Search has evolved into a mission-critical application for some. As our information storage grows, whether it’s pictures of the kids at home or Internet research content for school, our ability to identify and retrieve data that we need becomes more important and a growing challenge. Net Librarian tries to use specific searches across several search engines to return results stored in a local database.
Net Librarian is a web application interface for keyword input to internet searches saved to an Access 97 backend. In addition, searches are sent to the vendor to establish a cache or store of searches made. Searches with multiple engines and URL verification takes a little longer than a basic search. These searches are stored for reuse at the vendor’s site: websearchstore.com.
Installation & Setup (3 out of 5)
Installshield makes this standard installation simple.
The full version requires a simple code to enter. Though it is called an activation code, it is a simple serial key. There is no secondary activation. [See image 1]
The installation provides a summary of installation parameters chosen prior to installing the application. [See image 2]
The End-User License Agreement, or EULA, references vbaccelerator.com as the source of some of the code for this application dating back to the late 1990s. vbaccelerator.com itself hasn’t been updated since 2004.
There is no upgrade path from the trial version to the full version. Installing the full version of Net Librarian 8.0 returns an error stating that the installed version must be removed before the full version can be installed. [See image 3]
Installation was not possible on Windows Vista because of missing components, such as VB5 Runtime files. [See image 12] The product could benefit from an update.
User Interface (2 out of 5)
One of Google’s best decisions was using an extremely simple search interface. Net Librarian 8.0 utilizes a similar approach. [See image 4]
Net Librarian search results are returned to the screen in table format including the object types. It lists extensions and text type such as png, gif, jpg, phone, fax, keyw, http, and swf.[See image 5]
The interface for the menu option Application Preferences is terrible in its use of interface real estate. One of the options is to move the storage location for search databases. When selected, the interface to make that change goes off the screen to the side unless window is full screen, even as lots of whitespace remains. [See image 7]
Some aspects of the web interface are hard coded to an external IP address serving the application. This means that you need to be online to open the product. It will use cached files to render the main screen, but without that, it seems one can not use Net Librarian to even query the saved searches unless online. Using View Source in the context menu (as this is a web application) the HTML code shows several http calls to the IP address 188.8.131.52.[See image 15] The ARIN database indicates the IP address belongs to Scott Baxter of Huntington Beach, CA. This makes sense as he is the author of Net Librarian (originally from baxtsoft.com) and the owner of WebSearchstore.com. A FQDN should be used and not an IP address to allow portability of the http calls should that be required someday.
Product Features (3 out of 5)
Net Librarian saves searches to an Access database in the local file system.[See image 8] While this is not the best backend choice, it does allow some flexibility with what the user can do with the content. The saved search is both accessible and editable using a capable database application, especially Microsoft Access.[See image 14]
A search summary or index can be exported to Excel, or more accurately, to an .xls spreadsheet.[See image 10] The .xls file consists of three columns listing the returned searches - web site URL, title of the site, and some sample text.
If you use a browser other than Internet Explorer, Net Librarian will tell you that it works best with IE. It seems no effort has been made to ensure compatibility with Firefox, Opera or any other competing browser.
Performance (1 out of 5)
The complementary website that stores searches made with Net Librarian responds promptly to the same search made earlier with Net Librarian. The search site, www.websearchstore.com, is similar to the Net Librarian main user interface. [See image 11]
This application has been around for years. For example, Net Librarian 4.09 came out in late 2002, Net Librarian 6.1.2 was released in August 2004, and Net Librarian 7.0 in July 2005. It would seem to me then that by version 8.0 the product would be polished, effective, and efficient. Really, this is not the case. Though it is designed to work behind the scenes, it is still slow in its initial mining.
Searches can generate large .mdb files with accompanying .txt files. The track.txt files are sites matching the search that Net Librarian has not visited yet. The .mdb files represent the database of objects with URLs for the completed search. Searches can be resumed, but only if no other searches are implemented. A new search will clear existing .txt files and create new ones specific to the new search. The .mdb files are saved for future viewing.
Search results are not just websites. The websites that are returned from at least 20 search engines are parsed for various resources within them, essentially visiting each site and saving URLs for images, sounds, and documents. Net Librarian also identifies phone and fax numbers from the text of websites. This in-depth search functionality is the heart and soul of the product; however, overall the performance just isn’t good enough to meet the expectations of many potential customers.
Net Librarian is not an application to satisfy immediate searches. If instant gratification is important, then this is not your research tool. It is designed to work in the background supposedly assembling a comprehensive resource based on keywords. Searches are slow during the initial web search. They are also slow on subsequent searches even though they are stored at websearchstore.com for faster access by all Net Librarian users.
Price to Value (3 out of 5)
There may be value for some, but overall, the features are available in better applications elsewhere, such as Copernic Agent. That Net Librarian is very inexpensive may endear it to some, but really it is a reflection of its limitations more than its value.
Security & Privacy (3 out of 5)
Even though searches are not stored in user profiles, Net Librarian does recognize different users and does not expose other users searches to the interface. The files are available by default in %program files%\Web Search Store\Net Librarian 8.0\websearch\.[See image 8]
Help & Support (2 out of 5)
Net Librarian offers only 60 days of email support for the purchased product. This is only sufficient if supporting documentation was exceptional, but…
Documentation on the product is almost non-existent. Perhaps Net Librarian is thought to be so simple that it does not need significant support; however, there are many questions left unanswered by the FAQ. The FAQ, called Web Help in the Help menu is comprised of all of three questions.[See image 17]
Search Input (3 out of 5)
Net Librarian claims to use more than 20 search engines in compiling the sites for the application to parse, but there is no formal list in documentation or the web site. The URLs for the searches are accessible in the track.txt files.[See image 16] Working through the .txt file, which can get very large (over 10MB and more) the various search engines will be revealed. Mamma.com and Ask.com are represented.[See image 16]
There is no way to control the search other than providing the keywords and whether to stop and continue later. Restricting which search engines to use or not use is not an available setting through the application. The search results do not disclose which search engine(s) provided the hits for the objects saved in the search. The objects returned are fairly granular, but how it arrived there is not.
While Net Librarian queries for various components such as images, it has not evolved to include newer searchable objects, such as XML, RSS, or OPML files. The interface also does not employ any Web 2.0 technologies, which could potentially improve the usability of the application before and after searches. The application is tied to one browser, Internet Explorer, and therefore, in addition, tied also to one operating system (Windows). The backend for stored searches uses thAccess 97 database format and could benefit from a significant update. Finally, Net Librarian is not ready for Windows Vista with some components unavailable in the operating system by default.
Search has evolved greatly since the concept behind Net Librarian was released as an application and somehow Net Librarian has not kept up. The feeling I had while using this product is that Net Librarian was a very useful solution many years ago. It works in the background assembling detailed search results using websearchstore.com as a saved cache to expedite future similar queries. While there is some residual value in the methods used by Net Librarian, there are better options in the market for search and online research. If someone is well versed in earlier versions of Net Librarian, then maybe 8.0 offers more value.
Copernic, Google, Ask, Desktop Search, Yahoo!