Archive for May, 2010

Windows Logo Testing – What is holding us back?

Wednesday, May 12th, 2010 at 6:08 pm   - Posted by Subodh Bhide

We all test our products and have high quality standards. But is the end user aware of our standards? Does he trust our claim of delivering flawless products? I am sure that in majority cases, the answer is going to be a NO. Unfortunately, end users just don’t trust the manufacturer’s word on quality. However, there is one way to increase users’ confidence.. and that is through compliance testing.  Windows Logo Test is one such compliance program that is designed for ensuring strict compliance with Windows OS design and standards. From my personal observations, the buyer’s anxiety level is drastically reduced  when they see a Microsoft Logo artwork on the product. It gives them the confidence that the product will be of great quality, as Microsoft themselves have certified it.

So, Windows Logo testing is a critical stage in the life cycle of product development. However, on many products, such testing is considered as “optional” or “good to have”. Based on Harbinger Systems’ experience working with product vendors, we see that the Windows Logo Testing thought process goes through the following stages. Some vendors drop out midway through the process, while others do walk the entire distance and come out on the other side.

  • Awareness - We start our list with the lack of awareness. Many product designers are unaware that such test programs exist. Even if they have heard about it, they are not sure what these tests do and what their results conclude.
  • Cost Benefit – The next question that crops up is around the cost of certification. Obviously the program is not going to be free. But on the other hand, certification is an investment which when used to its full potential can help attain higher sales and better user reviews. This is because, after certification the product is robust and can sustain all critical user scenarios. Also important is that users trust Microsoft certification.
  • Knowledge – After agreeing that there is significant ROI for the logo testing and willingness to give it a shot, after reaching this stage of the thought process, many still don’t know how to proceed. Microsoft provides elaborate information on Winqual site regarding the program, and it is a great resource to acquaint oneself with logo testing.
  • Time - One of last reasons for not proceeding with Logo testing is the lack of time. Product cycles are so stringent that there is little room left for any kind of conformance testing. However, what the product companies should understand is that logo testing programs can be executed independently of their existing test cycles. If expertise bandwidth is an issue, then Microsoft permits the use of 3rd party test agencies for running these tests on behalf of the company.
  • Infrastructure – So far we addressed some key hold-ups like awareness, cost and time. Another show stopper is the test infrastructure. The question we often get asked is, ‘Is it worth having an elaborate test setup costing thousands of dollars for a product portfolio that has very few products?’ Well, the answer is not straight forward, but an easy way out is to use a 3rd party test agency that has dedicated setups with the right mixture of architectures, hardware and OS versions to perform exhaustive testing. This can be done without spending huge amounts on infrastructure.
  • Expertise - The final step is the expertise needed for such a conformance testing program. I wouldn’t say that executing and triaging such tests is a child’s play, but surely external help is available if the expertise cannot be cultivated in-house.

I am quite sure that many of you can visualize yourself at one of these stages when it comes to Windows Logo Testing. Hopefully, some of the pointers shared above will help you proceed to the next level.

Are you a product developer thinking about Logo Testing? Which stage of the above thought process do you find yourself in? Do let us know or contact us here if you have questions for us.

 

Search results, relevance and the emerging web: What is Google doing about them?

Monday, May 3rd, 2010 at 4:33 am   - Posted by Prachi Kulkarni

This is a new world for searches. More and more techniques are evolving to index, manage and present the tremendous amount of data available on the net, and make it fit for users’ need. Both old and new search players are trying to shape searches and refine results. More than ever before, relevancy and context are key to “organizing the world’s information” today.

Google has introduced many new initiatives to advance relevancy in its search results. Let’s take a look at few of the most critical ones of these:

  1. Latent Semantic Indexing: Google’s effective use of Synonyms to determine if a page should qualify in the Search Engine Results Page (SERP) of a particular search is the basis of this. With the help of synonyms, a search about “song words” or “song writer” will also include hits with the word “lyrics” or “lyricist” in the content. Google is going one step ahead, where it will not only look at the key words for relevancy, but also scan the entire content and look for keywords as well as synonyms. If the entire content satisfies the requirement, it will be judged as relevant. This is called Latent Semantic Indexing.The purpose behind this is to index searches closer to the way a human reader would rank the page. In natural indexing, actual information rather than repetition of words would be a factor in how well the page ranks. The use of related words (synonyms) would signify that the subject is thoroughly covered.

    As shown in the above example from the Google blog, the search for “pictures developed with coffee” will have all the search results with synonyms for pictures, and all the synonyms along with the search terms will be displayed in bold.

  2. Semantic Searches based on profiles, reviews and events: For over a year the concept of Semantic Searches is doing the rounds in full force on the search scene. RDFa and microformat have been around for quite some time, but now they have caught prime attention spot as the new wave of search.
    Last year, Google announced its support for Semantic Searches with Rich Snippet. As an initial step, Google did this for two formats, people and reviews, with LinkedIn and Yelp as partners. They followed it up with semantic support for Video searches as well as events.

    The concept behind this is quite simple. There are people associated with documents, pictures, companies, addresses – almost everything. Similarly, there are events associated again with a place, an address, a company, a play etc.

    So now with the help of Rich Snippets, it is possible to tag profiles, events as well as reviews for a particular search string. An example outlined again in the Google blog shows three links to three events on Jan 23rd, Jan 25th and Jan 29th when a user searches for Irving Plaza on Google (see image below). Needless to say, this provides a fast and convenient way for a user to determine if Live Nation has any events lined up that he or she might be interested in.

  3. Google Insight for Searches and Google Trends: While Google Trends show us the current search trends, Google Insight takes it one step ahead. It shows us the applicable (frequently used) keywords as well as the rising searches within a topic, so that we can make out what is the related ‘hot’ search topic. For example, if we search for “Tigers + animal”, in addition to the search trend information of these and the synonyms, Google Insight also shows us the top and the rising searches related to these words.
  4. There are other new initiatives from Google in beta stage, like Real Time searches, Comparison Advertisements, as well as new domains in Semantic Searches, but we will leave them to another post.

Of course, relevancy in search results has been a determining factor of the success of a search engine for decades. Google is a key player in the search world, and will definitely be an important contributor in defining the semantic web. With all the above as well as other semantic analysis techniques like context servers and usage of explicit user actions when indexing content, the search scenario for 2010 and beyond looks very interesting… And Google seems to be taking small but sure steps towards the intelligent web of tomorrow.