Digital music
Our site was originally built around the (back then) newly developed audio compression technology called MP3. I personally had my first encounter with this magical little acronym back in Autumn, 1996, when I was (ab-)using our school's Net access and surfing the wonderful (but quite small, compared to current one) World Wide Web. Found lists of anonymous FTP sites' addresses and went through them to see what I might find (for those of you who got to the Net much later on, I'd like to note that yes, there were online piracy years and years before the appearance of P2P networks, Pirate Bay and others -- actually, to be correct, there was online piracy years before the WWW too). Found sites that contained latest singles from popular artists, using a never-heard-of file extension, .mp3.
It was quite mind-blowing to download a good quality audio file that was well below 3 megabytes in size (as the school's "super fast" connection was probably a shared ISDN line, it probably took only an hour to download one track ;-). I fell in love. There was only one MP3 player software available back then -- WinPlay3 -- a software that allowed to play only one track at a time, didn't support playlists nor didn't have any other "useless glitter". Soon, the audio format's popularity grew and playlist managers and alternative GUIs for WinPlay3 began to appear. Shortly after, also alternative players began to emerge -- by Spring 1997, it was quite clear that one of them was well above others, one called WinAMP.
To somehow make it clear how difficult the MP3 format and digital audio formats generally were back in 1996, I must detail the process of ripping a CD disc: first of all, virtually none of the CD-ROM drives (yep, we didn't have CDR burners back then) supported something called DAC (i.e. Digital Audio Copy). Lack of this support meant that only way to "rip" the CD was to play it with your PC, output it with your audio card, loop the signal back to your audio card's input (using a 3.5mm headphone cable), then record that audio using Windows Audio Recorder. Of course, my Pentium 166Mhz didn't have enough oomph to compress the audio on fly, so the audio had to be recorded in "raw" WAV format, taking 700MB per CD (and back then, an average HDD was a 500MB drive). After you had recorded the CD manually (you had to cut the recording for each track too, if you wanted to have separate tracks), you had to compress (==encode) it into MP3 format. And of course, there was only one tool available for that task too, called l3enc. And yep, it was a command line program. It took about an hour to compress one 5min track into 128kbps MP3 file, using a good quality settings. Only in 1997, first GUIs for l3enc appeared that allowed auto-spawning new encoding processes after each track -- without that option back in 1996, you had to manually encode each track or to create a batch file to do the job. So, to summarize it all -- it was a pain, but it was fun.
Despite the small bumps in the road, the MP3 format's popularity grew. To emphasize this growth, the first hardware-based portable players began to emerge back in 1998 -- I managed to persuade my employer to buy me one, a Diamond Multimedia's Rio 300. It had the whopping capacity of 32 megabytes, it didn't show any artist information (of course, back then, ID3 tags -- the tidbits that contain the artist, album and track information for each MP3 -- weren't as common as they are nowadays) and it lacked quite decent pile of features that modern MP3 players do have. But it played the darn tracks we had so laborously ripped, pretty much continuously, for two years. That was the first time the Big Music reacted -- and they did it as they always do, they sued the Diamond Multimedia and tried to ban the player from the U.S. markets. Thankfully, Big Music lost that case -- a ruling that paved the way to the current situation where pretty much every imaginable device there is, supports MP3 format by default.
Year 1998 also marked a year when I first found Web sites that were solely focused on MP3 as a technology and as a phenomenom. To name two, Dimension Music and Cyberphunk come to mind. These two sites also played a very important role for AfterDawn, as their take on MP3 affected my thinking and crafted the plans for our site.
Our site's focus, as I mentioned, was originally very firmly on digital music and specially on MP3 technology. With our news, we covered bands that released their singles in MP3 format, reported on new CD ripping tools, allowed people to download WinAMP and other MP3 players from our site, etc. But things change in technology very, very rapidly. In this case, technology itself didn't really evolve -- MP3 is probably one of the oldest multimedia formats that hasn't really evolved at all since 1999 or so. Its not to say that it is "perfect" or that there aren't better, competing formats -- it is to say that it was -- and still is -- "good enough" for most people. And as it has become universally accepted format (again, can you name a modern product that produces noise that can't play MP3 files?), it is most likely to remain the king of the audio formats for years to come, even that lossless audio compression formats are gaining in popularity, NET connections are hundred times faster than a decade ago and HDD space is virtually free.
