This seems like an incredibly bad idea to me, at least if it were to become the dominant pricing model - but I highly doubt it will.
I mostly listen to artists that don't sell a ton of records, where a big success could be shipping 20,000 or 50,000 units compared to radio acts that can ship millions. I don't know how their model would work in reality, but let's assume these tracks might be 25% the cost of a big radio single. The process values popularity over all other factors, doubly reinforcing it. Not only would the popular act earn more money because they were shipping more units, but also they would earn more per unit. Assuming there are fixed production costs that get paid down (lower % per unit for more popular sales) why should a less popular artist be penalized in so many ways?
A record that only sells 10,000 copies to devoted fans shouldn't translate into less income for the artist than 10,000 shipped units of a pop act. The small time artist most likely needs to money much more than the one shipping a ton of units anyway.
Re:Fiat currencies have several problems.
on
Bank Run in Second Life
·
· Score: 3, Insightful
The gold standard is ridiculous and archaic It ended in 1971 after a thousand years of use. Neither ridiculous nor particularly archaic.
There's nothing wrong with fiat currency. It decreases in value every year destroying the value of savings. So just to be clear, you're saying there was no inflation in the USD prior to 1971?
Folks confused you with knee jerk analysis of the vista eula - All versions of vista can be virtualized, the clause that you are referring to allows some versions of vista to be run both as a guest and a host for that guest while only purchasing one vista license. This is a continuation of their program from windows 2003 (r2?) that allows you to run 1 host win2k3 and 4 guest win2k3's simultaneously with a single license as long as it's on the same hardware... This applies if win2k3 isn't the host as well, one license gets you up to 4 guest instances with one license no matter what the host OS is (esx, linux, etc)
As noted above the GUI does indeed work, check this picture for proof. What has been released so far is just buildable sources of the kernel and I assume some tools. Just like with linux you could then build [s/build/obtain in this case] the rest of the system yourself, or you can wait for a pre-built full install DVD - much like using ubuntu or whatever. As I understand it there are no barriers left (other than copyright) to producing a fully working full OS X install DVD - one or more should be out fairly soon. Look for a "JAS" release of 10.4.8 via bit torrent in the coming days.
So what you're saying is that you don't care what the actual limitations are, or what the intent of the authors of the documement is, or how the company itself would interpret the document, or what they might plan on enforcing. You only care about a single sentence in a 30 page document that appears to contradict me, even though you agree that if it does it wasn't their intent and would never be enforced? Do you actually think you're contributing anything by being so pendatic?
Since I've already once, I'll let you read it this time.
Where in the EULA does it specifically allow me to create an ISO image of a CD containing photgraphs I have taken and copy it onto my hard drive? If this is not expressly permitted by the EULA, does this mean you believe it's forbidden? Do you think Microsoft would tell you it was forbidden to this if they were asked? If not, where is the language written that applies to my photographs and not my legal backup? If thats not there, then where is the language explicitly forbidding the legal backup to be stored on my hard drive? If you can't find any of those, well then, you've got your answer.
I just read through the entire EULA because I just couldn't believe they had included "Home Basic users can't copy ISOs to their hard drives". Turns out I was right. As far as I can tell there is no restriction to ISO's per-se, instead the original author was attempting to infer a lack of a right of some versions to store a copy of the software [meaning, a copy of the vista DVD] on "network storage" based on the fact that this right is permitted for Ultimate. However, just because they grant a right to some versions doesn't mean you don't have that right when it isn't explicitly granted - for instance even if they only enumerated the right to backup copies for Ultimate you'd still have that right for all others, existing law generally grants it.
The translation to "can't copy [any] iso's" happened in the last step, by the comment submitter, and is as far s I can tell just a complete fabrication.
Some part of me wonders why a website full of people who swear to their grave that they'll never run a piece of software is so intent on discrediting it that they make up shit. Carry on though boys, have fun.
From what I understand, one of the only effective ways of limiting a 3rd party's access to a common AP without any administration rights would be to use spoofed 802.11 packets with the offender's MAC Address to send disassociation packets that will reset their connection, possibly causing them to have to manually cause a reconnect, and definitely causing all of their connections to dump and go through the process of reassociation and getting a new IP even if their client will automatically reconnect.
The above includes a number of programs related to or using aircrack-ng, one of which does this kind of disassociation and other nasty things. Due to driver issues I believe this kind of thing is only possibly in linux (*nix?) right now, and even then only with certain chipsets - the same ones that allow aircrack-ng's arpreplay attack. Out of the box the code will need to be changed to target only specific high usage MAC's - or there is code in the aircrack-ng base that does a disassociation as a "one off"
That is a thread on the aircrack forum discussing the tool on an older state to give you an idea of what it does out of the box. Note that running something like this is wholly anti-social, I trust you'll modify it appropriately and consider your actions carefully. I've never actually run this code base but I have every expectation that it would work as advertised - I have definitely disassociated 3rd party MAC's on wlans before and it does have the intended effect.
