That's makes a lot of sense and truly, fuck telcos.

Also, mechanisms like L4S for high-fidelity ECN feedback allow for higher link utilizations without end-to-end QoS impact.

IPv6 reducing routing table fragmentation should enable reasonable prefix based peering, along with billing senders according to the used peering. We don't need fast flat rates, we need cheap bulk traffic and flat rates for effectively-not-oversubscribed bandwidth. Maybe with a way to roll over unused priority bandwidth at some multiple into the next billing period to count against that period's bulk traffic usage.

Also, no one asked ATT to lose billions of dollars by buying Time Warner HBO and DirecTV. Or Verizon with Yahoo.

But they wanted to chase that Comcast vertical monopoly dream. And it will never yield any utility for customers.

* Self driving cars? No.

* Internet of Things? No.

* Metaverse? No.

* Remote surgical operations? You gotta be kidding.

Everyone in a big stadium can watch the show on their phones, though. If someone was willing to pay for the hundreds of tiny cell sites required.

We are already doing that, it's called WiFi. 5G SA is nothing but a souped up WiFi signal, and it should be treated as such and not how we have been treating 4G up until this point. Telcos need to change their business model otherwise the tech giants might end up eating their lunch

> Verizon is settling with the FCC over its use of an ad targeting technology known as a "supercookie," which tracks the websites visited by phones on its network. Supercookies allow websites to better target ads to visitors with Verizon cellphone service; but those visitors — for a period of time — weren't informed of the tracking or given the option to opt out. Because of that, Verizon will pay a fine of $1.35 million and will now have to receive customer permission before sharing tracking data with other companies or even within its organization, including with sites owned by AOL.

- The Verge

Underlying FTC press release

Full Title: FCC Settles Verizon "Supercookie" Probe, Requires Consumer Opt-In for Third Parties

https://www.fcc.gov/document/fcc-settles-verizon-supercookie...

> WASHINGTON, March 7, 2016 – The Federal Communications Commission today announced a settlement resolving an investigation into Verizon Wireless’s practice of inserting unique identifier headers or so-called “supercookies” into its customers’ mobile Internet traffic without their knowledge or consent. These unique, undeletable identifiers – referred to as UIDH – are inserted into web traffic and used to identify customers in order to deliver targeted ads from Verizon and other third parties. As a result of the investigation and settlement, Verizon Wireless is notifying consumers about its targeted advertising programs, will obtain customers’ opt-in consent before sharing UIDH with third parties, and will obtain customers’ opt-in or opt-out consent before sharing UIDH internally within the Verizon corporate family.

Source: PII data request of my own Verizon account, containing detailed demographic data, intent/interest data, and IP addresses of websites I had visited.

Edit: Anyone interested in doing the same, here's the link: https://www.verizon.com/support/download-and-view-vpd-file/

Cellular systems are designed for many more devices and, crucially, full mobility. These are orders of magnitude more complex than WiFi.

Companies like Rakuten have already shown it's cheap enough to distribute 4G femtocells for free [1] to your users while increasing your coverage

[1] - https://network.mobile.rakuten.co.jp/guide/rakuten-casa/conn...

Let's not conflate too many things. Maybe current deployments are in that frequency band, but "5G SA" just means the connection is legitimately 5G.

In theory all 4G frequencies will get converted to 5G, and it will be a moderate improvement. Separately 5G is expanding into the wifi arena and the ultra fast near-line-of-sight arena. But "SA" is not how you tell the difference. It will all be SA eventually.

On LTE this is called LAA. As far as I can tell, the term is being kept for 5G as well.

F1 used to have a deal with "Kangaroo TV" [1], where people at the race could rent a small portable TV. When I went to the Shanghai GP years back (it was the one where Hamilton beached his car coming into the pits), I rented one. With racetracks that are miles long so nobody can see all the action - and certainly not when sitting at Turn 1 in Shanghai - it's definitely a boon. I was able to see what was happening, even listen to the commentators. I was telling everybody around me what was really going on.

The thing is, it was fairly expensive. I don't remember the actual number, but something that you probably wouldn't just do on a whim.

And history tells us that it didn't work out. The "kangaroo.tv" domain name is apparently available for purchase now.

Too bad, because my experience is that, without this, going to an F1 race is inferior to watching on TV in most respects.

[1] https://www.autosport.com/f1/news/fom-joins-forces-with-kang...

There's value there, but not for consumers. It's large land monitoring where this gets value. Essentially competing with LoRaWan by building out a ubiquitous radio network any sensor can easily connect to.

5G over the same cmWave frequencies 4G uses still has a benefit - but mainly for the operator! It allows them more users and throughput for the same amount of spectrum. So in a rational market, the operators should be giving you discounts for using 5G instead of charging more for it!

So I guess some people have it, and use it? It’s basically always on around my neighborhood, but I guess I live in a nice city and Sprint’s old HQ is down the street so that probably has a lot to do with it.

You basically need line of sight for mmWave. Think of it like light. If you can’t see an access point, then you probably aren’t using mmWave.

~3.4GHz-4GHz (which is the 'primary' 5G band) in Europe can provide excellent speeds (2gbit/sec real world no problem to a phone) but without the terrible signal degrigation you get on mmWave.

mmWave will be used more in the future once the other bands get saturated, but it's a bit ahead of its time now.

Of course, some of are of the position of "You try to place remote access, control and delegation to critical pieces of my vital infrastructure, expect the most violent level of personal reactions".

I mean. We’ve all heard the marketing bs from the telcos that 5g opens up these possibilities. No engineer, worth the title, would be naive enough to actually believe that and put tire slip control into an rpc call...going over wireless. For real, does anyone actually believe this is a good idea?

I work for John Deere (of course do not speak for them), we want to analyze things 'on the cloud', but we are thinking about how to apply the results of data we gather latter. Already it is so the coop sprayer that comes in 2 months knows where the rows really are so it can figure out the most efficient path to drive their field and drive over the least crop. There are also what do I do in this field next year decisions that can be driven by data, and it is nice to not have to take some form of memory device to the tractor.

Companies with trucks are interested in things they can get from connected vehicles. Insurance also is interested to see if they can find the good drivers and give them better rates (governments probably want to force training on bad drivers or something). None of the above is real time critical.

Impossible (unfortunately) for reasons of privacy.

> No engineer ... naive

Your mistake is to imply that the "drive" lies with engineers - instead of a perverse economy in which some try to promote unneeded novelties to "have something new to sell - to people who already have everything". This was directly admitted by the involved.

> a good idea

Given the above, some believe it a potentially good idea "for the stockholders" (not for "function", as you seem to assume as a perspective).

However, I pay about 25 eur for my 5G telco (unlimited data), and 35 eur for my fiber connection (100mbit/100mbit). My telco allows dual SIM usage, so I am thinking about just cancelling my fiber connection and installing a 5G modem. The thing that is holding me back is that a good 5G modem is somewhere between 500 and 4000 eur, so it is quite an expensive experiment to do.

I was in Korea for work and it seemed the same there, I was even in eastern Europe visiting family and it was a pretty similar story. So I figured it was pretty universal. Although I know that Australia isn't in a good space with broadband simply because of politics and companies not wanting to work together, I'd assumed they were an outlier

    we have fibre to about 90% of the population

Real Question: On paper, NBN is an insane human development achievement. Why do so many non-urban Ozzies on HN complain about lousy Internet access? Honestly, I don't believe it. Send them to Germany for a year, and they will see how bad it can be. (HN is full of funny stories from Germans about how bad is their broadband.)

For example, my inlaws live in semi-rural South Carolina and a new cell tower was built about a mile away from their home ~2 years ago. T-Mobile's Home Internet product indicates there's no service at their address for even though I have what looks to be full signal strength indoors on their "5G UC" network (indicates mid-band & mmWave from what I can tell, but in this location it's likely just mid-band). Every speed test I've executed from my T-Mobile-connected phone (Galaxy S22) is in the 600-700Mbps range and it doesn't vary based on the time of day.

Verizon Wireless indicates that they're not in the service area either for their 5G home Internet product, although I don't have data points from any UE on their network in that area.

You have to spend more $$ to get "5G Ultra Wideband" on your plan, which is the real deal.

