My Blogs

This page contains all my blogs from the whole semester. It is sorted in reverse order of the date that the blog entry was last updated.


MetaSearch: Taking Search to a New Level

by roopakroopak (last edited on 13 Dec 2008 20:03)

Introduction

Metasearch is amazing! I had no idea such tools existed before. This was finally a tool that revived the excitement brought about by RSS and news searches. Metasearch is Google plus Yahoo plus Everything. As a result, it has capabilities that cannot exist with individual search engines.

Class Exercise

In class for BIT 330, we went through several metasearch tools and entered an identical query. We then looked at the top 10 results in each and determined how many were relevant. I used the query ["oil consumption" -intitle:prices] because this has been my most successful query with web searches for my project. Below are my results from the class exercise.

Search tools
Google Yahoo Search Info ixQuick Mamma Clusty DogPile
9 8 9 10 9 5 10 7

As can be clearly seen, most of the metasearch engines performed equally well or better than Google and Yahoo individually. If Mamma and Dogpile are factored out, it seems better to use metasearch tools than individual search engines.

Other advantages

In addition to providing more relevant results, metasearch engines allow you to see how many different search engines returned each result and allow you to use many different types of searches at the same place (news, video, image, academic, etc.). For instance, looking at the results of the query used for the class exercises in ixQuick below, you can tell how many times the results came up in the individual search engines by the number of stars next to the result. You can also see the ranking of each result in each individual search engine.

ixquick.results.bmp

This is very helpful in determining which results are likely the most relevant. Since results are sorted by number of stars, those sources deemed relevant by the greatest number of sources appear first. This can really allow you to take advantage of the benefits that come from using multiple search engines.

Dogpile provides a good example of a metasearch tool that allows you to use many different types of searches at the same place. It has separate tabs for Web Search, Image Search, Audio Search, Video Search, and News Search as shown below.

dogpile.bmp

This makes it very convenient for finding all different kinds of resources on one topic in one place. Yes, Google does have Web Search, Image Search, Video Search, etc. and you can access the different searches by clicking on different tabs or the dropdown arrow. The difference with Dogpile is that it incorporates different individual search engines in its search including Google, Yahoo, Live, and Ask.

Another type of metasearch tool

The tools above provide an integrated search for multiple search engines. Another type of metasearch tool provides a unified interface for separate search engines. Though I did not explore these as much, they are pretty interesting and I found these more useful for searches on topics that I just wanted some quick information about. For instance, Soovle allows you to see results from the one or two most popular search sites of each type at once. It gives you results from web search sites (Google and Yahoo), a video site (Youtube), and an encyclopedia site (Wikipedia), among others. Another site of this kind that I found particularly interesting is Joongel. This is basically a directory of search sites. It has the cool feature of allowing you to choose a category of sites (Tech, Finance, Movies, Shopping, etc.) for your query to search in.

Conclusion

Metasearch tools have many capabilities and allow you to perform a search in many different resources at once. They also sometimes point out the overlap between these resources allowing you to really gain from this collaborative approach. One of the best aspects of metasearch tools is that they introduce you to many new search engines that you probably didn't know about before. After using a metasearch engine several times, results from certain search engines may clearly appear more relevant. This would provide new information as to which search engines are best to use for which topics. Many of these search engines are not widely known simply because their functions are so specialized, but it is precisely for that reason that they are so valuable.

I will be using metasearch tools a lot in the future. Currently, I don't know about a wide variety of different search engines, so I am hoping that by using metasearch engines I will be introduced to new search engines that better cater to my needs. This way, I will be able to go to a specialized search engine for my finance queries, for instance, instead of just typing them into Google.


Ben Lewis: The (Google) Man

by roopakroopak (last edited on 27 Nov 2008 06:17)

Introduction

google.jpg

I am a big fan of Google just like most college students. You can type in almost any subject in Google Search and get all the sources you need for a solid paper in less than 1 second. Also, with Image Search, Youtube, and Google Calendar, Google pretty much serves all of my time-wasting and organizational needs. Because of this, when I heard that a speaker from Google was coming to speak to the class, I was naturally excited. Ben Lewis was that speaker.

What Ben Talked About

Ben was an interesting speaker who emphasized how Google was actually making its money from its products. He touched on the two most obvious ones (search and ads), but he also spoke to us about how Google Toolbar and Google Earth, as well as several other products, were also bringing in revenue. Perhaps more importantly, he specified the costs that are really associated with the products.

