Internet Television
Catherine Colman
Deep Dive Assignment

Producing and Delivering Content for Mobile Phones

A revolution has been occurring in the telecommunications industry that has the potential, not only to change the ways in which media is captured and shared between groups, but also to redefine the ways in which content is distributed and broadcasted to large audiences. This revolution has arrived in the form of increasingly sophisticated mobile phones attached to infrastructure that has the capability, not simply to transmit voice, text and photographs, but high quality video content as well. According to statistics from April, 2006 there were 2 billion mobile phone subscribers worldwide, 30 million of them with access to the more sophisticated 3G networks (mobile networks that have broadband capabilities able to support greater numbers of voice and data customers and which have higher data rates at lower incremental cost than 2G networks). Of these two billion subscribers worldwide, 850 million have camera phones , a number that, along with global subscription, is expected to rise significantly, projected by Lyra Research to reach 1.5 billion by 2010 . But the number already exceeds those with access to digital cameras worldwide and exceeds the 180 million who own personal computers. This pervasive phenomenon is not exclusively linked to more developed nations either. A rush to develop 3G infrastructures is considerably prevalent in developing nations in Africa, for instance, a continent which already has 83 million mobile phone users. Aside from the infrastructure, mobile phones themselves are only becoming more and more advanced. For instance, Nokia recently released two new additions to their Nseries line, the N93 and the N95, both of which, in addition to their high quality lens and extreme megapixel camera, have DVD quality video capture and playback.

But while the phones themselves become increasingly more complex and the potential of the infrastructure becomes considerably more sophisticated the actual implementation and successful use of these new technologies to broadcast various kinds of media in interesting ways has lagged significantly behind the expectation, particularly in the US. Unfortunately, there are many issues that have slowed the rapid evolution of this evolving form of content delivery. The first is bandwidth issues. The 3G, or third generation network, which promises broader bandwidth and faster deployment of rich media content, a key factor for most music and video applications, is still not significantly available in the United States. In order to deploy the 3G network, the entire existing mobile infrastructure has to be updated, as 3G networks are not upgrades of existing 2G (second generation) networks and do not operate on the same frequency spectrum. Rather, they are entirely new networks that need to be built with new frequencies assigned to mobile operators, all without interrupting service. Many major US service providers, including Sprint, Verizon and Cingular offer 3G capable phones, yet their networks do not yet widely support the enhanced technology. Another issue, more about the politics of content delivery structure then technical concerns, is who controls the gateway? Today, wireless operators are exerting the most control over media deployed to their handsets. But as more traditional broadcast companies and third party developers begin to execute mobile strategies, the concept of who is actually in control suddenly becomes an interesting question.

Another huge issue is pricing. Many different companies are employing varied strategies on how to charge for mobile content, some based on a given price per download, and some based on streaming or on-demand technology that charges the viewer flat package rates or individual rates for different kinds of content. Still other third party companies are either offering popular or user-generated content for free or developing solutions that use content viewers already own and then making it accessible to a variety of different mobile devices. It is so far still questionable as to whether users would prefer live TV-type broadcasts from networks they trust or downloadable media snippets; if they will mind embedded advertising or want commercial-free content given the price they may be paying for content delivery. There has also been a huge issue surrounding perceived consumer ambivalence. As aforementioned, the United States has long lagged behind other countries in mobile data usage, partially due to the lack of 3G technology and partially due to a backwards approach in creating user-friendly content delivery applications. Text messaging has been the biggest success, followed by downloadable ringtones, but many consumer surveys indicated for quite a while that there was little interest among US consumers in watching TV on a mobile phone.

The largest problem, however, is the multiple alternative technology platforms and standards fighting to become the preferred choice of mobile phone makers as well as the industry standard. Mobile infrastructure today is an alphabet soup of technologies and acronyms, which means that content may have to be customized for each wireless network, and sometimes for individual handsets. This makes it incredibly difficult to create media experiences that can be viewed by a large, cross-platform audience. In order to understand the technical specifications informing the encoding of video content to be delivered to mobile phones, it is necessary and quite relevant to understand the differences between different generations of mobile networks, which carriers are using which specific networks and the different standards that these networks use. As aforementioned, the term G refers to the different generations of mobile networks and devices. First generation (1G) mobile phones were analog devices. Second generation (2G) devices are digital and the networks they use are GSM, iDEN, CDMA, and TDMA. GSM stands for global system for mobile communications and is used in the U.S. by the Cingular and T-mobile service providers. iDEN stands for integrated digital enhanced network. It was developed as a proprietary 2G network by Motorola and is used by the Sprint/Nextel service provider and their various rather well known MVNOs like Helio, Espn and Boost Mobile. iDEN supports both three or six interconnected users (phone users) per channel, and either six or twelve dispatch users (push-to-talk users) per channel. Unfortunately, because Motorola and Nextel kept iDEN proprietary, it prevented it from becoming a serious competitor to GSM or CDMA. CDMA stands for code division multiple access and is also used by Sprint and by Verizon. TDMA stands for time division multiple access and is a common multiple access technique which is used by multiple protocols, including GSM, as well as in IS-54 and IS-136, although, due to the widespread popularity of GSM and CDMA, it is now being phased out.

2.5G generation networks include GPRS, HSCSD, WiDEN and EDGE. GPRS stands for General Packet Radio Service and is available to users on the GSM network. It provides moderate speed data transfer by using unused TDMA channels in the GSM network. HSCSD stands for High-Speed Circuit-Switched Data (HSCSD), and is an enhancement to Circuit Switched Data, the original data transmission mechanism of the GSM system. The user is typically charged for HSCSD at a rate higher than a normal phone call (e.g., by the number of time slots allocated) for the total period of time that the user has a connection active. This makes HSCSD relatively expensive in many GSM networks and is one of the reasons that packet-switched GPRS, which typically has lower pricing (based on amount of data transferred rather than the duration of the connection), has become more common than HSCSD. WiDEN stands for Wideband Integrated Dispatch Enhanced Network. WiDEN was originally anticipated to be a major stepping stone for provider Nextel and its affiliate, Nextel Partners. However, with the December 2004 announcement of the Sprint Nextel merger, it has been speculated that the Nextel iDEN network will be quickly abandoned in favor of Sprint's CDMA network. Although a complete roadmap of the merger's impact on the combined company's wireless networks has not been released, Nextel and Motorola have agreed to continue to maintain and expand the WiDEN network through, at least, December, 2007. EDGE stands for Enhanced Data rates for GSM Evolution, it is an enhancement to 2G and 2.5G GPRS networks and works within GSM networks. EDGE (also known as EGPRS) can function on any network with GPRS deployed on it, provided the carrier implements the necessary upgrades, which has yet to happen in most of the US to date. High-speed data applications such as video services and other multimedia benefit from EGPRS' increased data capacity.

3G networks include CDMA2000 and UTMS. CDMA2000 is actually a family of third-generation mobile telecommunications standards, including CDMA2000 1x, CDMA2000 1xEV-DO, and CDMA2000 1xEV-DV. They are approved radio interfaces for the International Telecommunications Union’s IMT-2000 standard and a direct successor to 2G CDMA. CDMA2000 is also a registered trademark of the Telecommunications Industry Association in the United States and standardized by the 3rd Generation Partnership Project 2 (3GPP2), a collaboration agreement that was established in December 1998 between ARIB/TTC (Japan), CCSA (China), TIA (North America) and TTA (South Korea). UTMS stands for Universal Mobile Telecommunications System. It uses W-CDMA, (Wideband Code Division Multiple Access) as the underlying standard, is standardized by the 3GPP, and is the European answer to the ITU IMT-2000 requirements. To differentiate UMTS from competing network technologies, UMTS is sometimes marketed as 3GSM, emphasizing the combination of the 3G nature of the technology and the GSM standard which it was designed to succeed.

