On this page, you will find an in-depth analysis of a the Scoopler search engine.
The purpose of this webpage is to provide you with my analysis of the real-time search engine Scoopler. I will take you through some features I liked and disliked, reconstruct a search experience and discuss special syntax, give an overview of the quality of experience and finally, rate the information and delivery on a variety of factors on a scale from 1-10. I encourage you to go to Scoopler yourself to check it out after reading this, or even as you read this.
Other helpful pages regarding Scoopler
Overview of Scoopler
Scoopler is a real-time search engine. Scoopler was created by AJ Asver and Dilan Jayawardane, and launched in Private Beta in April 2009. Scoopler gets its results real-time from a variety of services such as Twitter, Digg, Flickr and others. Scoopler is funded by Y Combinator.
It aggregates and organizes content being shared on the internet as it happens, like eye-witness reports of breaking news, photos and videos from big events, and links to the hottest topics of the day. It does this by constantly indexing live updates from services including Twitter, Flickr, Digg, Delicious and more. When you search for a topic on Scoopler, it gives you the most relevant results, updated in real-time.
- The site performs a real-time search of the major social networking sites (mentioned above)
- Search results continue to update in real time as long as the search results page is shown
- Users can view the most popular search terms being used on the site in real-time
- Users can filter the most popular results for the search term by All Content, Video, Links or Images
- Users can easily share a link from the search results with over 70 social networking sites - it shows up when you hover over the search results
- Users can use the Peek function to see a preview of the site corresponding to the search result without visiting the site
There’s a new trend that starting to sweep the web: Real-time. Everyone wants access to information as it happens instantaneously. FriendFeed recently went real-time and now Facebook is starting to embrace it. That’s the idea behind Scoopler is a search engine that searches many of the popular web2.0 websites and real-time sites all at once.
It essentially allows you to search on Twitter, Flickr, Digg, Delicious and more from a single place. It is no doubt that Google is real search giant however the Scoopler is better in a difference sense. It is a services that provide tools for monitoring trends or topics. Oh, did I mention that this is all in addition to content search like links, images, video?
I definitely recommend Scoopler as a source of valuable real-time information from the web. However, I believe its objective to be browsing for information to see what is being discussed right now, but not to conduct research due to the inability to more specifically define searches and unclear sorting of returned results. Here are a couple more (positive) reasons for why I recommend using it and would give it a score of 7/10 (10=highest).
- There are other tools like it out there, but many of these only index Twitter and there is a lot of additional information to be had from following sources like Delicious and Digg, like Scoopler does.
- Instead of just showing the top trending words or hash tags, Scoopler shows the actual headline so you get a glimpse at the story behind the trend. This adds a lot of valuable information.
- Scoopler combines content search (links, images, videos) although with real time information from sources like Twitter and Digg. It is a well laid out tool that makes browsing for new information easier through categories.
- You don’t have to leave the search results to check out a video, link or image. When you press the “peek” button which appears next to content when you hover over it, the content opens in a preview window in. This is very convenient.
- Peek's window is almost a full-size preview of the page being shared.
- Scoopler track the number of searches you've made , it also display the hot topics currently being discussed on Twitter and the best part is the ‘live update’ feature!
- After every 10 second or so Scoopler updates itself to see if there's anything new being added that matches your search terms.
Target Audience: Scoopler is aimed at people who are searching for eye-witness reports of breaking news, photos and videos from current events, and links to the hottest memes of the day.
Below is a more in-depth analysis examining Scoopler
Reconstructing the Search Experience
Let's walk through an actual search on Scoopler with the search term [Google] because it's been in the news recently about acquisition of Adcom. In fact, it is such a hot topic that on the front page, it is on of the seven hot topics to be discussed. But, let's do a query search with a simple [Google] first and see which syntax can be used in Scoopler to yield a search result similar to what is already provided to us.
