Updated: Apr 20, 2020
The nature of information and the formulation of unificatory theories of the nature of information and semantic information (when they are regarded as separate) are still regarded as being grounded in a set of open metaphysical, logical, metaphilosophical, and methodological questions. Information is a polysemous concept across domains of scientific and non-scientific application and across contexts in those domains. Thus, most contemporary theorists in the field recognize pluralism about the nature of information as a correct metametaphysical assumption and premise for the philosophy of information (T. W. Deacon, 2008, p. 150). However, many of these theorists also present unificatory conceptions, including what Luciano Floridi has called an ur-concept of information. These conceptions are often surprisingly reductionist.
Developing any unifying conception of information involves dealing effectively with foundational problems in the metaphysics of information. For example, I’ve argued elsewhere that a basic but important error often committed by philosophers of information is that, in discussing information transmission and the nature of information, they often confuse and conflate the two. A similar problem arises for theories of the nature of information that focus on information content. Information and information content are not the same thing.
One needs a conception of the nature of information before one can present a conception of information transmission, and defining information in terms of, or as, transmission of information, is clearly circular. This is often an indirect error, but is sometimes due to carelessness. The carelessness often comes with an ambitious attempt to present a grand-schematic metametaphysics for information, or else to secure a central naturalistic conception on a probabilistic or other basis (See Stegmann and Scarantino). Likewise for defining information in terms of information content: one needs a conception of the nature of information before one defines the nature of information content, unless one is just referring to the information in some structure, source, process, system, mechanism, or other phenomenon.
It’s not impossible that a central epistemically and metaphysically satisfying conception of the nature of information might come to be broadly ratified. However, if so it’s likely to be tempered by both theory defeasibility and pluralism across different explanatory levels and in different metaphysical, scientific, and explanatory contexts. Theory defeasibility is especially likely where and when conceptions of information are influenced by or grounded in scientific conceptions of information and information transmission, which is common. This is precisely because of the defeasibility of scientific theories in general, albeit on an optimistic meta-inductive basis. Pluralism is inherent at a metametaphysical level since more than one influential scientific theory is understood to provide a strong basis for a conception of information. Correspondingly, the main scientific characterisations of information – probabilistic-entropic and computational – have very different ontic and conceptual bases, and various efforts to unify them have been attempted.
Luciano Floridi is probably the best known philosopher of information to present both a framework for the accommodation of pluralism in different explanatory settings, and a (Kantian transcendental) unificatory conception of information that provides a basis for a general conception of information across numerous domains and explanatory levels (Floridi, 2008, 2011b). His work is notable for this system, based as it is upon the concept of levels of abstraction, and also for the fact that he breaks with approaches recommended by earlier theorists, some of whom pursued multifaceted models accounting for, say, cognitive and non-cognitive situations and epistemic and non-epistemic content, in order to provide a unified conception of information across multiple metaphysical domains. However, it's notable that Floridi's Kantian ur-concept of information and informational structuralism is coupled with a sophisticated application of the computer-scientific idea of levels of abstraction (LoA) for complex systems architectures. The ur-conceptual Kantian transcendental differentiae de re that constitute Floridi's basis of information and data, on a somewhat reductionist basis, must reconcile coherently with the pluralist LoA schema. It's by no means a comfortable marriage of concepts given Floridi's naturalistic but non-reductionist ontic commitments.
In keeping with the openness of the question of the nature of information and semantic information, the impression that there has thus far not been a convincing solution to either, and the awareness of and commitment to pluralism about information as an in-principle metaphilosophical and meta-metaphysical normative guide, a number of theorists have offered multifaceted and often triadic (on different bases) conceptions of the nature of information.
Tripartite approaches are sometimes influenced by Peircian semiotic or other premises, and sometimes by the perceived existence of different natural and other aspects of information: cognitive internal versus external representation, sources, and flow, for example. They are also often married with existing extensive or else grand schema philosophies, which I suggest is another mistake. Hofkirchner's (2009) tripartite Hegelian conception recommends that a unified conception of information has to account for what he calls the syntactic, semantic, and pragmatic features of existing conceptions of information and the theories that deploy them. However, the Hegelian commitments tend to greatly impede the clarity and aptness of the project, as does the normative assumption that information has a tripartite character, motivated as it is by longstanding concerns about how to account for semantic information content.
Terrence Deacon (2007, 2008, 2013, 2010) outlines another grand-schematic marriage of scientific theory, older philosophy, and a triadic model. It's a Peircian tradic-semiotic and Shannonian classical hybrid which normatively stipulates, and emphasizes the need to account for, physical, referential, and normative components of information. The Peircian model forces a-priori conceptual and stipulative premises into Deacon’s metaphysical schema where otherwise the naturalistic scientific metaphysical - or at least naturalistic and scientistic - approach he favours would not need such. For example, one wonders why Deacon's systems-theoretic conceptions of regularity, or self-organising regular behavior, or perhaps even just regular dynamics – or even “pattern or cyclicity” or else “merely a tendency to consistently exhibit some possible states more often than others” - are not enough to fulfil the functional and explanatory role he instead ascribes to the Peircian notion of habit.
Peircian habit is applied by Deacon as a means to imply naturalistic teleology for what Deacon calls teleodynamic systems, on a teleonomic basis. It also supports a counterfactual notion of causal power or influence for what's absent from the state of a system or signal, which in turn is again associated with a subjectivist teleology for systems. The accompanying subjectivism both requires and accommodates the necessary inclusion of a recursively autopoietic and evolved Peircian interpretant and interpreter. Such a convolution, though inventive, is simply not necessary to explain natural information content. Subjective agents, interpreters, and consumers are not necessary conditions for natural information content in physical systems and their structures. There is no objective nomic or natural necessary reason for normative stipulative assertions to the contrary. Naturalistic attempts to reconcile unifactory conceptions of information with pluralist premises are troubled by numerous challenges. Attempting to import or apply existing and dated grand-schematic a-priorist metaphysical and semiotic schemas creates problems both for the coherence of resulting theories, and for the reputation of the philosophy of information. A scientific metaphysical approach that heavily de-emphasises a-priorist premises is preferable.
 According to some conceptions, natural and physical information is intrinsically semantic by way of causal indication (Long, 2014)  I suppose my inclusion of the term ‘scientistic’, although not with negative connotations, is intended to signify more a-priori conceptual analytic approaches.
Adriaans, P. (2010). A Critical Analysis of Floridi’s Theory of Semantic Information. Knowledge, Technology & Policy. https://doi.org/10.1007/s12130-010-9097-5
Deacon, T. W. (2007). Shannon – Boltzmann – Darwin: Redefining information (Part I). Cognitive Semiotics. https://doi.org/10.3726/81600_123
Deacon, T. W. (2008). Shannon - Boltzmann - Darwin: Redefining information (Part II). Cognitive Semiotics. https://doi.org/10.3726/81605_169
Deacon, T. W. (2010). What is missing from theories of information? In P. Davies & N. H. Gregersen (Eds.), Information and the Nature of Reality: From Physics to Metaphysics (pp. 146–169). Cambridge University Press. https://doi.org/10.1017/CBO9780511778759.008
Deacon, T. W. (2013). Incomplete nature : how mind emerged from matter (1st ed.). New York: W.W. Norton & Co.
Dennett, D. C. (2013). Aching Voids and Making Voids A review of Incomplete Nature: How Mind Emerged from Matter . By Terrence W. Deacon. New York: W. W. Norton & Company. $29.95. xvii + 602 p.; ill.; index. ISBN: 978-0-393-04991-6. 2012. . The Quarterly Review of Biology. https://doi.org/10.1086/673760
Dodig Crnkovic, G., & Hofkirchner, W. (2011). Floridi’s “Open Problems in Philosophy of Information”, Ten Years Later. Information. https://doi.org/10.3390/info2020327
Hofkirchner, W. (2009). How to achieve a unified theory of information. TripleC: Communication, Capitalism & Critique. Open Access Journal for a Global Sustainable Information Society. https://doi.org/10.31269/vol7iss2pp357-368
Floridi, L. (2008). A defence of informational structural realism. Synthese. https://doi.org/10.1007/s11229-007-9163-z
Fodor, J. (2012, May 24). What are trees about? London Review of Books, 1–5. Retrieved from https://www.lrb.co.uk/the-paper/v34/n10/jerry-fodor/what-are-trees-about
Scarantino, A. (2015). Information as a Probabilistic Difference Maker. Australasian Journal of Philosophy, 93(3), 1–25. https://doi.org/10.1080/00048402.2014.993665
Stegmann, U. (2015). Prospects for Probabilistic Theories of Natural Information. Erkenntnis, 80(4), 869–893. https://doi.org/10.1007/s10670-014-9679-9