IPTV is a traditional way of delivering content over a managed, fully-provisioned network. Though the protocol utilized in streaming the video content is Internet Protocol (hence “IP” in IPTV), this is not the public Internet. It is a private network, not accessible externally. The video streams are delivered within that private network, and accessible only from devices (set-top-boxes) issued by the operator. Examples of IPTV services are Verizon FiOS or AT&T Uverse.
In comparison, OTT (over-the-top) video services use the publicly accessible Internet to deliver video streams. Such content is not just available via set-top-boxes, but also via any devices that can access the Internet – such as phones, tablets and smart TVs with a broadband connection. Popular OTT services include Netflix and Hulu. Examples of OTT services launched by service providers are Dish Anywhere by Dish Network, and Now TV by BSkyB.
Since IPTV runs on a private, fully-controlled network, operators can enable technologies like multicast for network resource efficiency. However, most playback devices (like phones, tablets, and PCs) would not support such proprietary technologies. It may seem that IPTV has the advantage of tight control and guaranteed (over-provisioned) bandwidth, but it also has the disadvantage of its inability to serve viewers who wish to play content wherever and whenever they want. As a result, OTT-delivered videos are becoming more and more popular. Operators that have deployed IPTV are increasingly adding OTT capabilities. Emerging operators are even bypassing IPTV and just implementing OTT.
The detailed differences are given in tabular form on Wikipedia as well : Comparison between OTT and IPTV
The historical differences between IPTV and OTT. However, in the current evolution of video distribution systems, these differences are starting to disappear as IPTV and OTT merge.
To better understand this, it is interesting to consider the striking similarities with the evolution of VoIP (Voice over IP). Initially, VoIP, as pioneered by Skype and others, was purely “over the top”, ie without any network awareness of the type of traffic/content, which had cost and reach advantages (free calls to anyone with a computer and Internet connection) but resulted in lower quality of service and convenience. From a business perspective, the telcos saw it as an inferior product that couldn’t compete with their offerings and most consumers saw it as a complement to, rather than a replacement for, a telephone line. Let’s call this Phase 1.
Later, we saw the emergence of pure-play providers such as Vonage that made it look more like a traditional telephone service: you could use a traditional device (a phone) and got a lot of the same features but with some extra benefits (eg you could get an email transcript of your voicemail) and you paid for it as a regular phone service. Telcos started to notice and a few offered a similar service coexisting with traditional telephony; mainstream consumers who were looking for a mature solution that felt familiar started to switch. This was Phase 2.
Eventually, VoIP became subsumed into the network, with telcos adopting technologies, such as SIP trunking, that use IP instead of PSTN (Public Switched Telephone Network). Today, when you place a call, chances are at least part of the connection will be over a packet-switched network. On the cellular side, LTE enables all calls to take place over IP and some cellular providers even allow you to seamlessly offload mid-call to WiFi/fixed Internet connections. I call this Phase 3.
So what does this have to do with OTT? In the Internet video space, we’ve seen a similar transition, from the early days when streaming meant low quality and buffering on your computer to the current service that is actually superior to traditional TV transmission. I would say we’re currently at Phase 2, with cable/satellite companies like Comcast and Dish acknowledging and embracing OTT and with pure-play providers such as Netflix and Amazon appealing to a mass audience with the ability to watch on TV sets (the traditional playback device) as well as other devices but with some extra features such as superior content discovery.
My expectation is that over the next few years we will see a transition to phase 3, where the technology currently being used for OTT will be integrated into the network. We’re already seeing some early indications of this, such as Netflix’s Open Connect initiative and upstream video transmissions to the head end (distribution point) happening over IP, to name some developments.
Several years ago I predicted this evolution in a white-paper and coined the term TVoIP (TV over IP) as a slightly tongue-in-cheek alternative to IPTV and as a nod to VoIP.