so, i've been spending the last few days playing with mixes. part of the problem is that the song sounds very different out of my mixing devices (alesis mixer, pod amp simulator), out of my soundcard (m-audio audiophile) and out of my laptop (some cheap integrated hd chip). every time i change something to sound better on one, something breaks on the other. i've come to the conclusion that i'm going to need to compromise a little...

i want to put that aside for the day and get most of the parts finished.

i think it's ultimately a render issue. for some reason, the out through cubase sounds differently than the playback in real time. i mean, different hardware is going to produce different results, but the changes are largely in the guitar tone and that's beyond natural playback differences. i'm thinking it might have to do with the way it's calculating some filters. that may actually mean that the out i'm getting is the more technically accurate version, but i want it to sound the way i'm mixing it, not as some abstract ideal i can't monitor.

i had this problem once before, years ago, and fixed it by hooking an out from the mixer into the in on the soundcard and just recording it in real time. but, i'm not sure i'm going to be able to do that with this piece because of the vstis. that other track was just all wave files.

but i'm not worrying about this for today, i'm finishing the recording as much as i can. the last thing i'm going to want to add is a bass, and i need to give myself a few days for it to get the tone right. i find bass to be a very difficult instrument to record. it's partly because i don't have any di cords. although, maybe i should take the opportunity to invest in one...

it's the primary reason i use so much synth bass. luckily, it tends to fit the style i'm writing in. but this track needs a bass guitar. the core of it was actually written on a bass guitar...

i actually did a huge amount of experimenting the last time i got to a bass part, which was way back in '08. yeah. well, i haven't done a lot of recording at all since then, and the bit i've done has worked fine with synths or sequencers. i went and picked up a separate bass limiter (boss lmb-3) and compressor (digitech dual band bass compressor) in stompbox format. the compressor actually has an amp simulator in it as well, so the two of these things together are fairly versatile. iirc, i need to connect it to a noise gate, though.

anyways, i think i have what i need this time to get it pretty good directly in, but i know i'm going to need a few days to play with it, too.

i actually ended up converting an old dod fx53 into a bass distortion, too. it was a $5 pickup somewhere, and i didn't know what i was going to do with it when i got it. as a guitar distortion, it's not what i want - i like really fuzzy stompboxes that push sustain and high gain. this is a weaker distortion. but it works well with a bass for exactly that reason - it's subtle enough that it doesn't kill the note clarity.

the bass (it's a mid 80s ibanez roadstar II, black pearl burst) also has active pickups, which is hugely important for recording, and why i bought it. well, that and the neck is small, which helps with my tiny fingers. the pickups in the early roadstars were designed so they could switch between p-bass and j-bass type sounds, but this one has two low impedance active soapbars in it. they're really the bass to look for if you want a playable, versatile and relatively cheap bass, although that factor may have boosted their price a little over the last decade.

actually, i have a mild hypothesis with this and it flips it around.

perhaps they're only counting featured track (they call it a "trailer" and promote it as "increasing subscribers", but subscribers are useless because nobody reads their feed - it's a featured track) plays if the user does run an adblock or flashblock type program. that way, the user is physically initiating the play.

that would explain why i'm only getting 30-50% of the actual hits - the rest of it is autoplaying. you can't turn the autoplay off, unfortunately.

therefore, you should install adblock.

although, i just tested one of the tracks under 300 with an embed (meaning i'm not signed in through it) and it didn't increment immediately like it ought to, indicating there's something deeper happening.

you'll see me on the street corner in five years yelling:

"i got 100,000 views but they didn't count them!"

while people toss quarters at me and tell me to shut up.

but, seriously. there's something really seriously wrong with the counter. i haven't seen hits on hardly anything in three or four days, but i know i'm getting upwards of a hundred visitors a day. it's not adding up...

if i were to get conspiratorial, i wouldn't blame it on my politics. there's no logical connection between my hit count and my rambling. that is, decreasing my hit count isn't going to stop me from rambling, and it has nothing to do with the reach of the videos i'm rambling on. if somebody is trying to shut me up by cutting out my hit count, they're completely incompetent.

if i was to get conspiratorial, it would have more to do with the fact that i'm NOT buying hits or paying for advertising. i have a hypothesis that it's ultimately google that's selling the hits, not some script kiddies. being able to generate traffic without buying into their system (via buying ads or buying hits) threatens their business model. i may have been flagged when i crossed some threshold, like 10,000 hits, maybe. i wouldn't put it above them.

but i can't shake occam's razor, and it's just that the counters are programmed badly. no matter how enticing the second conspiratorial option is, and no matter how much the circumstantial evidence leans that way, incompetence is simply a simpler explanation.

as mentioned, i can't let that get to me, no matter how irritating it is. i know i'm generating traffic. there's nothing for sale on youtube - that traffic needs to be routed to bandcamp for me to get anywhere with it. there's no chance i'm going to force somebody to watch an ad to listen to the song; the whole point is *i'm* advertising something, i don't want to give somebody else the space, that would defeat the purpose. and, i'm not going to generate this kind of traffic much of anywhere else. if they're going to throttle me, it just means it'll take longer to get to where i need the hit counts to be to generate that distribution interest. but, my argument relies on the hits eventually adding up, anyways. requiring six months rather than three is relatively irrelevant - i need total hits. so, i just need to forget about it and be patient....