I was interested in the auctioning system Google uses for ad space. Ben went through a search for a video game and showed us the four ads that were displayed on the right side of the screen. He then explained the how the many advertisers submit bids for certain keywords and the few highest bidders get to have their ads displayed. While doing this, he talked about how Google prevents ads with excessive exclamation points and other nonsense from showing up and allows only relevant ads in its ad space.

google-pagerank.jpg

I have never used the Google Toolbar and never thought much about how toolbars could help make money. Ben explained that the toolbar is beneficial for two major reasons. First, it keeps the brand at the top of mind. This made clear sense to me. Second, it has a PageRank feature that indicates how important a web page is by using clicks as votes. This is a useful service and it allows Google to track user data. Tracking user data leads to better information. Better information leads to better services and greater revenue. Google Earth offers Google Toolbar when it is installed, so it also provides the same benefits indirectly.

The order of costs Ben mentioned were about what I had expected. For every product, servers and bandwidth were the highest cost. The costs of creation and maintenance varied with each product, but the variable costs per use by a consumer were very low. The biggest thing that I had not previously considered at all was the fact that Google is losing money with each search that does not involve clicking on an ad.

What I Learned

I had not previously considered the level of strategic thinking that goes into Google products. It had crossed my mind several times that it makes no sense for Google spend its money offering free products. Google Earth in particular had always seemed like a really cool program that was useless for Google's bottom line. The idea of adding Google Toolbar as an option when installing Google Earth is an incredibly smart idea. Google creates a lot of goodwill because of these products and gets a lot out of them at the same time!

323106237_8b129e82d0_o.jpg

The support for creativity was something I had heard about on the news, but hearing it from a Google employee made it real in my mind. Ben said that Google does not put a cap on the amount of money that can be spent on a project. Ideas are funded as far as they can go! The fact that they allow creativity to work freely like this and that employees actually feel this freedom explained to me how a company can create such amazing things. Unfortunately, this freedom has become slightly limited recently due to the economic decline, but I feel that the culture that has been established at Google will continue to foster amazing new products.

On the flip side of this, I learned about the unsuccessful products Google has and the money it has lost in these products. Ben talked about Google Checkout and broke down how the fee it charges fails to cover the fees it must pay to credit cards and other related costs. He mentioned that there are actually a slew of products like this that turn out unsuccessful. Being the typical Google user, I had no idea about most of Google's products and only knew about the most popular, wildly successful ones.

The Point

Hearing from Ben was great because it gave me insight into the thinking that is behind the products I am familiar with. As an aspiring business professional, it was a good lesson in strategy and considering costs. The talk overall showed how Google uses its products to promote each other, and how it uses categorization to maximize its revenue, and how it grows through innovation.

logo-google.gif

Oil Consumption Search Engine

by roopakroopak (last edited on 09 Nov 2008 06:09)

Introduction

custom_search_sm.gif

A Google Custom Search Engine would be very useful for my project on oil consumption. New data comes out on a monthly basis or even more frequently on many sites. Also, trends in oil consumption tend to vary frequently and are hard to predict because they are so dependent on current events. Because of this, it would be important to have a means to search for the most up-to-date information. Below is my initial Oil Consumption Search Engine, which is linked to a few fact-based sites that are essentially databases for all statistics currently available about oil consumption.

The Sites

Energy Information Administration
I chose the Energy Information Administration website because it has been my greatest source of facts and figures regarding oil consumption so far. It provides weekly oil consumption data, U.S. Imports by Country of Origin, and many other very useful types of data. I used the entire site in my custom search engine because some of the oil data is scattered outside of the Petroleum page on the site. It will be very useful to any user looking for the most up-to-date information because it will have weekly updates while the my wiki site and related blogs will tend to be based on some past data.

Nation Master
I chose Nation Master because it has visuals of all different types of oil consumption data. Though it is not as up-to-date as the EIA website, it provides a bar graph, pie chart, and map for each type of oil consumption statistic it features. As I noted in my blog about Image Search, such graphs and maps are necessary to digest all the numerical information about oil consumption.

