{"id":"W79589210","doi":"","title":"ON THE DISTRIBUTION OF SOURCE CODE FILE SIZES","year":2011,"lang":"en","type":"article","venue":"","topic":"Software Engineering Research","field":"Computer Science","cited_by":16,"is_retracted":false,"has_abstract":true,"ca_institutions":"Queen's University; University of Victoria","funders":"","keywords":"Computer science; Source code; Log-normal distribution; Software; Code (set theory); Software quality; Pareto distribution; Statistics; Operating system; Mathematics; Software development; Programming language; Set (abstract data type)","routes":{"ca_aff":true,"ca_fund":false,"ca_venue":false,"about_ca":false,"invisible_to_affiliation_only":false},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0001510297,0.00003495216,0.00003745878,0.00001416145,0.00002656869,0.00001308193,0.0004985656,0.00001564425,0.0006892121],"category_scores_gemma":[0.0006981656,0.00002101987,0.00001985623,0.000151946,0.00002627252,0.00004870316,0.0001087777,0.00005877719,0.00009422173],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00001108298,"about_ca_system_score_gemma":0.00001265936,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00004251227,"about_ca_topic_score_gemma":0.000001155556,"domain_scores_codex":[0.9995767,0.00001762578,0.00005538398,0.00008851782,0.000158428,0.0001033571],"domain_scores_gemma":[0.9985895,0.00100039,0.00001272943,0.0003397805,0.00003473084,0.00002285177],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"theoretical_or_conceptual","study_design_gemma":"observational","study_design_scores_codex":[0.00001311588,0.0001748146,0.003521346,0.00002013543,0.00002797141,0.000004633835,0.001549415,0.0005097989,0.0004364829,0.69345,0.2784079,0.02188431],"study_design_scores_gemma":[0.000560673,0.0008496962,0.4870441,0.0001130734,0.000006185553,0.00001508957,0.0001295183,0.2147854,0.2335525,0.02335896,0.03900808,0.0005766542],"study_design_candidate":"theoretical_or_conceptual","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.07667085,0.000006070019,0.9208977,0.0001728537,0.000050353,0.00005803134,0.00001776739,0.0001498094,0.001976567],"genre_scores_gemma":[0.9960358,5.599898e-7,0.003120416,0.00003038517,0.000006661024,0.00000803238,0.000002897196,0.000002410451,0.000792828],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.919365,"threshold_uncertainty_score":0.7546389,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.0300936040446079,"score_gpt":0.2369343151585387,"score_spread":0.2068407111139308,"validation_status":"score_only:v0-immature-baseline","note":"Baseline scores from an immature model (maturity gate not passed). Scores rank; they never assert a category."}}