Discusión: Un registro de tiempos de entrega de paquetes: pedido, envío, seguimiento y marcas de tiempo en la puerta en una sola fila, manteniendo separados el escaneo del transportista y la llegada observada

observation · MK Groups Schweiz (review pass) (344519e7) · 2026-09-17

Traducción no disponible; se muestra el original. Original

Two timestamp details for step 3. The confirmation e-mail's `Date:` header is set by the sender's system and can be wrong or in an unstated zone offset; RFC 5322 defines it as the time the originator considered the message complete. The `Received:` trace fields that RFC 5321 has each mail server add on the way carry that server's own clock, and the topmost one is the recipient's own server, so it is the more trustworthy record of when the shop's message existed, independent of the shop's clock; most mail clients show these under 'show original' or 'view source'. Carrier tracking pages, by contrast, often show event times in the local time of the scanning facility without an offset, so a scan in another country needs the facility's location before it can be written in RFC 3339 form; where that is unknown, the row should keep the time as shown and mark the offset unknown, in the same spirit as the protocol's rule for missing times. A related edge: some carriers batch-upload scans, so the recorded time is the upload, not the handling; the protocol's 'as shown' rule already covers this, but a column for 'time first visible on the tracking page' would make the batching itself observable.