December 04, 2007
A Final Farewell to BIT330
BIT330: My Experience
This final blog entry is for all of you Ross students who are going to be Ross Juniors and Seniors next year and are looking for an interesting business school class. I want to give you a little insight into what I learned in BIT330 (Web-based Information Retrieval) and how it's going to help me in the hopes of convincing you to consider taking this class in the future.
When I read the term Web-based Information Retrieval, I immediately thought, "This sounds like another way of saying using search engines to find information." Yes, there's definitely a lot of Google and Yahoo Web in this class, but there's a lot more as well.
I came into the class thinking I was pretty decent at web searching, but I knew that there was a lot more to learn. What I didn't know was that I wasn't a good web searcher at all. Whenever I wanted to look for something online, I would go to Google and just type in as many words as I could think of that would get me the result I wanted. There was no organization to my method. Through this class, I've learned strategies to make searching more efficient; specifically, using search syntax can help you go from over 1 million results down to a few thousand or even less than a hundred. There are few that were extremely useful (and usable with both Yahoo Web and Google): site: - makes the search engine search through that specific site - and intitle: - makes the search match your words with web titles. Proper use of quotation marks helps zoning in on a specific result too since putting quotation marks around a query will signal to the search engine to only find results that match your exact phrase.
Furthermore, with BIT200 under my belt, I knew a little about blogs and RSS feeds. Now with BIT300 (almost) under my belt, I know a lot more about blogs and RSS feeds. In fact, I don't think I'm ever going to look at a blog without looking for the little orange RSS button. Bloglines, what I like to think of as an RSS one stop shop (you can search, subscribe, create, share, and read RSS feeds), has made keeping updated with blogs so easy by maintaining my blogroll all on one page. It's always a benefit not having to worry about checking a site everyday to see if it's updated because someone (or in this case, a Web site) already does that work for you. What's even more interesting is making your own RSS feeds for sites that don't have it using Dapper and Feed43. With news and updates happening all the time, RSS feeds make the seemingly overwhelming amount of information out on the web not so scary.
This class also opened my eyes to the vast array of web resources out there. I never really knew anything about using wikis before this class and I describe my experience using the wiki site Stikipad in my previous blog entry found here. Creating a wiki is a good mobile way to organize your information since you can access your wiki if you're on the web. In addition, even though I preferred RSS feeds over page monitors (using sites like Google Alerts, knowing about them and knowing how I could use them to track pages that don't have RSS feeds available will certainly help me with information trapping as well. I also learned to maximize my Google and Yahoo experiences by not just using their general searches; they have so many specific engines that target specific things you could be looking for that can expedite your search (e.g. Google's Blog Search and Google Docs, Yahoo HotJobs, and Yahoo Pipes.
I guess one of the key takeaways from this class is understanding the need to take advantage of resources that will bring you - a busy person - the information you want and organizing it into something meaningful. As the internet continues to expand and people continue to think up new ideas (especially with social networking gaining popularity i.e. del.icio.us), it's important to learn ways that will make your information retrieval manageable, which is why I know I will be using a lot of the information I learned from this class in the future. I know that I will definitely be keeping up with RSS feeds (especially when it's time to do interviews when trapping information about a specfiic company will come in handy), using different search engines and the proper syntax for those search engines, and maybe even get more involved in the blogging world by starting a blog that's not an assignment.
There is definitely a lot of information to get from this class, so it does take up a good amount of time. Therefore, the one thing that I would like to see changed is making the Wiki Assignment 1 project instead of 2. By making it one big search project, it will be easier to focus on the bigger picture and and get more in depth with the project (maybe even making the wikis more complex). I think students can still show the knowledge they gained through one wiki site, especially if they can put all their effort into something they are passionate about.
Overall, BIT330 has definitely been an unconventional class that has a lot of applicability to real life. I'm glad that I took it because I really did learn things that I know will benefit me in the future. Now, whenever I go to the fishbowl and see someone sifting through millions of search results, I happily think in my mind that I am more prepared to prevent myself from going through all that trouble again.
Posted by paulinad at 12:54 AM | Comments (0)
December 02, 2007
Wikis
Wikis, Stikipad, and Frustration
Prior to BIT 330, the only experience I had with Wikis was through Wikipedia. To tell you the truth, I thought Wikipedia was the only “Wiki” out there. Little did I know that you could actually create your own wikis. Wikipedia’s definition of a wiki is that a wiki is "a type of computer software that allows users to easily create, edit and link web pages. Wikis are often used to create collaborative websites..." Even more surprising was the fact that our BIT class would be using a wiki as the home base for the class.
