The dissertation project proceeds through three papers, analyzing issues related to software interoperability and respectively pertaining to one of the three following interdependent levels of analysis. The first level addresses the legal status of software interoperability information under current intellectual property law (focusing on copyright law, which is the main legal tool for the protection of these pieces of code), trying to clarify if, how and to what extent theses pieces of code (and the associated pieces of information) are protected erga omnes by the law. The second level complements the first one, analyzing legal and economic issues related to the technical possibility of actually accessing this interoperability information through reverse engineering (and software decompilation in particular). Once a de facto standard gains the favor of the market, reverse engineering is the main self-help tool available to competitors in order to achieve interoperability and compete “inside this standard”. The third step consists in recognizing that – in a limited number of cases, but which are potentially of great economic relevance – market failures could arise, despite any care taken in devising checks and balances in the legal setting concerning both the legal status of interoperability information and the legal rules governing software reverse engineering. When this is the case, some undertakings may stably gain a dominant position in software markets, and possibly abuse it. Hence, at this level of analysis, competition policy intervention is taken into account. The first paper of the present dissertation shows that interoperability specifications are not protected by copyright. In the paper, I argue that existing doubts and uncertainty are typically related to a poor understanding of the technical nature of software interfaces. To remedy such misunderstanding, the paper focuses on the distinction between interface specifications and implementations and stresses the difference between the steps needed to access to the ideas and principle constituting an interfaces specification and the re-implementation of a functionally equivalent interface through new software code. At the normative level, the paper shows that no major modifications to the existing model of legal protection of software (and software interfaces) are needed; however, it suggests that policymakers could reduce the Fear of legal actions, other forms of legal Uncertainty and several residual Doubts (FUD) by explicitly stating that interface specifications are unprotectable and freely appropriable. In the second paper, I offer a critique of legal restraints on software reverse engineering, focusing in particular on Europe, but considering also similar restraints in the US, in particular in the context of the Digital Millennium Copyright Act. Through an analysis of entry conditions for late comers and of the comparative costs of developing programs in the first place or reverse engineering them, the paper shows that limitations on decompilation imposed by article 6 of the Software Directive were mostly superfluous and basically non-binding at the time of drafting. What is more, the paper shows that nowadays new – and largely unanticipated – developments in software development models (e.g. open source) make these restraints an obstacle to competition against dominant incumbent controlling software platforms. In fact, limitations on the freedom to decompile obstacle major reverse engineering projects performed in a decentralized way, as in the context of an open source community. Hence, since open source projects are the most credible tools to recreate some competitive pressure in a number of crucial software markets, the paper recommends creating a simpler and clear-cut safe harbor for software reverse engineering. The third paper claims that, in software markets, refusal-to-deal (or “information-withholding”) strategies are normally complementary with tying (or “predatory-innovation”) strategies, and that this complementarity is so relevant that dominant platform controllers need to couple both in order to create significant anti- competitive effects. Hence, the paper argues that mandatory unbundling (i.e. mandating a certain degree of modularity in software development) could be an appropriate – and frequently preferable – alternative to mandatory disclosure of interoperability information. However, considering the critiques moved from part of the literature to the Commission’s Decision in the recent European Microsoft antitrust case, an objection to the previous argument could be that – also in the case of mandatory unbundling – one should still determine the minimum price for the unbundled product. The last part of the paper applies some intuitions coming from the literature concerning complementary oligopoly to demonstrate that this objection is not well grounded and that – in software markets – mandatory unbundling (modularity) may be a useful policy even if the only constraint on the price of the unbundled good is the one of non-negativity.

