The Amstrad E-m@iler, The Right Product With The Wrong Business Model

One of the joys of the UK’s Electromagnetic Field hacker camp lies in the junk table, where trash turns to treasure in the blink of an eye. This year I returned relatively unscathed from my few days rifling through the tables,but I did snag a few pieces. One of them is a wired telephone, which would be a fairly unremarkable find were it not for its flip-up LCD screen and QWERTY keyboard.

My prize is a 2002 Amstrad E-m@iler Plus, one of a series of internet-equipped telephones from the British budget electronics company. The device itself and the story behind it make for a fascinating tale of a dotcom-era Internet flop, and a piece of hardware that could almost tempt today’s hackers.

You’ve Heard Of The Dotcom Boom, But Have You Heard Of The Hardware?

In the late 1990s, everything was about the Internet, but seemingly few outside the kind of people who read Hackaday really understood what it was really about. I’ve written before on these page about how hype blinded the CD-ROM industry to the shortcomings of its technology, but while that had in reality only gripped the publishing business, the Internet hype which followed had everyone in its thrall. You’re probably familiar with the story of the dotcom boom and crash as startup companies raised millions on shaky foundations before folding when they couldn’t deliver, but in parallel with that there was also a parallel world for hardware. The future was going to be connected, but on what and whose hardware would that connection happen?

The Amstrad E-mailer on a desk.
This was considered stylish back in 2002.

It was clear that bulky desktop PCs weren’t exactly convenient, and the mobile phone manufacturers wouldn’t figure out the potential of their nascent smartphones for another decade, so there was a brief period when a host of internet browsing appliances attempted to grab the market. These were usually either set-top boxes or all-in-one stripped-down PCs, and were often sold through consumer electronics outlets rather than computer stores. Among them were even a few unexpected outings for familiar operating systems, for example the BeOS-powered Sony eVilla with its portrait CRT, or the Bush set-top box which would become sought-after among Acorn Archimedes fans because it ran RiscOS.

Many large consumer electronics companies dipped their toe into this market, and in the UK it caught the attention of Sir Alan Sugar’s Amstrad. At the time they were a huge name in consumer electronics in these isles, having specialised in the flashy-but-budget end of the sector. Their existing computer business had equipped Brits with CP/M word processors and proprietary PC clones for years, and it’s fair to say that Sir Alan had a keen eye for what was likely to work with consumers. The E-m@iler was a standard wired phone with an internet appliance built-in, at first as the name suggests as an email terminal, and then with later models featuring a web browser, and later on a camera for video calling. It was for its time a very well-thought-out product that was perfect for older or less technically-inclined  customers, and when it was launched in 2000 it caused quite a bit of excitement.

Such A Good Product, Shame About The Business Model

If the E-m@iler was a great piece of hardware for its time then, it had an Achilles’ heel in its business model. Sold at a loss, it relied on a subscription-based ISP with a premium-rate phone line and a per-email charge. It couldn’t be used with any ISP other than Amstrad’s Amserve, and thus although it won a small and loyal customer base, it hardly flew off the shelves. By the end they could be picked up in the UK supermarket Tesco for under £20, and despite those new models appearing, by 2010 they were gone. In a later interview Sir Alan revealed that the product did break even from the user base it had gathered, but was nowhere near a runaway success.

At the time I thought that the E-m@iler was a great idea well-executed far as the hardware went, even though its business model made no sense. A friend of mine’s elderly father had one, and it was easy for him to use it to keep in touch. Looking at mine in front of me today I still like it, but I want to know more. Time for a closer look.

Peering Inside, Is It Hackable?

the emailer mainboard, a green printed circuit board covered in surface mount components.
The mainboard is dominated by the Connexant telephony chipset.

My device appears to be unused as it still has protective plastic over its handset, but sadly try as I might I couldn’t get it to power up. Cracking it open I find its motherboard, centred around a Sharp ARM processor and a Connexant phone line and modem chipset. On the back there are serial and USB sockets as well as the modem, and though it’s not populated there’s also space for a parallel printer port. Meanwhile on the sides of the unit are a smart media card slot, a smartcard slot driven by a daughter board, and what looks like another serial connection for Amstrad’s pocket databank range.

Digging around online for pages from two decades ago I find that the later version with the video camera had a TI OMAP main processor and ran MontaVista Linux under the hood, but sadly there’s much less info to be found for my second-generation one. A TV advert on YouTube reveals it running Microsoft Mobile Internet Explorer so it’s either Windows CE or something proprietary underneath all the Amserve interface, thus perhaps it’s not as hackable as I’d hoped. Still, it’s not without interesting possibilities, as that slide-out keyboard has a PS/2 interface.

