Engineering Archive Data Record Software Interface Specification: Fill & Download for Free

As has been covered in the other answers, ProTools is considered the industry standard (inasmuch as there is such a thing anymore) more because of it’s widespread use than anything else. One answer above mentioned that in the hands of a skilled user it’s the fastest DAW for editing in, something I would have agreed with a few years ago but now largely believe is myth. I was a long time ProTools user and I can edit as fast in Logic as I ever could in PT. If you know your tools well you are going to be fast in anything. These days the actual featuresets of most pro DAWs are basically identical for baseline tasks.Regardless of that PT has been the most widespread DAW used in commercial facilities for two decades, and although it’s certainly starting to lose it’s hold (at the last two facilities other than my own that I was working in, more than half the engineers were on Logic and a scattering on Abelton or Cubase), it’s still the DAW of choice for most studios, esp if they need to support freelancers and work with other facilities.What no one has touched on and what I assume you really want to know is why it became that way. People have discussed promotions and marketing and such, but this was 20 years ago and is still now an industry where time is big money. If there wasn’t something superior about it, no one would have used it regardless of marketing.At the time PT/Sound Designer came about, the age of the common desktop personal computer was beginning it’s upswing. The music world had already long been able to use computers for sequencing MIDI information, and the solidification of MIDI as a communications standard across the MI industry was evidence that computers and music could play nice. The rise of more powerful computers and operating systems with (relatively) advanced GUI’s meant you could do more with an affordable, relatively small machine than ever before. Also we’d known about the systems for recording audio information digitally since the late ‘50s, and we were well into the CD era, heck we’d been making fully digital recordings since the ’70s.Thing is, most desktop computers couldn't handle the demands of digital audio.Enter two guys who were UC Berkeley grads who had developed software (Sound Designer) for editing sounds for the E-mu Emulator sampler and a universal file spec for transferring those samples between other sampling keyboards. This was originally to make their own lives easier but they soon figured it was something that would fill a more universal need and developed it commercially. At almost $1k for the software it wasn’t cheap, but it was a LOT more accessible than Fairlight and Synclavier systems of the time.They then started Digidesign together and released a more advanced ‘Sound Tools’ software, which was basically a stereo audio editor, and though it could do more, it was limited by hard disk speeds and computer power.The first versions of Sound Tools were their first attempt at building a direct-to-disk recorder with a software interface. The mac’s 8bit audio capabilies weren’t adequate however, and when they started working with a 16bit external audio card the computer choked on even simple editing tasks. You were, simply, better off with a grease pencil, a razor blade, and tape.But then in 1987, the Mac II came out complete with expansion slots (NuBus, very pre-PCI). These guys figured out that with a brand new type of DSP chip from Motorola, and these expansion slots, you could offload the heavy lifting of processing the audio data to a dedicated purpose-built processor, leaving the host computer to the more mundane tasks of running the interface and software. In 1991 ProTools had it’s debut. $6k got you 4 tracks of IO.The release of PT 2 and 3 in 1994 brought in the TDM architecture to further expand track counts (up to 48 simultaneous ‘voices’ depending on cards). By 95–96 PT hit the 24 track mark and was now in direct competition with 2″ tape, but with fewer drawbacks. By 1997 ProTools was offering configurations of up to 64 simultaneous ‘voices’, 24bit 48kHz recording/conversion/processing, as well as a healthy suite of plugins driven by the TDM DSP architecture.And here’s the crux of it… no one else was doing this at the time. Digidesign capitalized on the growing popularity and accessibility of the personal computer and on it’s potential as a tool for the creative professional fields. They got in first, yes, and they did it well, and they did it in a way that filled a need.At this point in history you have to remember that ‘home studios’ were by and large 4–8 track cassettes or open reels and maybe a crap ‘prosumer’ mixer. Commercial studios were still heavily invested in tape and other linear recording formats as well