GNOME Bugzilla – Bug 708452
Focus and Caret Tracking: Include preferences
Last modified: 2021-06-09 16:09:20 UTC
Created attachment 255388 [details] Magnifier Zoom: The first setting page for the mouse settings It is useful to be able to change between gsetting preferences for caret and focus tracking as well as just the mouse now. The magnifier is an a11y feature used mostly by those with low vision so they could struggle if they are only able to access the settings via the command line. I have some pictures of a UI I have come up with but I have not yet had time to fully integrate yet but I hope to be able to do that over the next few days or otherwise link to the wip. I would be greatful for feedback on the ui design though as this is my first attempt with glade ui and I am unsure whether I overlooked anything.
Created attachment 255389 [details] 255388: Magnifier Zoom: The first setting page for the focus settings
Created attachment 255390 [details] 255388: Magnifier Zoom: The fsetting page for the caret settings
I will let the design team give the feedback about the ui design itself, but I would like to add some comments about the text used. (In reply to comment #1) > Created an attachment (id=255389) [details] > 255388: Magnifier Zoom: The first setting page for the focus settings (In reply to comment #2) > Created an attachment (id=255390) [details] > 255388: Magnifier Zoom: The fsetting page for the caret settings Internally, both caret and focus tracking have the same three options, "proportional", "centered" and "push", so: 1. Taking into account that the three are the same, and the basis are the same, why the explanation is different on each tab? It is not only about changing caret for focus, that could make sense, but all the explanation. 2. The order is different on each tab. IMHO, for coherence it would be better to have the same order. 3. More important: in several cases it seems that the explanation about the effect was reversed. Example: "Focused item is pushed by view". Seems to suggest that modifying the view has as effect changing the focused item, when it is the opposite, changing the focused item pushes the view. I think that it would be better/easier to just reuse/adapt the text used at the Mouse tab. That would also help with having the tabs content more coherent (as I say on 2.)
I made a rough attempt at making both the 'caret' and the 'focused item' the subject in the sentence (though I just spotted that there is a typo I failed to see on the focus) but the mouse settings are doing it too I am afraid which is something that needs to be revisited in light of your observation Ideally (Joseph''s input is needed here since it involves his work on the mouse ui) I would like to decide what the subject of the sentence is and make sure that the agreed subject is the same for focus, caret and mouse. As you can see in the mouse settings the subject is only 'the magnifier cursor' in the last two gsettings Lets look at the possible usages of 'center' 'proportional' and 'push' once mor The following problem arises when we try and keep to a verb usage to describe our gsettings: 1. to center 2. to push 3. to be (proportional) 3. is the odd one out. We cannot proportion, it is never a verb. The verb there is 'to be', to make, to have etc. I think Joseph dealt with this by dropping the word proportional altogether to explain the setting and replacing it with the verb 'move' 1. to center 2. to push 3. to move The logic here is flawed because they all move. How do they move? Lets look at the adverbs: 1. to move central 2. to move (nope sorry can't make push into an adverb) 3. to move proportionally (works but is not very meaningful) Lets look at this another way:adjectives 1. A centered thing 2. A pushed thing 3. A proportional thing This works, for all three, great! So let's decide what the 'thing' is. It could be anything but we should try to be consistent. A focused item A View A zoom region A caret A mouse A document window The sentences will stop being consistent when we flip between difference sentence subjects and their objects (object as regards grammar not programming)as you pointed out already when you noticed that "Focused item is pushed by view" is misleading. I can suggest a solution but we need to decide whether consistency is more important than accuracy before I try because I suspect it won't be possible to achieve both unless we rename the gsettings to better describe what they do.
