Why aren't desktop OSes innovating anymore?
May 14, 2012 3:12 PM   Subscribe

Why is there so little innovation nowadays in desktop OSes (both Windows and Mac)? I would think features like a super-smart Siri, face and gesture recognition, and perfectly grandparent-friendly interfaces would long be standard by now, but that is far from the case. What is holding things up?
posted by shivohum to Computers & Internet (14 answers total) 5 users marked this as a favorite
 
"Super smart siri", depending on how you define it, is one of the hardest problems in computer science. It's one of those things where it's pretty easy to get 80% of the way there and the rest is very very hard.

What would face and gesture recognition be good for? It sucks for most things its used for now. What would you like to have the computer do with your face and gestures?

Grandparent friendly interfaces probably aren't really a priority, whether they should be or not. What people want, or at least what it's perceived they want, is everything faster, bigger, prettier, more featured, etc. Easy interfaces only benefit people who don't know how to use computers, and annoy pretty much everyone else.
posted by RustyBrooks at 3:15 PM on May 14, 2012 [1 favorite]


Because using touch at arm's length sucks, and folding/transforming laptops are incredibly fragile and have terrible battery life.

The iPad is a perfect "casual" computer for grandparents and non-intensive tasks, for real business and gaming, you still have a moust-driven OS.
posted by Oktober at 3:16 PM on May 14, 2012


Take a look at this article:
In theories of competition in economics, barriers to entry, also known as barrier to entry, are obstacles that make it difficult to enter a given market.

Then take a look at this chart:
Usage share of operating systems

That's 74 percent of the market still tied up in Redmond.
posted by Cool Papa Bell at 3:39 PM on May 14, 2012


Why is there so little innovation nowadays in desktop OSes (both Windows and Mac)?

Mostly because once someone figures out a very good way of doing something, everyone gets accustomed to that way and unless someone figures out something truly groundbreaking then it never gets replaced. So, for example, steering a car using a big wheel mounted on the dashboard is probably not the absolute best possible way to do it, but other possible methods are not so much better that everyone is going to switch to them. Similarly, a Qwerty keyboard may not be the best possible way to input text into a computer, but no other methods are so much better that they can easily supplant traditional keyboards.

I would think features like a super-smart Siri, face and gesture recognition, and perfectly grandparent-friendly interfaces would long be standard by now, but that is far from the case.

Recent interface simplifications for mobile devices like voice text entry and predictive text are actually steps backwards in terms of efficiency, they are necessary mainly because such a small device can't have a nice big keyboard that things like the Internet are designed around using. Similarly face and gesture recognition are helpful for a device like the Kinect that only has a camera for user interaction, but they are much less efficient means of conveying most kinds of information to a machine. There would definitely be ways to redesign an OS to be less complicated in the same way that using Siri is less complicated than typing a complex search query into Google is, but the tradeoff is that it's also going to be much less powerful in terms of allowing a lot of complex input from the user. Most desktop UIs are not designed around those sorts of simplifications (although you may seem them moving towards that as tablets and whatnot move in that direction) and if the complexity is still there then there's not a lot of better ways of doing it than what we already have.
posted by burnmp3s at 3:41 PM on May 14, 2012 [1 favorite]


perfectly grandparent-friendly interfaces would long be standard by now

I can speak to this one since I teach grandparents to use email every week. In short the issue is that there is no benefit to a business who creates tools like this. When the lowest price point for things like email is "free" and the standard for expected support is "nothing" it's almost impossible to compete in any rational way for a for-profit business.

Companies like Google have a really nice tool in gmail for superusers like me and the obvious downsides [the ubiquity of the google marketing machine] really only matter to a small subset of people. However, the same people who might benefit from SimpleMail are also not the people who would invest enough in something like this, even by paying a small monthly fee, and would be labor intensive in terms of support.

