3 ms·

> The bug was fairly large and consisted of state-of-the-art integrated circuits and single-bit core memory. It was completely hidden inside a hollow support br

> The bug was fairly large and consisted of state-of-the-art integrated circuits and single-bit core memory. It was completely hidden inside a hollow support bracket at the bottom of the keyboard mechanism, and was invisible to the naked eye, but also to the detection equipment of the era. Only an X-ray scan could reveal the presence of the device, which is shown in the image below. It contains special components to hide its presence even from non-linear junction detectors (NLJD). Holy cow. I wouldn't have thought this was possible in the mid-1970s. Also the article says that the NSA didn't trust the CIA or State Department to handle the investigation properly, which was very interesting to read. Had the NSA already been established as a "we do tech better" agency?

hwillis 9y ago
As an electrical engineer (but a new one, so take this with a grain of salt) it really is kind of staggering. This probably could only have worked with the selectric- they're quite special and if you haven't seen one in slow motion it's a thing of beauty[1]. The bug only had to measure magnetic disturbances in a few dimensions to recover text, and because it can measure the amount of time it takes to locate a new letter the field measuring error is much more acceptable. You wouldn't be able to do that with a more classic typewriter. The level of sophistication here reminds me of current research[2] using software-defined radios. I'd LOVE to see a diagram of this thing. Old circuits were elegant by necessity, and something this sophisticated has to be a work of art. > It contains special components to hide its presence even from non-linear junction detectors (NLJD) That's really amazing. NLJDs work by blasting broadband noise at a circuit- any internal wire trace gets turned into an antenna, and if it's attached to a transistor or diode it'll re-radiate at multiples of the frequency. Typically you can even find just chunks of metal touching a semiconductor with those... the fact that they evaded detection is very impressive. [1]: https://www.youtube.com/watch?v=izZ02t2UEGc https://www.youtube.com/watch?v=izZ02t2UEGc [2]: https://www.youtube.com/watch?v=VZ7Nz942yAY https://www.youtube.com/watch?v=VZ7Nz942yAY
- shriphani 9y ago
  Did these need qwerty keyboards? The mechanism seems to not need the qwerty arrangement - might have been a legacy thing?
  - hwillis 9y ago
    As in, were they fast enough to be used with other layouts? Yeah, but this was back when letters were dictated- secretaries didn't go switching to dvorak. Any typewriter could technically use any layout, since you'd just have to swap out the characters. The biggest advantage of the selectric was that you could swap out the ball to get different characters or fonts. The ability to get italics was a pretty big deal. Here's Bill Hammond talking about them: https://www.youtube.com/watch?v=bRCNenhcvpw https://www.youtube.com/watch?v=bRCNenhcvpw
Theodores 9y ago
You need to check out the electronics that the Soviets had on board the spacecraft after Sputnik. The kit they had to get to the moon, Venus and Mars was quite special in those analogue days, if you only know digital electronics then it is even more impressive. Hence bugging a few American typewriters the Soviet way results in such a cool solution. The task was probably relatively easy compared to space. The space race story we tell ourselves is not entirely true, the Russians didn't have 'better Germans than what the Americans had', actually their rocket science was far ahead of the amateur German efforts and they didn't need help from the Nazis. Meanwhile, in America they pushed their guys to one side and insisted that the Nazi work be what was the basis of the ICBM/space program. It was the American rockets that were relatively crude, not the Soviet ones. So cut them some slack, bespoke electronics with clever radio wave stuff was something that the Soviets were extremely good at, mass production a different story.
- dagenleg 9y ago
  > actually their rocket science was far ahead of the amateur German efforts That's not really what happened. At the end of the war, Germany had the absolute lead in the rocket development, except maybe the rocket artillery. Ever heard of V-2 rocket? It was the most advanced thing at the time. It's true though that USSR didn't use the German talent as much as Americans did, immediately after the war, but I think it's mostly due to the security concerns and ideological reasons. Why have Germans building rockets when you can use brilliant Korolev instead.
- azernik 9y ago
  The Soviet advantage wasn't really even in better rocket science - it was in a greater applicability of military knowledge. In addition to making earlier investments in ICBMs (because of their lack of bases for strategic bombers), the Soviets also couldn't miniaturize their warheads as much and so had to build larger-payload ICBMs to launch them. As long as the space race was all about Earth orbit, where repurposed ICBMs like the Soviet R-7 or the American Atlas and Titan-II were all that was necessary, the Soviets could do more piggybacking on their existing engineering. Once it came to designing a new rocket purely for space exploration, the Soviets didn't have the resources to pull off something like the Saturn V. This advantage played out militarily as well - when MIRVs became practical, Soviet missiles that had been built for the older larger Soviet warheads were capable of throwing more MIRVed warheads than American missiles.
gumby 9y ago
> Also the article says that the NSA didn't trust the CIA or State Department to handle the investigation properly, which was very interesting to read. Had the NSA already been established as a "we do tech better" agency? NSA may have been influenced by Angleton's belief that there was a mole (don't know if he was a "stopped clock" since Aldritch Ames didn't start leaking until the 1980s ARAIK). The more likely explanation is the simple "we learnt of the leak, we developed a remedy, we should just keep the need-to-know circle as small as possible". Once you need to liase with someone else there are meetings, counterparts, assistants, etc...

