Hacker News new | past | comments | ask | show | jobs | submit login
Twilio Super SIM (twilio.com)
461 points by galfarragem on May 11, 2019 | hide | past | favorite | 175 comments

I've been wondering for a while what the purpose of all the NB band LTE stuff is even for. The going rate is often around 40c/MB in most countries (lower on Twilio by the looks of it, but the same order of magnitude), which seems like there's some very specific high value-per-byte application intended. Some of the technologies people are touting for very long battery life have transmission latencies of 10+ seconds as a trade off, which makes the applications for it even more restricted.

It feels like the result is going to be that literally every device you can purchase will have a always connected LTE lojack attached. Imagine trying to firewall or restrict the network in your business when literally everything has its own backbone, it'll be a complete nightmare.

There are also LEO satellite networks on the way.

Is it feasible to use physical RF shielding for devices?

What’s interesting is a lot of new IoT devices operate or can fallback to a mesh or gateway mode configuration, for example, a group of common sensors that work together. So you’d potentially need to block Bluetooth, ZigBee, WiFi, LTE/GSM, GPS (messaging over GPS) and others to account for different radio technologies.

Shielding by enough to throw any common network off is certainly feasible. Metallized non-opening windows (reflect IR from the sun to save on HVAC), metal plates or thin meshes on the outer walls and either rotating doors or a combination of rf-absorbing walls and a convoluted path to the inside that prevent radio waves from getting in through the door. I'd assume you can feasibly add 60 dB shielding to a building with that. More is obviously possible, there are conference rooms with advanced shielding of >100dB against industrial espionage. They use fiber optic communication lines to the outside because you want all electrical connections to be very heavily filtered to prevent conducting radio waves through them.

Thank you! Are materials like these suitable: http://lessemf.com/fabric.html

If you have physical access to the device, it's probably cheaper/easier to modify it, replacing the antenna with a load to ground. And if you're going to do that, you could probably negotiate with the provider to sell you a cheaper one with no LTE radios.

I'm probably thinking of larger B2B though, individuals/SB probably won't have the expertise/influence

> replacing the antenna with a load to ground.

Would that be sufficient for an iPhone?

Twilit has been on an absolute tear over the last 2 years. I wasn’t sure what they would do next to innovate, but their evolution to customer support products, authentication solutions and now this are inspired.

What sets them apart is that, although they are a US company, they think globally. This product is for a global market, just like their app-based push authentication is (the huge market being Europe which has mandated 2FA requirements for finance companies coming into effect). Too many companies build and think about the US as the only market, then try to bring it to other markets which doesn’t always work well.

Counterpoint. While Twilio is interesting, I wonder what percentage of their business is simply multifactor authentication. Having previously used a wide range of companies for SMS connectivity (including multilingual, flash SMS, contact forwarding, and raw PDU features), and implemented VOIP systems with privately owned physical PSTN interface infrastructure, recently I looked at Twilio for inbound VOIP and SMS services across three Asian jurisdictions. It fell far short of our requirements, which were basic (phone and SMS on the same number, one number per country).

By being the default option for MFA and SMS automation, they are an approved and contracted vendor at a zillion companies. Getting a customer to upgrade an existing contract is 100x easier than wooing a new customer. Especially when you have a reputation for being rock solid at your specialty.

>phone and SMS on the same number, one number per country

Was this actually possible if you'd used the local carriers directly? There could be some regulatory reason, like differences between cell and landline numbering plans for billing.

mobile phone numbers can receive both sms and calls, so a mobile number would work, but you are right. it seems (in Ireland anyway) that getting a mobile number for VoIP stuff is fairly hard... you would need physical SIM cards and hardware...

Their API has been unchanged for years, which shows me they did it right the first time.

So has Amazon Marketplace Web Services and it is by a huge margin the worst API I've seen my share of bad ones over the years.

And yet it is still better than the Amazon Vendor Central API... which they forgot to build at all

Oh my. Vendor central is such a pita one can't help thinking they do it on purpose. (Also, I had to write several puppeteer based scripts for different reporting web apps that don't expose an API and vendor central was definetly the most painful to write)

And ads API, is just so bad. And data is only available for the last 3 months.

One of my clients has been waiting for over 3 months for a reply for access to that API. Right now they are stuck manually processing hundreds of orders per day.

