1: \begin{abstract}

2: It has been over 200 years since Gauss's and Legendre's famous priority

3: dispute on who discovered the method of least squares. Nevertheless, we

4: argue that the normal equations are still relevant in many facets of

5: modern statistics, particularly in the domain of high-dimensional

6: inference. Even today, we are still learning new things about the law

7: of large numbers, first described in Bernoulli's \textit{Ars Conjectandi}

8: 300 years ago, as it applies to high dimensional inference.

9: 

10: The other insight the normal equations provide is the asymptotic

11: Gaussianity of the least squares estimators. The general form of the

12: Gaussian distribution, Gaussian processes, are another tool used in

13: modern high-dimensional inference. The Gaussian distribution also

14: arises via the central limit theorem in describing weak convergence of

15: the usual least squares estimators. In terms of high-dimensional

16: inference, we are still missing the right notion of weak convergence.

17: 

18: In this mostly expository work, we try to describe how both the normal

19: equations and the theory of Gaussian processes, what we refer to as the

20: ``geometry of least squares,'' apply to many questions of current

21: interest.

22: \end{abstract}