BP Statistical Review 2008
I chose the BP website because its information is from a player in the industry. Since BP is an oil company, different types of statistics are presented than would be given by a disinterested source. Therefore, there is more information on methodologies, trade movements, and refineries than on the EIA website and Nation Master. Trade movements are very important in the discussion of how U.S. oil consumption relates the rest of the world. Methodologies and refinery information provide some insight into how oil consumption can be more efficient and the feasibility of alternative methods. The site also contains historical data about oil consumption for each country since 1965.

Though there is overlap between the three sites, each contributes a significant amount of unique information. Together, they provide a large pool of information that contains a large percentage of the useful statistics regarding oil consumption in the U.S. and the world.

Additional sites to be added in the future should be focused on experts' views on the current state of oil consumption. This could primarily come from blogs and news sources. Sites containing research papers on the topic would also be useful, but unfortunately many of these papers would be part of the Deep Web and therefore inaccessible through the custom search engine.

Conclusion

A custom search engine can help cut out many of the irrelevant sources that are found through a normal Google Search. For my topic, using the Oil Consumption Search Engine would help a searcher find relevant information much faster because it emphasizes results from the three sites above. It would be especially useful when used to find the up-to-date numbers for the statistics on my wiki, because these were the sources those statistics initially came from. Therefore, a custom search engine would be a great addition to my project and would make it a much more complete source of oil consumption information.


Image Search: Google vs. Flickr

by roopakroopak (last edited on 01 Dec 2008 00:32)

Introduction

Since I am attempting to show the problem with oil consumption, pictures are necessary for my project. There are many numbers out there, but there are usually hard to digest. In my own research, the only way I have been able to really get a grasp on the problem is through numerous graphs and charts. I decided to take a close look at Google Image Search and Flickr to find out the strengths of each and how they can help me with my project. I used the query ["oil consumption"] for both.

Google Image Search

images_hp.gif

Just like with all types of Google searches, I was happy with the way the results were displayed. Eighteen thumbnails were displayed in rows of 6 and I could see them all on my screen with no scrolling. A few words of description were under each thumbnail. Perfect!

The results showed graphs of many different varieties. There were even a few that I had never seen before. Below is a sampling of some of the different types of graphs I came across.

oil_consumption_by_country_1.jpg world_oil_consumption.jpe
Blog_Peak_Oil_2004_Scenario.gif

This variety of types allows for a very clear illustration of how oil consumption in various countries has grown (line graph) and how much of the world oil consumption is accounted for by each major oil-consuming country (pie graph). The stacked graph also provides an illustration of the overall oil consumption broken down by region. Through the use of these graphs and others of this sort, almost all figures relating to oil consumption could be converted into an easily understandable form. Google Image Search proved very helpful in this sense.

The results also illustrated oil consumption in map form. A couple of examples are below.

world-oil-consumption-map.jpg googearth_oil3.jpg

The first is a typical oil consumption map that simply uses a legend and different colors to illustrate where oil consumption is heavy and where it is light. This is perhaps the most clear way to break down oil consumption geographically. The second is a snapshot of a Google Earth visualization for oil consumption. This is amazing! You can get these 3D charts at the Google Earth Blog. These are easily the most unique depictions of oil consumption I have seen.

Overall, Google Image Search proved to be a great tool for images relating to oil consumption, and the results contained every type of visual that I had found so far in all of my research.

Flickr

flickr-logo.jpg

The display of the results in Flickr wasn't nearly as appealing as in Google Image Search. The images were in list form, requiring me to scroll down the page even to browse the first few. There was a large space next to each image for description, but there was only the date uploaded and some tags next to each image. There was no description!

The images were not nearly as relevant to my search as in Google. A couple of the results were wine classes and pictures of cars. Though the cars were consuming oil, this provided me no useful information. There was also some overlap with the images in Google Image Search, including the world map and the snapshot of the Google Earth visualization.

The relevant and non-overlapping images mostly consisted of line graphs. The only new type of information that was not available in Google Image Search was in the images shown below.

779992458_5f1cfe37a0.jpg?v=0 2898151295_dc13ff853b.jpg?v=0

The unique thing about these images is the focus on oil consumption in the United States. The first image provides an incredible depiction of how high U.S. oil consumption is compared to the rest of the world. The second provides a nice breakdown of consumption by state in map form. Though Flickr may lack in variety of graphs, this focus on the United States is very helpful in demonstrating the oil consumption problem.

