Hello All,
I'm looking for a way to add MythVodka to my current 0710 setup, so I can take advantage of the streaming media via Hulu and the BBC. I started working on it, but I realized that there isn't a mythplugin directory. Can anyone let me know what if there is way to make this happen. Below is the link to the MythVodka site for some background.
Thanks,
Ern
http://www.mythtv.org/wiki/MythStreams
I didnt even know this existed, so I am willing to take a stab at it. Cant even figure out the right collection of packages to install to get it to compile so far though...
I figured this would be a nice alternative for those trying to get rid high subscription cable/satellite costs.
I found the link below while trying to figure out a way to make this happen. The process described is a little lengthy, but it may offer some ideas.
http://forum.linuxmce.org/index.php?topic=1981.0
Perhaps I should rephrase my initial question. Will there be a way to install mythplugins in the 8.10 release?
No.
-Thom
Thank Thom.
Quote from: tschak909 on June 16, 2009, 10:57:11 PM
No.
With the greatest respect, may I enquire why not? Is it just due to the complexity of the system and lack of dev time, or is there an underlying reason?
I only ask as, whilst I accept and endorse the view that we shouldn't be adding "taped on" front ends to corrupt the whole integrated MCE look and feel, there are many back-ends that can enhance the product. MythVodka is one, but so is TVWish and I believe some plugins that can transcode recordings for the iPod / iPhone and publish as an RSS feed to iTunes. There are, I'm sure others that would greatly enhance the functionality of MCE without moving away from the existing (very nicely integrated) front-end.
Mostly because we are still trying to integrate MythTV better into Orbiter, and none of the plugins will be needed once more of MythTV's functionality is exposed to Orbiter.
Why don't you developers out there concentrate on integrating such functionality into LinuxMCE? All the pieces to do it are already part of the system, and can be done with much greater flexibility than what is present in these MythTV plugins that people keep pointing to.
-Thom
Quote from: tschak909 on June 16, 2009, 11:56:03 PM
Mostly because we are still trying to integrate MythTV better into Orbiter, and none of the plugins will be needed once more of MythTV's functionality is exposed to Orbiter.
Why don't you developers out there concentrate on integrating such functionality into LinuxMCE? All the pieces to do it are already part of the system, and can be done with much greater flexibility than what is present in these MythTV plugins that people keep pointing to.
-Thom
Hey Thom...you cant honestly blame people for wanting existing MythTV plugin functionality when they see it. If we integrate an application like MythTV, VDR...Asterisk even that all have a wealth of existing plugins then people will want to use them I'm afraid. The only way around this situation in my opinion is to directly implement a full live TV/PVR/IPTV capability directly into LinuxMCE and then encouraging people to implement extension functionality as DCE plugins will be a breeze ;-)
Andrew
Quote from: totallymaxed on June 17, 2009, 12:14:32 AM
Quote from: tschak909 on June 16, 2009, 11:56:03 PM
Mostly because we are still trying to integrate MythTV better into Orbiter, and none of the plugins will be needed once more of MythTV's functionality is exposed to Orbiter.
Why don't you developers out there concentrate on integrating such functionality into LinuxMCE? All the pieces to do it are already part of the system, and can be done with much greater flexibility than what is present in these MythTV plugins that people keep pointing to.
-Thom
Hey Thom...you cant honestly blame people for wanting existing MythTV plugin functionality when they see it. If we integrate an application like MythTV, VDR...Asterisk even that all have a wealth of existing plugins then people will want to use them I'm afraid. The only way around this situation in my opinion is to directly implement a full live TV/PVR/IPTV capability directly into LinuxMCE and then encouraging people to implement extension functionality as DCE plugins will be a breeze ;-)
Andrew
Besides, why re-invent the wheel if you don't have to? I'm assuming the whole reason Myth / VDR / Asterix et al were used in the first place was that it seemed silly to reproduce the effort those developers had already put in.
I, for one (as I KEEP saying) wholeheartedly agree that supporting front-end plugins would be a retrograde step and detract from the whole single-look Orbiter concept that is at the heart of MCE, but I honestly cannot see how integrating Myth (and / or VDR) into Orbiter is going to provide the back-end functionality provided by plugins. I have mentioned a few, there are others that would greatly enhance some people's setups but not change the look or feel of the system. Already mentioned here are MythVodka (allows streams from the likes of the BBC to be brought in), TVWish (list of films / series to look out for and record if they are shown) but then there's stuff like Myth Recording Extender (MRE) (Alters the end time of a recording to cope with overrunning schedules) and various iPhone integration - that's just a quick scan! On VDR, we have a similar range.