Furthermore, with the development of better MP3 software tools, the whole process of ripping a CD into a bunch of MP3 files became so easy, the software players all became so good, that the whole subject of audio compression and MP3 in particular lost its shine sometime in 2000 or so. And only to be replaced -- in consumers' minds and on our site -- with...
Digital video
As you might have gathered, I use the term "digital" to refer to non-physical-media formats here, so the term itself is slightly misleading, as audio CD was digital already back in 1980s and the digital video arrived to the mainstream in form of a DVD disc in mid-1990s. But this is about evolution of those formats in digital form, so lets focus on that.
Terminology aside, the "early adopters" were already enjoying their DVDs in 1998 or so -- after moving to the format from the much-praised semi-digital format, LaserDisc. My first encounters with digital video (in form that I associate the term to) were in early 1998, in form of a VIVO video clips and VideoCDs. Sure, VideoCDs -- or VCDs as they were commonly referred to -- were already extremely popular in certain Asian countries and in Chinatowns across the world. But the combination of technologies really started the wave of digital video to our PCs. First of all, PCs finally became fast enough to be able to play decent-quality video material, thanks to the evolution of CPUs and graphic cards back in late 1990s. Secondly, the awe of consumer digital video, in form of DVDs, brought the source of good quality material widely available. To merge these two, affordable DVD-ROM drives began to quickly replace CDR and CD-ROM drives in new PCs by the time of a new millennium.
Add the rapid growth of HDD space, availability of proper residential broadband connections and you can understand how and why the digital video phenomenom emerged. It might sound silly to say, but I strongly believe that most technological innovations of the recent decades have been partially sparked by piracy -- and digital audio and video are prime examples of this development, a development that has eventually created also huge, legit business opportunities too (imagine the combines sales figures of all portable MP3 players or DivX-capable DVD players or just blank CDR and DVDR media).
Anyway, back to the development of things. Unlike with audio, the first video encoding methods and formats weren't anywhere near "good enough" for most people and therefor, the development has continued from those early days and is still ongoing. In late 1990s and in very early of this millennium, the "king of the formats" was undoubtedly VideoCD. With its 320x240 (or 320x256 if you lived in PAL country) resolution, its quality was somewhat comparable to decent VHS quality, but was sub-par quality when compared to SVHS, LaserDisc or DVD. But one movie could be burned to two blank, cheap CDR discs and as a nice bonus, all DVD players were obliged by specs to play them back properly.
Sometime in 2001 or 2002, we lived a short period of time when VCD was replaced from its throne by its little brother, SuperVideoCD (or SVCD for short). Similar to VCD, it was also a format that standard DVD players were obliged to play (however, the support for SVCD wasn't as universal among DVD player manufacturers as it was with VCD support), but it had much better resolution (480x480 / 480x576) and it supported better video encoding format (VCDs were encoded using MPEG-1, but SVCD used MPEG-2, the same format DVDs use and most DTV broadcasts in the world use, even today). The problem with SVCDs was that they took quite a lot of space (most long-ish movies required 3 CDs) and the fact that a better contender was already making its way to the "throne".
Sometime in 2002 a format called DivX ;-) -- yes, the smiley is part of its original name -- made finally its breakthrough and became the most popular video format. The original format wasn't actually a format at all, but a mere hack of Microsoft's MPEG-4 implementation. Microsoft in its wisdom had crippled its MPEG-4 in a way that it required Microsoft's own WMA audio format to be used as the audio track. As the WMA was subpar choice for most people, DivX ;-) hack emerged that allowed Microsoft's MPEG-4 video to be combined with MP3 audio. Year or so after the DivX gained ground, the guy behind the original hack, added with financial backing and other developers, created a "proper" MPEG-4 based video format that they began to call as DivX (without the smiley face).
DivX's advantages over VCD and SVCD were obvious -- space and video quality. One average-length movie could be fitted quite nicely into one CDR disc while preserving the quality of the original DVD video in a way that it was quite hard (for us with non-20/20 eyesight) to tell the difference between the original and the re-encoded copy.
A failed co-operation between DivX, Inc. and an open source developer community back in early days of DivX's corporate life, a rift between the two emerged. That rift eventually turned into an open source alternative of DivX's MPEG-4 implementation, called XviD. For most people the difference between the two has never been very clear, as they both (specially in their later forms) followed the MPEG-4 specs quite strictly and therefor (in most cases) a video encoded in either formats, was playable in players supporting one or another.
