Comment by Dalewyn | Hacker News Reader

Comment by Dalewyn | Hacker News Reader

Parent Full thread

>(middle-aligned, bottom) large icon taskbar.

Wider and wider monitors that are higher and higher resolution, so the bottom left gets farther and smaller from the eyes. Nevermind multi-monitor setups.

>Being a fan of the "classic" labeled taskbar, taskbar items with text could provide a lot more info

Text requires translation/localization, which is extra work, which is extra cost. Pick a generic, ideally universal icon and it's once-and-done. Also, not everyone can or will read.

>the elongated click area is larger and more predictable.

While I agree, ultimately most people are going to find working with a mouse to be complicated and arduous because it's simply not a "natural, human" action.

>presenting just icons means there has to be a second level grouping if the same app has multiple windows.

You can have multiple buttons with the same icon, you know. A good example of this being the taskbar from Windows 10 and prior with window groupings disabled.

>Touch displays? Should be irrelevant in desktop mode.

The unfortunate reality is the desktop form factor has become a second-class citizen, best we (as in us nerds and graybeards and wizards) get used to it because that reality doesn't care for our chagrin.

>High DPI screens that demand dense rectangular items? The tradeoff seems to be big to be just that.

If anything, higher resolution screens demand circular things because they obviously look smoother than on lower resolution screens.

> Wider and wider monitors that are higher and higher resolution, so the bottom left gets farther and smaller from the eyes. Nevermind multi-monitor setups.

This is easily solved by having a 'bottom left' for every screen, something I wish Windows offered natively. OpenShell does, and I've configured my awesomewm instance to do so. It still makes the most sense to me and is the way I'm most productive.

> Text requires translation/localization, which is extra work, which is extra cost. Pick a generic, ideally universal icon and it's once-and-done. Also, not everyone can or will read.

Text is necessary to distinguish instances of open applications. Not having the ability to do that decreases productivity. I would think the amount of computer users unable to read text is an extreme minority, and I'm counting people using accessibility options to read text when I say that.

>I would think the amount of computer users unable to read text is an extreme minority

Can they read? Probably yes. Will they read? Probably not.

Incidentally, you demonstrated my point almost perfectly: Not everyone can or will read.

> Can they read? Probably yes. Will they read? Probably not.

If they don't want to read, as opposed to being unable to read, they are irrelevant to the point being made.

> Incidentally, you demonstrated my point almost perfectly

I read your point, I just disagreed with it. Still do.

I said that not everyone can or will read. This is the third time I repeat this, actually. You didn't read, neither the first nor second time, which demonstrated my point perfectly.

Text is meaningless when people will not read. Ask any programmer or customer service monkey demanding to know what error message a user is running across. Most people will not read, even when demanded to read what is on the screen word for word. "I get an error." "What is the error?" "I see an error." "What does the error say?" "It's an error, fix it!"

A lot of computer people don't like how Windows BSODs became more cryptic and meaningless ("simplified"), but this is simply because nobody reads them. Likewise Windows Update changelogs, nobody reads them.

I don't necessarily agree with the trend of removing text from everything, I find text much more descriptive than cryptic monotone hieroglyphs. But I can understand and respect why the computing world is trending this way: Text is expensive to make and maintain, and not everyone appreciates them.

> I said that not everyone can or will read. This is the third time I repeat this, actually. You didn't read, neither the first nor second time, which demonstrated my point perfectly.

Once again I did read your point, I simply disagree, and still do. Ironically it's only you who doesn't seem to be reading here.

If you're going to continue to insist that isn't the case, then there really isn't any point in trying to have a discussion with you since you can't do the bare minimum and assume good faith.

Let me put it this way then: You are disagreeing to the notion most people ("computer users", as you put it) can't read.

That's fine, and I even agree with you there; most people fortunately are literate today.

The thing is I also never made such a notion nor was it the only notion I put forth. Again: "Not everyone can or will read." Emphasis mine.

This is the fourth time I'm repeating myself, by the way. No, you still aren't reading. I do thank you once again for demonstrating my point, but this is getting a little ridiculous.

I'm not demonstrating your point for you, despite your continued insistence of that, and again, you really do seem to be refusing to read, not me. I agree it's getting ridiculous.

Let me repeat something I said earlier since you apparently missed it:

If they don't want to read, as opposed to being unable to read, they are irrelevant to the point being made.

Now, do you actually have a useful point to make in response? Or just more nonsense about how I'm supposedly demonstrating your point despite that not being at all the case?

If the latter, since you already have said it's ridiculous, and I agree, I'd ask you simply not respond to end this chain here.

Users who will not read are nonetheless absolutely relevant because they are still users of your program.

Development resources are finite, so it's natural and proper to excise the unnecessary and focus on what is truly worth developing.

If maintaining text is expensive (translations/localizations) and at least some users will not appreciate them anyway, why keep the text if just using icons is cheaper and can suffice? I don't agree with this trend, but I do understand and respect where it is coming from.

I agree with your last point. I disagree though that just icons can suffice, only in the sense that productivity is lost because without text, there is not a way to easily distinguish between multiple instances of a program.

The thing is, the text has to be maintained anyway, it's still there, for example in tooltips, so not giving users an option to show it has no real defense that I can see.

macOS doesn't really have multiple instances of running applications. Apps have a single instance with multiple windows, and the mechanism for managing windows and open apps is separate.

A lot of Windows applications now also behave like that behind-the-scenes, though that isn't the default behavior and requires some cross-process signaling.

> macOS doesn't really have multiple instances of running applications.

So if I wanted to start multiple instances of TeXStudio, I'm unable to do so?

I don't use a mac often, but when I have and I've put on less common programs that do have multiple instances, they have worked fine.

You can launch new instances from the terminal with

  open -n -a <Application>

If you want you can also launch a new instance using sudo to get root privileges in them as well.

So then it is possible, which again means I need text to tell them apart instead of just an icon.

Yeah as far as I know to change the name of the launching app you would have to duplicate the app bundle and rename it. But if you go that far you could also change the icon as well.

It's _technically possible_ but nobody does it, it's mostly available as an escape hatch to debug applications and run things as other users.

You'd open a new window within the currently running TeXStudio process with File > New (Window|Document). Some apps also let you open new windows through the context menu of their Dock icons.

You can force a new process by launching through the Terminal but this is not frequently done.

The program takes a parameter to open a new instance "--start-always', this is needed to load a different session. No idea if that is needed on mac or if a new 'window' would suffice.

Either way I need text to tell my instances, or windows apart.

> Pick a generic, ideally universal icon and it's once-and-done.

I think text is pretty essential, but if we're not going to be allowed that, at least make the icon actually representative of its function. One of the reason I find having the text so important is that icons have become increasingly inscrutable.

Reply on news.ycombinator.com