tech-ed collisions

Google's real time search

Well it was only a matter of time really. Here's a nice video of Google'e real time search: In the past I have talked about searching the past and the present, drawing a distinction between traditional search engines,which basically crawl 'historical' content (even though it may have been created quite recently, and searches on services like Twitter, which capture what people are thinking or experiencing 'now' (or pretty dammed closed to it anyway). These services have highlighted a bit of a challenge to the old search engines but obviously one that the engineers etc over at Google were up for. Cheers

technorati tags: , ,

Filed under  //   Google   search  

Searching the past and the present

Google's great as a search engine when it comes to the past but what about the present? I'm not simply talking about what you might find on Google News as they trawl through traditional media (eg newspaper) sites and link to articles that have appeared on those sites (quite recently) but more about what people are thinking and experiencing right now. When I want this sort of information I go to Twitter. This post from the Search Engine Watch blog covering the Web 2.0 Summit in San Francisco describes what Microsoft is up to in this area with Bing. Looks like they have signed deals with Twitter and Facebook. On the Twitter side of things, it seems that there will be a Bing Twitter search using information such as number of followers and retweets to help filtering. You might be able to check out the beta at http://www.bing.com/twitter/search. I tried and it didn't work - got a message saying 'Twitter search results not currently available' but at least you can look at some sort of Twitter tag cloud. On the Bing blog I think I found out the reason why - its only available in the USA at the moment. If so, the error message could have been a bit more helpful. When I first saw the SearchEngineWatch post I sort of jumped to the conclusion that the Twitter search results would be available in their general search but the Bing blog suggests that it is a separate interface just searching Twitter. If so, then I don't see what the big deal is as there are plenty of other ways to search Twitter.

technorati tags: , ,

Filed under  //   search  

Google vs Bing - which will you use?

My Twitter feed is starting to get a heap of activity about bing, Microsoft's new search engine, so I thought I better check it out. There's an interesting conversation going on over at Robert Scoble's friendfeed - must be great to have so many connections - but even with all the opinion going on, it's still good to kick the tyres yourself. On launching bing in my browser I was presented with a great image of Santorini and now I just want to go on holidays there. I tried a couple of searches and the first thing I noticed was the speed - its 'google-like'! It's also great to have a nice, clean, simple search interface, unlike some of the horrendous offerings on some of the portals that are seen as competitors to Google. I have only tried a few searches so its hard to comment with authority on the quality of the search results but they seem generally ok in the same way that Google's are generally ok. Context is still an issue for these search engines. Try doing a search on 'bing' and you will see what I mean. At the time I tested Google seemed to have better search results for me than Bing did when I used 'bing' as the search term. Next thing I noticed was the lack of ads down the right side which was a big plus for me. However, that was when I searched for 'eportfolio', a keen interest of mine. When I then searched for 'bing' there was a sponsored link at the top of the search results, and it was for something totally irrelevant to me (sorry Bing Lee whoever you are and whatever you are trying to sell on eBay). Ads can be annoying but when they are totally irrelevant and in a prominent position they can be really irritating. Some work to do there I think - at least its all Beta so it has a ready made excuse. There is an attempt at providing some context through related search suggestions (down the left side) but no joy there at all for the afore mentioned 'bing' search if you are interested in Microsoft Bing. Google also has some really cool ways of displaying results now (discussed previously here) that Bing doesn't match. So what else does it offer? There are a number of links to other Microsoft related services such as Windows Live and ninemsn. The latter one is a bit of a turn-off for me as I really don't like using a search engine provided by a major (in this part of the world) traditional media outlet. This is one of the main reasons I don't use the local Yahoo! offering unless I am getting no joy on other search engines. When I compare the various menu options available from Google and Bing I probably get to the real reason why it will be difficult for me to switch search engines from Google to Bing. I am a user of a number of Google's other services and it is just too convenient for me to have Google as not just my default search engine, but my default home page in my browser. This is going to be a real challenge to Microsoft - how do you convince people to switch search engines to you when it's not just search that people are invested in? I guess if you are a Windows Live user then this decision would look quite different. Unfortunately for me at the moment there doesn't seem to be a really compelling reason to switch search engines but I am really hoping that someone, maybe Microsoft with Bing, can offer some serious competition in the search marketplace.

