Jump to content

Singingfish: Difference between revisions

Edit links
From Wikipedia, the free encyclopedia
Browse history interactively
← Previous editNext edit →
Content deleted Content added
VisualWikitext
Line 71: Line 71:
*Because Singingfish came onto the scene at about the same time as [[Big Mouth Billy Bass]], many people in 2000 and 2001 assumed that the former was the maker or the name of the latter.{{Fact|date=October 2007}}
*Because Singingfish came onto the scene at about the same time as [[Big Mouth Billy Bass]], many people in 2000 and 2001 assumed that the former was the maker or the name of the latter.{{Fact|date=October 2007}}
* Singingfish used Oracle database technology for storing and aggregating multimedia metadata. Originally, the search engine itself was based on Oracle InterMedia Text. Later, for price/performance reasons, Singingish feed metadata into dedicated indexing and search servers hosted by [[Inktomi]]. Some time after the acqusition of Singingfish by AOL, the search function was rehosted on AOL servers using AOL's PLS/CPL search engine<ref>http://web.archive.org/web/19980113101346/http://www.pls.com/</ref><ref>http://www.searchtools.com/tools/plweb.html</ref>.
* Singingfish used Oracle database technology for storing and aggregating multimedia metadata. Originally, the search engine itself was based on Oracle InterMedia Text. Later, for price/performance reasons, Singingish feed metadata into dedicated indexing and search servers hosted by [[Inktomi]]. Some time after the acqusition of Singingfish by AOL, the search function was rehosted on AOL servers using AOL's PLS/CPL search engine<ref>http://web.archive.org/web/19980113101346/http://www.pls.com/</ref><ref>http://www.searchtools.com/tools/plweb.html</ref>.
* Many multimedia URLs used as seeds by Singingfish's crawlers came from cache logs from the NSF-funded National Laboratory for Applied Network Research (NLANR)<ref name=nlanr>http://www.nlanr.net/</ref> IRCACHE Web Caching project.<ref name=ircache>http://www.ircache.net/FAQ/</ref> These caches were orginally set up and operated by NLANR to test the viability of high-level Internet caching. NLANR created and maintained nine high-level Web caches located throughout the United States. These caches were directly connected to approximately 400 other caches, and indirectly to 1100 worldwide, receiving approximately 7,000,000 requests per day from the others. Since this was a publically-funded project, the IRCACHE logs were available free of charge to anyone who knew about them; Singingfish downloaded these cache logs each day and simply searched the logs for URLS containing the appropriate media types (e.g., .WMA, .RAM, .MP3) and HTTP success codes. Singingfish also used HTTP failure codes from these logs to prioritize it's queue of multimedia URLs to revisit, in order remove broken links. Once NLANR public funding expired, The Measurement Factory<ref>http://www.measurement-factory.com/</ref><ref>http://www.ircache.net/</ref> took over the operation and maintenance of the high-level caches. A number of companies, including Singingfish, paid a fee to The Measurement Factory, for continued operation of the system and daily access to the cache logs.
* Many multimedia URLs used as seeds by Singingfish's crawlers came from cache logs from the NSF-funded National Laboratory for Applied Network Research (NLANR)<ref name=nlanr>http://www.nlanr.net/</ref> IRCACHE Web Caching project.<ref name=ircache>http://www.ircache.net/FAQ/</ref> These caches were orginally set up and operated by NLANR to test the viability of high-level Internet caching. NLANR created and maintained nine high-level Web caches located throughout the United States. These caches were directly connected to approximately 400 other caches, and indirectly to 1100 worldwide, receiving approximately 7,000,000 requests per day from the others. Since this was a publically-funded project, the IRCACHE logs were available free of charge to anyone who knew about them; Singingfish downloaded these cache logs each day and simply searched the logs for URLS containing the appropriate media types (e.g., .WMA, .RAM, .MP3) and HTTP success codes. Singingfish also used HTTP failure codes from these logs to prioritize it's queue of multimedia URLs to revisit, in order remove broken links. Once NLANR public funding expired, The Measurement Factory<ref>http://www.measurement-factory.com/</ref><ref>http://www.ircache.net/</ref> took over the operation and maintenance of the IRCACHE high-level caches. A number of companies, including Singingfish, paid a fee to The Measurement Factory, for continued operation of the system and daily access to the cache logs.


==Notes and references==
==Notes and references==

Revision as of 09:21, 17 October 2007

Singingfish
Company typeSubsidiary of AOL
IndustryInternet Search
Founded1999, acquired by Thomson SA in 2000, by AOL in 2003
DefunctFeb 2007
FateAcquired
HeadquartersSeattle, WA
Number of employees
over 50
Websitewww.singingfish.com

Singingfish was an audio/video search engine that powered audio video search for Windows Media Player[1],[2], WindowsMedia.com, RealOne/RealPlayer,[2], Real Guide[3], AOL Search, Dogpile, Metacrawler[4] and Singingfish.com, among others. Launched in 2000, it was one of the earliest and longest lived search engines dedicated to multimedia content. Acquired in 2003 by AOL[5], it was slowly folded into the AOL search offerings and as of February 2007 has ceased to exist as a separate service.

Singingfish powered audio search continues to live on for the time being at AOL Search and other AOL properties. However, little if any development has been done since August 2006. Singingfish powered video search is no longer publicly available and is now being re-directed to AOL Video Search.

Overview

Singingfish employed its own web crawler, Asterias [6], designed specifically to ferret out audio and video links across the web. In 2003 and 2004, Asterias discovered an average of about 50,000 new pieces of multimedia content a day. A proprietary system was used to process each of the discovered links, extracting metadata and then enhancing it prior to indexing as much multimedia content on the web has little or poor metadata.