Typically, content is sent to a user’s handset in two ways, either through video streaming or through progressive download to be viewed in a stand-alone video player available on the handset. The main difference between these two is that in video streaming, the compressed video data is fetched from a server during playback, whereas when downloaded, the video clip to be viewed is stored locally in the memory of the mobile terminal or memory card, a problem for some less sophisticated phones which lack large capabilities for memory storage. The external session establishment of a video streaming service is implemented using the phone’s web browser or messaging applications (SMS or MMS viewers), which have the ability to launch external applications such as a player application residing on the user’s phone. When content is made available for streaming the viewer’s phone is directed to RTSP url. RTSP, or Real Time Streaming Protocol was developed by the IETF (Internet Engineering Task Force) and published in 1998. It allows a client to remotely control a streaming media server, issuing VCR-like commands such as "play" and "pause", and allowing time-based access to files on a server. One way that a video is made available for download is through the download of a .RAM file, which will launch the phone’s installed video player. A .ram file is a simple text file that contains the URL location of a RealMedia clip or SMIL file on a remote server.

Video for delivery to a mobile phone has to be compressed in a different manner then video being made available for web viewing, not only because of a mobile phone’s smaller screen size, but because of bandwidth constraints. The typical codecs used for mobile video are H.264 and MPEG-4 visual. H.264, also known as AVC/MPEG-4 Part 10, is a newer codec that phased out H.263. It contains a number of new features that allow it to compress video much more effectively than older standards and to provide more flexibility for application to a wide variety of network environments. H.264 uses advanced compression technology to provide better video quality from smaller amounts of video data, saving bandwidth and storage. It delivers the same quality as MPEG-2 at a third to half the data rate and up to four times the frame size of MPEG-4 at the same data rate. H.264 is supported by many of the 2G and 2.5G networks, including GPRS, EGPRS and W-CDMA. MPEG-4 was introduced in late 1998. It is the designation for a group of audio and video coding standards and related technology agreed upon by the Moving Picture Experts Group (MPEG). When generating mobile video content, the video must be output with one of these compression algorithms in a 3GPP compatible standard, most often with a file extension of .3GP. Generally the accepted maximum size ratio of content that will play on a wide variety of handsets is 176x144, at 10-15 frames per second and with a data rate of 50-60 Kbps.

Authoring mobile video requires the use of 3GPP compatible authoring tools, of which there are several widely used versions. Apple’s Quicktime Pro supports the H.264, MPEG4 and H.263 codecs and various input file formats, including AVI, MOV, DV, MP3, MPEG-4, WAV and MPEG-2 and outputs 3GPP and 3GPP2 compatible files. Quicktime Pro also allows for the creation of files for streaming in unicast or multicast transport interfacing with the Quicktime streaming server. Real Networks also has two video encoders, Real Producer and Helix Mobile Producer. Helix Mobile Producer supports streaming output files using the real media streaming server that are 3GPP and 3GPP2 compatible as well, and accepts the same range of file formats as Quicktime Pro. Real Producer from Real Networks outputs .ram which is intended for download, not streaming. There are also many free or open source authoring tools that output 3GP video and some of the mobile phone manufactures even have their own tools. Final Cut Pro, video editing software available for MAC users, can also output .3gp video.

However, it is not only the compression and size of a video that is of concern when creating for a mobile phone, but the actual look of the video as well. Because most mobile phones still have viewing screens that are incredibly small and since most users will potentially be viewing content while in motion, there are several generally accepted rules to follow to make sure that the content you create will have the best quality possible. Creators are generally advised to avoid using source video sequences with blank or black first frames (possibly resulting from a video edit), high scene-change frequency (one per 1- 2 seconds or fractions of a second) and fast horizontal, vertical or diagonal panning and fast zooming, as it seems to cause unnecessary motion irritation in users who may already be in motion, or unable to hold their handsets still in the way that a PC or TV remains stabilized. Creators are also often warned to avoid re-encoding from already encoded content and to always use the source audio and video sequences. This is due to fact that the large amount of compression needed to optimize for the phone already diminishes the quality of the original and performing this kind of compression on an already encoded source will only decrease the quality further.

