Hacker News new | past | comments | ask | show | jobs | submit login
Twilio Super Sim – Public Beta (twilio.com)
262 points by samdung on June 12, 2020 | hide | past | favorite | 142 comments



I've always thought it would be fun to make experimental hardware devices with this type of technology. An example:

Build a solar-powered device that you stick up in a tree in a forested patch near an urban center where kids go and have bush parties/bonfires/etc. The device would include a text-to-speech synthesizer. It's dormant most of the time, but designed to activate when people are around it at night. As the "admin", you get a notification when it turns on and can receive some low-res video/audio from the scene, and you can send it messages that are read aloud using the TTS synth.

I imagine it would be quite disconcerting if you were smoking pot in the woods and suddenly a voice in the trees started talking to you.


I can imagine such a device would be quite disconcerting to Winston and Julia as well, hidden away in a natural clearing; around a tiny grassy knoll surrounded by tall saplings that shut it in completely.


Orwell underestimated our ability to make surveillance devices tiny and sprinkle them everywhere


He also underestimated our willingness to buy our own surveillance devices and put them in our homes


Call the git repo "moses" or "burning-bush".


While kinda funny to imagine up I would hope somewhere in the process between developing the hardware and physically hanging it in the tree you would have a 'oh no!' moment and realize just how creepy what you're doing is :)


wow just let them be. they're surveilled enough as is without literally the trees having ears...


Attaching devices to trees is actually a serious challenge that people are working on in order to provide better fire detection systems. Major obstacle is charging and there were, unsuccessful AFAIK, attempts to charge them using the energy provided by trees themselves.


Brilliant! Get this man investments!


Who is going to fund a company whose business model involves making people hear mysterious voices from on high? Well, I suppose angel investors might.


Just replace the creepy voices with ads and watch the money roll in.


In college, some friends and I tried to build a detection system that would predict conversation lulls in public spaces - like when you're in a restaurant and it randomly gets quiet - so we could insert premium audio ads at those moments.

We never got it working, but when I see stuff like this sometimes I still think about it.


Eavesdropping is illegal in a number of states.


I worked on IoT projects for two major wireless carriers both trying to sell hardware integrated with data management. I think the idea was to sort of create value-added services on top of the dumb pipe.

Both companies invested heavily in the projects and both failed badly.

One of the major issues was just getting a dev setup using all the proprietary stuff involved. Onboarding developers was just one issue.

Both systems tied together a bunch of partners, and the folks running both shows had no technical background. Nor their managers.

Order the SIM from the console makes things pretty easy.

Twilio has consistently delivered easy ways to access cell stuff. This is def a step in the right direction. It would be awesome to tinker with this product for sensor data.


If you're interested in this space, Hologram does the same thing (and appears to have had a head start compared to Twilio): https://hologram.io


Was going to post the same thing. I wonder how the two compare.

Pricing-wise, seems like Twilio has higher fixed but lower usage costs:

https://www.twilio.com/docs/iot/supersim

| Starting at $2 per active Super SIM per month.

| Data Usage - Starting at $0.10 / MB.

https://hologram.io/pricing/

| $1 per device per month + $0.38 / MB.

Of course these prices depend on device volume. Twilio may be able to offer lower costs because of their scale, but perhaps Hologram can provide a better user experience because this is the only thing they do.


I'm frankly appalled at $0.10/MB. Are we still in the 90's?


I work for an IoT company, and I can tell you that data costs are high for IoT because the cost model is different. You purchase a 10GB plan from AT&T for your cell phone and they know that you likely won't use it all, so they can "oversell" data. With IoT, you have a much better handle on how much data you use so when you purchase a 10MB plan, they know you will use almost all of it. Additionally, IoT devices tend to have higher backend infrastructure costs as they often connect/disconnect from the towers as they wake up and go to sleep, or move around between towers /carriers a lot. Plus the "always roaming" statement above adds cost in exchange for flexibility.

It's just a different business altogether.


ATT offers an "unlimited" plan on their website right now with 100gb of "premium" (read unthrottled) data for $85/mo (goes down to $50 if you want 4 lines).

Twillio is charging >10x the rate per-byte. ($1e-8 / byte compared to the $1e-9 that ATT charges).

I don't know about you, but I use more than 1/10th of my data plan every month.


>Twillio is charging >10x the rate per-byte.

If you want to compare ATT to this offering, it would be fair to compare how much ATT for a single megabyte of pay-as-you-go international data. That price is $2.05 according to https://upgradedpoints.com/att-international-phone-plans/#:~....

They also won't give you a line for $2/month.


If you go to AT&T and say "I want 10,000 SIMs that'll let me use a few kilobytes of data a day, with enterprise management features", they either won't sell them to you or will sell them at prices comparable to Twilio (except they won't work when out of range of an AT&T tower, and certainly won't work overseas). Their minimum pay-as-you-go plan that includes any data at all seems to be about $30/mo - if you're using maybe 10 kilobytes a day, that's not feasible.


AT&T's IoT pricing seems to undercut Twilio and Hologram on price: https://marketplace.att.com/products/att-iot-dataplans-lte-n...

This platform has been around for at least 4 years iirc.


It’s showing $1.50/MB and $22/GB and it’s a network configuration that is AT&T except in areas that don’t have service it’ll allow regional carriers and TMo.

Hologram SIMs are price competitive here and allow devices to connect to the best network available without unfavorable network preference schemes.

Definitely difficult to parse.


You're only thinking about the data fees. There's much more to an IoT offering that is priced into that cost. There's plenty of enterprise features, for example. Bulk management of SIMs is a huge one: setting data limits per-SIM or per-fleet, activating/deactivating them for seasonal devices or new distributions. Also consider SLAs for your very important data, or dedicated support from the carrier. You don't need any of these things when you are managing one phone, or even a family of phones, but they are essential for fleets of 100's or 1000's of devices.

Also, keep in mind that is data being sold to you for use only on their network. Roaming fees will be exorbitant if you need to go outside their network, which may be needed to reach rural areas, or when you have an asset tracker traveling between states or territories. Keep in mint that "normal priced data package" has very predictable usage — your phone is almost always online, meaning the backend infrastructure costs are low since it's not constantly going offline/online. The tower going out and verifying an IoT device should be on the network (especially if going through a roaming partner) is surprisingly expensive.


Yes and no. Whether the phone is roaming or not, the cost of data is the very same (marginally, zero). The industry a s a whole has a major issue if they think they can sell data for 10c/Mb. What I'm saying is that this is chocking a potentially huge industry, where only the applications that transmit ridiculously little data (by today's standards) can be made economically. My feeling is also that the overhead of switching and accounting for 100 bytes or one gigabyte is very very similar....


These devices are essentially roaming wherever they are. The use case is that you're sending far smaller data than a megabyte in even a day - maybe a GPS update once an hour, that sort of thing.


A single visit to one of the apple's product landing page would cost - $5.

Doesn't seem too bad to me.


For hologram.io it's $1.50/SIM + $0.40/MB under 100 devices, but only $0.65/SIM + $0.33/MB over 5000


These are just self-service pricing options to get people started on Day 1.

When we price custom for enterprise (i.e. specific countries, carriers, roll-out plan), the pricing becomes a lot more flexible. Should be seeing updates to that messaging soon.



what are some of cool products built with this tech.


My mailbox is in a block of other mailboxes that is fairly far from my home (say, 500m) and out of wifi range without an especially large antenna. I was thinking of making a small battery-powered widget to tell me when the mail arrived and adding it to my Home Assistant (https://www.home-assistant.io/) installation.

Another use may be something like a device that warns if a locked, remote, storage area is accessed. In fact, almost any alarm system that lacks a cellular backup could add one pretty cheaply with this.


Here's an idea about how to start with the mailbox sensor: https://www.twilio.com/blog/iot-mailbox-sensor-m2m-to-sms-fu.... This one sends an SMS when triggered, but you could certainly add it into Home Assistant too.


LoRa WAN and such tech can be very useful for this.


Thanks! I hadn't seen this!


Now that I know this exists, I'm going to use it as a bag tracker to tell me where my baggage went when I travel. I'll probably just get one and use it as my car tracker to tell me when to move for street parking. Or maybe I can just inactive it.


How will you deal with charging? As far as I know connectivity has been a solved problem for over a decade now (on a small scale you could always just get a consumer SIM and use that), but power consumption is a problem. This is why Tile trackers for example use BLE tags and nearby phones instead of just embedding a SIM in their trackers.


The baggage only needs an infrequent duty cycle, right? Like once an hour should do the trick. I don't think I have to attach a large enough battery that I need to not put it in luggage. If it doesn't work out it doesn't.

For the car I have a ciggy socket.


https://www.eatabit.io is built on it. We were one of their first beta clients. We used every M2M carrier previously and Twilio is easily superior in cost & reliability. The cellular data space has alot of problems especially with roaming: cell phones transparently switch between networks but data SIMs have a very limited set of networks they can operate on. If Twilio can get VZW access on these SIMs they will have a very strong product. As it stands, coverage is limited.


I’m interested to hear how the multi-IMSI setup performs globally and how the devices handle the SIM profile switches.


Curious as to how does this compare to "Programmable Wireless" and whether it supports calls and SMS? It's not clear whether it's an evolution of Programmable Wireless or something totally separate. I tend towards the latter because the documentation is separate between both offerings.

Last time I tried Programmable Wireless it was very lacking. Calls were not supported on roaming at all (so I couldn't test it as I'm outside the US) and SMS was very flaky; some texts outright bypassed the SMS to API routing and were handled by the upstream carrier directly.

