Flying out of San Diego to Maui…
I have spent a bit of time molding my original playlists into something a bit more functional for me. As such, I thought I’d explain what has done. These are more of an “evolution” of the original idea I developed as I’ve been using iTunes and my iPhone for music.
I still have my five “groups” of Industrial, Electronica, Rock, Orchestral, and Miscellaneous. Some songs may end up in multiple groups, but those are the five main groups. I modified the system a bit so that when I am collecting music from my collection into my smart playlists, I have no more overlapping going on. I accomplished this by figuring out what groups I preferred the most, as well as which has the smallest number of songs in my collection. I set up the Smart Playlists to not look for “Grouping is …”, but rather “Grouping contains …” this makes it possible for one song to end up in potentially two, three, four, or all five groups. However, I want to be able to make sure that a song is only counted once. My solution was to redesign my smart playlists according to categories loosely based on the five groups. A table which outlines my scheme is presented below:
| Group | Industrial | Electronica | Rock | Orchestral | Miscellaneous |
| Industrial | |||||
| Electronica | |||||
| Rock | |||||
| Orchestral | |||||
| Miscellaneous |
Ont he above chart, the Green cells are absolutely necessary, the Yellow are optional, and the Red are not allowed. The rows each represent the categories for the song smartlists, while the columns represent the actual headers. How this works is with a set of “contains” and “not contains” statements for each smart playlist. Miscellaneous is the easiest. It is looking for songs where Grouping contains Miscellaneous. Industrial is the most stringent, it is looking for songs where Grouping Contains Industrial, but “not contains” any other group. So, for each category, I make sure I have these options (songs have to meet ALL of the requirements for each category):
This pattern means that if the grouping has miscellaneous in it at all, it will fall into the miscellaneous category (for example)… it helps even out my songs a bit, and it ensures that my songs only fall into one of those five categories.
My design for my playlists is still two-level. There is one level that does nothing but pulls the songs that meet certain requisites from my library into a sort of “holding list” — this helps populate songs I want on my iPhone, and is potentially as large as 1500 songs. For each of the five categories I listed above, there are four main collections or types of songs I am looking for:
This gets 5 categories of potentially 300 songs (thus, the 1500 songs in the list). Of course, if there isn’t enough songs for each category… well… this will cut down a bit.
I collect all of these songs into one large list. From that list, I collect songs from each of the five categories I explained above, in smaller quantities, and some slight deviation from the four sets I filled the list with:
Combining them in a randomized list, I get a playlist of up to 300 songs, nice and shuffled, ready for me in my “daily playlist”.
Building the lists as such has resulted in a rather interesting and dynamic blend of music constantly available, and ensures that no song is left unplayed, keeps me with an ever-flowing list of music. As it stands, I tend to play about 60 - 80 songs in a day (since I play music while at work), which keeps things shuffled, and the way I have this constructed, I don’t necessarily need to sync my phone every day — it can go a few days on nothing but the music I have sync’ed to it.
I may eventually make a category for my discoveries on iTunes, who knows… but anyway, I’ve been developing some complex “smart playlists” for iTunes, which also affects what songs are loaded onto my phone. My primary philosophy has been to evenly-distribute my songs in a playlist which is semi-random in nature, and keeps me with a relatively steady stream of different music.
First, I developed five “categories” of music that I work with:
I start with getting a collection of 300 of each of the above categories, or a total of 1500 songs. Mind you, it doesn’t quite end up being 1500, because I actually don’t have 300 songs rated at 5 stars. To add to the mix, I have said that for the “Most Played”, and “4 Stars” category, I ignore any song I have played in the past 10 days — this forces these lists to cycle new songs on them at a pretty wide interval, while the “5 Stars” category I ignore songs I have played in the past day. These are the songs that are actually loaded onto my iPhone.
I combine all 1500 songs (max) from these groups, then I set an additional limitation for all of the subsequent lists below — they are limited to only songs not played in the past 5 days or skipped in the past day (which forces those lists to update a bit quicker, while the main/core list above keeps the phone updated when I sync it).