Here in 2024 the idea of accessing the Internet through a wired phone seems laughable. Indeed the wired phone itself has become an endangered breed in many places. But a device like the E-m@ailer made sense two decades ago, for which I will give Sir Alan his due: I believe he was right. It’s sad then that it completely missed its opportunity in a rare case of Amstrad getting it so wrong on the subscription model. Now it’s little more than a paperweight, but I can’t help looking at it with speculative eyes. Should I put it back on the swap table at the next event I go to, or should I bring it back to life with a decent screen and a Raspberry Pi?

29 thoughts on “The Amstrad E-m@iler, The Right Product With The Wrong Business Model

  1. I have one in a box waiting for something weird to happen to it. You can run Linux on them; the OMAP is a bit underpowered but I remember having X running on them back in the day, and the debugging port is relatively well-understood. Amstrad sold them below cost, expecting to make it back on the service charges, so they really didn’t want people running their own software on them, and eventually they changed the boot loader to require authentication. The private key has since been long lost.

    There were three, the E1, E2 and E3, the last of which had a colour screen and a camera for videophone capability. I think the one in the picture is the E2. The E3 is the one with the secured boot loader — some work, some don’t. See for more information.

    The E3, BTW, was powerful enough to run a Spectrum emulator, and of course Amstrad owned all the Sinclair IP at that point; so they had a catalogue of about 40 games you could download and run, and play with the world’s worst gamepad.

    1. There was a prototype keyboard made during the design phase of the original E-m@ailer that mimicked the rubber-keyed 48K Spectrum.

      Maybe if they’d gone with that for production it would’ve had a better reception from the geeks :-D

    2. Hm. I’ve always had the opinion that Spectrum games looked best in black and white.
      The colors were so garish, CGA-like. In black/white, those games had looked higher def to me.
      In that hindsight the mono LC display probably was like an upgrade.

  2. Amstrad also developed a ‘TV browser’ very similar to the Bush box, but in the form factor of a keyboard (foolish idea, as it was tethered to the TV…). Sir Alan reputedly nixed it at pretty much the last moment.

  3. Amstrad outsourced all the development for the Emailer to a firm whose name I forget, based in Lewes IIRC, and STNC in Cambridge/Bury St. Edmunds.

    The Lewes folk designed the hardware, and wrote most (/all?) of the device drivers, and STNC did all the rest of the software. Some of my code is in the IP stack on it, I wasn’t part of the main team working on it, I was Software Manager in charge of all the engineering teams at the time.

    We (STNC) had been bought by Microsoft by the time it came out, which is why it shipped as Microsoft Mobile Explorer (MME). The “OS” is the underpinnings of the browser – MME was super portable, it could use OS features (like a memory manager :-) if they were available, but could also run on bare metal. On the Emailer it’s nearer to something like Arduino than anything like WinCE.

    1. Creative Technology in Uttoxeter did some of the development for it (presumably subcontracted) and it was through them that I designed the fonts for the Emailer. The only font work I ever got properly paid for.

  4. When these came out they didn’t sell hugely well. But when they sold them off really cheap they became hugely popular. Currys/Dixons (electronics shops) were shady gits though and didn’t check with customers if they were using it with a BT line as Amstrad had a deal only with BT… I worked for a cable company and probably spoke to a good few hundred people who suddenly got massive bills due to the stupid Amstrad emailer phone dialling out all the time to pull ads and retrieve email.

  5. Hm. I don’t get it. So this is a cross between a 1980s Minitel and an early 90s ISDN picture phone? 🤨
    What’s so special about that? And why does it need the internet?
    Wasn’t ISDN a more mature alternative?

        1. In Germany of ~1990, deployment of landlines was generally slow (analog/digital).

          The re-union thing caused extra work, because the east had a very bad infrastructure.

          Really, really bad. Noise, crackles etc.
          It was too bad to provide slow 1200 Baud/75 Baud connections as needed by BTX service (the German Minitel service, essentially).

          To give an idea, the ancient DBT-03 modem from the 80s had used this speed and was made from 1980 technology.

          I’d say that ISDN had a positive effect here, even.
          If the infrastructure had to be deployed in analog form back in the 90s, it may have taken way longer.

          Because you can’t really multiplex analog on a single pair of wires.
          Not with the given bandwidth of a typical twisted pair line.

          By deploying ISDN and fibre glass, much more simultaneous connections could be maintained.

          So even if ISDN didn’t make it into the households, digital connections were largely being deployed between the telephone exchanges.