as synchronizers etc… If you wanted more than 24 tracks of playback, it involved multiple tape machines and expensive synchronizer systems to make them all playback and locate properly. Tape machines had to be maintained and aligned regularly, and if you’ve ever had to re-cap the headamps on a 24ch 2″ machine well.. not fun and takes your studio out for a few days.Not to mention tape was expensive, required careful handling and storage to archive, and after too many passes ‘shedding’ would destroy your tape and everything on it. It was also relatively coloured. That wonderful old ‘tape sound’ that everyone chases now, complete with wow, flutter and hiss was all well and good, but by the mid 90s engineers had gotten skilled at trying to minimize many of these things when desired, not add more. Digital delivery formats, while initially a little shakey, were allowing artists and producers to put out incredibly revealing and detailed recordings. The idea of being able to work digitally from start to finish may not have been universally chased, but it certainly wasn’t unwelcome.By ‘96-’97 or so, ProTools had exceeded the track counts of most commercial studios with 24-64 voices, at 24bit resolution and was relatively cheap compared to the cost of two or three tape machines, a synchronizer and the ongoing cost of tape (remember tape machines were NEVER cheap hardware… esp the good ones. A 1984 Otari MTR90 can go as high as $10k now…). And that was if you only considered it as a recorder alone. But with the editing capabilities it offered and the audio processing plug-ins…. It became a no brainer for many big commercial facilities, esp post houses who needed high track counts, precision editing and fast non-linear locate on the regular. 96–97 is where you saw the big shift of commercial facilities and even some small project studios moving to computer based systems. The other big thing they did that helped was making the DAE core of their audio engine open to other developers. This meant you could use all of Digi’s big powerful hardware with any software platform that chose to integrate it.In short: Because Digidesign has always focused on offering a software and hardware solution, they were able to offer stable, fast, powerful, computer based audio workstations before anyone else. And although they were expensive, they offered solutions and enhancements to a large number of limitations of tape based systems. So studios jumped all over the platform. Knowing you could invest in the hardware and use it with multiple software hosts was a bit of a security blanket for the owners/managers as well. Add to the mix the fact that in industries like this small facilities want/need compatibility with big facilities and it starts to snowball pretty fast.Looking around at DAWs today and even in the last decade, it’s harder to see why PT and Digidesign were on top. In fact it’s specifically because of that adhearance to hardware/software packages that Digi/Avid are losing market share now with basically anyone BUT big studios. They had the reverse problem that as computers got more powerful and cheaper, fewer people could see the point in buying a DAW system that locked you into hardware, and was extremely expensive compared to everything around it. The TDM/HD processing power certainly helped it hold on as a ‘power’ tool through the early 00’s, but by 05–07 when even a basic computer could run 48+ tracks of audio and healthy plugin counts… well it’s a much harder sell. And Digi/Avid resisted going fully native for SO LONG, that the base of up-and-coming engineers that would have become the core evangelists when they became established, couldn’t understand why they had to pay more money for crippled LE systems shackled to interfaces that underperformed hardware and half the cost. And even than weren’t fully compatible with the big-boy systems.These days, Avid offers the PT software fully native, but it took them till 2010 to do it! The cost for PT is still high and Avid and Digi’s business model has been for many years all about nickel and diming you to death even if you bought the big systems. (Want MP3 import/export? $20 please. Sure we know you just spent $2k on an LE system. Oh you want 48 tracks on that 003? $450 please. Timecode? Another $450 for the DV Toolkit. Upgraded your OS? Please wait 6 months and then hand over a few hundred for the service pack for your HD system)Still, 20 years ago ProTools was the only thing around that could compete with the incumbent systems (tape) of the time, and not only compete with, but actually outperform and give you features that tape could never accomplish and tie into what was fast becoming the personal computer revolution… And you have the major components of why PT was the undisputed industry leader for almost two decades.

