Rejection of "ignore the" patch (split from the Evil_G unsupported build thread)

[Llorean]

Yes, TSOP should be used for "Artist" when sorting. In terms of ID3 tags, the Artist tag is actually the "Performer" tag.

[DefineByte]

Thanks.

I'm tempted to ask you all sorts of ID3 questions but I'll refrain.

[Llorean]

I'm hardly an expert on tagging. I just do research when a subject comes up, before I start on discussing the subject, and this is research I did months ago when the topics of "Albums with various artists" and "Sorting in the database" were discussed elsewhere.

Using the three stort tags would solve several problems at once, in a 100% multilingual way, without obvious negative side effects.

[lalittle]

Using the three stort tags would solve several problems at once, in a 100% multilingual way, without obvious negative side effects.

I guess the problem is that I honestly don't understand how a solution that requires the user to manually edit the tags on every artist, album, and song that starts with the word "The" is considered a "solution" when it requires a notable amount of user intervention.  This is especially true given that it would require most users to use an entirely new program since many if not most of the popular media management programs (iTunes, Media Center, etc.) don't allow access to the "sort" tags individually.  I think what's needed is a way of handling this with as little user intervention as possible.

That said, what if the sort feature:

1)  Sorted by ignoring articles at the beginning of the tag, UNLESS

2)  it saw a specific character pattern at the beginning of the "sort" tag (say a double tilde "~~" for example.)

This way, ONLY the "special case" artists/albums/songs would have to be manually edited -- i.e. you'd only have to add the double tilde to the beginnings of "sort" tags where you WANTED them to sort WITH the articles.  Since this is a such a small percentage of the artists/albums/songs, this would eliminate a HUGE portion of the tag editing necessary, and for "most" artists/albums/songs it would work "by default" with no intervention needed.  It would also only require the use of a new tag editing program for people that had "special case" titles in their libraries.

3)  The behavior could be turned off.

4)  The patch would be made to work on a "per language" basis, where it only ignores articles in the language currently being used.  This could either be done by have the patch check for what language is currently being used, or by allowing the user to select the language that this feature effected.

This would take care of the language situation, it would not require the editing of the majority of tags, it would sort "special case" titles correctly, it would not require iTunes or Media Center (or other programs) users to use another tagging program, and for "most" artists/albums/songs it would work "by default" with zero invtervention needed.

Would this be an acceptable solution to the developers?  It addresses all the issues brought up without necessitating a lot of tag editing, which would make it accessible to everybody, including people that didn't know about a seperate "sort" tag (which until I read this thread included me.)

Thanks for your feedback on this,

Larry

[DefineByte]

I'm hardly an expert on tagging.

I don't think anyone is. Disc subtitles, for instance, seem to flummox a lot of people.

[Llorean]

Why do you suggest doing it in such a huge hack of a way? Sorting without articles by default?

I've told you: THAT IS NOT THE DESIRED BEHAVIOUR. You're not going to convince me otherwise, so please stop trying.

Please, if you're going to continue beating your head against this wall, do it elsewhere. Convince Apple to add a feature into iTunes that duplicates the "Performer" tag into the "Performer Sort Order" tag, and removes the article.

A list of articles for each language is just a worst hack.

Again, I don't see how "My tags are bad" is an argument for "The Rockbox guys should do something they don't want to do."

YOU can fix your tags. We have our software how we want it. Why exactly do you think we should make something work a way other than how WE want it, when it's software designed for us, by us, and you don't pay for it?

Remember: You aren't the consumer here. Rockbox, ultimately, doesn't depend on users. Every now and then a decision will be made, and it's not a democracy. There are a few people (I'm not one of them) who have the absolute final say on how things go. Decisions get made, and in some cases people don't agree with them. You aren't paying the bills, and you're free to use any other firmware of your choice. Rockbox isn't invested in keeping you as a customer.

It would be nice if you stuck around, but keeping you around is not a reason for the core Rockbox devs to implement a feature that they feel isn't the right solution to the problem.

[lalittle]

Why do you suggest doing it in such a huge hack of a way? Sorting without articles by default?

I respect your request that I not beat this into the ground, but since you specifically asked, I'll answer this question.  Please understand that I'm not doing this in order to try and "convince" you of anything -- I'm simply trying to communicate my ideas on this subject more clearly.  I thougth this this may still be of interest to you regardless of what the end result is.

What we're disagreeing on is the set of parameters that are important in this feature, and what constitutes a "hack."  To me, one of the important aspects of this feature is that it should do the sorting without user intervention -- i.e. that it should not require a separate process of editing tags in order to achieve this behavior.  I understand that you don't agree that this is important, but it is important to me.  There is no "right" or "wrong" in this regard -- it's based on subjective opinion.  From my point of view it's therefore not a "hack" since your method does not achieve what I'm looking for -- it just doesn't address a specific condition that I see as important.  This NOT to say that you're method is "wrong" -- it simply is not a solution given my parameters, which are different than yours.

In other words, while we're both talking about the same END result, we're really not talking about the same thing.  To me, the process needs to include the concept of working with the NATIVE tags from popular programs.  Without this extra idea, the process doesn't achieve the goal that I'm personally looking for.  I fully understand that your method satisfies your needs, but our needs/desires are different.

I don't know enough about this to do it myself, but hopefully, others will continue to work on a patch that achieves the goal without requiring tag editing.  There is really no reason we can't both have the behavior we're looking for -- I may just have to use a patch that is not officially endorsed by the developers.

Thanks,

Larry

[DefineByte]

It should be easy enough to write a script in foobar2000 to create the TSOP tags for you (foreign languages not withstanding). Is the extra effort your only beef?

[GodEater]

From what I can gather, it seems *we* should implement a crappy solution in our software because either :

a) He can't find a tagging program that does the sort / display stuff
or
b) He can't be bothered to learn about one that does

[lalittle]

From what I can gather, it seems *we* should implement a crappy solution in our software

That's not fair.  I'm not saying you should "do" anything -- I'm stating an opinion that you happen to disagree with.  By having a discussion about this subject we can potentially understand something about the other's point of view.  Regardless of whether or not any body's opinion is changed, it's still productive to try to understand other points of view.  Calling it a "crappy" solution serves no purpose -- it's only creates an adversarial tone that makes it more difficult to have a meaningful conversation.  We can respect each other's opinions even if we don't agree.

because either :

a) He can't find a tagging program that does the sort / display stuff
or
b) He can't be bothered to learn about one that does

What's wrong with not wanting to be "bothered" with another step, even if it's a relatively simple one?  It still takes extra time, so it's still more "convenient " to not have to do it.  For example, I like being able to simply hit "record" on a TV program in order to record it rather than having to manually input record start/stop times.  It takes more code and an entire system to support it, but it's a lot more convenient to the end user.

I seen nothing wrong with wanting the software to accomplish "article free" sorting without requiring my intervention.  It's not because I'm "lazy," it's because it's more convenient to not have to go through the extra tagging step.  Please note that I'm NOT trying to convince you to do this -- you've made your position perfectly clear. I'm just responding to your post, which I personally did not think was a fair representation of my opinion.

From my point of view (which I fully understand is different than yours), eliminating the need for user intervention would be a fantastic aspect of this feature.  I merely expressed my opinion on this subject on an open forum where user feedback and opinions are welcomed.  It's one thing to disagree with me -- even strongly -- but please don't ridicule my ideas just because my opinions differ from yours.

As to your two conclusions,

You're correct that I don't want to have to use a separate program to tag my files.  I want to continue to use Media Center for my all my media needs (tagging, management, playback, etc.)  Even if Media Center DID offer the ability to edit the "sort" tags separately (which it does not), I would still rather not have to do this.  I don't call this laziness -- I call it wanting a more convenient solution that requires less work on my part.  It's simply about wanting a more robust tool.

And just to be totally clear -- I FULLY understand that the Rockbox developers do NOT have to take ANY of my suggestions to heart.  These forums are for the free exchange of ideas, and I'm simply offering my opinion for anyone interested.  When it comes to a project like this, I firmly believe that offering opinions and ideas are the cornerstone for positive developments.

Thanks,

Larry

[TedGamble]

For all you folks that are still depending on iTunes (or any other tagging software that falls way short of your expectations in terms of features), I have two words:  THE GODFATHER.  This FREE (my favorite F-word) software is loaded with just about every feature that you can imagine:  Batch tagging, batch rename, regular expression matching, and much much more.  Get this program, tweak your tags (in batch) and lay off beating up the Rockbox team because they don't want to do it your way.

http://users.forthnet.gr/the/jtclipper/

And, by the way, just because their website says "Beta", do not fear.  I've been using this fan-damn-tastik program for 6 months with NO, ZERO, ZIP, NONE problems.