Surely, it takes work to
remove the plugins from the standard Myth / VDR / whatever distro?
Dude, you can do whatever you want to your install, but we have to keep things supportable.
-Thom
Thom's points are extremely valid and are the basis why this is such a powerful platform.
The whole design of the system revolves around evrything being represented by devices. Those devices add capabilities to the entire system. The system should provide a user interface to consume those capabilities. That user interface should function the same way (it might look slightly different) no matter what way you access it - OnScreen, PDA, Mobile Phone, Web.
Core capabilities that the system provides at the moment are LiveTV viewing and PVR functionality. These capabilities can be provided by either MythTV components or VDR components. Ideally the user would have exactly the same experience regardless of what components are used. Currently - that is not ideal in that you see bits of the mythTv/VDR frontend when using the liveTV or PVR capabilities. Thom has been working on cleaning up some of the MythTV implementation and Andrew (and team) has been working on cleaning up some of the VDR stuff.
Most of the plugins for MythTV are additional capabilities that in some instances are already provided by linuxMCE (eg ; MythVideo, MythMusic etc). Additional capabilities (in particular Internet Streaming) need to be usable from linuxMCE through the standard way. There is an excellent framework for enabling this and as Thom keeps highlighting we need more and more developers embracing this so that we can add and support new capabilities as quick as possible.
So if you really need to use Hulu in the short term and want to manipulate your installation so that is accessible from the MythTV menus and UI then it is probably achievable but that by itself will not add the capability to 'linuxMCE' and cannot be supportable by the project (like Thom said). And the problem with doing it and putting it on the wiki (as a hack/duct tape) gives the perception that linuxMCE has support for that capability. This will detract from actually getting that capability into the system properly.
Thom's efforts to share knowledge that he has gained through his time investment(workshops/webcasts) are a great step to get more people exposed to the framework we have and how it is best extended to cater for new capabilities.
Anyway enough of my rant.
Darren
ps: I am getting much closer to being in a position to do some dev work again with the new bub becoming less demanding and I not have the hardware ready for a dev setup in the workshop... so fingers crossed.
Thanks for all the input guys. If there are no plans in the works for mythplugins, I can hack my current install to make work. I wish I had more to contribute in the way of development, but I'm just starting to get my feet wet in programming, and it may be a while before I can add that level value to the project/community.
thanks again.
E!
Quote from: darrenmason on June 17, 2009, 05:26:47 AM
Thom's points are extremely valid and are the basis why this is such a powerful platform.
The whole design of the system revolves around evrything being represented by devices. Those devices add capabilities to the entire system. The system should provide a user interface to consume those capabilities. That user interface should function the same way (it might look slightly different) no matter what way you access it - OnScreen, PDA, Mobile Phone, Web.
Core capabilities that the system provides at the moment are LiveTV viewing and PVR functionality. These capabilities can be provided by either MythTV components or VDR components. Ideally the user would have exactly the same experience regardless of what components are used. Currently - that is not ideal in that you see bits of the mythTv/VDR frontend when using the liveTV or PVR capabilities. Thom has been working on cleaning up some of the MythTV implementation and Andrew (and team) has been working on cleaning up some of the VDR stuff.
Most of the plugins for MythTV are additional capabilities that in some instances are already provided by linuxMCE (eg ; MythVideo, MythMusic etc). Additional capabilities (in particular Internet Streaming) need to be usable from linuxMCE through the standard way. There is an excellent framework for enabling this and as Thom keeps highlighting we need more and more developers embracing this so that we can add and support new capabilities as quick as possible.
So if you really need to use Hulu in the short term and want to manipulate your installation so that is accessible from the MythTV menus and UI then it is probably achievable but that by itself will not add the capability to 'linuxMCE' and cannot be supportable by the project (like Thom said). And the problem with doing it and putting it on the wiki (as a hack/duct tape) gives the perception that linuxMCE has support for that capability. This will detract from actually getting that capability into the system properly.
Thom's efforts to share knowledge that he has gained through his time investment(workshops/webcasts) are a great step to get more people exposed to the framework we have and how it is best extended to cater for new capabilities.
Anyway enough of my rant.
Darren
ps: I am getting much closer to being in a position to do some dev work again with the new bub becoming less demanding and I not have the hardware ready for a dev setup in the workshop... so fingers crossed.
