National Broadband Plan or Starlink

https://www.youtube.com/watch?v=h0Itx_TUOKA

Nice video to see the equipment and service, I know the price is expensive and while we here in Ireland are lucky to have the NBP going ahead this is going to be a life changer for so many people around the world.

The Cush wrote: »

In a Dáil written reply on Tuesday...

Typical COMREG/Dept of Communications (et al). They don't want to acknowledge that they were too restrictive with their original specification. Yet, they can no longer ignore Starlink, so they have to justify why it is not (yet) in play.

The Cush

Article in The Currency (behind paywall)

What will Elon Musk’s Starlink mean for the world (and Ireland’s costly and long-awaited National Broadband Plan)?

The founder of Tesla and SpaceX is working on ambitious plans to bring fast and cheap broadband to the world – and he has just got an Irish licence. As the country prepares to invest billions in building broadband infrastructure, could there be another way?

24th Nov, 2020

It is starting. Nearly three years ago, in February 2018, the Elon Musk-founded rocket company SpaceX put two experimental satellites into orbit. The satellites were delivered aboard SpaceX’s Falcon 9 Full Thrust rocket – probably the most successful rocket in human history in terms of launch cadence and reliability. Most famously, the Falcon 9 was the first – and remains – the only rocket so far developed that can deliver payloads to orbit and subsequently land and then re-use its expensive first stage booster. The company, which had for years been successfully putting payloads into orbit on behalf of its…

https://thecurrency.news/articles/29957/what-will-elon-musks-starlink-mean-for-the-world-and-irelands-costly-and-long-awaited-national-broadband-plan/

I assume Comreg has issued the licence but they don't publish satellite licence details on their website.

Gavin Sheridan has said on Twitter that the licence has been granted. Hopefully this is the beginning of an informed look at the options for more remote NBP sites.

NewClareman wrote: »

Hopefully this is the beginning of an informed look at the options for more remote NBP sites.

It's an article.... It's also written by a journo who has a long history of anti NBP history. No idea why they are continually anti NBP but they do seem to post alot about how great their own FTTH from Eir is.

Amusing...

listermint wrote: »

It's an article.... It's also written by a journo who has a long history of anti NBP history.

Gavin Sheridan has a track record of pursuing open government and looking for information on how government decisions are made. All of us who value democracy and transparency owe him a debt. We need more like him.

In the NBP case he submitted a Freedom of Information request for the contract between the State and Enet. This was fully justified, given the amount of money being spent, the fact that only one consortium was now involved and concerns as to how the evaluation process was conducted.

His FOI request was refused by the Department. He then pursued his request all the way to the Supreme Court. The decision in the Supreme Court was that public bodies must justify refusals to disclose confidential or commercially sensitive documents. Whether it will result in the NBP contract being released I don't know. What is certain is that it makes FOI requests more difficult to refuse.

NewClareman wrote: »

Gavin Sheridan has a track record of pursuing open government and looking for information on how government decisions are made. All of us who value democracy and transparency owe him a debt. We need more like him.

In the NBP case he submitted a Freedom of Information request for the contract between the State and Enet. This was fully justified, given the amount of money being spent, the fact that only one consortium was now involved and concerns as to how the evaluation process was conducted.

His FOI request was refused by the Department. He then pursued his request all the way to the Supreme Court. The decision in the Supreme Court was that public bodies must justify refusals to disclose confidential or commercially sensitive documents. Whether it will result in the NBP contract being released I don't know. What is certain is that it makes FOI requests more difficult to refuse.

Outside of his FOI work which I commend him for.

He has a chip on his shoulder about NBP in its entirety. And he continually shills starling only writing positive things about it.

I guess contrary makes your voice louder. Much easier to do on a 500mb line.

scunermac

theguzman wrote: »

https://www.youtube.com/watch?v=h0Itx_TUOKA

Nice video to see the equipment and service, I know the price is expensive and while we here in Ireland are lucky to have the NBP going ahead this is going to be a life changer for so many people around the world.

Those Pings are quite impressive for a satellite signal. I was always under the impression that satellite broadband was dead in the water because of the inability to achieve low ping times. I stand corrected.

dam099

I knew LEO would give better pings but that's more impressive than I expected.

The speeds are obviously great if you have been on a rural connection with only a few Mbps but not really comparable to what NBI will be able to offer.

Whats with the weird stutter on progression of the graph line for the downloads and uploads though? I usually would have seen a nice smooth progression across for DSL/Cable/FTTH and even Mobile, that seems to stop and start (I would have thought bad latency but the ping seems to belie that). Often the graph line might bounce up and down as speed fluctuates but not stop and start like that.

Gooey Looey

You can't really trust a speedtest. Satellite has always used TCP spoofing, where the station on the ground sent back early acks to quickly ramp up transfer speed and I'm sure there are more modern smoke and mirror tricks going on here. This is what made satellite unsuitable for anything in real time like VPN, VoIP or gaming

ED E

Gooey Looey wrote: »

You can't really trust a speedtest. Satellite has always used TCP spoofing, where the station on the ground sent back early acks to quickly ramp up transfer speed and I'm sure there are more modern smoke and mirror tricks going on here. This is what made satellite unsuitable for anything in real time like VPN, VoIP or gaming

It seems very probable as they're talking about software updates fixing latency. Unless their basic pathing sucks at the minute that sounds like traffic manipulation.

The Cush

scunermac wrote: »

Those Pings are quite impressive for a satellite signal. I was always under the impression that satellite broadband was dead in the water because of the inability to achieve low ping times. I stand corrected.

Low altitude combined with lots of ground stations to get those pings. To maintain those pings and qualify for rural broadband funding they have applied to the FCC to amend their filing to use lower orbital planes between 520-570 Kms, the current orbit is 550 Kms.

Gooey Looey wrote: »

You can't really trust a speedtest. Satellite has always used TCP spoofing, where the station on the ground sent back early acks to quickly ramp up transfer speed and I'm sure there are more modern smoke and mirror tricks going on here. This is what made satellite unsuitable for anything in real time like VPN, VoIP or gaming

The US Rural Digital Opportunity Fund, would be somewhat comparable to the NBP, in terms of intent. This is part of SpaceX's submission.

While LEO companies such as SpaceX, were allowed to apply for the low latency (<100mS) tier, the FCC were however sceptical that this latency target could be met. (See pages 36/37 in the linked document.) On October 13 the FCC released a list of “qualified bidders” for the RDOF funds. That list included Space X, so that concern must have been allayed.

I have no idea whether technology-specific test criteria were applied. However, the standard FCC approach to assessing broadband performance, measures round trip delay. That leaves no room for traffic manipulation.

ED E wrote: »

It seems very probable as they're talking about software updates fixing latency. Unless their basic pathing sucks at the minute that sounds like traffic manipulation.

It's very probable that they're doing software updates on a frequent basis. I suspect that latency is only one of their many concerns. The complexity of the Starlink network routing is mind boggling. At least it is to my poor mind.

) Mark Handley published a YouTube video, nearly a year ago ago, which gave some idea of what could be involved. However, the complexity in the space segment doesn't end with high speed dynamic routing. Privacy concerns add a whole new layer of complexity, as explained in this Phys.org article.

Given the above, it is likely that frequent software updates, as they refine their approach, will be the norm for some time. It also explains why SpaceX are so guarded in the technical information they publish.

You wouldn't want to under-estimate the bould Elon Musk. He has revolutionised the car industry - the old behemoths are in a mad, expensive scramble to catch up. Tesla is now worth more than GM. Some achievement for a fellow who began his car manufacturing business from scratch, just a few years ago. He is the classic disruptor.

He has also upended the satellite launch business, with his reusable first stage section of the rockets. He now has that market cornered, on a worldwide basis.

He may just disrupt the incumbent broadband providers. The initial beta testing is proving positive - 50 to 150 Mbit/s download, with respectable upload and latency figures too. 150 Mbit/s is more that adequate for 99% of households, who want stream Netflix and connect to the office IT system. And he says that he will improve on those figures. He could truly revolutionise the world by making broadband available throughout the globe, in about 3 years.

The cost of the dish seems high: but if you think about it this way... the NBP plan will serve some 500,000 premises and is costing billions. Give the 500,000 premises a Starlink dish @ €500 a pop: this would cost the taxpayer €250 million - a small fraction of the NBP cost.

Of course, I am not advocating this - his project has yet to prove itself on a far greater scale than the current beta testing that is currently going on. But, as I said, don't underestimate the man. If anyone can pull it off, he can.

Poulgorm wrote: »

You wouldn't want to under-estimate the bould Elon Musk...
He may just disrupt the incumbent broadband providers...

I would never underestimate Elon Musk, he's an amazing innovator. But Starlink is no threat to the incumbents, even he acknowledges this. He said earlier this year (at satellite 2020) “I want to be clear, it’s not like Starlink is some huge threat to telcos. I want to be super clear it is not”. Look at the video at 16min in.

However, he also said: Starlink will likely serve the “3 or 4 percent hardest-to-reach customers for telcos” and “people who simply have no connectivity right now, or the connectivity is really bad..." “So I think it will be actually helpful and take a significant load off the traditional telcos.”

That would make it the ideal solution for the most difficult to serve NBP sites. Even if the service is subsidised for the end user it could still work out less expensive. It could also be a far faster solution for those waiting.

If you're trying to understand the difference between Starlink and 5G, this article won't help you.

NewClareman wrote: »

However, he also said: Starlink will likely serve the “3 or 4 percent hardest-to-reach customers for telcos” and “people who simply have no connectivity right now, or the connectivity is really bad..."

The economics of the Starlink project will be very interesting. Billions of people in Africa, South America and vast tracts of Asia have poor or no broadband. However, a lot of these people are relatively poor and couldn't afford the monthly charge that the current handful of customers are paying.

The capital costs of the Starlink project must be enormous - he is talking of up to 30,000 satellites within a few years. But, say, if he just gets a billion of these people and charges them, say, $25 per month, Starlink would have an income of $25 billion per month (or 300 billion per annum)

The man could be onto something...

Poulgorm wrote: »

The economics of the Starlink project will be very interesting. Billions of people in Africa, South America and vast tracts of Asia have poor or no broadband. However, a lot of these people are relatively poor and couldn't afford the monthly charge that the current handful of customers are paying.

The capital costs of the Starlink project must be enormous - he is talking of up to 30,000 satellites within a few years. But, say, if he just gets a billion of these people and charges them, say, $25 per month, Starlink would have an income of $25 billion per month (or 300 billion per annum)

The man could be onto something...

If anyone thinks this project is being build for rural people in countries all over the world they are not seeing its point.

This is a long term project specifically for driverless vehicles and avoiding any or all the costs associated with dealing with thousands of Telcos and ISPs around the world for it.

Rural connectivity is a side show to that.

The Cush wrote: »

At $500 or its Irish equivalent plus VAT and shipping costs, the equipment cost is very high as an interim solution for the domestic Irish market and will probably be prohibitive.

According to Business Insider (behind paywall) SpaceX is paying Swiss company ST $2.4 billion to manufacture 1 million terminals. If true, it means that each terminal costs over €2000, ex installation. That would reinforce your point above. If Starlink is to have any role, it would have to be a long-term one.

According to the article, SpaceX signed the agreement with ST a few years ago. This video gives an idea as to why the dish is so expensive. This is the original patent.

In the Business Insider article linked, Dave Stehlin, the CEO of the Telecommunications Industry Association, said he "said he was excited about the prospect of a new, relatively low-flying satellite-internet provider that hopes to offer rural Americans a high-speed broadband-like service; an oft-stated goal of SpaceX". He also said that he would expect the production cost to decline, later in the production run.

listermint wrote: »

This is a long term project specifically for driverless vehicles and avoiding any or all the costs associated with dealing with thousands of Telcos and ISPs around the world for it.
Rural connectivity is a side show to that.

I doubt very much that rural connectivity is a sideshow. SpaceX went to great lengths to be considered fo the US Rural Digital Opportunity Fund. In the short-term this area, along with mobile connectivity and military applications are likely to be the main revenue sources.

I do agree that driverless cars is one opportunity, among many, that SpaceX appear to be targeting. How long it takes is another matter. There is yet another layer of complexity and lots of issues to be solved.

Copy of the agreement between Space X and Ector County Independent School District. The amendment to the original contract halved the equipment cost to $499, the price SpaceX picked for its public beta. Other online sources indicate that this figure is based on what SpaceX believe the US market will bear.

The FCC has just published the results of its Rural Digital Opportunity Fund Phase I auction. SpaceX is to get $885 million to provide rural broadband to 642,000 premises in 35 states, using Starlink. (See page 24 onwards, of Appendix A.)

The criteria for selection were tight, so the FCC must be satisfied that they can provide a decent service.

It's worth noting that a total of $9.2billion in funding was released. This is expected to support high-speed broadband to over 5.2 million unserved premises in rural USA. 99.7% of these locations are expected to get broadband with speeds of at least 100/20 Mbps, with over 85% getting gigabit-speed broadband.

NewClareman wrote: »

The FCC has just published the results of its Rural Digital Opportunity Fund Phase I auction. SpaceX is to get $885 million to provide rural broadband to 642,000 premises in 35 states, using Starlink. (See page 24 onwards, of Appendix A.)

The criteria for selection were tight, so the FCC must be satisfied that they can provide a decent service.

About €740 million to serve 642,000 rural premises.

Great value.

Poulgorm wrote: »

About €740 million to serve 642,000 rural premises.

Great value.

It represents a Govt subsidy of around €1,150 per house for Starlink which required zero infrastructural build across the continental United States, we are spending €4,950 per house here in our NBP Fibre to the Home subsidy which is open access, future proofed and represents a huge positive investment into the countries economy. If NBI asked for a €1k connection fee or similar to the ESB's connection fee I would gladly pay it if I thought I'd get fibre within a reasonble timeframe of say 3 months.

