Search

Language researchers are usually aware of intellectual property and personal data (PD) requirements. The problem, however, arises when these two legal regimes have conflicting requirements. For instance, when copyright law requires the acknowledgement of the author, but personal data law enshrines the data mini-misation principle. It is a practical question for a language researcher whether he should namethe author of the text used for, e.g., building a language model, or followthe data minimisation principle not to name the author. The access right that a data subject has introducessimilar conflicts. The question is what the scope of the access rightis. Does it cover only processed personal data,or does it extend to data derived from PD?

Publisher Copyright: © 2021 Estonian Association Applied Linguists. All rights reserved. The study focuses on the sharing of linguistic data containing personal data, which is the processing of personal data. Therefore, the requirements of the General Data Protection Regulation (GDPR) must be complied with. Compliance with these requirements is the responsibility of the controller, who may use the assistance of processors to process the data. In international practice, it is not clear how the responsibility for the processing of personal data is divided between a specific researcher and a research institution. For example, the French and German models differ from the Estonian, Lithuanian and Finnish models. The Greek model offers a unique approach. In general, the employer (organization or legal entity) is responsible for the processing of personal data. At the same time, research has its own specificity, which is characterized by academic freedom and as well as mobility of researchers. The situation is further complicated by the sharing of language data through research networks such as CLARIN. In the present study, the authors analyze the division of duties and responsibilities between the researcher and the research institution (including research networks). The authors also analyze how data sharing should take place from a personal data protection perspective. It is important to clarify whether the data provider and the data recipient are both controllers, joint controllers or the recipient is the processor and how the responsibility for data sharing is shared. The authors of the analysis have an interdisciplinary and international background, covering different areas of law (data protection, labor law, contract law) and jurisdictions (Estonia, Italy, Greece, Lithuania, France, Germany, Finland) and language technology.

Artikli eesmargiks on selgitada, millises ulatuses mojutab keeleandmetele kohalduv oiguslik režiim keelemudelite arendamist ja kasutamist. Autorid lahtuvad oma kasitluses protsessiskeemist, alustades algandmetest ning lopetades keeletehnoloogiat sisaldavate valmistoodetega (nt koneliidesega kulmik). Keeletehnoloogias kasutatavad algandmed sisaldavad tihti autorioiguslikult kaitstavaid teoseid, autorioigusega kaasnevate oiguste objekte (esitus, salvestus) ja isikuandmeid (isiku haal, isiku kohta kaiv muu info), mida sailitatakse annoteerimata ja annoteeritud andmekogudes. Keelandmete oiguslikke kusimusi on juba varem uuritud. Oiguslikult on labi uurimata aga keelemudelite oiguslikud aspektid. Autorid on seisukohal, et reeglina ei mojuta keelemudelite edasist oiguslikku staatust kasutatud algandmete oiguslik režiim, sest autorioigusega kaitstavad teosed mudelis pigem ei saili. Kull aga voib oiguslikke probleeme tekitada isiku haale kasutamine keelemudelis. Autorid analuusivad voimalikke lahendusvariante nende probleemide uletamiseks. Artiklis vaadeldakse ka uue autorioiguse direktiiviga kehtestatavat andmekaeve regulatsiooni ja selle rakendamist keelemudelite loomise kontekstis. *** Influence of legal regime of language data on language models This article aims to explain the extent to which the legal regime applicable to language data affects the development and use of language models. In their approach, the authors follow a process chart, starting from raw data to finished products containing language technology (eg a refrigerator with a speech interface). The raw data used in language technologies often include copyrighted works, objects of related rights (performances, sound recordings) and personal data (voice, other information about the person) stored in non-annotated and annotated databases. The legal issues of language data have already been studied. However, the legal aspects of language models have not been throughly explored. The authors are of the opinion that, as a rule, the legal status of the language models is not affect by the legal status of the used raw language data, since copyrighted works usually do not remain in the model. However, the use of a person’s voice in a language model can create legal problems. The authors analyze possible solutions to overcome these problems. The article also outlines the regulation of data mining introduced by the new copyright directive and its implementation in the context of development of language models.

Kirjandusteadlaste ja arvutilingvistide koostöös katseprojektina valminud Johannes Semperi ja Johannes Barbaruse kirjavahetuse korpus on nii kirjanduslooliselt kui tekstilingvistiliselt huvipakkuv digitaalandmestik. Kirjandusteadlastele avab kaasaegsete digitaalsete meetodite kasutuselevõtt huvitavaid uurimisperspektiive ja vanade uurimistulemuste ülekontrollimise võimalusi arvutuslike meetoditega. Korpuslingvistidele on aga väljakutseks ajaloolise ja isikupärase keelekasutusega, erinevatest keeltest kubiseva ja rohkete koha-, aja- ja isikuviidetega tekstimaterjali ettevalmistamine rikkalikult märgendatud korpuseks. Artikkel peatub üksikasjalikumalt nii käsikirjalise materjali digitaalseks tekstiandmestikuks ettevalmistamise kui ka analüüsi- ja märgendamisprotsessi probleemidel ja nende võimalikel lahendustel. Kasutajatele tutvustatakse ka korpuste päringusüsteemi KORP võimalusi sarnaste tekstide uurimiseks. Abstract. Marin Laak, Kaarel Veskis, Kadri Vider, Neeme Kahusk, and Olga Gerassimenko: Turning from digital document to text corpus: conversion of correspondence between Semper and Barbarus to a machine-readable unit in KORP. The article describes a joined pilot project of literary scholars and language technologists that resulted in a correspondence corpus of Estonian avant-garde poets Johannes Semper and Johannes Barbarus. The corpus is an inspiring digital dataset both for literary and linguistic researches. Contemporary digital methods allow literary scholars to find new interesting research perspectives and to revise the old research results with computational methods. Corpus linguists can find interesting challenges in historically and personally unique language use of the correspondents, in multiple languages used for citations and language play, in multiple references to places, events and persons in the textual material that was transformed to an annotated corpus. The article describes the preparation of typed-in manuscript material for a digital dataset in detail, problems of annotation and analysis and their possible solutions. The reader will get an insight to the possibilities that corpus query system KORP offers for the research of similar textual material.

The article focuses on determining responsible parties and the division of potential liability aris-ing from sharing language data (LD) containing personal data (PD). A key issue here is to identify who has to make sure and guarantee the GDPR compliance. The authors aim to answer 1) whether an individual researcher is a controller and 2) whether sharing LD results in joint controllership or separate controllership (whether the data's transferee becomes the controller, the joint control-ler or the processor). The article also analyses the legal relations of parties involved in data sharing and potential liability. The finalsection outlines data sharing in the CLARIN context. The analy-sis serves as a preliminary analytical background for redesigning the CLARIN contractual frame-work for sharing data.

Sometimes legal scholars get relevant but baffling questions from laypersons like: "The reference to a work is personal data, so does the GDPR actually require me to anonymise it? Or, as my voice data is personal data, does the GDPR automatically give me access to a speech recognizer using my voice sample? Or, can I say anything about myself without the GDPR requiring the web host to anonymise or remove the post? What can I say about others like politicians? And, what can researchers say about patients in a research report?" Based on these questions, the authors address the interaction of intellectual property and data protection law in the context of data minimisation and attribution rights, access rights, trade secret protection, and freedom of expression.

The authors address the legal issues relating to the creation and use of language models. The article begins with an explanation of the development of language technologies. The authors analyse the technological process within the framework copyright, related rights and personal data protection law. The authors also cover commercial use of language models. The authors’ main argument is that legal restrictions applicable to language data containing copyrighted material and personal data usually do not apply to language models. Language models are generally not considered derivative works. Due to a wide range of language models, this position is not absolute.

7th International Conference of the Faculty-of-Law-of-the-University-of-Latvia on Legal Science - Functions,Significance and Future in Legal Systems, Riga, LATVIA, OCT 16-18, 2019 The purpose of this article is to explain the extent to which the legal regime applicable to language data affects the development and use of language technology (LT). The main focus of the paper is on EU law. The article also maps possible text and data mining (TDM) issues. The authors focus on TDM for research purposes outlined in the Digital Copyright Directive 2019/790.The authors follow a process approach of LT development, which starts from raw data collection and leads to LT products such as a refrigerator with a speech interface. Particular attention is given to language models.The raw data used in LT often include copyright-protected works, objects of related rights (e.g., performances) and personal data in the form of person’s voice or other information stored in non-annotated and annotated databases.The authors’ main argument is that the legal regime of language data does not usually affect the use of language models since copyrighted works are not likely to remain in models. In the process of developing a language technology application, language models are the first intermediate result that can be free from legal restrictions affecting language data. The use of a person’s voice as identifiable personal data in a language model can create legal challenges. In some cases, developers of language technology must be careful how to address issues of processing of personal data contained in models.

The authors address the transformation of research data into open data. The article draws on the experience in four countries: Sweden, Finland, Estonia and Lithuania. The transformation process presents several challenges where legal, organizational and individual aspects influence the process. Research data often contain personal data. Research data could also be covered with intellectual property (IP) rights. This means that personal data and IP regulations should be integrated into the dissemination model. While there is a potential conflict between the policies for open data that aim to make data freely available and those of an entrepreneurial university that emphasize commercialization of research results, these policies need to be made compatible. Researchers producing data are vital for reconciling the two, but they currently lack the motivation to contribute towards the implementation of the open data policy due to missing career incentives.

Kui keeleressursside autorioiguslikke kusimusi on eelnevalt pohjalikult uuritud, siis isikuandmete kaitse temaatika keeleressursside kontekstis on suurelt osalt katmata. Teema on aktuaalne, sest 25. mail 2018 hakatakse kohaldama EL-i isikuandmete kaitse uldmaarust. Kaesolevas artiklis analuusitakse isikuandmete kaitse kusimusi keeleressursside loomisel ja kasutamisel teadustoo eesmargil kahes kontekstis: 1) andmesubjekt ehk isik, kelle andmed keeleressursis sisalduvad, on andnud oma andmete tootlemiseks nousoleku; 2) nousolekut ei ole ning isikuandmete tootlemine toimub teadustoo erandi alusel. Artiklis analuusitakse keeleressursside loomisel kasutuses olevaid nousolekuvorme ning tehakse ettepanekuid, kuidas keeleressursside kontekstis isikuandmete kaitse regulatsiooni jargida. *** Data protection issues relating to the development and utilisation of language resources Copyright issues related to language resources have received considerable attention. Personal data protection in the field of language research is an area which is not thoroughly explored yet. This topic is very relevant since the General Data Protection Regulation is applicable from 25 May 2018. The authors analyse the issues relating to personal data protection in the context of the development and utilization of language resources. Firstly, the authors analyse the cases where there is a consent which allows processing of personal data to develop and use language resources for research purposes. Secondly, the processing of personal data without consent for language research purposes is explored. The article is meant to provide practical information to language research community and create a foundation for further discussion.

The development and use of language resources often involve the processing of personal data. The General Data Protection Regulation (GDPR) establishes an EU-wide framework for the processing of personal data for research purposes while at the same time allowing for some flexibility on the part of the Member States. The paper discusses the legal framework for language research following the entry into force of the GDPR. In the first section, we present some fundamental concepts of data protection relevant to language research. In the second section, the general framework of processing personal data for research purposes is discussed. In the last section, we focus on the models that certain EU Member States use to regulate data processing for research purposes.