There was also a "bug" where you couldn't send SMS to the SIM using an arbitrary originator number. I say bug in quotes because I'm sure there is some obscure reason for it (or is it to cover a limitation?) because the error code returned was custom and seemed explicitly made for this reason (and was not mentioned on any docs).

I understand it was in early beta when I tried it but none of these issues were resolved even years later.


Hi there, Matt Jackson from Twilio here. I’m the Product Manager for Super SIM. It’s great to hear that you tested out Programmable Wireless and your feedback is appreciated.

_How does Super SIM compare to Programmable Wireless?_

Super SIM is a separate offering from Programmable Wireless. Our first IoT connectivity solution, Programmable Wireless, was developed in partnership with T-Mobile, allowing us to connect to T-Mobile’s global partner network and run on top of T-Mobile’s mobile core infrastructure. For Super SIM, we developed our own cloud scale mobile core that allows us to connect with multiple partners, offering a comprehensive, guaranteed list of tier-1 networks for your devices to connect to. Moreover, with Super SIM, we’re able to extend control to you as the developer by letting you choose which networks your Super SIMs can and cannot connect to. This is really important for IoT use cases where your hardware may not be compatible with all of the networks that are available in a country. While we offer a lot of our networks at the same price, there are some that may be more costly but can offer you more redundant coverage or coverage in more remote regions. You can choose which networks work best for your use case and your customers and take control of your connectivity.

_Does Super SIM support voice calling and SMS?_

Super SIM was designed with IoT use cases in mind that primarily use data. You can use our SMS Commands API to send machine-to-machine SMS between your device and your cloud but Super SIM does not support sending or receiving SMS from other phone numbers. Super SIM does not support traditional calling such as with a smartphone’s native dialer.

_Why can’t I send SMS with an arbitrary originator number with Programmable Wireless SIMs? Is this a bug?_

This is the intended behavior. When you send a SMS from a phone’s messaging apps with Programmable Wireless SIMs that message gets handled by Twilio’s Programmable SMS APIs. You cannot set the from number on those messages to a phone number that is not a Twilio phone number that you own. This prevents number spoofing which while it has valid uses, it’s often used for SMS spamming which Twilio takes very seriously so this is a limitation to that feature.


> This is the intended behavior.

You are misunderstanding. I am not talking about sending SMSs from the SIM to the outside world. That would indeed open you up to spoofing and all kinds of abuse.

I am talking about the the other way around. I have a message from the outside world (whether received through a Twilio number but handed to my own application for processing so Twilio's context of the original sending number is lost, or from a different source like Slack or Telegram) and want to send it off to a SIM, using an arbitrary sender number to distinguish between conversations. This fails too, despite there not being any obvious abuse that I can think of.


Super SIM is on Twilio’s own core networking stack rather than being a shim on top of T-Mobile. Presumably these issues would be resolved with Super SIM. (I don’t know if they have been yet)


Looks like a killer of all other IoT connectivity solutions (eg, Verizon, AT&T, Particle). Those IoT platforms better find a way to support third-party SIMs, if they don't already.

Source: Consulted AT&T on bringing their IoT Connectivity solution to market. Their huge cell network was the main selling point; the platform and services were add-ons.

Edit: Another commenter mentioned Hologram, which already has a global Sim and therefore a lead start. I haven't watched this space for a few years and forgot about them. Good for them!


Particle has a number of devices that already support third party SIMs (their Electron and Boron lines in particular).

Particle also has partnered with Twilio in the past. I know there was some legacy integration with Twilio SIMs.

Particle's value prop is an IoT platform where cellular connectivity is a component. If anything, this seems like it would be complimentary and not a killer to their value prop.


Thanks for the mention! I'm an employee at Particle, and this is exactly it. I'll be honest and say that dealing with carriers sucks. If you want to bring your IoT device to market, you want to focus on building that device — not negotiating with AT&T or Verizon. We protect our users from that so you can focus on your core competencies and shipping your products.

We leverage our large deployed base of devices for negotiations in a way that you cannot if you are a single IoT business. We've actually had customers use Particle, decide they wanted to do their own thing for cost/complexity reasons, start a relationship with the carriers, and come running back because it was intractable for them.


Valid point. I mean this kills their connectivity offering, which in my AT&T example was the core thing. Sounds like Particle now treats connectivity as a commodity and focuses on the platform. Good decision.


I know the Verizon one does. It just is not on their pages. The platform they bought from Qualcomm to do it was in that same space as Twilio. Been a few years though they may have ripped it out by this point. 10 cents per MB and 5 cents per SMS. Ouch!


@matjaxon - A lot of your comments here are getting deleted. If you're not deleting them yourself, I suspect it's because you're making a lot of comments with a new account and getting caught in the spam filter.

Someone from HN should undelete your comments.


Thanks appreciate the heads up! I reached out to HN and they verified my account. Always ever been a lurker and made an account today. You were spot on about why they were getting caught in the filter.


Viewing this site took 4.5MB with trackers.