[lalittle]

It should be easy enough to write a script in foobar2000 to create the TSOP tags for you (foreign languages not withstanding). Is the extra effort your only beef?

Thanks for the suggestion -- I appreciate it.

Yes -- the extra effort is essentially my only beef.  Unfortunately, extra effort and time would still be needed even with a foobar script.  Having to use foobar -- even with a script -- would require extra time that I just don't have.  I know this may seem strange to some people who do this sort of thing all the time, but I have a limited amount of time to spend on this sort of stuff, and I'm already WAY over my limit.  My preference is still for a system that takes care of this automatically.  Others here call this a "hack," but to me, it would constitute a "solution" given the parameters that are important to me.

That said, if a script was written for foobar that created the "sort" tags automatically from the existing tags, wouldn't it have to use the EXACT same logic that I was talking about Rockbox using?  Rather than having a program ignore articles directly, we'd have a situation where one program rearranges the words in the tag in such a way that another program sorts them correctly.

Thanks for the suggestion,

Larry

[lalittle]

lay off beating up the Rockbox team because they don't want to do it your way.

I NEVER intended to "beat up" on the Rockbox team in ANY way -- I honestly hope my comments were not taken this way (although I fear they may have been based on some of the reactions.)  I appreciate everything the developers have done with this project, and I have a great deal of respect for their expertise and dedication to Rockbox.  They've put a LOT of time into this, and they've done an incredible job.  I'm simply voicing my opinions and ideas on this subject since I truly believe that this can be helpful to a project like this.  I FULLY understand they may not agree with me, and I have no "expectations" with regard to them utilizing any of my input -- I merely offer my opinions for the purpose of giving feedback.

Larry

[Llorean]

Just as a note, I do fully respect your right to offer your opinions on things, and if I've come across as somewhat defensive I apologize. It was beginning to feel like an argument to me, since it felt you were repeating the same reasons for doing it as had been said before.

Yes, the exact same logic would apply in the foobar script. But the logic itself isn't purely bad: If your collection contains only English songs, with a set list of known articles, and all songs need the article removed, the logic is fine.

Having the logic handle more languages, or a longer list of articles (the list of articles skipped would, for a fully functional feature, need to be independent of the current display language) would be part of a "properly" working feature. As well, if the TSOP patch feature were included, having an article skip feature would be redundant on the assumption that a user has properly tagged files (Rockbox already makes this assumption in several other places, so please don't use the argument that a user may not have properly tagged files: if they don't, it's their choice, and their choice alone). There are many reasons not to waste binary space on redundant features.

In the end, the TSOP patch offers the greatest flexibility regarding this feature.

I know it's not ideal for what you want, but a batch script could be used with Foobar as has been said, and I'm not sure about the "I don't have the time" argument on that: If you convinced someone to make the batch script (and I'm sure someone will at some point if the TSOP tag support gets written) then it's a matter of a batch process. You leave your PC running, and it scans the performer tags, and generates TSOP tags without the article. Heck, someone could even write a plugin for Rockbox that would (much more slowly) generate sort tags off the existing tags.

In my opinion though, sort tags would be an "optional" feature. You add sort versions of the tags if, and only if, the "Sorted by" information should be different from the display information. So for all artists not containing an article, and all artists you wish to continue to include the article in the sort, you don't have to add additional tags. Just run the batch script overnight on the other artists while you sleep, and you're done.

[saratoga]

It's not the right way to solve the problem.

The proper solution is this:
There are in many tagging formats two tags each for Title, Artist, and Album. One for display, and one for use in sorting. If you want The to be ignored, for the sorting tag, you put "Blah, The" instead of "The Blah" (which you use for the display tag).

So a correct solution would be for someone to make a patch to use all of the tags effectively, rather than for it to intentionaly ignore a string of characters.

I don't think adding more ID3v2 fields is really the best solution.  Besides the normal issues with incompatibility between custom fields across different software, theres also the practical problem of actually updating a large amount of music with custom fields potentially across many different formats.

IMO there should be a preprocessor that examines tags when they're loaded.  A language specific text file could define what strings should be parsed out, thus allowing intelligent behavior across many different languages while still allowing consistent behavior with other software.