After reading some with regard to the quantum computing gates and the comparison to classic gates it seems that the quantum computer, at least for the time being, is not viewed as a programmable element of a "hardware circuit" but more as a hardware device that combines functions, as in what is termed in "classical computing" ASSP - application-specific signal processors. Is this view right?
Today many people believe that programming means coding on some language like python; this is not true. The early classical computers were programmed by inserting junctions which connect logical elements of an electronic scheme, and this is also programming.
I believe modern programmable quantum computers are programmed like that: a programmer is given a certain number of gates, and allowed ways to connect the gates; his task is to connect the gates to implement some algorithm.
(...) depending on what you attempt to compute you "design" your circuit - this is what is called ASSP (Application Specific Signal Processor) where the input is Signal which are processed by the Circuit (processor) to create the processed output - the measure gate. [Source]
Reading the question more carefully and after your clarification in the comments, I think it has some merit. It is indeed the case that quantum gates are not fully programmable in most architectures; the gate array (quantum circuit) has to be designed according to the specific algorithm and specific problem we're trying to solve (what you call ASSP), which certainly is a nuisance.
However, note that in Google's latest architectures like the Sycamore, the gates are indeed fully programmable (at least, that's their claim), as is expected of a general-purpose quantum computer. I'm not sure what your impression of "programming" quantum gates is, but it's like manipulating a sequence of laser or microwave pulses on qubits in ion-trap and superconducting architectures respectively. See Peter Shor's answer for a nice explanation.