Discussions about shaping, QOS and traffic control are obviously the appropraite play for administrators, but I think your question was what to do as a user without any other access. This is completely unsuited for a provider. But since you asked about TCP resets - this will be dramatically more effective with no impact on the other users when modified to run in a single MAC targetted mode. Whether it's right to do it, well, you're a left to your own decision. I just thought you might appreciate a substantive reply instead of hand-wringing.
err? You seem angry but about what I'm not even sure you know. I never said anything about anyone knowing about backdoors that might be in place. And now checking a serial number against a validity database is obviously actually a tool for the US government? HOLY SHIT THE NSA HAS BEEN HACKING MY BRAIN SINCE HALF LIFE!
Active outbound control firewalls like zone alarm already interfere with US (and I'm sure other jurisdictions) use of key loggers that publish to the net. I'm pretty sure they haven't caught any heat about that. In fact it was a US LEO that encouraged me to start using that kind of technology. Requiring anti-malware technology to skip over it would essentially be akin to law enforcement requiring back doors be put in their software, something I'm pretty sure wouldn't fly.
Re:Is the big fat memory leak fixed?
on
Firefox 2.0 RC2 Review
·
· Score: 2, Interesting
For what it's worth I believe they still have a ways to go. I write this with RC2 having launched it somewhere between 24 and 36 hours ago. It was using 237MB of memory, but when I closed my other tabs it dropped down to 193MB. Obviously if I closed it and relaunched it on this page it would only be using a fraction of that ram. Compare that to other processes currently running and it dwarfs them all - VMWare virtual center server 2.0 is using 44MB, Sql Server 42MB, Outlook 2007 36MB, Apache 32MB, IE 7 RC2 27MB. IE would definitely be using more if it had gotten more use, but many of the other applications have been running for at least a week or two.
From the launch of rhapsody until now, rhapsody has always used microsoft DRM. Why they are using home rolled DRM now I can't say, but I'd guess that it probably has at least as much to do with product differentiation and branding than any of the technical reasons mentioned. Plays for sure products are a very crowded space with little differentiation and no one making any money. But that's probably not the platform's fault.
as someone who regularly uses NAS storage for streaming video over 802.11g I can confirm in the real world that only some HD content will run real time well enough to watch. Higher bitrate content approaching 20Mbps while still underneath the real world transfer rates of that kit becomes unreliable, even when there is no other b/g traffic being broadcast in the area. Transfer rates don't stay as a solid line, and video streaming needs some headroom for buffering and catch up if anything glitches or something else gets broadcast. The same content is rock solid on 100Mb ethernet or the matched pre-N stuff I used. Not at all unlike mounting single layer DVDs that are less than 5Mbps and not being able to stream them well over 11b. I haven't rushed out to buy any pre-N though, but I'll be happy to use it when it's more reasonable.
Saying "no one can use it" about network bandwidth right night is kind of like saying no one can use it about RAM in the 80's - you're assured of being wrong much quicker than you think. Hell, some people's consumer internet runs faster than 11g can now.
After researching the subject of "net neutrality" I found that two considerably different definitions of the term are in use.
The first is the idea of preventing providers from shaping or blocking traffic based on the source or destination corporate entity - ie making google traffic super slow while making msn traffic extra fast. This is obviously troubling and should be subject to oversight. A vast majority of consumer broadband is already subject to regulatory oversight though, either through city franchise agreements or through state PUC's. While I'd support federal laws to curtail this if needed, wouldn't it be better to let the existing structures work if they have the means?
The second definition was in use heavily by the technical communities that are researching or providing data about net neutrality. This definition includes the first definition but also adds on basically any kind of traffic shaping or port blocking based on protocol or port, irrespective of the public WAN side source or destination. Examples of this are shaping to reduce the network impact of peering systems like bittorrent or other heavy users like NNTP and IPTV, and the policies blocking some services universally inside a tier such as not allowing inbound connections to server ports, outbound PPTP, VOIP over cellular data etc.
Shame on those technical folks that are trying to substitute the second definition for the first, they should know better. Trying to legislatively micromanage decisions every provider has to make to make for network usability and completely banning all forms of QOS would be a serious mistake. While I'd be pretty upset if i woke up tomorrow and found i was unable to use VPN protocols, I'd rather have to complain to my city about the franchise or switch providers than end up with a situation where washington banned a whole set of core network management technologies that have been in use for decades without which the internet would be much worse off.
Every study that i saw that included statistics or hard data actually fell under the latter definition and not the former. The reason is that it is relativey easy to detect port blocking and protocols that have different throughput characteristics and examples are fairly common. Trying to programatically detect shaping based on corporate entities or netblocks would be very hard unless it was extremely blatent - what are you going to do, measure connections to thousands of different content providers? Even then how could you tell if the bottleneck was put in place by your edge network or was just due to host side network capacity?