I have categorized my music into five semi-distinct groupings as well:
I then compile a set of 25 “sub categories” — a kind of “matrix multiplying” of the two above sets of 5 categories/groups. So, I end up with sub-groups of “Industrial/4 Stars” or “Rock/Most Played”. I restrict each of these subgroups to a max of 10 songs for each.
I am able to re-combine the 25 sub-categories into a set of 5 50-song playlists for each of the first set of 5 categories (randomly shuffled), and another set of 5 50-song playlists for each of the second set of 5 groupings (randomly shuffled).
Finally, I take the 5 50-song playlists for each of the groupings, an put them together randomly to a single 250-song “Daily Playlist”. The end result is a nice, random assortment of songs, somewhat evenly distributed across all of the categories I specified. All in all, I end up with about 4.6 - 5.0GB of music being stored on my phone, regardless of how much music is in my library, and it updates/fills in new music to the phone whenever I sync (as well as removing anything that doesn’t apply to the lists defined above).
I’ve had a few people mention that my site tends to focus a lot on World of Warcraft. Well, that’s a game I really enjoy playing, and tend to spend a good portion of what little free time I have actually playing.
One of the hot topics I’ve seen appear from time to time in WoW, and well other MMOs I’ve played in the past as well, is the notion of “class roles”. Typically, there are a few different roles people have categorized classes into — basically, stereotypes for those classes. I’ve seen things like “puller”, “ranged Damage Per Second (DPS)”, “Area of Effect (AoE) DPS”, “Tanking”, etc. However, it more or less can fall into three specific categories:
These three roles are more or less what different classes can fall into. different classes in World of Warcraft have different strengths and weaknesses with these roles, my view of these classes per role is outlined in a simple table below:
| Class | Tanking | Healing | Damage-dealing |
|---|---|---|---|
| Druid | Yes | Yes | Yes |
| Hunter | Somewhat | No | Yes |
| Mage | No | No | Yes* |
| Paladin | Yes | Yes | Somewhat |
| Priest | No | Yes | Yes |
| Rogue | No | No | Yes* |
| Shaman | Somewhat | Yes | Yes |
| Warlock | Somewhat | No | Yes |
| Warrior | Yes | No | Yes |
I marked the Hunter, Shaman, and Warlock as only “somewhat” when it comes to tanking, because in all honesty, these are not classes relied upon for being a “main” tank in any capacity, but can provide “off-tank” roles (that is, tanking a creature that gets added to the fight, holding it at bay until the main group tank can take over). Shamans can do it face-to-face, hunters potentially as well — these two classes have the second-tier armor in the game (plate being first-tier or “best” armor, mail is second-best). Warlocks and Hunters also have pets which can be used in an off-tank capacity, something I’ve become particularly familiarized with over time, having two level 70 hunters, and a level 45 warlock (my roommate, guild mate, and long-time best friend has a level 70 warlock). While these definitely cannot be relied upon for any real main-tanking capacity, they can provide some assistance during a fight that other classes cannot.
I marked the Paladin as only “somewhat” when it comes to DPS, because right now, the Paladin has the absolute lowest and worst capacity for DPS of any class. The Paladin’s best DPS build still pales in comparison to other classes spec’ed for DPS.
Besides the rogue and the Mage, every class can provide at least something to two of the roles. This is often referred to as “hybridization”. It does not necessarily mean that every shaman is both a good healer and good damage-dealer, or that every druid is good at everything. It just means that there are builds, gear, and abilities available to the classes which makes them viable in those roles. There are literally only two classes which are not hybridized — mages and rogues. These two classes have only a single role in the three-pronged class-role system I’ve defined above — they are Damage-dealing (or DPS) classes. Two other classes, the Hunter and the Warlock are mostly exclusive to DPS, but can provide off-tanking capabilities when needed.
According to Blizzard at BlizzCon 2007, Mages and Rogues are the “top dogs” when it comes to DPS, in their respective fields. Rogues [are supposed to] have the highest single-target DPS possible out of any class. Mages [are supposed to] have the highest multi-target DPS possible out of any class. From those two, then each class falls below them, logically, looking at the role distribution, warlocks and hunters are next in line, as they do have a somewhat limited (but not quite so restrictive) set of roles.