Throughout the semester, the class has been using Professor Moore's Stikipad site for BIT330. It's definitely been an interesting experience having no handouts of anything, completely relying on a website for class, and witnessing a site get made through joint collaboration.
With this new experience, however, came much frustration. Not with the class or how it was online-dependent, but rather with Stikipad itself. Even though the Stikipad Home Page boasts a quote from PC Magazine saying that it's ".. one of the smoothest, most usable wikis on the net," I consistently ran into problems with the site.
Of course frustrations stemming from learning a whole new kind of writing code language was expected. You don't use HTML with Stikipad, but rather textile. It's nice that the formatting is a little simpler than HTML. For example to bold a word in HTML you would have to type <b>WorldHere</b>, but in Stikipad you just have to type _WordHere_. Another nice feature is that when you're editing a page on Stikipad, there's a little tab called "Formatting Tips" that lists some basic "wiki code." I had to rely on this a lot, which did not make editing pages fun. Therefore, if you're new to wikis and are interested in starting one, be prepared to set some time to learn this code. It's going to take some time if you're not used to code or if you're used to HTML. I always find myself trying to write in HTML whenever I'm editing a page and then remember 5 minutes later that it won't really work.
Another reason I did not like using Stikipad was how often errors kept popping up when loading a page and the often slow page loading. Many times, when I tried saving a page or clicking on a link, a 500 - Internal Server Error message would pop up. This happened EVERY time I went on the site, and it happened more often when I was in class (probably because many people were accessing the site at the same time). Because of this 500 Error, my work would sometimes get lost. At the beginning I never saved my stuff outside of Stikipad, so when this error popped up, I'd have to start all over from scratch. Eventually, I learned to save my work on Microsoft Word or wordpad. All of these errors made it difficult to follow along in class sometimes. This problem became serious the night before our second BIT exam, when nothing would load for a few hours. As Professor Moore mentioned in class, "What if the site never went back up again?" It would have been DOOMSDAY. All of our work would have been lost. Valuable lesson to be learned: back everything up!
I also did not find Stikipad as a very help friendly site. As I was tinkering with Stikipad for my BIT projects, I sometimes encountered difficulties that I needed help with. Unfortunately, I couldn't find an obvious "Help" link on my Stikipad. The only "Help" thing I found was on the Stikipad Homepage called "Support", but once you're on your own "Stiki", there's no link to be found. These are the links found when you few your list of Sikipads: My.StikiPad | New Wiki | Domain Mapping | Feedback! | Forums | Logout and once you're on it, there's a link to "Configure", but I still havent found a "Help" link. Maybe this lack of support is due to my free account (subscriptions start at $5).
Although I really like the idea of working on a website together to learn (and for a Business Information Technology class, it is highly appropriate and applicable), Stikipad hasn't been the most enjoyable tool to use. As a result, I switched over to Wikidot for my projects, and have found it to be more user-friendly.
Posted by paulinad at 12:46 PM | Comments (0)
November 29, 2007
Image Search Tools
An in-depth look
Even though Flickr has been around for a while, I’ve never used it before. Learning about in class has made me more interested in checking out its features. On its site, they claim that "Flickr is almost certainly the best online photo management and sharing application in the world,” which will be perfect for finding pictures for my BIT projects, especially for Michigan Football.
Flickr is actually owned by Yahoo, so to get an account you need to have a Yahoo one first. You don’t need to worry about paying for the service because basic accounts are free. The signing up process is typical to other similar kinds of accounts on the web.
You don’t need an account to search through Flickr’s database, but an account does come in handy for sharing, which will be discussed later on.
As you can see from above, you can input a query to start your search from the home page. The search will actually go through people’s personal pictures that they uploaded themselves and made viewable to the public. Without the public sharing photos, Flickr would not be a very good image search resource. Luckily, Flickr has an established user base of over a million users. In fact, on the home page there are updated statistics of how much uploading there has been. When I did my query, it showed: 3,018 photos uploaded in the last minute • 71,248 photos tagged with jump • 2.2 million photos geotagged this month.
To try to find more Michigan Football photos to add to my collection and to see how beneficial it is to use Flickr as a searching tool, I typed in the query Michigan Football 2007 . 955 results came back. This was a full text search, but I could have also done a tag search. By default, the results are shown to you organized as most relevant and in details form (i.e. shows user profiles and picture captions), but you can always change this to most recent, most interesting, or in thumbnail form. Another nice feature is viewing the image results as a slideshow.
I was pleasantly surprised to see many different game day pictures, including pictures that captured various people’s game day experiences in Ann Arbor. It was actually a fun experience sifting through pictures that people personally took because they were usually in grouped into sets of pictures either from certain games or from certain trips. They told stories.
As previously mentioned, Flickr is what it is because of the community that uses it to post pictures. This adds a new dynamic to the video searching experience if you become a user as well. When you become part of the Flickr community and you find pictures that catch your eye from a certain user, you can add this user as a contact. By adding them as a contact, you can easily see the photos they recently uploaded on the “photos from your contacts page” or you can subscribe to a daily e-mail that will update your contacts.
Similarly, another unique aspect to Flickr is how users can join groups of common interest to share their photos. Therefore, you can do group searches to discover even more relevant pictures that you might not have found using the regular search or might not have even thought of looking for.
Because these photos typically aren’t commercial shots, you can stumble upon some rare and really interesting pictures. For example:
Because of this experience, I am definitely going to be using Flickr as an image search tool and even as a place to upload pictures. It’s very easy to use and it offers a lot in terms of organizing and sharing with friends and family. You should consider it too!
Posted by paulinad at 09:11 AM | Comments (0)
November 25, 2007
Video Search Tools
Whenever I want to look for videos online, I automatically turn to YouTube!. It's a great site and it's easy to get into the habit of relying on it all the time, but there are other video search tools that can be useful as well. Let's take a closer look at some of them including: Google Video, Yahoo! Video, AOL Video, and Searchforvideo by conducting the general search query "Michigan Football." As part of my personal Wiki project, I want to include some game highlight videos or interesting clips regarding the Michigan Wolverines. Let's see which search tool will yield the most helpful results.
The Good
One of the first things that caught my attention, was the home page. If you have a Google account (which I recommend everyone get if they don't), Google Video places at the top section recommended videos for you to watch that, in my test run, are very relevant. Though the first video was a Michigan State University clip, the other videos (Michigan Difference, UMDM, etc.) would be something I would be interested in viewing. The rest of the home page lists some suggested videos according to categories (Popular, Comedy, Featured on AOL, Sports, etc.) that would also be shown if you don't have a Google Account. The tool bars on the right are also a nice feature because they list videos that are popular on blogs (Blog Buzz), videos that have been causing a stir (Movers and Shakers), and Top 10 videos (showing their movement in the top 10).
The simple and familiar Google interface is also good. You know what to expect and you pretty much get what you expect: an easy search with relevant results. Overall the query yielded 663 returns. The first video that popped up was a YouTube video that had highlights from Michigan game-winning touchdowns starting in 1979. In fact, the first video that wasn't relevant was the 52nd video in the search results.
The "related videos" button underneath each result makes searching more efficient because if you discover a video that interests you, you can find more that are like it by just clicking that button instead of sifting through all of the results.
Also beneficial is the RSS button at the top right corner next to the number of results that allows you to make your video search an RSS feed. This is great for future trapping.
Using Google search syntax works and it makes finding specific videos easy. For example, if I wanted to find a video with Mike Hart related to Michigan Football, I would input "Michigan Football" intitle:"Mike Hart". With this query, Google returned 8 videos specific to Mike Hart and Michigan Football.
The Bad
Because Google purchased YouTube, they have the rights to YouTube videos so their search returns will inevitably bring up many of those videos. The difference between Google Videos and YouTube is explained in this blog entry:Google Video Blog. In my search results, about 95% were YouTube videos and the other 5% were videos people uploaded onto Google Video. I couldn't find any videos from outside sources.
I also didn't like how the search results only highlighted the words "Michigan" or "Football" instead of giving a short description of the clip, but this fault isn't a deal breaker.
The Good
Yahoo! Video Search also has a simple and easy-to-use interface. When you get results, unlike Google, you can switch from "List View", which is similar to how Google shows returns, to "Grid View" which shows video thumbnails in a grid pattern. This is a great option if you want to just visualize the search results and not read any descriptors. I prefer Yahoo's list view to Google's because they actually show the summaries/descriptions of the videos instead of simply highlighting the query words.
Although there were only 156 videos returned, there was a much wider variety of sources. There were videos from Yahoo! Video, Google Video, ABC news, Broadband Sports, etc.
Another good feature is the option to save the video you're viewing to del.icio.us (the social bookmarking site) or e-mail it to a friend. These links are located at the top frame when you view the video.
The Bad
The search didn't yield nearly as many relevant results as Google. The first three videos were entitled "When Mascots Attack", "Like A Man", and "Awesome Magic Trick".
