Computer Troubleshooting for Librarians

Troubleshooting is a specialized domain of problem solving that is mostly concerned with computers, electronics and mechanical systems. As librarians we have to deal with all of these as a part of our job and in our private lives.

The principles of troubleshooting are usually specific to a field, electronics engineers do one type, programmers use another set of tools. There are general principles that apply across all disciplines, and I’ll try to bring out a couple of specific techniques for computers that can help us solve problems on our own.

First, keep it simple, stupid. Most computer problems are solved by restarting the computer. Write down any error code the computer shows. Connection points are the most likely places for any connection to fail, so check the cables by pulling them out and putting them back in carefully.

Second, use your resources. Have any of your co-workers had this problem before? Ask a question on a listserv, or read the manual. If your manuals are electronic, use your search skills. A good desktop search program, like Copernicus or Google Desktop Search, limited to just that set of files, will give you better results than the built-in Adobe or Microsoft search.

Third, just Google it. Operating system errors and program errors will have happened to other people, many of who are more experienced at troubleshooting than you or I. This will not work for your proprietary systems like your ILS, especially if you are using something more than 3 or 4 years old. Use advanced search techniques, you’ll often get better results than the IT department. (Yes, this is what they do.)

Fourth, remember that this computer problem is just a research question. Write down the terminology. Map the concepts, Draw the connection between widget A and widget b. Like any good subject librarian, you need to know a broad overview. Don’t worry about programming unless you’re interested in it. Start with a binder. Move on to a wiki if you have the interest.

Fifth, and finally, only solve the problem once. If you discover a solution, or IT helps you solve it, write your steps. Just like finding aids for students, these pathways will get you thinking when you run into the same or related computer problems.

Organization & the Digital Collection

I’ve been spending a great deal of time recently harmonizing our database lists between our different online locations. I recently completed a first draft of a unified database list. Tech limitations have been the main driving force behind this exercise in excessive librarian geekiness. While updating our OPAC requires learning web design circa 2002, it is simple HTML. The “easy to use” backends of vendor tools have been causing me significantly more problems.

On our legacy OPAC we have divided the databases up onto general categories to point faculty and students toward general sources and more specialized databases by program. This is the setup most of our patrons are used to. I am working on duplicating and standardizing these categories across all of our services, including our OpenURL resolver and our federated search. It is in these tools where most of my complications have shown up.

With Serial Solutions recent update and merger with WebFeat we gained a lot more control over how our data is displayed. Nonetheless, there are some frustrating gaps in the configuration options.

While I can create the display order for our databases, I can’t actually create categories on the “search by database” page. Where I can create categories, the names of the databases are obscured.

Serials Soloution, in their efforts to make the backend easier to use are determinng policy and practice for our library. We’re engaged in a decades long experiment in throwing Ranganathan’s First Law right out the window. Our libraries are now digitally chaining our items, limiting knowledge to forms, digital locations and uses permitted by the content owners. Some of this is ignorance, but much of this change from physical ownership to digital licensing is being pushed past us because Librarians are not paying attention to the implications of licensing arrangements.

What’s the soloution? I’m not entirely certain, but I have some ideas. If you have a chance to work with a consortium, do so. If your purchasing budget is such that you can effectively bargin with vendors, choose those with more open policies. Work with open access programs at your University to reduce the stranglehold vendors have over some types of content. Complain about poor design decisions, the lack of cutting edge options and the paucity of open API’s to your software vendors.

Our new book chains are not the only option as we transition from physical books to a digital age.

Thoughts on Chrome OS and Libraries

With the Chrome OS announcement and reveal today, a few thoughts cropped up on this as a possibility and a threat to libraries and freedom of information.

First, in the spirit of Cory Doctrow’s ‘Little Brother’, if these catch on there are some significant concerns. Whitelisting could be embedded in the OS at a level where it would be very difficult to subvert. Our laptops and desktops today are vastly overqualified for the tasks we use them for. That’s both a problem and an opportunity.

Second, I kind of want to lock down my OPAC terminals to that level. The speed issue is great too, especially if it can take under 1GHz  computers and make them start up that fast. You could replace the current low-end enterprise desktop models with a ChromeOS thin client for the web. You could provide secured kiosks for the library system with a very minimal financial outlay.
Third, why no integration with Android apps? That seems like a major oversight. I don’t think I’d trust a web based encryption program, nor am I particularly impressed with the current state of video or photo editing online. I don’t even want to talk about the current state of the ILS ecosystem in this context.

Ditch the Twitter Client – Use Firefox

I twitter a little bit, and I use it to integrate my status updates for Facebook and LinkedIn. I’ve tried a few twitter clients, but they don’t feel very intuitive. My current social media setup is TwitterBar with the RSS Ticker addon used to pull updates from Twitter, Facebook and LinkedIn. This works extremely well for me because Twitterbar comes bundled with the One Riot search engine, when I’m looking for realtime information, a quick search is as easy as a Google search.

So, if you want my setup, here’s how to do it.

1. Install Twitterbar
2. Install RSS Ticker
3. Add the RSS feeds you want as Active Bookmarks in Firefox
4. FACEBOOK – Check out this link to figure out how to get the RSS feed
5. LINKED IN – Check out this link to figure out how to get the RSS feed
6. Twitter… if you can’t find this RSS feeed, turn in your Firefox card
7. Success!

Enjoy!

Edit – The loose nut behind the keyboard missed some links. Thanks Ahniwa.

The Scroll and the Codex: Two Different Views on Ebooks

Due to the rather unique collection in our library and our location, we depend pretty heavily on electronic journals and ebooks. I am dealing with a lot of problems with pdf ebooks for research at the moment, and I also read a lot of fiction (Mobipocket and an old Palm T|X at the moment) in ebook format, and I’m struck at the differences.

We are at a transition period in how we consume information, and it appears that we have simply carried forward the information access issues that existed at the turn of the last millenium.

I’m talking, of course, about the scroll and the codex.

A scroll is a single long sheet with words printed from top to bottom. A codex is a book as we would recognize one. I’m dealing with both in interfaces optimized for the scroll.

PDF ebooks have been particularly difficult to deal with. To have the text at a readable level, you have to zoom in significantly. This puts the top and bottom of the page outside of the viewing window of most modern computer screens, especially with the standard 17 inch screens at most workplaces. AJAXy interfaces like Library Press Display make even less sense on small screens. This is because our standard screens work best with an internet-pages, which imitate a scroll.

The small screen on my palm T|X suffers from the same problem, however using the Mobipocket reader converts everything into a scroll.

The scroll is fine for end-to-end reading. It has it’s own built in bookmarking feature. However, it is useless for reference purposes.

The codex is great for reference purposes. It bookmarks well, but not as good as the scroll. It’s more portable and durable than a scroll.

Why are we forcing something so revolutionary and different, viz the electronic book, into these two formats? Using PDF files, which force a codex form on a scroll, or plain text/mobipocket/EPUB files which roll out like a scroll, but suffer from a serious lack of precision in layout and destroy the benefits of the codex.

So, what’s the solution? I’m just a librarian, not a software engineer. Stick with the format that fits. Scroll-type ebooks are great for novels.Forced-codex PDF files are good for making sure that the text and pictures look good when they’re printed out (and very little else). Paper books still have a pretty good advantage over both despite the weight  and size problems. We still don’t have a format for printed words out there that takes the advantage away from the evolutionary, revolutionary, incredible paper book.

A summary of what I am thinking about and reading today

• 00:49 trying out twitterbar #
• 01:24 Please Snark This tinyurl.com/yhs8fzt #
• 12:28 testing the facebook connection #
• 12:57 Ironic that the bastion of the Canadian free market can’t manage it’s own finances. tinyurl.com/yf452px #

For more, check out my Google Reader share at http://www.google.com/reader/shared/01139605564737264142

Backup. Backup. Backup. Backup. Backup. Backup. Backup. Backup.

Did I repeate that enough?

Backup.

I just got the details on how our ILS server is backed up. I’ve been making a backup every time I make changes to the OPAC. And that’s saving my butt.

I’ve been making extensive changes to our online catalog, most of which are cosmetic. Cosmetic changes in the world of ILS software can have consequences far beyond what you think of  when you think ‘I’m changing the look and feel of the interface’.

I moved an inline set of searches to the sidebar and moved the small, embedded search boxes into the body of the page. I commented out a section of the sidebar that displayed the contents of the cart, intending to move this to another section of the website.

It broke the whole freakin’ use of the cart.

I, of course, had no idea that’s what I’d done, so to get a white screen of death (Cannot Open Error Screen) every time I tried to use the cart was pretty scary.

Furthermore, as I’ve been working on these changes since August, I didnt’ have a good idea as to when I’d done this.

Remember though, I’d backed up before every change I made. I restored each section of the website from the initial backups I’d taken, then copied over the backup of changes I’d made, rebuilding and testing the website each time. This allowed me to narrow the changes down to one template file. I restored this file from a backup and re-did the changes I’d made, testing and confirming that commenting out this cart display broke the whole cart.

The whole story is something straight out of The Daily WTF.

Breaking a display breaks the whole cart?

The whole library profession really needs those vaunted next generation ILS’s the cool kids keep talking about…

Systems Basics – The Password List

In the initial days of my job, I ran into a serious snag that took several weeks of work to resolve. Almost all databases have an admin section where you can get usage data, create links to other databases and OpenURL resolvers and update the look and feel of the databases.

You guessed the problem…. I had no passwords.