And I dive in, defending the indefensible, giving voice to those who already won’t shut up.I am going to open myself up to ridicule when I say that the web—HTML, CSS, JavaScript, the browser, et al.—is not the problem. The problem is, like so many things in software, a management problem. This is not unique to software. My old CTO would often say “all problems are management problems.” That’s not to say that all challenges are management, but challenges are what businesses and organizations naturally face. It’s part of being in business. Challenges can become problems when bad management gets thrown into the mix.Our troubles flow from many wellsprings, but I think it safe to generalize and say that the core of the problem is a disconnect between perceived value, the actual effort required to create good software, and the consequences when it all fails. To illustrate what I mean, a history lesson.The web started off as nothing more than an interconnected web of text documents. Once you learned the correct mark-up (HTML stands for Hypertext Markup Language), you could write a page and link it to other pages easily. The idea of a web “developer” didn’t make sense. In the earliest days, what was required was so simple that the first sites were put together by people in their spare time, or by secretaries and other document managers.As such, when sites grew larger and required greater levels of maintenance, pay and prestige were inherited from those who had previously done the work. It’s no surprise that the earliest tools for creating more complex sites (PHP, Drumbeat, Dreamweaver) were aimed at this demographic. The idea of programmatic quality didn’t exist because there was no program. And the idea of highly-skilled practitioners didn’t exist because people saw websites like Word documents.The inertia from this first stage is still with us today, both culturally as well as technically. Culturally, web developers are seen as “less than” and are systemically disrespected. Women are more likely to be front-end developers than middleware, back-end, or data. Engineers that work primarily with HTML, CSS, and JavaScript are often claimed to not be “true” engineers. And viewing the front-end as a near-afterthought is endemic among the programming cognoscenti. It’s no surprise that people who seek recognition and respect are driven out of front-end work.Further, it is no surprise that front-end jobs are among the lowest-paid programmer jobs on the market. Many company leaders were leaders in the 1990’s, when these prejudices first took root. So even though it is understood that websites are important, and web applications are valuable, there is a never-ending push to produce these things at costs more in line with perceptions based in 1997. Yes, there’s also problems related to demonstrating value, which is why analytics are basically an addiction, but I argue that the problem is not one of valid business concerns, but of a perception that businesses think they should be able to get this work for cheap.Technically, the cheap WYSIWYG tools of yore persist in the form of frameworks and libraries everywhere. Dreamweaver, as can be seen in the above Google Trends chart, is a shadow of its former self, but today has simply been replaced by tools like jQuery (red), React (yellow), and Angular (green), which can be seen in the chart below. We are thus at a point where it is standard operating procedure to deploy a framework or library to achieve nearly everything on the front-end.Further, the choices of protocols, formats, and rules for new technologies were necessarily influenced by the technology that came before. It’s why, by 2009, AngularJS labeled itself as “HTML if HTML had been created with applications in mind.” It’s why a simple UI library, Bootstrap, is the most popular product repo on Github.[1] It’s why the aforementioned jQuery became a near-default inclusion on every website in the world for a decade. And more contemporarily, it’s why Web Components continue to be discussed even though they’re basically dead. They are shields against the inherent difficulty of software development and a desperate cleaving to the paradigms of the past. Every stage of front-end evolution has been a progression away from simple markup and toward “real” programming.Concomitantly, every stage has been a progression away from a perspective of front-end developers as lower-skilled, less-valuable engineers. The engineers suffering under these systemic misconceptions are, I think, a big reason why front-end engineering has grown far more complex than it needs to be. It is the proverbial pendulum swing, with developers going overboard to prove their bona fides. This drive will persist for as long as companies fight to keep salaries suppressed and use exploitative outsourcing firms.Not all of the increase in complexity is entirely arbitrary though. The demand for more robust tools and user experiences necessarily demands greater front-end complexity. Basically, it is getting harder and harder to effectively compete on the open market with low-quality developers building the user interface. As the companies that wish to compete on the open market face the headwinds of developers not answering recruiters, quitting, or straight up ghosting, minds must change and are changing. But just as the technological prejudices continue to infect modern tools, social prejudices resist social change, thus giving us a messy, confusing march into the future, where we as an industry are forced to deal with the consequences of the past in the present.The Internet was done so well that most people think of it as a natural resource like the Pacific Ocean, rather than something that was man-made. When was the last time a technology with a scale like that was so error-free? The Web, in comparison, is a joke. The Web was done by amateurs.