Firefox OS for TV: UX Overview
blog.mozilla.org
blog.mozilla.org
Surely Mozilla have native English speakers who can at least edit their blog posts? Anything that goes out to the public affects Mozilla's reputation, and should therefore be taken seriously.
> Firefox OS for TV is a new direction that base on the vision of Connected Home from Mozilla.
I mean, c'mon. It's hard to take the post seriously when they didn't even bother to proof-read.
Certainly, but we also try to let teams and individuals operate fairly autonomously, and I believe our open, honest humanity is one of our greatest strengths as both a global community and as a company.
Let's put this in context, though: This isn't a press release, nor is https://blog.mozilla.org/ux/ a press channel. Rather, it's someone openly sharing the UX team's ongoing work on their team blog, much the same as you'll find at https://blog.mozilla.org/creative/ or https://blog.mozilla.org/services/. Could Shiqi's posts have used a pass by a native English speaker? Certainly. But I'm happy to be part of an organization that shares openly at the occasional cost of mistranslations.
Plus, I'm sure the UX team will learn from this and adjust their review process accordingly. :)
Keep up the good work.
> Let's put this in context, though: This isn't a press release, nor is https://blog.mozilla.org/ux/ a press channel.
That distinction may be well known internally, but nobody else knows or cares to learn about it; we're going to base our judgement on impressions: The page says "Mozilla UX" in big letters at top and it's on the mozilla.org domain; to me and I suspect to almost everyone else outside Mozilla, it's official Mozilla communication.
Traditional TV service has the advantage of shortcuts by way of being able to jump to a channel directly using the number pad, but the top set-top boxes don't have that affordance because they generally lack number pads.
TVs aren't really great for interacting with - so a better approach is not to interact with them at all and do the interaction on a different, closer touch device.
This might be the former TV-addict in me speaking, but would you be able to change channels and adjust the volume without taking your eyes off the TV screen? Because remote controls in their present form are just perfect for that, and have been since at least 20-30 years. In other words, don't get rid of physical buttons unless you really, really need to.
That being said, I'd prefer my TV controls to be at least mostly tactile/physical. At least for the TV operations. I'd be thrilled if instead of menu arrow button hell I could just set things from my smartphone.
[2] https://www.youtube.com/watch?v=TyYBJTFVOo0 [1] https://www.fan.tv/
I've only tried it in passing, but it's definitely a big step up from the standard remote.
1: http://www.cnet.com/news/samsung-smart-remote-hands-on-with-...
I personally use a wireless mouse with a trackball on the PC running my living room TV (Slackware, XBMC); previously some generic Chinese handheld, but now I use a Logitech M570 (which is what I also use on most of my desktops). Works much better than having to mash directional buttons all the time.
http://i.imgur.com/XEn3vBk.jpg
Note the overlay over the TV image, and the salmon colored pointer near the vudu icon. No grid and no directional arrows.
On the other hand that stylish ribbon design looks like it would work well with directional arrows, 1 right for Netflix, 2 left for CinemaNow, quickly gets into your muscle memory.
I really wish the trend was TVs becoming less smart, more like a monitor than a computer, especially now that TV "pucks" like Apple TV, Roku, etc are gaining popularity.
How so? Ever since I switched to a Smart TV I pretty much removed all other boxes that were standing around. My Samsung solves all the problems I had separate boxes for before. The only thing that is still connected is a PS4.
Let me chose the smart box I want, and be the best dumb display for that box you can be, rather than giving me a bunch of poorly engineered over-expensive cruft that I neither want nor need.
For some reason they keep adding stuff that I don't want to the device. And "smartness" is the worst of it. I plan to keep the display far longer than the Roku box (or whatever).
Consider: is your TV going to be able to play H.265 content when it become available? Chances are your manufacturer won't even upgrade your software - they've got a dismal record for this. But even if they do, you probably don't have the CPU power to handle it. To avoid replacing the display, you'll wind up buying a Roku or Chromecast, and (as was said up-thread) fighting the TV's UI to get to your new box.
I just use a basic home-theater-in-a-box receiver for 5.1 sound and because it has a built-in Bluray player (which I have yet to use) and like 3 or 4 HDMI inputs. So I run the cable box into one input, Chromecast into another, game console or HTPC into another, etc. and then output video from that into the TV.
So in the end, I only use my TV as a display even though it has other features built in. Having all sources go into the receiver means I can just switch inputs and audio follows video. If I want to watch Youtube or Netflix or listen to streaming music or watch something on my NAS via Plex, I use the Chromecast. If want cable TV for some weird reason, I switch to the cable box, etc.
But yeah, a large format display should last me longer than the life cycle of the "smart" capabilities or streaming services. When I bought this TV, there were several streaming sites that didn't exist yet or I didn't use. I want to be able to add or remove hardware as I need it and I want to use services that are currently valuable to me (and not have to skim past all of the integrated ones I never use or don't exist anymore).
Right now the two options I see are to use a low-powered HTPC that is still more functional than anything in a smart TV and can be upgraded, or something small and cheap like a Chromecast or extender that lets me tap into sources on my network or on the WAN.
These don't seem to be available.
I have one, connected it to the internet once, disconnected it and did a firmware reset. Never again.
That particular thing - a Samsung, mind you - installs adware/crapware/shit when you first connect it. We can argue about the featureset in general and if the 'smart' part is usable in general (I .. don't think so), but it is especially annoying if you look at the device, connect it and see it installing the worst of the worst stuff, pinning it to the most prominent places in your menu.
Tech support (yes, I was desperate to call them) said "Yeah, that's a feature", "No, you cannot prevent that from happening", "No, you cannot hide/remove/move those applications and icons". Disconnect, firmware reset, no Samsung TV in the future, ever again.
I understand/guess that this is a regional thing (Germany here). No clue WHY Samsung things it can sell out their customers like that, but all of the "content" was German crap.
I only ever go into the smart functionality by mistake now - I find the Apple TV to be a way better experience (and I have a few iDevices so AirPlay a fair bit of stuff also).
But mostly, as it is still Firefox OS, I don't see how people will be able to do cool multimedia applications, without access to native code and threads. Implementing DLNA, SMB, AFP shares, listening to network devices, hardware decoding, perfect audio/video synchronization requires threads and probably native code.
Limiting to HTML5 videos is going to be a very light. Or they will re-implement all the media center and multimedia code themselves, supporting all codecs and formats (and subtitles)?
I could somehow understand the limitation on the phone, but on a TV, multimedia playback is the massive use case.
Generally on Smart TV's the video decoding isn't handled by the browser/render, instead handled by the underlying media player and hardware decoder. Pretty much all the TVs (Samsung, LG etc) use HTML5 video with much wider support for codecs then you see in browsers. HLS/H.264 is commonly supported.
Similarly they tend to use mechanisms such as custom plugins and JavaScript interfaces for DLNA/SMB etc. Not sure why you need threads in particular, there's already a lot of JavaScript mechanisms (events, async calls) to support these types of features.
If you want extra codecs, media players or media center, you need to write them in native.
Matchstick developers have gathered quite a lot of support for their croudfunding campaign, by promising an open software and hardware TV streaming stick due this February, and for a low price ($16+). Then, come February, they pissed off the vast majority of their supporters by suddenly announcing the 6-month delay in favor of things like hardware upgrade and DRM support, which none of their backers actually asked for. If it turns out that they were not in communication with Mozilla and their vision for Firefox OS TV's is incoherent with Mozilla's, then this project will truly end up a poor application of users' money and trust, bordering on a scam. Actually, a lot of supporters have lost faith in crowdfunding, and, presumably, Firefox OS, as it is; demanding their money back - not that Kickstarter supports it, after the project's been funded.
I participated in the kickstarter. What I am interested in is less a physical product and more an open platform, but I suspect we won't actually be getting that.
You seemed to indicate that the DRM addition would make it not an open platform so though I'd share that as it still seems like a potentially positive situation.
*at least in Firefox
In regard to pictures 4 and 5, Roku tried the fully horizontal menu for a while, but wasn't great. It's too limiting. Think about using a vertical left and a grid right like they do. It fits more and given the right control, works pretty well to navigate, even if it isn't the prettiest.
And, if you only do one thing, remember the rule of simplification and reduce the number of visual elements required to convey the same information. Examples...
In picture 6, you shouldn't need icons strawn about the window left, right, and bottom. All kinds of space wasted for no apparent reason.
In picture 7, just say no to the advertisement looking things in the top bar and the useless title and description at top. If you must get their attention, make them go through a screen to have full attention on your singular announcement. Don't make noise.