Dm me I have access and can help. I’m specializing it helping amazon sellers.

I wish Twilio would figure out some way to accept SMS from short codes. It’s just annoying when I can’t use my VoIP number for 2FA with certain companies (and I realize it’s the companies fault, not the VoIP provider).

Most companies prevent VOIP numbers from signing up as it is a very common fraud vector.

What causes that? Uber (for example) requires the device ID, then I can't figure out how to call the driver, when (s)he's apparently lost...

This is very cool tech, but it seems like a niche use-case. It's not made for consumers, and the vast majority of IoT devices barely move, let alone between countries.

But, if I were working for a haulage company and was interested in tracking, say, truck location, speed etc across the whole of Europe I could see this being a great way to do it.

PM for the Super SIM here. The problem we're trying to solve here isn't so much that the devices moves between countries, that's somewhat solved, especially in the EU. We want to help companies that deploy devices globally simplify their logistics and supply chain by being able to use a single SIM regardless of the destination of the device.

Will there ever be a consumer offering along these lines? (i.e. i am travelling around and don't want to have to get travel SIMs in every different country I visit).

We won't build specifically for consumers, we're targeting customers with 1000+ devices. My plan is to move my T-Mobile consumer subscription to my eSIM on my iPhone X once T-Mobile supports it and then put a Super SIM in my second slot for global data.

Google Fi is great for this. Wherever I land abroad I'm connected within a minute and pay the same data rate that I do while at home.

What phone do you use for this or is it multiple phones, and what SIM or is it multiple SIMs?

With Google Fi you have to have a Google Pixel Phone, and it just works. Google is a MVNO so they can buy tower time from any carrier.

Edit: And yeah it's a single SIM.

Unfortunately it's only available to US customers.

This exists, but pricing can be exorbitant and availability is heavily location dependent.

Amazon kindle used to do this back in the day. Then people started tethering them and then all the kindle 3's and before got limited to 50 mb/month. (Later ones just had no internet outside amazon/wikipedia iirc)

Unlimited 3G everywhere was why I bought mine. Never tethered it either, but it was limited just the same :/

Your carrier doesn't offer this already?

This. My wife's company builds devices for ag tech. This technology is enabling to reach a global market.

I didn't get that from the product page. Perhaps it should say "Simplify your logistics supply chain with a single SIM solution regardless of the final destination for your device".

Totally missed what market you were going after.

Even if your IoT devices don't move, do you want to deal with 1-3 companies per country you deploy in and follow market changes there, or is that something you're willing to pay someone else to worry about?

You could sell something like 3g internet connected Kindles and without the overhead of some arrangement with a local carriers which Amazon had to set up with their WAN option (which also works globally).

> But, if I were working for a haulage company and was interested in tracking, say, truck location, speed etc across the whole of Europe I could see this being a great way to do it.

If you wanted to do that in the EU you just need a sim card with data from any EU country. EU rules on roaming mean that your sim will work across the EU just as it does from its home country. I agree this would be a great use case if you're not only tracking in the EU though.

If your routes are restricted to only EU countries, sure! Stray over the Croatian border to Montenegro, for example, and you'll be racking up the bills.

Source: knowingly burnt £10 trying to find somewhere child friendly for lunch.

The EU roaming rules are only valid for a certain time period.

I can't get a polish SIM card and use it for an entire year in Germany, for example.

? Yes you can. I use a Polish SIM card(with Play) and use it all year round in UK, except for the 2-3 weeks or so when I'm back in Poland for holidays.

While you probably can. The law is written to prevent use cases like this, to avoid a race to the bottom between operators in different countries. Which makes sense since infrastructure costs can vary largely between countries, so it'd be unfair competition.

Interesting that they do this because they know infrastructure costs vary between countries but they don't give a shit about the price of electronic goods, services, etc in the single market.

If you use it for more than 120 days outside your home country your phone operator is allowed to query this and charge you.

Sure. This hasn't happened yet in all the years I've been doing this. Maybe it's because I don't use data on this card, only calls.

The relevant legislation has only been in force since 20170615 [0].

[0] https://en.wikipedia.org/wiki/European_Union_roaming_regulat...

Bkess you for using ISO 9660, my friend.

I was contemplating doing something similar, especially with eSIM on Apple Watch.

