Now in 2009 the practices reported here are well established and accepted. Already in 2006 the same practices emerged in the sw dev team of the F1 Racing Team I was part of and was reported in the keynote speach at the Italian Agile Day 2006.
Oggi queste pratiche descritte qui sono note e riconosciute da (quasi) tutti. Già nel 2006 queste pratiche sono emerse spontaneamente nel team di sviluppo sw del team di F1 di cui ho fatto parte e sono state citate nella keynote del Italian Agile Day 2006.
How it works in Google as reported by Mark Striebeck from his keynote speach at XpDay 2009 London:
Come funziona in Google, dalla keynote di Mark Striebeck agli XpDay 2009 in Londra:
Google has one single code repository that contains all the source code of all the systems developed by Google. All systems are released independently and still everything is built from HEAD without any binary releases.
Google ha un unico repositori del codice per tutte le sue applicazioni. Ogni applicazione è rilasciata indipendentemente e tutto viene compilato e rilasciato dalla Head (no brench) senza reference a altri binari compilati.
Two interesting slideshow on this topic:
Due presentazioni interessanti sul tema:
3 team practices / 3 pratiche di team
-
release only from the head of the repo (no branches)
-
check-in only shippable code
-
use Latent Code Patterns to check-in shippable code even when the development of user story is in progress
Update: Google also for Continuous Deployment, the next frontier:
Cerca con Goocle Continuous Deployment, la sfida successiva: