MétaCan
Menu
Back to cohort
Record W2996114301 · doi:10.25200/slj.v8.n2.2019.408

Information automatisée et nouveaux acteurs des processus journalistiques

2019· article· fr· W2996114301 on OpenAlexaff
Laurence Dierickx

Bibliographic record

VenueSur le journalisme About journalism Sobre jornalismo · 2019
Typearticle
Languagefr
FieldComputer Science
TopicCultural Insights and Digital Impacts
Canadian institutionsWorld Federation of Science Journalists
Fundersnot available
KeywordsHumanitiesPolitical sciencePhilosophy

Abstract

fetched live from OpenAlex

FR. Le développement des technologies d’automatisation de la production d’informations dévoile, sur la scène médiatique, de nouveaux acteurs qui ne sont pas traditionnellement liés au monde du journalisme. On retrouve, parmi ceux-ci, des sociétés technologiques où sont employés des linguistes et des informaticiens. Bien qu’elles ne considèrent pas « faire acte » de journalisme, elles participent néanmoins à une chaîne de production éditoriale, jusqu’alors dévolue aux seuls professionnels de l’information. Cela implique de nouvelles formes de collaboration où vont cohabiter des agents sociaux aux cultures professionnelles différentes : à la rationalité de la technologie, sera opposée la subjectivité du journalisme. Mais il ne s’agirait pas de réduire un processus automatisé à une technologie qui se bornerait à transformer des entrants (des données) en extrants (des textes ou toute autre forme de représentation visuelle). Ce processus participe à une logique éditoriale traditionnelle, caractérisée par une succession de choix. Cet article vise à une meilleure compréhension de cette dynamique, à travers une étude de cas réalisée dans le cadre de la conception d’un système d’automatisation visant à soutenir les routines quotidiennes de journalistes attachés au service boursier d’un média belge francophone. Dans ses principaux enseignements, cette expérience témoigne de la nécessité d’un profil qui soit à la fois technique et journalistique, dès lors qu’il facilite les échanges entre les mondes du journalisme et de la technique. L’implication active des journalistes dès les premières heures du projet va apparaître comme un préalable, dès lors qu’ils disposent de l’expertise du domaine d’application et des compétences éditoriales qui forgeront l’artefact dont on ne peut considérer qu’il soit d’abord technologique. Pour les agents sociaux du monde de la technique, cela implique un changement de paradigme : il s’agit désormais d’admettre qu’ils participent à un processus éditorial qui suppose, tout au moins, de développer une « pensée journalistique ». *** EN. The development of news automation technologies is uncovering new actors on the media scene who have not traditionally been linked to the world of journalism. Among these are technology companies employing linguists and computer scientists. Although they do not consider themselves as “producing journalism,” these companies do participate actively in an editorial production chain that was hitherto the purview of news professionals. This implies new forms of collaboration between different professional cultures; where the rationality of technology is opposed to the subjectivity of journalism. An automated process cannot be reduced to the transformation of input (data) into output (texts or any other form of visual representation), however. This process relies on traditional editorial logic, characterized by a succession of choices. This paper aims to better understand this dynamic by way of a case study conducted within French-speaking Belgian media, where an automated system was designed to support the daily routines of stock market journalists. The main lesson gleaned from this experiment was the need for a profile defined by both the journalistic and the technical fields, as it is meant to facilitate exchanges between these two worlds. The active involvement of journalists from the very beginning of any such project appears to be vital, as long as they have an expertise in the field it will be applied and the editorial skills to shape the program, which cannot be considered primarily technological. For social agents involved in the world of technology, this calls for a paradigm shift in which they accept that their participation in this new editorial process implies, at minimum, the development of a form of (automated) “journalistic thinking.” *** PT. O desenvolvimento de tecnologias de automação de notícias está revelando novos atores no cenário da mídia, que não estão tradicionalmente ligados ao mundo do jornalismo. Entre eles, existem empresas tecnológicas onde linguistas e cientistas da computação são empregados. Embora não se considerem fazendo jornalismo, essas empresas participam ativamente de uma cadeia editorial de produção, que até então era confiada aos profissionais da informação. Isso implica novas formas de colaboração onde coexistem diferentes culturas profissionais: a racionalidade da tecnologia será então oposta à subjetividade do jornalismo. No entanto, qualquer processo automatizado não pode ser reduzido à transformação de entradas (dados) em saídas (textos ou qualquer outra forma de representação visual). Esse processo se apóia em uma lógica editorial tradicional, caracterizada por uma sucessão de escolhas. Este artigo objetiva uma melhor compreensão dessa dinâmica, por meio de um estudo de caso realizado na mídia belga de língua francesa, onde um sistema automatizado foi projetado para apoiar as rotinas diárias de jornalistas ligados a um serviço do mercado de ações. Em suas principais lições, essa experiência atesta a necessidade de um perfil fundamentado tanto no campo jornalístico quanto no técnico, pois facilitará o intercâmbio entre os mundos do jornalismo e da tecnologia. O envolvimento ativo de jornalistas desde as primeiras horas do projeto aparecerá como um pré-requisito, pois eles têm o conhecimento deste domínio de aplicação, bem como as habilidades editoriais que forjarão um artefato que não pode ser considerado puramente tecnológico. Para os agentes sociais envolvidos no mundo da tecnologia, supõe uma mudança de paradigma ao admitir que o envolvimento deles em um processo editorial que implica, pelo menos, desenvolver uma forma de "pensamento jornalístico". ***

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.

How this classification was reachedexpand

Full frame distilled prediction

Teacher imitation

Not calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.

metaresearch head score (Codex)0.004
metaresearch head score (Gemma)0.001
Version: codex-gemma-dda1882f352aValidation status: machine_predicted_unvalidated
Candidate categoriesMeta-epidemiology (narrow), Science and technology studies, Scholarly communication, Research integrity, Insufficient payload (model declined to judge)
Consensus categoriesMeta-epidemiology (narrow), Scholarly communication, Insufficient payload (model declined to judge)
DomainCandidate signal: none · Consensus signal: none
Study designCandidate signal: Not applicable · Consensus signal: none
GenreCandidate signal: Empirical · Consensus signal: Empirical
Teacher disagreement score0.606
Threshold uncertainty score1.000

Codex and Gemma teacher scores by category

CategoryCodexGemma
Metaresearch0.0040.001
Meta-epidemiology (narrow)0.0020.002
Meta-epidemiology (broad)0.0020.001
Bibliometrics0.0010.003
Science and technology studies0.0020.001
Scholarly communication0.0240.072
Open science0.0030.001
Research integrity0.0010.004
Insufficient payload (model declined to judge)0.0010.001

Machine scores (provisional)

The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.

Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.

Opus teacher head0.084
GPT teacher head0.295
Teacher spread0.210 · how far apart the two teachers sit on this one work
Validation statusscore_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from it

Classification

machine, unvalidated

Machine predicted; both teacher heads agree on what is shown here.

Study designNot applicable
Domainnot available
GenreEmpirical

How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".

Quick stats

Citations6
Published2019
Admission routes1
Has abstractyes

Explore more

Same venueSur le journalisme About journalism Sobre jornalismoSame topicCultural Insights and Digital ImpactsFrench-language works237,207