Calling you must be a bit of a pain. I just check and calling abroad from Lithuania is still 0.5 euro per minute...

No, I just have a dual-sim phone(OnePlus 5T) - for data and calls in UK I have a proper British sim. But whenever I need to call my family or anyone in Poland, I just dial from the second(Polish) sim, and call for free, using the minutes that come with my Polish contract(and they can call me for free on my Polish number as well). It's also pretty useful whenever I'm somewhere in the UK that doesn't have a good signal on my main network - then I can use the Polish sim to connect to any network that has the best signal and just use data off that card temporarily.

Then we leave EU, alas and no of the consumer friendly law applies.

Well, yes, it remains unseen what happens once brexit actually happens. But this solution works at the moment, and I can only imagine it can also work in other EU countries.

I am from Norway and we love you, Polish people. <3

Note that the EU rules apply differently to data than they do to voice and SMS usage.

A provider offering unlimited data in the UK can cap your data usage in Belgium or charge e.g. £8 / GB after you use 20GB of data.

That does not sound unreasonable.

Europe > EU

> it seems like a niche use-case

A niche market in a massive pool could still result in hundreds of millions in sales. Imagine every Tesla that needs to communicate with a server, every RaspPi, heck even iPads or what have you. The potential is pretty huge for a tech like this one so hurray for a more expansive tech!!

I am experimenting with GPS luggage tracking hardware with simple live-updates. This is potentially a great use-case.

That's cool! How do you deal with tracking while it doesn't have GPS signal?

WiFi I suppose, using a mapping from BSSID -> lat/long, like Mozilla Location Services or Combain.

What's the power consumption like from attempting to get a GPS lock? Luggage presumably spends most of it's life indoors. I imagine a WiFi scan uses less power than attempting to acquire a GPS signal.

I'm a hobbyist when it comes to hardware/IOT, I've been working on something similar but for RV's instead of luggage. Much harder constraints when it comes to luggage! RVs have power and room to spare.

Oh, you could put a small generator on the luggage wheels, to charge it a bit while wheeling the luggage around!

Sounds like a fun project.

Even if your IoT devices don’t move there’s value in putting them on a mobile network. Particularly for security focused devices if you put a SIM and a battery in it then you become resilient to someone cutting the internet line and power before breaking in.

the vast majority of IoT devices barely move, let alone between countries.


One of the things holding back the development of truly mobile IoT is ease and ubiquity of connectivity. This tries to solve that.

Think of it as Amazon's Whispernet, but with far more penetration.

I have a few Symbisa and Particle units that work like this.

I'm assuming the reason this is positioned as an IOT product is because they need to sell some product based on the network access they are able to negotiate. If they get allowed to use the 911 and voice network, I can imagine the product will shift to consumer pretty fast.

I'm using it in Canada and although data doesn't work (despite their claims) I'm paying one tenth the cost of a local bargain cell plan so I can SMS.

They could move across countries/providers if you think of different use cases: devices on airplanes, boats, weather balloons, long range drones, shipping packages, luggage, etc.

Having a device not tied to a certain provider could allow a shipping container to pass between different owners without needing a proprietary device/plan.

There's at least one payment terminal company out there with a world SIM in their POS system.

It's not the cheapest way to do that kind of system, but deployment is an order of magnitude easier. Unbox it and go.

I was thinking the opposite. It'd be nice not having to purchase a new SIM in each country you visit (if the rates are decent, anyway).

This product is intended for use in IoT devices and not really for 'iPhone use'. A few things to consider:

First of, in the telecom market, there are two ways to get a discount: Buy more SIM cards, or buy more data. Let's say you offer a track-and-trace product, have deployed 1000 GPS trackers, you can easily get the monthly fees and the usage fees down to cents. Same goes for your 'GB-per-month' personal phone subscription, you buy more data, you get it cheaper.

But since roaming will always be costly, even for wholesale carries, SIM card subscriptions that can be used globally are almost always A. expensive, or B intended for IoT (low usage). (GPS, trackers, sensors, feedback buttons, etc). Twilio is focussing on the latter with this SIM.

That brings us to NB-IoT and LTE cat M.

