Streamlining the browsing tree & multiple configs

[DavidHB]

By way of a change, I'll not respond in-line.

Firstly, I don't think that there is any significant difference between us about the value of index- or metadata-based browsing. Where my thinking, I believe, is different from yours is that I tend to regard the file system as just another set of metadata, one that is more universal within the computer machine but less well adapted to the specifics of a particular task (in this instance the creation and maintenance of a music library) than the metadata sets designed for that task. However, the very generality of file system metadata has value; as I have said before, the general and the specific are complementary and not mutually exclusive. I learned this during a decade of involvement with digital photography; the thought seems to be equally applicable to digital music.

If the file system is 'just' metadata, it can be changed like any other metadata. The genre level in my own library was an afterthought (I'm not necessarily quick to grasp the logic of my own conclusions!), but the lifting and shifting required to implement the change was accomplished in twenty minutes. Would it have taken much longer with, say, 600 CDs' worth of folders and files, rather than the 80 or so I did move? Actually, I don't think so. But I did do the moving around on a copy of the library that was not in use, and then used SyncToy to mirror the changes in the 'live' library on the NAS (and of course re-started MinimServer to get the library re-indexed).

Another conclusion from my digital photography work was that it is essential the correct data structure, software and hardware infrastructure in place before allowing the library to grow. If your experience is like mine, it is during the course of ripping and managing the first 100 or so albums that you learn what the 'correct' infrastructure actually is. I'm glad that I had mine in place before I committed to the purchase of the Akurate.

I take your point about browsing long data elements (of any kind) on devices with limited screen real estate. I could install install Kinsky on my Android phone, but I doubt that I will. A high resolution (Nexus 10) tablet is an essential part of my infrastructure.

You and I seem to have reached diametrically opposed conclusions from our shelf scanning past. You want to "throw off the restrictions" (and so the shelf paradigm must seem to you like backward thinking). I want to take the past as a starting point, but use new opportunities to make the system more capable, thereby (I believe) also removing the restrictions. Is one approach inherently better than the other? I doubt it (so long as you can at least find your files in something other than a Control Point ).

Are there circumstances where metadata-based browsing does not work? Well, yes. The obvious one is where tagging-type metadata is inaccurate, inconsistent or just absent. Getting the data right, even using the best software applications, is challenging and labour-intensive. Classical music throws a curved ball at the metadata writer from (I'd guess) at least one album in three. When, as from time to time it will, the tagging proves to be faulty, it is good to have the dear old omnipresent file system to fall back on. And ripping and tagging always involve a great deal of file system use and maintenance, so it is as well to devote as much care and attention to this as to other forms of metadata input and editing.

Your comments about JRiver interest me. I agree that the cost is not exorbitant, but, looking at the media-rich but content-poor website, I find it hard to form a view as to whether JRiver does anything I need for which I do not already have a capable application. Certainly, it deals with aspects of media (video, for example) in which I currently have little interest. But if it can tag files more easily than, say, MP3Tag (which is very powerful, but presents the new user with a pretty steep leaning curve), I might be interested.

David