The Poly came out in 1981 or early 1982 and was intended for educational use. Much was made of its facility for networking. It seems that number of Polys could be wired together and kept in sync by the teacher. Although this now sounds pedagogically quaint, a few schools signed up to buy them.
In 1983 a networkable Poly cost $8090, but it appears the complete teaching kit ran into the hundreds of thousands. In response to the Poly, Apple reduced the price for schools of an Apple II from $4812 to $1200. Customs decided that Apple was "dumping", which seems to have
had a precise technical meaning in those days of import regulation. They imposed a duty of $820, bringing the price up to $2020. Apple changed their price to $2020, rather than pay the duty (Customs could not complain, because their duty was defined as setting the fair
Money was valuable then. I recall you could get lollies valued at fractions of a cent. $8090 would get you a largish McCahon or a few houses in Wanaka, so even if Apple had stuck to their original price, the Poly may not have taken off. Nevertheless, Polycorp was blaming
Apple, Customs, and the Government for their impending failure.
By way of comparison, here are prices for other computers advertised in Bits and Bytes, vol 1, nos 1-3 (retail prices, some educational discounts offered but not revealed):
Atari 400 $1295
Atari 800 $2695
Dick Smith 80 $1295
BBC Micro $1595
The Apple II was designed in 1976. It was inferior to the 1982 BBC Micro in every possible regard, despite retailing at three times the price. The IBM PC had been out for a couple of years, and the Mac was due in January 1984. In these circumstances, $1200 actually sounds
like a reasonable price for the Apple II, but of course you don't buy from Apple expecting value or quality.
The Poly used the 6809 processor. This was an advance on the 6502s and z80s that everyone else used, although it was admired as much for conceptual elegance as increased performance. Unfortunately the 6809 was a year or two late for the 8 bit era, and hence was mainly used in
idiosyncratic late machines like the Poly, the Dragon (Welsh) and the Peach (Japanese).
Bits and Bytes mentions some system of layered video, allowing graphics and text on the same screen. 21 colours could be used at once, exactly 5 more than was common.
Its networking system appears to have involved an ad-hoc protocol over a serial bus. As far as I can tell, the Poly system was technically inferior to ethernet, which had had a decade of development, but was was just then becoming widely available, with the first Apple II
ethernet card released in 1982. The reporter didn't seem to understand much about networking.
It seems that what the Poly developers really cared about was software, or rather, a particular mode of software use and development that they imagined would suit schools. They thought that, with the Poly, teachers could easily develop teaching software that could be shared between schools. In 1983 only a tiny portion of the curriculum had been Polyised, but they didn't think the rest was far off.
Polycorp's mistakes are obvious enough. They wasted effort on irrelevant details, fighting rather than harnessing the flow of commodification. If they had resisted the urge to reinvent hardware and had targeted their software at (say) ethernet connected IBM PCs, they might well have got it into a few more schools, and lasted as a company for a few more years. There is something to be said for waiting until the appropriate technology arrives, rather than trying
to force it.
I think this is a general and well learnt lesson of the last 25 years, not hindsight specific to Polycorp. Now most people have a sensible, incremental approach to changing the world with computers. For example, Vicki Smith and Karla Ptacek have had school kids collaborating across the world, using generic hardware, operating systems, and network protocols, with only a little bit of specialised (buggy) software. And now there is Skolelinux which perhaps has similar aims to Polycorp, but actually saves schools money relative to the alternatives, by borrowing as much as possible from existing ideas and technology.