OSD Roadmap&History

From The Neuros Technology Wiki

Jump to: navigation, search

Neuros Technology | Products: LINK, OSD, Tablet | Developers

Ant small.jpg
    THE INFORMATION ON THIS PAGE SHOULD BE ENTERED INTO BUGZILLA
    See also Bugzilla_OSD for more information.

Contents

Background

The vision for the OSD is pretty well summed up by its name, the open source device. I personally have a very strong belief that we can build a business around an open embedded device that can serve as an embedded multimedia hub or media center. Something that sits squarely between the closed dedicated, embedded devices such as Tivo, slingbox, the Digital Media Adapters, Sonos (all TMs of their respective companies) and a full blown, loud, expensive PC.

What this really means to the product roadmap is a dedicated, focused effort on developing strong software for the device. You will see that some of the enhancements below are significant projects. If Google or Microsoft or any other "software company" launched them, you would call them new products.

I believe this is really is a big key to our success. We need to be able to launch new software "products"? quickly, and we can do so without new hardware or housings or FCC approval. These "products"? will build on the value of the device and expand its market, and really support the "media center"? position of the Neuros. At the same time, we also internally must view these as real products. There is something of a 50%/2% rule in that you can get a "feature"? working 50% completely in 2% of the time that it takes to make a real product. If we treat something like an online video integration project as a feature and just get it working, but not very well, we won't learn much. We have to commit a period of months to a project to make it a real product that we can send to reviewers and not be embarrassed about.

In addition to all the below, there will be dozens of details that will need to be addressed that don't even make the roadmap. Bugzilla is the right tool for bugs and many enhancements, this page is intended to outline the direction of the product in "broad brush strokes" to realize the vision (in the short-term) for the product as an "embedded media center". JoeBorn Jul 4, 2006 23:32 CST


    NOTE For a longer term, more freeform view from the community, check out OSD 1.0 Ideas.  


In loose chronological order, here is initial roadmap of the OSD. I suppose we'll cut and paste these to a history or features section as they are completed.

Electronic Programming Guide (EPG)

Our first priority needs to be making the EPG/Blaster realistic for everyday use. This is a key enabler to broaden the appeal of the device.

  1. Choose Guide Provider- There are a number of them, some free, some charge, some work for different regions than others. We need to research which one is right for the OSD
  2. IR Blaster/Serial enabled “Fast Learning� IR Blaster With intuitive UI for easily programming Blaster-

An IR Blaster is currently an intimidating, daunting thing for the average user (although not necessarily for the folks reading here perhaps). We need a simple UI where the user just points the tuner remote at the OSD and hits a few buttons (guided by an on screen display) and the OSD “learns� the codes. This can be a big selling point. An IR blaster could potentially allow “blind� attachment of the OSD allowing use in some cases where the TV is actually the tuner (no set-top tuner box).

Networking (including security)

The EPG will work through Ethernet, so we'll need to get this up and going.

  1. Wireless- Engineering and Sophia need to chime in on this one
  2. Source Wifi Dongle
  3. Get Drivers Going

Enhanced APIs

Framework Additional hooks, whatever per developer requests This is largely an ongoing project, but I put it explicitly on the list to emphasize that this is vitally important. Supporting the community is vitally important, and we need to support our community just as we do internal employees. Its clear that we have some volunteer programmers that are more talented and productive than even some of our own internal people. We as a company need to make sure we treat them accordingly. I consider this an important part of my job -JoeBorn. There was considerable discussion of this at the IRC meeting between Raster and Bob, look about halfway down 6-13-06 IRC Log. Someone needs to fill in this vague notion with some details.

libao is an excellent starting place, allowing us to leverage many existing software packages. It shouldn't be hard to either patch in or (once again) partner with Xiph.org to bring a version with support for the neuros media API. Srobertson

(Note: libao seems well underway, see "Build your own Music Server with the OSD"

Local Network Streaming

Create UI for enabling easy file browsing/streaming from the network of photos, music and video. We need to implement a UI that allows for file modification, moving and deletion.Ideally this happens without requiring PC software, for windows machines can a computer on the LAN just "share" a folder to make it browsable by the OSD? JoeBorn

For music, the Digital Audio Access Protocol is gaining traction, especially because implementing that allows one to read straight from iTunes, no configuration necessary on PC or Mac. Or so they say, I've never tried it. For photos and video, we have a couple of options. Samba, which supports Windows-native file sharing, is a *beast*, but it's common. I'd have honest concerns about the overheads of full-strength Samba on a DM320, though. Alternatives include (but are not limited to) XMBSP, the Xbox Media Center protocol designed specifically for streaming this kind of stuff (PC, Mac, Linux daemons already out there, but must be installed on host PC's before use, plus no write support). Srobertson

Video Sharing

Online video is big and getting bigger. There have been some interesting articles comparing youtube to a TV network in terms of show views (it's not apples to apples, but you get the idea). They have millions of users, and all are tethered to their PCs. Youtube doesn't allow downloads, so there's not really any other way to view the videos. While there are lots of other things that the OSD could do, such as connect to myspace, etc, most of them are interactive and give the edge to a computer because it has a keyboard. Mouse etc. online video is different, but it's less interactive and more browsing oriented, a much better fit for doing on the couch with a remote. It's not hard to imagine it being a big hit among you tube fans. This is certainly an example of a “product� as discussed above. I'm sure it will be easy to get some semi-manual hack working, but for it to be meaningful we'll have to implement a lot of interesting features of the service. Research will have to be done and specs developed (or at least understood in someone's mind). Why do people use this? What will be different about doing it on a TV, etc. Above all else, a requirement will be some kind of open, published format from the website. The last thing we need is to get embroiled in some kind of business development negotiation. The minute that happens, suddenly you see snails whipping by as your pace slows to a crawl as you feverishly fly around the country accomplishing nothing. Fun! Apparantly youtube uses macromedia flash? Any other insight?

I've written a proposal for one way to do integration with YouTube and/or Google Video, the two biggest online video providers. Technically, it shouldn't be too hard, as the content is stored in well-documented (if proprietary) formats that should be easy to use (or even break if necessary). Licenses are fairly permissive for those two services, so we shouldn't have any major trouble, but it'd be nice to ask them first anyway. (We may have to distribute this module as an aftermarket add-on, tho, because of non-commercial clauses in both of those sites' license, which would mean finishing the Package Management Proposal, which is at this time horribly half-baked.) The only forseeable problem is in relying on screen-scrapes, as that puts us at the mercy of a site redesign, but that can't really be avoided. Srobertson

Photo Sharing

This one is very much like the video sharing, make it work well and automatic and with all the features you want and it's great. Do it halfway, it's worthless. Remember the PC is only a room away, the minute it gets a little bit slow or cumbersome, it's worthless. from Flikr, etc

Internet Radio Streamer

--Probably shoutcast or some open standard, again think of this as a product.

Personal tools