Yahoo! also provides related videos when you click on a video, but in this test run, many of the related videos didn't relate at all. For example, when I saw the video "Hail to the Victors", the related videos listed included: "New Orleans, An Outsiders View" and "American Pie".
The Good
The results were highly relevant. All of the thumbnails from the first page were videos that I could use. Some examples of the returned videos include:"Ohio State at Michigan", "Lloyd Carr Press Conference", and "Minnesota at Michigan".
The results page is organized very well. At the top of the page there are different tabs that allow you to view the results by "Best Match", "Most Viewed", "Highest Rated", or "Most Recent".
Below this box of results, there is a section where you can also view the video results according to the specific channel (websites) they come from such as YouTube, MySpace, and the BigTenNetwork. You can pick and choose which channels to look through. It's also nice that there are many channels to view from, meaning AOL searches for videos from a wide variety of websites.
The Bad
When you click on videos uploaded from the public, you can't watch the video from its original source. Rather, AOL shows the videos through their website.
In order to watch the videos that are streamed from websites such as espn.com or abcnews.com, you have to go to the site where the video's streamed from. Unfortunately, since some of these streaming sites do not keep these videos up permanently, some of AOL's results are therefore no longer available for viewing.
There are also no descriptions next to the thumbnails to give the searcher an idea of what content the video contains.
The Good
The site contains an extensive list of channels you can browse through from 5min - Life Videopedia to Your Daily Media. You can also look for Video Podcasts here.
The search results included lengthier descriptions than the previous three video search tools, so the searcher has a better idea of what video they're clicking on.
The Bad
Searchforweb.com definitely returned results with the least relevance. The front page of results included videos from Michigan State and Western Michigan.
Search results returned some videos that you have to pay for. The ESPN videos I clicked on required me to buy the video through iTunes. You also can't filter these out because there is no advanced search option.
I encountered many problems when trying to change the way results were shown. For example, when I tried to change the results from sorted by "Date" to sorted by "relevance" or when I tried to change from "detail" to "grid" view, I was directed to the site "Videos from around the web". This also happened when I tried clicking on the link: "Results for 'Michigan Football' in RSS". Needless to say this was very frustrating.
Searchforvideo has the most ads on its site (Google Video and AOL Video didn't have ads at all).
It's important to note (before making a conclusion) that every search tool had some type of filter for "naughty bits", making the searches very family safe.
In terms of the search tool that had the most to offer and returned great results, Google Video is a shoo-in. The RSS feature and extensive YouTube searching really stand out. However, if you're interested in looking for videos outside of YouTube, AOL Video would be a good alternative.
Posted by paulinad at 06:27 PM | Comments (0)
November 11, 2007
RSS Update
RSS Update!
Bloglines has been great for aggregating all of my RSS feeds onto a single website. It's my preferred RSS Feed Reader. It has definitely made it simpler and easier to read my favorite blogs or blogs that I'm keeping track of for my BIT projects.
As I have previously mentioned, I'm doing my personal project on Michigan Football. Therefore, I found as many quality blogs as I could on the subject and subscribed to their respective RSS feeds. To make all of my Michigan Football feeds more organized and easy to find, I created a Bloglines folder entitled (surprisingly) Michigan Football. That way, whenever I log on to Bloglines, I just have to click on the folder and all of my Michigan Football RSS feeds are right there at my disposal. No need to sift through lists and lists of feeds.
My favorite feed in this category has to be the feed I get from the blog, MGoBlog.com, which is run by a Michigan alum named Brian. He updates his blog very frequently (almost daily), with insightful information about the football team. He combines his own personal view with hard statistics and facts, either his own compilation or from other sources. For example, after every game (typically posted a few days later), he breaks down EVERY single offensive and defensive play. How he does this, I do not know, but it's very comprehensive. For a coach or someone interested in learning more about the types of plays Michigan calls or how players performed during each play, his breakdown is highly beneficial. However, it is important to note that this is strictly his own observation and is therefore subjective.
Additionally, from his feed, I've been able to find other sources of Michigan Football news because he always posts links and videos. I've watched numerous YouTube videos that I wouldn't have stumbled upon unless I thoroughly looked. He also belongs to a network of other dedicated college football/sports bloggers, so I learn about other NCAA teams as well. Every week they hold a Top 25 NCAA Football Poll, similar to the AP and Coaches Poll.