While talking about class roles, I should point out that each class provides a set of utility functions for groups, through various abilities they have. Rogues and mages have some of the most extensive arrays of utility abilities out of any class, while Druids tend to have the least… Deduction seems to imply that the more roles the class can fill, the fewer utility abilities that class has to offer for a group, but this is not a hard-set rule.
Now, the immediate, “surface” logic after looking at the role break-down, is to ask which class is the best healer, or best tank — after all, there seems to definitely be a hierarchy for the classes when it comes to DPS, with Mages and Rogues supposedly being at the top, right? Well, Blizzard had initially designed a hierarchal structure for tanks and healers. The problem is…. every group *must* have a tank, and *must* have a healer, then any number of various DPS slots beyond that. The tank and healer positions are mandated, or else the group will utterly fail.
As such, when there was a rigid hierarchy for the healers and tanks, where there was an obviously “best” tank (warriors), that meant everyone who was building a group wanted a warrior as the tank. After all — who wants the second-best tank, when you can get the best one? What did this do? It marginalized Druids and Paladins, bumping them out of groups as tanks because they weren’t as “effective” in the tanking role as a Warrior. At the same time, it also meant that Warriors that were designed around being DPS, or at the very least were not purely tanking, were also equally ignored or mocked for not being tanks. Tanks were effectively seen in short supply, because only tanking warriors were wanted, and it took a lot of coercing to get a druid or a paladin to be used as a tank, other than by a group of close friends who knew the character’s strengths and weaknesses first-hand. Healers had the same problem, with Priests seen as the best (or Paladins arguably for the Alliance only before Burning Crusade, for both factions more after Burning Crusade’s launch).
The problem is…. those two roles are mandatory for any group to function. You can build a group with any combination of healers, tanks, and DPS as long as you have at least one tank and one healer in the group, and will do at least passingly well in an instance. If you don’t have at least one tank and one healer though… your odds of being successful drops to zero in all but the lowest level instance runs. By placing a hierarchy on healers and tanks, this made only one class be an option for tanks, and one (arguably two) for healers.
Blizzard seems to have resolved this, and still are working on resolving it by making all the tank classes become equivalent in tanking capacity. Notice I said equivalent, not equal, or identical. They are on equivalent footing — they can be interchanged in a group as the main tank role, but the approach each class takes towards tanking is different. The effect of this was that Warriors, Druids, and Paladins became more viable as tanks, and it increased the number of tanks available to choose from for a group. It also had the effect of making more freedom of choice for Warriors, and DPS Warriors became more and more popular as a result, no longer being forced into the pure tanking role. The effects were more flexible choices for group designs, and character builds. healers are also on their way to being along this line I believe, which is a good thing. since every group must have at least one tank and one healer, having there be more options for both roles is a good thing.
Now, there is a counter-argument I have seen to this line of thinking. The idea that some propose that the more roles a particular class does, the less effective that class should be at each role. The “jack of all trades, master of none” concept, which does, on the surface, have some merit. After all, if a class can fill two roles, it would be twice as popular as a class that only fills one role, right? I mean, a class that can cover healing and tanking would be more popular than a class that is just a healer, so the class that is just a healer would need “something” to be more appealing to a class… the obvious choice would be it should be a better healer…
There is a flaw in that reasoning. First, just because a class can fill multiple roles, it does not mean every character of that class fills all roles equally. Thanks to gear and talents, characters begin to become widely diverse in what roles they are able to fill within the class. A Fury-spec’ed DPS Warrior just doesn’t have the same capability at tanking that a Protection-spec’ed tanking Warrior does. A Feral Combat-spec’ed Druid won’t make for a very good healer, and so forth. In fact, a DPS-built Warrior, or a healing-built Paladin, or a spellcasting DPS (Balance) or healing-built Druid wouldn’t make for a very good tank, they’d be only marginally better than what the Shaman, Warlock, or Hunter could potentially provide with the right builds in each. so, really, each character can fill one role, regardless of class, and based off of their gear and Talent build.
