Streaming video, particularly over-the-top (OTT), is something. A thing. Arguably, the thing.

Various research houses, which have been watching the trend closely, report increasing penetration and consumer engagement for streaming video and OTT.

The Diffusion Group reported last week that OTT is gaining ground in consumers' hearts and minds. TDG said that among consumers who take both OTT and traditional pay TV services, if forced to choose between pay TV or OTT services such as Netflix (NASDAQ:NFLX), 57% of dual-service users would choose pay TV, and 43% would choose streaming services. The OTT preference is skewed toward younger viewers, as one would expect.

Similarly, Parks Associates recently released a couple of related research pieces suggesting that OTT and Internet-delivered video are becoming the new normal. One indicated that 59% of U.S. broadband households surveyed subscribe to Netflix, Amazon or Hulu, with other OTT services trailing far behind the big three. The other indicated that more than 50% of U.S. broadband households now watch Internet video on a TV screen - long the turf of traditional pay TV services - rather than just on PCs and mobile devices. Parks believes the data indicates a fundamental change in the video marketplace.

Traditional video service providers of course are aware of the trend and actively looking for ways to buck it. Most of the big players currently offer streaming video services of one sort or another, and smaller ops are adopting them as well. For example, C Spire recently launched an app-based streaming video service in its markets in Mississippi.

Perhaps more interestingly, AT&T (NYSE:T) is planning a multi-year transition of all of its video services to a streaming model, starting with DirecTV NOW in the next year and with other video services to follow in the years to come. The wholesale changeover - which promises to be a big, expensive job - suggests that AT&T sees streaming as the wave of the future. The company has also been virtualizing as much of its infrastructure as it can, and a streaming model makes lots of sense in that context.

For other service providers with large "sunk" investments in legacy gear, changing to all streaming and virtual functions makes less sense in the short term. Longer term, a phased transition based on end-of-life of deployed legacy gear seems a better approach, particularly if streaming helps them cut costs.

A bigger question, though, turns on the video marketplace as mentioned earlier. Though this quarter's numbers aren't all out yet, the trend for several years now has been one of declining pay TV subscriber numbers, escalating content costs, and low consumer satisfaction. At the same time, OTT providers' numbers have been growing by leaps and bounds. In that context, dumping a ton of money into new video technology may not appeal to smaller or cash-strapped operators, which in turn basically creates a vacuum into which OTT fits nicely.