There's one simple reason why we don't have space telescopes the size of yo momma (like, kilometer-sized) in orbit right now, and that's because they're just too big and too fat to launch. A company called Tethers Unlimited has received a pile o' cash from NASA to figure out how to build giant structures in orbit instead.
Launching stuff into space isn't just expensive, it's stressful. And I mean stressful in the physical sense: anything that you want to send up has to be able to survive around three gravities. This means that when you build a structure that's going to spend its lifetime in space, you have to massively overengineer it to be able to survive the launch, and then all of that additional size and weight is wasted once it's in space. Also, since anything going to space has to fit inside the fairing of the rocket that's going to take it there, there's a hard limit on maximum sizes.
What Tethers Unlimited wants to do is throw the whole build-it-on-Earth-launch-it-to-space method out the window and just skip right to building giant structures in space instead, using automated processes and fused-deposition modeling.
SpiderFab, then, would fab whatever a SpiderFab fabs, potentially including "multiple high-gain antennas in Earth and solar orbit to support a deep-space communications network, long-baseline interferometry systems for terrestrial planet finder programs, and submillimeter astronomy of cosmic structure." These structures would be able to be sent up in the form of raw materials, ultimately reducing launch mass and bulk by a factor of 30 (!), and allowing for the construction of space telescopes that are one to two orders of magnitude (that's 10x to 100x) larger than anything we've ever seen before.