Can somebody please explain to me why the WiFi Alliance is spewing marketing bullshit? - Rant 1 of n

I am probably using LinkedIn incorrectly (a conclusion that can be reached in many different way) but ultimately it has become a place where a lot of virtual discussions at length on matters relating to 5G are taking place. Both because the character limit is very low and because I have a propensity towards verbosity. A response to a reply to a reply in re-convention (or something like that) quickly exceeded the allowable limit of a single post. Ordinarily I'd split the post but in this instance I thought it would be easier to make into a blog-post by including the initial discussion:

I want to say the following to the 5G people who continue to claim that 5G will replace #wifi: This is complete nonsense. There is no evidence to support the claim. On the contrary, I believe 5G will be the mobile tech carrying the least amount of traffic - 90+% will be offloaded to #wifi.

The first problem is that the author is building a straw man. There is no serious claim being made that "5G" will replace wifi. On the contrary the IEEE working groups involved in the most recent iterations of wifi technologies present 802.11ax as meeting the MAC/PHY requirements for 5G indoor hotspots. There is a common claim that wifi technologies will fully converge within a 5G landscape, a claim that is uncontroversial except it seems to a marketing team that wishes to pitch 802.11ax as more than it can possibly be. Branding 802.11ax as "WiFi 6" is one part of an inherently dishonest marketing effort. There is related oddity in the term "mobile tech" - wifi has never been a "mobile tech" it is a WLAN tech and 5G is not an exclusively "mobile tech" either - the whole idea of FWA is that it is not mobile. A claim of "I believe that radio technologies defined by the 3GPP using licensed spectrum will carry less than 10% of traffic and that wifi will be the dominant radio access technology" is a very different claim to what is being made.

The second problem coming out from the entire WiFi 6 marketing exercise is that it appears to be selective as to whether to include the IEEE mmWave technology (802.11ad and ay) branded as WiGig- there is a conceptual disconnect between 802.11ax and 802.11ay. In light of the fact that the bulk of discussions around 5G NR surround mmWave deployments there is a lot of room for the most disingenuous of arguments and marketing spin that pivots the faux conflict between "5G" and "Wifi 6" into a fight about sub 6Ghz versus above 6Ghz.

The term WiFi 6 has emerged from the WiFi Alliance to refer to 802.ax which is not the variant of WiFi which actually runs up against 5G NR in the mmWave (more precisely 5G NR FR2). In terms of capacity 802.11ay is a lot more important. In discussions about WiFi based access versus "cellular" I explicitly include the significance of all of the WiFi family of technologies whereas Claus (of Wifi Now) seems to want WiFi 6 (a useless marketing term) to mean one thing when it suits him to escape the fact that we have a network ecosystem being designed around integration into a 5G core. Yet he wants it to mean another thing when that suits him. Thus the biggest problem with Claus insisting on saying " I believe 5G will be the mobile tech carrying the least amount of traffic - 90+% will be offloaded to #wifi" is that it represents a total lack of understanding of what 5G is. If traffic is "offloaded to" 802.11ac or 802.11ax or 802.11ay access media the offloading is either done as a form of peering though a router (so adding latency and representing a distinct WAN and WLAN separation) or the offloading is being done such radio access technology is part of the "5G enabled network".

A more meaningful and real discussion is possible with predictions of lining up as to how different technologies will contribute to an ecosystem.

I therefore stated:

Wifi takes a severe beating in conference settings with thousands of devices. Wifi works wonders for home automation connections of a couple of dozen sensors but will not meet the criteria for many organizations access control systems. At the moment there are countless instances where an organization has wifi APs deployed for certain devices but other mission critical devices are connecting to an MNO network. 

More importantly (as Mohamed Abdel Monem partly raises)  the 3GPP releases forming the basis of 5G do not only cover a radio access means and you need to look properly at what a 5G core entails. The most successful deployments of 802.11ax are going to be done by vendors deploying a "5G" core technologies.