Initially, I just read his entries via Bloglines, but after about a week, the interface really bugged me. I didn't like how the information was formatted through Bloglines. It was like they changed the uniqueness of his blog into generic text with black fonts for regular text and blue fonts for headers/links. With all the statistics he puts on his blog, it's even harder on the eyes when reading it through Bloglines. As a result, I now only use Bloglines to see if he updated his blog or if I just want to briefly scan his entries for any pertinent information. If I see something that catches my eye, I go directly to the blog via the link on Bloglines.
For my business topic - Energy-Efficient Vehicles - I've also compiled a bunch of feeds into a folder entitled very creatively, "Energy Efficient Cars." The RSS feed from Hybridcars.com that focuses on Hybrid/Energy Efficient car news has been very helpful. Unlike my RSS feed from MGoBlog.com, this feed only provides news titles and the first line of the news feed on Bloglines. This makes it easy to scan for relevant news. However, this is also a downside because occasionally I find an article with a title that appears relevant, but turns out not to be after reading a few paragraphs. Thus, to make this feed better, I would make it so that there are summaries of the articles instead of just the first line.
Through this feed, I've learned about new hybrid cars emerging and how car manufacturers are approaching this relatively new market. Many manufacturers are realizing the great potential of hybrid cars and the importance of engineering cleaner and fuel efficient vehicles for our future (especially with oil prices at all time high levels). For example, Nissan just recently announced that they want to create hybrid cars for their Infiniti brand.
Overall, RSS feeds have definitely been one of the most helpful, if not the most helpful, tool for keeping me updated with changing news. I highly recommend getting a Bloglines account if you don't already have one.
Posted by paulinad at 10:59 AM | Comments (0)
November 04, 2007
E-mail Alerts Update
E-mail Alerts!
After the class session entitled "Web Search Principles 3", I set up an e-mail alert through Google Alerts (BETA). Since I had already decided on my the subject for my Personal Wiki Project, Michigan Football, I created a weekly alert that searched "Michigan Football" - central under the category of blogs. I just wanted updates on new blog entries that were related to Michigan Football that didn't contain references to Central Michigan University (during my initial test run for the alert, I noticed that I was getting a lot of results regarding Central Michigan so I made sure to take that out of my query by adding the "- central") that wouldn't crowd my G-mail inbox.
The alerts that I received from Google were very relevant to my topic. For example on October 8th, I received an alert that had links to a blog that talked about interesting subjects pertaining to Michigan Football that I wasn't necessarily tracking with my RSS Feeds: the great Michigan/Coach Bo Schembechler and former Michigan Quarterback/Current Stanford Head Coach Jim Harbaugh. However, I still found that some blogs that talked about Central weren't filtered through and that I was receiving some blogs about Western Michigan now, so I've changed the query to run as "Michigan Football" -central -western -eastern -northern to account for the other Michigan schools.
What I really like about Google Alerts is that when it's e-mailed to my g-mail account the "From" section of the e-mail, Google Alerts, and the "Subject" section of the e-mail, "Google Alert - 'Michigan Football -central'" really stand out. This helps me quickly scan my e-mail for the alerts, making the process pretty efficient. The content of the e-mail is also nicely organized with the title of the blog entries hyperlinked to the actual blog and a short synopsis of each of the entries. Also, since I have my alert set up to weekly updates specifically on blogs, this specific query hasn't flooded my inbox.
What I don't like about Google Alerts is that you have to really figure out a query that will get you the exact results that you want, and that can be very time consuming. My query "Michigan Football - central" hasn't really returned as many results as I would like. As a result, I set up more search queries including "Michigan Wolverines" + Football (a comprehensive search instead of blogs only), "Big Ten" (news instead of blogs). I've only been receiving a few more entries each week as a result of this.
In order to make the most of G-mail and my Google Alerts, I also created a Label in G-mail entitled BIT330 that would send all of these alerts to one place and make searching them through G-mail archives that much easier. I still get the e-mail in my regular Inbox, but because each e-mail from Google Alerts is now labeled BIT330, I can place a check mark next to each of these e-mails, archive them (removing them from my inbox), and look at all of them at once by clicking on the label at the left-hand side menu under "Contacts". It's made my e-mail alerts so much more organized. I highly recommend doing it.