hwillis 9y ago

As an electrical engineer (but a new one, so take this with a grain of salt) it really is kind of staggering. This probably could only have worked with the selectric- they're quite special and if you haven't seen one in slow motion it's a thing of beauty[1]. The bug only had to measure magnetic disturbances in a few dimensions to recover text, and because it can measure the amount of time it takes to locate a new letter the field measuring error is much more acceptable. You wouldn't be able to do that with a more classic typewriter. The level of sophistication here reminds me of current research[2] using software-defined radios. I'd LOVE to see a diagram of this thing. Old circuits were elegant by necessity, and something this sophisticated has to be a work of art. > It contains special components to hide its presence even from non-linear junction detectors (NLJD) That's really amazing. NLJDs work by blasting broadband noise at a circuit- any internal wire trace gets turned into an antenna, and if it's attached to a transistor or diode it'll re-radiate at multiples of the frequency. Typically you can even find just chunks of metal touching a semiconductor with those... the fact that they evaded detection is very impressive. [1]: https://www.youtube.com/watch?v=izZ02t2UEGc https://www.youtube.com/watch?v=izZ02t2UEGc [2]: https://www.youtube.com/watch?v=VZ7Nz942yAY https://www.youtube.com/watch?v=VZ7Nz942yAY

shriphani 9y ago
Did these need qwerty keyboards? The mechanism seems to not need the qwerty arrangement - might have been a legacy thing?
- hwillis 9y ago
  As in, were they fast enough to be used with other layouts? Yeah, but this was back when letters were dictated- secretaries didn't go switching to dvorak. Any typewriter could technically use any layout, since you'd just have to swap out the characters. The biggest advantage of the selectric was that you could swap out the ball to get different characters or fonts. The ability to get italics was a pretty big deal. Here's Bill Hammond talking about them: https://www.youtube.com/watch?v=bRCNenhcvpw https://www.youtube.com/watch?v=bRCNenhcvpw

Theodores 9y ago

You need to check out the electronics that the Soviets had on board the spacecraft after Sputnik. The kit they had to get to the moon, Venus and Mars was quite special in those analogue days, if you only know digital electronics then it is even more impressive. Hence bugging a few American typewriters the Soviet way results in such a cool solution. The task was probably relatively easy compared to space. The space race story we tell ourselves is not entirely true, the Russians didn't have 'better Germans than what the Americans had', actually their rocket science was far ahead of the amateur German efforts and they didn't need help from the Nazis. Meanwhile, in America they pushed their guys to one side and insisted that the Nazi work be what was the basis of the ICBM/space program. It was the American rockets that were relatively crude, not the Soviet ones. So cut them some slack, bespoke electronics with clever radio wave stuff was something that the Soviets were extremely good at, mass production a different story.

dagenleg 9y ago
> actually their rocket science was far ahead of the amateur German efforts That's not really what happened. At the end of the war, Germany had the absolute lead in the rocket development, except maybe the rocket artillery. Ever heard of V-2 rocket? It was the most advanced thing at the time. It's true though that USSR didn't use the German talent as much as Americans did, immediately after the war, but I think it's mostly due to the security concerns and ideological reasons. Why have Germans building rockets when you can use brilliant Korolev instead.
azernik 9y ago
The Soviet advantage wasn't really even in better rocket science - it was in a greater applicability of military knowledge. In addition to making earlier investments in ICBMs (because of their lack of bases for strategic bombers), the Soviets also couldn't miniaturize their warheads as much and so had to build larger-payload ICBMs to launch them. As long as the space race was all about Earth orbit, where repurposed ICBMs like the Soviet R-7 or the American Atlas and Titan-II were all that was necessary, the Soviets could do more piggybacking on their existing engineering. Once it came to designing a new rocket purely for space exploration, the Soviets didn't have the resources to pull off something like the Saturn V. This advantage played out militarily as well - when MIRVs became practical, Soviet missiles that had been built for the older larger Soviet warheads were capable of throwing more MIRVed warheads than American missiles.

gumby 9y ago

> Also the article says that the NSA didn't trust the CIA or State Department to handle the investigation properly, which was very interesting to read. Had the NSA already been established as a "we do tech better" agency? NSA may have been influenced by Angleton's belief that there was a mole (don't know if he was a "stopped clock" since Aldritch Ames didn't start leaking until the 1980s ARAIK). The more likely explanation is the simple "we learnt of the leak, we developed a remedy, we should just keep the need-to-know circle as small as possible". Once you need to liase with someone else there are meetings, counterparts, assistants, etc...