Therefore what we can expect is access points are centred of 5G NR in the mmWave and access points that support various access measures and are built to form part of a RAN that is cloud native (so a 5G core)

Which I didn't think was particularly controversial but Dean Bubley (who I think is mostly right but is missing a very important part of the mark here) at this point disagrees.

At venues like conferences, many devices are Wi-Fi only, and the economics of device manufacturing means that tablets & laptops won't be embedding 4G or 5G modems by default anytime soon.

Secondly, a good proportion of visitors are travellers, or don't have flatrate data plans. There will be a need for free/amenity connectivity in a huge number of locations - conference centres, hotels, airports etc. At the moment, nobody has created venue-specific "free 4G or 5G" for visitors. It's technically possible (and easier with private venue-specific cellular), but the BSS/OSS isn't there & won't be easy

Thirdly, a combination of spectrum, core networks, complex RAN installation etc mean that private venues aren't often capable of cellular deployment, and MNOs aren't interested in paying.

Unless & until private 5G becomes as cheap & easy as WiFi to deploy (and has the 100's of thousands of certified engineers to do it) it's a niche play.

Come back in a decade & we might be closer

I think he is jumping over things and confusing what some operators are trying to sell the market on with what the technology is actually doing. The IEEE in their public relations material goes to pains to present ax and ay (and a bunch of other specifications in the family - bd ...) as part of the 5G landscape. A brief scan of documentation around 802.11 development makes it abundantly clear that the people developing - as opposed to marketing - WiFi related products are acutely aware of the fact that WiFi is an access technology (a RAT - radio access technology) that either integrates into an ISP world ethernet or will form part of service orientated network using a "cloud native" "5G core". 

In the process he commits the major fallacy that the operators want people to commit. He is assuming that 5G is designed as a salvation for the inefficient operator - it isn't. The development of a 5G ecosystem over the last 3 years has been an exercise in making different 5G functions and services cheaper and easier to ubiquitously deploy.

I have my doubts that 802.11ay is going to beat out mmWave NR (as developed by the 3GPP) and instead predict that the successor to 802.11ay (whatever the lettering ends up being) will likely be significantly better than mmWave NR. ETSI and the 3GPP will then have to decide whether to absorb it further as the principal radio technology or to continue developing whatever new new radio will be. 802.11ad has been quite a limited success and I don't believe ay will shift the landscape that much. If my assumption here is correct then (as I've predicted before) devices combining 802.11ax with 5G NR FR2 capabilities will be the backbone of high density indoor deployment.

The real question is what the rate of adoption of 5G core network principles and designs into the non-telco network.

My argument - and ALL equipment announcements made thus far reinforce this - is that the higher priced tier "WiFi" enterprise equipment is very much going to be built to service a 5G approach to networking. Therefore you will within 3 years be able to fully deploy a "5G network" at enterprise level that does not include a single MNO world radiohead (so no NR, LTE, LTE-A etc ...) and that enterprises are going to do this if the MNOs don't learn to play ball.

In the space which 802.11ax occupies (WLAN) it (802.11ax) is going to be huge and if Claus was simply saying that "anybody who thinks 5G NR is going to carry 1/50th of the traffic that wifi will carry in the long term is wrong" I'd be the first to agree with him and say that he is being conservative. The short to medium term it is a little bit more complicated because of the investments in older WiFi infrastructure and the "5G effect" and business cycles. But that isn't what Claus is saying, what Claus is saying is "WiFi 6" (a very limited little thing) will outpace an entire ecosystem.

