Part of the reason for a command line is being able to reproduce commands exactly. Going back through the command history is an important part of that.
A GUI with good usability can let you repeat commands exactly if required. They use last used values as default. If people in needed that often we’d see more of it in GUI apps. There is often more useful functionality that get prioritised though.
It is done, but it comes in specifically packaged form for each use case.
Postman is essentially a GUI app for reproducible curl commands. There’s nothing in Postman that couldn’t be done by a good shell script, and it will spurt out a curl command if you want. To get this in a shell environment, you need to have the knowledge to put together tools that are sitting right there. To get it in a GUI environment, you need a team of people who probably know how to do it in a CLI, but then create a complex memory hog GUI to do it otherwise. None of that development effort is directly transferable to any other problem space.
That’s not necessarily so. There are all sorts of legacy reasons people give for making poor software. From lazy monopolies to programmers with little understanding of usability. To people without the big picture. It will change.
Sadly, big business, techies without imagination and community FOSS without enough capacity are the ones that control what is available. Nothing will suddenly change. Usability is way down the priority list.
For a while I was using a VPN where the connections we all via command line. I loved it, because once you type it, it’s just up up up to find that location again and swap around.
Now I have to use the mouse and click like a damn fool!
In my first professional programming job writing custom software for clients in 1995, one of our standard sales pitches to clients was the idea that a GUI-based application would do away with the need for command prompts. This was always met with applause and great rejoicing. It’s kind of remarkable that command prompts are still going strong thirty years later. I’m sure nobody would appreciate having this phenomenon compared to the Amish so I won’t do it. But I think it’s pretty cool that the Amish are still around doing their thing.
A GUI makes sense if you want things to be approachable. A CLI makes sense, if you want things to be well customizable.
E.g. compare LaTeX script with setting formatting in an office program. Yes i see directly how the formatting is. And if a good Formatting standard is set, it can be used quite well nowadays. But if just want to put one Paragraph differently in order to fit a picture or table or something then things quickly get messy.
I thought command line users like typing things. I avoid typing where possible, and dont use the command line on Linux.
Part of the reason for a command line is being able to reproduce commands exactly. Going back through the command history is an important part of that.
A GUI with good usability can let you repeat commands exactly if required. They use last used values as default. If people in needed that often we’d see more of it in GUI apps. There is often more useful functionality that get prioritised though.
It is done, but it comes in specifically packaged form for each use case.
Postman is essentially a GUI app for reproducible curl commands. There’s nothing in Postman that couldn’t be done by a good shell script, and it will spurt out a curl command if you want. To get this in a shell environment, you need to have the knowledge to put together tools that are sitting right there. To get it in a GUI environment, you need a team of people who probably know how to do it in a CLI, but then create a complex memory hog GUI to do it otherwise. None of that development effort is directly transferable to any other problem space.
That’s not necessarily so. There are all sorts of legacy reasons people give for making poor software. From lazy monopolies to programmers with little understanding of usability. To people without the big picture. It will change.
Why would it suddenly change now, decades after the first GUI operating systems were available to average people?
Sadly, big business, techies without imagination and community FOSS without enough capacity are the ones that control what is available. Nothing will suddenly change. Usability is way down the priority list.
For a while I was using a VPN where the connections we all via command line. I loved it, because once you type it, it’s just up up up to find that location again and swap around.
Now I have to use the mouse and click like a damn fool!
I work as a software developer. My team mate once said he was a “mouse and click kinda guy” when asked why he didn’t use the terminal for git.
Ugh, I bet he never pressed CTRL + C or P in his life.
In my first professional programming job writing custom software for clients in 1995, one of our standard sales pitches to clients was the idea that a GUI-based application would do away with the need for command prompts. This was always met with applause and great rejoicing. It’s kind of remarkable that command prompts are still going strong thirty years later. I’m sure nobody would appreciate having this phenomenon compared to the Amish so I won’t do it. But I think it’s pretty cool that the Amish are still around doing their thing.
GUI is great, unless you want to automate something… Then you either want a CLI or API
Nah, companies found solution to that long ago.
…
…HOW’S IT GOING JOHNY, GOT THAT LIST PARSED YET?
A GUI makes sense if you want things to be approachable. A CLI makes sense, if you want things to be well customizable.
E.g. compare LaTeX script with setting formatting in an office program. Yes i see directly how the formatting is. And if a good Formatting standard is set, it can be used quite well nowadays. But if just want to put one Paragraph differently in order to fit a picture or table or something then things quickly get messy.
People do not just point and nod. Using words and not just guestures makes life easier in many, but not all situations.