Conclusion

After taking a close look at both Google Image Search and Flickr, I was able to conclude that there was something to be gained from using both. Though I would say Google Image Search is more useful and easier to use, Flickr does provide some unique images that are very useful for my oil consumption project. An important thing to note is that I learned more about the current state of oil consumption from looking at the images I found than by looking at numbers from sources like the Energy Information Administration website. The images reminded me of the necessity of providing visualizations for something as difficult to digest as the state of U.S. and world oil consumption and the problems which will arise in the future. Image search is certainly a very useful tool that should be utilized in any research project.


Interview Prep

by roopakroopak (last edited on 17 Oct 2008 23:13)

Introduction

470557013_0823e5f44f.jpg

Interviews are right around the corner and there are fewer and fewer jobs to be had! For this reason, I have decided to stray from my usual topic of oil consumption and focus on how to deal with the coming battles that face us all. Interview preparation is difficult because it seems like companies can ask you anything and everything and in whatever way they want. In my recent discussions with seniors, I have heard stories ranging from an interviewer playing around with his cell phone to four interviewers walking into the room and turning down the lights. Despite how crazy an interviewer may act during an interview, all you can really do is know your stuff and hope for the best. Since the markets are going nuts and students are going crazy, I decided to see how the tools we have used in BIT 330 can help prepare for the madness to come.

Company

jp_morgan_logo_02.jpg

I started by using the tools to find information about J.P. Morgan interviews. I specifically picked J.P. Morgan because it hasn't had much bad publicity and I feel like it will still exist in the coming months. I used Google Search, Bloglines, and BNET to see what I could find. I figured that I would get a good variety by using a web search, a blog search, and a deep web search. For all three, I used ["J.P. Morgan" interview blog] as my query. I thought the best resources would be blogs specifically tailored for interview prep.

In my first few Google results, I found a nice page about J.P. Morgan on Indeed.com that had a basic description of the company and some basic information (revenue, number of employees, stock symbol, headquarters location) that was good for getting started. There were also links embedded within this information for further research. The tenth results was a forum in The Student Room where there was some specific information about the interview process given by various people. Details included the number of interviews per round, the numerical test, and how technical the questions are. Just below this result was the Vault.com excerpts of surveys of current and former employees in various positions. Overall, Google seemed like a great place to get important numbers about the company and some detailed information about the application process.

Unfortunately, Bloglines and BNET didn't yield nearly as helpful results. The first result in Bloglines did state that J.P. Morgan sets aside money for temp or temp-to-permanent jobs. Other results had no information specifically about interviewing/working at J.P. Morgan, but had news stories and interviews with people who did not belong to the company. BNET only returned results from Zacks Analyst Blog Highlights and Schaeffer's Daily Market Blog which contained articles about the stock. Of the three tools, Google Search seemed the best and Web Search in general seemed like the best method for this research.

Job Field

interview_clipart.jpg

I decided to do a more general search next on investment banking interview preparation. I thought this would be more applicable to the average student because this is where the most prep time would be devoted. Seniors have told me that knowing specific information about a company is not always necessary, but you are always quizzed on the basics. I entered ["Investment banking interview"] into each of the three tools and hoped for the best.

The first result from Google was an excellent resource! It was an investment banking interview prep blog with various links to IB flash cards, professional networking sites, and company profile pages along with many other resources. On the home page of the blog, there was also a recent entry by the host about his internship training week. He included a brief overview of each day of training and his thoughts on the process. This site seemed extremely helpful! Other results within the top 5 included Investment Banking Interview FAQ from Cornell University's Management School and two pages from Mergers & Inquisitions (an investment banking site) with a Superday Guide and information about the selection process. This simple search in Google seemed to give me all the resources I needed to get a good start on interview preparation.

The first result in Bloglines contains a great blog entry within the Mergers & Inquisitions site about how to answer the most important interview question: "Walk me through your resume." Other good results included "What Are The Big 5s in Investment Banking Interview?" and a Brainteaser Q&A. Bloglines seemed to be a better place to find out the best way to answer specific questions that come up in an interview, while Google appeared to be a better resource to find out about the interview process.

Unfortunately, BNET found no results and simply returned a list of companies which were its top tags.

