Thread by @chiwawa_42

France is a bit late in #5G deployment due to an ongoing "debate". Or is it so ? Let's update you on the situation. A #thread : 1/43 cc @EU_Commission @Europarl_EN @DSMeu @ThierryBreton @FTTHCouncilEU @5GPPP @EU_Growth @EP_Industry @BERECeuropaeu @Arcep @sorianotech

We'll focus on the mainland. Oversea territories fall under different rules and actors. 550 000km2, 65M inhabitants, 4 mobile carriers, over 20 MVNO, about 20 fixed consumer ISPs with the 4 MNO dominating the market. 2/43

While #5G rollout hasn't started yet, we're pretty far ahead in terms of fixed fiber deployment thanks to a 12 year old regulation allowing any licensed operator to rent underground ducts and cable poles from the incumbent. 3/43

Though I may come back to the FttH topic later on, because as it goes, many foreseeable problems arise and public authorities, while informed by a few engineers years ago, are not taking actions yet. 4/43

Back on the ongoing #5G debate : we have two sides. On the pro-side, we have… next to no one. Neither carriers, vendors or regulator are willing to take a public beating by an angry mob. 5/43

On the Con-side, we have nearly every politicized attention-seeking speaker on the floor. Green-washers, opposition parties, self-entitled experts on climate, health issues… But no specialists. Guess why ? 6/43

I know I'll take some hits and bruises with this thread, but I want to clarify a few things nevertheless. It's all about facts and science, not opinions from now on. Let's review the arguments. 7/43

1) "5G is going to use new, higher frequencies, and no public study has demonstrated its absence of health impact". That's partially true, but mostly false. 8/43

5G will essentially run on existing frequencies used by previous generations, by re-farming the spectrum and a new technique allowing multiple generations to run simultaneously as to not force early terminal's replacement. 9/43

"New" frequencies can also be used as to cover dense areas more efficiently. Well, those are hardly new : 3,5GHz was used for years as fixed wireless access (WiMax) and low millimetre waves is common for back-haul. 10/43

As to the health impacts, numerous studies have been conducted and found none. It doesn't mean there is none, it's just, as Science works, that we cannot detect them because if they exist, … 11/43

… they may be of such low significance, probability or amount, that even the best instruments are not accurate enough to find it. 12/43

So, should we "wait and see" as the "caution principle" mandates ? Well, that's where it comes to opinions, so I'll let you make your own. 13/43

2) "5G will require more antennas to be deployed". That's a big nope. By re-farming existing frequencies, coverage will be strictly identical with the same antennas. But, when needed, we could - more easily - add more to get better coverage… 14/43

…with a lot less power. Because more antennas mean less radio power. You don't have to push harder for signal to reach farther terminals, so the overall energy consumption and radio power is reduced by a denser coverage. 15/43

3) Associated with the previous is "5G will draw more electricity" : well, it might, if we use it as it's fullest potential, but in the short term it consumes less because it has better tricks to reduce its footprint. 16/43

For instance, #5G allows to reduce the number of MiMo (parallel) radio chains, thus reducing available bandwidth, when there's less traffic on the air. That's a 15-20% energy savings, not shutting down the network. But there's more ! 17/43

#5G runs on newer equipments which have more efficient chips (see Moore's law) than the previous, and some of them are already deployed in advance, currently running 4G network stacks, so we don't even have to replace all of them. 18/43

At last, the modulation is more efficient - when background noise allows it - than previous generations, allowing for a 15% increase in bandwidth for the same energy consumption. But it allows for gigabit level speeds where 4G allows for less… 19/43

4) "With more pipes to fill, we'll consume every last bit of it, so in the end it'll be a waste". Historically that's true : empty pipes attracts creativity and waste. But… 20/43

…with the ongoing simultaneous sanitary, economic and environmental crisis I *think* (sorry for the opinion) we might *at last* work on websites and app's efficiency and sobriety, so in the end, we'll do more with less. 21/43

This doesn't account for innovative uses of this network, things we can't just grasp yet, because that's the whole point of innovation. 22/43

