Earlier History of Fourier Series and Trigonometric Series

D. Bernoulli, D’Alembert, Lagrange and Euler, were led by problems in

mathematical physics to consider and discuss heatedly the possibility of
representing a more or less arbitrary function with period 2π as the sum of a
trigonometric series, and they were persuaded that an arbitrary and
discontinuous function could never be resolved in series of this kind. (1740)

Fourier announced his belief in the possibility of such a representation in 1811.

His book The Analytical Theory of Heat contains many particular instances of
such representations and makes widespread heuristic use of trigonometric
expansions.

Poisson & Cauchy & Dirichlet & Dini & Jordan (various test for convergence of
Fourier series)

It seems fair to credit Dirichlet with the beginning of the rigorous study of
Fourier series in 1829 and with the closely related concept of function in 1837.
(The development of classical analysis and the history of Fourier analysis are
closely connected to each other)

From the beginning a basic problem was that of representing a more or less
arbitrary given function defined on a period interval as the sum of at least one
trigonometric series, together with a discussion of the uniqueness of this
representation.

With the passage of time the mean of representation underwent modification

in at least two ways. (Reason: The Dirichlet kernel is not a good kernel)
1. almost everywhere convergence
2. other summable methods (e.g. Cesaro summable method)

Although it is undeniably of great intrinsic interest to know that a certain

function, or each member of a given class of functions, admits a pointwise
representation by some trigonometric series, it must be pointed out that
without delay that this type of representation leaves much to be desired on the
grounds of utility. A mode of representation can be judged to be successful or
otherwise useful as a tool in subsequent investigations by estimating what
standard analytical operations applied to the representation function can, via
the representation, be expressed with reasonable simplicity in terms of the
expansion coefficients. This is, after all, one of the main reasons for seeking a
representation in series form.

Great achievements: Du Bois-Reymond 1876 & Menshov 1916 & Kolmogorov

1923 & Carleson 1966 & Hunt 1967 & Fefferman 1973
Acknowledgement: Lots of sentences are quoted from《Fourier Series A
Modern Introduction》Volume 1 (2nd edition) by Robert E. Edwards.