That's half a dollar of data. Now I understand that IOT stuff would likely have smaller payloads. But that really looks like a great way to rack up costs with a run away program submitting or retrieving data in a bad loop.


If you imagine "IoT" as "a bunch of Linux boxes pushing around multi-MB JSON payloads over fast network pipes" then yeah, this would be a bad deal.

For datagram-based protocols like CoAP which are explicitly designed to let you control bandwidth usage down to the packet level this should be fine. 1kB of data (which is honestly large for a normal telemetry or sensor payload) is really cheap, even at $0.10 per MB.

Many of the devices and network architectures in this space are made to literally wake up a few times a day and push/pull data, not keep idle connections up and running. Your "runaway loop" would kill the device's battery and cause a bunch of other red flags in such a topology long before it racked up substantial bandwidth charges.



Hi there, Matt Jackson from Twilio here. I’m the Product Manager for Super SIM. With Super SIM you can configure a data limit for your SIMs as a guardrail against potential runaway applications. You can set this to a value as low as 1MB. If you know that your use case doesn’t require a lot of data, such as occasionally sending a little bit of JSON with some sensor readings, you can set this to a low number to keep your costs down in case something does go awry.


I really appreciate Twilio's features that limit unexpected costs. I have a personal Twilio account I use to tinker, and to run a simple phone system for my small business, and I would be super worried about doing that if it was like AWS where the best I could get was alerts. Instead Twilio let's me charge my account upfront and disables it if I hit that cap. It's great peace of mind for smaller users.


Is there any way to contain or firewall the SIMs such that they only have access to specific networks? That seems like the holy grail of avoiding people buying IoT devices for their SIM (e.g. what happened with early kindles and many other devices).


Great question! Super SIM has a feature called Network Access Profiles that lets you pick exactly which cellular networks you want your devices to be able to connect to. A lot of other cellular solutions, Twilio’s other cellular connectivity solution included, give you really rudimentary control such as do you want access to the United States (yes/no) and do you want the rest of the world (yes/no). Network Access Profiles lets you pick exactly which countries and which networks inside those countries you can connect to so you can build access just the networks that you want to.


What I mean more is, once those devices are connected to "the network" is there a way to limit the usage of those SIMs so they could only connect to my systems, and not connect to "the internet" at large.


Seems like that's something you would design into the device itself (or rather, its software), whitelisting allowed domains or IPs.


I've done exactly this before when building an IoT device.


Some companies implement IoT by using protocols other than HTTP. For instance, CoAP is one I'm familiar with which can significantly reduce bandwidth consumption.


MQTT is common.


I wonder if this will be like Programmable Wireless where you can get cheaper data rates for different levels of commit. Their pricing page says "starts at" and doesn't really go into how they handle bulk discount.

If this really is the pricing, then it's definitely IOT focused instead of Programmable Wireless's more general focus.


Hi there, Matt Jackson from Twilio here. I’m the Product Manager for Super SIM. Glad you're familiar with Programmable Wireless, our other cellular connectivity offering!

The “starts at” that we have on our pricing page is for the cheapest priced networks we have in each country. For example, in the UK we have a number of networks that are $0.10 per MB but if you want to use EE it’s $0.20 per MB. With our Network Access Profiles, you can pick exactly which networks you want your Super SIMs to be able to connect to. If EE is the only network that services your area or works best with your hardware and is worth premium to use, you can enable it for your Super SIMs.

For now, we just have a pay-as-you-go pricing plan for Super SIM whereas Programmable Wireless has a number of different quota plans that let you commit to a minimum monthly spend per SIM in exchange for discounts as a way for you to self service into lower data costs for hungrier devices. Our pay-as-you-go prices are meant to just get you started. We don’t expect any of our developers to scale up their IoT business at these list prices. Once you’re up and running, you can connect with one of our IoT sales specialists to discuss volume discounts. Promise they don’t bite!


Pretty sure you could set limits or alerts on bandwidth usage to avoid that.


Could this be used as a sort of sporadic consumer phone service? For example signing up for services that require phone validation but want to protect your real phone for privacy. Activate the SIM, signup, verify, then deactivate. Repeat if they don't recognize your device or when you need to re-verify. After hardware costs you would be looking at $2/mo any time you needed it.

Most other services I have looked at require a constant subscription. Or they are not a "real" number like google voice that can be detected and rejected. Or when you cancel and re-activate your previous phone number has changed so you can't re-verify without maintaining an active subscription. If the number is assigned to the sim then you can potentially rotate or have multiple numbers as well.


I suspect the phone numbers will still show up as a voip number, which is probably be an issue for sms validation.


There are carriers out there that sell "real" numbers at least here in the UK. I'm sure the same exists elsewhere.


This will be an issue, fraud prevention systems targeted at reducing IRFS aren’t exactly a fan of VOIP numbers.