          By late 90s, most gray telephone boxes on the street corner had bern digital. It was merely the last mile to the house that was still analog.

          Using the huge, old analog cabins of rotary switches would never had sufficed here.

          But yeah, ISDN wasn’t very successful among private users.
          Maybe also because of cost and initial non-availability.

          Personally, I’ve never been a die hard “fan” either, even though I found the technology fascinating.

          Most PC users did instead keep using telephone modems, which in practice did merely reach a quarter of the rated speeds.

          Again, it was the analog phone quality of the landline that decided about connection speed.

          The ~6KHz bandwidth (?) simply wasn’t much. It was comparable to medium wave broadcast radio.

          ISDN rather was like a PCs serial port (COM port, 115 200 Baud max) by comparison.
          Of course that’s very slow by today’s standard, but still faster than a telephone modem.

          Anyway, I think it’s just a bit sad that the whole VOIP stuff these days is less being thought through than ISDN from the 1980s was.

          ISDN had almost instant connections between two callers, it allowed group conferences, displayed the caller ID, it had digital FAX, the video phones could do video telephony without special software or an ISP etc.

          It also allowed direct access to X.25 networks and the ISDN PBX boxes at home or in an office could be used to control various relay contacts (used for gate opener) etc.

          All you needed was an ordinary stationary phone (or a DECT handset if you will) and the phone’s keypad.
          No app was needed, no cloud. No viruses.

          (The German national ISDN also had offered a permanent connection, which the Euro-ISDN didn’t have, bzw.)

          It was so much of an cleaner approach, all in all.
          It was very direct, like the analog landline was.

          That’s also why users in Germany are right now so sad that the latest Fritzbox router (7690?) drops support for the S0 bus. It simply was a neat thing.

          Nowadays it needs GSM/2G modules to perform such similar remote control tasks (machine control; use in elevators etc), because the internet just doesn’t cut it. It’s not reliable.

      1. I see, I didn’t know this. My bad. I had naively assumed that UK was similar to us. 😅

        Anyway, we did have BTX terminals, the Multitels before ISDN was around.

        A Multitel was like French Minitel terminal, essentially. A phone with a built-in monochrome CRT.

        Some Multitel models had a built-in 1200 Baud modem, some needed the infamous DBT-03 modem.

        Even in mid-80s, it was possible to send “messages” between users with these devices.

        Other users, ordinary citizen, could buy BTX decoders for their TV set.

        BTX was similar to the technology used by Videotext/Teletext (both used CEPT glyphs).
        In the UK you called this Ceefax, I think?

        Anyway, that’s why these decoders were affordable, I suppose.
        The technology didn’t need much resources to work. Low bandwidth and little computing power.

        All in all, the concept was not different to Internet set-top boxes of late 90s.
        Such boxes already existed in the 1980s in more simplifed form.
        The Canadians had their NABU set-top boxes, essentially.

        Also interesting is that the underlying technology came from the UK, essentially:
        Both Minitel and BTX and other Videotex (no t) systems were based on British Prestel system. 😃

    1. I don’t know if that’s useful context.
      Amstrad was founded in the 60s by Alan Michael Sugar as AMS Trading, they always made low cost tech products. They started out as cheap stereos and TVs and moved into home computers in the 80s. Their computer lines were pretty successful, but they faded away under the spread of PCs.
      To most people the Amstrad e-mailer was a strange device, like it was trying to recreate the world of 10 years prior.

  6. I worked on the backend systems for this for a major UK ISP.

    Had one on my desk, plugged into the corporate PBX. It had a PBX dial prefix you could configure, and our PBX let you define shortcodes that would call out, so I set the shortcode to a geographic number for the inbound modem bank, and the PBX prefix to call the shortcode. Result: a working emailer without the massive bill.

    My fondest memory is the outbound dialler, which would call your emailer from a known number when you had new mail, that would then cause the emailer to call in and collect it. The emailer wouldn’t ring, but since it would ring any other phone on your line, there was a curfew built in to the system to prevent it callingovernight. One evening, I was testing it, so I went round the office and entered every desk phone number into the call queue, then waited for the curfew, which I’d modified, to come off. The whole place erupted as 60 outbound lines started calling every phone in the office like my own miniature Lawnmower Man ending.

  7. This made a fabulous crown to a costume I created once. I taped it to the top of my bicycle helmet. Really wanted to hack it but never found the time. It’s still waiting in a box for me to find the time :)

Leave a Reply

Please be kind and respectful to help make the comments section excellent. (Comment Policy)

This site uses Akismet to reduce spam. Learn how your comment data is processed.