I'd expect any browser plugin that was built would do the same. While it would be useful to know what blocking and shaping you are subject to, trying to group it under of the umbrella of anti-competitive practices is highly deceptive.
eh? So you think labels used to pay artists 30 cents an itunes download, 40-45% of their gross revenue? That would be quite a breakthrough even if it's gone - I thought labels were assumed to be evil in general, or is it that three years ago labels were extremely generous and they only turned evil in the last couple of years?
You are right, I don't know if apple changed from taking 35% to 30% in the last year or so. It would make a reasonable amount of sense if it was true though, as that's about the fat they have reportedly had in their share lately - if they dropped their fees by four or five cents they'd presumably be back to break even or close. ITMS has always been considered a promotional device for their high margin portables so it would fall in line there. I admit it looks like my info is out of date, but bottom line though is that 30% or 35% doesn't make a big difference to my argument. Majors still aren't paying anyone 30 cents a track - even if apple passed on reduced fees to labels in the middle of multi-year contracts (seems like they would if they are eschewing profitability). Moreover the quoted bit in wired was about historical pricing - i doubt he was talking about six months ago - more likely he meant a few years ago, right? So that would be back with my outdated info.
A few random comments:
CD-Baby quotes their fee as 9% and their average pass through as 60 cents putting their average gross proceeds per song over all services at 66 cents. Average could mean a dozen things here but it's likely a vast majority of sales come from ITMS, even more than the industry average as unsigned/true indie acts do very poorly on the subscription/hybrid services.
I'm surpised the website you quoted hasn't gotten flack from apple or others - I find it pretty much impossible to believe apple has dropped the pricing non-disclosure terms from their ITMS contract. Not that that means i don't believe the numbers.
TuneCore: wow! $7 setup and $8/year per album, no sales percentage. That $15 initial price for itunes US apparently includes merchant processing fees, chargeback costs, online database self-service, mail receiving, CD or lossless archiving, CD ripping to one or more formats and bitrates, audio quality assurance (ony real way is real person listens to the whole thing), match to and quick check meta data, due dilligence for checks for copyright/trademark issues in audio or metadata, clearance confirmations, UPC assignment, manual submission to itunes (no bulk submitter available), acting as an intermediary in content disputes, monthly accouting data and usage pass through, no fee monthly electronic payments, no fee customer support, free manual on demand takedowns after 6 months or at the end of a non-renewed contract. God bless them but more than half of these things listed individually costs a business more than $15 - Quite the loss leader, with noi apparent way to recoup costs to even pray to break even. Not to mention a lot of sunk costs like servers, ripping stations, custom software, workflow, contract negotiations, office space, utilities. Must be a labor of love by some guys who are independantly wealthy?
What happens when your iTunes broker goes out of business?
You got me. While there's a ton of media out there everywhere if it falls below a certain noteriety your chances are slim at best. I had a similar experience with a record that should have been much easier to find. Purchased it when it came out brand new on CD for $10 - caveat was that it was made by (well known) san francisco artists and didn't have a national distribution deal. CD got stolen when my car was broken into, copy was stolen another time my car was broken into, lost my PC copy due to a bout of digital music apathy. Tried half heartedly for a few years to replace it but it never hit me in the face. 6 years after it was released I got serious about replacing it - record was out of print, label seemed to be defunct, no distribution deal = no legal online availability, way below the radar for illegal trading it seemed, local used shops claimed demand far outstripped supply. In the end I shelled out $65+shipping for a used promo/review copy from new york city making it by far the most expensive single album i've bought as I've never been a collector.
As far as screwing the artist, I'm sure most people including musicians would agree that morally you're in very different waters if the record is out of print and only available in secondary markets. Of course the irony is that it's quite hard to find anything being pirated that is actually out of print outside of archive or niche destinations it seems that the attention span for works worth stealing is lower than the attention span for works worth selling.
Full albums are available from many ftp sites, irc channels, public and private trackers, usenet and your neighbor down the street are all available in many formats for zero dollars and zero cents. So are you smart for paying 10% of the licensed price, or stupid for paying someone to screw over your favorite musicians when you can do it yourself for free?
It depends how you define label. If you define label as someone who owns the product like a traditional label then no, apple has done business with content consolidators/distributors who broker 3rd party content to itunes since they launched. If you define label as someone who brings a wide array of content to the table under one contract then yes, itunes will only deal with you directly if you are bringing a fairly large basket of wanted content to them. I believe all of the consolidators itunes does business with were labels first in that they own some of the rights directly, i'm sure apple doesn't care - if you brought together 100 bands, a few with ok national sales, and a good amount with at least regional or niche sales, i'm sure they'd be just as happy to work with you as any other "label". Will they strike a direct deal with you as a band though? Nope.
A few comments: First, russian pirate mp3's for sale is not a grey area - it's black and white. Not only do they not have to pay two thirds of their revenue in direct licensing (instead they pay 0%) they also don't have to spend money on content negotiation, contracts, promotion aimed at providers, software that implements DRM at the label's insistance or license reporting management and payment infrastructure. A very small amount of the business is fixed distribution costs, as they obviously prove.