This DivX/XviD harmony existed for years, only to be changed by yet another technological advance in the field of digital video: high definition. The arrival of high definition video meant challenges to the existing formats, due the fact that -- to put it simply -- a five times better resolution also resulted into five times bigger filesize. First commercial HD broadcasts made across the world were using MPEG-2 -- a format dating back from 1980s -- and requiring massive amounts of bandwidth and storage space. Yet another format was needed, and industry soon adopted MPEG-4's h.264 layer as its de facto new standard for video encoding. And of course, video enthusiasts and pirates followed.
The digitalization of TV broadcasts meant also that there was no longer need to "rip" and encode the TV signal to another format to be used online -- instead, simply capturing the broadcasted h.264 (aka AVC) video stream would be enough. Added with the fact that quickly after its appearance, Blu-ray adopted h264 as its de facto encoding standard, the success of AVC is easily understood. But the sheer amount of existing "standard definition" material means that for most parts, people are nowadays using two main video encoding methods -- DivX/XviD for SD material and AVC for HD material.
AVC/h.264 has actually made its way to much wider scale that any of its predecessors -- almost all online video services use AVC as their standard video encoding format (YouTube and Hulu to name two), virtually all modern digital video cameras have shifted from tape-based DV recording to AVC-based recording, etc. But as we've seen in the past decade alone, it is quite foolish to say that AVC will be "the Format" for years to come. More likely, something better, maybe something open sourced, even something "near lossless" will appear in coming years and change the landscape again.
One sidenote that must be mentioned here is the evolution of the container. For Joe Average, he doesn't know what format the video is using, whether it is DivX, XviD, WMV or something else. For him, the video is either "AVI" or "MKV". And that is the big change that has happened since 2006 or so. Before that, a PC-readable video for Joe Average was always an "AVI", but now it can be also an "MKV". Video files are actually bundles of "streams" -- there's video stream, containing the video itself, encoded in some format (like DivX). Then there's audio stream, encoded in MP3 or MP2 or AC3 or something else. And then there can be additional stuff, like alternative audio streams, subtitle streams, etc. All of these are packed into one file -- and that file is called as "container". AVI, recognized by .avi extension, was the de facto standard for PC video for well over one decade. It lacked features and had some major problems in its original design, but it was universally supported. Only the HD material's arrival pushed the requirements of video beyond of what AVI could elegantly handle. That paved way for Matroska (recognizable by a file extension .mkv), a container format developed in 2004, but truly adopted as late as 2007 or 2008 by most people.
Both digital audio and digital video and their evolution, requirements and people's preferences link pretty tightly to the true invention of the past decade...
P2P
As mentioned earlier, at the dawn of the WWW and online mass media, FTP was the most common way to distribute digital audio files -- as well as other relatively large legal and illegal files -- due its efficiency and technology. Back in mid-1990s, the best way for anyone to "randomly" find illegal files to download was to obtain lists of anonymous FTP sites' addresses and go through them, one by one, and see what can be found from each site. Obviously, the best "beef" could be found from private FTP servers, that didn't allow anonymous access (just like today), but finding them and getting access to them was next to impossible for most people, even back then, when the online world was much, much smaller than what it is today. FTP was accompanied by IRC and DCC transfers -- back then, the best way to transfer large files between two individuals -- a predecessor to "trusted P2P" if you like.
First attempts to make it easier to find MP3 files were built shortly after the whole MP3 phenomenon was in its infancy. Best known and most widely used solutions were specialized online search engines that went through hundreds of known anonymous FTP sites and handful of Web sites that allowed downloading pirated MP3s. Such solutions put enormous pressure to the servers, as thousands of people tried to access to servers that could realistically handle less than 100 concurrent users at the time. They also put such sites to the limelight, drawing unwanted attention from media industry -- first online piracy busts cases I personally remember involved busting some of such FTP/Web sites back in late 1990s.
However, a small application and a novel thinking changed all that. In June, 1999, a student called Shawn Fanning, released a software called Napster, that didn't rely on distributing the files from a single server, but instead, turned each user of the software to a new server. Such approach spread the cost of bandwidth across all users of the software. But more notably, it allowed people for the very first time to search for individual songs and files from a network that was bound to be online and available all the time, no matter what level of usage it received, as each new user also added a new "server" to the system.