I have tried the same with their previous offering called "Programmable Wireless" and both voice support was lacking abroad and SMS had an issue where you can't send arbitrary caller IDs for SMSs sent to the SIM (so if you receive an SMS on a Twilio number but process it in your code before deciding whether to send it to a SIM, or receiving an SMS on a non-Twilio number).


Have been using these and they work really well. One thing though is because Twilio is the carrier the internet breakout is via AWS so you will have AWS public IP addresses. We went back to the original Twilio wireless because the breakout is on T-Mobile IP ranges.


Do you know what the latency is and whether they're doing some magic around roaming where the data session is terminated locally?

Their previous offering always went out through T-Mobile in the US which always added an unnecessary round trip across the Atlantic if you were in Europe.


I really don’t know the answer to that. The latency wasn’t noticeable for me but I’m in SF. Breakout was coming from us-east-1. I believe traffic will always go via Twilio anyway but I don’t know if they have a multi region setup.

Interesting data point the “home network” on the SIMs we used was Telefonica.


Pricing:

* Initial cost: $3/SIM

* Monthly cost: $2/SIM (you can inactivate it at any time)

* Data: $0.10 / MB

* SMS: $0.01 / SMS

Looks good for applications that only sends a small amount of data.


How does this compare to buying some Internet with Legs sim cards which offer 200MB per month for free(includes Europe and USA) https://www.ebay.com/itm/10-X-THREE-3-NETWORK-INTERNET-WITH-...


Could a potential use case be soldering these onto consumer electronics and phoning home telemetry even if the user doesn’t connect to WiFi?


Yes. So long as you have a subscription service (or are willing to eat the cost). This is how many home security systems handle things these days. No need for a phone line (which can be cut) or wifi (which is lost with a power outage). A control panel and the sensors have batteries that continue operating despite power loss and network outages.


A powerful use case here is at-home medical devices (think blood glucose monitors). With a SIM and cellular service, they can work anywhere, be agnostic to consumer WiFi issues (oops, someone changed the WiFi password!), and give "real-time" updates to a doctor or other health professional. Getting patient data has a huge barrier today and there is a ton of room for improvement, leaving the door open to better/more proactive care.


Potentially yes. This is just a SIM card, so one would additionally need to have a modem, consistent power source and space for all of these extra items. The value of the telemetry would need to be worth several dollars / month over several years to justify this.


Is this intended for consumers or corporations? Could someone clarify the use-cases? It looks like it's $100+ for 1GB of data


Hi there, Matt Jackson from Twilio here. I’m the Product Manager for Super SIM, and I can try to answer these questions:

_Is this intended for consumers or corporations?_

It's for developers! We try not to look at our offerings as being strictly for consumers or corporations, but rather for the developers who use them, regardless of where they work. We want any developer with an IoT idea to be able to pursue it, whether their team is one person or one hundred. (This is a common philosophy at Twilio and was the thinking behind offering pay-as-you-go pricing on nearly all of our products.)

_Could someone clarify the use-cases?_

Super SIM is designed for cellular IoT, where we often see use cases like fleet tracking, industrial field monitoring, or micromobility. But there are a lot of use cases for cellular, with new ones showing up every day. An interesting trend we’ve seen is the use of cellular connectivity for things like point-of-sale systems in places where you normally expect there to be Wi-Fi. By using cellular connectivity, these point-of-sale platforms are able to eliminate the variability that comes with each different customer’s Wi-Fi setup, offering their customers a much better out-of-the-box experience because their customers can simply turn on the system and get connected.

_It looks like it's $100+ for 1GB of data_

Our data prices start at $0.10 per MB but this pricing is meant to just get you started. We don’t expect any of our developers to scale up their IoT business at these list prices. Once you’re up and running, you can connect with one of our IoT sales specialists to discuss volume discounts.


I imagine it's best-suited for corporations making devices that can be deployed to arbitrary countries. It looks like you're paying an enormous premium on data for the ability for this sim to "just work" internationally.

Another use case might be for tinkerers creating very low-bandwidth applications (e.g. a few megabytes/texts per month).

For anyone else I imagine it makes more sense just to buy a prepaid SIM and swap it out as-needed.


IoT devices don't use SIM cards for gigabytes of data; they use them for bytes and kilobytes of data. That's why this is billed by the byte.


This is likely aimed at $device_manufacturer that wants to put a cellular modem into their device for remote telemetry (uploading a tiny payload every 24 hours for the lifetime on the device). There's a lot of attention in the docs towards "fleets" — they aren't expecting you to just buy one.


This is not intended for direct use by consumers. This is intended for devices that require internet connectivity, but might not have reliable access to a WiFi or wired network.

Examples include: fleet management vehicle trackers, meteorology sensors, centrally controllable streetlamps

These are commonly referred to as Internet of Things (IoT) applications.

These devices typically use a very small amount of bandwidth (relative to a mobile phone user) but need the reliability of being able to roam across multiple networks. Hence the relatively steep $100/GB.


Surfroam is 10x times cheaper for data! Is 0.01/MB https://surfroam.com/pages/rates No fancy online platform. But is a LOT cheaper. The card is 15€ but has 15€ of data!


@Dang, I am seeing lots of valuable replies from @matjaxon from Twilio being shown as [dead]. I presume all these replies wont show up normally as most have their HN settings as default.

I guess this is happening because it is a new account. ( Likely to combat Spam ).



Hi there, Matt Jackson from Twilio here. I’m the Product Manager for Super SIM. You’re right, we currently don’t have support for Verizon but we hope to be able to add it in the future. We understand that different developers have different connectivity preferences, so we designed our system to be flexible to adding new partners to the platform. As we continue to add networks, the networks of the future should become available to Super SIMs bought today.

With Super SIM you can also have access to multiple networks at once. For example, if you’re connected to T-Mobile and move to an area where there isn’t coverage, your device can automatically switch over to AT&T. This gives you redundant connectivity too in case any one network or tower goes down.


Matt - off topic, but since I have someone from Twilio here ...

Please, please, for the love of god, please add an "email" verb to the twiml language.

You can make it super restrictive if you need to - like you can only email addresses that you can prove you own ... or something.

All I want is to be able to cc: an email with certain SMS alerts and it takes a sendgrid account and write some functions and validate email @ sendgrid with real names and addresses and on and on and on ... just to send an alert to an email address.


Hi there! That's a bit out of my wheelhouse but I shared your message with some Twilio Sendgrid email folks to take a look.


Vouched for Matt’s reply to overcome HN spam filter.

Matt, I’m not sure what the spam filter is latching in to, but you may want to try varying your opening line a little bit for each response.


Are there any geo areas where ATT+TMO/Sprint (combined due to merger) isn't at parity with VZ?


I live in a mountainous area, and there is a very large difference. So much so that VZ is basically my only option.

I think what happens is that tower leases in some protected areas can be hard to negotiate, so the cost/benefit ratio gets out of whack for new towers (amongst other issues with power, construction, and RF propagation in remote, mountain areas). Maybe someone with more knowledge can chime in.


Towers are typically owned by a REIT (real estate investment trust) or other entity that develops the tower and telco/equipment shelter and then leases out space on the tower to mobile providers (cost is higher to the mobile provider the higher you want to be on the tower).

In some locations, there just isn't any land to site the tower due to regs (national parks comes to mind, nothing except my Iridium hotspot works in parts of the Smokey Mountains), so there are no providers or a single provider who was able to swing a special case on an existing tower (water tower or similar).


Yes, many. Verizon is still the only option for reception in many areas—I live in Los Altos Hills, and the only provider with reception for much of the town is Verizon.


What's the difference between an SMS command and just sending data? It says they charge 10 cents for a Mb but 5 cents for a data command. Can I just send a command that is less than a Mb and save money?


Twilio Developer Evangelist here. Our Super SIM Product Manager @matjaxon had his comments held up since he has a new HN account so I'm passing this along from him: "You could send the same information to and from your devices using either SMS Commands or data but it would likely cost you a lot more to use SMS Commands. While our prices start at $0.10 per MB, we bill you in bytes.

For example, if each time your device checks in you exchange 50 KB of data, you could do that 20 times for $0.10. You can only communicate with your device twice using SMS Commands at $0.05 per SMS Command. SMS Commands are often used as a way of configuring some IoT hardware solutions where you can set configuration values over SMS or to communicate with your device if it seems like something is going wrong with its data connection. It's an extra way of interacting with your device that may be deployed into the field thousands of mile away from you."


Ah ok, so it's not a thing you'd use for constantly sending commands.


It appears they charge by the byte, so it'd only be efficient to use SMS if the payload is greater than 512KB


SMS payload is 140 bytes. (160 characters with 7-bit characters)


It is interesting to see how many telcos have failed at building IoT connectivity products that work. If you work for one of these, feel free to reach out.

We will be open sourcing a server (written in Go with a Vue frontend) that can run on a regular private APN to provide developers with self service management and connectivity to their NB-IoT and LTE-M devices.

If you want to have a look around you can take a peek at a beta of the software running at https://nbiot.engineering/


A couple of suggestions for the landing page:

1) Explain what the product is, at the top of the page.

2) Remove the distracting animated background.

3) Switch out the graphic for "Shop" from a woman holding bags to a non-gendered icon. The page currently has three men working and one woman shopping. Some may think it portrays sexist attitudes.


btw, why do you say "woman"? What if that is a person with non-binary gender identity?


Thanks for the feedback. I’m afraid we spent more time on the pages behind the login page and the backend code than the landing page. This is a useful reminder that we have to take more care when designing the new landing page.


Think cargo ship containers and other similar internationally transient use cases. Sounds like an exciting product/service. Looking forward to seeing how this is used.


Trucks, containers, ships, cars...