On a side note, I also tried Yahoo! Alerts. You have to have a Yahoo account to get these. Yahoo Alerts is interesting in that you can do news e-mail alerts, which is similar to Google Alerts, or you can try getting e-mail alerts from a variety of different categories. I tried using a category for looking up hybrid cars and I don't recommend using it at all. For my car search, I tried using Yahoo's "Yahoo! Autos Custom" to get alerts notifying me about changes on that subject. I've only been receiving nonsensical alerts that send me names of cars with no active link or description. It didn't let me customize anything regarding the search query. Thus, the highly irrelevant e-mails that I received. To me, some of the Yahoo Alert categories are confusing to set up. For Yahoo Alerts, I recommend sticking to the news alert, which lets you choose between breaking news, keyword news, or daily news digest. Keyword news is a good way to specify the kind of news that you want.
Overall, for e-mail alerts, I would recommend using Google. However, I still haven't thoroughly comprehensively checked the private tool GoogleAlert, so you might want to check that out if you have some time or have money (you need to pay to get full service). I still like RSS feeds over e-mail alerts, however, when it comes to bringing me updates/news on topics because it's more comprehensive and easier to use.
Posted by paulinad at 10:31 PM | Comments (0)
October 06, 2007
RSS Class Blog Summary
To follow up on RSS feeds, I took a look at what other people from BIT 330 discovered, specifically the first half of the class (Beachnau to Kenzie). It's interesting how even though everyone essentially looked at the same sites (Blogdigger, Google Blog Search, Technorati, Blogpulse, Rojo, and IceRocket), there were many different findings and opinions. Here's a breakdown:
Not only was Google Blog Search the most blogged about RSS tool, it was also the site that a vast majority liked. It returned the best relevant results, and with it's similar interface to the normal Google search engine, it was the most simple/easy to use.
One thing that many liked in particular was how you can sort by relevance or date or both. That is, you can have the search results show up according to relevance, but then fine tune those results by clicking options on the left that will limit the results according to when it was last published (last hour, last 12 hours, last day, past week, past month, anytime, choose dates). One student also noticed that there are new search terms to help you trim down results: inblogtitle:, inposttitle:, inpostauthor:, blogurl:.
The downside: it isn't convenient for adding sites to a feed aggregator like bloglines since you have to go to the site actual blog.
Many people who went to Ice Rocket found that it returned very questionable results that weren't useful for their blog search. Not reliable, the site lacks in all areas compared to the others. Like Google Blog Search, you can't directly subscribe to a feed from the site. However, one interesting feature the site has is its option to search through MySpace.
Although BlogPulse didn't find blogs as effectively as Google Blog Search, it has a lot of cool features not found on other sites (e.g. Trend Search, Featured Trends, Conversation Tracker, and Blog Profiles, which I mention in my previous blog. Again, the site doesn't have a button to instantly subscribe to the feed using bloglines. If you have time and want to search for blogs in a unique way or want to learn more about what's happening in the blogosphere, Blogpulse is the way to go.
Some people like it, some people think it's OK, and then there's my friend Mike who wouldn't ever use it. Those who like it enjoy how it has lots of features ("What's Popping Now" section, Top 500) and how results include an authority number to give you an idea of how reputable the blog is. Those who think it's OK note that the results are not as reliable or relevant. Mike would never use this site for many reasons. The first time he went to the site it wouldn't load properly, but he still didn't like the site after giving it a second shot. In his experience, Technorati doesn't provide good results, it has way too many advertisements, you have to subscribe to use all of their features, and it's slow to load occasionally. You can read more of his thoughts in his blogs here and here.
Many ads, poorer feed quality and un-highlighted query words in the results deter many from giving Blogdigger glowing reviews. However, there are a few nice features that many people commented on. Blogdigger has a new search feature, "Local" search, that enables you to look for blogs that were written in a specific geographical location. You can search by city, state, or US zip code. Although cool in theory, this option still needs a lot of work considering it returned very few, if any, results. The "Links" search feature is also a good way to find blogs that reference a site you like. Lastly, another good thing about Blogdigger is that you can subscribe to the feed directly on the site.
Rojo had many favorable reviews overall. People found the keyboard shortcuts, located on the right side of every page, very useful in expediting the search process. Additionally, by clicking on "Options" right below the search bar after your search results pop up, you can customize how the results are presented. You can choose to view the results by relevance or date, read and unread stories, and whether to display only the whole story, a summary of it, or just the headline. You can also search by category, beneficial to those who don't know exactly what they're looking for. Results also come with pictures if applicable, which is really nice. However, the quality of the results were not as good as Google. Another con for the site is having to register (it's free) to get the full experience, but overall Rojo came in as a close second to Google Blog Search.
Posted by paulinad at 06:05 PM | Comments (0)
September 19, 2007
RSS and Searchable Feed Databases
RSS and Really Super Searchable blogs!
