"It's like you're working in the music industry."

Heh, I guess you didn't read all my prior posts in which I stated that part of my preference for MP3 was its superiority as a sharing format on the various private music trackers I visit. As a frequent uploader, this is important to me. If I uploaded Ogg Vorbis, I would get few if any downloads.

I use FLAC for my music archive. I listen to FLACs in foobar2000 with my DAP/Headphone Amp/Sennheiser HD-580s. I also transcode FLACs for sharing.

I use MP3 for the sole purpose of sharing with others. I never listen to the MP3s. I have a DAP but barely use it. My collection is too large to fit on my DAP, and my portable use doesn't justify an upgrade. I listen to classical music which is difficult on a portable due to ambient noise. (the lower dynamic range makes it more difficult to hear the soft parts)

So, I choose the format that the vast majority of people want - MP3. My friends can play it. My gf can play it in iTunes. And it's preferred on all the trackers I frequent.

Perhaps I should be more clear. I would be quite supportive of the use of Ogg Vorbis for internet radio and for streaming video, where licensing costs are a concern.

Ogg Vorbis is an excellent format. While I wouldn't use it for my own archive due to the network effect advantage of MP3, that is simply not an issue for streaming audio and video.

Ogg Vorbis is a good choice for HTML5. Theora is not. I'm hoping Google's new codec acquisition is superior.

Well, I think the difference in treatment has a great deal to do with the differences between audio and video, and human perception.

For most people, LAME V2/V0 is transparent, as is 256 kbit AAC, q6 Ogg Vorbis. Thus, there is no perceptible difference between any of these formats.

It is MUCH easier to tell quality differences in video because the amount of compression is so much greater (50-100x vs. 5-10x). For example, I don't know anybody that can ABX DVD-Audio/SACD against CD. 24/96 audio is technically superior but virtually nobody can hear the difference. The difference between DVD (480p) and Blu-Ray (1080p) is like night and day for those that care to look.

The choice of encoder makes a lot more difference in the video arena, as does the choice of codec. I'm relatively indifferent to lossy audio codecs because they all sound the same to me, but x264 CLEARLY produces output superior to Theora or xvid/divx at any given bitrate. It retains more detail. It retains more grain. It's just better - and obviously so. In fact, the differences between H.264 encoders can itself be dramatic. x264 is VERY GOOD.

1) Theora isn't "marginally worse" than x264 in encoding quality. It is a lot worse. In certain use cases it may be acceptable, but it simply can't retain the level of detail that x264 can for a Blu-Ray sourced 1080p encode. The same is true for Xvid/Divx. H.264 demands greater CPU cycles for encoding and decoding but it can achieve much better results at a given bitrate. My 1080p encodes at 12-14 Mbit are virtually indistinguishable from the source.

I don't think that was ever the case for Xvid encodes of DVDs at any bitrate. Xvid just couldn't retain all the detail, even at very high bitrates (relatively).

H.264 is also the preferred video format for all high definition encodes, due to its clear superiority and the quality x264 encoder.

Thus, the H.264 v. Theora comparison is different from the MP3 v. Ogg Vorbis comparison for these reasons:
1. H.264 markedly better than all competitors at any given bitrate (VC-1, Xvid, Theora, MPEG-2 etc.)
2. Most people can easily tell these differences without prior training, or seeking out flaws (many MP3 artifacts can be detected only after training yourself to listen for them)

2) Ogg Vorbis is marginally superior to MP3, but since both are transparent to most users at 192-256 kbit range (when using LAME V0 or V2) that advantage doesn't matter. MP3's ubiquity and "good enough" sound quality takes the cake.

NOTE: Theora may be "good enough" for online video sites such as Youtube, but it will never be good enough for high-quality Blu-Ray and DVD rips. However, I doubt even this use. Youtube videos are very bitrate starved. This is where codec differences make the most impact. I just don't think Theora can compete with H.264 at very low bitrates.

And as home users' bandwidth continues to increase, true high definition video streaming will become more common. I definitely care more about video qualtiy when I watch HD than SD. High Definition video provides so much more detail that minor flaws become more apparent.

The current HD streaming on Youtube is terrible since the H.264 videos are so horribly bitrate starved and thus blocky as hell. When this changes, our expectations may increase markedly. I just don't think Theora is in the position to compete.

The Rockbox firmware has relatively simplistic failings. For example, look at the screenshots on Wikipedia or their site. The text is so small that it's relatively difficult to read. Apple's firmware is certainly less powerful.