Interesting! The first time I ever remember my phone claiming it had a 5G connection was late 2020/early 2021 (I forget the exact month, but it was within a few months of me moving to a new city). I had taken an Uber downtown, and when trying to do the same to return, I was surprised to see that I couldn't get anything to load, and my phone claimed it had a strong 5G signal. I couldn't find any way to tell it to try to use LTE instead, so I ended up having to walk maybe 10 minutes back in the direction of my apartment before my phone finally could load anything, which coincided with it showing LTE and not 5G.

never knew exactly what had caused the issue, but I suspected it might have something to do with a cell tower reporting 5G to my phone but there not actually being infrastructure to support that in the area. I wonder if something similar to what you describe was happening, and due to an OS bug or something my phone had different expectations of what communication would look like when the other end reported that 5G was supported. Does "5G SA" require any different behavior on the phone's side of things, or would any communications it receives be indistinguishable from those of "proper" 5G?

For that, you want fast and low latency and that means a dense amount of cell towers. The fact that “real 5g” has a shorter range is actually a positive. It means you can make your shit super dense.

In scattered environment the range is too short to excuse the cost of covering it with 5G

There is no niche for it except for "my landlord didn't install proper internet so I use 5G" coz it happened to be alreadhy in range

You're welcome to get your own contract, but if they don't have permission by the owner to dig into the ground and drill into walls to run the lines to you and a place to put meters and equipment, you're not getting the internet/gas.

> everyone rents and you are dependent on landlords for your internet contract as well as your actual ethernet connection

We have mandated infrastructure sharing both for fixed and wireless networks so the competition is pretty good with virtual operators almost always being cheaper than the brand name owning the infra.

... if you even have the choice. Many Americans are only served by a single provider, and this isn't the "too rural" excuse but also with many metropolitan areas. There are exceptions, such as with Utopia Fiber (https://www.utopiafiber.com/) in metropolitan parts of Utah, but this isn't the case even in California, New York or Texas.

It's very different in the US/CA/AU and many other regions of the world. In the former, you are quite likely to rent from a corporate landlord. In many other regions, there are very few corporate landlords.

I am a sole proprietor landlord with one property, but also a shareholder in a large company which lets properties. I can be agreeable with my direct tenants in my own property. I always let them make small modifications to the property - fiber lines, hanging things on walls, swapping out furniture to an extent. And I don't mind discussing ideas like a complete remodel of the property if my tenants are long-term. But it is tremendously difficult for me to be so agreeable to my indirect tenants through the corporate landlord in which I am a shareholder.

Fiduciary responsibility to maximize profits is enshrined in law in the US since the original precedent in Dodge v The Ford Motor Company in 1919. Similar laws exist all over the world. They create a perverse and probably sociopathic environment for managers working in corporate landlords. They must maximize profit and do it this quarter. If a tenant would increase their rent payments on the condition that they can fund their own fibre line to the property, then it is in the interest of the business' managers. If a tenant demands that the business fund a fibre line and they are unwilling to pay more rent, then showing such charity is not in the interests of the business managers. In fact, strictly speaking, such a courtesy to the tenant is against the law.

With that said, shareholders can waive this fiduciary responsibility of their managers to maximize profit. But in my limited experience, in the US, every shareholder has to independently agree to this. And in very large corporate landlords with private and corporate investors, it is practically impossible.

All in all, rent from a small landlord and it won't be a problem. In most of the world, nearly all landlords are private. Proportionally, there are markedly more corporate landlords in the US, Canada, and Australia than in most other regions of the world.

I might be wrong about some of this so I would welcome a discussion. But this is what I've observed.

Recently they put in ‘fiber to the node’ to a nearby crossroads, I now have two 18 Mbps lines which, fortunately, give about 100% of the rated performance so I can’t complain much. I can even use Docker now, YouTube, Netflix, and online gaming are all pretty good. (The Samsung TV, outside of game mode, and some monitors, like the monitor on an Alienware ‘gaming’ laptop, add more lag than the DSL does.). I have a load balancer that sends different connections on different lines so collectively the system meets the FCC broadband definition. Still the people down the road get better speeds without the load balancer for less money.

I’d say it is not just a ‘lack of competition’ but ‘pernicious competition’ that is going on here. For instance, the cable company has less reason to serve my valley because people who want TV can get it from satellite. There are plenty of old and poor people who would rather spend $65 a month for 1 Mbps DSL than $70 for 300 Mbps cable and they will go to public meetings to try to keep the better service away. So in some sense competition from legacy satellite and legacy telco makes service worse.