- Alan KayThe Alan Kay quote posted by Garry Taylor, and again quoted above, demands a direct response. Yes, the web was built by amateurs, as I expounded on above, but it is unfair to compare them to the foundations of the Internet and earlier computing developments and practices. Because it is the very reverse of the above mentioned prejudices, processes, and forces that is why software in general, including the Internet, is seemingly so much better than the web.The early days of computing and software happened under the shadow of war and were instilled with the grave importance demanded by such a genesis. The first commercial computers cost so much money, such as those by Eckert & Mauchly, that the only businesses that could afford them were those that were willing to spend an immense amount of money on raw calculation abilities. That meant banks, financial firms, and the government, and that meant a focus on reliability and accuracy. It also meant that those who were most likely to even be interested in computers were those who already understood how difficult the whole endeavor was, but also understood how important it was. Basically, the entire early industry was filled by genius visionaries backed by far-sighted organizations. It’s no surprise that the things they built have lasted.For a long-winded, but so totally completely worth it explanation of these socio-technical mores, watch the seminal documentary from 1990, The Machine That Changed The World. I have embedded episode 2, which discusses Eckert and Mauchly’s attempts to build the first commercial computer.By the 1970s, software was beginning to eat the world. The people who had laid down the foundation on which the industry was being built were still around, and the lessons learned were still salient, but business pressures and the drive for greater productivity were increasing. In the below video at the 1:47 mark, John R. Mashey talks about how there are simply too few people to write all of the software that is needed. He reiterates it in a sister video filmed at the same time.[2] These videos were recorded in 1981-1982.The date is important I think because of something else that happened in 1981: the first PC running Microsoft’s DOS was released by IBM. DOS, and Microsoft in general, would spend the next twenty years being lampooned by 1337 developers as a catastrophe that was systematically destroying programming. Many of the exact same criticisms were applied to the foundation and ecosystem that grew up around DOS and then Windows. It was lambasted as a confusing mess of files, protocols, and standards. Indeed, the now-famous Next computer and NextStep OS touted their superior developer experience in basically every presentation, constantly selling fast development of complex applications. Executives listening to these presentations of course translated “fast” to “cheap.” The more things change, the more they stay the same, and all that.The attacks on the web are no different than the attacks on other popular platforms in the past, because often the success of a platform has little to do with its quality. The specifics are different, but the ultimate consequences about which people complained are the same. Just as with today, the fundamental problem then was businesses that wanted cool stuff, but didn’t want to pay the amounts required to create truly high-quality cool stuff, and this had ripple effects across the industry. There were some intelligent advancements like those discussed in the Unix video above, but ignorant businesses would always go far beyond the intentions of the creators of those advancements, demanding both productivity increases and unrealistic cost reductions.This is mass market software development, forever by and by.With all of that circumlocution out of the way, I want to again turn to the key point I am trying too make. You’ll notice that I haven’t mentioned anything about HTML, CSS, or JavaScript specifically. Nothing about silent errors or whether AngularJS was actually a good product or not. Nothing about global CSS. Nothing about HTML being overly rigid. I have focused entirely on business motivations, personal motivations, and the history that affects us all. These are, in my mind, the reason for our current troubles.As such, I see it as a false savior to look to other technologies and languages to deliver us from evil. Sadly, it also means that the promised land is actually harder to reach. We can’t simply convince our architect or product owner to adopt a new framework or language. We need to convince them to conceive of our entire process differently, and that is a fight that is likely not winnable. Because even if you convince them that something is wrong, they will just look to another silver bullet, and the bullet du jour is of course Agile, which has its own raft of problems that cause people to hate their jobs.[3]We are humans. Our problems are human. And our solutions must also be human.Ain’t that a bitch.But much of that is by the way. This question was about web development, so let’s talk web development. Are these problems exacerbated by the nature of web development? Yes, I agree with that. JavaScript, which was itself a creation of ignorant business direction, allows bad developers to write code that is worse and more error-prone than it would be in other languages. HTML is literally a text formatting syntax. And CSS isn’t as bad as people make it out to be, but it’s a styling structure bolted onto a text formatting syntax which, even now, causes all sorts of problems in edge cases.Further, the demands on this tech triumvirate are greater than ever before. Applications in the past were driven primarily by functionality, but JS/HTML/CSS are driven by a desire to deploy applications everywhere, have the UI automatically adapt, have functionality adapt to manifold screens and platforms, and in many cases load instantly de novo. These demands are then being pressed onto technology never meant to fulfill these needs. It is no surprise at all