Well, all except one. Druids can fulfill two roles with the same talent build and gear. They can be good tanks or good DPS. Now, they can’t be both simultaneously, this is an either-or sort of option that they do have a bit more flexibility with, but can’t fulfill both at the same time, relying on their shape-shifting. This sort of exclusive-or approach to the class still forces the druid to be either a tank or DPS, and switching mid-fight tends to come at an incredible cost, in energy, rage, and mana. Not something done lightly.
Now, there is some good logic in this argument. Like I said, there are two classes which only can possibly fill one role, the Rogue and the Mage. They deserve to be better at DPS than other classes, since they can’t fill any other role, and since nearly everyone could potentially be a DPS class, it would be important to distinguish them from the other classes. This is done with the DPS they can provide, and they do have rather impressive arrays of utility abilities, making them often very sought after, in some capacity, in any group.
All of this, when taken together, results in a simple concept; healers and tanks should be equivalent to each other in their respective roles — those two roles are absolutely mandatory in any group, and therefore if all of the classes that are tanks or healers are equivalently capable (with the right builds) to handle those positions, this simply opens up group construction to becoming more organic and flexible. DPS can have a bit of a hierarchy, as it is far more popular, more classes can fulfill it (all except for the Paladins really, and Paladin DPS is slated by Blizzard to be looked at in more detail). There also is the merits of utility abilities, which help counterbalance the number of roles that a class can do (or so it appears) , but utilities tend to be very specific to certain situations and conditions. At least on a PvE end, classes seem to be getting balanced with these dynamics and this logic behind them. There shouldn’t be a healing class better than others, or any that are worse than others (other than the classes which simply cannot heal), same with tanking — heck, I’d suggest that shamans, warlocks, and hunters might be given more capability with tanking, to fill in the holes there and make things even more organic, but that’s just me.
Over the years, people have asked me what my taste in music is. My problem is, it isn’t really something I can quantify in very easy terms, even though I definitely have certain trends in the choice of music I seem to make. since I got my iPhone, I’ve made a sort of project in getting music on it that I like… which has become a rather slowish process I’ve developed of getting music I have on my computer, organizing it, loading it into iTunes, getting the cover art for the various CDs, and then transferring them over to my phone. Where then I filter out any song that is a duplicate of another song (which I have a number of songs that fell into that particular category), and go through the process of coming up with a rating for my music. Apple has a five-star rating system for iTunes that I use for this.
Typically 1 star means a duplicate (and thus removed), 2 stars I usually use for songs I just don’t care for, 3 stars is any song I’m not entirely certain if I want to keep or not, 4 stars I mark as worth keeping, and 5 stars is the songs I really love. I made some basic playlists, grouping all of the 0 stars songs as “Unrated”, 1 - 2 stars as “To Remove”, 3 as “Uncertain”, and 4 - 5 stars as “My Top Rated” — when songs first get put on my phone, they have 0 stars. I typically then go through that list, marking what rating they get. I occasionally connect my phone up to the computer, and remove anything I’ve marked at 1 - 2 stars. My goal is to keep only the songs I mark as 4 or 5 stars, anything at a 3 star rating gets reviewed, and if I don’t like the song (for whatever reason), it gets moved to 2 stars, otherwise I put it to 4 (or occasionally 5) stars.
Using this process, I’ve gone through literally thousands of songs, and have been coming up with a list of songs on my phone I actually *want* to listen to. It isn’t complete CDs, just the songs from CDs that I like, ditching the rest. Certain CDs never make it to my phone though… as I tend to not care for the “Live” CDs much… I want to hear the song as it was recorded in the studio, not listen to it distorted on a sound stage with people cheering & screaming throughout it, so “Lives” tend to get put on the chopping block right away.
So far, I have music from a number of my favorite groups (there’s still several more batches of songs I am going to process and find the “good songs” to keep from them), but for the most part, I have a working list from the following groups:
Things I am planning on adding soon: Beastie Boys, Crystal Method, Chemical Brothers, Prodigy, Offspring.
So, for people who are curious what kinds of music I like, that’s a rough rundown so far of what I’m wanting on my phone — that is, music I want to have readily available and on my person all the time.