5) "But people will always find bad ways to waste resources" that's an opinion, not a fact, as I said, because times are changing fast and we might as well be more cautious starting this year. But let's sum it up anyway… 23/43

Speculating on 8k mobile porn is nonsensical, your eyes ain't accurate enough to count pubic hairs at that resolution, so there's no point in going over full-HD. 24/43

Remote surgery doesn't need 5G, and must not rely on a shared-media network anyway. Fixed fiber already does the job. Though there might be better ways to federate health networks for that, such as @IXPSantX . 25/43

Smart Cities is just another bad excuse to feed consultants with public funds, I doubt it will ever take that much traction, and if it does, that should be for savings of other resources and might worth it. 26/43

Internet of Things may be of some use, and we already have networks for that, like some designed with low-power as its first rule such as @Sigfox 's. 5G might help, though, for some other uses currently unfit for it. 27/43

There might be other speculations, but I think you got the point : #5G gives more with less, or more with more if we need it, but that's up to the users, not the network's purpose. 28/43

6) "It's a revolution, we'll be forced to replace every smartphone !" That's also a negative. 4G will still run for years, you'd get the newest iPhone long before that. 29/43

Or break your feature phone, unless it's a famously unbreakable @Nokia 3310, but 2G is planed to be kept operational just for it - for the time being, so there's no "forced replacement" of any kind. 30/43

I'm joking, 2G will be kept running for many other purposes, such as existing industrial and safety equipments, doorbells, eCall, but you got the point. 31/43

7) "We might as well give the house' keys to the Chinese Government"! Digital sovereignty deserves a thread by itself, but let's wrap it up quickly if you will. 32/43

We don't need Huawei's or ZTE's to deploy #5G, we have two perfectly fine vendors of our own, with @Ericsson (Sweden) and @Nokia (Finnish, German and French know-how combined). 33/43

I think that covers it all, but if I missed any point, please tackle me so I'll add it to the list. Now we know why we will deploy #5 because it has only advantages, let's see how we could do it even better. We're back to opinions. 34/43

There's a thing about the European Union : every member state' government systematically tries to steer a ruling's interpretation as to best fit its own agenda. On the telco market, this as led to opening the market… 35/43

…generally in a really inefficient way, by privatizing historical networks and forcing new actors to build redundant networks of their own. Well, nearly every state f****d it up, only the Brits did it right… 36/43

…but only on the fixed network access, by splitting the incumbent in two companies, one for the "network" and the other for "services". Mobile there also runs off multiple redundant networks. We'll miss them… Or not. 37/43

This has led to the creation of stable oligopolies to replace a monopoly, often with the incumbent leading the pack and have really attentive ears from the regulator. Barely what I would call an open market. 38/43

Deploying a common, public or private, cellular network wasn't feasible with early technologies. Only 4G started to make it viable. 5G goes even further. But a cellular network is a complex beast, 39/43

and sharing it amongst multiple service providers might weaken it if one was to do nasty things. So for the sake of redundancy, as nationwide coverage is essential, two networks might be the optimal approach. 40/43

Going from 3 or 4 networks to two won't cut the overall energy consumption in half, because another advantage of 5G is that it has less fixed (20-35%) and more variable (load-following, 65-80%) energy footprint. 41/43

But it would reduce the fixed-part of that footprint, and either reduce the amount of capital to invest in that natural upgrade, or build a denser coverage, that choice is mostly up to regulators. 42/43

It would also allow for fair access across the EU of both existing and aspiring MNOs and MVNOs to revive competition as the EU rulings mandates. #5G allows for that to work, and even more, let's jump on it ! 43/43 /#Thread

Thanks for reading this thread @EU_Commission @Europarl_EN @DSMeu @ThierryBreton @FTTHCouncilEU @5GPPP @EU_Growth @EP_Industry @arcep @@BERECeuropaeu @sorianotech and everyone who will share it !

cc @Collabblues

Y'a ptet un truc à dire sur l'épisode 1992-2000 de régulation telco en EU, #WithFlags.