Obviously, the arrival of Napster -- in the same month as our site went online ten years ago -- finally brought the term "MP3" to the mainstream. The hobby of mine and my friends for several years by then was quickly brought to the main evening news briadcasts and to the newspaper headlines across the globe. Imagine having launched a site covering that exact subject at the very same time the whole phenomenom began to reach mainstream media.
Of course, such attention brought even more people to the world of peer-to-peer (a term coined shortly after Napster made headlines, later abbreviated to P2P) and expanded the scale of music available from Napster. Shortly after, it was safe to say that Napster had the largest digital audio library available in the world, easily surpassing even the largest music stores. The demand for "free" music also pushed the dial-up connections to the extintion very, very quickly and began to push the speeds of broadband connections further and further, as people finally had the "killer app" in their hands that actually made use of faster Net connections. Napster's use peaked in February, 2001 at 26.4M worldwide users.
Big Media quickly drew its attention to Napster and big part of 1999-2001 period on our site was devoted to the coverage of Napster's court case and its proceedings. Eventually, as suspected, the company had to file for a bankruptcy, as the court case drained the company financially.
However, despite Napster falling, the concept of P2P was popularized and as always with technology, improved by each new attempt to create the "ultimate" P2P application. Even during the Napster's reign, there were several other P2P networks available, but none of them grew "big enough" to get the kind of mainstream attention each network truly requires to get the availability of files to the level where "pretty much everything" can be found easily. The fall of Napster paved the way to new "Big" P2P solutions to become mainstream. The one that emerged as the winner of this slightly chaotic era, was FastTrack-based Kazaa, a software that caused quite a lot of controversy due its use of integrated ad elements, some spyware-ish elements in its client and the company's shady ownership structure.
As expected, it didn't take very long until Kazaa was sued by the Big Media, too. The court fight was quite interesting, due Kazaa's weird ownership structure, distributed across the world. But as with most such cases, U.S. courts eventually blocked the service. During the last months of the Kazaa's reign, already several P2P applications were fighting for their limelight -- and this resulted to a short period of time in P2P history, when world was divided into several large-ish P2P protocols/networks. In some countries, Direct Connect and its open source successor, DC++, became the largest P2P apps -- and in other countries, eDonkey and its open source counterpart eMule, became popular. Furthermore, several Gnutella-based apps cherished. As these P2P protocols weren't centrally owned or managed, shutting them down by shutting down one single operator was impossible. But they had several shortcomings in technical terms and otherwise -- all this made room for yet another P2P technology to rise -- BitTorrent. BitTorrent's approach was completely different to those P2P apps that had held the throne before it -- instead of bundling all "logic" required to running a P2P network into a one application, BitTorrent separated the tracking of peers, sharing files and finding files into separate portions, thus distributing the P2P model itself even further from the original Napster-style "one app and one organization to rule it all" approach.
BitTorrent's rise paved way to commercially run search services that allowed indexing of P2P files -- s
uch services that we nowadays know by names of Mininova, The Pirate Bay and others. It also somewhat undoed part of the P2P ideology first made available by Napster, as most people wouldn't anymore share "all I have", but instead, preferred to download/share one file at the time, making it significantly harder to find "alive" hard-to-find files on P2P networks. On other hand, the ever-increasing popularity of P2P has obviously made up some of this, but arguably not all.
BitTorrent has also made impossible possible and has gained some significant support also outside the pirate circles -- several Linux distributions, game demos, etc are nowadays distributed using BitTorrent, allowing authors to lower their bandwidth costs by utilizing the best of P2P, a distributed mechanism of transferring large files efficiently.
After ten years, it is obvious that P2P is a technology that is here to stay -- despite the fact that its name has been labeled by piracy -- as it is the true "killer app" that originally launched the enormous demand for faster and faster broadband connections, thus paving also way to other bandwidth sensitive solutions, such as online vide, video-on-demand and "rich" Internet applications.
AfterDawn.com
I began my work career as a web developer back in 1998 in a small Finnish web development company, building websites for variety of companies. Having learned the basics of web development, I quite quickly became frustrated by the tasks assigned to me ("build a feedback form for us", "lets build a product catalog for users to browse", etc) and wanted to create something on my own terms. That desire added with my growing interest towards digital audio, highlighted earlier in this article, created the groundwork for the site that was eventually to become AfterDawn.com.
