Print

Please login or register.
1 Hour 1 Day 1 Week 1 Month Forever
Login with username, password and session length

[AndrewFG]

Lossless compression offers data integrity checking and tagging.  Not to mention it saves half the space.

I am inclined to agree with these benefits..

The only mystery for me is why is there no native support for Flac in any mainstream operating system? It is not natively in Windows, Apple, or Linux. And it is also not natively in DLNA either.

[mwillems]

I am inclined to agree with these benefits..

The only mystery for me is why is there no native support for Flac in any mainstream operating system? It is not natively in Windows, Apple, or Linux. And it is also not natively in DLNA either.

I'm not sure what you mean with regards to Linux not providing native support for FLAC?  Of the distros that furnish media players and codecs, most of the ones I'm familiar with support FLAC out of the box (e.g. Mint is one of the most popular distros right now, and it ships with banshee which supports FLAC).

And for the rest, Linux is modular, and several distros don't ship with any media player or codecs at all.  So if "native support" means "only using software installed by the installation medium," many linux distros don't natively support MP3 and some don't even natively support audio playback.  If you look at available packages, though, almost every distro offers software that will play back FLACs in their repositories.

Maybe I'm missing something?  Not that the existence or absence of native linux support would affect your assertion that there is no native support for FLAC in any mainstream OS   (at least in the desktop space)

[Hendrik]

For Apple its kinda obvious, they don't care about anything they didn't invent themself, and thats ALAC, which is in direct competition to FLAC.
Technically Microsoft also has its own lossless format, WMA Lossless, however with Microsoft i just think they don't care, period, at least not on the desktop.

For hardware devices, its hard to say. Most likely there is just no great market in consumer-space for lossless.

[JimH]

Split Standards

[Frobozz]

dBpoweramp includes a custom Windows file explorer extension that adds a property page for FLAC files where you can see the tags and other info in the file.  Also adds a popup if you hover over a FLAC file that displays tag info and file info.

The dBpoweramp extension makes it feel like FLAC is more native in Windows than it is.

Similar extensions are how we'll get file formats to feel more native in various operating systems.

[AndrewFG]

My basic question is this...

If no major PC operating system supports it out of the box, and if no major tablet operating system supports it out of the box, and if the only media hardware interworking standard (DLNA) in town does not support it, and if there is no sign of these issues ever becoming resolved, then why does everybody still continue to claim that Flac is the de facto standard?

I think we are all deluding ourselves..

[Hendrik]

Android supports FLAC out of the box, as of 3.1 anyway. No multichannel or high resolution though.

[JimH]

I think eventually Microsoft will add it, but Apple probably won't.  For either, it would mean admitting defeat.

[mbagge]

DLNA will stream a Flac file just fine if the rendering device supports it.
I am using the DLNA function in MC to stream Flac music to my Denon receiver and to my Onkyo before that.
Just to get that one straight.

It is a valid point to argue that you have to do all sorts of installing and configuration to achieve 'built-in' support for the Flac format and sometimes even this is not an option for a specific device (smartphones, tablets ..)
At the same time we are installing other specialized apps on our devices to do stuff for us that didn't came right out of the box without any second thoughts.

Personally I never doubted the advantages in using the Flac format.

[fauxfreshness]

I think eventually Microsoft will add it, but Apple probably won't.  For either, it would mean admitting defeat.

Having worked as an engineer at Microsoft, I can safely say that it wouldn't mean admitting defeat.  Having VC-1 (the CODEC they pitched for HD video on optical media, and is still used in their streaming) not take off is much more of a defeat than some Open Source container format that houses AIFF or WAV source solution formats (which they co-developed the latter).  Adding FLAC means very little in that case to them, since they know the VAST majority of people that care enough will have an application or API extension that can deal with it.

The bigger issues might be:

• Not enough paying customers care
• It's dependent on an open source project, and developer

The first drives the second, but the second is the highest gating point of all.  Many might not know this, but the TCP/IP stack for Windows 2000 was taken from BSD, and was heavily improved upon in later generations.  They had one in NT 4.0, but it wasn't going to scale for their directory service (also partially based on genuinely free and open standards).  There are a few shining examples of how for-profit companies have had major issues using open source code and have been exceedingly careful with it.  Such as...

