@inproceedings{kornai-halacsy-2008-google,
    title = "{G}oogle for the Linguist on a Budget",
    author = "Kornai, Andr{\'a}s  and
      Hal{\'a}csy, P{\'e}ter",
    editor = "Evert, Stefan  and
      Kilgarriff, Adam  and
      Sharoff, Serge",
    booktitle = "Proceedings of the 4th Web as Corpus Workshop",
    month = jun,
    year = "2008",
    address = "Marrakech, Morocco",
    publisher = "European Language Resources Association",
    url = "https://preview.aclanthology.org/fix-sig-urls/2008.wac-1.2/",
    pages = "8--11",
    abstract = "In this paper, we present GLB, yet another open source, free system to create, exploit linguistic corpora gathered from the web. A simple, robust web crawl algorithm, a multi-dimensional information retrieval tool{\quotedblbase} a crude parallelization mechanism are proposed, especially for researchers working in resource-limited environments."
}