What are you trying to say? I don't understand your point if you have one. From here it just seems like you're trolling by writing an essay about how RSS feeds' claims can't be verified by an index ...
You are right that I argued something you never claimed, and I am dropping it.
Your point is narrower than how I read it: two databases want to swap an edge and agree on what it means. Publisher feeds are the one provable link between feeds, so they are the natural carrier. The note about Fountain in this thread states 330 album feeds each naming a publisher feed and 70 publisher feeds listing their albums back. That is a clean set of edges, and today every index stores its own guess about what each edge means. When your database and Oscar's exchange those edges, the same edge arrives carrying two meanings and nothing on it says which one applies. A named link type makes the meaning travel with the edge instead of being reconstructed on the far side by whoever guesses first.
One detail I would want settled before the exchange is worth much: whether an absent value means "no answer" or "this feed predates the tag". Both write nothing, so the two look identical in the XML and two indices will disagree about which one they are reading. An explicit value for unknown fixes an ambiguity that the omission creates.
That part I would use. The rest of what I wrote belongs to a different argument, and you were right to say so.
Your point is narrower than how I read it: two databases want to swap an edge and agree on what it means. Publisher feeds are the one provable link between feeds, so they are the natural carrier. The note about Fountain in this thread states 330 album feeds each naming a publisher feed and 70 publisher feeds listing their albums back. That is a clean set of edges, and today every index stores its own guess about what each edge means. When your database and Oscar's exchange those edges, the same edge arrives carrying two meanings and nothing on it says which one applies. A named link type makes the meaning travel with the edge instead of being reconstructed on the far side by whoever guesses first.
One detail I would want settled before the exchange is worth much: whether an absent value means "no answer" or "this feed predates the tag". Both write nothing, so the two look identical in the XML and two indices will disagree about which one they are reading. An explicit value for unknown fixes an ambiguity that the omission creates.
That part I would use. The rest of what I wrote belongs to a different argument, and you were right to say so.
1