Thanks very much for doing that! (I haven't actually downloaded your new versions yet, though.) Did you use a tool to do the A/B versions or was it a manual process, or a bit of both?I've combined the A/B versions of the last two editions and tidied up the reception errors as best I can. I've not gone through and tidied any of the others up though so all the other little glitches are still there. I don't like to tinker around too much without the page checksums to check against!
In hindsight I can't believe I didn't write my own download utility which would keep retrying each page until the CRC was correct. I did have poor reception a lot of the time but would probably have got a good CRC eventually, and if I cared enough about this stuff to be archiving it it would have been worth leaving the machine on for a while (maybe even overnight) to get a good capture. Ideally I'd also have stored the correct CRC in the file along with a bad version, if that was all I could get, so there was at least the prospect of fixing it up later and "verifying" it against the CRC as you mention, even if the fix up happened years later.
And I can't help but think - though this might be expecting too much of teenage me - that even on a BBC, saving multiple bad copies of the same page and using majority voting for each of the 1K bytes would be doable. There'd be room in memory for ~20 copies of the same page, and by that point surely a simple majority at each position would be likely to give something pretty decent unless reception was utterly borked.
I was only a teenager though, and as I remarked in a different context I think it's easy to forget just how hard information was to come by back then as well. And what it was like having a single-tasking machine which I was doubtless itching to use for other things instead of leaving it running capturing teletext pages for ill-defined reasons.
Statistics: Posted by SteveF — Thu Oct 10, 2024 1:20 am