History

File:Singingfish GroupPhoto 2000-02-05.jpg
January 2000 group photo

Singingfish was founded in mid-1999 by John DeRosa, Eric Rehm, Ken Berkun, and Mike Behlke. By January of 2000, it grew to 17 employees. A public alpha version of the search service was unveiled at Streaming Media East in New York City in June 2000.[7]By the Fall of 2000, Singingfish had grown to over 50 employees. In November of 2000, it was acquired by the French company Thomson multimedia[8] (now Thomson SA). Singingfish announced its first customer, Zurich-based Internet AG[9], in December 2000. Singingfish Search first appeared on Internet AG's "Swiss Search" and Infospace's Dogpile and Metacrawler sites late in the summer of 2001.

By this time Singingfish was clearly the dominant multimedia search engine. No competitor indexed as much of the audio and video content available on the web or provided more relevant query results. In the fall of 2001, Real Networks signed a deal with Singingfish for the latter to provide audio and video search for RealOne Player. In January of 2003, Microsoft signed a similar deal[10], in this case to provide audio and video search for Windows Media Player and WindowsMedia.com.

Singingfish downsized dramatically under Thomson multimedia (Summer 2001) and then slowly continued to shrink in size through its acquisition by AOL in October 2003. However, Thomson multimedia continued to support research and development into multimedia metadata standards such as MPEG-7[11][12] that were incorporated into Singingfish technology.[13]

Soon after acquiring Singingfish, AOL integrated its audio/video search service into AOL Search, adding yet another big-name to the stable of search products powered by Singingfish. As of August 2006, Singingfish continued to power multimedia search for both Microsoft and Real, and was being fully integrated into the AOL search and directed media business unit.

The last members of the Singingfish staff were laid off in December 2006[14]. As of February 7, 2007 all web hits to Singingfish are being redirected to AOL Video.

Features

Supported Formats

Partner feeds

The content providers below added their audio and video content to Singingfish's search index using regular feeds. In addition to providing high quality meta data to be indexed, these direct feeds also enabled Singingfish to produce highly current results for news related queries.

Trivia

  • Because Singingfish came onto the scene at about the same time as Big Mouth Billy Bass, many people in 2000 and 2001 assumed that the former was the maker or the name of the latter.[citation needed]
  • Singingfish used Oracle database technology for storing and aggregating multimedia metadata. Originally, the search engine itself was based on Oracle InterMedia Text. Later, for price/performance reasons, Singingish feed metadata into dedicated indexing and search servers hosted by Inktomi. Some time after the acqusition of Singingfish by AOL, the search function was rehosted on AOL servers using AOL's PLS/CPL search engine[15][16].
  • Many multimedia URLs used as seeds by Singingfish's crawlers came from cache logs from the NSF-funded National Laboratory for Applied Network Research (NLANR)[17] IRCACHE Web Caching project.[18] These caches were orginally set up and operated by NLANR to test the viability of high-level Internet caching. NLANR created and maintained nine high-level Web caches located throughout the United States. These caches were directly connected to approximately 400 other caches, and indirectly to 1100 worldwide, receiving approximately 7,000,000 requests per day from the others. Since this was a publically-funded project, the IRCACHE logs were available free of charge to anyone who knew about them; Singingfish downloaded these cache logs each day and simply searched the logs for URLS containing the appropriate media types (e.g., .WMA, .RAM, .MP3) and HTTP success codes. Singingfish also used HTTP failure codes from these logs to prioritize it's queue of multimedia URLs to revisit, in order remove broken links. Once NLANR public funding expired, The Measurement Factory[19][20] took over the operation and maintenance of the IRCACHE high-level caches. A number of companies, including Singingfish, paid a fee to The Measurement Factory, for continued operation of the system and daily access to the cache logs.

Notes and references

  1. ^ http://www.econtentmag.com/Articles/ArticleReader.aspx?ArticleID=3923
  2. ^ a b http://www.clickz.com/2175491
  3. ^ For queries on Real Guide, Singingfish search results appeared in the section "Audio & Video on the Web". The Singingfish logo was only displayed on the second and later result pages.
  4. ^ For queries on Dogpile and Metacrawler, results from Singingfish were annotated next to the relevant result(s).
  5. ^ http://searchenginewatch.com/showPage.html?page=3110961
  6. ^ http://www.janetsystems.co.uk/Articles/WebRobots/tabid/82/itemid/41/Default.aspx
  7. ^ http://www.internetnews.com/bus-news/article.php/392941
  8. ^ http://www.streamingmedia.com/article.asp?id=5991
  9. ^ http://www.prnewswire.com/cgi-bin/stories.pl?ACCT=104&STORY=/www/story/12-12-2000/0001385098
  10. ^ http://searchenginewatch.com/showPage.html?page=2165251
  11. ^ http://portal.acm.org/citation.cfm?id=357744.357887
  12. ^ http://www.us.design-reuse.com/articles/article2312.html
  13. ^ http://www.medialab.sonera.fi/workspace/MPEG7WhitePaper.pdf
  14. ^ http://www.trenchmice.com/scoop/2/
  15. ^ http://web.archive.org/web/19980113101346/http://www.pls.com/
  16. ^ http://www.searchtools.com/tools/plweb.html
  17. ^ http://www.nlanr.net/
  18. ^ http://www.ircache.net/FAQ/
  19. ^ http://www.measurement-factory.com/
  20. ^ http://www.ircache.net/
Retrieved from "https://en.wikipedia.org/enwiki/w/index.php?title=Singingfish&oldid=165148394"