Second, while I understand what you are saying something like monopoly even if you put near in front of it. While they do have a majority of the market, there are at least a dozen major competitive services that have access to the same markets and products at similar fixed and variable costs. Why does itunes maintain such a big lead then? Only supplier for the ipod, customers dislike competitor's subscription models, best client software, most (re)investment into the product line.
Woosh you are awful angry. Somewhat niavely though I'm going to assume you are actually interested in answers to the questions you asked. Note that as far as I understand them none of the known or rumored facts match up with the grandparent's post.
First off though, mostly your questions are almost impossible to answer given your conditions. This is because AAPL is well known (some would say notorious) for not including P&L statements in their quarterly or yearly SEC filings. In fact they only report gross revenues by segment, other costs, fees, expenses or margins are only reported for the company as a whole.
This is very much on purpose. Sometimes individual segment details are given during the conference call - but only when they have something positive they want to highlight.
This conference call is one of the only times you can find a first source itunes profit comment.
Operator: And our next question we'll take is from Arik Hesseldahl with Forbes.com.
Arik Hesseldahl: Hi, Steve. Always concerned about -- not concerned, I guess, but wondering -- one of the previous questions was about revenue. I'm wondering if iTunes has reached the break even point yet.
Steve Jobs: Yes. The iTunes music store had a small profit this past quarter.
I am a bit perplexed by your question about marketing costs and if they should subtract from profit. It is of course universally accepted in the financial world that they do - you didn't make a profit if you spent all your money. As you can see though as early as a year after launch ITMS was profitable. If you don't buy the "universally accepted" bit, consider that most costs for digital distribution are not fixed - that is if you sell 20% more quarter over quarter your total expenses won't rise 20% in lock step with it, or anywhere close. If you choose to reinvest that profit in the business unit in terms of marketing, development or upgrades that don't directly relate to capacity - well, that's your business. And possibly that's because those costs directly improve sales in a business unit you DO regularly release profit and margins information on.
[this part you should cover your ears during and yell la-la-la as it will never be released directly by apple]. Financial news and industry sources put ITMS's cut of an online sale at 35%. Net profit on their cut in early 2004 was believed to be about 3%, or 1% of gross revenue. More recently, net profit has been pegged at about 11% or 4% of gross. [you can stop with the la-la-la now]
Of course you are correct, initial digital content did sometimes have wild price disruptions. It wasn't intended to be a fool proof argument, just an excercise to help you evaluate whether what wired was quoting was true or not, in absence of hard numbers. Consider that none of the majors launched with itms with a majority of their content - if some of the contracts had unintended pricing, it would have been easy for them to withold it. Even in the case of unforseen pricing traps it would still be incorrect to imply that this was the going rate. Consider a case where due to some kind of marketing snafu best buy was forced for a week to sell a new album release for $1 at retail instead of the $10 they intended - you certainly would think copy saying "Best Buy raises prices on some albums 900% this week" or "Best Buy used to sell albums for $1" was deceptive at best.
While you were joking, quoting it and accepting it as a base perpetuates their myth, maybe even more than them writing it to begin with. Why does wired choose to lie about the financial details of online music? I haven't a clue, but this is hardly the first time. Perhaps it's just pandering to an audience, as I have no idea what direct benefit they could gain.
The better question is if there has been a change at all. While the nature of major label contracts means that it is very, very uncommon for the terms of them to be public, I worked in digital music both pre-ITMS and post launch and am very sure that bands on a major label were never close to averaging thirty cents a sale. There may have been an example or three of this, and probably still are but it was never close to the statement that "record companies once offered artists about 30 cents for each song sold". In fact, any averages that came close to this figure would only have _ever_ been for the situation where some smaller itunes content providers offered consolidation deals where they repped 3rd party or unsigned content to apple for more or less pass through costs. These situations never included things like promotion, development or recording costs on the part of the ITMS supplier.
Again, due the the nature of the contracts involved it's nearly impossible to cite sources for this, the same reason it is easy for a wired reporter to make up facts in their article. But consider this logical argument: It is well known that ITMS takes thirty five cents on every dollar on sales (3rd hand citation but other sources are common). That leaves about sixty five cents to the content providers. Even if you have limited knowledge of the music industry it should be easy for you to realize that no major label contracts passed on nearly 45% of gross income from their products to the artists. Whether you like that fact or not, wired is plain wrong in saying that "it used to be so much better" - and I'd bet that probably both the reporter and the editor involved knew that was an intentional distortion. From what I know, majors typically pass on between eight and sixteen cents per track to the artists, and that number hasn't changed much since the ITMS launch.
If anything I believe artist's gross revenue per unbundled song has had slight upward pressure though nothing very dramatic. As I understand this owes the the fact that artists gross revenue per customer with unbundled tracks is understandably down versus typical sales that are bundled (even singles shipped with at least one or two extra songs). Though for all the same reasons I can't cite that so you might as well ignore it.