As an afterthought one other way that works is also if we just use the verb 'to be' for all settings the question is: what is it that is pushed what is it that is proportional what is centered in each (focus caret and mouse) case? Also should the default setting be at the top? I think yes because an accessible feature should be as easy for its users as possible and I think it is important that a user would be able to guess without being able to see it easily where the default setting radio button is located just after a fresh install/update. But you noted again there is an inconsistency and this is because each setting has a different default gsetting. I think it is more important to pick the default that is most likely to be used but I am not overly confident that the default settings are the right ones for the majority of use cases. A proportional caret is much easier for me to write something but I do not have low vision. However, I think there is an advantage to using caret tracking for users with reading and writing difficulties who do not require high levels of magnification (or any magnification) at all so they might not be disadvantaged if the default was on but in the 2nd radio down from the top instead of being the first one. For mouse and focus tracking I cannot think of a use case other than low vision. I cannot comment on tunnel vision (I do not know the correct term so if the term I used offends someone, this is unintentional and I apologise) so but I do not really have a strong opinion and I suspect joanie should probably be a good person to judge whether we have the right defaults. I think that we should decide whether what we want to achieve is possible without filing a gsettings but first though.
bug not but! sorry.
I was asked to give a description of the tracking modes. My apologies if this duplicates aspects of the above. Also, sorry for the length. All of the tracking modes affect the visible contents of the magnified view. With any degree of magnification, less than the entire desktop is visible within that magnified view. For example, with a 2x magnification factor, only half of the screen is visible at any one time when the magnifier occupies the entire screen. Even less is visible if the magnified view occupies only the top half of the screen. With that in mind, for purposes of explanation, the following describes the tracking modes with respect to the focused item, but similar comments apply to caret tracking. When in centered mode, when focus changes, the focused item is to be displayed at the centre of the magnified view. In this mode, as the user TAB- or arrow-navigates the user interface, the contents are shifted to move the focused item (button, menu item, checkbox, etc.) to the centre. When in push mode, the magnified contents are shifted just enough to bring the focused item into view. Sometimes no shifting occurs in this case, specifically, when the focused item is already in view before it acquires focus. Proportional mode is the most difficult to describe. Here, the location of the focused item in the magnified view provides a cue as to where the associated unmagnified item is on the desktop. Since the user cannot see the entire destkop within the magnified view, they lose visible reference points such as the edges and corners of the desktop. Placing the focused item in a location that is representative of where it is on the desktop provides a degree of visual orientation. For example, if the unmagnified focused item is towards the bottom right of the screen, the contents of the magnified view are shifted such that the magnified focused item is towards the bottom right of the magnified view. If the item is at the centre of the desktop, then it is displayed at the centre of the magnified view. And so on. One way to handle the radio button labels is in terms of a command to move the contents of the magnified view: () Center the focused item. () Move contents so that focused item is in view. () Place focused item so that ??? For the text input caret: () Center the text insertion point. () Move contents so text insertion point is in view. () Place text insertion so that ??? As you can see, it's difficult to come up with a succinct label for the proportional mode (at least for me). Hope that helps.
(In reply to comment #4) Just to clarify: > ... > The following problem arises when we try and keep to a verb usage to describe > our gsettings: > > 1. to center > 2. to push > 3. to be (proportional) > > 3. is the odd one out. We cannot proportion, it is never a verb. The verb there > is 'to be', to make, to have etc. I think Joseph dealt with this by dropping > the word proportional altogether to explain the setting and replacing it with > the verb 'move' For mouse tracking, my suggested labels were (way back when): "Zoomed image moves with the mouse pointer: () Always () To keep the pointer visible () To keep the pointer centered" Those labels were modified during the review process to what they are now. In short, I didn't drop 'proportional' in favour of 'move'. I dropped 'proportional' early on because it isn't exactly user-friendly. Later, the heading ("Zoomed image moves with the mouse pointer") was removed, and "Always" was replaced with "Magnifier cursor moves with contents".
(In reply to comment #7) > I was asked to give a description of the tracking modes. My apologies if this > duplicates aspects of the above. Also, sorry for the length. > > All of the tracking modes affect the visible contents of the magnified view. > With any degree of magnification, less than the entire desktop is visible > within that magnified view. For example, with a 2x magnification factor, only > half of the screen is visible at any one time when the magnifier occupies the > entire screen. Even less is visible if the magnified view occupies only the > top half of the screen. > > With that in mind, for purposes of explanation, the following describes the > tracking modes with respect to the focused item, but similar comments apply to > caret tracking. > > When in centered mode, when focus changes, the focused item is to be displayed > at the centre of the magnified view. In this mode, as the user TAB- or > arrow-navigates the user interface, the contents are shifted to move the > focused item (button, menu item, checkbox, etc.) to the centre. > > When in push mode, the magnified contents are shifted just enough to bring the > focused item into view. Sometimes no shifting occurs in this case, > specifically, when the focused item is already in view before it acquires > focus. > > Proportional mode is the most difficult to describe. Here, the location of the > focused item in the magnified view provides a cue as to where the associated > unmagnified item is on the desktop. Since the user cannot see the entire > destkop within the magnified view, they lose visible reference points such as > the edges and corners of the desktop. Placing the focused item in a location > that is representative of where it is on the desktop provides a degree of > visual orientation. For example, if the unmagnified focused item is towards > the bottom right of the screen, the contents of the magnified view are shifted > such that the magnified focused item is towards the bottom right of the > magnified view. If the item is at the centre of the desktop, then it is > displayed at the centre of the magnified view. And so on. > > One way to handle the radio button labels is in terms of a command to move the > contents of the magnified view: > > () Center the focused item. > () Move contents so that focused item is in view. > () Place focused item so that ??? > > For the text input caret: > > () Center the text insertion point. > () Move contents so text insertion point is in view. > () Place text insertion so that ??? > > As you can see, it's difficult to come up with a succinct label for the > proportional mode (at least for me). > > Hope that helps. Yeah it is a tricky one, essentially we have to adapt every sentence to best fit the kind of movement we want people to be able to expect and it is hard when you already know. I agree with your decision to remove proportional from the explanation even if it is the true name of the setting itself. I liked 'x moves with y" for proportional best I think but I still find it difficult to describe what I see for proportional and push focus. Do you think we could try the list for feedback on some ideas given there is a bit more time for this one?
(In reply to comment #8) > (In reply to comment #4) > Just to clarify: > > > ... > > The following problem arises when we try and keep to a verb usage to describe > > our gsettings: > > > > 1. to center > > 2. to push > > 3. to be (proportional) > > > > 3. is the odd one out. We cannot proportion, it is never a verb. The verb there > > is 'to be', to make, to have etc. I think Joseph dealt with this by dropping > > the word proportional altogether to explain the setting and replacing it with > > the verb 'move' > > For mouse tracking, my suggested labels were (way back when): > "Zoomed image moves with the mouse pointer: > () Always > () To stay visible > () To stay centered" > This to me actually seems a lot better. I am afraid of being too wordy and yet the description is important or else the mode labels would not mean much to most people. Yet, I am not sure I can cope with two the same and one different (I am not sure why), I prefer when they look all different, or all the same. Ideal situation (in my opinion) is one where you can read the label without having to actually read ie. it is short enough to read at a glimpse. If for no other reason, it is something most people are likely to expect. I still do not feel like the following set is right but it is a bit easier to pick up what the selected view is going to move like: Movement () Always () Just visible () Keep central I think this sort of thing should even work for all three. I am defintely going to have to revisit the gsettings though, just to check I am right about what they are doing for each case. > Those labels were modified during the review process to what they are now. In > short, I didn't drop 'proportional' in favour of 'move'. I dropped > 'proportional' early on because it isn't exactly user-friendly. I am curious about how hard it would be to have a 'clickable' icon next to each radio that when clicked, pops up with an animated gif (or even something cool looking like the gnome getting started animation) to quickly demonstrate the zoom movement in some sort of pop up window so users can click it to get a short popup demo. That might be awkward to implement but it does not seem beyond possible, I guess the first question would be: Would something like that be useful for your average magnifier user?
GNOME is going to shut down bugzilla.gnome.org in favor of gitlab.gnome.org. As part of that, we are mass-closing older open tickets in bugzilla.gnome.org which have not seen updates for a longer time (resources are unfortunately quite limited so not every ticket can get handled). If you can still reproduce the situation described in this ticket in a recent and supported software version, then please follow https://wiki.gnome.org/GettingInTouch/BugReportingGuidelines and create a new bug report at https://gitlab.gnome.org/GNOME/gnome-control-center/-/issues/ Thank you for your understanding and your help.