However, I think you missed my basic point:

If you use Ogg Vorbis, you have to make your DAP purchasing decision based on your audio codec choice. This simply shouldn't be the case. You shouldn't have to severely limit your hardware choices because of your chosen audio codec. Ogg Vorbis provides some advantages over MP3, but none that are significant enough to justify replacing my current DAC.

Furthermore, preferring a popular format due to network effects is perfectly reasonable. If I want to share an album with a friend, providing an MP3 will ensure that it will be supported by his music program and DAP of choice. MP3 is universally supported, and thus the simple choice for file sharing.

I don't even think Ogg Vorbis is "nerdy". The reason people use MP3 over Ogg Vorbis is not because they view the latter as "nerdy", but rather because Ogg Vorbis provides no significant advantage over MP3, and the popularity and universal support for MP3 makes it a perfect format for sharing, while at the same time ensuring your files will be playable for some time.

Anyway, the market speaks for itself. There are many Ogg Vorbis evangelists. Despite the outspoken support, there is virtually nobody sharing Ogg Vorbis content on bittorrent, Usenet etc. Sure, it's often used in games due to the licensing costs of MP3, but even most internet radio stations use MP3.

I am certainly not telling you to switch. There are many nice attributes of Ogg Vorbis. It certainly can achieve transparency at lower bitrates and can achieve a higher overall level of audio quality than MP3. If these factors are paramount, or you simply want an open source lossy format, by all means, use Ogg Vorbis.

But I think it's important to recognize why MP3 is and will likely continue to be so popular. Your view that choosing a format due to its popularity is somehow reflective of ignorance of technology or anti-nerdiness is flawed. Network effects are real. MP3 is a more valuable format because it is universally supported on all DAPs and software players, and therefore can be shared with near 100% certainty that it can be played with little effort.

While licensing issues have introduced issues on Linux, it's relatively easy to install MP3 support on Ubuntu, Debian and others. And Linux users tend to have a higher level of technical competence anyhow.

But if I want to send an album to my girlfriend, I am not going to do so using Ogg Vorbis. She will complain that she can't play the files in iTunes. It's just not worth the effort.

Would I use iTunes? Hell no. I think it's a bloated piece of crap. I use the highly configurable foobar2000, setup to my liking. However, my format choices are somewhat dependent on the choices that others have made. Since many users DO want to use iTunes and DO want to use their iPods and Touches, it makes sense to use a format that they can easily use. This is especially true considering LAME V0 is transparent to my ears anyhow.

But please explain how an audio codec can be "cool" or "nerdy". I don't use FLAC or MP3 because they're cool. I use them because they are the most functional for my purposes. I assume you use Ogg Vorbis for the same reason. The fact that my choices are more mainstream doesn't make your choices somehow superior or better informed. We simply have different interests.

I am quite sure that if you were on a private music tracker, for example, you would rip and upload in one of the more popular formats. Even if you wanted to download in Ogg Vorbis, you would have great difficulty finding anything to download.

Sorry for the spelling errors and such. I wrote my response quickly.

Your points are good ones. However, you and other make reference to "low bitrate MP3s". Isn't that somewhat besides the point? If you're going to compare Ogg Vorbis q6 or q8 to MP3, you should compare to the best and most widely available codec and encoding options.

By far, the most popular MP3 encoder is LAME, and the preferred format is V0 or V2. V0 generally has the most downloads on the popular private bittorrent trackers, but V2 is the choice of "scene" release groups.

Neither is low quality. V2 averages around 192 kbit/sec and V0 around 245 kbit/sec. There is very little perceptible difference between LAME at V0 v. 320.

While Ogg Vorbis using the aoTuv encoder can achieve transparency at a slightly lower bitrate, from my ABX tests, and reading on Hydrogenaudio, you save, at most around 5%.

I grant that this may be valuable on a DAP. While there are DAPs with Ogg Vorbis support, lets be honest here. What's the most popular DAC on the market? Apples iPod or Touch/iPhone, most likely. The vast majority of people I see with DAPs use an Apple product. Unless you use the third party Rockbox firmware, which substantially reduces the UI quality, one of the reasons people pay a premium for Apple products, you're not going to have Ogg Vorbis support.

So, while it is possible to buy a DAP that supports Ogg Vorbis, you have to make your hardware purchase based on your codec choice, which really shouldn't be the case.