Of course there's nothing to stop those 13 year olds from acting in terrible harry potter fan made films either - but have them paste together full chapters of the books and share them, well then you'd have a lawsuit. So you were saying?
Do you know what these networks are like for latency and packet loss? Trust me, if you are gaming on it you are going to get rather annoyed long before you get a TOS complaint.
Slashdot Mirror
This seems like an incredibly bad idea to me, at least if it were to become the dominant pricing model - but I highly doubt it will.
I mostly listen to artists that don't sell a ton of records, where a big success could be shipping 20,000 or 50,000 units compared to radio acts that can ship millions. I don't know how their model would work in reality, but let's assume these tracks might be 25% the cost of a big radio single. The process values popularity over all other factors, doubly reinforcing it. Not only would the popular act earn more money because they were shipping more units, but also they would earn more per unit. Assuming there are fixed production costs that get paid down (lower % per unit for more popular sales) why should a less popular artist be penalized in so many ways?
A record that only sells 10,000 copies to devoted fans shouldn't translate into less income for the artist than 10,000 shipped units of a pop act. The small time artist most likely needs to money much more than the one shipping a ton of units anyway.
Folks confused you with knee jerk analysis of the vista eula - All versions of vista can be virtualized, the clause that you are referring to allows some versions of vista to be run both as a guest and a host for that guest while only purchasing one vista license. This is a continuation of their program from windows 2003 (r2?) that allows you to run 1 host win2k3 and 4 guest win2k3's simultaneously with a single license as long as it's on the same hardware... This applies if win2k3 isn't the host as well, one license gets you up to 4 guest instances with one license no matter what the host OS is (esx, linux, etc)
As noted above the GUI does indeed work, check this picture for proof. What has been released so far is just buildable sources of the kernel and I assume some tools. Just like with linux you could then build [s/build/obtain in this case] the rest of the system yourself, or you can wait for a pre-built full install DVD - much like using ubuntu or whatever. As I understand it there are no barriers left (other than copyright) to producing a fully working full OS X install DVD - one or more should be out fairly soon. Look for a "JAS" release of 10.4.8 via bit torrent in the coming days.
So what you're saying is that you don't care what the actual limitations are, or what the intent of the authors of the documement is, or how the company itself would interpret the document, or what they might plan on enforcing. You only care about a single sentence in a 30 page document that appears to contradict me, even though you agree that if it does it wasn't their intent and would never be enforced? Do you actually think you're contributing anything by being so pendatic?
Hehe, ok.
Since I've already once, I'll let you read it this time.
Where in the EULA does it specifically allow me to create an ISO image of a CD containing photgraphs I have taken and copy it onto my hard drive? If this is not expressly permitted by the EULA, does this mean you believe it's forbidden? Do you think Microsoft would tell you it was forbidden to this if they were asked? If not, where is the language written that applies to my photographs and not my legal backup? If thats not there, then where is the language explicitly forbidding the legal backup to be stored on my hard drive? If you can't find any of those, well then, you've got your answer.
I just read through the entire EULA because I just couldn't believe they had included "Home Basic users can't copy ISOs to their hard drives". Turns out I was right. As far as I can tell there is no restriction to ISO's per-se, instead the original author was attempting to infer a lack of a right of some versions to store a copy of the software [meaning, a copy of the vista DVD] on "network storage" based on the fact that this right is permitted for Ultimate. However, just because they grant a right to some versions doesn't mean you don't have that right when it isn't explicitly granted - for instance even if they only enumerated the right to backup copies for Ultimate you'd still have that right for all others, existing law generally grants it.
The translation to "can't copy [any] iso's" happened in the last step, by the comment submitter, and is as far s I can tell just a complete fabrication.
Some part of me wonders why a website full of people who swear to their grave that they'll never run a piece of software is so intent on discrediting it that they make up shit. Carry on though boys, have fun.
From what I understand, one of the only effective ways of limiting a 3rd party's access to a common AP without any administration rights would be to use spoofed 802.11 packets with the offender's MAC Address to send disassociation packets that will reset their connection, possibly causing them to have to manually cause a reconnect, and definitely causing all of their connections to dump and go through the process of reassociation and getting a new IP even if their client will automatically reconnect.
P SESSID=62e86b03ba6476a407065a1ffec82800&topic=172. 0
http://homepages.tu-darmstadt.de/~p_larbig/wlan/
The above includes a number of programs related to or using aircrack-ng, one of which does this kind of disassociation and other nasty things. Due to driver issues I believe this kind of thing is only possibly in linux (*nix?) right now, and even then only with certain chipsets - the same ones that allow aircrack-ng's arpreplay attack. Out of the box the code will need to be changed to target only specific high usage MAC's - or there is code in the aircrack-ng base that does a disassociation as a "one off"
http://tinyshell.be/aircrackng/forum/index.php?PH
That is a thread on the aircrack forum discussing the tool on an older state to give you an idea of what it does out of the box. Note that running something like this is wholly anti-social, I trust you'll modify it appropriately and consider your actions carefully. I've never actually run this code base but I have every expectation that it would work as advertised - I have definitely disassociated 3rd party MAC's on wlans before and it does have the intended effect.
