[ziyadb]

"That leaves us with the riddle that every software developer faces at some point: how best to make a UI simple enough to be intuitive for a consumer who wants a solid, fast browser that just works, and yet is customizable and extensible so that power users can add the features they want?"

You don't. Taking both user groups into consideration and weighing them equally assumes that they have the same bearing on the usage of your product. Once it reaches mass market, the tinkerers (occasionally referred to as the early adopters or evangelists) are no longer the largest nor the most profitable segment of your user-base.

Therefore, you can theoretically afford to disregard them--however, in practice, this has implications that can prove disastrous. For instance, they could seek out an alternative and start the disruptive cycle all over again, increasing its appeal and reducing yours, which propagates and ultimately ruins your perception, though this can probably be avoided.

[Jare]

I can't imagine a world where the unsophisticated "consumer who wants a solid, fast browser that just works" doesn't just use the default browser that comes bundled with the OS. Safari and IE are more than capable to fill that role these days.

[fpgeek]

Some of those unsophisticated consumers are still using Windows XP (or Android Gingerbread or ...).

[Jare]

To clarify, the theory is that there is a significant market in people who are still using XP, are sophisticated enough to want a 'better' browser, have knowledge and IT permissions to install one, yet at the same time don't want a featured interface.

I'm sure there's a few people like that, but a market big enough to model your product around them? I.... I just can't imagine that being the case.

[PommeDeTerre]

GNOME 3 is a perfect, and very recent, example of what you're describing.

It was open source programmers who made GNOME 2 and earlier a successful, usable desktop environment. These were very technical users who created software that generally worked well for their purposes, but incidentally also worked well for other, less-technical users, too.

Yet at some point, we saw the number of "designers" and "UI experts" start to grow, and eventually overwhelm the number of programmers. The "make it usable by everyone" mantra took over, but without the skill and sensibility we saw from programmers alone.

The end result has been GNOME 3. At best, it can be considered one of the worst open source disasters of all time. Its designer-driven development process and subsequent lack of usability and quality has alienated many programmers. These are exactly the kind of people who shouldn't be alienated, because they did things right in the first place. Aside from a few odd cases here and there, GNOME 3 is generally hated by anyone who uses his or her computer to do real work.

And what you say about seeking alternatives, or even creating them, is very true. KDE and Xfce benefited immensely from the GNOME 3 debacle, accepting many of the talented and valuable GNOME "refugees". MATE and Cinnamon have been born to try to fix the problem, as well.

GNOME 3 isn't the only example. Firefox is another good example, and obviously quite relevant to Opera. It has been systematically dumbed-down basically every release starting with Firefox 4, which has rendered it far less useful than it once was, causing many advanced users to abandon it.

Opera does indeed appear to be following this same path, unfortunately. Opera 15 has the same feeling that GNOME 3 does, in that it is a regression from the previous versions. Discarding so many things that make Opera 12 and earlier such a great browser just isn't a good way forward.

[brusch]

I don't know - the longer GNOME 3 is around, the more people like it.

Sure they've lost some people ("I can't install compiz fusion any more") - but in the last 6 month I'm reading lots of "it's fine - first it was a big change, but now I really like it".

I like GNOME3 and would never want back to GNOME2. But everyone his own's.

[inthewind]

I want to like Gnome 3. Every now and then I give it a spin, but ultimately end up back with Xfce. It just doesn't really give me anything extra, (apart from better font rendering in Firefox for some reason, and touch pad clicking! ) Their browser (web) doesn't work for me. It also kept freezing my keyboard for some reason or another, so it isn't much use.

The biggest irritant other than the above is that a couple of the taskbar widgets were missing. What I take away from that is that one little feature can influence us quite heavily. And it's pretty much the same story for Opera. I totally relied on the fit-to-width button and I love their spatial navigation - but have found myself using Firefox - just because I can make it more readable with a couple of easy modifications (though I still have issues with that.)

Give me a one line or paragraph summary of what Gnome 3 is, and why you'd want to use it over Gnome 2! Then try the same with Opera vs Chrome.

[brusch]

Gnome3 is a major update of Gnome2. They've replaced many aging libraries with their new counterparts. It includes 3d graphics (tasteful in my opinion), have dynamic workspaces and I really like the function when you're pushing the super key (windows key). The most important function is that it hardly gets in my way and I had to install less things to get the functions I really like (super - start typing to start an application, transparency...). I had to install GnomeDo and configure Compiz Fusion to get this in Gnome2 (and I still couldn't use super alone for this function). With GNOME3 I hardly every click with the mouse to start an application. I can't say anything about web (their browser) - cause I'm still using Firefox.

I think it is always hard, but sometimes open source projects have to make a cut to take a step in the right direction (and yes - some of the configuration stuff is hidden in Gnome3 and this really sucks). There are rough edges - but I think it's bold and good that they've made this step.

[inthewind]

As I said, I've used it. But it doesn't really give me much back (apart from the headache of rendering my computer useless.) Modern libraries (!?!), and 3D graphics and window transparancies isn't really a game changer. They uttered something about having a semantic desktop at first, but those ideas fell from the wayside. Synapse and Xfce give me easy application launch, I find it less jarring also. I hate that transition in Windows 8. I can get by with Gnome 3 though (when it works,) it's usable.

You can't even configure the appearance of your apps (colour scheme etc) that easily in Gnome 3, it was pretty sucky in Gnome 2. This was pretty much customisable in Windows 98!

Anyway you didn't rise my challenge of trying to summarise what Gnome 3 is! But thanks nonetheless.

[cuu508]

> GNOME 3 is a perfect, and very recent, example of what you're describing.

Agree, in a sense that it has abandoned tinkerers and focused on mainstream users. I'm not sure the result is disastrous, though. In Google Trends GNOME has maintained its position relative to KDE:

https://www.google.com/trends/explore?q=gnome+kde#q=gnome%2C...

If you have more relevant stats, please share!

[Sprint]

I share the dislike of Gnome 3 with many but I also see the potential of it for others so this is meant as a humorous countergraph: https://www.google.com/trends/explore?q=%22xfce%20sucks%22%2...

(Google Trends is not a measure of anything but Google Trends!)

[inthewind]

https://www.google.com/trends/explore?q=gnome+issues#q=gnome...

[richtr]

Since we seem to be doing Google Trends here is a relevant one:

https://www.google.com/trends/explore#q=%22chrome%20sucks%22...

How did that work out again?

[olimp]

It's because now more people search how to solve Gnome problems vs KDE problems. I'm not joking.