A consequence of that may be that Google decided to move Google Realtime Search to a new address. It is not clear from the announcement if the realtime results in the main Google search engines remain; We were not able to find any at this point in time though.

Realtime search is currently rolled out in 40 different languages, and it will take some time before every user can access the new address www.google.com/realtime.

A temporary address has been created to give everyone the ability to test Google Realtime Search right away. Users just need to follow this link to do so.

On the new homepage you’ll find some great tools to help you refine and understand your results. First, you can use geographic refinements to find updates and news near you, or in a region you specify. So if you’re traveling to Los Angeles this summer, you can check out tweets from Angelenos to get ideas for activities happening right where you are

google realtime search

In addition, we’ve added a conversations view, making it easy to follow a discussion on the real-time web. Often a single tweet sparks a larger conversation of re-tweets and other replies, but to put it together you have to click through a bunch of links and figure it out yourself. With the new “full conversation” feature, you can browse the entire conversation in a single glance. We organize the tweets from oldest to newest and indent so you quickly see how the conversation developed.

Finally, we’ve also added updates content to Google Alerts, making it easy to stay informed about a topic of your choosing. Now you can create an alert specifically for “updates” to get an email the moment your topic appears on Twitter or other short-form services. Or, if you want to manage your email volume, you can set alerts to email you once per day or week.

google alerts

So, Google Realtime Search in its current stage is based 100% on Twitter. It is probably safe to say that this has an impact on the reach and quality of results. Twitter is not popular in every country on the globe which reduces the functionality and use of the search engine for users from those countries.

We do think that Google made a good decision to move realtime search to its own address, especially if that means that realtime search was removed from standard Google search.

What’s your take on this development? Let us know in the comments.

Update: Realtime Search on Google has been removed, at least from this website address. There does not seem to be a new address for it currently.

Factery Labs uses a different approach. The realtime search engine tries to concentrate the results on facts rather than popularity or time.

Factery has developed FactRank technology to mine real-time feeds and web content for the purpose of providing simple, short answers to user queries

A user looking for prices for Apples’ tablet computer iPad for instance will see lots of results that detail those prices. A list of preloaded searches and results are constantly updated automatically with the option to pause the updates. Custom searches can be added to the list of searches that are constantly updated on the right side of the website.

The concept itself is interesting but the results are sometimes not related to the search query. One of the default searches for instance was “where is chuck norris” with the answers concentrating on the who instead of the where.

The Factery Labs search engine can be useful for users who want to access facts about a specific term in short time and it could become especially useful for searches that static websites cannot provide answers for.

All of these realtime search engines have in common that they retrieve data from various sources upon the user’s request. The data is usually fetched from services such as Twitter or Friendfeed but also from blogs, social media sites and podcasts. Differences mainly exist in the way the results are presented to the user and the filtering options that are provided.

Feedmil

Feedmil offers a few features that the other realtime search engines do not offer. It comes with a slider that can be used to define the popularity range of the results. It is for example possible to concentrate on popular sources for the results only. The service can also filter results by type so that only results produced by a certain type of website or service are displayed. The types include blogs, micro-blogs, podcasts and social media among others.

Feedmil divides the search results page in two columns. The main column displays search results based on time and popularity while the sidebar column displays hot feeds related to the search term. The presentation of the results is excellent although they seem to be filtered before they appear on the page (we would have expected more Twitter results of the past four hours for a popular search term such as Google). It would also have been nice if the pages would not have to be reloaded when switching between the various news sources.

OneRiot

OneRiot looks almost like a conventional search engine. It mixes search results from various sources on the search results page. Sources used include blogs, Twitter, forums or Friendfeed and popular websites such as Userscripts or Digg. The results can be sorted by Pulse or Realtime with Pulse displaying popular results while realtime concentrates on fresh results. The results page looks a bit messy which can be mainly attributed to the avatar icons and the two different link colors that are used by OneRiot. The results look almost like Google Adsense ads which by default also use two different colors for their links.

Scoopler

Scoopler divides the results in three different columns with two showing results related to the query and one displaying the hot topics of the day. The main column displays the search results in realtime which becomes apparent as the posting time is usually less than five minutes ago. It will display results from a variety of sources including Twitter, Youtube and blogs. The main column gets updated that frequently for popular search terms that it is almost impossible to keep up.

The right column displays popular results and can be filtered to display only images, videos or links.

The auto updating page strongly emphasizes the realtime aspect of the search engine. It would have been nice if there was a way to pause the search results from being updated or to add certain search results to a personal area where they could be accessed at a later point in time.

Collecta

Collecta is the final search engine of this article. It will also display realtime search results and auto-update the page just like Scoopler. It comes with a pause function which is very handy as the results are displayed in rapid succession if a popular search term has been entered. The service uses four different categories that can be filtered in the search options: Articles, Comments, Pictures and Updates.

The results are displayed in the main column of the page. A little bit irritating is the orientation of the page as it is aligned on the left side of the screen and not centered. A click on a result will display the full message in the right column from where it can be viewed and processed further.

A way to add multiple results to the right column would have been nice. As would have been if the source of the result would have been displayed in the main column.

Verdict

Realtime search engines have one problem: They can either display all results which are impossible to keep up with for popular search terms or display only popular results from sources that they have picked (e.g. Twitter accounts with a follower count of at least 200). Both options will dissatisfy some users. It would have been nice if the services would offer a better customization of the results by offering the user to add individual sources to them. Letting a user add Twitter, Youtube or Friendfeed users and channels, blogs and websites or social media sites could be interesting for many users as this would take the RSS reader concept to the next level by mixing it up with other sources that have been handpicked by the user.