Why allow a user to render a window full-screen, but not to maximise it?
An example is:
#!/usr/bin/env sh
kdialog --title '' --msgbox ''
Without knowing why, I err on advising that both options be disabled, by default, if they can be re-enabled via .KWinRules.
Perhaps because your example is a Dialog, not a Window, designed for short-lived, transient interactions/prompts?
Itâs based on QDialog class, which traditionally doesnât have fullscreen options in itâs default window decorations - theyâre supposed to be small, focused pop-ups.
Fullscreen would block other content creating a poor user experience.
Iâm pretty sure we all know what a dialog is and donât need to discuss the sense or non-sense of maximizing them, it was clearly just an example.
The real question seems to be why weâre allowed to put them into full screen, when maximizing is disabled. For the sake of consistency both options should be disabled.
Possibly - but the option is not actually present in the window, only through the menu, or perhaps via direct keyboard shortcut. Thereâs an option to pin, or minimise, but no âmaximiseâ to click in the titlebar.
Perhaps it was felt that such a buried inconsistency wasnât worth worrying aboutâŚ
The ability to force a window into fullscreen is a lower-level feature which can evidently bypass the dialogâs own design where maximising is disabled.
Iâd suggest that nobodyâs particularly bothered with it, and I canât imagine it would ever cause an issue.
FWIW a better example (with some saner fontsâŚ):
kdialog --title 'This is just a test...' --msgbox 'The quick brown fox jumps over the lazy dog'
A maximized window takes on its maximum proportions allowed on the desktop - which may be smaller than the size of the desktop due to the presence of panels and/or docks - whereas a full-screen window supplants the desktop in the display buffer.
@ben2talk, that appears to be a verbose way of merely explaining that maximised windows do not render over plasmashell, whereas full-screen windows do. Consequently, have I not noticed a nuance that you intend to communicate? If not, that distinction was already fundamental to my explanation, so I fail to understand why youâve re-stated it.
Also, Ben, and Iâm going to bring this up here, because itâs contextually relevant:
Although I promise youâve no obligation to affirm this, I would expect that @amilias would agree that that is very strange to say after someone politely informs you that youâre wasting time on trivial pursuits.
Wow, I always thought ben2talk was just some weird old guy who loves derailing threads by rambling about his personal preferences, but I guess he really is just a huge jerk (too polite to use a different word) after all.
Haha well I was just struck by it, at first I was surprised by the âbugâ and couldnât figure out how it could be a problem - it just seemed to be something outside the normal scope of kwin.
I also noticed that developers didnât seem to interact with it, and that you posted a link to other bugs that you had created in the past which you deem relevant, but which you also posted.
So I was just curious to find out if anyone would support or bring context that I could understand.
There was no disrespect intended, but having communicated with a large range of members - many of whom do tend to be neuro diverse and intensely focussed on details which are irrelevant to the majority, I was curious if anyone would bring up any interesting information - other than that âfullscreenâ tends to be a lower level function, that doesnât respect the same rulesâŚ
If it had brought anything interesting, I could have brought that back here.
So now it appears that @amilias is offering very postitive input here:
From your reddit link, we see the point of THIS thread is actually related to someone elses desire to have a âFULLSCREENâ button available in the titlebar (perhaps frustrated that the maximise button is not) because they use it quite a lot.
Fullscreen is a function which I actually consider a separate issue from KDE WIndow manager functions - itâs a standard feature across different WMs.
In KWin, we have the option to actually hide the window decorations in the maximised state - giving a very similar experience to âfullscreenâ but without taking over the screen.
But there are better ways to manage fullscreen - traditionally F11 for the vast majority of cases.
Then the right-click context is an âadvancedâ use case. Firefox might be considered an example which manages its own fullscreen behaviour with CSDâŚ
Then with SSD the FULLSCREEN button would disappear as soon as you entered fullscreen - thatâs something that I actually would consider a serious bug.
But in this case - the âmaximiseâ button is simply not available because the window is not intended to be maximised - and the âproblemâ that it is possible to bypass this by going into a menu and fullscreening the window isnât a âproblemâ.
It could marginally be described as ânot totally consistentâ but I feel that it is not able to create a problem at all. You can use that same menu to undo your fullscreen action.
Can you describe an example where this might actually prove to be important?
I seem to have missed the part where you apologize for trying to dictate which topics are ârelevant enoughâ for people to discuss while bad-mouthing them in your private echo chamber, basically calling us mentally ill for having the audacity of taking interest in a minor UX issue, just because you donât care about it. How about ignoring threads that donât interest you next time?
