woensdag 30 september 2009

Is Autonomy a Microsoft Target?

Must be something going wrong with implementing Fast.

Talk is swirling around the largest listed company in the Cambridge cluster of
high tech companies, Autonomy
Corporation plc, that it is an imminent target of a takeover offer from
Microsoft.
Talk in financial markets, (reported here by Reuters),
is that the Seattle giant may make a bid at around 2,800p a share, which would
value Autonomy
at about £6.7 billion, about 75% above its current share price of 1513p and
market capitalisation of £3.9 billion.
Autonomy's
share price has been up between 3.5% and 5% in Friday morning trading.

Source: http://www.siliconfenbusiness.com/articles/Is-Autonomy-a-Microsoft-target-at-north-of-6-billion/734

dinsdag 22 september 2009

Google and Meta tags...

Today the whole blogosphere buzzed about the fact that Google does not take into account the fact that website owners use keywords in the HTML meta-tags to describe the content.

Even Beyond search has quaked about it.

Google has reacted to this news very quickly, stating that their Enterprise solution does use this kind of metatags.

The reason for not using user generated meta tags in the ranking mechanism in Google websearch is simple... it is sensitive for manipulation.
Google websearch relies on the popularity of webpages and websites. Furthermore Google believes in the volume of data, not per sé the quality, to be able to find usefull and recent subjects that determine the relevancy of content for users.

As we al know, the volume and popularity factors are not very usefull in an enterprise environment. There, it revolves around the finding of one specific document or relevant information that has been revised or tagged by other humans.

In the enterprise metadata is simply necessary to categorize and give context.

Still we must make sure take this kind of news does not make the enterprise user even more "dumb" by expecting to find relevant results in there company while they don't invest any effort in "upgraing" the quality of the data....
They all want it "Google-like" but in this case they do not understand what that means.

woensdag 16 september 2009

Google Fast Flip is essentially Google News in a different format

Today Google announced a new Labs experiment: Google Fast Flip.

Google already has an outlet for news, namely Google News.

In my opinion Fast Flip is the same news outlet, only in a different interface. An intuitive interface I must add.
But the Google News interface already has a "newspaper-like" shape. To scan the headlines I like Google News better. This also has to do with the fact that I can add subjects and content myself.
Further more I can arrange the frontpage the way I like it.

What's your opinion? Just another funny interface the guys from Google are playing with, or is it really revolutionary???

dinsdag 15 september 2009

Reasons for choosing Solr on all for good twisted

Today I read a blogpost on the Lucid Imagination blog about the fact that a Google employee chose Solr as the search engine for the site.
The blogpost cited a part of the testimonial on the allforgood site on with I have to disagree with the reason for choosing Solr

The problem they had was:

One of the top concerns we’ve been hearing from nonprofit organizations who list
volunteer opportunities on All for Good is that their opportunities aren’t
updated on the site as frequently as they need. This happens because All for
Good doesn’t directly receive volunteer opportunities from nonprofits – we crawl
feeds from partners like VolunteerMatch and Idealist just like Google web search
crawls web pages. Crawlers don’t immediately update, they take time to find new
information.

The solution stated:

Today, we’re rolling out improvements to All for Good that will help solve this
problem and improve search quality for users. The biggest change, which you
won’t see directly, is that our search engine is now powered by SOLR, an
incredible open source project that will allow us to provide higher quality and
more up-to-date opportunities. Nonprofits should start seeing their
opportunities indexed faster, and users should see more relevant and complete
results.


Now... why do I disagree with the way the choice for Solr is argumented?

It is the fact that the use of Solr solves their problem of "latency". Remember the biggest problem was that the indexed information was not up to date.
Solr doesn't solve that. Solr is just a service around Lucene. Solr doesn't take care of the crawling part of the problem.

Us experts on search applications and information access solutions know that it is the combination of crawling frequency, the accessibily of the source that has to be indexed (RSS, Web, document repositories, databases etc.) the preprocessing of those diverse formats that determine the speed of the indexing and thereby, search process.

In this case probably Nutch will take care of the crawling part, so the frequency with which updates are processed rely on the speed of that part of the solution. Not the fact that Solr is used...

donderdag 3 september 2009

Solr next new thing for Europian Parliaments

According to some representatives of European Parliaments (UK, Europe, Belgium) , Solr is the next new thing in the field of search solutions for their websites and intranet search.

Today I was at a workshop focussed on the topic of search. The workshop was organized by the dutch Parliament and there were representatives of 8 parliaments of other European countries (Belgium, UK, Denmark, Norway, Sweden, Finland, Austria, Israel).

What stroke me there was the real interest in Solr as a solution for there search needs. I know that Solr is really upcoming but still has a "techy" ring about it... so I thought.

It seems that Solr is on its way to becoming a standard in government land. Faster than I had thought it would be.

Driven by the need for Open Source software and lower license costs, many organisations are willing to try out Solr. They take the absence of support from the vendor (there is no) for granted and are investing in own knowledge or that of implementation partners.

One reason for this "gamble" was very sensible in my opinion:
The field of search is really evolving. Not only from a technical point of view but also from a vendor and marketing angle.
We choose Solr because it is cheap, it is still being developed very fast, and our need for functionality is simple now. At the time we want more functionality, the product probably will have it. It therefore grows with our needs...
When after 3 years it seems that we have made the wrong decision we still haven't thrown away much money. We can than always switch.
I must agree with the choice. Why pay EUR 100.000,00 for a product that promises everything while you are using part of it.
Of course Solr is free but the implementation will cost you money. But even if you spend 50.000,00 on customising Solr it it still half as cheap.

Seems like the way to go for me...