LTE cat M(machine) is basically a stripped down LTE modem, while an NB-IoT modem (or LTE cat Narrowband) is a stripped down LTE cat M modem. Network parameters for these types of connectivity have been tweaked in favour of power consumption at the costs of low bandwidth and high latency. You could say that LTE-M is used for 'MB's per month'-devices (smartwatch for example) and NB-IoT is used for 'bytes per month' devices (is my container full already).

NB-IoT is cheap, extremely power efficient, but is not intended for moving devices.

LTE-M can be used for more demanding applications, like smartwatches, panic buttons (including voice) etc. It can also be used in roaming applications, but that's it. No streaming video etc.

So if you are looking to deploy a large scale IoT application, you want cheap, simple and energy-efficient hardware, which is why you need NB-IoT or LTE-M.

You could go for other technologies like Sigfox or LoRaWan, but they are inferior to NB and LTE-M if you ask me (roaming, availability of network/hardware).

In my view, Twilio does a great job of offering all kinds of telecom services (one stop shop), a common strategy in telecom called 'bundling'. But if you know what you want and where to look, you can get much better rates... Margins are great on this, I can assure you that :)

How is this different to Hologram[1] or EsEye[2]?

Hologram seems to offer an api with as much access as twilio's (data usage report, switching between "networks" (what package you buy, which includes certain countries and pricing...) and monthly payment of a fleet of chips. I'm failing to see the added value of twilio, but the more competency the better!

[1]: https://hologram.io/ [2]: https://www.eseye.com/

it looks like hologram's per-MB cost is 4x twilio's

But much cheaper per device. I guess it depends how much data you are expecting to transfer.

Does it provide voice connectivity?

Not that I'm aware of, but it might be possible. I don't have a chip with me right now so I can't test it.

At my current job we use it for gps trackers and IoT devices attached to vehicles, so we don't use voice.

I wish folks would abstain from making "how is this different to X?" comments. They add very little to the conversation.

I actually like alternative solutions listed when something is posted on HN that I know nothing about. It's obviously best when the poster does some homework and lists the largest differentiators. I've seen the question "how is this different to X?" used, more often than not, in more of a combative context however where the user defends an alternative and then points negatively at the original.

Just my $0.02 and I"m sure there's a better format that could be used, just not sure what that is exactly :)

Are you kidding? They're one of most useful types of comments, as they both inform the readers of potential differentiators and alternatives.

I didn't click the links for this one. But often they provide an existing similar service, available today. This is great if the idea shown is so great you need it yesterday, and you hadn't seen it before.

It is the difference between innovation and prior art/similar business, for me that makes sense to discuss.

At a first glance the pricing of $0.10 - $0.025 per megabyte in Europe seems somewhat expensive. The wholesale rate is €4.50 per gigabyte or $0.005c per megabyte. I assume the markup covers the cost of maintaining the account and the access to many networks worldwide. Can anyone who is already using GSM data in "IoT" devices comment on how much this compares to what you pay now?

€4.50/GB is absolutely insane for Europe. In Italy you can get 50GB for €7 From Kena or ho.mobile (€0.14/GB) or 50GB for €8 from Illiad (€0.16/GB). All at 4G speeds.




Wholesale rate is for businesses where they pay for the actual usage, and consumer-oriented providers, which you mentioned, oversell network capacity.

There's overselling and then there's assuming that the average user is only going to use 3% of their allowance! That seems pretty optimistic. Having said that presumably the wholesale rate varies quite a bit between countries, and I assume some contracts are offered as loss leaders.

One of my relatives in Italy is using Kena. It's solid. Even though they advertise 30 Mbps, we clocked it at 68 once. Generally it's more than enough to watch Youtube, which is where most bandwidth is spent, unless you hotspot to a laptop.

The point about the overselling / oversubscribing for consumer rates actually goes counter to your point since every customer (consumer or business) all have the same priority in European networks.

€4.50/GB is the maximum what EU operators are allowed to charge each other for roaming data transit right now, and what an external operator like Twilio would probably have to pay them. Of course in each country you will find operators who offer better deals on their own networks - but Twilio can't reasonably build out their own worldwide mobile network.

Despite this, I can see this service being popular among companies who are willing to pay a bit more per GB to avoid negotiating with operators and getting SIM cards all over the world.

Depends on the country, no? In Spain you get like 1 to 3 GB (depending on the operator) for 7€. And that's with very cheap operators with poor support.