Discussions about shaping, QOS and traffic control are obviously the appropraite play for administrators, but I think your question was what to do as a user without any other access. This is completely unsuited for a provider. But since you asked about TCP resets - this will be dramatically more effective with no impact on the other users when modified to run in a single MAC targetted mode. Whether it's right to do it, well, you're a left to your own decision. I just thought you might appreciate a substantive reply instead of hand-wringing.
err? You seem angry but about what I'm not even sure you know. I never said anything about anyone knowing about backdoors that might be in place. And now checking a serial number against a validity database is obviously actually a tool for the US government? HOLY SHIT THE NSA HAS BEEN HACKING MY BRAIN SINCE HALF LIFE!
Active outbound control firewalls like zone alarm already interfere with US (and I'm sure other jurisdictions) use of key loggers that publish to the net. I'm pretty sure they haven't caught any heat about that. In fact it was a US LEO that encouraged me to start using that kind of technology. Requiring anti-malware technology to skip over it would essentially be akin to law enforcement requiring back doors be put in their software, something I'm pretty sure wouldn't fly.
For what it's worth I believe they still have a ways to go. I write this with RC2 having launched it somewhere between 24 and 36 hours ago. It was using 237MB of memory, but when I closed my other tabs it dropped down to 193MB. Obviously if I closed it and relaunched it on this page it would only be using a fraction of that ram. Compare that to other processes currently running and it dwarfs them all - VMWare virtual center server 2.0 is using 44MB, Sql Server 42MB, Outlook 2007 36MB, Apache 32MB, IE 7 RC2 27MB. IE would definitely be using more if it had gotten more use, but many of the other applications have been running for at least a week or two.
From the launch of rhapsody until now, rhapsody has always used microsoft DRM. Why they are using home rolled DRM now I can't say, but I'd guess that it probably has at least as much to do with product differentiation and branding than any of the technical reasons mentioned. Plays for sure products are a very crowded space with little differentiation and no one making any money. But that's probably not the platform's fault.
as someone who regularly uses NAS storage for streaming video over 802.11g I can confirm in the real world that only some HD content will run real time well enough to watch. Higher bitrate content approaching 20Mbps while still underneath the real world transfer rates of that kit becomes unreliable, even when there is no other b/g traffic being broadcast in the area. Transfer rates don't stay as a solid line, and video streaming needs some headroom for buffering and catch up if anything glitches or something else gets broadcast. The same content is rock solid on 100Mb ethernet or the matched pre-N stuff I used. Not at all unlike mounting single layer DVDs that are less than 5Mbps and not being able to stream them well over 11b. I haven't rushed out to buy any pre-N though, but I'll be happy to use it when it's more reasonable.
Saying "no one can use it" about network bandwidth right night is kind of like saying no one can use it about RAM in the 80's - you're assured of being wrong much quicker than you think. Hell, some people's consumer internet runs faster than 11g can now.
After researching the subject of "net neutrality" I found that two considerably different definitions of the term are in use.
The first is the idea of preventing providers from shaping or blocking traffic based on the source or destination corporate entity - ie making google traffic super slow while making msn traffic extra fast. This is obviously troubling and should be subject to oversight. A vast majority of consumer broadband is already subject to regulatory oversight though, either through city franchise agreements or through state PUC's. While I'd support federal laws to curtail this if needed, wouldn't it be better to let the existing structures work if they have the means?
The second definition was in use heavily by the technical communities that are researching or providing data about net neutrality. This definition includes the first definition but also adds on basically any kind of traffic shaping or port blocking based on protocol or port, irrespective of the public WAN side source or destination. Examples of this are shaping to reduce the network impact of peering systems like bittorrent or other heavy users like NNTP and IPTV, and the policies blocking some services universally inside a tier such as not allowing inbound connections to server ports, outbound PPTP, VOIP over cellular data etc.
Shame on those technical folks that are trying to substitute the second definition for the first, they should know better. Trying to legislatively micromanage decisions every provider has to make to make for network usability and completely banning all forms of QOS would be a serious mistake. While I'd be pretty upset if i woke up tomorrow and found i was unable to use VPN protocols, I'd rather have to complain to my city about the franchise or switch providers than end up with a situation where washington banned a whole set of core network management technologies that have been in use for decades without which the internet would be much worse off.
Every study that i saw that included statistics or hard data actually fell under the latter definition and not the former. The reason is that it is relativey easy to detect port blocking and protocols that have different throughput characteristics and examples are fairly common. Trying to programatically detect shaping based on corporate entities or netblocks would be very hard unless it was extremely blatent - what are you going to do, measure connections to thousands of different content providers? Even then how could you tell if the bottleneck was put in place by your edge network or was just due to host side network capacity?