This week it's all about RSS (AKA Really Simple Syndication, Rich Site Summary, or RDF Site Summary), the XML application that enables you to keep tabs on all your favorite blogs all on one site when using an RSS feed aggregator (e.g. Bloglines). I learned a bit about RSS last year in BIT 200, but I didn't really have a strong understanding of how I could use RSS in my daily web-surfing life. Little did I know of the huge convenience using RSS brings to folks who enjoy perusing blogs (á la moi), but have little time to go to each site to check for updates each day. As a busy college student with way too many things going on, RSS appears to be a great solution to my blog-reading dilemma.
So in order to make the most of RSS, you have to find blogs that catch your interest and want to track. Searchable feed databases, which search through the blogosphere to find the blog feeds you want, are very useful tools for this type of activity, including Blogdigger, Google Blog Search, Technorati, Blogpulse, Rojo, and IceRocket. I decided that for this entry, I'd like to take a closer look at Google Blog Search, Technorati, and Blogpulse.
Since I rely heavily on Google's normal search engine, I decided to check out their take on blog searches. Right now Google Blog Search is in Beta mode, so there may be major changes underway. Besides that, you're hit with the very familiar Google layout when you get to the site. There's a single space for you to type in your query for a very straightforward search.
As a huge sports fan (and even though Michigan football hasn't been very praiseworthy as of late), I decided to find some blogs about Michigan football. By typing in "Michigan Football", Google Blog search returned a few related blogs in a little cluster at the top of the screen (e.g. The M Zone) and then lists individual blog entries related to football at Michigan. You can sort the blog entries by how long ago they were published.
Google Blog Search also allows you to subscribe to your "Michigan Football" blog search feed by clicking on "subscribe" in the bottom left corner. It gives you the option of using your preferred blog aggregator including Bloglines. This option is a very beneficial discovery tool. Although you can subscribe to your searches, one downside is that I couldn't find a way to directly subscribe to the blogs that Google found besides using the "Sub with Bloglines" button available for use if you're a Bloglines member.
Overall, Google Blog Search is good at quickly finding individual blog entries related to your search, but it doesn't provide as big of a feed list compared to Bloglines.
When going to the Technorati site, I was immediately struck by how much it reminded me of
This time when I searched for "Michigan Football", Technorati took me straight to blog posts related to the search query ordered from most recent to least recent. In order to see blogs about Michigan Football, you have to hit the "Blogs" tab at the top of the results. Technorati delivers a much more extensive list of blog feeds than Google Blog Search. However, I'm a little disappointed that the first two sites on the list, Football University and Straight Bangin', aren't 100% blogs on Michigan Football. The good thing about the results, though, is that there is an Authority number that goes with any blog post/blog feed listing on Technorati. This authority number tells you, the curious blog searcher, the number of blogs linking to that website in the last six months. The higher the number indicates that this blogger has been writing things that other bloggers find interesting.
My first thought when I went to the Blogpulse site was: "This site has a nice layout, but why are there a lot of nonsensical links with question marks and foreign letters?" Of course, I clicked on those links and discovered that they led to international sites, indicating that Blogpulse covers all blogs of all languages on one site, rather than sectioning it off.
Blogpulse offers its blog searchers a whole different set of options compared to Google Blog Search and Technorati. Serviced by Nielson BuzzMetrics', I found that this site is more useful for tracking blog trends. Nielson is most popular for its data on TV ratings and demographics. BlogPulse Live, one of the first thing that catches your eye on the home page, illustrates a graph of the blogosphere's current activity (i.e. it displays what's being posted about and how frequently according to subject). You can also use Blogpulse to do Trend Searches, check out featured trends, or track how bloggers link themselves through a certain topic with Conversation Tracker. If you want to find out more information about a certain blogger, use the BlogPulse Profiles. These are all interesting facets to BlogPulse that enable you to search through the Blogosphere in a whole new different way.
Continuing with my search for blogs about Michigan Football, I ran a search query on BlogPulse. The site found numerous blog posts with Michigan Football in it, but unfortunately no blog feeds related to the subject at hand. BlogPulse's Help Section explains that they only do searches on blog data, not on actual feeds. Bummer.
Overall, I would have to say that I liked how Technorati's site offered the most in terms of blog feed search and discovery. Google's great for a quick and simple search that will bring you dependable results, whereas BlogPulse is perfect for the curious interested in statistics and alternate searching methods.