Once a mobile phone compatible file has been created a content provider will then have to utilize a specific technology to broadcast it to the viewer’s handset. Currently, the technology in this space is also a battle of acronyms, with DVB-H, T-DMB, and MediaFLO plus a few others dipping their feet into these unexplored waters. DVB-H, or digital video broadcast-handset, works by simultaneously sending a signal to multiple users with compatible receivers on their phones. Users find programs via an on-screen guide, select the program to watch and view it when it is broadcast by the provider. DVB-H can offer a downstream channel at high data rates which can be used standalone or as an enhancement of mobile telecom networks, which many mobile phones are able to access anyway. Time slicing technology is employed to reduce power consumption. IP datagrams are transmitted as data bursts in small time slots. Each burst may contain up to 2 Mbits of data. The front end of the receiver switches on only for the time interval when the data burst of a selected service is on air. Within this short period of time a high data rate is received which can be stored in a buffer. This buffer can either store the downloaded applications or play live streams. DVB-H can operate on a variety of network standards, including GPRS and UMTS. Currently, DVB-H is the broadcast technology that has been the most widely used and tested. In Finland, the license to operate a DVB-H network was awarded to Digita in March 2006. In May 2006 they announced that they had signed a contract with Nokia to use its DVB-H platform for the service. The network will cover almost 30% of the country by the end of 2006, with the service launching in most of the major markets. Initial trials in Helsinki, showed that 41% of all users testing the DVB-H technology would pay for the service, while in the US several companies are already having some success transmitting TV signals to mobile phones through the Sprint, Verizon and Cingular networks. In addition, Modeo, a company owned by Crown Castle Mobile Media, is in the process of launching broadcast service in a select number of US cities this year that will be independent of any specific provider and supported by several handsets made by Nokia, Motorola and Samsung.

The main competitor of DVB-H is Qualcomm’s MediaFLO technology, which claims to deliver content more cost-effectively, using multicasting technology to broadcast to many users over the bandwidth required for a single ultra high frequency TV channel. The "F-L-O" in MediaFLO stands for Forward Link Only, meaning that the data transmission path is one-way, from the tower to the device. The MediaFLO system transmits data on a frequency separate from the frequencies used by current cellular networks. In the United States, the MediaFLO system uses the frequency spectrum at approximately 700Mhz, which was previously allocated to ultra high frequency TV Channel 55. On December 1st, 2005 Verizon Wireless and Qualcomm announced partnership for the launch of the MediaFLO network. Verizon Wireless expects to deploy MediaFLO in half of their 1xEV-DO markets in the United States by the end of 2006. Another potential technology is the DMB (Digital Media Broadcasting), which has had some success in Asia. Also known as DMB-T, Digital Media Broadcasting - Terrestrial, it is the youngest major broadcast standard and provides the best reception quality for the power required. It evolved from the European Digital Audio Broadcasting (DAB) standard, which was widely supported by Korean mobile chip makers. DMB-T uses MPEG-4 Part 10 (H.264) for the video and MPEG-4 Part 3 BSAC or HE-AAC V2 for the audio. The audio and video is encapsulated in MPEG-2. Like DAB, DMB-T is made for transmissions on radio frequency bands, band III (VHF) and L (UHF). Because the United States has so far failed to allocate these two bands as most of the rest of the world has, DMB is still unavailable here.

