I'm used to the definition of Affine Variety as $$V(f_1, ..., f_s)=\{a=(a_1, ..., a_n) \in k^n \hspace{2mm}| \hspace{2mm} f_i(a)=0 \hspace{3mm} \text{for every} \hspace{3mm} 1 \leq i \leq s\}$$ for some field $k$. However, I've read recently (on a French book) the definition of an Affine Variety $V$ as a subset of some vector space such that, for every $x, y \in V$, the line that intersects $x$ and $y$ are entirely contained in $V$. Formally speaking, an affine variety $V$ is a set such that $$ x, y \in V, t \in \mathbb{R} \implies (1-t)x+ty \in V.$$ These two definitions, of course, are different, but could someone explain me why there is this conflict in definitions? It seems to me that in French Affine Varieties are called Algebraic Affine Varieties, and what they call by Affine Variety is what I just wrote here. What would be the equivalent definition in english?
Thanks.