SOFTWARE INTEROPERABILITY: Issues at the Intersection between Intellectual Property and Competition Policy / Morando, Federico. - (2009). [10.6092/polito/porto/2615059]

SOFTWARE INTEROPERABILITY: Issues at the Intersection between Intellectual Property and Competition Policy

MORANDO, FEDERICO
2009

Abstract

The dissertation project proceeds through three papers, analyzing issues related to software interoperability and respectively pertaining to one of the three following interdependent levels of analysis. The first level addresses the legal status of software interoperability information under current intellectual property law (focusing on copyright law, which is the main legal tool for the protection of these pieces of code), trying to clarify if, how and to what extent theses pieces of code (and the associated pieces of information) are protected erga omnes by the law. The second level complements the first one, analyzing legal and economic issues related to the technical possibility of actually accessing this interoperability information through reverse engineering (and software decompilation in particular). Once a de facto standard gains the favor of the market, reverse engineering is the main self-help tool available to competitors in order to achieve interoperability and compete “inside this standard”. The third step consists in recognizing that – in a limited number of cases, but which are potentially of great economic relevance – market failures could arise, despite any care taken in devising checks and balances in the legal setting concerning both the legal status of interoperability information and the legal rules governing software reverse engineering. When this is the case, some undertakings may stably gain a dominant position in software markets, and possibly abuse it. Hence, at this level of analysis, competition policy intervention is taken into account. The first paper of the present dissertation shows that interoperability specifications are not protected by copyright. In the paper, I argue that existing doubts and uncertainty are typically related to a poor understanding of the technical nature of software interfaces. To remedy such misunderstanding, the paper focuses on the distinction between interface specifications and implementations and stresses the difference between the steps needed to access to the ideas and principle constituting an interfaces specification and the re-implementation of a functionally equivalent interface through new software code. At the normative level, the paper shows that no major modifications to the existing model of legal protection of software (and software interfaces) are needed; however, it suggests that policymakers could reduce the Fear of legal actions, other forms of legal Uncertainty and several residual Doubts (FUD) by explicitly stating that interface specifications are unprotectable and freely appropriable. In the second paper, I offer a critique of legal restraints on software reverse engineering, focusing in particular on Europe, but considering also similar restraints in the US, in particular in the context of the Digital Millennium Copyright Act. Through an analysis of entry conditions for late comers and of the comparative costs of developing programs in the first place or reverse engineering them, the paper shows that limitations on decompilation imposed by article 6 of the Software Directive were mostly superfluous and basically non-binding at the time of drafting. What is more, the paper shows that nowadays new – and largely unanticipated – developments in software development models (e.g. open source) make these restraints an obstacle to competition against dominant incumbent controlling software platforms. In fact, limitations on the freedom to decompile obstacle major reverse engineering projects performed in a decentralized way, as in the context of an open source community. Hence, since open source projects are the most credible tools to recreate some competitive pressure in a number of crucial software markets, the paper recommends creating a simpler and clear-cut safe harbor for software reverse engineering. The third paper claims that, in software markets, refusal-to-deal (or “information-withholding”) strategies are normally complementary with tying (or “predatory-innovation”) strategies, and that this complementarity is so relevant that dominant platform controllers need to couple both in order to create significant anti- competitive effects. Hence, the paper argues that mandatory unbundling (i.e. mandating a certain degree of modularity in software development) could be an appropriate – and frequently preferable – alternative to mandatory disclosure of interoperability information. However, considering the critiques moved from part of the literature to the Commission’s Decision in the recent European Microsoft antitrust case, an objection to the previous argument could be that – also in the case of mandatory unbundling – one should still determine the minimum price for the unbundled product. The last part of the paper applies some intuitions coming from the literature concerning complementary oligopoly to demonstrate that this objection is not well grounded and that – in software markets – mandatory unbundling (modularity) may be a useful policy even if the only constraint on the price of the unbundled good is the one of non-negativity.
File in questo prodotto:
File Dimensione Formato  
Federico_Morando-PhD_Thesis.pdf

accesso aperto

Descrizione: Tesi dottorato Federico Morando, versione integrale
Tipologia: Tesi di dottorato
Licenza: Creative commons
Dimensione 1.61 MB
Formato Adobe PDF
1.61 MB Adobe PDF Visualizza/Apri
Pubblicazioni consigliate

I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/11583/2615059
 Attenzione

Attenzione! I dati visualizzati non sono stati sottoposti a validazione da parte dell'ateneo