(TV of course is a great case study of ‘pernicious competition’ because of the structure of the industry. Your local OTA TV station tries to keep a secret that you can tune in free with an antenna because they get cash when you subscribe to cable. There is cable, multiple satellite providers, plus a number of IPTV providers but they all provide the same bundle at the same price. There is no way I can say I want CNBC and Bloomberg and the BBC but CNN, Fox and MSNBC can all go to hell. Given that they get paid no matter if people watch they have no need to maintain quality or a brand promise. CNN tried to go direct-to-consumer but decided it didn’t want to threaten the current cash cow even though it may be a doomed business…. It’s hard to tell because there are powerful forces that will try to rebuild the bundle out of Netflix, Hulu, Disney Plus and all that.)

In contrast, I personally do not derive any pleasure from liberating my home internet gateway from a cable that goes to the wall.

Sure, for the use cases where, for example, you're moving too much, wireless is good

But you get an inferior experience on most other cases, for example, switching it from one device to another, not to mention spectrum crowding etc

I almost always have a better experience switching from one device to another. It’s zero extra taps (just play/pause of some content on the second device) on my Jabra earphones or zero to one taps for my AirPods (one if they’re already playing media on the original device).

Compare that to wired headphones, which I would need to unplug and then snake around my desk to the second device.

Yet many of the same people demand wired internet. Hence I don't believe wireless internet and headphones are comparable.

I don't game, so don't care about ping.

So, where I am, using 4G for my home internet is faster and cheaper, and I can gift data to my phone. I have not had any issues with reliability.

We've had 4G for ten years or more now, I'm not ready to put my faith in 5G and don't expect to be any time soon.

I almost signed up for a new office a mile from the centre where the best cable speed on offer was 20Mbps and the thought of relying upon mobile internet was enough to make me think twice.

I'm currently in semi-rural location, no mobile service at all indoors but, 200 metres from the fibre cable cabinet, enjoying cheap and reliable 60Mbps without a hitch.

3UK have thousands (maybe 10k?) new masts in planning to really improve 5G coverage (and they have the spectrum to actually deliver 2gig/sec to phones in the real world). Annoyingly the (vast in many cases) majority will are being rejected on suprious grounds by the council.

Conspiracy theories regarding harmful radiation can also be a factor.

Several other companies could provide similar speeds at similar (some lower) costs through a fibre connection, which was installed throughout the building.

5G home broadband seems like a fix for when the market for wired options is broken.

I've got access to 4 seperate FTTH providers now in my apartment (all using different infrastructure, not including resellers): VM, Openreach (BT) FTTP, Hyperoptic and now Community fibre.

The limiting factor on speed now is WiFi, which cannot manage 1gig even closer to the router.

As long as the manufacturer foots the eSIM bill (and it's profitable) they can serve ads.

But mature 5g isn't supposed to be any of those.

Not really. The link layer is encrypted, and you're almost certainly running SSL on top of it.

It’s 20-30x faster than my VDSL2 line and only £8/month more expensive

And apart from the day of the Queens funeral it’s been really reliable

On the day of the funeral throughput was like 28kbps, whereas 4G from the same tower was 100Mbps so I presume the teleco had an internal network issue with their 5G backhaul / network (probably sold all the capacity to broadcasters)

With wireless internet, disruption can be caused by many things. That makes diagnosing a problem so much harder. Incidentally it also tends to be less reliable.

Besides that, the massive complexity of the technology stack makes wireless harder to manage, configure, figure out compatibility for, and more prone to software bugs.

But they both have trade-offs. Wireless allows many connections at once rather easily at the cost of interference and (potentially) speed and reliability, cable is more expensive up front if infrastructure isn't already there.

Sounds like short sighted view of the decade

5G has an option for higher frequency towers with signals that drop off much quicker, allowing you to pack towers closer together. Hence giving city centers much more bandwidth.

Down in the suburbs you can still get away with fewer towers with larger range and still give everyone bandwidth.

Out in the countryside is where the problem you propose starts occurring. When the people are so spread out that a single cell is going to be woefully underutilized, and hence not built close enough to your home. But that cross-over is beyond the suburbs, not around the city center.

I expect the future is that rural moves to mostly wireless and inner city areas continue to be connected to fiber.

In Australia we have mostly failed to provide good wired internet to rural areas so they are moving to starlink while the cities are largely connected to fiber now.

