Your boss comes to you and says..."I need you to write the spec for a new product to address XYZ" ... and you have no idea what XYZ is. What do you do?
- First, admit to yourself and your boss that you don't know anything about XYZ.
- Second, ask for some time to become acquainted with the subject
- Determine if your company already has anybody expert on XYZ. If so, make them your new best friend.
- Failing that, it's time to "hit the books".
Actually, in today's world, it's more appropriately "pound the keyboard". If XYZ is something relating to a mature technology, you may just have to dig out that old college textbook you've been hording. Or maybe make a quick trip to the local bookstore. More likely, XYZ didn't even exist when you were in school and is obscure enough that you're not likely to find anything at the local bookstore.
In the world of online research it's all about the keywords. Once you've got a few combinations of relevant keywords, you're on your way. Did your boss give you any papers on XYZ? Did they have a bibliography or set of references? Is there a common thread there that you can exploit when using search engines?
If not, then go to Google and search on "XYZ".Depending upon the term, you might get thousands or even millions of hits (Googling the literal letters "XYZ" yielded 37.8 MILLION hits when I tried it!). Scroll down until you find something, anything, that relates to the subject you are actually interested in.
Once you've got that first thread, you're on your way. Things to think about include:
- Look at references and footnotes to find more threads to follow.
- Did any companies dealing with XYZ come up in your search? If so, see if their website has any educational material on the subject. Most do.
- Try your search terms on CiteSeer. The nice thing about CiteSeer CiteSeer is that each paper is cross-linked to papers it referenced and papers which referenced it. It's like tapping into a disembodied stream of conciousness on the subject of interest.
- Assuming its an EE-related subject, make sure that you try out the ACM and IEEE digital libraries. You may need to do this at work in order to get access at no cost to you. Alternately, stop by the local university's engineering library if you find references to a paper, but no access.
- Find out if there are any magazines or journals that deal with the subject. A good place to start is the "References" section of this website. Don't forget to search some of the general interest sites such as EDN or EE Times. A two or three paragraph filler article here can often be your starting point.
- Is XYZ something likely to be either taught at or the subject of research by a university? I've always been interested in systems engineering. The Berkeley web site shows up in my online research fairly often.
- Master's and PhD thesis are commonly placed on the web, add "thesis" to your search terms.
- Other good keywords to toss in to narrow your search are: tutorial, reference, introduction and application note.
- Are there any applicable standards? Which standards bodies? Do they have a website?
- If XYZ is a new product or process you're considering, has someone else already patented it? Try out the United States Patent and Trademark Office Search. They have an online search engine to search U.S. patents issued since 1790. You will have to register to download the full patent, and there is a small fee ($3 on the sample we tried). You might also see if your company already subscribes to a patent service, in which case the fee becomes a non-issue.
By now, you've got the idea. Don't expect to become an expert overnight. But if you're lucky, you might score a tutorial on the subject and at least be able to start discussing the subject. If, after reading this tutorial, you want to know still more about research methods, check out William Badke's online text "Research Strategies"
Also, don't neglect traditional publishers. Web searches may get you started on a subject, but often nothing replaces a paper book for bringing structure and depth to a subject. So don't neglect Amazon . You may have to spend some money on books, but I've always found it to be a good investment. Usually I find that I combine the two approaches. For instance, during development of this website, I've made frequent searches on various topics related to web site development (Java, CSS, HTML, ...). At the same time, my desk is loaded down with two books on Perl, three on cascading style sheets, and others on Java, HTML and XML. Each source brings a different perspective to a problem, and they all have their place.
Finally, be tenacious. If you truly need to become expert on a subject, the web is a great place to be. All you need is time and a good connection to the net...
- Mike Stanley -
- IEEE Xplore
- ACM Portal
- Google Advanced Search
- Academic Business Literature Digital Library (SmealSearch)
- Dr.Dobb's (computer science related)
- United States Patent and Trademark Office Search ($)
- arXiv.org - an electronic archive for research papers
- Scirus science-specific search engine
- Datasheet Locator
- IC Master
- Electronic Engineers Master
- Application Reference Materials (Hearst Electronics Group)
- Semiconductor Applications (Hearst Electronics Group)
- Electronic Products Reference Design Directory
- NSSN Search Engine for Standards
- Defense Technical Information Center Search Engine
- Units & Constants at the National Institute of Standards & Technology
- MEMS World Online: "covering all primary conferences, journals, and patents of the field published since 1966."