4 ms·
Ah, this explains why don't remember seeing Teletext: > Adoption in the United States was hampered due to a lack of a single Teletext standard and consumer res
by aazaa 5y ago
Ah, this explains why don't remember seeing Teletext:
> Adoption in the United States was hampered due to a lack of a single Teletext standard and consumer resistance to the high initial price of Teletext decoders. Throughout the period of analog broadcasting, Teletext or other similar technologies in the US were practically non-existent, with the only technologies resembling such existing in the country being closed captioning, TV Guide On Screen, and Extended Data Services (XDS).
https://en.wikipedia.org/wiki/Teletext https://en.wikipedia.org/wiki/Teletext
- joezydeco 5y agoIn the 1980s the company that owned the Chicago Sun-Times, Field Communications, tried to bootstrap Teletext (the BBC's "CeeFax" branded as "KeyFax") in the USA. Their plan involved broadcasting read-only teletext on their company-owned UHF station after normal broadcasting hours. Take a loving gander: https://www.youtube.com/watch?v=lVpUSaUMgbY https://www.youtube.com/watch?v=lVpUSaUMgbY
- _Nat_ 5y agoJust skimming the Wikipedia page, it sounds like almost all Teletext was read-only, which the entire protocol was designed around. But since you stressed their plans being "read-only", was there something else to it?
- zapzupnz 5y agoRead-only as in KeyFax was literally a broadcast image of Teletext, rather than actual Teletext. That way, people without a compatible set or decoder (everybody) could watch it. That had the knock-on effect of users not being able to call up a page at will; they had to wait until it was broadcast.
- iso1631 5y agoNormal teletext you select a page (typically a 3 digit number from 100 to 999, although hexadecimal pages also existed) Your decoder would then look at the vertical blanking lines with the data on, and wait for the page to match the one you selected. This could be fast (typically pages like 100 would be broadcast every second or so), or slow (page 693 might be sent once every 15 seconds) The decoder would then show it. The decoder would only need a 1 page memory. "Read only" teletext didn't include a decoder at all, it was the output of a decoder at the TV station which was sent via normal UHF means, typically with background music. As memory got cheaper, "Fast Text" was introduced, each page would have shortcuts to 4 more pages (red, green, yellow, blue). These would then be stored in the background when they were broadcast, so the decoder needed 5 pages of memory, but the experience of the user was much better as there was less waiting for the next page they'd usually want. This also allowed "hidden" pages like "35F" to be used. Towards the end of analog programming, teletext brought in on a PC capture card was stored and decoded by your media player. Computers obviously had enough memory to store every page so response was instant.
- mrvenkman 5y agoIt was a great hack. ;)
- chaoticmass 5y agoI wonder-- is this why my Sony HT Receiver remote has four unlabeled buttons on it, colored Red, Green, Yellow, Blue?
- iamtedd 5y agoA better description is non-interactive.
- joezydeco 5y agoYes. Thank you. Totally got that wrong.
- zwischenzug 5y agoUK TV used to do this during down hours. I remember watching teletext pages early in the morning before our TV had teletext. I also remember the sheer excitement when we bought a TV with teletext on it, and my dad telling me he had friends who didn’t buy newspapers because of it. I was amazed.