Non-monetary data usage threatens Bitcoin’s efficacy
Mon 6 Feb 2017
Researchers in Italy, investigating the trend towards using Bitcoin’s blockchain for non-financial purposes, have concluded that the ongoing practice is likely to impact the Bitcoin system in the near future.
The study An analysis of Bitcoin OP RETURN metadata, by Massimo Bartoletti and Livio Pompianu of Italy’s Universita degli Studi di Cagliari, examines the growing use of Bitcoin’s OP_RETURN metadata feature, which was instituted in the cryptocurrency’s scripting language as a repository for ancillary data, and to prevent users from instituting the processor-intensive workarounds which let transactions deal with data unrelated to money.
The latter practice occurred because Bitcoin had no assigned field for arbitrary data when it was created. Workaround methods included writing data as a parameter of an out-script instruction, such as the public key. Storing data in this fashion has very limited redundancy potential, since the values must be calculated, at the expense of processor time, in order to validate the entire transaction. Since the computed values have to be stored in RAM as well, the extra data also threatens memory resources.
Thus OP_RETURN was introduced in March of 2014 – a field which was pre-marked as ‘unspendable’ data. Though the field data limit was originally set to be 80 bytes, the first release allowed only 40 bytes, prompting discourse in the Bitcoin community about whether nodes should be permitted to block OP_RETURN transactions. Ultimately the allocation was finally extended to 80 and then 83 bytes.
At the volume and scalability of the Bitcoin blockchain, that’s still a significant potential drag on the efficacy of the entire ledger system.
Researching use of OP_RETURN since its launch, the Cagliari researchers identified six possible categories of use for the field: Assets, used to certify ownership, value and exchange of real-world properties; Document notary, which establishes timestamps and propriety of a document, such as might be necessary for establishing original creation and ownership; Digital Arts, which operates around DRM-style access rights to digital files such as music or photos; Other – a catch-all category including specious usages such as blockchain messaging systems like Eternity Wall; Empty (zero-data fields); and Unknown – where the researchers could identify no obvious protocols.
Because users require higher storage capacity, i.e. for large numbers, than the OP_RETURN field allows, the data often ends up split between a higher number of transactions, in a similar manner to segmented archive recombination (from the days where large files would be split across multiple posts in newsgroups, with the user tasked with their reassembly after downloading).
Users’ co-opting of the Bitcoin blockchain for ancillary purposes in this way is as inventive a use of a protocol intended for other, more limited purposes, as binary newsgroups were undertaking in the pre-broadband age. But the effect is more deleterious, because latency and efficacy is such a critical factor for the future of the Bitcoin blockchain.
The researchers note that it took a year for uptake of the OP_RETURN field to have any noticeable impact on Bitcoin, and conclude: ‘We think that the main motivation for not using cheaper and more efficient storage is the perceived sense of security and persistence of the Bitcoin blockchain. If this trend will be confirmed, the specific needs of these applications could affect the future evolution of the Bitcoin protocol.’