About xyzzy! search enginexyzzy! — which rhymes with whizzy — is a new web search engine with an emphasis on providing quality results for “sentient beings”, especially Yorkshire folks. It is currently undergoing development and testing, and was opened up to the public for the first time on Sunday 1st January 2017.
xyzzy! is actually more of a discovery engine than a search engine, because you're not going to be able to find all the information you want here, but you will discover a lot of nuggets and treasure along the way as you search.
We aim to cover all sorts of wholesome topics, and topics off the beaten path, such as community, volunteering, spirituality, psychology, philosophy, science and technology, health, disability, education, literacy, humanitarian concerns, the environment, culture, the arts, multiculturalism, geography, history, traditions, folklore, alternative lifestyles, creativity, craft work, innovation, humour, and good news.
We crawl and index all resources ourselves. Unlike many, it's not “powered by” google, yahoo! or bing, nor funded by big money, and we do not aggregate results from other search engines. We do however use the Wikipedia API for the Wikipedia search facility, as we feel that this is a useful resource for our clients.
Dancing to a different drumAs Jacob Nordby wrote in Pearls of Wisdom:
“Blessed are the weird people:
poets, misfits, writers,
mystics, painters, troubadours,
for they teach us to see the world through different eyes.”
Yorkshire flavourWe're especially interested in items that will appeal to our lovely local friends in Yorkshire, in the North of England.
Wholesome contentThe emphasis is on the word “wholesome”, so suggestions pointing to pornography, commercial products and services, mainstream consumerism, partisan politics, pseudoscience, conspiracy theories, prejudice, clickbait, scams, or fake news, would be a total waste of time for all concerned.
Our aim is not to scour web sites, but rather to produce a quality resource from the individually hand-curated suggestions and recommendations of users.
We can crawl and index plain text and html (web pages); pdf, doc, docx and odt (office documents), and epub and mobi (e-books), as long as the content is freely-accessible and not protected. We're especially interested in long-term trends, and well-written and thoughtful articles that give the reader something useful to add to their toolkit, or repertoire, and take away with them.
Unfortunately, due to their restrictive terms and conditions, we can't crawl and index social media, such as Facebook; YouTube videos or animated images; nor material behind subscription paywalls.
Please note that obfuscated or shortened web addresses will not be crawled and indexed.
We're not interested in atom or rss newsfeeds, or sitemaps, since we're not going to harvest every link to other documents, or index whole web sites. Nor are we interested in news and current affairs that will soon pass their sell-by date. Most often we'll follow-up on suggestions you kindly make, and index a web page and the web site's home page, if different, or anything else that catches our eye when we're reviewing your suggestions.
Please remember that these are early days, and xyzzy! is very much a work in progress. Many thanks for your patience, understanding and good humour!
Your suggestionsWe're very interested in your suggestions, so please feel free to submit them to us for review! Don't forget, we're looking to index individual, quality web pages and documents with readable text, not whole web sites. Thanks in advance.
The databaseFor the technophiles, we maintain database tables for suggestions, sites and urls, and now use Lucene as the main indexing and search engine (using node.js, Edge, C#.NET and Lucene.NET for indexing, and PHP, Mono, Java Bridge and Lucene for search).
We also maintain a cache of favicons, and converted copies of documents are cached on-demand whenever a user chooses to view a cached copy of a web page, document or e-book.
SearchingThe search text box has a maximum length of 100 characters and it allows you to search for up to five terms. Each word outside quotation marks counts as one term. If a phrase is in quotation marks, that phrase counts as one term.
The radio buttons beneath the search text box allow you to search the web in general (which we have crawled and indexed ourselves). Or you can search Wikipedia (which queries Wikipedia's own database).
You can search by word frequency (the number of times word/s are used in a document). Or word proximity (for documents containing each word located within 10 words of the others). Or closer (proximity within 6 words of the others).
The results returned are ranked by score, with the highest scoring document matches appearing first. Up to 200 results are returned, in up to 10 pages containing up to 20 entries (with links to the pages at the bottom of the results). If there are more than 200 results, the additional results will not be shown, and you may need to refine your search.
Dictionary and thesaurusIf a word or phrase (eg love affair) is found in our dictionary (WordNet and VerbNet) or thesaurus (Roget's), then matching entries will be shown at the top of the first page of results. These can be expanded and hidden using the 'See more' and 'See less' links.
Creditsxyzzy! is powered by Windows 10, Lubuntu 18.04 ‘Bionic’ LTS Linux, Apache, Tomcat, PHP7, node.js, Redis, SQlite, Lucene, C#.NET, Mono, LibreOffice, Calibre, WebODF, epub.js, WordNet, VerbNet, and Roget's Thesaurus 1911 (Augmented). And it is proudly made in – and served from – Selby, North Yorkshire in England.
ContactIf you want to get in touch, please email Eric Twose using eric.twose AT btinternet.com (replacing “ AT ” with an @ sign).