I had a python script running on a handmedown computer running Ubuntu 8.10 that just periodically checked an email account, parsed the command, flipped the relay as necessary, read the temperature rise, and replied.
We had an actual server in the garage, with a X10 firecracker on one serial port and a temperature sensor on the other. We used a 220V relay with a 110V pull, hooked up to a normal X10 switch controller. So from an X10 remote in the house, or via an sms gateway, we could turn on our hot tub remotely.
This had a few advantages including: 1) this setup was more accurate than the ancient thermal sensor/control on the hot tub 2) we never again had a large electric bill because someone forgot to turn it off and 3) really nice for when a date is going well and you can turn on the hot tub ahead of time :)
I work with HVAC software and so many of our customers want important alarms on their mobile devices. They don't want it in email because nobody has notifications on for email so they want a text and they are aware of the carriers SMS gateway. Five years ago this worked ok but with the rising spam wars these similar alerts we send out get caught up in spam filters easily and customers blame us.
It is very simple to use something like pushover and API alarm delivery to them, and so much more reliable. Still it's shocking to see; wait that requires us to pay pushover? Nah, we'll take our chances with email@SMS for our multi million dollar chiller plant in our hospital alarm notifications.
/spam: my friend runs https://messente.com that specializes in global SMS delivery and they are, honestly, quite good at it. (Let me know if you ever start using them so I can let him buy me a referral-beer at the pub)
So put that on top and you can ignore the remaining 90% of the article.
It doesn't even mention that this method could lead to costs for the recipient.
For instance in germany you have to activate that feature for t-mobile numbers and they charge you for received emails per SMS
The only info I felt was missing is how to retrieve the carrier when given a phone number.
I don’t think security is necessarily the most important thing to these companies.
About a decade back, we used to use the email approach at a non-profit I volunteered at. Annually we would need to send a bunch of texts of a short time period. Think "once a year event".
Eventually we had to stop using that solution. Most texts went through, but there were enough failures that we had to switch to twilio. I can't remember when; I guess 5 or 10 years ago? Getting old.
(We didn't immediately switch to twilio. There was one year during which we used my personal phone to send out an obscene number of texts from my personal phone number. Figured out that was a bad idea pretty quick.)
1. Latency. We really needed messages to go out to everyone all at once.
2. My personal phone became unusable for the rest of the event because people thought that number was a general event organizer contact mechanism. This was a big deal because I was a key member of the operations staff and really needed my phone to be useful.
Solving (1) and (2) was certainly doable, but twilio solved the problem for us and I had higher impact IT stuff to be working on for the org.
I worked with an app that sent weather alerts to subscribers and had used these gateways. It wasn't very high volume, only sent messages during active storms in the subscriber's area, and not more than one every 10 minutes or so. We hardly ever sent more than 100 messages in an hour for all subscribers, even during a heavy storm period. Still we encouraged users to use a push messaging app like pushover which we integrated with, since sms through smtp often didn't deliver or took hours to arrive. We considered setting up a real sms service like twilio but the push messaging apps had better features than sms anyway, so the sms over smtp option was just a hack for users that really wanted sms.
In 2011 they worked pretty well, but by 2018 they were frequently down or didn't work properly. Pretty sure that messages sent through them are much more likely to hit spam filters or get rejected for whatever reason.
The amount of traffic that goes through these gateways probably isn't worth charging for, given the amount of headache it would cause for incidental uses like the above. But it would probably cut down on gateway spam, which seems to have increased lately (at least on AT&T).
In the US the recepient of a mobile call and obviously also an SMS pays. Either per use or as part of a package.
In the rest of the world the recepient pays absolutely nothing, the orginator pays either per use or by package.
So such gateways make sense for US-providers. But no sense in the rest of the world. Very few did always exist during the last 20 years even in the rest of the world. I tried to use them years ago and failure rate was something like 90% or more. If something is too good to be true it's typically not true.
Tracfone targets themselves very narrowly at the "elderly and only want to pay for what I'll use/just want an emergency phone" demographic. If the subscriber is going to use the phone approximately at all, those are garbage rates, but if you only want to keep the lights on, $20/3 per month is about as cheap as it gets.
Do you think their pricing page is incorrect, because it doesn't include the price the recipient might have to pay, for each possible carrier they might be using?
I mean, where do you stop? When you pay your ISP, should the bill show "pushing the button to activate your connection", rather "internet service"? Should coffee shop include the cost of a bottle of water, because drinking coffee will dehydrate you? I don't understand what you expect.
As to the second point, yes - I did mean free for the sender. I assumed that people were aware that the recipient can still be charged. I mean by the same logic, you might as well add the recipient's phone bill (if they don't pay per SMS message individually) and even the cost of their phone if you want to be picky. I don't think anything is inherently free. We just have different interpretations.
At the end of the day, I just thought the idea was cool, so I wanted to share it hoping someone might find it useful.
Still costs money, but if you're sending a lot of messages, this would substantially reduce the costs.
curl -s -X GET "https://lookups.twilio.com/v1/PhoneNumbers/$number?Type=carrier&Type=caller-name" -u $accountsid:$authtoken | /usr/local/bin/jq '.'
Again, thank you for putting this into a post!
I personally do carrier lookups in the unix shell with a 'curl' command using the Twilio API (see my other comment under you).
However, you can just look this information up on a number of different websites:
... et. al ...
She was also hooked on the brand new "Kindle Keyboard 3G", which came with unlimited cellular data so you'd be able to buy ebooks at any time, without wifi. It also had a primative web browser that was capable (just barely) of running the gmail web interface... so voila, it's actually an e-ink tablet, with email and SMS capability! She'd have to manually load the kindle browser to check for messages -- great, really, because who wants to be interrupted with a notification while reading a book?
But after a few semi-urgent texts went unnoticed, I dreamed up a solution: change the Kindle's name based on the number of unread texts. This wouldn't interrupt reading, but it would be visible on the kindle's home screen, a good compromise.
So I hacked up a python/applescript/firefox script running on my mac mini home server. every 5 minutes it would log into her gmail account, check for unread text messages in the gmail inbox, and then log into her amazon account and update the kindle's name from "Wife's Kindle" to "Wife's Kindle (x)". It was beautiful. It worked!
I think with all the technological abundance we have now, we often forget the fun/creativity that limitations can impose on us.
It's meant for trial users to be able to test our software without having to have a subscription or GSM modem.
But those gateways saw / see a lot of abuse. Even though we clearly state that each trial user only has 10 free messages we still regularly get an e-mail from an angry 'trial user' because only 10 messages out their 50K+ phone number mass mailing arrived.
So whenever a pump was misbehaving or broke, we would observe it all on a webpage and a technician would be alerted to fix it.
Later on some ML was strapped to the SMS data to predict failing pumps before they failed.
Good stuff. sms can be very powerful in the right setting.
Don't know what the author means by free here though. Sms is something you pay for, one way or another.
Edit: One funny thing is that iMessage will detect gateway traffic from/to email addresses that are tied to iMessage accounts, and put those into a "blue bubble" chat instead of the normal SMS/MMS "green bubble" chat.
On the receiving side: there isn't any real delivery guarantee for SMS/MMS gateways. You're basically depending on a mostly forgotten and probably neglected service (and probably physical server) at each carrier to receive potentially important information.
Edit: You can also get around the gateway identification problem by spamming every gateway with your message. That worked pretty well when I tried it last (around 5 years ago). But I wouldn't be surprised if continually doing that earns you a hellban from every carrier.