The big deal is that if you ask 100 people what the top five things are that they need their email program to be able to do (for example), you'd get fifty answers. So you wind up with systems that are able to do fifty things, but aren't really optimized for any of them except keeping people's eyeballs inside the Google panopticon, for example.

Or, at a more basic level, the reason Yahoo Mail isn't easier to use is because there is an essential conflict between showing people advertising and showing people their mail. [see also why we don't have a la carte options for cable television which is all anyone on the planet really wants most of the time]

Easy interfaces only benefit people who don't know how to use computers, and annoy pretty much everyone else.

Also this. And the fact that many if not most people who want easy interfaces at some point graduate to wanting to be able to do more with their systems and you have a planned obsolescence track that, again, is not worth investing in. It's easier to take the functionality of something like gmail and skin the crap out of the top of it to make it more of an ideal SimpleMail solution and then let people unskin it as they get more functional. Serving the information poor under capitalism is about as attractive as serving the cash poor. It's not particularly rewarding, it's rarely profitable and there's very little push for these solutions from people who don't even know what their options are.
posted by jessamyn at 3:53 PM on May 14, 2012 [1 favorite]


I don't think your examples are so advanced, but those operating systems have definitely stagnated. It's because engineers and designers are not as well educated today, work in environments toxic to creativity and neither possess nor are pushed to possess the self-control and drive necessary to prevent technological stagnation. OSX is not much more than a modified version of an operating system created in the 60s, running on hardware based on designs from the 60s, and Windows is worse. I believe Richard Hamming is correct about the reason for this:

Coming out of the war, coming out of Los Alamos where we built the bomb, coming out of building the radars and so on, there came into the mathematics department, and the research area, a group of people with a lot of guts. They've just seen things done; they've just won a war which was fantastic. We had reasons for having courage and therefore we did a great deal. I can't arrange that situation to do it again. I cannot blame the present generation for not having it, but I agree with what you say; I just cannot attach blame to it. It doesn't seem to me they have the desire for greatness; they lack the courage to do it. But we had, because we were in a favorable circumstance to have it; we just came through a tremendously successful war. In the war we were looking very, very bad for a long while; it was a very desperate struggle as you well know. And our success, I think, gave us courage and self confidence; that's why you see, beginning in the late forties through the fifties, a tremendous productivity at the labs which was stimulated from the earlier times. Because many of us were earlier forced to learn other things - we were forced to learn the things we didn't want to learn, we were forced to have an open door - and then we could exploit those things we learned. It is true, and I can't do anything about it; I cannot blame the present generation either. It's just a fact. (source)
posted by michaelh at 4:01 PM on May 14, 2012


This is, by the way, a subject of great interest to me. There is a concept in the design of things that it's best if the thing you're designing documents or explains itself, because it's really annoying otherwise.

Often you have to choose between "the documentation is somewhere else, you're expected to read and understand it before using this thing" and "let's plaster this thing with tips on how to use it, which will help someone using it the first time but piss off everyone else". If it's something you will use thousands of times, then the 2nd method is VERY ANNOYING because the plastering only helps you 1/1000 times and the other 999 it's just junk.

Sometimes there's no choice - weird road configurations are an example, because the cost of being wrong is too high. So you'll see street signs that inform almost no one ever, just the hapless person coming through there the first time.
posted by RustyBrooks at 4:02 PM on May 14, 2012 [1 favorite]


There was, back in the 90s, a Windows POP-based email client that was built around that kind of simplicity. I can't remember its name, but it never got any traction for reasons mentioned above.

A variation on the theme: the innovation is happening elsewhere, and you're not quite noticing it. When you talk about new adopters in 2012, as opposed to 2002, they're coming in from their mobile phones and their Kinect-enabled XBoxes and their iPads, or running everything out of a web browser; it's not simply that they're engaging less with the desktop OS and its accoutrements, it's that the entry level brings with it ways to interact (e.g. Facebook messaging instead of email and IM) that aren't bound to the old desktop model.