• OpenSSH source code breach where vendors didn't even bother to check the MD5 sums on the site to know that the code they were downloading for their own applications and operating systems (Apple, Solaris/Trusted Solaris, etc) was compromised.
• The XFree86 project that died almost overnight when the founder/owner decided to make some additional requirements to the existing GPL v2 licensing (and why Xorg got a huge boost from being based on an earlier XFree86 without the licensing garbage and why Apple and others moved to it).
• Remember the ISO mounting tool Microsoft made for burning DVDs and making bootable USB media for Windows 7?  Remember it being pulled off the their site shortly after for having GPL v2 code in it but not having its own source code released?  There are still "discussions" and training internally about what is allowed for use and when to involve legal departments for reviewing possible code and how it's been released.  BSD licensing is actually "free," meaning you can do whatever you want and then make it proprietary and sell it.  It's another reason why Apple's underlying codebase, "Darwin," is BSD based.  Well, basically entirely BSD.  They have supplied code back to branches of Webkit and Xorg, but from devs I've conversed with it's a mixed bag of usefulness.
• Route of legal recourse and support.  Microsoft CEOs have gone on the record as stating, repeatedly, that they will support any customer that has licensed their software and found another "company" (or patent troll) trying to sue the customer for using or infringing on said patents.  If FLAC got major backing from a company like Microsoft directly, that could "wake up" other companies and patent bodies to take interest and line up at the door demanding money.  This is partly what MPEG-LA does, and even though Microsoft has agreements in place for AAC support (remember downloading the additional CODEC for .m4a files to work on Xbox 360 consoles? This is part of the license inventory system so they knew what to report.  Nero does something similar, along with others.)  MS is making billions from Android handset makers on patents that many of which haven't been publicly specified or challenged directly.  In other cases, there are cross-licensing agreements in place or even support and marketing agreements in place to "entice" possible litigation to be dropped and promote something else (Windows Phone, anybody?)

There are a lot more moving parts to this, but in my case I'm just glad this isn't being tested legally.  I mean Ogg Vorbis/Theora have been around, and almost nobody cares.  But as soon as Google announced its own video and audio compression solutions MPEG-LA immediately stated that they'd "monitor" the situation, further claiming it would be "impossible" to have a compression algorithm that wouldn't infringe upon at least some of the patents the body holds. 

As a result of seeing this from multiple sides (I'm currently a software dev/contractor) with companies and government entities, I'm actually glad that FLAC is around and blooming, but not getting direct backing in scale from anybody that matters.  Maybe this whole concept of patenting mathematical truths (which you aren't supposed to, right?  ;-)) will wind down further.  Or not...

[AndrewFG]

^

A very interesting post. Thank you for sharing your knowledge and your insights.

[fauxfreshness]

No problem.  :-)  My interest with FLAC only came about because of making music and having most MP3 CODECs screw up the master's performance and creating weird gains or shifts in bass/mid-bass frequencies, or creating a TON of ISC (Inter-Sample Clipping) points where the plotted solution exceeds 0 dBFS.  So I went with AIFF as it's fantastic and has great tagging support, but Android doesn't have native support and can act really dumb with playlists using them.  I read a bit on FLAC, and it made sense for my usage in this case and my 24-bit/48 kHz masters were handled gracefully by BandCamp when I tested downloading them.  FLACs have always sounded great, but I didn't have a real reference without source material that I knew performed a certain way on my fixed reference combination.  Oddly, I've had portable apps for playing FLAC files forever, but I just never got around to using them until recently.  ;-)

[candycane]

Yes fauxfreshness, a very interesting and insightful post, thanks for sharing.  IANAL, but with my limited exposure to legal stuff in this area, and the apparent fact that Josh Coalson began development of FLAC no later than 2000, I think that the risks of "Route of legal recourse and support" at least are manageable in the U.S. at this point.  FLAC is supported in lots of paying software and to my knowledge none have been attacked by patent trolls yet, and with the 20 year limit from filing for U.S. patents filed after June 8, 1995, it would seem patent trolls might not see enough potential blood money to attempt anything worrisome. With fair certainty, any troll trying to assert a U.S. patent filed after 2000 will have a significant "prior art" issue.