I'd expect any browser plugin that was built would do the same. While it would be useful to know what blocking and shaping you are subject to, trying to group it under of the umbrella of anti-competitive practices is highly deceptive.
eh? So you think labels used to pay artists 30 cents an itunes download, 40-45% of their gross revenue? That would be quite a breakthrough even if it's gone - I thought labels were assumed to be evil in general, or is it that three years ago labels were extremely generous and they only turned evil in the last couple of years?
You are right, I don't know if apple changed from taking 35% to 30% in the last year or so. It would make a reasonable amount of sense if it was true though, as that's about the fat they have reportedly had in their share lately - if they dropped their fees by four or five cents they'd presumably be back to break even or close. ITMS has always been considered a promotional device for their high margin portables so it would fall in line there. I admit it looks like my info is out of date, but bottom line though is that 30% or 35% doesn't make a big difference to my argument. Majors still aren't paying anyone 30 cents a track - even if apple passed on reduced fees to labels in the middle of multi-year contracts (seems like they would if they are eschewing profitability). Moreover the quoted bit in wired was about historical pricing - i doubt he was talking about six months ago - more likely he meant a few years ago, right? So that would be back with my outdated info.
A few random comments:
CD-Baby quotes their fee as 9% and their average pass through as 60 cents putting their average gross proceeds per song over all services at 66 cents. Average could mean a dozen things here but it's likely a vast majority of sales come from ITMS, even more than the industry average as unsigned/true indie acts do very poorly on the subscription/hybrid services.
I'm surpised the website you quoted hasn't gotten flack from apple or others - I find it pretty much impossible to believe apple has dropped the pricing non-disclosure terms from their ITMS contract. Not that that means i don't believe the numbers.
TuneCore: wow! $7 setup and $8/year per album, no sales percentage. That $15 initial price for itunes US apparently includes merchant processing fees, chargeback costs, online database self-service, mail receiving, CD or lossless archiving, CD ripping to one or more formats and bitrates, audio quality assurance (ony real way is real person listens to the whole thing), match to and quick check meta data, due dilligence for checks for copyright/trademark issues in audio or metadata, clearance confirmations, UPC assignment, manual submission to itunes (no bulk submitter available), acting as an intermediary in content disputes, monthly accouting data and usage pass through, no fee monthly electronic payments, no fee customer support, free manual on demand takedowns after 6 months or at the end of a non-renewed contract. God bless them but more than half of these things listed individually costs a business more than $15 - Quite the loss leader, with noi apparent way to recoup costs to even pray to break even. Not to mention a lot of sunk costs like servers, ripping stations, custom software, workflow, contract negotiations, office space, utilities. Must be a labor of love by some guys who are independantly wealthy?
What happens when your iTunes broker goes out of business?
You got me. While there's a ton of media out there everywhere if it falls below a certain noteriety your chances are slim at best. I had a similar experience with a record that should have been much easier to find. Purchased it when it came out brand new on CD for $10 - caveat was that it was made by (well known) san francisco artists and didn't have a national distribution deal. CD got stolen when my car was broken into, copy was stolen another time my car was broken into, lost my PC copy due to a bout of digital music apathy. Tried half heartedly for a few years to replace it but it never hit me in the face. 6 years after it was released I got serious about replacing it - record was out of print, label seemed to be defunct, no distribution deal = no legal online availability, way below the radar for illegal trading it seemed, local used shops claimed demand far outstripped supply. In the end I shelled out $65+shipping for a used promo/review copy from new york city making it by far the most expensive single album i've bought as I've never been a collector.
As far as screwing the artist, I'm sure most people including musicians would agree that morally you're in very different waters if the record is out of print and only available in secondary markets. Of course the irony is that it's quite hard to find anything being pirated that is actually out of print outside of archive or niche destinations it seems that the attention span for works worth stealing is lower than the attention span for works worth selling.
Full albums are available from many ftp sites, irc channels, public and private trackers, usenet and your neighbor down the street are all available in many formats for zero dollars and zero cents. So are you smart for paying 10% of the licensed price, or stupid for paying someone to screw over your favorite musicians when you can do it yourself for free?
It depends how you define label. If you define label as someone who owns the product like a traditional label then no, apple has done business with content consolidators/distributors who broker 3rd party content to itunes since they launched. If you define label as someone who brings a wide array of content to the table under one contract then yes, itunes will only deal with you directly if you are bringing a fairly large basket of wanted content to them. I believe all of the consolidators itunes does business with were labels first in that they own some of the rights directly, i'm sure apple doesn't care - if you brought together 100 bands, a few with ok national sales, and a good amount with at least regional or niche sales, i'm sure they'd be just as happy to work with you as any other "label". Will they strike a direct deal with you as a band though? Nope.