One of them is the mm wave which is very high performance over line of site, makes the most of diffraction so it often does well near-line-of-site, then quits working. It is particularly unsuited for handsets because your hand and other body parts can block the antenna. This is a great answer for FWA in favorable geography.

The other 5G is similar to 4G but a bit better performing and my understanding it is actually a bit cheaper for the carrier. 4G infrastructure manages a higher level of spectral efficiency than people thought possible before because of the very close coordination between towers, they actually pipe the baseband signal (around 100 MHz bandwidth) by fiber optic to a central location and process the signals collectively. My understanding is that 5G has a system for coordinating the towers that is still efficient, but not so crazy expensive. So it is ironic that carriers market 5G as something that costs consumers more money because the real driver for it, from the telco perspective, is lower costs.

I always wondered if it wasn’t enough to have 1 infrastructure provider like with streets or anything else. Why build the infrastructure twice or three times?

[ed] Government broke them up in the early 80’s; overnight we saw innovation like slimline touch tone phones and non-acoustic modems.

You do NOT want to return to that. And the only reason that you like monopolies on utilities is because it’s been like that for generations; we honestly don’t know what innovations would happen if utilities and roads were privatized and not enforced by government monopolies.

I'm not sure your assessment is reasonable. You're putting up a strawman by claiming that the only way to have shared infrastructure is to have an abusive monopoly that stifles innovation and screws consumers over. This is false. It's quite possible to have a single infrastructure provider that does it's job, such as a consortium of telcos shipping in to deploy and run shared infrastructure. There are plenty of examples of those types of initiatives in other areas, such as shared ATM networks, which provide a far better service than "everyone for themselves" deployments.

Some scrappy company comes in saying they can offer 10gb fiber for the same price or less, but it takes an infrastructure upgrade. The consortium of incumbents shoot down the idea because it will “cause problems” or whatever. So they’re never forced to spend money to upgrade the shared tech, and your service never improves.

I don’t doubt it. But do you have any examples of this having ever happened? Every case I’m aware of has gone much like the person your replying to describes.

For example, I moved to England and spent the better part of ten years trying to get British Telecom to install a working telephone and internet connection at a house in the center of a large town. They’d send a truck out a few times a year to fix the same line fault, which consisted of having a guy climb a pole, unplug somebody else’s house from the “good” connection, and plug them into one of the bad ones. Presumably that neighbor would get the truck to come back at some point, starting the cycle again.

There are no other providers. Or rather, the other providers all subcontract from BT, so you eventually get the same truck to come out. But now you’ve added another layer of pain to the equation.

If there were a second option, they would get all the business. But there is not, so you’re stuck with what you’re given.

Yes, there are whole Wikipedia lists of this having ever happened.

https://en.wikipedia.org/wiki/Interbank_network

This is no rocket science. I mean, even utilities grids in open markets work like this, for stuff like power and even gas. Nowadays even railway networks in Europe are decoupled from operators. Why do people still cling to old monopoly tales?

Sounds hauntingly familiar.. "How can we guarantee the security of the phone if owners can just side load anything onto it."

https://www.theverge.com/2021/6/23/22546771/apple-side-loadi...

They don't have to be at the cutting edge of future innovation. If customers in the USA or somewhere else, who are paying much more for their broadband, start benefiting from some new technology that requires changes to the shared infrastructure, then Singapore can implement it at that time.

I hope/imagine that a lot of fiber backhaul is shared between operators that share towers.

I agree with this. For the most part, LTE is enough for everything from casual web browsing to streaming video. There just isn't that much of an improvement jumping to 5G in the end-user experience. The is very different from the 3G to 4G jump. I still remember using an original 2G iPhone. Maps basically couldn't load over 2G.

You can just buy a SIM card and you’re on.

Eh.... nitpick. Let's ignore mmwave for a minute and focus on bands that matter for most people. T-Mobile and Verizon both have midband spectrum which is overall most useful. Verizon's is C-band 3.7ghz T-Mobile's BRS 2.5ghz midband is mostly backhaul limited where it's deployed, it'll do ~700mbit/s down but the actual spectrum is capable of at least double this. (in my market I can get 100mhz + 20mhz of N41+N41 aggregated, this is 2.5ghz spectrum). It doesn't matter if I'm next to a tower, it's limited by the backhaul.

