I agree #3 would be recording features, part of which is the rules.
Reliable Series recording is a major issue. Without reliable EPG data, and rules to identify that a program belongs to a particular series, series recording is difficult.
Several PVR/DVR suppliers over the years have solved this problem by providing the EPG themselves, or forming a close relationship with the supplier, and implementing a "
Series ID" for every series ever created. Zap2it, Rovi, (and by implication, Microsoft), IceTV, and I believe earlier software solutions used this method. TheTVDB does this, but it is maintained by volunteers (well, mostly it seems), and is usually updated after the fact, rather than before a program is shown, which is when recording rules need to select the program to be recorded.
I was looking at Rovi recently. The company, their vision, products, and geographic coverage. I can see why Microsoft shifted to them. Integrating with their product set would solve both the EPG issue, and possibly the recording rules, series recording, and add a discovery process for new media, suggestions and recommendations, and lots of additional information about the media. Not to mention that Rovi covers all media sources, including new streaming sources etc. They don't provide content, just information about the content. I would recommend that someone looks at what they do, and consider whether JRiver should become a partner. I believe that they provide their products to other companies, which then supply a service to end users. It may fit well with MC.
The alternative to implementing a Series ID that is available for query before a new series starts is to have much stronger recording rules, and maybe to use more sources to enhance what data is known about scheduled programs. Some of my quick thoughts;
1. A recording rule should have its own unique name defined by the user, which is
not used as the search criteria for a program.
2. Any recording rule should actually be made up of one or more rules, where some rules may include a program for recording, but others may exclude it from being selected for recording. This is partly supported currently, with title matching as the primary selection criteria, but also repeat avoidance for both scheduled and already recorded programs, selection by channel, range of channels, or all channels, and selection by a range of time criteria. I think this needs to be extended in the area of the primary selection criteria. ie: One rule to select Hockey, one to include Boston Bruins, and one to exclude any location other than XXX.
3. It should be possible to create a rule with simple or complex rules. Ideally, I think there should be three levels: a) Simple word or part word match, on any EPG data field. b) Boolean matching, with AND, OR, NOR etc. operators, where a true result means the program is recorded. c) Regular Expressions (or similar) for complex disassembly of program Titles and Sub-titles, and possibly descriptions, categories, actors, years, etc., in fact all available EPG fields.
4. I think there needs to be some fuzzy logic around criteria matching, such as matching abbreviations, poor spelling, phonetic matching, different forms of names, and so on.
5. For all the above to work, the more accurate information about programs the better. The EPG used by MC needs to be improved after it is sourced, and before programs are selected for recording. Looking up TheTVDB after a program is recorded is too late. TheTVDB, and maybe the IMDB, themoviedb, and others, should be looked up to enhance the EPG, where there are gaps in major fields, such as categories, keywords, season and episode numbers.
Requirement #4 would be to improve the information available in the Guide through
colour coding of program bars (Movie, Sports, Documentary, etc.), different coloured spots or icons on program bars (to be recorded, recording, will be recorded as part of the padding around another program, etc.), maybe a pop up details on a program when it is highlighted. Also
channel logos in addition to or instead of the names. Probably lots of other things to turn the EPG data into information. Again, this largely depends on the quality and volume of EPG data, but there is also a lot of information that MC knows which could be communicated visually.
I am sure that there is much more that current WMC users could contribute. Overall, except for encrypted content, I think MC has the presentation of content that is on a users system covered. But the future is streaming, IRTV, and other sources, and that is where MC could go beyond WMC capabilities. I haven't looked at what the newer XBoxes are supposed to be capable of, but perhaps that is a yard stick to look at as well.
My observation of WMC users is that, like any population, some people want a one click setup and never to have to think, and other people want to fiddle and fine tune everything. I have noticed though that many users spend a lot of time and effort to get WMC to do what they wanted, and were willing to learn how to do that.
Sorry for the big block of text. Just another brain dump I'm afraid.
