Toll

Where Toll fits

Machine traffic is not one thing.

Some of it only reads. Some of it copies pages into a model. Some of it answers a question and may cite the page. Some of it calls an API. Some of it fills a form a thousand times.

Those should not share one switch.

Tools that sit in front of the whole site can allow a known crawler, charge it for the fetch, or block it. Some pay publishers later, when a page is actually used in an answer. Some let an identified agent pay per API call. Those products serve sites that already have an audience, and buyers that will say who they are.

Toll looks after the writes. It does not ask the caller to identify itself. It does not charge for a public article. It does not replace a crawler deal, a license, or a pay-per-citation program. A publisher can use those and still put Toll on the comment box.

The site stays open. Spam takes another road.