But also anything that is centrally manufactured or managed but distributed globally. Imagine being able to buy and manage SIMs from one place regardless of whether your [connected thing] will be deployed in United States or in Timor-Leste.


What type of IoT device would I install this SIM in? Are there SBCs that come with a place to plug in a SIM and everything else needed to run it? I thought all of those types of things were made for phones.

Edit: After some googling, found out about cellular IoT and GSM/GPRS modules for things like Arduino. Will that work with these Super SIMs?


Yep Arduino's or RaspberryPis are solid platforms to get started with. We have a couple quickstarts that you can checkout.

https://www.twilio.com/blog/super-sim-quickstart-adafruit-fo...

https://www.twilio.com/blog/super-sim-quickstart-arduino-mkr...


Just to confirm, but a GSM module won’t work for AT&T (but will for the time being for T-Mobile) in the US, right? To my knowledge, AT&T has shut down most (all?) of their GSM network and is only LTE (in its various flavors).


Just to circle back here, seems the official date of AT&T shutting down their 2G (GSM/GPRS) happened in Jan 2017. Their 3G (HSDPA/UMTS) network is tentatively scheduled for about 1.5yrs from now.

So my suggestion, if you’re building anything IoT like on AT&T, make sure you are using LTE-A, LTE-M, or NB-IoT for your radios or you’ll have a short operational life, if at all.


Yes ... I work for Twilio as an IoT SE - there are a lot of great and inexpensive modules that you can use. Stay away from anything 2G/3G as they are being shut down as bands are being reallocated. Also, avoid the "quad band" solutions that are advertised.

I usually buy SIMCom modules from Ali Express. The SIM7000A or G, SIM7070G. The A is usually North American bands, and the G is global bands. There are E versions for Europe. These are CAT-M1 and NB-IoT modules, and I usually use the CAT-M.

NB-IoT is a great idea, but depending on where in the world you are it might not exist. Also, Super SIM does not currently support NB-IoT. We have a different T-Mobile SIM for that. If you really want to go NB-IoT, then the SIM7020{x} is a good module to develop with, or the Quectel BC66.

There are also some great projects now running code directly on the cellular modules. Search github for Wiz-IO. Georgi is brilliant in what he's doing, and supports numerous languages running directly on the Quectel modules.


I’ve been using Sierra Wireless Cat-M1 modules for the projects I’ve been prototyping as they include an ARM Cortex A7 and GPS too in a single package. Bit expensive on the BOM, but convenience makes it worth it. Will test them with this once I get a chance.


Apologies if this is a dumb question - can I use this on a person phone for consumer use cases (voice/text/data)?


Hi there, Matt Jackson from Twilio here. I’m the Product Manager for Super SIM. This is definitely not a dumb question. In fact, all of us on the Super SIM team have really enjoyed reading this thread and learning what you have questions about!

Super SIMs can be put into any device that has a SIM card slot, including almost any cell phone, but they are not designed to be a replacement for your cell phone’s SIM card. Super SIM is specialized for IoT: it doesn't support traditional circuit network components used for traditional voice calling or SMS so these won't work like you expect. However, any apps that use data instead for chat or VoIP calling, such as iMessage or WhatsApp, will work. Overall, we wouldn't recommend using your Super SIM as your regular SIM for your consumer cell phone, although some people at Twilio have put Twilio SIMs into their phones to see what this is like :)


Last time I tried Programmable Wireless, iMessage explicitly didn't work; the activation text (sent in the background to Apple to verify your number and activate iMessage) was being swallowed by the underlying carrier and ended up verifying the internal number of the SIM (a T-Mobile number) instead of being delivered to me through HTTP (so I could route it myself through the Twilio number I wanted to use as outgoing).

I have tried raising it (as it was mostly an oversight and could've been sorted by just asking T-Mobile to make changes to their config) but my impression was that nobody even cared looking into it and just declared "iMessage is unsupported".


You can use iMessage if you use it with only data but that does limit your ability to only using iMessage with other iMessage users. When you put the Programmable Wireless SIM in your phone you'll be prompted to associate that number with iMessage, if you deny it, then you it won't pick up that number on the SIM with iMessage. I personally use this solution when traveling out of the country with a Programmable Wireless SIM in my iPhone. It "works" but unfortunately in a limited manner (i.e. data only).


> Wireless SIM in your phone you'll be prompted to associate that number with iMessage

That was my problem with it. The “number” it will associate is some internal T-Mobile number and not the Twilio number you would have assigned in the Twilio console.


> it doesn't support traditional circuit network components used for traditional voice calling or SMS so these won't work like you expect.

Does this mean that the number associated with the Super SIM would be classified as VOIP and rejected by certain systems that require a "real" number. For example using the new SIM in any phone number verification system that rejects google voice and other providers.


Good question. Super SIMs actually don’t have a phone number that is publicly addressable. You could use a Twilio phone number and our voice SDK to wire up VoIP calling to and from a device; however, Twilio phone numbers will be treated as VoIP numbers and can often not be used for verification systems.