Conclusion

Though interview preparation is grueling, there are a lot of resources on the internet that make it a lot easier. People have compiled all of the information they used and blogged about their experiences so that we can learn without as much pain. From my two searches, I uncovered a lot of resources that I was previously unaware of and will definitely benefit me in the coming months. My recommendation, as I stated above, would be to use Google Search for factual information on a company and information about the interview process. I would recommend using Bloglines to find out the best way to answer specific questions. Looking at other web searches and blog searches could provide greater insight into the best way to use the internet to prepare for interviews. The deep web was once again disappointing to me, but I still believe that there is a chance it could be useful if the right deep web search engine is used for the right purpose. I plan to subscribe to RSS feeds of Google Search results and Bloglines results from the queries I used. I hope my searches will be a good launchpad for your own interview preparation searches.


The Deep Web?

by roopakroopak (last edited on 13 Oct 2008 12:51)

Introduction

I naturally have a tendency to be skeptical of web resources that I have never heard of. Before our class on The Deep Web, I had no idea that the Deep Web existed. The idea that these were resources that a normal Google search could not access interested me slightly, but I assumed the information couldn't really be that useful. I decided to go ahead and give the "Deep Web" a try and explore what it had to offer. I thought it might change my perceptions like RSS feeds and news searches had.

Unfortunately, I was not really wowed by this new resource and it didn't alter my perceptions. I went through each of the Deep Web resources given in class to conduct exercises and submitted the query 'oil consumption' (with quotes). This was a relatively simply query that I assumed would produce similar results across the board. My assumption turned out to be horribly wrong. I have included the top 5 results returned by each resource below and evaluated each resource based on my results.

Deep Web resources

Scirus

Environmental Defense - Putting the Brakes on Oil Consumption
Plasmatron could cut oil consumption, emissions - MIT News Office
03-Engine Oil Consumption Measurement Brochure
The Determinants of Sulfur Emissions from Oil Consumption in Swedish Manufacturing Industry
SwRI measures real-time wear and engine oil consumption in motorcycle engines

My first observation with Scirus was that the format of the results was similar to that from a normal web search in Google. Each result had a title and a short description below it. This was very simple and straight-forward.

Taking a look at the results, I noticed that each result was relevant in some way to the search. They talked about either measuring oil consumption or reducing oil consumption. They were large websites with many resources and did not talk about current news. Instead, each result was one webpage that discussed something related to oil consumption. Also notable was the fact that of the top five, one result was from 2006, three were from 2007, and one was from 2008. I would not consider this a particularly useful source for my project. Though it did not unlock any wealth of information about oil consumption, it did at least provide a few interesting pages.

Google Scholar

Consumption of Olive Oil and Specific Food Groups in Relation to Breast Cancer Risk in Greece
Fish oil consumption and decreased risk of cardiovascular disease: a comparison of findings from …
Oil and the Macroeconomy since World War II
Olive-oil consumption and health: the possible role of antioxidants.
The Effect of Olive Oil and Fish Consumption on Rheumatoid Arthritis-A Case Control Study

Once again, the format of the results was similar to that from a normal web search in Google. This was simple to use just like with Scirus above.

Unfortunately, the results returned were terrible! I think that might actually be an understatement. This was easily the worst resource I had found so far in the class! Just looking at the results made me laugh and think that this might be a joke. Olive oil? Fish oil? How could these possibly be the main topics of 4 of the top 5 results? The only result that actually spoke about the type of oil I was looking for spoke about World War II.

Could this possibly get worse? Somehow it actually did. The publication dates for the results in the order they are listed above were: 1995, 1986, 1983, 2000, and 1991. Professor Moore mentioned during our class on the Deep Web that sometimes the publication dates are wrong in Google Scholar. I'm not sure if this is the case with these results, but I am assuming that the majority of them have the correct publication dates. They just happened to be not at all relevant to what I was looking for and from at least 8 years ago. All I have to say about Google Scholar is that it is horrible and I probably will not be relying on it for any help.

BNET

Cyprus Solar Thermal Market and Technology Assessment
Palm oil consumption increasing, says council, BUSINESS TIMES
BP SHOWS AREAS OF MOGAS GROWTH; FLAT GLOBAL OIL CONSUMPTION
BP: WORLD OIL DEMAND DIPS IN 2001; U.S. REFINING CAPACITY CREEPS HIGHER
Too Soon to Re-enter Oil Stocks?

The format of the results from BNET was slightly different from the two resources above and a normal Google search, but it had the same degree of simplicity and straight-forwardness. It also contained tags below the results, which I found rather convenient.

Besides the article on palm oil, these results were relevant. Some of the results linked to excerpts from the papers or articles on the BNET site and some linked to other websites. It was a bit annoying that most of them required registration to view the actual paper or article. However, these articles were very informative and BNET seemed to be a good resource for extensive research on a topic. Just like with the previous two resources, BNET's publication dates for the top 5 results were not very recent. The fifth result was from the day I conducted the search (08/13/2008), but the other results were all from 2002 and 2003.

Turbo10

Energy Information Administration - International …
NationMaster - Oil > Consumption (most recent) by …
Dodge, Chrysler DaimlerChrysler Vehicle Defects
Oil Industry Statistics from Gibson Consulting
Petroleum - Wikipedia, the free encyclopedia

Now this was a good example of bad formatting of results. The default was 5 results and there was no description below the titles. There were also no publication dates listed. The only information that was given besides the title and the domain name was the source that provided to result to Turbo 10 (for 4 of the results this was alltheweb.com and for 1 of the results this was dmoz.org). This was completely ineffective.

The results were not bad in terms of relevance. The first result was an excellent resource with all kinds of statistics about oil. This was the same resource I had gotten most of my data for my project from at this point. The second and fourth results were both very informative resources as well. As far as quality of results goes, this was the best of the Deep Web resources so far. I must note that I found it amusing that the fifth result was simply a link to the Wikipedia entry for 'Petroleum.'

Librarians' Internet Index

AAA Fuel Cost Calculator
Corporate Average Fuel Economy (CAFE)
How Gas Prices Work
Oil Spiel
Fueleconomy.gov

This was the best of all of the resources in formatting of results. Below each result title was a description, the url, the topics (i.e. Automobiles, Oil & Gas), and the date the site was last updated. I could have no complaints about this. It was even better than Google.

The results themselves did not make me quite as happy as the formatting, but they weren't bad either. The first and fifth results were completely irrelevant. The middle three contained interesting information about fuel economy, how consumption is taken into account to determine gas prices, and peak oil. All three of these were potentially useful sources. It is important to note that the first three results were last updated in 2005, the fourth was was updated in 2006, and the fifth was last updated in 2008. Though this was a good resource overall, like with most of the others, the results were not very current.

Conclusion

I did not find any of the Deep Web search engines to be especially useful for me. I can say, however, that I can see each of them being useful except for Google Scholar. I did only submit one query to each of the Deep Web search engines, so naturally my results cannot be taken as conclusive evidence of how useful they will be in general. However, I believe that some of the observations I had can provide good clues as to how useful they will be for research on other topics. If you would like to take a look at the full lists of results returned by the Deep Web search engines for my query ('oil consumption'), here are four of them: Scirus, Google Scholar, BNET, and Librarians' Internet Index. For Turbo 10 you will actually have to type in the query ('oil consumption'), because the url does not change in the address bar when you conduct a search.

050822b_DeepTrawl.gif

Yahoo! News v.s Google News

by roopakroopak (last edited on 07 Oct 2008 01:07)
YahooNewsLogo-large.jpg
google_news_logo.gif

In my last post, I talked about how amazing I think RSS is. This post is about how to apply the usefulness of RSS to news search engines. I decided to compare Yahoo! News and Google News because these are the two news search engines I plan to use the most in my research about oil consumption.

The competition

I looked at the results from both Yahoo! News and Google News for the query "oil consumption" (including the quotes) for the last day and for the last week. I felt that looking at the last week would give me a good overall view of how relevant recent results from each search engine would be to my project. I looked at the last day to get an idea of how many relevant results I would receive daily in my RSS feed using this query.

For the last week, I looked at the first 20 results of each sorted by relevance. Here are my results expressed in percentages:

Last week
Yahoo! News Google News
Yahoo! News 35 10
Google News 30

As you can see above, the results from Yahoo! News were slightly more relevant than the results from Google News. The 10% shown at the intersection of Yahoo! News and Google News is the percentage of results that were BOTH relevant and overlapping. This was to give me a gauge of how many of the results that I would consider useful were actually appearing on both. It is clear that a lot can be gained from using both news searches as opposed to one.