The US is really scrimping here considering the massive sums of money they have at their disposal, the only real winner here is Elon Musk getting a subsidy for twice the cost of what he sells his product on the free market for already $500 equipment cost, plus the end user stuck on a monopoly $80 - $100 per month of a subscription. I really hope for Starlink that it can deliver the data and throughput and does not develop into a low earth orbit version of Imagine, however the laws of physics and finite spectrum will ensure that Starlink too will fall victim to congestion and the usual traffic chokes that occur.

I am most happy that we signed the NBP contract last year before a) The Pandemic struck and the countries finances subsequently collapsed, b) Corrupt Fianna Fail got into Govt, c) Starlink began to emerge as a rural alternative. d) A brown envelope to Fianna Fail would see some company Ltd. stuffed with Fianna Fail cronie directors probably getting €1-€2billion in an even worse value deal for Ireland, like how FF and the Greens wasted millions on the earlier NBS subsidising Three's 3G network and useless two-way VSAT for rural dwellers.

Using Starlink for the NBP would be akin to the Govt saying back in the 1950's here is a €1,000 to each house go buy a generator and fend for yourselfs instead of investing in the necessary crucial infrastructure of the ESB and rural electrification.

I am not anti Starlink, quiet on the contrary I am following it very closely and am enthusiastic for it, I may even end up getting it for my own remote property next year if I don't have NBP FTTH before then. I am a fan of Elon Musk and his disruptive nature. I have travelled the world and been in places where I can imagine the effect starlink will have on their society where they don't even have a phone signal or reliable electricity.

I can see huge growth in remote payments by phone and general e-commerce in Asia and Africa where traditional internet solutions are not available. This where Starlink and the other touted LEO Sat providers will excel plus cruise ships, in-flight-wifi, driverless vehicles, remote rural monitoring and military applications. The fact that Jeff Bezos is building his own competing LEO fleet shows that Amazon want to bring another billion+ consumers into the world-wide web and cents on the dollar of each purchase each year equals mega bucks for them. When developing economies can't build proper internet infrastructure themselves Starlink etc. will bypass this drawback to global growth. We are one of the most developed economies in the world with 3rd highest GDP and the highest discounting microstate taxhavens, we should have had FTTH back in 2010 nevermind be talking about it now and Fibre is the future for Ireland and certainly not starlink.

The circumstances in rural USA are different to here. Many (most?) of the non-coastal states there have a population density which is much lower than rural Ireland. Providing an FTTH grid to all the rural areas there would require a subsidy which would dwarf our NBI subsidy. So that will never happen. So, the €1,150 subsidy per house probably represents good value for their situation.

At this stage, nobody knows the ultimate capacity of the Starlink scheme.

But as NewClareman says, it may have an application here for some the more remote locations.

Nice to know that theguzman believes that FG are so honourable...

Among the updates announced are improved NAT types, bug fixes to Xbox Live, improved latency, and a snow melt mode which is to improve signal on particularly snowy days.

NewClareman wrote: »

Among the updates announced are improved NAT types, bug fixes to Xbox Live, improved latency, and a snow melt mode which is to improve signal on particularly snowy days.

Some people in the UK have reported getting invites for the better than nothing beta over on Reddit.

https://www.reddit.com/r/Starlink/comments/kg4oyu/uk_invites_happening/

As we can see from the map of the starlink fleet Ireland has a bird within range almost most of the time now also, this will of course improve as more and more sats get launched.

https://satellitemap.space/

If you are stuck on a crappy wisp or some thing with no alternative then this will be a great option if you can afford the expensive price-tag and it looks like it might available here long before the fibre will arrive.

theguzman wrote: »

As we can see from the map of the starlink fleet Ireland has a bird within range almost most of the time now also, this will of course improve as more and more sats get launched.
https://satellitemap.space/

That map strengthens the case for Starlink to be considered for the more remote NBP sites. At any time there is at least one satellite where the only populated region it can target is Ireland. I cannot imagine why SpaceX wouldn't want to exploit this capacity by giving favourable wholesale rates for a guaranteed customer base, in this region.

NewClareman wrote: »

That map strengthens the case for Starlink to be considered for the more remote NBP sites. At any time there is at least one satellite where the only populated region it can target is Ireland. I cannot imagine why SpaceX wouldn't want to exploit this capacity by giving favourable wholesale rates for a guaranteed customer base, in this region.

I would see nothing wrong with it as a stop gap measure until fibre arrives in the meantime but with the contracts signed and the NBP heavily underway it is unlikely. Put in Starlink for a man in Donegal or West Cork etc. who won't get NBP FTTH until 2024. Subsidised Starlink would be a monopoly and Elon won't be giving access to the charlatan WISP's who will be hopping onto the NBI as a life-raft for their businesses once the penny drops for those reliant on WISPs across the country.