As long as we're discussing pricing, let me plead for an option with $0 monthly fee, even if the data itself is expensive. I have a couple products in mind where mobile data would be a necessary backup, and spending $2/month/device for that won't work for the market I have in mind.

The value Twilio's offering is when you need coverage also outside Europe, in remote countries. Yes, you can get cheaper data plans to cover the EU, but those often come with very high data rates in other areas.

I get $10/GB on Google Fi and can have 9 data only sim cards on my plan for $20/month.

I don't think you can really compare the prices of this kind of low-data, many-sim offer with consumer stuff.

Recently I talked to a national (Portugal) operator who offers M2M SIM cards and their price for national only cards is 0,19€/MB, international one similar to this 0,27€/MB.

Twilio's price is pretty good compared to what they've got.

This looks pretty expensive compared to prices in South East Asia too. You pay between $10 - $30 for a prepared SIM with 10-15 gigs which comes down to $0.001 - $0.003 per megabyte.

With the plans you mentioned, how much do you pay per month if you use only 1 MB?

I see your point

Twilio: If anyone there is listening, please, please give us an <email> verb in twiml ... it's the simplest thing and would be tremendously useful.

I don't want to host offsite code or write complicated functions just to CC: an SMS to my email inbox ...

A <email> twiml verb would be simple and useful.

Didn’t Twilio buy an email provider that would make this straightforward to implement?

Edit: SendGrid!

As of the completion of the sendgrid acquisition, there is no email verb for twiml. Your only options are to host code on a third-party and call it from Twilio (sendgrid could be that Third party) or write a Twilio function.

But the easiest way to do this, by far, would be a simple email verb in Twiml. No dice so far…

Until they actually ship and make it GA, I reserve the right to call this vapor ware like the rest of the Twilio announcements in the past year. When is VPN for programmable wireless going to ship? Never?

PM for Super SIM here. We're running this as a closed beta now, hard at work getting it to GA.

Don't announce a beta unless it's available for public sign up. Or just wait until the product is GA...like a regular company. Otherwise it leads to customer disappointment and disillusionment.

Just pull VPN for programmable wireless from your site. It's clear that it won't launch within a year or two from being "announced".

> Don't announce a beta unless it's available for public sign up.

Many betas are closed. That’s not unusual.

It's not unusual but it's not a good idea. Just keep it confidential until ready, or ship a MVP and launch it as a real thing.

I just signed in to my twilio account and ordered a "sim starter pack" thinking it was this, but if this is still in closed beta, what did i order? how does super-sim differ from the programmable wireless product?

Twilio PM here. You ordered a starter pack of our IoT SIMs we launched in partnership with T-Mobile. They connect to TMO in the USA and have broad coverage globally. They don’t have access to ATT or VZW in the USA like the Super SIM.

cool, thanks for the reply

Can this do voice calls? I'm in Canada where every cell plan is more expensive than any other in the world. I have friends who purchase plans from Europe and then use the SIM locally with roaming - turns out it be 10x cheaper, the only drawback being that they have a foreign number

PM for Super SIM here. We're not planning on doing voice calls for the Super SIM, we want to really nail the IoT use case first.

I use a data-only cell plan ("tablet" plan from Fido), $15/month for 3 GB.

For voice calls, I use a SIP account from voip.ms, which mostly goes to voicemail (SIP kind of drains cellphone battery). I rarely use voice, and people usually text because calling, and most use Signal/Whatsapp anyway. The voip.ms account is also useful for doing SMS, and associating apps with it (such as Signal/WhatsApp).

This is exactly what I do too, but unfortunately I don't think you can get the tablet plan any more without attaching it to a "real" plan.

This isn't for personal use, so no calls. The data rate is probably around 2-5c/MB for Canada (though its not on the pricing list, that seems to be average), which puts it above even Rogers consumer plans. Google Fi is intended for consumer use and is 1c/MB in Canada (up to a cap, and then unlimited).

"This isn't for personal use, so no calls."

Very little of Twilio is for "personal use" but you can still deploy it as such and there is another class of Twilio SIM product (programmable SIM) that you can do anything (voice, sms, etc.) you want with.

I have built my own little personal telco out of twilio and couldn't be happier ...

Well, except for an <email> verb in twiml - that would make me much happier ...

It's hard to tell if this is similar to their programmable SIM offering or not...

I have programmable SIM deployed in my iOS test phones and use it to place calls and texts from the phones (no contract is nice). I also pay about 12.5$/GB because I have the higher commit turned off

I can relate. I just paid a total of CA$50 for a 1GB prepaid SIM card. Crazy stuff. I don't understand how nobody appears offering half prices, it's clearly possible given they are overcharging a whole lot right now... and doing so seems that would win a lot of customers instantaneously

They say voice won't work in Canada but that data does; I haven't been able to get either to work. But for $3 USD + usage per month I can use SMS.

How expensive is a cell phone plan in Canada?

On the main carriers, a 1GB data plan with unlimited Canada-wide calling is $100 after tax. (Occasionally, there will be a discount that all of the carriers do simultaneously where a few GB is added to each plan for free, and you can drive the price down a bit if you've been with them for a while and threaten to leave.)

Each of the big 3 carriers (Rogers, Bell, Telus) also has a "budget" subsidiary brand (Fido, Virgin, Koodo), where the same plan is $50 or $60. However, customer service is absolutely terrible and they ding you for absolutely everything. For example, if you ever need to call customer service, there's a $35 service charge.

This is for Ontario; other provinces have local carriers (Sasktel, Videotron, etc.), but I'm sure pricing is pretty much the same.

Oh, and roaming in the US costs an extra $10 per day to use your regular plan, even if you only send a single SMS. Best not to use it at all or get a local SIM if you're going to be there for a while.

Is that $100 CAD? If so, that's <$75 USD, which is more than we pay in the US, but not insanely so -- most people I know pay 50-60USD (not including any device payments).

Verizon charges about $80 for unlimited data.

I got really confused when I first saw this and though it was something for consumers not IoT devices (Sorry twilio team for the application for access shrug).

I really wish someone would nail this for consumer phones. I'm coming up on a year away from home in various countries and the constant sim swapping is beginning to get annoying.

I actually bought a dual sim phone so that I can keep my Australian sim card in for 2FA security sms messages (etc).

Before someone suggests it, Google Fi is US only, unless you want to arrange for a reshipper (and even then, I'm not sure if it works).

I was just going to say Google Fi absolutely nails this until I saw your last line and realized they're still just catering to the US market.

The past two years I've been a digital nomad living about 90% time outside the US and traveling to a lot of countries. Fi has not let me down in a single place! I keep maxing out data cuz of the $60 data cap for upto 15gb.

Wish I had a suggestion for a service like this for people based outside the US but looks like no one's trying to serve this market like Google is. I'm sure there's a lot of money to be made if a business just starts focusing on this segment.

Telstra does this (depending on which countries you're in, but it's a pretty solid set). Pricing is exorbitant though ($200 Aussie/mo) and international data is low (10GB/mo).

Still, it's an option, particularly if you're not paying your own phone usage and need to maintain a consistent number.

I’m totally with you.

The dream would be to have a single SIM which roams globally and to which you could attach multiple numbers from various countries.

Closest I’ve found is https://www.truphone.com/ but they seem not to offer individual plans.

Why do you want multiple numbers? Doesn't something like Whatsapp or FaceTime help prevent you from needing them?

Mostly for business. Maybe some people are comfortable to Whatsapp you, but it's inappropriate for a lot of situations.

Also, in a lot of places you need a local number to get things done. Like when making reservations at restaurants or you leave your number in the POS to have them text when a table is ready, etc.

As it is, I have a dual SIM Android phone as well as an iPhone and have 3 postpaid plans in different places. Massive pain.

Got it, so it's the recipient who matters. If whatsapp let you send it out as a plain old SMS from a number you can select, and receive their SMS as a whatsapp message, would it solve your problem?

That works for people you meet socially, but may be less accepted for business. There's also the case of e.g. Uber drivers calling you to ask for directions to where you are.

Slightly off topic. I’m trying to find a SIM for my LTE modem (AT&T/T-mobile bands) that I got as a backup for my Comcast service. I’m looking for pay as you go with no or minimal monthly fees, a reasonable 4g price, and a fallback to unlimited 3g.

Does anyone know of a data plan like that? I tried to use Twilio Wireless but they don’t have the fallback option.

