bugs and comraderie

Since I just spent the last day doing electrical work around my house (I now have not one, not two, but eight new grounded outlets, plus a light in my closet and living room), I am a bit peaked and have little of interest to spout on about.  So, I am going to show just how geeky I am (pretty darn, yet surprisingly unskilled) and talk about the bugs I have encountered using free online tools for work.

It seems like eons ago that I first noticed del.icio.us’ problems with capital letters.  All of a sudden one day, I stopped getting stuff in my network–and since I was getting regularly told about all the great stuff sig. other or b.f. at work were sending me, that was a bit of a problem.  I figured out what was going on fairly quickly, because the change happened right after del.icio.us implemented some upgrades to the networking capabilities.  Whenever someone posts something, the people they watch or have ever sent something to appear as for: tags in the posting screen.  So, theoretically, clicking on this tag will send the person that link.  Well, all of those tags are in lowercase letters, even if, like me, you have a capital letter in your username.  And, guess what?  If you have a capital letter in your username and someone tries to send you a link without editing the for:tag, you will never get the link.  I naturally contacted del.icio.us about this, but it appears to be a bug they can’t or haven’t resolved.  The capital/lowercase makes a difference in regular tags, too–Libraries is not the same as libraries, and Google not the same as google.  That’s actually a really good feature, if you are the type of person who wants to make a distinction in terms by using capital letters.

I’ve also noticed that del.icio.us has some problems with intersections.  Very, very rarely, you might run across a link where a combination of terms, say social+networking+libraries, doesn’t pull up every link you tagged with that combination of terms.  For the most part, it works perfectly, though, so I continue to use it and trust that I don’t miss anything.

At work, I have been working on a project to put resources on a web site for faculty.  The dean who wanted me to help develop this web site was looking for a static list of books and articles on various topics, like professionalism and leadership.  Being the rebel that I am, I just couldn’t stomach the idea of putting up a web site that was so 1995.  I decided to go for syndicating RSS feeds on the site: one for books and one for articles.  The articles would be PubMed RSS feeds (so not chosen by hand), but the books would be the library’s new holdings.

I tested out a number of tools for this project.  For the PubMed feeds, I looked at just about every RSS to HTML/javascript tool I could find.  (David Rothman, rock star that he is, pointed me to couple more and even built a page demoing all these tools after I asked him for his advice.  So now, everyone can have it a little easier.)  I found that Feed2JS worked well, except for it cut off part of the word “articles” in Related Articles for each PubMed citation–but just in IE (even in our install of this, fixing it seemed more complicated than finding a better solution, so I pressed on).  Grazr was great, but since it is in its own little widget thing, I was fairly sure that wasn’t going to pass muster for a web page at my institution.  Feedsplitter was the one I ended up deciding on, after an awesome work colleague of mine found it and installed it on a test server for me.  It runs a little slow (i.e., is not getting the feed updated so fast as it should), but works well in IE and Firefox, and is simple to use with easily customizable CSS, so it blends into the web page.

But then I realized that the feeds weren’t caching–in any tool I tried.  Every single time the feed updated, the old stuff would disappear.  If there was one article in that update, only one article would show.  So, I hit up my RSS expert pals again, who in turn hit up their RSS expert pals (I really love the biblioblogosphere, let me tell you).  I tried FeedJumbler (supposed to have a built in caching tool)–didn’t work.  FeedCatch (a FeedShake product that may not exist anymore)–also didn’t work.  That one wouldn’t even recognize that my PubMed feeds existed, due in large part to the fact that they don’t pass an RSS validation.  I ran it through FeedBurner–no dice.  Using Feedsplitter and Feed2JS’s native caching also didn’t work.  I am still working on this problem, which is primarily a problem because I am really not a systems person and have no clue whatsoever about any of this stuff.  Which is why I rely so heavily on the advice, good will, and kindness of others.  I am pretty sure the best solution will be to stick the RSS feed into a blog and then yank the feed from the blog.  I know that the PubMed to blog thing works, as I have seen it in action several times now.  I just have to butter up the people who help me with this at work a little more.  Perhaps a nice cheesecake or something…  I also plan to try this out with Yahoo! Pipes, especially since I haven’t gotten a chance to test it yet myself.

Then, there are my trials and tribulations with the book side of things.  I was going to use LibraryThing (costs a very minimal sum for non-profits, but even said small sum probably wasn’t going to be approved by the library, since it wasn’t even a library project).  Then, I decided BlinkList would be even better.  And, it would have been–if it worked.  BlinkList states that their badges (widgets, javascript includes, whatever) are designed specifically to highlight books, movies, etc.  They should be able to grab all Amazon images and display them.  So, I thought, great!  I am a big believer in the book jacket phenomenon, as I have mentioned before.  Well, the first few books I added to my account worked fine, but after a while, absolutely none of the “blinks” picked up the images.  So, that was shot.  I suppose I could have lived with that, but then I realized that the ratings in the badge were either not being listed or were wrong.  I have tried Listal as well, and that seems to work, but I am more worried that one will disappear.  Maybe I should be really persuasive about LibraryThing at work.

But, whining about this to my library technology expert pals brought me yet another amazing act of kindness.  A friend built me a WordPress blog that accomplishes everything that BlinkList had–book jackets, ratings, tags, description fields (for the call number), and more.  I haven’t quite decided to go with that solution yet, because of the need to install an enterprise version of WordPress here first, but it is simply unbelieveable to have people be so willing to help me out from across the country.  (Thank you!)

And, now, I’ll wrap up today’s installment by again thanking everyone who helps me so generously every day–starting with the H.D. boyz (C., A., P., and the other C.), and ending with my three most constant technology saviors and co-enthusiasts (D., B., and J.).  Thanks.


Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s