Context. The 4th release of the SDSS Moving Object Catalog (SDSSMOC) is presently the largest photometric dataset of asteroids. Up to this point, the release of large asteroid datasets has always been followed by a redefinition of asteroid taxonomy. In the years that followed the release of the first SDSSMOC, several classification schemes using its data were proposed, all using the taxonomic classes from previous taxonomies. However, no successful attempt has been made to derive a new taxonomic system directly from the SDSS dataset. Aims. The scope of the work is to propose a different interpretation scheme for gauging u0g0r0i0z0 asteroid observations based on the continuity of spectral features. The scheme is integrated into previous taxonomic labeling, but is not dependent on them. Methods. We analyzed the behavior of asteroid sampling through principal components analysis to understand the role of uncertainties in the SDSSMOC. We identified that asteroids in this space follow two separate linear trends using reflectances in the visible, which is characteristic of their spectrophotometric features. Results. Introducing taxonomic classes, we are able to interpret both trends as representative of featured and featureless spectra. The evolution within the trend is connected mainly to the band depth for featured asteroids and to the spectral slope for featureless ones. We defined a different taxonomic system that allowed us to only classify asteroids by two labels. Conclusions. We have classified 69% of all SDSSMOC sample, which is a robustness higher than reached by previous SDSS classifications. Furthermore, as an example, we present the behavior of asteroid (5129) Groom, whose taxonomic labeling changes according to one of the trends owing to phase reddening. Now, such behavior can be characterized by the variation of one single parameter, its position in the trend.