Comcast offers Xfinity Mobile to their internet customers for free. You pay as you go for the data you use ($12/GB). You could probably tether a smartphone for data use.

What do you mean "for free" ?

A Project Fi data only sim is probably your best bet.

Unfortunately that also requires a regular Fi account to be active ($20/mo)


Can I use this in my mobile phone?

What's with the poor data pricing? These plans seem like a non-starter for _any_ application that's remotely data intensive. At $50 / month you're going to get a total of 4GB in the USA. That's just unworkable for streaming video for e.g. remote CCTV applications.

It seems like if you're doing anything interesting that uses more than insignificant amounts of data (beyond e.g. reporting GPS coordinates) you're better off taking out a business contract with a mobile network and using the roaming allowances built in there. E.g. you can get 3x the 4GB data that was billed at $50 by taking out a £15 business SIM contract from Three which incidentally will give you up to 1TB in the UK.

> These plans seem like a non-starter for _any_ application that's remotely data intensive.

Yes, and this is by design. LTE NB-IoT is very different from the regular broadband LTE people are used to. It's made for devices that send a (short) message once in a while. Really, it's best to think about it as for small message oriented applications --- no data stream.

To be low cost, a NB-IoT has only one receive antenna not two (or more) for regular LTE. Also, to reach into basements and harsh conditions, NB-IoT supports many repetitions. In theory up to x2048 times regular LTE, although most networks won't go that high. This means that the spectral efficiency of NB-IoT is much, much lower than for regular LTE. One byte over the air can cost a lot of resources, but it's ok because it's made for application with very little data to transmit: a meter level, a position... The data plans telcos sell for NB-IoT will forbid using a lot of data, it's just too inefficient. What NB-IoT needs to be is 1) low cost and 2) low power. For a meter sending one report per day, one can target a lifetime of 10+ years for example.

So although NB-IoT can use the regular LTE channels, it really is a completely different kind of animal. LTE-M is similar, but a bit bigger / higher throughput. Still for small IoT applications, but with more data than NB-IoT can accomodate.

If NB-IoT is a scooter, regular LTE is a truck. If you can fill the truck, it's more efficient. If you can't, because all you have to deliver is a pizza, then a scooter is best. Use the right tool for the job.

>remote CCTV applications.

This is not for that kind of use case.

>Super SIM allows IoT devices to connect to multiple operators. Get instant global connectivity on a single SIM. >Your device can travel around the world using one SIM...

Think of a ballon traveling over Eurasia and sending back data.

I could be wrong, please correct me if I am, but doesn’t hologram offer all these same things?

> Add new IMSIs through over-the-air updates on a platform with global resilience and redundancy.

How big is the IMSI namespace? How quickly will Twilio put IMSIs back into a shared pool? Will they retain logs of IMSI-eSIM mapping?

This sounds very similar to the Truphone SIM: https://www.truphone.com/us/consumer/sim/

I wonder how a card-based solution will compete with e-sim. Seems like twilio’s advantage will be reduced to demand aggregation once e-sims are popular. Anyone have thoughts on differentiation in this space?

$90/gb! Yikes!

Yay! We are approaching the SIMless world where you connect to mobile networks without a SIM or a universal SIM.

Kinda. I have eSIM phone and was hoping to sign up for a temporary plan while spending 3 days in Singapore - NAH! "Please enter your national ID number or come to our office to use eSIM". Meanwhile SIM cards are being sold while on plane, 100GB for 12 euros...

Ok, I guess I can then easily sign up for this while back in Europe? Nope, every local website has no signs of english sign up form, many require ID card...

The hope is I can do it via my phone menu, but even then calling me is going to be a rip-off (0.5 euro per minute when people been video calling for free for over a decade...)

You make it sound like this is a trend, but this article is the first instance I've ever heart of the concept of a universal SIM?

The technology has been around for several years now. There was reluctance (naturally) by the telcos in letting this happen, but applications are demanding it now bringing it more into the mainstream from the niche it used to occupy earlier.

Hologram is another example, and there are many more but enterprise-focused.


At scale, will the price get to a point at which this can be a second interface?

Maybe used as an OOB management plane interface for OTA updates, remote diagnostics etc? Perhaps used for some_other_use_cases in which second global interface would be useful?