In picture 8, having the blue under each black and white icon is a little distracting, and the text and icons have features too small to read or recognize easily.
While grids are nice they have a problem. They're great at navigating two dimensions but they aren't easy to get out of. Where do you put the settings and search buttons? Do I have to scroll through all the categories (like netflix does) to get to search?
Similarly how do you do subcategories?
This system actually solves both of those problems. What it lacks is information density. Something like this linking _into_ a grid could work.
Note I say this while our company that does this developed an app with this exact navigation scheme for Comcast cable boxes.
http://adifferentengine.com/cases/sochi/
But we had a very strict set of content so we knew it would work for our use case.
My opinion:
1. They go on the remote controller itself as magnifying glass and gear buttons, if you have that luxury. If there is no UI pointer, there is no need to provide anything else.
2. If you have a UI pointer, icons (gear, magnifying glass) or word links (settings, search), upper-right is most intuitive but lower-right may work also.
Its an annoyance to repeatedly use arrow keys on remotes.
Also, the Dashboard is a separate app. Like, um, in a car, right? Just like the radio and temperature controls are away from the user, so they have to look away from the road (their focus) to access it, right? Right.
What this UX actually needs, instead of being an app menu that can access TV or whatever, is TV-centric design. It needs a focus. Provide a viewport that apps can draw to, and allow switching of apps without having to "look away" from the content.
Folder tabs? Search? (Please don't add a) favorites menu? They're all acceptable ways to provide apps to users, but this is none of them.
You must be forgetting that most people use TVs for the video content. I'm not saying "oh, get rid of YouTube and Netflix," I'm saying "constrain yourself to video output because no one really cares about the web on a TV box." Yeah, cool, you can go to webpages that have totally broken design and are totally illegible on TV (hi, HN!). I mean, an RSS reader makes sense, because then the TV (and thus, the user) has control of how the content's displayed. But "FF on TV" sounds like a really bad idea to me, because the Web sucks on anything that isn't a computer.
> The usage of applications on smart TV is getting huge, and it needs an re-framing of the UI structure to make the application as important as TV channels, and as input sources.
No, it isn't. No, it doesn't. Unless you mean you're going to make my Netflix another channel just like Fox News, which would be pretty nice. But that differs in the fact that Netflix is a video focused application that acts a lot like a channel or something. But that means you had better add a unified UI for video playback. Netflix and YouTube and everything should share the same UI widgets.
A dashboard that can be activated on command is cute and can possibly be a good idea. Like, maybe show it for a moment during that period where the TV is adjusting to a new signal or something to allow the user to peek at it by default so they can see alerts ("hey, Hillary Clinton's running for Pres! I need to pull up the CNN Channel!"), and then allow them to later press a button to show a dashboard that doesn't force them out of their content. But if you push them out of the content, the illusion is broken. It's no longer a TV, it's a computer. A big, complicated computer, with lots of little nodes in its huge and complex menu. And only us nerds want that, because us nerds like intricate and complex things. (there's nothing wrong with catering to nerds, but I don't think that this is supposed to be directed towards them, because they could just as easily grab a Raspberry Pi and load up XBMC on it without having to invent their own new flavor of Linux)
PS: adding a focus means your UI will be necessarily shallow. It will force you to get creative when you want to get complex in writing a good UI. Snapchat, for example, has a focus on the camera. Swipe from the left, you get your incoming Snaps. Swipe from the right, you get your Stories. Swipe from the top to see stuff about you and your settings. And that's the extent to which anyone uses Snapchat.
Unless you mean you're going to make my Netflix another channel just like Fox News, which would be pretty nice. But that differs in the fact that Netflix is a video focused application that acts a lot like a channel or something. But that means you had better add a unified UI for video playback. Netflix and YouTube and everything should share the same UI widgets.
I think that's a big part of it. The distinct between "channels" and "apps" is artificial. On my Roku, most of the "channels" can be interacted with in at least some way, making it difficult to draw a line.
Most channels on Roku are offering on-demand features, and you interact with the channel to locate a feature. The Weather Underground channel has controls allowing search for locations, and viewing maps and radar. Some really big things, like Tablo, include features for programming recording; and Plex is pretty much everything but the kitchen sink.
I think if they saw that the apps/channels distinction is artificial, the rest of the design would collapse into a bunch of irrelevancies, and they'd be forced to confront the question for real.