In this short article I will examine the positions of foundationalism and coherentism and will argue that a form of weak foundationalism is the most satisfactory option as a valid theory of justification by knowledge and is therefore a viable way to avoid any kind of vicious regress problem and skepticism. Foundationalism addresses the infinite regress problem in the following way: if person O is to be justified in having a belief X, X must be justified by a further belief Y, which must inferentially justify belief X. Furthermore, Y must be justified by another belief, Z. Instead of following this line ad infinitum, the foundationalist holds that ultimately there must be a foundational belief that is self-justifying and capable of justifying any belief that relies on it in turn for its own justification. Consequently, for example, belief Z must be able to justify itself without belief Z1, and be able to justify belief Y at the same time. Therefore, inferential justification must be possible for non-foundational beliefs; non-inferential justification must also be possible. How one arrives at noninferentially justified beliefs is a major problem for the foundationalist. In my opinion, it is difficult to accept that there can be self-justifying foundational beliefs because, as has been pointed out, if a belief is to be self-justifying, person O must "know" in some sense what feature of that belief makes it self-justifying, otherwise the belief it is arbitrary. So, assuming that feature Q is what makes belief P justified, person O should then know that belief P has feature Q. Therefore, belief P is no longer non-inferentially justified because it is based on knowledge of person OR the IQ characteristic.... . half of the paper......there are some fundamental beliefs that possess some degree of intrinsic justification, but as has been noted, accepting these beliefs as completely self-justifying is difficult to accept. Therefore, these core beliefs that possess a low degree of justification may rely on other minimally justified beliefs for support, consequently creating a kind of coherent foundation. In the end, the pursuit of an unproblematic theory of justification seems like a Sisyphean task. All major theories have obvious weaknesses and it seems that a compromise between some elements of the different views will be necessary to get closer to a more acceptable view. In this article I have tried to show some of the main weaknesses of the two main theories of justification and to look for an alternative and weak foundationalism, which seems to me to be a legitimate way forward..