I totally agree with all that but... the fact remains that if we want to subsume the UI of MythTV, VDR or any other app providing some functional user level component of the system (ie a piece that needs a Ui for a human to interact with) then we should not be using apps that assume 'they' are the UI. To enable what you describe cannot be archieved with either MythTV or vdr in their current form with a ton of the very 'Duct tape' Tom is always moaning at everyone for using, as both these apps cannot be fully decoupled from their UI's. The only way is to build of the the low level libs that for the basis for these two apps (for example v4l etc) and build TV centric capability directly into LinuxMCE...in my opinion that is the only way we will have a 'Duct tape' free implementation.
Until that time...trying to make MythTV or vdr be something they were never designed to be will inevitably involve ever more 'Duct tape' to hide it!!
All the best
Andrew
Andrew,
Understood, but I think at least MythTV has the potential to be well integrated (not familiar enough with VDR to comment). Mythtv completely seperates the backend from the frontend. There are various other 'frontends' that talk to the backend. I believe linuxMCE may need to entirely replace the frontEnd but there is no theoretical problem with that as far as I can see. At the moment we are relying on functionality that is only found in the frontend but perhaps that can change.
regards
Darren
Quote from: darrenmason on June 17, 2009, 11:50:08 PM
Andrew,
Understood, but I think at least MythTV has the potential to be well integrated (not familiar enough with VDR to comment). Mythtv completely seperates the backend from the frontend. There are various other 'frontends' that talk to the backend. I believe linuxMCE may need to entirely replace the frontEnd but there is no theoretical problem with that as far as I can see. At the moment we are relying on functionality that is only found in the frontend but perhaps that can change.
regards
Darren
Sure vdr can be run as a backend too... but the fact is you cant build a suitable UI (at least I dont believe you can) by then just building a few new Orbiter screens etc and 'plumbing' them in. Our feeling increasingly is that the Orbiter needs significantly extending to allow for this as...and thats a non-trivial endeavour ;-)
Andrew
Andrew - you have more experience than I in actually attempting this integration so I respect your view on that, however it is an interesting discussion that although getting OT I think is worth exploring more....
Some of my views (could be flawed?);
The system provides three core capabilities that are implemented by MythTV/VDR.
Live Viewing:
With the main system capabilities being;
* Display the selected video stream
* Controls (Pause, review, etc)
* Change Channel/ Source
* Record current
* Information display (typically from EPG)
Schedule [PVR]:
Main system capabilities;
* Browse EPG
* Create recording schedule
* Edit recording schedule
Recordings:
Main system capabilities;
* Browse / search recordings
* Display the select video recording
* Controls (play, pause, rewind etc )
* Manage ( keep, delete etc )
Looking at each of these. LiveTV is interesting and I think that this capability should extend to the viewing of any form of video feed. The display of the video stream is the bit where I think we need to be interfacing directly with the MythTV backend. I think this is not beyond the scope of current orbiter. The next three are all capabilities that don't require UI aspects and I believe the Information Display should be possible if the information can be effectively passed as data from the backend. The PVR software in this instance could be one of many linuxMCE plugins that provide these capabilities. Other plugins could also provide the same capabilities from other sources - like Internet streams, WebCams, Captured video feeds from settop boxes etc. Perhaps a separate plugin/player combo is needed to sit on top of these and group together all the channels/sources and manage what plugin is serving up the video at what time.
Scheduling is where perhaps we would struggle with constraints in Orbiter for EPG browsing but from what I gather Thom has made some advancements here using the dynamic datagrids etc. The create/edit recording capabilities might involve some work - but not insurmountable
Recordings should entirely fall into the same mechanism as our current 'Video' capabilities. This might need some additional features/functionality to get it to the state that MythTV is but I don't think that Orbiter would limit this too much (could be wrong)
The capabilities that still present a problem (in my mind) are features such as commercial skipping and the big one (for me) is support in the Recordings area to watch/stream recordings that are still recording. If we are entirely using the 'Video' functionality then media DB would need the incomplete streams added and the player (Xine) would need to be able to play these.
I am keen to explore a few of these ideas whenever time permits. My knowledge is currently deficient in some of the HADDesigner stuff. I have only explored adding/moving a few buttons here and there. I am interested in how datagrids can be used.
Anyway - certainly going OT and is better aligned with a development thread but makes for interesting discussion I think
Darren