Would there be isolation and security benefits to that type of second interface?

Their programmable SIM is super cool to me. I have been considering making my own DIY version of a light phone. Super SIM is interesting for IoT applications where you have something that crosses a wide geographic area (like freight). I'm very curious to see where this goes.

Don't know if anyone else has, but I've seen so called "global SIM" with multi IMSI type functionality from so called "secure phone" providers in the past.


So could one use this as an international works anywhere sim?

I don't think this class of SIMs that Twilio offers are voice capable ... they do provide tmobile SIMs that are voice capable, but they call them something different ...

You can always use OTT voice.

The business use that immediately comes to mind is tracking, such as for vehicles. Often cars are stolen in Europe and moved between countries.

Forget IOT, I want this for my phone!

It sounds almost too good to be true.

Google Fi has $10/GB global LTE for US residents. Multiple SIM cards sharing the same plan, works on non-Google devices too, including mobile hotspots. In the US, they use T-Mobile.

Yes Google Fi seems great, I just wish they wouold offer it to non US residents.

It's amusing that they went through all the pain of making a global SIM to make it available only for the US residents.

You want $25/GB for your phone?

This is great (one plan worldwide, but in places with weak currency like India, paying in dollars might turn out to be expensive), but overall isn't the industry moving in the direction of eSIMs (no physical SIM required)?

Moreover, once OneWeb, Starlink and the like launch worldwide internet service, it might pretty much render the telecom industry as it is today irrelevant?

I think we can expect Starlink to have similar coverage to GPS: great in the country, useless in my bedroom in a multi-story building. Also the expected receiver size for Starlink is an order of magnitude bigger than the one for a cellular network.

For phones, tracking, etc the cellular network would still be superior. Something like a car might go with both starlink and cellular because anywhere where one is weak the other is likely to be strong.

> great in the country, useless in my bedroom in a multi-story building...

I'd think it'd work a bit like cable/DSL often does in an apartment company - one receiver for the building.

Or put in on the side of the building like regular satellite dishes. But realistically any apartment building that's not in the middle of nowhere will be better served by fiber anyways. Once you have multiple customers in the same building they will quickly feel the limit of how much bandwidth you can push to a small area throught the air from 500km away.

Maybe, but there is a huge market of people that do not have an eSIM. They could use this, right now: this will allow Twilio to get a head-start in this market, rather than waiting until eSIMs reach critical mass (probably years away).

eSIM, in general, is a software functionality to change subscriptions. Embedded or pluggable or any other SIM form factor (software or physical) can have the so-called eSIM technology.

Google Fi should roll out something like this. They already have good global coverage and prices

I’d be curious to hear from the folks at Particle.io on the offering/competition.

Any idea what the approximate pricing per GB is and how it scales with more devices?

I will wait for WiFi via Starlink, connectivity will be everywhere.

Isn’t that exactly what Hologram does?

This can be a game changer.

From a product perspective, this is really confusing.

Twilio make business-facing APIs, is this supposed to be a consumer product? Why would I want API access to my own usage data? This feels like a far cry from their strategic strengths, except for that it’s phone-related.

This is for Internet of Thing usage, the customer is a company who is making Internet Widgets with some need to be able to communicate no matter where their Internet Widget is operating, without necessarily having to configure or prepare for the destination.

Agreed until you consider that there is a massive iot market and they're going after it with pretty cool innovation. We will probably never really hear about this project unless we're specifically in the iot world.

That said I use a twilio SIM on my test phone and it is great. Was thinking of dual SIM with it but there's a few limitations I didn't want to live with.

> Super SIM allows IoT devices to connect to multiple operators.

It’s for people who deploy IoT devices.

This is still B2B. It's aimed at IoT devices

Nope. Not going to happen. I don't care about a multi-purpose SIM. I want cheap data.

I keep my T-Mobile SIM for when I fly into the US, I reload it for $30-60 and use it for a month. I'm rarely in the US.

I get back to Indonesia, I go to the local SIM store and buy 50GB of data for ~$13. With Twilio rates I'd be paying $1,250-5,000. Pass.

You do realize that this is not a product for end-users at all?

Guidelines | FAQ | Support | API | Security | Lists | Bookmarklet | Legal | Apply to YC | Contact