When you are looking at LTE indoor deployments you are looking at the space where the idea of a "private LTE" deployments became possible within the "4G" era. The fact is that the migration from "3G" where you have telco centric vertically integrated deployment of a data network over which commodity internet run to "5G" involves a long term multiple parts evolution. The MNOs world vision is very strongly found in the "private LTE" deployment approach and there is this idea of flogging LTE-A and LTE-U against 802.11ac which is an interesting comparison in a similar vain to Display Port versus HDMI. Ultimately though to offer a private indoor LTE based alternative to WiFi you need to be spending insane amounts of capital for a limited number of additional benefits. Most importantly as a venue you could simply not deploy LTE and let the telcos put base stations near and in your venue (and possibly get some rental).

But thats not what happens when you look at venue neutral hosting infrastructure shift. 

The real fight in terms of market share is going to be between vendor specific non-5G Core enterprise offerings in the SD-WAN and "cloud management" space (think things like Juniper Mist).

So here is how I see the market playing out:

In terms of "indoor" (which includes stadiums) traffic volume I expect in the medium term it to be a 75% carry over on 802.11ax. 802.11ay with 5G NR (especially mmWave) and legacy other radio technologies carrying 25% of traffic (my strong suspicion is ay will carry a lot less than NR but it actually doesn't matter one way or the other because its going to be on a 5G core). Over time AV equipment and walkie talkies and the like will gel into the broader ecosystem - but it will take a while before we see plain old wireless microphones really shift.

With regard to mobile phones the case for supporting 5G NR is made by the business case of the MNOs in a NSA approach to building out their network. Mobile phones connecting to the indoor network will take up some of the traffic BUT the bigger pile of traffic will be IoT and automation applications where determinism and security (real or perceived) plays a bigger role in decision making where I expect NR to dominate in both traffic and spend.

Laptop manufacturers would be daft to support 5G NR rather than ax and ay and as a result its going to be a vanity feature on overpriced devices. - but never under-estimate the impact of the "Prosumer" - SCSI, Firewire and Thunderbolt are good testimony's.

Tablets will generally be without 5G NR support entirely although "phablets" if they make a comeback will follow the phone route.

Even if the 75% estimate is wrong for the medium term as long as there is a 10% segment for mmWave based NR in a conference centre environment we will see AP deployments will be built to support 5G both as a core and as part of access points - I anticipate a situation where you will have 1 access point with NR capabilities for every 2 cheaper access points only supporting 802.11ax.

Diving into the IoT space I anticipate a lot more mmWave NR devices than 802.11ay or 802.11ax with previous wifi connectivity dominating (802.11 a/b/g/n probably the most common). Further there are a dozen 802.11 standards in development for low powered and the like - - all of which need to connect into an edge centric cloud native architecture, and all of which at this juncture assume that core network architecture to be a "5G Core" or at the very least to be pretty advanced. Again looking at the 802.11 development work being shunted through the IEEE there is a very clear recognition of the 802.11 family as being a collection of RATs for a 5G ecosystem (a debate between radio engineers over respective RATs is a different one all together) so it might well happen that successors to 802.11af and ah gain a lot of traction.

Back to mobile phones generally. I expect 802.11ax and WiFi technologies in sub 6Ghz to be fully supported by mobile phones with support for mmWave WiFi technologies (initially ad and ay) being less common and more flagship and niche (a "multimedia" or "gamer" phone with WiGig support - offhand I don't know if any mobile phone on the market support WiGig right now). This will in part be from pressure from some of the MNOs wanting to keep some form of lunch. One of the results is that while you will continue for a while to have this space in which wifi will be ideal for in venue use for broadband connectivity but stepping into things like AR and VR starts to become a little more complex.

I've posted in another exchange that I expect the 5G NR in conference venue space to be a much thinner market than the MNOs are pitching it to be but that because of the audience which it captures I anticipate it to have a sufficiently high level of success as to influence general adoption. Think of the relative success of EPOC PDAs, not a massive market but enough to tip over ARM development (and look at ARM's ubiquitous nature today). Basically the sort of person who in the early to mid 90s would have had a Psion netbook (which didn't have wifi built in) or Palm Pilot and who in the early 2000's (before teenagers came to roost) had a Blackberry or a Nokia Communicator - tomorrow is going to have a communications device set that is built around 5G NR FR2 (and possibly FR1 as well). If half a percent of of the laptop market is ultrabooks with 5G NR capability that market segment is also a market segment who are in conference venues a lot.

