# rondelmo.it archive: downloads Snapshot of 2026-10-23 of the website of the Computational Linguistics Laboratory, Ca' Foscari University of Venice. Each `.tar.gz` extracts to `rondelmo.it/`, with paths that match the URLs of the site. Rights stay with the original authors; nothing here is licensed by the archive. ## rondelmo-stilven-20261023.tar.gz STILVEN: Veneto language resources. Materials of the STILVEN project as the lab published them: three CD-ROM directories (CD-STILVEN1, CD-STILVEN2, CD-STILVEN3) and the downloads listed on the STILVEN downloads page (morphology, minimal pairs, a Veneto corpus and an English corpus, a lexicon, frequency lists). The site describes STILVEN as a project approved in December 2007 whose task was a computational infrastructure for the analysis and translation of the Veneto language. CD-STILVEN3 holds venparallel.txt and engparallel.txt: two UTF-8 files (with byte-order mark) of 667,960 lines each, whose first lines, line 5000 and line 400000 pair a Veneto entry with an English one (single words and short phrases in the lines sampled), so they read as a line-aligned bilingual list rather than sentence pairs. Only those lines were inspected. CD-STILVEN2 has an Italian README listing MOSES training material, conference papers, new dictionaries and morphology work. 128 files, 206.5 MB, sha256 `c6f58dcbb8ade099b9cf6b0828c15806562abcefd15bffd2511c00293711eb8a`. ## rondelmo-resources-20261023.tar.gz Software and datasets from the downloads pages. Software and datasets the lab distributed. res/deep_getaruns.tar.gz is GETARUNS 1.10 (readmeGT inside); res/LinuxUbuntu.zip, res/WindowsXP.zip, res/MacOSX10.4.11.zip and res/MacOSX10.5.5.zip each hold the VENSES 3.7 application together with a SWI-Prolog installer (VENSES: Venice Semantic Evaluation System, a textual-entailment system built on a reduced GETARUNS and a partial parser, per the lab site). res/rte_devset.zip and res/test_texts.zip are test material; res/sharedtask holds output of GETARUNS for a shared task, one folder per text, with a README. resource/ holds a SWI-Prolog version of Berkeley FrameNet 1.4 (README_FNet_Prolog.txt), resource/VIT with a Browser-VIT folder (VIT is the Venice Italian Treebank, the subject of docs/Venice italian treebank.pdf), and two archives named PROSO_DOS.rar and Proso_Mac.rar whose contents were not inspected. 572 files, 128.5 MB, sha256 `3c520db385a0e7fc44f4041105bf7d542bfa81d34dc7467a742a9658d44c55f5`. ## rondelmo-events-20261023.tar.gz Conference and workshop sites. Static pages, programmes, papers and photographs of events hosted on the laboratory site: ARQAS, ICALL2004, SLaTE2011, STEP2008, TATAE, plus the master and papers folders and the EVEXTAL/VEXTAL/GFS-IX pages. Registration forms that other people filled in and uploaded to the WordPress site are not part of it. 734 files, 150.0 MB, sha256 `b5515505611c1a9772ced4b9349eb7f7f5582a3c8b4aa48de3219caf30e533e4`. ## rondelmo-nlpcs2014-20261023.tar.gz NLPCS 2014 workshop: static files only. The static files of the nlpcs2014/ directory: slides, theme assets and uploads of a second WordPress install. Its pages are not preserved: the install was configured for a MySQL database called "conference", which is not in the 2026 dump of the host nor in the 2022 copy of the MySQL data directory. 1,561 files, 90.1 MB, sha256 `51c0d38e233a4f84d34068e23dc9b3949b60aaee1b0d14b8c4b577a60f9e41c8`. ## rondelmo-documents-20261023.tar.gz Lecture notes, papers, CVs and the Italian and English pages. The rest of the docroot: docs/ (lecture notes for linguistica informatica, papers, CVs), paper/, slim/, TATAE/, video/, images/, ling-inf-1_files/, and the root-level pages in Italian and English, the word lists eng2.txt and ven2.txt, and app/ (the static files of a small PHP application). 169 files, 32.1 MB, sha256 `1caa33fdd09291afd8a70ce1087fc9f38723d442d92507dbe81b434465b5add8`. ## rondelmo-wordpress-content-20261023.tar.gz Main WordPress install: uploads, themes, core assets. Everything static under wordpress/: the papers, images and slides uploaded to posts (wp-content/uploads), the twentyten theme and the WordPress 4.6.1 core scripts and styles. The posts themselves are in the wordpress-pages bundle and in the fulltext file. 92 files, 13.1 MB, sha256 `67a616eb68afd4b760b15631850cef3b2a5d22833b106e0e3cd876753e8e5261`. ## rondelmo-wordpress-pages-20261023.tar.gz WordPress posts, pages and archives, frozen as HTML. The 157 HTML files of the WordPress site (the home page, posts and pages, attachment pages, and the author, category and month archives), rendered by a throwaway WordPress 4.6.1 container loaded from the 2026 database dump, crawled with wget, and cleaned for the archive (comment forms, search, feeds, analytics and avatars removed; links made relative; the 55 approved comments kept; the 324,755 unapproved comments, nearly all spam, left out). File names follow the old query strings: index.html@p=ID.html is what was served at /?p=ID. 157 files, 0.2 MB, sha256 `5fa3a9b4fa39a398865f8eabf1da8710d126fdac13d163333d80c610fe0a9b43`. ## rondelmo-fulltext-20261023.jsonl.gz Plain text of every HTML page and PDF, one JSON record per document. JSON Lines, gzip-compressed. Fields: path, url, type (html or pdf), title, lang (the html lang attribute, else null), bytes and sha256 of the file in the archive, chars, text, extractor, origin, and a note on PDFs with almost no extractable text. HTML text skips the navigation, sidebar and footer of the WordPress theme and the archive banner. PDF text comes from poppler pdftotext with no layout mode and is not cleaned: expect hyphenation, running headers and, for scans, nothing. 929 files, 2.6 MB, sha256 `adb401d84a7592ef0a506ac3abea444e19cb119356944442b24ae0dea242c73d`. Verify with `sha256sum -c SHA256SUMS`. `MANIFEST.tsv` has a sha256 for every member of every bundle.