A few comments: First, russian pirate mp3's for sale is not a grey area - it's black and white. Not only do they not have to pay two thirds of their revenue in direct licensing (instead they pay 0%) they also don't have to spend money on content negotiation, contracts, promotion aimed at providers, software that implements DRM at the label's insistance or license reporting management and payment infrastructure. A very small amount of the business is fixed distribution costs, as they obviously prove.
Second, while I understand what you are saying something like monopoly even if you put near in front of it. While they do have a majority of the market, there are at least a dozen major competitive services that have access to the same markets and products at similar fixed and variable costs. Why does itunes maintain such a big lead then? Only supplier for the ipod, customers dislike competitor's subscription models, best client software, most (re)investment into the product line.
First off though, mostly your questions are almost impossible to answer given your conditions. This is because AAPL is well known (some would say notorious) for not including P&L statements in their quarterly or yearly SEC filings. In fact they only report gross revenues by segment, other costs, fees, expenses or margins are only reported for the company as a whole.
This is very much on purpose. Sometimes individual segment details are given during the conference call - but only when they have something positive they want to highlight.
This conference call is one of the only times you can find a first source itunes profit comment.
I am a bit perplexed by your question about marketing costs and if they should subtract from profit. It is of course universally accepted in the financial world that they do - you didn't make a profit if you spent all your money. As you can see though as early as a year after launch ITMS was profitable. If you don't buy the "universally accepted" bit, consider that most costs for digital distribution are not fixed - that is if you sell 20% more quarter over quarter your total expenses won't rise 20% in lock step with it, or anywhere close. If you choose to reinvest that profit in the business unit in terms of marketing, development or upgrades that don't directly relate to capacity - well, that's your business. And possibly that's because those costs directly improve sales in a business unit you DO regularly release profit and margins information on.
[this part you should cover your ears during and yell la-la-la as it will never be released directly by apple]. Financial news and industry sources put ITMS's cut of an online sale at 35%. Net profit on their cut in early 2004 was believed to be about 3%, or 1% of gross revenue. More recently, net profit has been pegged at about 11% or 4% of gross. [you can stop with the la-la-la now]
Of course you are correct, initial digital content did sometimes have wild price disruptions. It wasn't intended to be a fool proof argument, just an excercise to help you evaluate whether what wired was quoting was true or not, in absence of hard numbers. Consider that none of the majors launched with itms with a majority of their content - if some of the contracts had unintended pricing, it would have been easy for them to withold it. Even in the case of unforseen pricing traps it would still be incorrect to imply that this was the going rate. Consider a case where due to some kind of marketing snafu best buy was forced for a week to sell a new album release for $1 at retail instead of the $10 they intended - you certainly would think copy saying "Best Buy raises prices on some albums 900% this week" or "Best Buy used to sell albums for $1" was deceptive at best.
While you were joking, quoting it and accepting it as a base perpetuates their myth, maybe even more than them writing it to begin with. Why does wired choose to lie about the financial details of online music? I haven't a clue, but this is hardly the first time. Perhaps it's just pandering to an audience, as I have no idea what direct benefit they could gain.
The better question is if there has been a change at all. While the nature of major label contracts means that it is very, very uncommon for the terms of them to be public, I worked in digital music both pre-ITMS and post launch and am very sure that bands on a major label were never close to averaging thirty cents a sale. There may have been an example or three of this, and probably still are but it was never close to the statement that "record companies once offered artists about 30 cents for each song sold". In fact, any averages that came close to this figure would only have _ever_ been for the situation where some smaller itunes content providers offered consolidation deals where they repped 3rd party or unsigned content to apple for more or less pass through costs. These situations never included things like promotion, development or recording costs on the part of the ITMS supplier.
Again, due the the nature of the contracts involved it's nearly impossible to cite sources for this, the same reason it is easy for a wired reporter to make up facts in their article. But consider this logical argument: It is well known that ITMS takes thirty five cents on every dollar on sales (3rd hand citation but other sources are common). That leaves about sixty five cents to the content providers. Even if you have limited knowledge of the music industry it should be easy for you to realize that no major label contracts passed on nearly 45% of gross income from their products to the artists. Whether you like that fact or not, wired is plain wrong in saying that "it used to be so much better" - and I'd bet that probably both the reporter and the editor involved knew that was an intentional distortion. From what I know, majors typically pass on between eight and sixteen cents per track to the artists, and that number hasn't changed much since the ITMS launch.
If anything I believe artist's gross revenue per unbundled song has had slight upward pressure though nothing very dramatic. As I understand this owes the the fact that artists gross revenue per customer with unbundled tracks is understandably down versus typical sales that are bundled (even singles shipped with at least one or two extra songs). Though for all the same reasons I can't cite that so you might as well ignore it.
Of course there's nothing to stop those 13 year olds from acting in terrible harry potter fan made films either - but have them paste together full chapters of the books and share them, well then you'd have a lawsuit. So you were saying?
Do you know what these networks are like for latency and packet loss? Trust me, if you are gaming on it you are going to get rather annoyed long before you get a TOS complaint.