technorati tags: , ,

Filed under  //   Google   search  
Posted June 1, 2009

Nice new search options from Google

Just the other day I was trying to find some reviews on a particular software product but was really only interested in quite recent ones. As the product in question has been around for some years there were plenty of reviews about it but I had a really difficult time finding recent ones with Google. When I got to work this morning and checked my Twitter feed I found an interesting tweet leading me in the direction of Google and their 'Searchology' event held this week. At the event they announced some new features including 'Search Options' and 'Rich Snippets'. Search options looks really interesting and provides some great ways of examing your search results and helps you get answers easily to questions like my above mentioned problem. The following video from Google provides a nice overview of all that this new feature does: As well as looking at recent results you can look at videos, forums and reviews. Results can be displayed in different ways such as visually or on a timeline (for me its interesting to look at the timeline search and compare it to how we have implemented a similar feature for one of our customers). Rich Snippets looks interesting in that they:

extract and show more useful information from web pages than the preview text that you are used to seeing. For example, if you are thinking of trying out a new restaurant and are searching for reviews, rich snippets could include things like the average review score, the number of reviews, and the restaurant's price range"
Rich snippets rely on other webpages using microformats and RDF to provide information to Google. Nice to see them looking at open and existing specifications here. For our own benefit, I think we will need to have a look at how this feature can enhance the services we provide for ourselves and our customers.

technorati tags: ,

Filed under  //   Google   search  
Posted May 15, 2009

Great discussion on openDSM and open source

A couple of us at education.au have blogged previously on openDSM, our distributed search manager software that was released recently as open source and is housed on Google Code. If you would like to know more about this software, the problem areas that it addresses and the philosophy behind it and some of the work that we do, here is a really interesting podcast with the current lead developer/architect, Nick Lothian. There is a great discussion on open source more generally too. Cheers, Jerry.

Media_httpblogseducat_frcjd

technorati tags: ,

Filed under  //   search  

Open source distributed search manager ready to go

For a number of years now education.au has been providing 'search' services to the education community in a number of its projects. The best known of these is 'edna'. edna's search functionality started out as a collection of metadata records describing web resources that would be of value to people in the education and training sector. With the explosive growth of the Web it soon became obvious that many more high-value and relevant resources were being created than could possibly be catalogued (using agreed metadata application profiles and controlled vocabularies) by a single organisation. At the same time a number of other, related organisations were populating their own repositories with similar records. edna, in agreement with the owners of those collections, started harvesting metadata from them and making all these resources discoverable from a single search. For a number of reasons, it was not practical or possible to harvest metadata from a number of other repositories however the notion of having a single search interface to many collections was, and is, very compelling. education.au set about developing software to achieve this. The result was a software application known as DSM (Distributed Search Manager). A number of our other stakeholders were facing the same challenges and DSM was deployed to address these challenges too. DSM is now used in a number of Australian education related search services and has also been used in New Zealand and other industry areas (such as natural resource management) where the same issues are being faced. The projects we have engaged in have enabled us to gain considerable experience with a number of repository interfaces and also standards/specifications that are commonly used in Search and repositories (especially in the education sector). The software has evolved to a point where we believe it could be of great benefit to other organisations facing the challenge of searching across multiple repositories, databases and search engines. We are very pleased to release this distributed search manager as open source and it is now available and known as openDSM. openDSM is located on GoogleCode at http://code.google.com/p/distributed-search-manager/. We encourage those that are interested to have a look at DSM, see it in use(the demo application simply accesses three open collections however we have developed adapters for approximately forty different collections so far), deploy it or even contribute towards the ongoing development of it. We hope that this software will be the first of a number of technologies that we can contribute as open source to the education community in particular.

technorati tags: , ,

Filed under  //   search  

Metadata still matters

Metadata is used in many applications and has potential in many more emerging areas but I am going to concentrate simply on 'search' here. Those of us in the repository space will be familiar with the metadata vs text indexing argument and comparisons with Google. This blog has been active for a few months now and it is interesting to examine how people are discovering it. There is no real surprise for me that the top referring source is our company website so at least some of that website's navigation could be seen to be effective. The second largest referring source is (again, no surprise) Google. However this is where things start to get interesting. The most popular search term from Google to this blog is 'Youtube'. Given that I loosely comment on technology and its application in educational environments this is probably ok. However, the second most popular search term that people are finding their way to this blog is 'how to braid hair'. Now one thing that this blog is not about is hairdressing however in one post I mentioned a site called viewdo which features community contributed videos on how to do things. In that article the words 'french braid hair' appears as an example of what you can see on viewdo. So what does all this mean for me? Firstly, Google is indexing my blog which is a good thing. However, there appears to be no real context applied to the text that is being indexed. Using metadata, I can apply that context and would be very unlikely to have any metadata or tags mentioning braiding hair when describing the content in this blog. Given that the blog is being indexed by generic search engines I need to be careful in my choice of words. The amount of times I have now mentioned braiding hair, the distance between occurrences of those words etc is probably going to result in this post being accessed by any number of aspiring hairdressers (that word is going to cause problems now too) and parents who may not have much interest at all in technology and education. For those unfortunate enough to have been misdirected here, I hope they/you manage to read enough of the article to find out about viewdo and can ultimately find what you are looking for there. It's a pity I don't have any downstream analysis as I will never know whether they did or not. By the way, Wii was the third most popular search term. At least I have an idea of what people are looking for.

technorati tags: ,

Filed under  //   search  
Posted March 1, 2007