For the last day, I looked at all of the results since there were fewer than 20. I recorded my results below using the actual number of results:

Total results from Yahoo! News: 11
Total results from Google News: 15

Last day
Yahoo! News Google News
Yahoo! News 4 3
Google News 4

It appears from these results that on a daily basis, Yahoo! News and Google News will give me about the same number of relevant sources in my RSS feed. A few more days would have to be looked at to draw a strong conclusion about this, but looking at the last day should give us a decent idea. The 3 shown at the intersection of Yahoo! News and Google News is the percentage of results that were BOTH relevant and overlapping. It can be seen easily that on a daily basis most of the relevant results from Yahoo! News and Google News are the same. If I had an RSS feed of these results, I would only get one additional relevant source from looking at both as opposed to one. Obviously, once again a few more days would have to be looked at before drawing a strong conclusion on this.

The conclusion

The important thing I have taken away from this for my project is that by having RSS feeds from both Yahoo! News and Google News using the query "oil consumption," I will get at least a handful of results that will keep me updated with relevant news about oil consumption.

As for my personal use, the combination of these tools has given me an even better way to prepare for interviews in the coming months than RSS alone. In my previous post, I mentioned that RSS gave me what I took BIT 330 for. The use of news searches with it will help enormously in preparing for interviews, because it will deliver the daily news about the companies I am interested in right to my RSS reader. Here's a list of some of the daily searches I plan to subscribe to that could help get you started:

Company Yahoo! News Google News
Goldman Sachs Y G
Morgan Stanley Y G
Deutsche Bank Y G
Citi Y G
BOA Y G

Subscribe to RSS feeds for each company you're interested in to keep you up-to-date and prepared!


RSS strikes oil!

by roopakroopak (last edited on 02 Oct 2008 11:03)

This post was truly inspired by RSS. It has shown me that everything might be ok. I know this sounds odd, but read on.

Introduction

google-reader-logo.JPG

To be honest, after BIT 200, I was not very impressed with RSS. We were required to use Google Reader for a period of time as an assignment, so I just clicked on popular feeds like The Wall Street Journal, New York Times, and whatever else seemed news-related. I thought of it as a way for me to get the news items for newspapers that I SHOULD read but never actually do. What happened? Well, I never read these feeds on Google Reader either. There were thousands of news items piling up that were about hundreds of topics. I wasn't really interested in any of it but felt obligated to get a feed for something that sounded intellectual for the class.

The main reason I initially decided to take BIT330 was so that I can easily get large amounts of specific information about companies when interviewing time comes around. I found the classes on Search Techniques and Web Directories somewhat interesting and saw myself using the information in the future, but they didn't give me exactly what I took the class for. I thought the classes on RSS would potentially give me this, but was apprehensive after considering the thousands of unfiltered news items that were in my Google Reader from BIT 200.

The search for oil

Two things separated this experience with RSS from my last one:
* Search techniques
* Subscribing to search results

For my term project, I am giving detailed background and following the news on oil consumption. This does not mean watching oil prices! This means monitoring how consumption is affected and keeping up-to-date with plans to remedy the oil shortage.

I started out as indicated in the RSS Lab Exercises by searching in Bloglines. I naively submitted the query: oil consumption. Go figure, 83,300 posts! Even of the first five, not too many looked very useful. This was more than mildly discouraging. I thought about it for a second and then went with: intitle:'oil consumption'. This came out a bit better, but I felt like the results were still not relevant enough. Finally, my perfect query hit me: intitle:'oil consumption' -prices. There it was!

bloglines_home_1.gif

Within the descriptions of my first six results, there was:

  • "World Crude Oil Consumption by Geographic Region"
  • "Oil Consumption by state"
  • "reduction in oil consumption will increase the inelasticity of remaining demand…"

My search had given me all kinds of background information on my topic and it seemed as if the results were tailored for me! After looking at the next step in the exercises, I subscribed to this search.

The spoils

I am not a tech-savvy person and BIT 330 has been a bit outside of my comfort level so far. It took me a LONG time to figure out how to navigate all around the Wikidot course website and I am very slowly becoming comfortable with formatting within Wikidot. This experience showed me that the term project, which had previously seemed quite overwhelming, might not be quite so overwhelming. If I can have multiple feeds from various sources with search results tailored to my project, I should easily have enough information to develop a good term project (in theory).

man-sitting-clip-art-silhouette.jpg

Beyond the project, this experience gave me what I came to this class for. It gave me an easy way to get large amounts of specific information about companies. This is something I know I will use well after I leave this class and may very well help me land a job!

Unless otherwise stated, the content of this page is licensed under Creative Commons Attribution-ShareAlike 3.0 License