16 comments

[ 0.23 ms ] story [ 50.0 ms ] thread
Why is this poorly formatted word document opinion piece on hacker news right now?
I guess the fact that in the age of mobile-first responsive content this guy decides to publish as word-exported-to-pdf tells you everything you need to know about his ideas of "good user experience"
I'm sympathetic to your point. But publishing in the Portable Document Format isn't the worst.
Something can be unacceptable without necessarily being the worst possible case.
True, let me rephrase.

Publishing something in the Portable Document Format (particularly on a non-web subject) is not unacceptable.

HTML is nice, but it does have its drawbacks. For example, HTML doesn't have pagination or standard TOC support. It's hypermedia, so you have to link out to a bunch of other files, making it much more cumbersome to transmit and distribute. Using a zip, I suppose? (Yes, you can use data URIs in HTML5 compliant viewers, for inflated sizes, and mandatory full file reads before rendering.)

I probably wouldn't have done a PDF either, but I think you're reading more into it then you can.

OK, we shouldn't get too hung up on this; "unacceptable" is obviously hyperbolic! I won't bother with a point-by-point rebuttal, just summarise by saying that this is clearly a blog post, and PDF is completely wrong for a blog post. For the author to get something relating to usability so fundamentally wrong in a post about usability just entirely undermines their argument.
You know what, on second thought, you're right.

Given the context and content, it's pretty ironic.

He considers this a "whitepaper", rather than the more accurate "blog post" (see the Documents section on his website)
(comment deleted)
I don't understand, 90% of the clutter would disappear if invest some effort into pimping your emacs. Imagine how fantastic it would be if you published this with org-mode.
The question that pops up in these interface discussions, and it's pretty much applicable to all users of GNOME classic and KDE, is this: who still launches applications using the mouse? I haven't done that in 5 years. It is five times faster to launch apps using the keyboard, and this saves infinite screen space. Seriously, when I boot up, I hit the Super key, T, enter, super key, S, enter, and Super, F, enter, and the terminal, sublime, and Firefox instantly appear. I quickly search for other applications with a few more key strokes. I don't need a window list cluttering my screen, because a quick hold of alt and tab shows what windows I have open. This is how people launch apps and open files on GNOME3, OSX, and even Windows, now, and it is objectively superior to the old click to launch, click to select interfaces.
I find launching regularly used applications much quicker with the mouse if my hand is already on it. Especially on Windows 8/.1 where the action is just to aim at the right tile and can be done using muscle memory.
Whew, flashbacks to Office 95 and its toolbar program, with launcher icons across an entire edge of the screen. Those were the days.

I just don't understand it. He talks about having icon launchers for 6 IDEs, 40 directories, 150 documents, 25 web sites, 4 games, and 50 utilities--all visible on the screen at all times.

When I used Windows 95/98/2000, I had lots of GUI launchers and desktop icons. Even when I first started using Linux, I had a bunch.

But now I only have a few launchers in my panel. I launch programs with krunner, the application menu, or through a terminal. fasd lets me quickly access directories I've accessed before. Fish shell history and percol/fzf/etc let me quickly rerun past commands. Emacs and recentf and helm-locate let me quickly access any file on my filesystem by typing a few characters.

I have one panel, about 44 pixels tall, across the top of the screen. My window titlebars are about 32 pixels tall, and they are hidden when the window is maximized. 98% of my screen is devoted to Emacs or Firefox or a shell, LibreOffice, Dolphin, etc. No window borders. Decoration is minimal and gets out of the way. UI is mostly hidden until I need it.

But--one of the great things about Linux/FOSS/etc is freedom. This guy can customize his UI the way he likes it. GNOME2/MATE, KDE, TDE (long live KDE 3.5!), etc. let people make their systems work for them, not the other way around.

In contrast, Apple, Microsoft, GNOME3--they all say, "Have it our way."

I miss that little "drawer" function. Y'know, the icon that you could hide other icons in, so you could have your shortcuts inside a single drawer instead of plastered all over your status bar.
I'm stuck using Gnome 2 at work for the time being, and I must say I really don't dig it. I think various pieces of 'bloat' that have been added in the years have made my life better.

I find myself taking longer to find the open window I want. Instead of doing an Expose effect then clicking the window I want, I have to look through a list of windows. I find myself frequently clicking through the list in order to find the window I want. The fact the order isn't consistent doesn't help.

That's just one example I know, but after hearing Gnome 2 hold outs continually claim Gnome 2 is still better, I found it really shocking how much I don't like this experience.

Soon we will get support for Unity and I honestly can't wait.