that there is chafing.And perhaps that’s the point of the tweet: this chafing is an existential problem— that ignorant developers see the web as the only way to distribute an app as far and wide as possible. We should instead be creating a new platform that will cater to all of these new requirements.The problem that I have with this perspective is that the web is the best way to deliver an app far and wide. This isn’t a case of ignorant developers ignorantly using tools to grind out something that the ignorant business wants. In many cases, they are bringing to bear a great amount of old wisdom onto modern problems. I do not think that this is a problem that can be solved by “thinking better” about technology.Indeed, I think that some of the chafing is caused by wisdom. A good example can be seen in React and its Higher Order Components, or HOCs. An HOC is a component that takes another component as an input and returns a new component. This allows component pipelining. This has serious problems at scale and has been a subject of constant discussion, even motivating the release of a new feature in React to help deal with the problem.[4]But that pipelining idea has been around for decades. The above Bell Labs video includes a segment with Brian Kernighan using shell pipes to complete a task, taking advantage of what is known as stdin and stdout. It is a structure and conception of computation that comes from those who work in pure data and unsurprisingly works best with pure data. UI elements are technically pure data, with the UI being the output of a function in React, but to truly conceptualize a UI as pure data is immensely difficult. These pipelines are wisdom, but they are, I argue, essentially incompatible with a UI that is more industrial design than programming. I do not see an easy path ahead to reconcile the concerns of industrial designers and developers, which is why the conflict between design and development is basically a running gag in the software world.That said, I’m not telling others that they shouldn’t try to develop a new platform. Give it a shot! I am just not optimistic because the technological benefits will likely not outweigh the social and business considerations. Basically, HTML/CSS/JS is good enough. This epiphany is what Google had when they decided to stop waiting for a better language and dump money into making JavaScript as fast as possible with Chrome. They could have tried an entirely new platform push. They even made motions in that direction with the development of a Dart run time. But in the end, they stuck with what we had. Microsoft has done similarly with a truly stunning level of investment into JavaScript applications and the creation of TypeScript.Moreover, we already had platforms built specifically to distribute applications, all doomed, with only our current tools left standing. We had Java Applets. We had ActiveX. We had Flash, Flex, and Air. We had Silverlight and XBAP. We had NaCL. And soon we will have WebAssembly. As many on Quora have prognosticated, they do not expect WebAssembly to ever be a complete replacement for JavaScript, meaning that even arguably very superior languages will not be enough to upend the current paradigm.And at the risk of sounding like a broken record, the reason for this is because our biggest problems are not technological. No matter the technology we use, front-end development will suffer under the weight of the past. All of the manifold considerations and demands that I mentioned above will remain, and just as they do today, businesses will demand more from less. They will take a youthful developer population — I base this statement on the fact that speakers at JavaScript conferences appear to have an average age of 7[5] — and impose immense pressure to deliver features faster and more cheaply than other layers of the application, and the end result is not unexpected.Today, we have an almost comical level of developer tooling available: tools to analyze code, tools to write code, tools to delete code, tools to auto-refactor, tools to test code, tools to test tools. These tools and the technological edifices built up around them are often designed specifically for rapid feature delivery and nothing else. Lacking tools to achieve an end, developers are leveraging libraries like they’re going out of style. It doesn’t matter what you are hoping to achieve with your UI, there is a library that you can npm i and start running. Thousands of other developers are keen to make their name and will create NPM packages for public use, which further encourages the entire process. This is all distinct from the technology underlying the process.These same helpful engineers, appreciative of the pressures under which front-end teams are operating, are pushing JavaScript out into places which were unimaginable just a few years ago. React Native, Node, Electron, Native Script, Ionic, Cordova, and a million think pieces covering each one: it’s as though JavaScript is subsuming the developer world.This makes almost zero sense when viewed from a technological perspective. But when we look through the lens of organizational balance, it clicks. By that, I mean analyzing how to best achieve application consistency and feature delivery across platforms. All organizations need web sites and web applications, meaning that all organizations have JavaScript developers, meaning that they have the hiring pipeline, organizational wisdom, and social infrastructure to support them. It is easier to leverage this existing structure and draw upon an ever-larger developer population than it is to teach smaller populations required skills or translate their tech onto the front-end. We have many groups nonetheless attempting this, with GopherJS, PureScript, JSweet, et al, but their success has been small, and a tiny fraction of the success of Microsoft’s TypeScript, which tries to make the best sausage possible from bad meat.As