Artikkel keskendub keeleressurssidega seonduvatele oiguslikele kusimustele. Aluseks voetakse protsess, mis algab keeleressursside loomisega ning lopeb nende levitamisega. Keeleressurss on oma olemuselt autorioiguslikult kaitstav andmebaas, mida katavad kahetasemelised oigused (oigused autorioiguslikult kaitstavale materjalile, mida ressursside loomisel kasutati, ja oigused ressursile endale), millega tuleb levitamise protsessis arvestada. Keeleressursside loomine saab tugineda kahele voimalikule oiguslikule mudelile: 1) litsentsipohine mudel ja 2) erandipohine mudel. Molemaga kaasnevad positiivsed ja negatiivsed aspektid, mida kaesolevas artiklis ka analuusitakse, vottes arvesse arenguid Euroopa Liidus. Enne kui keeleressurssi saab levitada CLARIN-i vorgustikus, on oluline, et oigused levitada oleks ule antud monele Eesti Keeleressursside Keskuse konsortsiumipartnerile. Oigused ressursile lahevad ule kas toosuhtes voi siis lepinguga. Sel eesmargil on tootatud valja CLARIN-i deponeerimislepingud. Enne kui kasutaja saab ressursile ligipaasu, peab ta noustuma CLARIN-i teenusetingimustega. Artikkel selgitab CLARIN-i tuuplepingutega seonduvaid kusimusi ning tutvustab CLARIN-i lepingute viimaseid parandusi. Regulatory framework determining the development and utilization of digital language resources in Estonia and its compatibility with CLARIN infrastructure The article focuses on legal issues relating to language resources. These issues are analyzed throughout the process of the creation of language resources and their subsequent distribution. Language resources can be defined as copyright protected databases, which are governed by two tiers of rights: 1) rights in material used to create the database and 2) rights in this database. The creation of language resources can be subject to two models: 1) the contract model and 2) the exception model. Both models have their advantages and disadvantages. The models are analyzed in the light of developments at the EU level. Before language resources can be distributed, it is necessary to transfer copyright covering resources to a partner of the Center of Estonian Language Resources. The rights are transferred within employment or contractually. For this purpose, CLARIN deposition license agreement templates have been developed. Before end users can access language resources they have to accept CLARIN Terms of Service. The scope of the article is not confined to the explanation of CLARIN agreement templates. The article also highlights the latest amendments to these templates.

Artikkel käsitleb teaduspoliitilisi ja õiguslikke küsimusi, mida tuleks arvestada keeleressursside litsentsimisel, ja määratleb sobivaimad litsentsimise tingimused, pöörates tähelepanu eelkõige lepinguõiguse ja intellektuaalomandi litsentsimise küsimustele. Samuti võrreldakse artiklis USA, Euroopa riikide ja Eesti keeleressursside litsentsimise õiguslikku raamistikku ning tutvustatakse levinumaid praktikaid avatud juurdepääsu (ingl open access) ja vabade (avatud) litsentside vaatepunktist. Artikli praktiliseks eesmärgiks on selgitada Eestis keeleressursside loomise ja kasutamisega seotud teadlastele ja teadusasutustele intellektuaalomandi (IO) objektide litsentsimise põhialuseid ja jagada näpunäiteid litsentsitingimuste valimiseks. Artikkel on sündinud keeleteadlaste ja juristide koostöös ning mõeldud esmajoones keeleteadlastest lugejaskonnale.DOI: http://dx.doi.org/10.5128/ERYa9.21

Proceeding volume: 123 The article focuses on the regulatory and contractual frameworks in CLARIN. A process analysis approach has been adopted to allow an evaluation of the functionality and shortcomings of the entire legal framework applicable to language resources and technologies. The article discusses and provides background information to amendments of key provisions of CLARIN license templates. The authors also address issues relating to the research exception allowing for the development of language resources without the copyright holder’s consent. The article introduces some practical information on a new version of the license category calculator. The article reflects the authors’ personal understanding and insights gained by examining the legal aspects of language resources and technologies in Estonia and Finland.

The chapter deals with the analysis of sentence structure on the layer where semantic role categories such as AGENT, PATIENT, OBJECT, etc. play a crucial role. The main objective is to find out which regularities prevail in categorization of spatial characteristics of motion events in Estonian in terms of these categories, and which morphosyntactic means are used to encode them. For the analysis a (mini) corpus containing sentences with a motion verb as predicate was automatically created. First, a brief overview of verbs of motion in Estonian is given. After that, the chapter concentrates on the analysis of semantic roles and their encoding in motion sentences in Estonian. We found it necessary to differentiate between the following roles: SOURCE, GOAL, ROUTE, LOCATION. The realization of each category by morphosyntactic means of Estonian is described in detail: NPs in certain case forms, PPs, adverbs. One of the results of the chapter is the fact that there is a statistical asymmetry in the use of SOURCE, GOAL, and LOCATION: GOAL is the most frequently encoded category of these three.