I get between 600-700mbps real world if I stand outside my house. I live in semi-rural area and the tower is about 6 miles away nearly line of sight. Indoors the attenuation hits and I mostly use the other bands to reliable pull ~150mbit/s, which is fine for me (and traffic goes mostly over LTE).

> - Telcos desperately need a killer app or use case that drives 5G adoption. They havent got one.

The 'killer app' is called home internet in markets that have trash CLEC's which haven't upgraded infra in 20+ years. Mine is Blightspeed selling 'up to 3mbit/s' copper DSL in 2023. Other option is Comcast, which works but is ludicrously overpriced because there's nobody el> Verizon have some high bandwidth spectrum, but it doesnt propagate as well as lower (and slower) frequencies like what Tmo has.

https://imgur.com/gallery/BFTTMRS

Measurement was taken a decent distance from the city in the suburbs.

I rarely see speeds that fast too; downloading a game pass game is usually 1MB/s for me

There seems to be some phone and plan that gets better coverage than another but it is by no means documented. Maybe it just ‘works better the more money you spend’ but without any explanation of what the rules are I am afraid of spending a lot and still getting bad coverage.

Not to worry, they'll just pass the cost to customers or ask for a bailout and get it. Being a giant corporation is awesome.

FWA is a spotty solution anyway. Economically it works best in the ‘WebPass model’ where you server a whole apartment building in an urban area that has a clear line of site.

Rural carriers are averse to maintaining their copper networks, never mind installing fiber. If you are fortunate you have a line of site to a tower but probably you are less fortunate. When you head south from Ithaca, NY you encounter a dissected plateau at an altitude around 2000 feet. There is no ‘highest hill’. A tower on one ridge can sorta see into the immediately adjacent valleys (though the sides may be steep enough to still make shadows) and under 1 GHz maybe into the next valleys over. Heading South down to Alabama the mountains get bigger but it is basically the same story, where FWA goes to die. It is delusional that it could be cost-effective or even effective universally because in a lot of cases there isn’t even a road or a power line at the top of the ridge.

FWA really does create competition for people where geography is favorable but it is no answer for the many people still stuck with slow DSL.

My rural area is served by an independent telco/ISP. They would agree with the part about maintaining copper, but installing fiber has actually been their solution due to the lower cost of ownership over the long term. I wish more carriers saw it that way.

The lack of a corporate coffee store does not necessarily mean it’s a rural area. Been to plenty of small population counties and there was at least one starcrack store in the area, including a McDiabetes, and other low cost crap.

Usually a rural area is defined upon the population size and density.

Telcos are (in theory, at least) private utility companies, so why does it make sense for them to create the application/use case? Do you mean that Telcos like Verizon are reeling from not creating their own platforms (say, apps or app stores) that charge for use of LTE etc.?

Basically, why and how do you think that Telcos are reeling from not capturing LTE value? Didn't they already by charging for use of their network?

Side-note: If I'm not wrong in my assumption of what you meant, then you conjecture a rather depressing idea that companies feel the need to capture value from an increasing number of market segments and crave hegemonic monopolistic control, for example Microsoft since forever.

In their ideal world, they would put up the investment for the infrastructure (or even better: have gov't pay for it) and then turn on the money printer.

In the days when all that you transmitted were telephone calls, this was easy. Then people started sending IP packets and it got a little harder, but ok, the telcos would just charge a per byte fee. The money was in charging to carry your phone calls and SMSes anyway.

Then the smartphone exploded and suddenly consumers did not give a hoot about phone calls. The sole reason to have a cellphone plan was to get data. The telcos had to expand the network capacity by several orders of magnitude for a customer base that had absolutely no desire to pay by the byte. This blew up their model. Moreover, telcos saw that people were paying $$ every month to folks like netflix, watching video on their phones, chewing up network bandwidth, and the telcos werent getting any piece of that action. Hence the whole net neutrality thing.

The aim isnt that telcos have app stores. They know they cant compete with that. The holy grail is that a telco finds some use of the network that commands a premium price that users will accept. Verizon trades on having the best-est network. AT&T tries to bundle, eg HBO. Tmo goes after the younger customers with discounts.

Now 5G is here and telcos are desperately, desperately trying to find uses for it that cant be satisified by the existing network and biz models. Will they succeed? I wish I knew.