time passes, this equation gets even more unbalanced in favor of using JavaScript everywhere. An organization hires web developers, uses new tools to push them into other areas, which causes the JS world to respond by providing greater tooling, which accelerates feature delivery, which increases demand, which creates more web developers, which eases hiring bottlenecks, and on and on. It’s no surprise that the industry supporting desperate developers is enormous and grows ever bigger, thus keeping the snowball rolling. As long as this remains mostly true, web development will be the platform you can’t kill.But just as I went to great lengths to explain above, this is a product of success and not peculiar to the web. Oracle, Microsoft, IBM, Google: all of them sell products that take something that was previously — and honestly should be — expensive and makes it cheaper. Talk to anyone who does full system architecture to hear horror stories about Oracle. Web development gets a great deal of attention simply as a result of the web’s success.This is the great tragedy of web development’s fame and fortune. It is so widespread that enabling its practitioners is incredibly valuable, which causes it to become more widespread, even into areas where it should not be. Most experienced developers know full well that native platform code is best. We’re sure as hell not using JavaScript everywhere because it’s a good idea. It’s only a good idea when we’re desperate to ship features with a limited team in a limited time frame. And most of us know that! In every shop I’ve worked, every developer has had their pet language in which they would prefer to be writing software. That language is never JavaScript. But we also know that mixing and matching JavaScript packages is a crazy-fast way to grind out an application and bolt on new features with a minimum of thought. Just as John Mashey in the Unix video above discusses, in hardware, no one expects a telephone to turn into a microwave, but in software, that’s exactly the sort of behavior we demand.And nowhere is that demand as bad as on the web, the mutant bastard child of a thousand different visions, pushing millions of different developers, to create products for billions of different people. In that regard, the fact that the web works as well as it does is just as great a miracle as anything else.Thanks for watching. I’m Aaron Martin-Colby. Good night.Footnotes[1] Gitstar Ranking - Top GitHub users and repositories[2] UNIX: Making Computers Easier To Use -- AT&T Archives film from 1982, Bell Laboratories[3] Why do some developers at strong companies like Google consider Agile development to be nonsense?[4] Introducing Hooks – React[5] Stack Overflow Developer Survey 2018

Absolutely! Others are actually creating great music as I write this. As Robert L. Gerrard explained, Garageband has vast capabilities as does any other modern DAW. Think of Garageband as a tool. Think of Pro-X, Fruity Studio, Pro Tools and every other DAW in existence as tools. These all have a different GUI, and a different workflow, but all of them have most of the same capabilities.Consider building a house… How much does it matter if your hammer has a wooden handle, a steel handle or a Fiberglas handle? They can all drive nails. The handle could be painted paisley. It doesn’t matter.What makes the difference between an amateurish production or a professional one? - You do! - Your knowledge of music. Your sense of taste. Your abilities to get a great mix.Regardless of what DAW you may have, to get professional results, the music must have a pleasing (often this will be subjective) structure… rhythm, melody, chord changes… transitions, emotional content. Any mix will need to employ well executed equalization and compression. Tasteful ambiance, either reverb or delay set just right. Often some volume or panning automation can add life and movement to your mix. You can certainly do all of this using Garageband.My advice: Keep well organized archives of all of your musical compositions. And backups! Maybe even backups of the backups. At some time in the future, you may want to revisit them. Maybe just for nostalgia, but also maybe as time goes on and your mixing abilities improve, you may actually want to go back and revamp the mix. Get a good sized USB drive or two for storage. Or burn your archives to data DVDs. Same goes for your software and effects. Back them up! That way even if your computer totally dies, your music won’t.More advice: Get some good studio monitors. And good headphones. If you can afford it, get your mixing space acoustically treated by a professional. Then get Sonarworks Reference to calibrate your monitors to your mixing space, and also optimize your headphones. You can’t mix what you can’t hear properly. If you don’t already have one, get an Audio Interface. Using a computer sound card, even a good one you’ve added, won’t sound nearly as good, because Audio Interfaces have pre-amps designed specifically to professionally record music. And most provide 48V Phantom voltage for condenser Mics. Get a few different kinds of really good microphones. Learn which Mic works best, in what situation.If you ever require the services of a professional mixing engineer, generally they’ll be using Pro-Tools. But no biggie. As long as you have rendered all of your tracks to WAV files, they can work with those. Provide untreated tracks and also stems, so they can get an idea of what direction you wanted to take.Assuming you’ll be mixing everything yourself, get knowledge! Google “Mixing tips and tricks”. “How to mix drums”. “How to mix vocals”. You get the idea… Whatever you’re doing at the moment, that you can’t get to sound just right, Google it. There’s lots of help out on the internet. On YouTube, I really like the videos from “Produce Like a Pro”, hosted by Warren Huart, a well known L.A. based mixing engineer. His advice and insight is priceless!