Re: Well said...

[Marcus Rhodes <marcus1 thinqware com>]

Thank you, Sergio.

"That is why I rebuild gnome2.32 from scratch using an archlinux distribution (642 modules compilled over a 45 days, with the help of the FreeBSD ports)
and gnome2.32 is back running in kernel 3.9 with systemd.. compilled with gcc4.8.  runs very well in a US$300 lenovo g475.. I setted up even a distribution repo to work with... Several (about 120) users uses it... (it is distributed in a new samsung momentus 320Gb HD, installs in 5 minutes).
have 10Gb size.. "

You should create an installer.  You could become famous, a hero even.

"Remember that new 3D interface for unix/linux that have a cube and the user could have 6 desktops running on each side of the cube??? Who uses it??? it is fantastic but no one uses it any more just because they (the people) can do things they want using the old interface."

Excellent point!  I would add, who uses it for anything truly practical?  I've had a lot of people try to convince me that they really use the cube, but I've never seen any benefit to it even when they use it.

And this shows that the UIs were already beginning to push some strange and largely useless eye-candy on us long before the advent of this touch-mania.  Fortunately, such stuff could just be ignored or disabled without repercussions, but how does one do that with Gnome3?  The 'classic' mode?  Hardly.  That's nothing more that a half-hearted nod to tradition.  Hardly a usable productivity tool.  I guess we should have seen the media-center mavens on the march even then.  And maybe that's why LXLE, Xfce, and Enlightenment were started: Their developers could see the handwriting on the wall way back then.

"I know that world must evolve, no problem, create a new gnome3, but do not kill gnome2 (or are you afraid of gnome 2.32??) If gnome3 gets so better, fast, easy, to use have more features than gnome2, people will move without complains..."

Exactly!  I've heard the Gnome DE head say that Gnome2 had reached the end of its life, and that it just couldn't be further developed.  But my question is, developed into what?  More spinning cube absurdities?  What exactly did it need to do?  Because, so far, I'm not seeing anything in 3 that 2 couldn't have done just as well.  Could it be that UI designers have just become so focused on the tool that they've forgotten that it is not an end in itself, but rather the means to an end?

Let me breach another example of how this touch/media-center mentality senselessly spreads like cancer.  Look at gedit.  I know it wasn't perfect.  Even I had some recommendations.  But whatever I tried to suggest simply got thrown back in my face.  Like, for example, adhering closer to the common standards by using, say, Ctl+F4 to close a tab/document instead of Ctl+W, or at least giving the users the power to configure that binding.  Instead of that very sensible change, we got a new, touch-oriented search 'dialog'.  Seriously?  In a text-editor?  Was someone thinking perhaps that users would be editing text via a touch-screen?  Or was the decision just to make that feature more harmonious with the surrounding UI?  Either way, it makes no sense.  Of all the many, many improvement that could have been made to gedit, this is the one that made the cut?!  Really?!

I'm donating to the Mate project right now, and from now on.  Maybe they'll take my suggestions seriously instead of mocking and ridiculing me.  I advise everyone else who wants a keyboard-friendly, minimalist DE to do the same.  What the media UI developers clearly must feel are the millions, and even billions of users filling their in-boxes with requests, even demands for the latest, coolest, gee-whizziest, touch-driven, spinning-cube eye-candy should, of course, feel free to support their favorite DE, too.  But, apparently, they already do, or we wouldn't be in this situation.  Unless, of course, those hordes of eager users begging for touch features are just imaginary.

Oh, and a word about icons and other GUI elements...  Do you know where they really came from?  I mean aside from the obvious if pathetic attempts at skeuomorphic familiarity like trash-cans and cassette-player controls.  It offered third parties a way to push their logos into our eyeballs.  And that really doesn't strike me as fitting into the Linux paradigm.

Think about it.  Do you always know what those GUI symbols mean?  Or an unfamiliar logo?  No.  Of course not.  But how do you look them up?  You can't.  So the UI developers added text labels right from the start.  And where labels didn't seem appropriate (tool bars and such), they added text balloons that appear when you hover over the icon.  At least, when they remembered to do so.  I'm looking at the Evolution composer's toolbar (which, for some genius reason, I can't disable) (any of them) (so much for less-is-more, eh?) right now wondering how on earth my wife is supposed to know that the magnifying glass = search, but the magnifying glass and pencil = replace when she doesn't even get a text-balloon to tell her what they mean.

I mean, some things are good, great even.  I love scrollbars (which Unity bizarrely eliminates).  I don't use my mouse on them, but the eye acquires from a scrollbar very quickly what used to have to be rendered something like: "You're seeing 345-361 of 1232 lines."  So I'm not against GUIs.  I really love them.  And I love the standardization they brought to the applications.  But I'm against taking it too far, and all this touch stuff is far beyond too far, especially when the keyboard is rendered next to useless, and well organized, text-only menus are no longer available.  At all.

Here's a thought.  Lose the toolbars (or at least make them optional), rationalize the menus, ensure mnemonics (you know, those little underlined characters in the text labels) are available everywhere, and exploit accelerator/short-cut keys (Ctl+O, Ctl+P) more fully, and win back your fans.