It's because engineers and designers are not as well educated today, work in environments toxic to creativity and neither possess nor are pushed to possess the self-control and drive necessary to prevent technological stagnation.

"And back in the day, we pushed around packets with our bare hands."

Sorry, not buying it at all: the ferment of technological innovation in that period was possible because a small-scale experimental environment allows you to break and rebuild things in ways that aren't possible once something is adopted by the masses. It's no different from the interface for driving a car, where there was a period of huge experimentation, then consolidation, and now we just have revisions at the edges or in limited/experimental environments.

Consumer desktop computing isn't quite a done deal, but it is an area where you're likely to see less and less radical innovation outside of restricted experimental spaces.
posted by holgate at 4:19 PM on May 14, 2012 [3 favorites]


Two words: backwards compatibility.

The general expectation that an OS will be able to run programs and open documents created for previous versions, and do so pretty seamlessly, vastly, vastly multiplies the time, cost and complexity of any major interface innovation.
posted by psycheslamp at 4:47 PM on May 14, 2012 [1 favorite]


Linux is innovating like crazy. Take a look at Unity or Gnome 3. Whether or not this innovation is actually an improvement is a ongoing issue of discussion in the Linux community.
posted by COD at 5:27 PM on May 14, 2012


Keep an eye on Apple's OSX. They're taking a bunch of cues from their wildly successful iOS (which itself is outside the scope of your question) and introducing ideas from there. Whether this is progress is hotly debated at the moment.
posted by ChurchHatesTucker at 6:21 PM on May 14, 2012


I would think features like a super-smart Siri,

Think about this for a moment-- Siri might not be "super-smart", but it's pretty smart, right? How often do people use it? I use it just occasionally. There is a bunch of stuff it's simply not suited for (eg, browsing).

There has been an awful lot of work done in new and different interfaces, tangible interfaces, 3d interactive interface devices/controllers, gesture recognition, etc., etc., etc. Outside of niche uses or specific applications (eg, Nintendo Wii), it just hasn't taken off. At a certain point, it's like asking why cars still use steering wheels.

I think the difference comes in how we interact with general-purpose computers and how we interact with purpose-specific devices. It's the latter category where you're going to see the most interactive innovation. I think the Desktop OS is going to start fading out in favor of iPad-like devices or other consumer devices that will have innovative interfaces. Not that I think the Desktop OS is bad, just that it's at an evolutionary endpoint.
posted by deanc at 7:16 PM on May 14, 2012


As COD explains, there is a lot of innovation going on in Linux, and especially in Ubuntu. The new release of the operating system includes a tool called the HUD (Heads-up Display) that I think really has the potential to change the way we interact with computer programs. Essentially, when running any program, you can tap Alt to open the HUD and then just search for what you want to do, rather than having to click through a menu.

For example, if I'm browsing the web and I want to find that bread recipe that I bookmarked, I just tap alt and type "bread" into the search box, and it pulls that up as an option. This is really useful when working with complex programs with a lot of sub-menus - it allows you to find things without having to remember where exactly they are. Also it can help with discovering features: you just type what you want to do, and the HUD will bring up what it thinks are the relevant choices. It's not perfect, but this is just the first iteration, and it will get better for sure.
posted by number9dream at 8:18 PM on May 14, 2012


It's not perfect, but this is just the first iteration, and it will get better for sure.

And it's an implementation of concepts found in Quicksilver on the Mac, and, going a lot further back, the Remembrance Agent in Emacs.

I take your point: there's a shift of sorts away from nested file/menu hierarchies towards unified incremental search with natural language components (assisted by its presence on the web) but it's somewhat marginal: the dream of the desktop filesystem 'going away' hasn't been realised.
posted by holgate at 9:22 PM on May 14, 2012


« Older How do I thank multiple interviewers for a single...   |   Startup Weekend questions Newer »
This thread is closed to new comments.