> Twilio phone numbers will be treated as VoIP numbers and can often not be used for verification systems.

Hi Mat, obviously you aren't responsible for Programmable Wireless but I expect you probably work with the same folks. It would be really nice be able to identify a given number as wireless. I've been using Twilio as my everyday wireless provider for a couple years now (not that it's the typical use case or that I'm your typical customer) and this is one of the few issues I have with it and the only one that really impacts personal use. In general, you guys have created an incredibly powerful and flexible platform, thanks for all your hard work.


Yes but the data is crazy expensive for that use case.

Companies that offer this aggregate connectivity through carriers across many countries which can vary widely in price, so they have to price in order to protect themselves in case someone only uses it in (insert most expensive region).

So you're paying premium for it to work many places. And it's intended for small data chunks that would be sent by things like IOT sensor clusters.

If you know you're only going to use it in (these N regions) you could possibly work with them to get a better rate and have them blacklist your traffic in the more expensive regions.


you can use their normal programmable wireless for that. I did and pay $1/month to have a sim sit in a $30 Nokia dumbphone that the kids can use when they ride their bikes to the park.

There are ... quirks. Turns out the SIM cards have a "real" phone number that you don't know about, so even though you're routing calls and SMS through your twilio number and outbound dials only show as the twilio number, you will still get random calls that don't hit the twilio network. My sim's "real" number is a detroit number, and I used to get a handful of collections calls on it from the previous occupant of that number.


You would probably rack up hundreds or thousands of dollars per month in data charges.


Do these services just support 1 or 2 sims? I can think of a few hacky projects to use these for. I think T-Mobile was going to start offering NB-IOT starting at $5 but I don't know how many minimum units or other stipulations they have.


Hi there, Matt Jackson from Twilio here. I’m the Product Manager for Super SIM.

Our other cellular connectivity solution, Programmable Wireless, which we developed in partnership with T-Mobile, has a NB-IoT SIM that you can purchase from us if you’re interested in checking out T-Mobile’s narrowband network. If you don’t have any hardware, you can purchase a Developer Kit for the T-Mobile USA NB-IoT network from our Console which comes with some hardware and a NB-IoT SIM.

https://www.twilio.com/wireless/narrowband


Do you support accounts with 1 sim?


We don't have any minimums on accounts but there are normally a minimum number of SIM's come in an order depending on which item you're looking to get.

The 2 flavors of Programmable Wireless SIMs, the normal one and the NB-IoT one, can be bought in "Starter Packs" that come with 3 SIMs that can be registered to any Twilio accounts. They don't have to be used with the same account so you can share them if you know anyone else interested in checking it out. As you scale up your solution you can get SIMs in orders as few as 10 that come preregistered to your account so you don't have to type in the registration codes for each one.

If you need hardware to get started with you can buy the Developer Kit for the T-Mobile USA NB-IoT network from Console that comes with some hardware and 1 NB-IoT SIM that you can register to your Twilio account.

For Super SIM, because we're in Beta, you can buy as few as 1 from our Console. That will come pre-registered to your account. Super SIM doesn't support Narrowband IoT but it does support an array of networks that do have LTE-M such as AT&T if you're interested in a similar but more widely adopted low power technology.


Looks great, we're using a similar product for building a massive guerrilla network of solar powered IOT devices in strategic locations collecting some alternative data for hedge funds.


Great service, a lot of ideas for usage.

Does anybody use Twilio for banking 2FA? Many banks still use SMS and SIM card hijacking is easy nowadays.


Any examples of getting one of these plugged into a Pi and having a script listen for SMS events and respond to them?



What hardware would you recommend?


Arduino's or RaspberryPis are solid platforms to get started with. We have a couple quickstarts that you can checkout too.

https://www.twilio.com/blog/super-sim-quickstart-adafruit-fo...

https://www.twilio.com/blog/super-sim-quickstart-arduino-mkr...


RasPi or Arduino


So this does not provide the wireless stack just the sim. But with software SIMs these days why is a piece hardware needed? Is it a bit late to the party? Particularly for IoT devices, given the wireless stack is anyway needed along with the RF, it makes sense to cut out other SIM in favor of software.


What do you mean by software SIM? I haven't heard or seen any such modules. eSIMs exist but still hardware. Maybe something based on SDRs can emulate them.


So, it's a VPN for cellular data, right?


How did you arrive at that?

It's a SIM to enable IoT devices to connect to (almost) any available network. It is not cloaking the origin of the data, or even if it is, that's not why companies will buy this.


That's what a virtual private network is - an overlay on transport networks to allow a group of devices under the same ownership to communicate. While VPNs are often used to cloak IP addresses, they were originally a corporate thing. Your in-house net could be transported over the public Internet while keeping its in-house address space and some degree of privacy.


So, human beings are basically apes, right?




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

Search: