Monday, October 29, 2007

Eight laws of library technology

by John Miedema

Greetings. I have worked in the information technology industry for over a decade, mostly as a web developer for IBM. One day I was in my local library, looking at the library OPAC, thinking, ‘Why isn’t this more like Amazon?’ That thought took me to library school. It turns out librarians were thinking the same thing, and they are busy reinventing the OPAC. To my surprise, what I learned at library school was that I was less interested in library technology than librarianship. I have recently launched a new blog,, where I intend to focus more on reading research and practices in libraries and in culture. But I have a number of thoughts on information technology that I have not unpacked. I wanted to do small justice to them by summing them up in a single post. I hope they are useful to somebody in the library field.

1. It all comes down to data and rules. Anyone beginning their journey into information technology must feel overwhelmed. The industry is so diverse, with so many vendors and products and jargon. No one can learn it all. So just dive in and learn something; in time you will see that it all comes down to two things: data and rules. Remember in the eighties when you learned WordPerfect 5.1? Then they made you switch to Word. Turns out the switch wasn’t that hard; the products were fairly similar. In short order, you found yourself learn Excel, then intuitively picking up on other software programs. Next thing you knew, you were building web pages. There is a snowball effect, and it keeps going. Maybe you’re at the point of designing an Access database, or picking up some JavaScript. Before long, it all blends. You hear talk about .Net, Web 2.0, web services, SOA, and the semantic web — cakewalk. In the end, there is not much new under the sun. It all comes down to data being sloshed around by the application of programmatic rules. Content and syntax. I hope that helps describe the big picture, and gives you courage to try anything in the field.

2. Organized information is handier than disorganized information. Just like closets. It sounds obvious, almost a definition of cataloging. Now let me offer this slant — any degree of increased order is helpful. There are many methods of increasing order: back-of-the-book indexes, full-text indexes, controlled vocabularies, taxonomies, etc. The thing all of these tools have in common is that they reduce the state of disorder or chaos in information to some degree. This is especially helpful as the quantity of information being managed increases. The task of a librarian in the information age isn’t necessarily to bring high-end classification systems to the web. Things like social tagging are catching on because they bring just a measure of order to large bodies of content. Do you have any innovative ideas that simply bring a little more order to information? You may have a major invention on your hands.

3. The rate at which data is being recorded is accelerating faster than our ability to manage it. Call it information entropy or info-glut. It seems that the ability to record information has necessitated it. How did organizations ever get by without all the data they now record, one has to wonder. The information technology industry keeps inventing new ways to cope with the situation — content management, business intelligence, tagging, and so on — but there is another practical option: collect less information. Will we be less informed? Not if we apply an old-fashioned solution, the scientific method. Scientists collect a finite number of observations from the natural world, apply scientific rigor (repeatability, etc.), and make valid conclusions more often than not. They don’t try to record everything. I’m just waiting for the day that some vendor clues into this, and packages it up as the next best thing. Maybe library science should pick up on this first!

4. Librarians should not build their own software systems. Librarians should experiment with every new technology out there. Librarians should become very technically literate … in order to know what they want, and what they are getting when they go to a vendor to purchase a system. System development is deep water. If a new technology makes it easy to get started, it will be all the harder to finish. You can’t build a system out of Web 2.0 widgets. Forget technology; go conceptual. Think very hard about what patrons want; most don’t know. There is an old joke among developers that systems would work great if it weren’t for the users. Users complain more about font-size than function. In truth, design is a two-step between users and experts. Once you designed it, hand it over to the vendors and their code-jockeys for development. I’ll make one significant exception; open source development has the potential to harness all levels of development skills into a worthy product; it just takes longer.

5. These days there is only one way to acquire a system: buy a package, and two, custom build it. No one does custom builds anymore, right? It’s too costly; buy a package. It’s the 80-20 rule: get 80% of what you want, and configure it for your organization. That’s the sales pitch. More often you get 50-50 or worse. Just because it’s shrink-wrapped doesn’t mean it’s a package. Think configuration. Ask your vendor how much configuration is required. Is it custom programming in disguise? That’s where the dollars drain out. Coding is not just keystrokes. Don’t treat developers like mechanics and they won’t treat you like business executives. The technologies are young compared to the automotive industry and there are few truly standard solutions; everyone is learning on the job. They are not building cars, they are building the assembly line.

6. RSS and XML are cooler than you think. RSS is a simple Web 2.0 technology that completely changes our relationship with the web. Instead of having to go to the web, the web comes to you! If you learn nothing else about Web 2.0, learn RSS. It’s a great step toward what’s coming next. If you want to learn the next most important thing, learn XML, god’s gift to the web. XML is a character based data format that allows disparate systems to talk to each other. It is the heart of Web 2.0, which is righteous on so many levels. It is easy to get started; at no cost anyone can micropublish through a blog. These technologies are just the beginning. Keep your eye on these buzzwords: web services, service-oriented architecture, and the semantic web. Librarians are already talking about semantic libraries. There’s lots coming down the pipe.

7. Print is the next evolution in information technology. If technology evolved in the order of its importance, then print would be the next big thing. There’s no question that digital technology is better for finding information. Scan those books, bring it on. But finding information is only half the picture. What is preferable for reading information? People talk about the continuum of data to knowledge. Data is something out there, on the web perhaps; knowledge is something in your head. We go through of process of taking information that’s out there, and internalizing it. That’s where print is so important. When it comes down to serious reading, especially of challenging material, there is no equal for print and books. There are many more examples where print has persisted where personalization matters. The business world still prefers print for signatures. Print has something that the greatest quality e-Book cannot have, fixity, the quality of unchangingness that we need to evaluate ideas. In the final analysis, we require something in our hand to make it real. It’s just the way we’re made.

8. Library technology is less interesting than librarianship. It is important to remember this. It is becoming a more distant memory now, but remember that not so long ago it was believed by many that digital technology would replace libraries. Librarians were told they could become knowledge workers in the private sector. I’m glad that seems quaint now. In the final analysis, information technology is just infrastructure. Want to code; go into computer science. Think back to why you went to library school. Librarianship is much more than technology. I’m sure you can speak to that.

No comments: