Renaissance of the Shell?

effective-shell.com

61–69 of 69 posts

Re: The Renaissance of the Shell?

#61

Earlier quoted context omitted.

If it had meant the oil company it would have omitted the definite article: "The Renaissance of Shell". Or perhaps even included the full name: "The Renaissance of Royal Dutch Shell".

Thank you, my native language lacks articles.

Ah, in that case yours is a perfectly understandable interpretation. May I ask which language?

Re: The Renaissance of the Shell?

#62
post #56
post #48

Earlier quoted context omitted.

ls -al rm -r. /* git checkout -b thisIsFine Couldn't give you a sed or awk example, because, fortunately, did not have to learn either yet.

The fact that some CLIs have terrible UX does not invalidate the idea of CLIs.

It does invalidate the parable though.

Re: The Renaissance of the Shell?

#63
post #62
post #56

Earlier quoted context omitted.

The fact that some CLIs have terrible UX does not invalidate the idea of CLIs.

It does invalidate the parable though.

How so?

I read the parable as being about the infinite number of ways icons and GUIs can be interpreted, thus making their inherent discoverability and clarity low.

I don't see how listing CLI commands with terrible UX invalidates that point.

Re: The Renaissance of the Shell?

#64

Most of ones work should be in the shell all the time, because automation tasks are every day phenomena (or should be). Otherwise, if its GUI oriented, you will still have to learn CLI variant sooner or later and not only its harder or almost impossible to automate, but such automation is usually flaky as well. Shell should be considered basic and most important interaction with computer with anything else deemed as…

> Otherwise, if its GUI oriented, you will still have to learn CLI variant sooner or later and not only its harder or almost impossible to automate, but such automation is usually flaky as well. Is it? Or is it that us compsci and computer engineering folks aren't used to it and don't know much about it? A Japanese instructional YouTuber I watch did a tour of his setup, and in it he shows off a lot of automation he u…

I don't assume. You assume stuff about me. :)

The GUI is heterogeneous world. Linux doesn't have tools AFAIK work mentioning because it didn't need them due to CLI. Tools like xdotool are very limited.

Windows had much better tools, such as AutoIt, AHK etc. This was because they all were created prior to PowerShell when there was no other option that encompasses all apps. And PowerShell doesn't deal with GUI too, although you could force it.

However, big part of those scripts depend on current theme, window sizing, even positions. To create good script is not easy.

Re: The Renaissance of the Shell?

#66
post #63
post #62

Earlier quoted context omitted.

It does invalidate the parable though.

How so? I read the parable as being about the infinite number of ways icons and GUIs can be interpreted, thus making their inherent discoverability and clarity low. I don't see how listing CLI commands with terrible UX invalidates that point.

Because there's really no difference between terrible command line argument names and terrible icons for the same commands.

Re: The Renaissance of the Shell?

#67
post #66
post #63

Earlier quoted context omitted.

How so? I read the parable as being about the infinite number of ways icons and GUIs can be interpreted, thus making their inherent discoverability and clarity low. I don't see how listing CLI commands with terrible UX invalidates that point.

Because there's really no difference between terrible command line argument names and terrible icons for the same commands.

That may be true, but I think the parable is about well-chosen cases, not the bad ones.

Icons don't have an intrinsic meaning the way words do, so some percentage of people will find brand-new interpretations for icons, to a larger extent than will happen with words.

That was my interpretation, at least.

Re: The Renaissance of the Shell?

#68
post #67
post #66

Earlier quoted context omitted.

Because there's really no difference between terrible command line argument names and terrible icons for the same commands.

That may be true, but I think the parable is about well-chosen cases, not the bad ones. Icons don't have an intrinsic meaning the way words do, so some percentage of people will find brand-new interpretations for icons, to a larger extent than will happen with words. That was my interpretation, at least.

> Icons don't have an intrinsic meaning the way words do

This is completely backwards. Icons can have intrinsic meanings - they can literally depict an object or an action from the real world. Words are pure abstracts humans made up, and can not exist outside of the context.

Re: The Renaissance of the Shell?

#69

Earlier quoted context omitted.

Ah, in that case yours is a perfectly understandable interpretation. May I ask which language?

Russian

Thank you. Now another question: how would you make the same distinction in Russian? I mean an explanation of it not just the actual words, I had three weeks of Russian in high school nearly fifty years ago but none of it has stuck.
Post reply on HN