i just hope it *is* incompetence - because if it's not, it still is.

and that's far more frightening than a bunch of bad programming.

as i mentioned before, though, i'm going to be pissed if i get a negative growth curve from july to august as a result of this, considering i know that i'm generating increased traffic. i had a nice curve going on there.

and, what if it's a ploy? a sales trick? see, i bet google has an army of psychiatrists working for them...

what if they key on users that are generating hits, then take those hits away in the hopes that it will coerce them into buying advertising? what if they even inflate the hit counts with the desire of building a sort of addiction to the metrics? the first hit's free, then we start charging.

i've been clear in the past that i can't buy advertising. the music is far too abstract. i mean, i probably wouldn't anyways - i'm both remarkably cheap and far too idealistic. but, i know it's not going to work. what i'm doing is far more effective.

but, that's kind of the point. i'm breaking their system...

see, i might be breaking their system, but it's not the case that just anybody could do this. i don't really pose them a threat.

first of all, this is a full time job. the sheer amount of posting i'm doing is not really feasible for most people. second, it requires some serious talent, which in this case reduces to intelligence. i'm a realist, and it's the truth of it - you need a brain to troll like this. it's an art. so, if you put those two things together, you're looking at a pay rate considerably over minimum wage and a difficult and expensive recruiting process attached to it. nike or coca cola or whatever can't just hire some high school kids to do this. they'd really have to headhunt.

it's consequently not feasible to adopt as a model. could some other individuals pull it off? sure. and you can't stop them, either. but i'm not a threat to ad revenue.

so, like, back off.

but, on a larger scale, i think the world needs to get ready for something: the trolls will inherit the earth. in the last generation, it was the nerds. everybody hated them and looked down on them, but they became the most successful gen xers. in the future, expect the trolls to overcome.

wait for it.

in that sense, maybe i'm blazing a path....

 ....but this is too specialized to replace traditional advertising, and it's not what consumers want, either. it may generate traffic, but it would hurt sales, for most companies.

i mean, the crux of what i'm doing is three-fold :

1) piss people off. then they have to listen - and in an agitated state, where they're likely to respond. companies won't want to do this, obviously.
2) say something controversial that pisses off most people but is exactly what somebody else wants to say and is afraid to say because it will piss off most people. that works for obscure artists that are going to build fan bases off the solidarity, but it's almost always a bad idea for a company to take a stand like this.
3) provoke thought. which is again counter-intuitive for selling a product, but good for distributing art.

so, even if the odd company can find a trolling genius, and is willing to pay her, what she can actually do for the company is quite limited. and consumers will quickly grow tired of it.

there's also a saturation point inherent to the process of trolling that trolls need to be wary of. i can troll at will right now because nobody knows who i am. even a company that might be willing to be antagonistic with it's product, like sony or nintendo, can't do that. if the logo is visible, it's just an ad. if it isn't, it's deceptive - and will backfire.

so, you have to go back to the determined individual. with enough time and hard work some other person with the smarts could do this, maybe better than me. but we're rare. 1 in some big number, anyways.

....and not a threat.

i don't like that i'm constantly bitching about youtube metrics, either, but i wish they'd be more clear in documenting rejected hits. i ultimately remain convinced that their algorithms are broken, but i'm wondering if i'm being throttled in some way.

i've had 94 hits a day - exactly - for almost a week, now. that's one less than 95. or < 95.

i mean, i watched the hit count rocket to 301 from 133 over about 36 hours, from thursday night to friday afternoon. i was averaging close to five hits an hour - just on the track on the front page. that's quite a few hits when you think about it, about one every 12 minutes.

the number on the top track clearly increased by well over a hundred hits over the course of thursday. yet, only 40 of those were counted.

so, what was the problem with the other 60-80 hits, exactly? wrong country of origin? did they not watch long enough? were they using a proxy? tor? adblock?

the counter increased by close to 120. i could see up to a quarter of those (30) being bad hits. but 50-67% of them? it just strikes me as unlikely, and i again have to conclude bad programming....

the flip side of this is that it seems like i'm getting 250-300% more traffic than is being recorded, based on what i observed yesterday and how much of it youtube actually picked up. so, that's positive.

i guess that whatever's broken for me must be broken for everybody else, too.

in the end, the basic relationship of more hits ----> more views remains in tact.

but the numbers are not remotely accurate....

i just wish more of those hits were translating to bandcamp views, because that's the actual point of the exercise. well, sort of. if i end up with a youtube profile with 500,000 hits a year or two from now, i'm thinking that should generate distribution requests. but it would help a lot if i could say "x people actually bought something", and x>0.

it's going to be another full year before i can say "look, my introductory demos generated 20,000 hits, why don't we talk about distributing some of the newer stuff?" and that's really just about all that youtube has the potential to be useful for. for an established artist, it's absolutely useless...

i mean, i wouldn't even consider burning copies of these first two demos. they're only available as mp3 downloads for $6 (the first is a double) and $3, respectively. but i couldn't imagine anybody buying them unless they're fans of the later stuff. they're not very good, really.

what i'm trying to do is demonstrate i can build an audience. it's the whole emulating a natural process of the past thing...

that's why everything's ready to print.

(or in the process of being ready to print)

which reminds me: i'm expecting a nasty surprise in the mail soon, and should make some calls on monday.