Overall, Xiph (which has taken over managing FLAC from Josh) seems rather sophisticated and pragmatic in managing intellectual property issues, and managing any issue I've seen in general, so if MS was open to the possibility of natively supporting FLAC I'm somewhat optimistic a deal could be worked out. I would agree that the current support/development model for FLAC is sketchy at best (especially considering the personality of the lead person for FLAC development at the moment, but that's another story), but my gut says Xiph would provide a more robust support model with at worst a reasonable financial contribution to cover support and development costs that would be negligible to MS.

FWIW, FLAC libraries are apparently covered by BSD, but whether that's sufficient or not for enough of MS's concerns to make this worthwhile is beyond by expertise.

In short, while I think the "not enough paying customers care" could be the more significant factor, I'm hopeful the actual costs could be low enough that the downside risk is low and upside potential benefit is worth the effort.  But I'm an optimist, so I could be wrong.

It would be cool for someone with the connections to try to connect MS and Xiph to explore this, noting my complete ignorance if this has been tried already.  Is there anyone listening that could try to set up a "speed date"?

[fauxfreshness]

I think there's a misunderstanding in scale here.  I'm talking millions of sales with the promise of an on-going installed base.  This is why interface patents and other stupid things don't make it to court unless the entity suing has a strong business case for it (no, not necessarily legal case) and the ability to reference the existing units that were sold and shipped.  The defendant then has to determine how much of a loss they're will to take if they think they can't win, or if they think settling out of court (hence the strong business case from earlier) is a better way to go.  Hell, Microsoft patented the rotary interface the iPod used after the iPod was out.  Guess what company got really, really into patenting things and suing others, even though its psychotic founder claimed great artists steal...

After that, you get the oh-so-fun "America Invents" Act, which among other things drops prior art in most cases and simply defers to the party that filed the patent first.  Period.  Have you noticed even more wonky and stupid patents getting dug up from Apple lately?  As a company, you normally have attorneys that research if there are existing patents that cover something, or should be referenced in the new filing.  They would also (in many cases) perform "best effort" searches for prior art to see if it would stand up to a challenge.  Now, why bother?  The US Patent and Trademark Office lets just about everything go through, so you just annotate internally any concerns, but nothing with "intent" to indicate you knowingly copied anything.

In my case, I'm more worried about Copyright at this point (I lost interest in filing patents while at MS and one other company) since I'm writing music.  It's a little more straight forward, although making money is a whole other issue altogether...  ;-)

[candycane]

I completely agree on the scale issue from a risk perspective, but the specific timing in the case of FLAC makes me optimistic. With respect to the  "America Invents" Act, it actually expanded the definition of "prior art", which is good in this specific case, but otherwise generally only applies to patents filed after March 16, 2013.

You'll never get me to defend the U.S. patent system as currently executed and enforced, but I'm actually relying on the shrewdness of patent trolls to support my optimism.  More specifically, trolls will use "expected value" theory to calculate the risks and rewards of going after firms (as will firms like MS in assessing efforts that might attract such trolls).  My assessment (based on having run such analyses for firms on both sides of such issues) is that going after any paying software supporting FLAC *at this point* will not be a good investment.

[fauxfreshness]

I brought up the "America Invents" Act specifically because of this.  Have there been any patents filed for FLAC?  I didn't see anything on the site when I was going over the specs for it, and we'd clearly be after the date in question.  Given the licensing being used, it appears counter-intuitive to have a patent (or collection of patents) in place just to license them for free so you can defend them if you need to.  Just like MS patenting an interface that already existed, or Apple patenting a "rectangle with rounded corners," it's more about when you committed the design to paperwork and how hard you can defend it.  Which is a shame.