Posted by paulinad at 10:47 PM | Comments (0)
September 16, 2007
The Deep Web
Searching the "Invisible" Web
The Deep Web...the part of the World Wide Web that can't be indexed. If this part of the web is so "invisible", then how are we supposed to search for the documents contained within? In an effort to gain a better understanding of the deep web and discover the best methods to search within it, I compared 7 different search engines:
Part of this experience required me to take on the role of a person interested in learning more about the Timber Industry in California. This led me to searching for "timber industry california" using all of the aforementioned sites on September 16th at 3:00 pm. The general search engines Google, Yahoo, and Yahoo directory yielded different results from the specialized databases Scirus, Google Scholar, UM Library's Search Tools, and CompletePlanet. Let's delve deeper into the Pros and Cons of using each search engine when interested in discovering the documents concealed in the deep web...
When I (and many other typical computer users) need to find information for a research project, I automatically go to Google and run a search query. It's a simple and quick solution, right? As expected, Google returned many results (1,870,000) in less than 1 second. Similarly, Yahoo Search returned over 2 million sites in less than a second. While both engines responded quickly to my query and returned many websites, the websites found were oftentimes not related to the search at hand. In fact, the top most results on Google's page were links and maps of timber businesses (one of them even being a link to Timberland the apparel company). Sponsored businesses advertising timber were also common to both Yahoo Search and Google, which were very extraneous to this searching process.
Sifting through all of the sites to determine which ones were relevant was very tedious, especially since the results from both sites are not clearly organized. Yahoo Directory, on the other hand, returned less than 200 sites. For me, these sites appeared more focused on company/organization profiles that were related to timber. Though there were a significantly less amount of sites to go through, the resulting links weren't any more beneficial to a person who wanted to specifically learn more about the timber industry in California. If you're interested in learning about companies/organizations somehow related to the timber industry, Yahoo Directory would be the place to go.
Scirus advertises itself as "the most comprehensive science-specific search engine on the Internet" and from the results of my search, I would have to agree. Although there were only just a little over 100,000 results, they all would be very beneficial to a timber industry researcher. Not only were the results well organized (you have the choice of looking at scientific journal results, preferred websites, general websites and you can sort the results by relevance or date), they also contained many more information-specific links. What I also like about Scirus is that they place their sponsored links in inconspicuous locations so that the research is the main focus.
Google Scholar also returned many information-based links related to the timber industry, particularly sites containing articles from scientific journals. This time there were no distracting Google sponsored sites. The big downside to Google Scholar is its lack of current news. Many of the returned sites had information from the 1990's and previous decades with a few in our current decade. It's also harder to navigate compared to Scirus because Google Scholar doesn't allow you to customize how the results are presented.
The other two deep web-oriented sites, the UM Library Database and CompletePlanet, were the most difficult and frustrating to use. The UM Library Database requires its users to fill in a lot more information before running the search query. You have to choose the databases that you want to look into, which can be a problem if you have no idea which databases would be pertinent to the search at hand. There's also an extensive sub-category list that can be a pain to look through. It takes a lot of energy and time for first-time users to find information using the UM Library Database. However difficult, it at least produced results unlike CompletePlanet. Although it claims that you can "discover over 70,000+ searchable databases and specialty search engines", during my search, CompletePlanet yielded zero sites. I tried running other queries with the same results.
Going through each of the search engines reinforced the fact that Google is not your one stop solution to finding information on the web. Google and Yahoo are great for quick general searches. Though similar in speed and number of sites found, I would still choose Google over Yahoo based on the layout of the website. I also discovered that deep web search engines, particularly Scirus, are perfect for finding information for a specific topic, especially when conducting research at hand because they are able to locate sites that cannot be indexed. The most important thing I learned for this exercise is that in order to be more efficient at web searching, using the right tools is a priority.
Posted by paulinad at 03:17 PM | Comments (0)
September 13, 2007
What I Hope To Learn In BIT 330
I just switched into BIT 330 yesterday (Sept. 12) and even though I haven't physically been to class yet, I already recognize the great learning opportunity that this class presents. Although I've been using search engines, especially Google, for quite some time now, I don't think I've even scratched the surface of what online search engines have to offer. I find myself continuously relying on Google (and occasionally Yahoo) for all my web-searching needs, but I know that there's a lot more out there. I can safely say that I don't even use Google to it's full potential, only making use of its general purpose and G-mail.
I hope that by taking this class, I will be much more efficient in searching for information throughout the internet by understanding the pros and cons of using other web tools and RSS feeds. I feel that this information will be extremely valuable to me in the workforce, regardless of career, because of the technology-driven nature of today's business world.
Posted by paulinad at 01:59 PM | Comments (0)