Many of these passwords were stashed in files, in purchase orders and in other librarians offices. I was able to assemble some passwords and a basic outline of our databases over a few weeks before the staff went on vacation. The rest, I had to send sheepish emails to our vendors to get those passwords reset.

That was an unpleasant experience, one that I would prefer to avoid in the future.

My first solution to this problem was a basic list. I created 3 major lists, ILS passwords, Service passwords, and Vendor passwords. This allowed me to consolidate the many little pieces of paper scrawled with passwords that lay cast upon my desk. This list remained my ’scribble list’ as we canceled databases, added new users and discovered (many, many) expired logins.

The second, and more permanent solution is KeePass.

KeePass is a password manager, a password generator and a safe, permanent way to control all the passwords a systems librarian has to keep track of.

For personal use, KeePass is an excellent program to use if you need to control a large number of complex passwords. For librarians… well, let’s just say a half-way competent hacker could make a total hash of any ILS currently on the market, especially with the way most of them remain illconfigured and unupdated. KeePass is secure, but it also allows one to export the data in a number of different ways, back up the database, and it’s portable. I keep two backups of my password database, one encrypted in my work email and two paper copies in separate locations. This may be paranoid, but it’s saved my data on three separate occasions. Paper copies don’t get corrupted.

You can carry around all your passwords on your memory key, or keep them on the remote drives provided by most IT departments. If you’re using a memory key, I also recommend the very cool launcher known as geek.menu. For the paranoid among you, this app is tightly integrated with truecrypt for an extra level of protection.

On Visualizing Projects Beyond Your Ken – Freemind

Don’t you love that phrase? Beyond your ken? It’s a Scottish idiom meaning beyond your ability to comprehend.

It’s very much how I’m feeling in approaching some of the projects here at CNA-Q.  Just doing edits to the OPAC requires editing several files at the same time, then aut0-generating the html, testing it with (at least) two different browsers and tracking those changes over a number of days. This is in addition to keeping track of admin passwords, login details, ip addresses, and even the structure and location of folders on the server.

I’ve been depending very heavily on Freemind to do a lot of this. It’s a mind manager, available as a portable app, which allows me to install it when and where I feel like it. I’ve mapped the entire library network, with our library server as the center, as well as two sub-maps of users and employees. Here is just a sample,  this is a map of our online databases.

This is just a screengrab, from the little icons you can see there are notes about each database, as well as hyperlinks to the library admin pages. The color coding is temporary, and relates to our implementation of 360 Link.

Green is ‘Not done Yet’. Yellow is ‘Working’.  Blue is ‘Full Text and doesn’t need it’

There is no way I can keep all these connections in my head. Notes are helpful, but being able to map out the different systems within the library makes a huge difference. I keep Freemind open behind my work window, when I need to navigate to a new area of the network to work on another project, I pop it open and click on the links embedded there.

One Month In – The Williams’ in Qatar

We did survive the trip! Flying a family, plus two cats from Calgary to Toronto to Zurich to Doha all in one 19 hour fun fest was one of the more difficult things we’ve ever done. For future reference, some things to keep in mind whilst traveling this far.

1. European and Middle Eastern airlines often have different requirements for carryon baggage. Going from Air Canada with a 2 carryon bag limit, plus carseat to Qatar Airways with a 1 bag, 7kg limit with no carseat cost us a lot of time and a few bucks.

2. If you are moving internationally with pets, give yourself 2 hours padding on top of the normal 2 hours you would give to get on the plane. We gave ourselves one hour extra and almost missed our Calgary-Toronto leg.

We’re living in the Al Jazeeraland 1 compound and we arrived just in time to bid the whole compound a happy vacation. This has left us the clubhouse and the pool largely to ourselves.

Doha is a beautiful city that seems to be constantly under construction. To Canadian eyes, especially to someone who has been immersed in the greenery of eastern Canada for the past few years, the landscape is very stark, variations of brown, dun, sand… it’s a desert. After a few weeks, you start to see a lot of differences, especially down by the Cornice. The desert is still everywhere though, as we discovered last week when a gigantic sandstorm blew in and filled our villa with sand.

Now, CNAQ… Wow. Beautiful campus with air conditioning set at such a level as to freeze a penguin. As the systems librarian, I’m spending most of my time parked in front of my computer trying not to kill our server or the IT department… joking guys… don’t erase my user account. When I step out for lunch, the 45 degree heat is a relief.

My summer is shaping up to be somthing of a hermit’s existence, I’ve been describing my job as ‘the html gnome’. I’m hard at work implementing our OpenURL resolver, our Federated search tool and once those are up we’re going to be putting Aquabrowser on top of the whole system.

So! If you’re brand new (or an old fart) at CNA-Q and you need library help, it’s Brett.Williams@ the normal CNAQ email address. Soon you too will be fighting with your office phone and getting locked out of your computer while freezing in the magnificent ice machine that is the entire CNA-Q campus.