Cannot get this to work. Installed Alfred workflow, but on triggering – 2nd Alfred option tells me I need to enable permissions. I have already done so under Acccessibility, but no Wooshy option is presented under Screen recording, nor is Alfred for that matter. Any suggestions?
currently a license is not needed, until it’s in alpha/beta. if you want to support though, you can from within the app, the License Pane. there’s a link there that sends you to the subscription page. thank you!
any chance you’ve skipped the Screen Recording the first time? in macOS it’s possible to ask the Accessibility permissions multiple times, but not the Screen Recording. if you denied one time, or if it went unnoticed, then it’ll not show up again. even if we try to prompt it programmatically.
see this: How do I reset Screen Recording permission on macOS Catalina? - Ask Different
the bundle for Wooshy is mo.com.sleeplessmind.Wooshy
but that’s a good point, thanks for reporting. i’ll add some precisions in the README.
EDIT: you can of course add the Workflow or Alfred yourself in the Screen Permissions without having to wait or need the macOS dialog.
Edge support added to a11 coming later today. but eww. i’m on Ventura that detects stuff install in the background and Edge is installing several stuff without even asking or letting the user decides. eww eww eww. Microsoft doing Microsoft stuff again. the implicit rule is to ask for users at install, and/or let them switch on stuff in the Preferences. eww.
Haven’t tried wooshy.
I’m using Superkey which works well.
I like that it doesn’t overlay letters on the screen like shortcat does.
It seems wooshy does the same, which is good.
sorry for the late answer, i can only write three posts it seems, and wait for someone to answer.
- buttons that do not have labels may still have metadata. currently the way to check this is to use the Accessibility Inspector that comes with Xcode. before the beta there’ll be a feature within Wooshy.
- yeah, i see what you mean. personally i use Wooshy combined with kindaVim and reach the top target with
ggor the last withG, and navigate if needed withjk. if you don’t have kindaVim, you can use the standard macOS keyboard shortcuts (up, down, command up, command down, control up, control down).
the hints are not in the roadmap, deliberately. maybe Shortcat is a better match. the new version has both the typing AND the hints.
1 Like
Superkey and Wooshy/Homerow/Shortcat are pretty different actually. Superkey uses OCR and only recognizes text. in Wooshy and the others, you can ask for “checkboxes” (or you can just type “check” or maybe only “ch” depending on the context) and it will show you all the checkboxes available. Wooshy (and the others) also reach icons, images, stuff without text, etc. Superkey doesn’t. but Superkey can reach stuff that lack Accessibility data. Superkey is also quite slower.
overall they fit quite different needs.
1 Like
yes, Shortcat is back from the dead 
it’s getting maintained more actively, and it changed quite a bit. it has now searching through text + the hints. the way Wooshy and Shortcat will diverge is in their philosophy. Wooshy intends to use as little screen estate as possible. it’s also for typists. personally i don’t do well with hints that i have to look at, and decipher. i work much better by typing what comes in my brain. hence the current features of Wooshy.
1 Like
but don’t get me wrong, Superkey is great. it’s just that those apps are for different uses. some stuff will be caught by one but not the others and vice versa. for example Superkey grabs text in PDFs, while Wooshy/Shortcat/Homerow will not.
you can think of it as Superkey being really about what you see on the screen (it is OCR), while Wooshy/Shortcat/Homerow would like some sort of AI. you type “button” and all the buttons get targeted, even if there’s no text “button” on the screen.
yes, Shortcat is back from the dead
it’s getting maintained more actively, and it changed quite a bit. it has now searching through text + the hints. the way Wooshy and Shortcat will diverge is in their philosophy. Wooshy intends to use as little screen estate as possible. it’s also for typists. personally i don’t do well with hints that i have to look at, and decipher. i work much better by typing what comes in my brain. hence the current features of Wooshy.
(replied to the wrong person due to the 3 posts limit, sorry 
)
1 Like
I like this intention. I get overwhelmed when I see hundreds of shortcut labels, and it takes too much attention for me to find/read/input those shortcuts. I also like the tiny (although adjustable) input window.
FWIW, I do find myself looking for a “match count” as a reliable/stationary indicator of number of matches as the input is refined. As opposed to relying solely on scanning for colors/color changes.
interesting. i never felt the need myself. i personally don’t even look at The Input. i just type, i stare at the place i wanna reach, and somehow i feel a sense of when i’m close to reaching my target. then i may navigate a bit with jk or G, usually. but i could add a count as an option, like at the same location as the loading indicator, once that one is gone. will have a look. thanks for the idea!
1 Like
narrowers will be done in alpha 12. fuzzy matching is coming instead, that makes the whole thing more natural, and it’s easier now to find your ultimate target. the words orders will not matter anymore. like the type can be at the start, the end (like the narrower before), in the middle, etc.
the only big “difference” is that you’ll not be able to type the “middle” of a word at a search term. the search term(s) need to be the start of words. else there’s too much target pollution.
this is still subject to change as it is an alpha. i’m still evaluating what is the best, most natural option. and i gather users’ feedback.
as an example, now, a button with the title “Draft new release” could be targeted with:
- “new”
- “n rel”
- “butt rel”
- “dr butt”
the idea is to never have to “correct” your search term. that was the primary idea with the narrower: you start typing, then you narrow down. not it just works overall. you may start typing a word, stop, and realize you still have too many targets, then start typing a bit more, the “bit more” being another word (doesn’t have to be the exact following one), a type, etc.
the doc will be updated.
word boundaries will be improved in alpha 13, so that you can skip the https, /, . etc.
Seems that I’m late to the party but just wanna say thank you for your replies in this thread. I didn’t know about Whooshy but have been interested in the recent trend of “mouseless” utilities; your explanations and comparisons helped me learn a lot.
Personally I prefer Homerow (got mentioned somewhere in the thread; used to goes by “Vimac”). But for its vim-ism, one important consideration is that as a Chinese speaker, it’s hard to focus elements in Chinese simply by typing their names — Chinese characters aren’t matched by Latin transliterations — so using generated labels as a proxy is the only option. That said, your approach definitely is innovative and has its own audience.
there’s some possible work to be done to match pinyin to Chinese characters. sounds insane i know but it’s possible. but this is way down the priority list sorry, small ROI. lots of other stuff to be done in Wooshy first 
1 Like
the count is there but having troubles finding a proper UI. always feeling a bit off currently.
1 Like
Where you have it seems helpful but unobtrusive. No?
yeah, that’s the latest iteration i have. still a couple of issues that you can’t see in the GIF tho. anyways. WIP! will probably come properly at a later stage.