You suggest that Ogg Vorbis support is only a problem if you're looking at cheap DACs. This clearly isn't the case. If you want to use an iPhone, Apple Touch, an iPod or any other popular DAC, you have two lossy choices - MP3 or AAC.

So, while you do gain some space by using Ogg Vorbis, IMHO it's not worth the limitations in hardware and software choice.

As for desktop usage, conventional hard drive storage is so cheap, that it is really inexpensive to simply rip to FLAC and your choice of lossy format. In fact dBpoweramp can rip a CD to both FLAC and any lossy format of your choosing in one simple step, and it will use the Accuraterip database to confirm the rip was accurate.

Or you can use EAC, which is free, and then use dbpoweramp or another tool to batch convert at a later point. While my i7 makes the process very quick, even a mid-range Core 2 Duo can transcode a large FLAC collection in a few hours, and certainly overnight.

With 1.5 TB drives at $100, hard drive space simply isn't a meaningful concern.

The market for lossy audio formats on private bittorrent trackers shows the strong preference for MP3. A recent statistical analysis at what.cd showed that MP3 was by far the most popular lossy format, accounting for at least 98% of all downloads.

Ogg Vorbis had very few downloads, and Ogg Vorbis uploads were the most likely to have 0 downloads. It's just not a popular sharing format for the reasons I raise above.

That doesn't mean nobody is using it. FLAC has become much more popular as of late, and lossless formats provides the benefit of being able to transcode to any lossy format retaining tags and album art.

I still think downloading, and certainly, riping to FLAC makes the most sense. You then can purchase any DAP of your choice, and know that you'll be able to convert your music to a supported lossy format. Even a relatively old computer can rip to MP3 relatively quickly. Foobar2000 even has a plugin that will do this for you automatically, for DAPs, including the iPod that don't support your format of choice.

BTW, I'm only referring to encoding for personal use. Obviously, when widespread distribution is a concern, you must consider licensing costs, and size becomes a greater issue.

As a matter of practicality, I don't see why FOSS advocates continually argue for Ogg Vorbis and don't talk much about FLAC. It is probably because MP3, and lossy formats in general are considered to be so popular. But, from my experience at what.cd, a private music tracker with over 120,000 members, FLAC is most certainly popular.

It only has about 1/3rd of the downloads of MP3, but by filesize, it's about equal to MP3. Despite the 300% file size difference, many people strongly prefer FLAC. I listen to classical music, and for many albums, FLAC is so preferred that there are simply no MP3s available, or any other lossy format for that matter.

FLAC is open source and not patent-encumbered. It also has an array of great features - wonderful error detection, frame-accurate seeking, extremely fast integer-math decoding (4-5x faster than MP3 making transcoding very quick).

FLAC is undoubtedly the most popular lossless codec. And from what I can tell, as the cost of hard drive space has dropped, and user upload speeds have increased, FLAC has rapidly surged in popularity. I wouldn't be surprised to see FLAC replace MP3 in 5 years as the format of choice for online file sharers.

There are many good reasons for lossy codecs, but as an archive format, lossless will always reign supreme for its audio quality and flexibility.

The problem with this argument is that it somewhat misses the point. MP3 is "good enough" for the vast majority of users at LAME V0/V2. I would venture a guess that 95-99% of persons couldn't ABX at V0 in perfect conditions (expensive amp, DAC, and high-end headphones), yet if we're talking about the use of a DAP and earbuds, it is quite clear qualtiy isn't relevant.

MP3's primary advantage is its effective standardization and universal support in all hardware and software. This single advantage far outweighs any benefit Ogg Vorbis can provide. An MP3 can be played on any DAP, on any operating system (with the right codecs), and all music software. It's therefore the preferred lossy sharing format. On the large music trackers, Vorbis makes up fewer than 1% of lossy downloads by file size and # of downloads. MP3 is the clear preference.

The fact is that, while Ogg Vorbis, may be better than MP3 quality-wise at V0 or 320 CBR, this is not the main point of lossy audio. If your primary concern is quality and archival, you shouldn't use any lossy format. You should use FLAC - it is open source and has superior error detection features (MD5/CRC for each frame, use with Accuraterip to verify any disk).

I use FLAC on my desktop and only download EAC rips with 100% logs, or try to, at least. This ensures that my downloads are "perfect rips", and the encoding process has not reduced quality at all. With a single click, I can verify my FLACs against the Accuraterip database to ensure they are perfect.