1. Facebook Design Flaw Abused, Hundreds of Groups Hacked
2. Google Acquires Mobile Ad Network Admob for $750 Million
3. Tweetdeck for iPhone Adds Facebook, Landscape, and Video
4. Happy Veterans Day to all that have served and those…
5. BAD Romance O Clipe Mais Perfeito DO
6. Call Of Duty Modern Warfare 2
7. Invites Google Wave
Query Test: Variety & Usefulness of Special Queries
The syntaxes tested below seem to pertain to both the content and live search portion of the website, but I only focus on the results turned by content in this example
Entering [Google] yields this results page.
The first return is "Google Cuts Prices of Cloud Storage, Increases Cap to 16 Terabytes", not wanting to see any information about it, I tried [google -cloud] and sure enough, the results page got rid of the news including cloud.
Since the objective is to look for news concerning adcom, I entered the query [google +admob], while this will return 83,100 results on Google itself, in Scoopler, it gave me this. It gave exactly the same returns for = [google and adcom] or [google AND adcom]. This is because in Scoopler, a space is already viewed as an "AND", its URL shows the query http://www.scoopler.com/search/#google+admob for a basic [google adcom] search.
Queries that gave different results include [google acquires adcom], [google, adcom]. Two things bugged me about these queries:
- First, when I spelled "acquires" improperly, say as [acquir] it does not give me any spelling or similar suggestions but just tells me Nothing Yet.
- Second, the query [google, adcom] in Scoopler apparently means exactly THAT, the comma included. For me, typing the comma is a reflex sometimes, and having it seems to list very different results than the regular [google adcom] syntax.
Let's try some other queries. It is difficult to compare it to Google or other search engines in terms of number of returns because it does not tell you the number of search results returned. In fact, you are always limited to just one result of content (while you can see "older content" in the "live post" portion to the right). However, besides [and], [+], [-], and ["…"], none of the following works. They all lead to the "Nothing Yet!" page.
- contains the word in the page title [intitle: google]
- contains multiple words in page title [intitle: "google admob"]
- information on acquisitions for google [~acquisition google]
- search only one website [google site:news.google.com]
- URL contains this term [google inurl:org]
- pages that link to this page [link:news.google.com]
- Although not relevant, to this search, syntaxes like definition, weather, information about stock symbol also do not work
This supports my recommendation that it is not intended for search purposes, but only for browsing. So since it is a real-time. Scoopler's search engine is very simple.
In terms of variety & usefulness of special queries, there aren't very many, but the useful ones (+, -, "…") pertaining to browsing the news do exist. If I have a general idea of what I am looking for, I can find it.
Quality of the Experience
The Scoopler site has a very Web 2.0 look and feel. The home page is dominated by a search bar, with links to the seven current hottest topics and some hot images and videos to the right. The site is very responsive and the results are returned in surprisingly little time for a real-time search. A given search at a given time can take as long as 10 seconds, however, depending on the activity on the site and the activity for the search term. The Peek and Share functions work quickly and seamlessly. Navigation of the site is intuitive.
Responsiveness (7/10) I guess I am a bit spoiled by the fast speed of Google, but all things considered, Scoopler does a good job in terms of speed and I like the "loading" bar it has which will allow me to visually engage how much time is left, which is usually no more than a few seconds
Paid Links (10/10) There is no advertising displayed on the Scoopler site.
Information Availability & Delivery,
Content (8/10) As mentioned in the demo, the site breaks its header search bar into eight categories, which is conducive to browsing for news. I dislike the inability to search for specific content within a particular category however. That is, for instance, if I am looking at the business category, I cannot do a search for "German soccer" within that category. It'll just perform a general search
Format of Results (9/10) For more information on this, please see the demo. The division between content and Live Post which updates in real-time much faster than the content is helpful. The content is further divided into all, link, image, video.
Delivery Form (6/10) This is underlying technological infrastructure within which the information is delivered (email, text message, Web page, or RSS feed). Scoopler is in web page form, but with a "feed" component as the Live Posts and contents are being updated constantly.
Automation (2/10) There is no automation directly associated with Scoopler. In BIT330, we learned about Google's its Google Alerts function or any other similar type of function but Scoopler, as a relatively new engine for one, and more importantly, as a real-time engine does not have such a feature. Because the content and especially the live posts are constantly updated, it would be better for the user to just go to a saved page if they want to see new information. Scoopler does save the last seven search queries you enter, but because you don't have an actual account, these are lost on your next visit if your cookies are cleared.
Frequency of Update (10/10) For the Live Posts, frequency is very apparent, especially when Twitter is involved. That is why I am thankful that Scoopler has a "pause" button for that portion so I can take more time to look through them. Again, I wish the search technology could be more refined such that I can search specifically within those Live Posts. For content, frequency is also very good. For the topic above [Google Admob], the results are as frequent as something posted 4 minutes ago.
Targets on the Web (5/10) Video seems to be dominated by Youtube, image dominated by Flickr, links have diverse sources from news to blogs. I would like to see more variety in the image and video searches.
As I mentioned in the demo, it is unclear how the search returns are structured (it's a combination of time of update and number of shares). For content, there is only one page of returns at any given time, but the system tries to filter the best results by having the most recent and most shared at the top followed by the most recent news.
Quality of Coverage (6/10) It is good in that it covers links, videos, images, and real-time posts such as Twitter. However, because of its sorting process, it is likely that all the news stories surrounding a recent event are likely to say the same thing. So coverage is good in terms of breadth, but not necessarily depth.
Opacity (6/10) Clear for Live Post, slightly less clear for content. The Live Posts are from a clear set of websites, the most popular ones being Twitter, Digg, Delicious. However, it is uncertain where content comes from.
Search Targets (7/10) It searches for images, news, blogs, videos. It does not seem to return results for pdf, books, financial, map, scholarly etc. However, I think this is expected given its intended function.
Indexed Information (Full Text vs Directory?)
It is a full text search engine.
Relevant vs Retrieved(10/10) Because of its filtering methods, all of the link results are very good for the search Google Admob. The image and video are understandably lacking for this particular query.
Precision (9/10) The results of a search engine query are said to have high precision if a high percentage of the retrieved documents are relevant to the user's question. Scoopler has high precision because it uses the number of people who "shared" the article as a form of determining what the searcher wants.
Recall (3/10) The results are said to have high recall if the search engine returned a high percentage of the relevant documents from the whole document base. I cannot tell how many total search returns are there given that it only lets me view one page.
Scoopler is a very good real-time search engine when it remains within its specified realm. specifically, I believe its intended functions are:
- Generate information about what is going on in the web right now.
- Explore what people are talking about and what they're saying
- Monitor change in what is hot
Its interface is easy to use and information is easy to find. I would recommend it to anyone interested in exploring what people are talking about in the cyber web.
The website offers a wide variety of features, such as a search for videos, images, and links in addition to features unique to real-time search engines, such as searching Twitter, Digg etc. Due to the existence of RSS feeds however, many of the information I want on the internet already comes to me. Thus, I do not know how useful a real-time search engine truly is to someone in terms of structured content. However, it is interesting to see what people are talking about on social websites and view, for instance, news from a different perspective.
Scoopler gives you live updating real-time results across a variety of services. These include Twitter, Flickr, Digg, Delicious and others. You enter a query and the middle field on the page returns auto-updating results based on information coming in. It also gives the hot search topics and popular content from around the web. It’s a pretty nice view of what is happening on the web at any given moment.
The problem here is that a lot of popular real-time results are completely dominated by Twitter. When this happens, it’s nice to have the results auto-refresh when new updates come in, but really this isn’t much better than actual Twitter Search.
My favorite element that’s really nice, is the “Peek” feature, which allows you to take a look at pages being linked to in an overlay on top of the results, so you don’t have to visit the actual page. There is also an easy way to share a result right from Scoopler.
If you are real resource lover, Scoopler saves you the trouble by aggregating information from around web, but I would love it more if I could clearly define where my search is being conducted. Otherwise, Twitter will dominate all the real time searches. It use an Ajax technology so you never need to refresh the page. It will update search in every second.