In addition to a battle over industry broadcast standards, there is also an ensuing battle over who will ultimately aggregate and provide the content., involving discussions over what kinds of content viewers want. Do they only want extentions of traditional popular content or would user-generated content be just as successful? Currently, many of the larger US carriers have partnered up with traditional media companies to provide various different packages of media available to users for monthly fees. Verizon, for instance, calls their service VCast. It allows those with compatible phones and the VCast plan to select from a variety of news, sports, weather and entertainment clips produced by several traditional TV networks. Cingular provides a similar service, partnering with HBO to provide their content via the HBO Mobile and HBO Family Mobile packages available for an additional monthly fee. Sprint, on the other hand, has focused much of their attention on sports. They currently have an NFL package that is free to a subscriber with any of their data packages and allows the viewer to watch game highlights and receive customizable alerts. Many traditional media companies have also tried to broadcast their content on specific branded handsets with limited success. ESPN mobile, for instance, was a project initiated by ESPN for sports fans using the Sprint network. However, the ESPN phone was a costly branded accessory that limited a user’s access to other Sprint web features. It was recently announced in a press release that ESPN will, as of December 31, 2006, ‘cease providing wireless service’ and that they are ‘working to deliver our content experience through another nationwide carrier’. Disney also released a branded mobile phone on the Sprint network that, in addition to providing child friendly content, will use the GPS capability of handsets on the Sprint network to help parents locate their kids.

There are also several independent third party providers of content that have aggregated popular content or released applications for mobile handsets which allow users across different networks to utilize a similar service. MobiTV, for instance, is a pay service that provides content from MSNBC, ABC News Now, CNN, Fox News, Fox Sports, ESPN 3GTV, NBC Mobile, CNBC, The Discovery Channel, TLC and The Weather Channel across the Sprint and Cingular networks. Basically, a user with a compatible handset would sign-up for the service directly through their Cingular or Sprint phone without having the complication of having to download or install a separate application. Another third party provider, Shozu, which has had a free J2ME application for posting user-generated pictures and videos to various online aggregating platforms including Flickr, Textamerica and YouTube for quite some time and supports many mobile handsets, recently announced version 2.0 of its application, which will incorporate a new functionally called Zucasts. Zucasts are particularly interesting because they will allow users to subscribe to various feeds of content from several less traditional content providers including the incredibly successful video blog Rocketboom, Buzznet, an online community centered around bands and music, Webshots, an online photo-sharing platform and various material from media partner Warner Brothers. Popular online video mixing community, Eyespot, which supports mobile posting to its site, also recently announced development of an application to work with the Qualcomm BREW mobile operating system, used by Verizon handsets, which would also allow users to view various kinds of content, including user generated mixes on the site itself.

Also providing a ‘free’, meaning the only cost associated is the carrier charges, traditional TV viewing experience coupled with a user-generated component is FreeBe TV, a 2.5G GPRS multi-channel mobile TV service that, despite its recent launch, has already enjoyed well over 100,000 hits to date on its website. FreeBe TV is available in the US and the UK and has 10 channels at present, including Monkey News Network, real live news read by monkeys. It has also just launched it's newest channel called You-Made-It, which is set to be a mobile version of YouTube where users will submit content and the best submissions will be broadcast. To sign up for FreeBe TV, viewers go to the website where they receive a WAP url and a pin access code. Unfortunately, FreeBe TV is only currently supported by 24 handsets, although they say they are working on supporting over 100 handsets by the end of the year. Another new service, Melodeo, Inc., also employs content aggregation from major media sources such as ABC, CBS, HBO, MSNBC and CNN, but it is set up and marketed as a social networking video experience. Melodeo allows users to attach their service to their already existing mobile account, supporting a wide variety of phones from many of the popular carriers, including Sprint, Cingular, Verizon and T-Mobile. Once users have attached the service to their account, they can subscribe to various available content or upload their own content in podcast format via the site’s web interface. Once particular content is subscribed to, Melodeo allows users to share that content within their defined social network of friends and contacts. The site also allows users to search for content within predefined categories and ranks different podcasts based on the most views, the most plays or the most shared. On Oct 2nd, 2006, Melodeo announced a partnership with Cingular.

