Los primeros 500 códigos sku

A SKU — stock keeping unit — is a retailer's own name for a thing it sells, which is nearly the whole definition. No standards body issues them, no registry deconflicts them, and two shops can use the same string for different products without either being wrong. The freedom is the point: a SKU serves the person picking stock in a warehouse, so it is usually built to be read, with segments for category, size and colour.

The numbers that are governed came out of the grocery trade. In 1948 Bernard Silver, a graduate student at Philadelphia's Drexel Institute of Technology, overheard a food-chain president asking a dean about reading product information automatically at a checkout, and took the problem to his friend Norman Joseph Woodland. Woodland left Drexel for Florida to keep working on it and, by his own account, formed his first barcode in beach sand: "I just extended the dots and dashes downwards and made narrow lines and wide lines out of them." They filed on 20 October 1949 for "Classifying Apparatus and Method", granted as US 2,612,994 on 7 October 1952. Their preferred symbol was a bullseye of concentric circles, readable from any direction, and premature by two decades.

What settled the format was a failed trial. RCA, which had bought the rights to Woodland's patent, tested the bullseye in a Kroger store in Cincinnati from July 1972, and it lost for a mundane reason: printers smeared ink, which ruins a circular code, whereas a linear code printed along its stripes merely grows taller. On 3 April 1973 the industry picked the IBM design instead — twelve digits, the last a check digit — drawn up by George Laurer at IBM's Research Triangle Park, where Woodland himself was by then employed. The first commercial scan followed on 26 June 1974 at the Marsh supermarket in Troy, Ohio: a ten-pack of Wrigley's Juicy Fruit gum, at 8:01 a.m., for 67 cents. A facsimile of it sits in the Smithsonian.

Europe then built EAN as a thirteen-digit superset, UPC became the twelve-digit case of it, and the administering bodies came together under the GS1 name in the mid-2000s. Both are now GTINs: a UPC-A is a GTIN-12, an EAN-13 a GTIN-13. A manufacturer part number sits between the two worlds — assigned by whoever makes the thing, stable across resellers, but unique only with the brand alongside it.

Propiedades clave

• A SKU is internal to one business. No registry assigns them, so the same string can mean different products at two retailers without either being wrong.
• A UPC-A is twelve digits and its last digit is a modulo-10 check digit computed with alternating weights of 3 and 1.
• An EAN-13 is a thirteen-digit superset of UPC-A: a UPC-A is a GTIN-12 and an EAN-13 is a GTIN-13.
• A manufacturer part number identifies a part as its maker numbered it, and is unique only together with the brand — which is why product feeds ask for brand and MPN as a pair.
• UPC number-system digits 2 and 4 are both reserved for local store or warehouse use — 2 for items sold by variable weight, such as meat and produce packed in the shop, and 4 most often for loyalty cards and store coupons — so neither needs a registered company prefix.
• The pattern fields are {CAT:n}, {SIZE:n} and {COLOR:n} for the attribute lists, {SEQ:n} for the padded counter, {YYYY} and {YY} for the season year, {N:n} for random digits, {A:n} for random letters and {X:n} for random letters and digits.
• Matrix mode emits the cross product of only the lists the pattern mentions, varying the last dimension fastest, so the same lists always give the same codes in the same order.
• The codes on this page are SKUs and not GTINs: there is no check digit, and they will not scan as a UPC or an EAN.