Sunday, January 27, 2019

Autoreaders are here

I forgot to mention, along my travels in the web I an across a great product, it summarizes your favorite web sites, an auto reader!  Basically a dedicated construction of join, a Watson-lite. Neat idea, dedicated web reader.

Gets us back to the point, what is the point of XML? That is another wonder of duplication, comma separated files with a few key words can do the same function, and almost everywhere the first software action is to remove the tags, bots don't really read them, in the sense of look them up.

I have a better idea.  Use the xml tags to describe you lay out, then use a bit of intelligence to map arbitrary text to your layout, put it in receivers hands. So, reader needs it in report format, for example, bingo his browser base rectangle is set. Then use has a slightly intelligent lexical analyzer that can pretty much get headers, footers, body, summary and the rest; just flow that text into the preset layout. Then keep all those tags off the net, especially the bot nets, they cause useless spaghetti everywhere.

No comments: