[vdr] VDR is obsolete

[Benjamin Forgeau <forgeau@tu-harburg.de>]

	Hi,

	Sooooorry for the subject of this mail! I just read parts of the 
historical discussion between Linus Torvalds and Andy Tanenbaum about 
the pros and cons of the internal structure of linux:
	http://www.oreilly.com/catalog/opensources/book/appa.html

	Some weeks ago I posted a very long message, without too much 
reaction. That's why the subject of this mail is this provocative, 
please don't consider it too seriously...

	Good, but the point is similar to the one of this historical debate:


	***	VDR is too monolithic! 	***


	Sure, VDR is robust now, the user interface exists and allows a lot of 
things, the plugin system is smart and well-documented, now VDRnews is 
gathering everything about VDR in one place (such a great Idea!!). VDR 
reaches completely the goal it wants to achieve: being an extensible 
Video Disk Recorder application.

	But VDR shouldn't be the only big application around, even if it 
implements the main DVB activity: watching and recording TV.

	As long as VDR does not use any supported *separate* library, it's 
almost impossible for such a library to survive, because it's useless 
to support and modify a library which is not used in the main 
application.

	Sooooo, that's my question:


	*** Why not putting non-VDR-specific code in a separate library? ***



	Or what about using libdvbsak as base library, merging parts of VDR in 
dvbsak?? Is there a real good reason, apart from the substantial 
re-organization time overhead???


	In a separate library, I would await:

	* Finding the DVB device and opening it

	* Tuning the card on a transponder taken from a global list

	* Parsing stream info

	* Starting and stopping video play

	* Other easy basic things in order not to re-invent the wheel


		Thanks for your attention, and thanks for any answer!


			Regards

 			ben


--
Info:
To unsubscribe send a mail to listar@linuxtv.org with "unsubscribe vdr" as subject.