{
  "site": {
    "title": "Silly Dataset Release Review",
    "owner": "ShmuggingFace downstream demo",
    "visibility": "Public mock demo",
    "reviewerHint": "Use this fake release to check whether a downstream project can generate believable Hugging Face and Kaggle-style review surfaces before a real dataset publication.",
    "primarySlug": "sock-drawer-benchmark"
  },
  "datasets": [
    {
      "slug": "sock-drawer-benchmark",
      "title": "Sock Drawer Benchmark",
      "owner": "laundry-labs",
      "subtitle": "A synthetic tabular benchmark for matching clean socks without claiming the missing ones ever existed.",
      "license": "CC-BY-4.0 mock",
      "task": "tabular-classification",
      "language": "English",
      "updated": "2026-05-15",
      "downloads": "1,337",
      "likes": "42",
      "kaggleUsability": "9.1",
      "kaggleMedals": "Socksilver",
      "description": "A deliberately fake dataset for testing metadata, row previews, downloads, and review comments. Rows describe sock colors, patterns, and pair likelihoods.",
      "tags": [
        "tabular",
        "classification",
        "synthetic",
        "mock-release"
      ],
      "files": [
        {
          "path": "data/train.csv",
          "size": "18 KB",
          "kind": "CSV",
          "sourcePath": "data/train.csv"
        },
        {
          "path": "data/test.csv",
          "size": "7 KB",
          "kind": "CSV",
          "sourcePath": "data/test.csv"
        },
        {
          "path": "README.md",
          "size": "5 KB",
          "kind": "Dataset card",
          "sourcePath": "README.dataset.md"
        }
      ],
      "columns": [
        "sock_id",
        "pattern",
        "dominant_color",
        "pair_probability",
        "split"
      ],
      "rows": [
        {
          "sock_id": "sock-0001",
          "pattern": "stripes",
          "dominant_color": "navy",
          "pair_probability": "0.98",
          "split": "train"
        },
        {
          "sock_id": "sock-0002",
          "pattern": "dots",
          "dominant_color": "yellow",
          "pair_probability": "0.73",
          "split": "train"
        },
        {
          "sock_id": "sock-0003",
          "pattern": "plain",
          "dominant_color": "black",
          "pair_probability": "0.91",
          "split": "train"
        },
        {
          "sock_id": "sock-0004",
          "pattern": "argyle",
          "dominant_color": "green",
          "pair_probability": "0.64",
          "split": "validation"
        }
      ],
      "discussions": [
        "Does this mock make the file downloads obvious?",
        "Synthetic-generation note looks clear.",
        "Try this as a closed reviewer app next."
      ]
    },
    {
      "slug": "office-mug-poses",
      "title": "Office Mug Poses Mini",
      "owner": "breakroom-vision",
      "subtitle": "Synthetic metadata for mugs facing left, right, or optimistically forward.",
      "license": "MIT mock",
      "task": "image-classification",
      "language": "n/a",
      "updated": "2026-05-15",
      "downloads": "256",
      "likes": "19",
      "kaggleUsability": "8.7",
      "kaggleMedals": "Porcelain",
      "description": "This fake release checks image-dataset language, file grouping, and Kaggle/Hugging Face page structure without including actual photos.",
      "tags": [
        "images",
        "computer-vision",
        "synthetic",
        "mock-release"
      ],
      "files": [
        {
          "path": "images/sample_manifest.csv",
          "size": "11 KB",
          "kind": "CSV",
          "sourcePath": "images/sample_manifest.csv"
        },
        {
          "path": "images/thumbs.zip",
          "size": "1.4 GB",
          "kind": "ZIP",
          "storage": "Git LFS",
          "downloadUrl": "https://github.com/ShmuggingFace/silly-dataset-release-demo/raw/main/large-files/thumbs.zip",
          "downloadLabel": "Open Git LFS"
        },
        {
          "path": "README.md",
          "size": "4 KB",
          "kind": "Dataset card",
          "sourcePath": "README.dataset.md"
        }
      ],
      "columns": [
        "image_id",
        "pose",
        "background",
        "handle_visible",
        "split"
      ],
      "rows": [
        {
          "image_id": "mug-101",
          "pose": "left",
          "background": "desk",
          "handle_visible": "yes",
          "split": "train"
        },
        {
          "image_id": "mug-102",
          "pose": "forward",
          "background": "kitchen",
          "handle_visible": "no",
          "split": "train"
        },
        {
          "image_id": "mug-103",
          "pose": "right",
          "background": "shelf",
          "handle_visible": "yes",
          "split": "validation"
        }
      ],
      "discussions": [
        "The fake project name is clear enough.",
        "Add thumbnail preview support in a future core slice.",
        "Good small image-dataset smoke case."
      ]
    },
    {
      "slug": "elevator-small-talk-corpus",
      "title": "Elevator Small Talk Corpus",
      "owner": "awkward-nlp",
      "subtitle": "Tiny synthetic text snippets labeled by how quickly everyone starts staring at the floor numbers.",
      "license": "Apache-2.0 mock",
      "task": "text-classification",
      "language": "English",
      "updated": "2026-05-15",
      "downloads": "404",
      "likes": "23",
      "kaggleUsability": "8.9",
      "kaggleMedals": "Lobby",
      "description": "A harmless fake NLP release for testing text previews, label descriptions, and download affordances in pre-release mocks.",
      "tags": [
        "text",
        "nlp",
        "synthetic",
        "mock-release"
      ],
      "files": [
        {
          "path": "corpus/train.jsonl",
          "size": "14 KB",
          "kind": "JSONL",
          "sourcePath": "corpus/train.jsonl"
        },
        {
          "path": "corpus/labels.json",
          "size": "2 KB",
          "kind": "JSON",
          "sourcePath": "corpus/labels.json"
        },
        {
          "path": "README.md",
          "size": "6 KB",
          "kind": "Dataset card",
          "sourcePath": "README.dataset.md"
        }
      ],
      "columns": [
        "utterance_id",
        "text",
        "label",
        "split"
      ],
      "rows": [
        {
          "utterance_id": "talk-001",
          "text": "Weather is certainly happening today.",
          "label": "weather-neutral",
          "split": "train"
        },
        {
          "utterance_id": "talk-002",
          "text": "This elevator has excellent vertical commitment.",
          "label": "situational",
          "split": "train"
        },
        {
          "utterance_id": "talk-003",
          "text": "Mondays arrive with confidence.",
          "label": "calendar",
          "split": "validation"
        }
      ],
      "discussions": [
        "Nice for testing text dataset previews.",
        "No real people or chats are represented.",
        "The mock notice should stay visible on every page."
      ]
    }
  ],
  "mockNotice": "This is a ShmuggingFace review mock. It is not Hugging Face, Kaggle, or a real dataset release."
}