Every article in the guide, and where it is drawn

Because nobody arrives at a dog-care library wanting to know what was published most recently. They arrive with a word, bloat, itching, the dog pulls, and the only arrangement that answers a word without asking you to already know its category is the alphabet. Every other shelf here is arranged by an argument: health guides by the region of the body, diseases by whether a vaccine exists, conditions by how fast they need acting on, behaviour cases by the complaint that brought the parent in. Those arrangements are better than an alphabet once you know where you are. This page is for before that.

The first part is the shelf: COND is the conditions sheet, HLTH the health guides, DIS the diseases, BEHV the behaviour casebook, RSCH the research records. The second part is the band that shelf is genuinely divided into on its own page, the conditions sheet really does sort by urgency, so “emergency” is where that record sits when you get there. Nothing in a locator is decorative; each band word is read from the same typed record the destination page renders, so this index cannot file something under a coordinate its own shelf would disagree with.

It is the pilcrow, the printer’s paragraph mark, and the number after it is how many questions that record answers inside itself. Opening it lists those questions, and each one links to the exact place in the record where it is answered rather than to the top of the page. Entries without a pilcrow, the behaviour cases, are pages that do not divide into numbered questions, so the mark would be claiming a precision that is not there.

Because this index files articles, records that answer a question, and a breed page is a profile rather than an answer. The breeds have a better arrangement of their own: the register at /breeds sorts several hundred of them by shoulder height, group and origin, which is how somebody actually compares breeds. Folding them in here would also have meant an alphabet where breed names outnumbered every written guide, disease record and behaviour case put together, and the reader who arrives with the word “bloat” would be reading past Beagle and Bloodhound to find it.

The site holds several hundred breed-by-condition probiotic pages, generated from a grid of breeds against conditions. They are real, they are in the sitemap and they answer a real search, but they are produced from a matrix rather than written one at a time. Filing them here would have meant an alphabet in which the great majority of entries were near-identical machine-written variants, which advertises volume rather than extent. They keep their own hub, linked at the foot of this page.

Because a title and a headword are different things. “Why Dogs Eat Everything Off the Ground” is a good article title and a useless index entry, an index files that under S for scavenging, which is the word the reader arrived holding. The same rule moves a leading article to the end, so “The itch” files under I where somebody would look for it. Where a word is genuinely ambiguous or has a common alternative, it gets its own cross-reference line instead: look up GDV and the index sends you to bloat.

No, it complements it. The alphabet files names, and it is the better tool when you know roughly what the thing is called and want to see it in the company of its neighbours. The finder at the top of the index is the better tool when you arrived with a sentence rather than a name. It reads inside every record, so typing a phrase from a question will surface the record that answers it even when the headword looks nothing like what you typed.