We explore some issues and considerations around the current copyright ownership landscape, pertaining to blockchain and new technology's industry integration
Registration of content attributes [M] to said company’s control or at least registered database. Hosting media itself grants controllers large influence with possible financial benefit both in respect to user-data, pending remuneration percentages and or corporatized control
If visible online this may also bestow URL destination popularity for search engine listings and or similar rankings. Highly impactful possible delineations of [M] for profit too remains murky 
In place of the more presently ubiquitous text entry, reverse-image lookup can be basically compared to using a picture as a search term. Rather than typing ‘sailboat’ into Google, Bing or Yahoo! one may upload a picture of a sailboat to find existing appearances or algorithmically related appearances of such
This could include analysis of image attached or derived meta-data. As with Google, Bing or Yahoo! the quality of any search result depends on the size of the searched database. In short, if someone is unlawfully using [ILLEGALM] but not indexed on the database or set searched then [ILLEGALM] cannot be found
Whereas software solutions may potentially include artificial intelligence, presently some companies specify the presence of a watermark, [Mw], would prohibit successful identification of [M] as also being [Mw]
Further complications come to mind when considering slight variations of [M2], photos taken [PM] from different angles [A2M], after Photoshop alteration [PSDM] and so on. Fixation of copyright protections to variations of [M] may thereby be potentially ruined 
In data analysis various and intricately complex algorithms exist  As evidenced, host’s violation notices of digital audio and or video content copyrights are possible. Even in cases of YouTube notices for some 400 hours of video being uploaded every minute  in-house databases to binary-quantifiable sets for immediate cross-comparison, e.g. ‘0101’ with ‘0101’, do perform their function. Yet when compared to a human’s natural analysis image recognition is far from equally robust 
Presently even facial algorithm matching technologies on personal devices can be fooled  To date no program may exist that would, with assured accuracy, be able to conclusively differentiate a ‘Spanish galleon’ from a ‘sailboat’ tattoo. Furthermore the same ‘sailboat’ tattoo placed on different body areas may be considered unique. In consideration of visual artwork as a field and in terms of quantity the problems of quick identification compound exponentially 
Search requires comparison against a given set or range. The larger the comparison set, concurrently, the longer time required. Complexity of the search of course further dictating processing speed  Over 11 billion websites are indexed  with numerous more typically created every second  Fully analyzing a miraculous 5,000 existing websites per-second, identification of either an approved or violating appearance of the robustly identified single media-content [M] would require weeks to complete. When considering a database of thousands, the time required could be measured in decades
The internet cannot be held as static. Either aforementioned search case then necessitating continued, ongoing cycling for identification of recent appearances. This is further omitting the fact of a search potentially being unable to identify if the copyright violator used [M] offline in say a print newspaper, poster advertisement, marketing brochure, physical portfolio, on merchandise or the like
This also does not consider search omitting downloads to personal devices, inclusions in say digital media compositions hosted by various sites and or sets of conceivably edited [M2] variations in alternate channels alongside transmissions of an encrypted copy of [∑M]
Let us imagine that a truly altruistic company hosts [M], has the most advanced software ever developed to accurately match [M] with [Mw] alongside all its possible variations
And let us imagine that this futuristic algorithm identifies a third party using [Mw] in violation with the creator’s copyright. Unless the violator has registered with the same company and authorized payments for [Mw]’s use, voiding the need for identification software in the first place, the copyright violation must be pursued using centralized legal systems
This violation must also be deemed as legitimately under specific violations of proved ‘rights’ assigned within said central jurisdiction 
Centralized legal pursuance would firstly entail identifying the violator as an individual, say through IP registration. Yet what if [Mw]’s appearance was from a separate blog posting that was just re-posted? Or perhaps [Mw] was provided to them by another individual who claims to have copyright permission?
If pursuing the violator as a company could it be argued that [Mw]’s appearance was known to be a violation definitively proved as implemented for financial benefit without proper media-content source credit listed?
What if the violator is identified yet unreachable or simply doesn’t respond to the registered mail from the company’s lawyers? As proposed by one company an offer of post-license purchase would be made available for the ‘offender’ to purchase so as to deter ‘loss of reputation’  which seems a wholly nominal inducement considering the size of the digital landscape
From identification, proof of intent, international jurisdictions, cost valuation or quantification of violation to then actually gaining remuneration, there exists a plethora of human-labor intensive and expensive steps  Rudimentary cost-benefit analysis shows the requirements to complete such tasks do not add up to a sustainable service as may presently be implemented
Staying with fantasy, let us say that the same altruistic company hosting [M] did indeed receive remuneration from a successfully pursued copyright violation. What would be the profit from said remuneration received for media-content that the company itself ‘does not own’?
Already enacted by film and music industries in such copyright pursuance, mass litigation may be held as the single feasible option  Compounding the precarious tenability of the entire endeavor theories of achieved remuneration now entail large-scale identification of violators
By definition, digital content is perfectly reproducible. In an act of astounding historical blindness when recalling Napster to current bit-torrent file sharing networks  roles of media-content ‘gatekeepers’ have thus far been evidenced to offer little more assistance than disorganized digital bounty hunters  Decentralized efficacy extends to a securely verifiable transactional accounting solution pertaining to transactions that are intended to be securely recorded. This speaks nothing to say, out of many other possibilities, the use of encryption
A perfectly possible blockchain transaction could record the purchase of an encrypted transfer which in reality details an illegally reproduced piece of content being sold for profit in violation of an owner’s intellectual property rights. Despite decentralized purchase recording, due multiple factors no restorative action could be held as imminently pending