Also, I have no clue what even is up with the latter part of that post, I donât see any reddit threads posted here and youâre again just digressing into a completely different topic, if not into straight-out lunacy.
Neurodiverse people who focus on the small details have contributed hugely to humanity.
Those who focus on trying to belittle and humiliate others, not so much.
I never did anything of the kind - not even close.
By using the term âusâ - are you implying that you are on some kind of spectrum and are sensitive about it?
The fact is that Iâve experience with many forum members of many kinds - and many of my closest associates tend to be on the autistic spectrum, sometimes they get bogged down with small details, but sometimes they are very good at finding things that the rest of us leave behind or dismiss.
Hence I just wanted to ask their opinions about the topic - and as it seems, not many were interested to comment beyond the clarification of maximise vs fullscreen, regarding it as a minor âimperfectionâ as it were thatâs likely not worth the trouble of fixing.
Sometimes I ask such opinions and I get very good feedback bringing up real-world issues that are frequently caused by such issues.
You might also notice that I never did anything of the sort, indeed I made sure to comment as to the qualifications of the OP, rather than dismissing him as a person I was merely curious about the subject of the post.
I actually really appreciated the way he put his address in his profile, and intended no disrespect whatsoever.
We might also note that nobody else seemed interested in the topic at all, and so I also was motivated by the possiblity that I might get some responses which might lead to a positive view on the subjectâŚ
Perhaps also interesting is that the first two people who responded on that thread are firmly set on the âspectrumâ and frequently bring very sharp and lucid overviews to subjects which I might overlook. I often value their opinions more than my own - though not necessarily in this case (I think AwwsumGuarav responded a little harshly).
Then look again, itâs clearly quoted.
Well Iâm not sure who suggested that - it did interest me, it initially surprised me and later on with such threads we do often find real issues that really do need sorting out.
The best way of finding out is by finding actual real-world issues that cause problems - to decide if such irregularities actually deserve attention.
I am still open to the idea that this could be a valid complaint - but I think it must be tied down to real-world issues - Iâd say itâs very application dependent.
This is, indeed, the purpose of this thread - the rest is just wildly off-topic.
So now this is more of a witch hunt, going off the rails. If thereâs no further interest or explanation relating to the actual original topic, I have no further wish to engage in off-topic banter.
@ben2talk, that was my Reddit OpenGraph embed (which may be why @amilias was confused by your assertion), and wasnât the impetus of this discussion regardless.
Albeit, not when browser.tabs.inTitlebar: 0.
I vaguely challenge that, considering that dwm.exeâs 1:1-equivalent context provides basic functionality that is widely-utilised, too:
However, considering how such a designation is subjective, I can continue as if it is.
I agree that that is a problem, in retrospect. However, is it on-topic for this discussion? If not, I advise that you create a separate Discourse or Bugzilla thread about it.
Considering my immediately precedent quotation of yours, how can that be so? Specifically, you insightfully assert that a user being unable to easily access full-screen mode without a keyboard may be problematic, yet simultaneously assert that being able to access full-screen mode for a window that is not designed for it is not.
Even if you were to somehow convince me of this, what could be the possible problem with having this option disabled when the maximisation option is disabled? I donât understand why you expend so much effort to convince us that it is of no importance.
Although what I have aforequoted appears to demonstrate that your intent was not merely this - instead, that you also believe that it is not a problem - to answer your question, the reason that this might be problematic is the same reason that maximisation is disabled. Specifically, that it can cause the window to render in a manner that it is not designed to. An example is that I have had windows crash with a segmentation violation, when resized, after informing the compositor that they may not be. Considering that merely F11 can invoke the full-screen mode, that may cause data loss. Plasma Mobile explicitly does not render such windows full-screen, by default.
I tend to play devilâs advocate - making suggestions with a weak basis to see if there is any pushback or more solid reasoning to be had to support the claim.
Hold on - so yes, if a user right clicks to access the menu and goes fullscreen (no keyboard) they canât get out of it unless they have an alternative, so that is indeed a problem - not just the fact that they can get fullscreen. Simply going fullscreen or maximising isnât a problem per se - it should simply never be necessary.
There would be no problem if both of these options were disabled - but I fail to see the problem of the hidden fullscreen option not being disabled - but I was not aware that it could lead to segmentation violations and crashes, then that is a valid reason to consider the option to remove it - if itâs feasible.
Plasma Mobile is a scenario Iâm not familiar with, and I defer judgement on that topic.