But the original question regarded normal OS usage, or rather it including CODEC support by default.  I think we've covered that pretty well by now.  I basically expected Google to do it because, well, they act like a scrappy small startup, except with billions to back it up and the ability to acquire or crush anything in its path.  Except the recording industry.  That's still a work in progress.

[candycane]

I agree having patents or pending would conflict with an open source approach.  FLAC has not applied for any patents or been granted any patents to my knowledge, which strengthens the case that their library licensing under BSD is not a sham or anything else Xiph represents is not a sham. In other words, I see nothing to indicate that Xiph is a patent or license troll or a troll-in-waiting.

Google does offer native FLAC support in Android, since 3.1, which came out more than 3 years ago.

[fauxfreshness]

Yeah, I bothered with FLAC since it's natively supported in Android since Android 3.1 and I figured I'd give it a shot for reviewing my masters before SoundCloud murders them.  I have all 4.4.x devices now, although I'm not sure which one my FiiO X5 I just got is running, but it's been pretty good so far today.

But I wasn't inferring that Xiph was a patent troll.  Think about it.  What if someone else files a new patent, or tries to amend an existing one (extend your coverage under patent by filing little tweaks here and there and getting new ones with new spans of coverage) using that existing work.  This is what I'm talking about, and why I had hope since Google was supporting it themselves because, quite frankly, they want to start up as many fights as necessary to show how flawed the various systems we have in place for inadvertently regulating the tech industry (like the mockery of bids for the 700 MHz spectrum).

Hell, Steve Jobs even claimed that Blu-ray's patents and licensing were "too complicated" for distribution with their gear, even though they sat on the governing board for the standard.  It's more "we don't want to expend the resources to get involved, and cut into our existing profit margin; if third parties want to handle it, then great."  Since I rip all my Blu-rays using AnyDVD HD in Windows, I still prefer having a main Windows workstation for this.  Since I've found tons of other software I was going to use on a Mac that works great on Windows, I'll probably not bother getting that Mac Pro afterall.  But a MacBook Pro is still feasible.  :-)

[Vocalpoint]

If no major PC operating system supports it out of the box, and if no major tablet operating system supports it out of the box, and if the only media hardware interworking standard (DLNA) in town does not support it, and if there is no sign of these issues ever becoming resolved, then why does everybody still continue to claim that Flac is the de facto standard?

My answer would be - who cares if the OS supports it or not? As long as Media Center supports it and I can run MC on a PC - there is no reason for the OS to ever support it. This is no different that a Excel spreadsheet You cannot open and edit an Excel spreadsheet on ANY windows OS without Office installed.

As far as tablet goes - JRemote streams my FLAC files in full resolution to my iPad - so who cares if the iPad doesn't know what a FLAC file is.

And for portable use - sure it would be nice to drag all my FLAC files around but there is no portable device (and probably none coming) that would ever satisfy the "space" issue - so it's simply easier to use MC to create a nice MP3 version of the FLAC file for car play, iPod play or what have you.

I have committed 100% to FLAC here and have never had an issue with anything. And it's clear that's it's open source pedigree, superb metadata handling and awesome sound (with space savings!) make it the new world standard for a high res music container - honestly I can't see an issue at all. If I did - I would not use it.

VP

[6233638]

As far as tablet goes - JRemote streams my FLAC files in full resolution to my iPad - so who cares if the iPad doesn't know what a FLAC file is.

If these devices had enough storage that I could keep my entire library on one, I wouldn't mind. It's a pain to convert files every time I want to make a change.

[Vocalpoint]

If these devices had enough storage that I could keep my entire library on one, I wouldn't mind. It's a pain to convert files every time I want to make a change.

True dat.

But my FLAC library is in the TBs now...there won't be a device that can ever handle that. So I am okay with trans coding to MP3 for mobile use. I find the MC does it nice a quick and quick - it's no problem to load up the iPod from time to time.

Cheers,

VP