Ok this is a little OT but I thought it was a nice change of pace. The BBC has recently released a new search engine. Apparently they did a survey and discovered that alot of people are dissatsified with some services out there.

What is nice to see is that the front end is in Perl. I can't speak for the back end. Given the direction a lot of companies are taking (.NET, Java etc) its good to see Perl still being taken seriously. Shame it doesn't list Perl monks yet :P

You can try it out

Or read their news article about it

It made my advocacy in the work place a little easier.

Replies are listed 'Best First'.
Re: OT: BBC releases search engine
by ignatz (Vicar) on May 02, 2002 at 22:12 UTC
    Interesting idea about the frustration with US centric results, but not an impressive execution. I would think that they would at least spider through their own pages. One of my sites got listed in an article of their's in 2000. They might at least have saved a little face and ala Yahoo had a result fallback to Google if their results turned up nothing.

    I'm wondering how you would create a decent UK centric search engine. Look for British idioms... European domains... Upcasting results with Commonwealth keywords (Cricket, Yorkshire, Euro), downcasting results with US style idioms (WWF, NFL, CIA), emphasising non US media outlets including countries such as India, Australia, and Pakistan. Tough one to pull off. 

    ()-()
 \"/
  `
[reply]
Re: OT: BBC releases search engine
by simonflk (Pilgrim) on May 03, 2002 at 05:34 UTC
    What is nice to see is that the front end is in Perl. I can't speak for the back end. Given the direction a lot of companies are taking (.NET, Java etc) its good to see Perl still being taken seriously.

    The BBC use perl on quite a lot of the site. In fact betsite was written by someone in the BBC Education department and is used on every page for the "Text only version" which renders the page without graphics and javascript in a format useful for blind people with text2speech software. You'll see that Perl is used in quite a few other places too.

    Shame it doesn't list Perl monks yet :P

    It does. Just click on the "Results from Web" tab. Or click here it's not blazingly fast, but I'd expect it to improve over time.

    I agree, it's great to see a big corporation using open source technology and giving back to the community at the same time. It's a shame there's as yet, no perl success story about the BBC.

    <sig>

    -- simonflk

    </sig>
[reply]
      Yeah I discovered the 'results from web' thing almost as soon as I hit submit and so couldn't edit it afterwards :P.

      Shame its a little slow though.
[reply]
Re: OT: BBC releases search engine
by beebware (Pilgrim) on May 03, 2002 at 22:53 UTC
    I'm 100% certain that the backend is Google somehow. Do a search on the BBC homepage and compare the results to those from Google UK. I tried it with a few select phrases and I got identical results (even the same text 'quoted' at each section). If only Google or the BBC would come forward and say 'yep, it _is_ Google' it would have been a lot more open...Saying that, there's no advertising on the site (the BBC isn't allowed to accept commercial advertisments) so I'm guessing that Mr+Mrs.licence payer have indirectly paid a nice hefty sum to Google and the BBC doesn't want people to know...
    That's my 0.02p anyway :)
[reply]
      beebware said:
      ... "I'm guessing that Mr.+Mrs. license payer have indirectly paid a nice hefty sum to Google and the BBC doesn't want people to know"...

      I should think that Google's Search Solutions are cheaper than the hardware, software development costs necessary to regularly index billions of web pages. I also don't think it matters that they advertise the fact. I don't know what brand of toilet paper they use in the Beeb. After all the licence fee pays for that too doesn't it?

      http://www.bbc.co.uk/info/bbc
[reply]

Back to Meditations