3 ms·

1) "It's a sign of poor application design when you have to learn it instead of merely using it." No, what this actually is a sign of is disingenuous arguing.

1) "It's a sign of poor application design when you have to learn it instead of merely using it." No, what this actually is a sign of is disingenuous arguing. I have conceded that vim is harder to learn. What I do not concede is that it is otherwise hard to use. You understand what I am saying, but being purposely difficult. 2) If vim's menus were to cover everything vim can do, then we would have ourselves a usability issue. That menu system would be absurdly deep (which is really just an argument in favor of vim if you think about it critically). 3) Blogs are not "technical documentation", which is what you said. Not to mention that any developer in the world should be able to use compat mode without reading anything. The rest of vim's feature set is large enough that it will require some type of research to uncover, no matter the presentation. Textmate can avoid this by having a dramatically smaller feature set.

phillmv 15y ago
1) It's just been my experience, when designing and discussing user interfaces, to include "how discoverable" an application is as a factor to how "usable" they are. A LOT OF FEATURES that experts can use does not necessarily make something usable. If this falls outside of the purview of what you consider to be satisfactory, that's fine by me. 2) Not really. You'd have a few context menus, and maybe an inspector window might be necessary. Obviously you're not going to represent all of the keyshortcuts but there is no good reason why I can't right click on a split and reorganize it relative to my other buffers. Or select text and hide it using whatever feature "zf" is called. Etc, etc. Maybe I have an extremely dim of the full, non plugin augmented, vim featureset but… I think it would be possible. 3) Now you're the one splitting hairs. A blog post covering a specific vim key combo… is documentation of a technical sort. C'mon now. Few people are reading that to relax after work. >The rest of vim's feature set is large enough that it will require some type of research to uncover I disagree with this. I mean, if you want to compete at vimgolf, of course you need to know the internals inside and out. But there ought to be no reason why I need to memorize five commands off the bat just to split a window. Or set a macro. Or blah blah blah blah.
- burgerbrain 15y ago
  "Few people are reading that to relax after work." The popularity of vim screencasts might suggest otherwise.

phillmv 15y ago

1) It's just been my experience, when designing and discussing user interfaces, to include "how discoverable" an application is as a factor to how "usable" they are. A LOT OF FEATURES that experts can use does not necessarily make something usable. If this falls outside of the purview of what you consider to be satisfactory, that's fine by me. 2) Not really. You'd have a few context menus, and maybe an inspector window might be necessary. Obviously you're not going to represent all of the keyshortcuts but there is no good reason why I can't right click on a split and reorganize it relative to my other buffers. Or select text and hide it using whatever feature "zf" is called. Etc, etc. Maybe I have an extremely dim of the full, non plugin augmented, vim featureset but… I think it would be possible. 3) Now you're the one splitting hairs. A blog post covering a specific vim key combo… is documentation of a technical sort. C'mon now. Few people are reading that to relax after work. >The rest of vim's feature set is large enough that it will require some type of research to uncover I disagree with this. I mean, if you want to compete at vimgolf, of course you need to know the internals inside and out. But there ought to be no reason why I need to memorize five commands off the bat just to split a window. Or set a macro. Or blah blah blah blah.

burgerbrain 15y ago
"Few people are reading that to relax after work." The popularity of vim screencasts might suggest otherwise.