No lossy format provides this benefit. If I want to put the music on my iPod, I can convert it quickly with my Core i7 (45 seconds an album). I can convert my entire collectoin in several hours.

So, why use OggVorbis over FLAC?

Matroska is a non-patent encumbered container which is much more flexible and can with zero effort, support virtually any audio, video and subtitle format. Nobody opposes OGG because it's patent-free. They attack it because there are superior options - namely Matroska.

Except of course you're ignoring the fact that a monopoly created on the basis of customer preference/superior service is not a violation of the Antitrust Laws. Maintaining their search monopoly by continually adding features and increasing the quality of their product, thus preventing competitors from gaining a large enough market share to compete effectively, simply isn't a violation of the Sherman or Clayton Acts.

And I don't see them using their monopoly unfairly to expand into other markets. They do link to their other services, but often the top hit for most of my searches is Wikipedia. The top hit is almost always the most relevant, or at the least, a highly relevant, source.

Why shouldn't they provide trailers on Youtube. It was the most popular provider of online video clips when it was purchased, and continues to be so today. Would you force them to link to another site, even where that site is inferior? Customers want links to Youtube. Also, they do provide links to other popular video sites, if there are relevant hits. Obviously, there will be more Youtube hits, on average, because of the site's popularity. I just don't see any attempt by Google to suppress their competitors.

Being big + better than your competitors =! Antitrust violation

I read the article (shocker, I know) and I am quite certain the case HAS NOT gone to trial. The article states that "An Erie County jury has been asked to decide whether a 12-year-old girl was burned . . ". The court "has been asked' to decided. The article does not even state whether the case has made it past a dismissal motion for failure to state a claim or a summary judgment motion.

From what we know, the claim is very early in the legal process - so early that the defendant would not even have had an opportunity to file a motion to dismiss.

Claims are usually dismissed before going to trial for two basic reasons (assuming that technical requirements such as standing, personal and subject matter jurisdiction, and venue have been met, which are outside of the scope of this response)-

1) Failure to state a claim: There is simply no cognizable legal claim raised by the plaintiff. To win a motion to dismiss, a defendant must show that even if the court accepts all material facts claimed by the plaintiff, there would be no legal wrong. That standard does not apply here, because IF her claim is factually accurate, she has suffered a legally cognizable harm.

2) Summary judgment: This occurs before trial, and requires a showing that the plaintiff has failed to make a sufficient showing that she can satisfy the factual or legal burden of her claims. This case will almost certainly be thrown out on summary judgment because it is basically inconceivable that any amount of discovery or case development will allow her to factually establish her claim.

Thus, I would be willing to bet a large sum of money that:

1) This case never makes it to trial
2) If it does make it to trial, the plaintiff will lose
3) If the plaintiff wins at trial, the trial judge will direct a verdict for the defendant, or will overturn any monetary damages on the basis that no reasonable jury could find in favor of the plaintiff OR the judgment will be overturned on appeal.

I am not a lawyer (I am a 3L at Yale Law School). This post should not be construed as legal advice or the creation of an attorney-client relationship. If you need legal advice, consult an attorney.

I agree. I think we should scrap the Electoral College as its an antiquated, undemocratic, and unnecessary intermediary. It's only positive function is that it theoretically gives smaller states more influence, but it arguably gives too much power to small states. And, in any case, I think we should move beyond the idea of representing state sovereigns rather than persons.

Thanks for the correction.

Actually, ignore what I said about federal elections. The US Constitution defers to state standards as to the handling of elections for federal office, unless Congress intervenes to the contrary.

In general, Congress has deferred to state law.

"In virtually all other elections it's just 50%+1 of the popular vote."

I should have been more clear about the popular vote, but your claim about other elections is false. Very few state or or local elections, and no federal elections, require more than a plurality of the popular vote. Some states have run-off elections, but in the vast majority of elections, it's sufficient to obtain a plurality. We would have a great deal more run-off elections otherwise, because many close elections with third party candidates result in a winner who obtained less than a majority of the vote.

As for the Electoral college, while it is true that a president could theoretically win with a mere 10-25% of the popular vote, I only recall two elections in America's history where the electoral vote diverged from the popular vote.

You don't even need 50.1% to win an election in the United States. You just need a plurality of the votes. In fact, it was considered noteworthy that Obama won more than 50% of the vote because it had been decades since a Democratic president had won an outright majority of the popular vote, rather than a mere plurality.