During the late 1990s, running a professional website was actually quite costly task -- even the cheapest adequate webhosts costed more than $300 per month. Having just began my career and being in my early 20s made it quite clear that I couldn't take the financial risk needed, even when we're talking about couple of hundred bucks per month. Furthermore, I was -- and still am -- very aware of my shortcomings in web development certain areas -- and I also understood that the site of a scale that I imagined could not be run by one person alone. Thus, I contacted group of my friends back in early 1999 and told them about my idea for a website and wanted to know whether they'd be willing to participate to the project, financially and with hard work. I managed to grab a team of seven people around me to back the idea and the groundwork for the site was made.
The name of the site was chosen in a "IRC meeting" set up with the founding team in March, 1999. I asked all participants to come up with 5 domain names they thought might be good names for the site -- first checking that such domains were available (with .com obviously). After several rounds of voting, the name AfterDawn was chosen. It was suggested by our design guy, Teemu, and even today I'm extremely happy that we chose that particular name among all the suggested names, as the name itself doesn't really "mean" anything and therefor, it has allowed us to expand the site's scope easily without conflicting with the name (imagine a site with "MP3" in its name, focusing on Blu-ray or mobile phones).
First years of the site were laborious -- trying to find the focus of the site and trying to drive traffic to the site. Things looked quite good in late 2000, when traffic was slowly increasing and so was the income of the site (however, we were still losing money -- meaning that we earned less money from the advertisements than what we spent on hosting bills). And of course, those years were also hard on personal life -- first going to your "real life job" for 8-10 hours each day, coming home, turning your computer on and spending next hours on developing AfterDawn, writing articles to the site, etc.
Then the online advertisement market collapsed in 2001 when the "dot com bubble" burst, leaving us in a situation where our traffic was increasing, hosting costs increasing and advertisement income going down the drain. This miserable situation continued for almost two years -- while we gathered more and more visitors to our site, site made absolutely no money whatsoever, making it an expensive and very laborious hobby for all of us. Thankfully, sometime in late 2002, things started to turn for the better -- and quickly. We finally began to earn more money with the site that what it cost to run it -- still not paying a single penny for any of us in salaries. This turn of tide made it logical for us to turn the site into a proper company in May, 2003 -- with all but one of the original founders participating as site's owners. In true democratic fashion, we decided to split the shares of the company based on the work done by each individual during the first, hard, four years -- a situation that is in place even today.
Turning the site into a company didn't mean much at first, but as things improved even further since 2003, in 2004 -- five years after launching the site -- we were able to hire our first employee, Dela, who joined our editorial team in August, 2004. Furthermore, in early 2005, I was finally able to quit my day job as a web developer and began running AfterDawn as its CEO. Same applied also to Ketola, our CTO, and Teemu, our design guy. In 2005, our first office was opened -- making it the first time ever when the site wasn't run anymore from our bedrooms and controlled solely over ICQ and IRC discussions. Since that day, we've put every spare dime that has come to our coffins, to the development of our site and its content. And nowadays, our little hobby employes either full-time or part-time, appx. 15 individuals.
During the ten years, our site has grown to somewhat of a big-ish IT/Tech site. Our English site reaches almost five million unique visitors during its peak months (typically January) and our Finnish version of AfterDawn ranks as the country's largest IT site (reaching appx. 10% of our country's population each month). We have almost 900,000 registered members on our forums, with more than 4 million forum posts. And our software section has delivered appx 120,000,000 software downloads to the day. All this highlight also the approach we've had from the day one -- unlike most sites, we don't focus only on one content area (such as news, or software downloads), but have tried to make the site a one-stop shop for all your tech needs, offering wide selection of software downloads, daily updated news, large discussion forums and wide range of articles and how-tos.
Anyway, this is my personal summary of the past ten years -- some (or many) of the dates mentioned here might be inaccurate, timelines of events might conflict, I have surely missed several important aspects of development of things, etc. I do apologize for those errors, but also hope that you understand that all this text is really just "my memoirs of the past ten years", not a historical account of things :-)
With all this blabbering, I wanted to shed some light on the past decade, highlight things that have been important for us and our site. Also wanted to show what time and level of motivation it can take to build a site like ours. And of course, to highlight the efforts made by the excellent team of ours, people who trusted my instincts and wanted to be part of building it into a reality. And of course, to thank all of those who have visited our site during the past ten years, supported us, told their friends about our site and sent us feedback and ideas on how to improve our site even further.
Thank you all for the past ten years.
-Petteri Pyyny, CEO
AfterDawn.com
Written by: Petteri Pyyny @ 10 Jun 2009 9:10