More importantly the difficulties which conference venues - and conferences - are having with Wifi demanding users takes a beating is not (as you might be able to tell from reading this rant from the beginning to now) because 802.11n and ac are bad as access technologies but rather because of the complexity which density introduces and this complexity requires a smarter network. Some companies have pushed out vendor specific cloud management platforms but most companies who are interested in an ecosystem have either on the side (hoping to get some MNO business) or as integral to their strategy embraced the umbrella of 5G (either by pointing out the IMT-2020 / 5G doesn't belong magically to the MNO world, or by embracing the 3GPP and its processes). Sadly though there seems to be a marketing faction out there that would like to differentiate 802.11ax from the ecosystem, slap a name onto it and then encourage the development of management systems to clean up the mess that such an approach creates. Quite simply the fight is about brand recognition, people who've hitched their wagon onto the "Wifi" brand (remember that the WiFi Alliance has trademark property around WiFi as a brand) have something to loose by "brand 5G" subsuming WiFi technologies. At the same time we have MNOs who have a lot to loose by "brand 5G" being seen as less about MNOs and more about how converged, Internet communications advances saying stupid shit. The fact is that when people say self-serving stupid shit they need to be called out on it.

I don't expect MNOs to find their investments in mmWave small cells paying off, and many MNOs will do the calculations as to whether to deploy and discover that with their business models it won't work and so they don't deploy in the first place. I think most MNOs are going to move into the market in the most idiotic way possible. There ARE business models which existing MNOs can adopt which will be beneficial to them ( Cellc - or more importantly Blue Label Telecoms, who are exposed to CellC - for example could if they take the initiative actually disrupt the market to their benefit). Key to such a business model would be to render accessing the network in a venue over any RAT by an authorized device in a "best effort" configuration to not result in additional charges. If I am at an airport my connectivity to "wifi" whether using 802.11ax or LTE-U or 5G NR FR2 should be free unless I am connecting to QoS guaranteed services or the like.

(This ties to the central thesis of my presentation at AfricaCom last year: https://telecoms.com/493620/best-effort-should-be-considered-as-critical-for-the-telco-of-tomorrow/ (quoting Jamie):

To validate the best effort approach, telcos have to appreciate there are two ways to deliver connectivity in the future. There is the commoditized delivery of the internet, and value added services with enhanced connectivity. One has a dumb pipe, while the other has a smart one which can be glorified with the network slicing euphoria. This is where ‘best effort’ should be applied, and is built on the demand for latency.

I don't expect it to happen because well MNOs behave stupidly and then blame the government or the consumer for their failure - they know how to reach me though if they are interested). Ultimately however 5G is not good for monopolisitic minded MNOs. I say this all the time - as do many analysts and the like in the 5G space.

Equally I think venue managers who invest as a neutral host provider extensively in mmWave NR simply to play in the hype train are going to get burnt. But looking at how many venue management organizations make decisions it is not unreasonable to conclude that a lot of money is going to be put down on the need to appear more innovative and ready than other venues and the drive will definitely be there. 

The most rational strategy for venues to adopt will be to focus their efforts on 802.11ax ("Wifi 6") access points in accordance with the recommendations for the IEEE which means a "5G Core" and to not lock themselves into a vendor cloud management solution that doesn't support scoping into the 5G ecosystem. But all of this comes back to a very REAL problem which the marketing folks over at the WiFi Alliance have decided to become hellbent on creating. Excising one technology from the 802.11 family (802.11ax) giving it a bullshit fancy name and then pitching it as superior to an entire ecosystems of technologies is the sort of dishonest stupid thing we expect from MNOs.