Another interesting provider of mobile access to media content is Orb. Orb employs its own brand of technology that allows a subscribed user to remotely and securely access all their home-based media resources, from TV channels, to video clips, music and photos or any live or recorded media on their PC from their mobile phone. Orb also supports viewing from live webcam feeds, which can allow subscribers to monitor their homes, their children or child-care providers any where that their phone can access the web. All users have to do is log-in to their account from a web browser. The most interesting thing about this service is that, when connecting remotely to a user’s home-based media, Orb automatically detects connection speed and the upstream bandwidth available from the device attempting to connect, and based on these, determines the best delivery format, codec and bitrate to use to stream the content. This means that content does not have to be formatted for the phone itself or for streaming capability. Orb also allows a user to add friends, co-workers or family members to their Orb account, allowing each person to customize their Orb preferences without impacting others in the group. This a particularly nice feature because it provides an effective means for multiple users to access a centralized database of content and simultaneously allow these users to identify their own particular preferences. Unfortunately, the Orb interface that allows a user to aggregate content is only available for windows, although the content can be viewed by any phone with a Web browser and either Windows Media Player, TCPMP, RealPlayer, or 3GP Player. Orb is currently partnered with Vodafone and Nokia.

Many companies are also creating platforms and architecture that emphasize easy solutions to generating revenue from broadcasting mobile media content, without the user or company having to worry about the various different kinds of technology implemented. MediaSphere, for instance, is a solution that allows its users to upload content, optimize for mobile viewing using the site’s optimization tools, enforce copyright protection by encrypting files using DRM, and control content usage, pricing, bundling, number of plays, and subscription rules directly from the site’s web interface. MediaSphere also allows users to utilize their technology surrounding billing solutions, delivery strategies and customer tracking. While MediaSphere is geared more toward companies seeking mobile broadcast solutions, Momo, created by Activefone, seems to be more friendly to less corporate minded individuals. Unlike many of the other platforms, Momo emphasizes the use of video and media actually taken with mobile phones and delivered to users connected within the site’s online social networks. Momo allows individual users to create communities on Momo’s web interface which they control. Once a community is created, other users join the community and can upload video they have taken with their mobile phones. This content can then be syndicated and made live by the community creator. Once the content is live it generates revenue for the community creator and the mobile service provider every time a user downloads a file. Bliptv.com (not to be confused with blip.tv) is another website that allows users to upload short video clips under 20mb and in either .wmv, .avi, mpeg or .mov format. If a user’s video is chosen for broadcast, the user will be paid an unspecified amount for their contribution. Bliptv is a proprietary broadcast solution that only transmits to the Sprint/Nextel network.

Overall, it seems that broadcast solutions which offer a wide range of content, including that which is user-generated, may be the wisest way to go in the future. Content that is proprietary to certain carriers, handsets or media companies may be interesting initially, but has already proven not to be as successful as many had originally predicted it would be. The value of delivering content to a mobile phone is, in my opinion, very similar to delivering content online, in that it should rely more on social networks, sharing within these networks and user-generated or niche created content rather then on traditional broadcast media models. Applications such as Momo, which utilize social networks created online, emphasize user-generated content and provide a means for generating revenue could be the more interesting and effective way to deliver content. But the question then becomes, will third party applications of this nature become the most successful, and will similar models be adopted by the various carriers or will they continue to favor delivering content from traditional media sources? None of this speculation is necessarily attempting to infer that traditional broadcast companies should not be in the business of creating mobile content. It is exciting to finally see that some of them are recognizing the mobile platform as a viable and necessary options for content delivery. Perhaps what is really needed then is a combination of sources, such as what Melodeo does, which provides traditional content from trusted media sources, alongside user-generated podcasts in a social environment. The only thing that is certain is that the development of mobile media solutions and the mobile phone infrastructure itself is still in the early stages. Before content can be widely accessible there needs to be more widely accepted standards, both within the mobile infrastructure and within the various broadcast delivery methods. 3GPP and 3GPP2 only went so far to create a standardized arena. Much more work is needed.