The history of over the wire/air updates and embedded devices

[e100]

From a historical viewpoint has being able to do over the wire/air firmware updates been a feature of embedded devices forever, or only something that appeared in the last 15 years or so?

I'm thinking of those devices that are smart enough to 'phone home', query to see if an update is available, download it and reprogram itself, rather than relatively dumb devices that have to be forced into a special bootloader/update state by a parent device that pushes the update.

[pcprogrammer]

I guess this became possible when FLASH memory was created. You need the ability to erase the memory electrically to reprogram it, and the early embedded systems used PROM's or EPROM's  and needed to swap these with newly programmed ones. In case of the EPROM it was of course possible to erase it, but one needed an ultraviolet light source for it.

The EEPROM's that followed could and can be programmed in the field, but I don't know if these where used for actual firmware.

[Berni]

More that 15 years ago most of these devices didn't have a internet connection in the first place.

At the same time we did need to get cheep flash technology. Sure we had flash way before but firmware storage was still some kind of ROM.

Roughly 15 years ago is also when the PlayStation 3 was released. It had a Ethernet port built in since release and it could download game patches to its internal hard drive.

[Gribo]

In-applicaiton Firmware update became relevant when flash devices became common, late 80's, early 90's. E(E)PROMS were not good enough for this function - usually requiring a technician to update.

[PaulAm]

Automotive modules such as ECMs and BCMs have been updateable for at least 20 years, although only at a dealer using brand specific tools.

Wireless updates have only started showing up in the last 5 years or so.  Particularly with cars the security challenges are significant.  It also really annoys customers when they park their car and an update fails overnight resulting in a bricked vehicle in the morning (and yes, that has happened).

[MT]

From a historical viewpoint has being able to do over the wire/air firmware updates been a feature of embedded devices forever, or only something that appeared in the last 15 years or so?

I'm thinking of those devices that are smart enough to 'phone home', query to see if an update is available, download it and reprogram itself, rather than relatively dumb devices that have to be forced into a special bootloader/update state by a parent device that pushes the update.

There was not much of a "public" internet in the 70 and 80ies to begin with. Chips (MCU's) "phones home" up via modem? Well in the 70 and 80'ies that gives you a whole bag of interesting problems,
one major one is it's not very cost effective on chip level. On product level, inbuilt modem and hoopla? Perhaps, depends on application........maybe for 100 units, but millions of units trying to call?
In 1986 a song was made about "phoning home".

[ace1903]

2008-2009 I worked on software for updating mobile phone from memory card (SD and some Sony MS protocol if I remember well). Update over USB was available and developed much earlier.
Roughly that period some mechanism for updating phones from network existed. Mobile operator was able to push some updates to phone firmware but only on radio and network stack.
We had IP based FOTA but can not remember when it was developed.

There was some company that patented all the mechanism and logics for producing optimized update packages.
Later found that they sued google and Mozilla for their update mechanism.
Probably even today they can sue everyone that makes update over the any medium.

[Berni]

There was some company that patented all the mechanism and logics for producing optimized update packages.
Later found that they sued google and Mozilla for their update mechanism.
Probably even today they can sue everyone that makes update over the any medium.

Yeah that sounds like one of those overly generic bullshit patents that should have never been granted in the first place. Similar case is Red Digital Cinema patenting compressed raw video recording and then suing the pants off everyone. The patent is not about there own RedCode raw format, it is not even about any particular compression method, it is just about the concept of recording video as a sequence of compressed raw frames.

[tooki]

There was not much of a "public" internet in the 70 and 80ies to begin with.

There wasn’t any public internet whatsoever until 1989 (and even then, just barely). The commercial internet as we know it was born in 1991. There had been other online services for a while, but the internet was originally limited to institutions.

[Whales]

Do early computer programs broadcast over radio stations count?  ie instead of using a tape player to load a program into memory you use a radio.  Albeit this isn't "automatic".

[MT]

There was not much of a "public" internet in the 70 and 80ies to begin with.

There wasn’t any public internet whatsoever until 1989 (and even then, just barely). The commercial internet as we know it was born in 1991. There had been other online services for a while, but the internet was originally limited to institutions.

What i said! Geee!

[tooki]

There was not much of a "public" internet in the 70 and 80ies to begin with.

There wasn’t any public internet whatsoever until 1989 (and even then, just barely). The commercial internet as we know it was born in 1991. There had been other online services for a while, but the internet was originally limited to institutions.

What i said! Geee!

”Not much of” implies the existence of some public internet during that period, but there was none at all except for the last few months of 1989, and even then only in one country and with enough limitations that 1991 is often considered the actual birth of the public internet.