Reading: Design is the Problem

I’ve been reading Design is the Problem: The Future of Design Must be Sustainable. LibUx was doing it as some kind of book club thing, and while I’m horrible at book clubs, I *am* open to book recommendations. So, Design is the Problem.

It makes pretty decent bedtime reading–engagingly enough written that I can stay awake for more than a couple paragraphs, not so thrilling that I will glance at the clock and realize that I should have been asleep hours ago.  I have a surprising amount of difficulty finding that balance, so it’s worth noting.  Also worth noting: I don’t have my ILLed copy of the book at hand, so we’re just going to pretend that my intent was to capture broad ideas and not worry about details.

Things I like as of right now, when I’m maybe 3/4 of the way through the book:

  • The focus on systems.  Systems make me so ridiculously happy.  Don’t just think of the system you’re designing in–think of the systems that system is a part of.
  • The idea that in nature, one system’s outputs are another system’s inputs.  No waste!  What would that look like in a library?  I keep trying to figure that out, and coming up with things like–I produce lots of data when I’m doing usability testing.  I do my best to share it, but I can’t force others to reuse it, and I don’t generally have time to reuse it myself.  Imagine if that were part of a larger process that I could dip into when I needed to and/or contribute my data to, and know that it would flow onward to someplace where it would see further use.  Being able to reuse others’ output, whether in usability data or anything else, would mean I don’t always have to start from scratch!  (And neither does anyone else.)
  • Building on that idea–the idea that a subsystem’s “waste” could actually improve the larger systems it’s part of.  Imagine a factory that purified a river instead of polluting it.  What would be the library equivalent?  Must figure this out.
  • The idea of river systems vs. lake systems.  In a river system, inputs come in, flow through, and pass right back out.  In a lake system, inputs come in and get used and swirl around and get reused, maybe several times, before they pass out of the system.  They contribute value to the system instead of just passing through.
  • Recognition of the fact that there are many kinds of sustainability: social, ecological, financial, etc.  Come to think of it, there are probably as many kinds of sustainability as there are kinds of systems.
  • The case studies.  I really enjoy the case studies embedded throughout the book; they take the abstractions and make them concrete.

Things I don’t like:

  • The footnotes.  I mean, I like footnotes in general.  Footnotes are good, right?  They often contain citations, and I like citations.  But the sources this guy is citing in his footnotes–!  Wikipedia, repeatedly.  And news articles.  I mean, nothing against Wikipedia and news articles, but seriously, these are the chief sources you use for your book?!  You couldn’t even be bothered to chase down the sources those sources cited?  My inner librarian is crying.

    Every time I say to myself, “Oh, that’s interesting!  I wonder where he pulled that from?” and then look and see it’s Wikipedia, I immediately begin to doubt the point he just made.  Is it actually true?  Maybe, but it just got added to the “Interesting but unverified” column in my brain.

Anyway–it’s worth a read, and possibly also worth further research to verify all the things the author relied on Wikipedia for.


Software for Visualizing Data

It’s been a rough couple of weeks, posting-wise–but all is not lost!  I am still learning fabulous and ridiculous things.  My latest rabbit hole is visualization software.

See, a year or two ago I volunteered to write a script to pull more meaningful data out of the spreadsheets of data generated by my library’s reference stats software.  The spreadsheets were in Excel, so I channeled high school computer class and wrote Visual Basic scripts.  Those scripts counted stuff up and threw it in tables.  Not nearly as fun as the Visual Basic project I did back in the day where you pressed buttons and an image moved around the screen, making annoying noises… but definitely more useful.

The problem?  We keep coming up with different questions to ask of our reference desk data.  I’m not particularly fluent in Visual Basic, and as much as I do enjoy coding, I don’t really enjoy coding in Visual Basic to manipulate spreadsheet data.  Plus the boring tables of numbers that are all I was able to output are, shall we say, less than inspiring.

As a result, I spent much of yesterday messing around with software that would help me slice, dice, and visualize my data.  For the record, “business intelligence (BI) software” is at least as good a search term as “data visualization software,” at least for what I was doing.

I started off with Google Data Studio, because I read about it somewhere recently and it seemed nice and approachable.  It did have an unintimidating interface, but I spent *way* too long fighting with it over whether my many columns of dates were actually dates.  I reformatted those suckers at least five times trying to convince Google Data Studio that, yes, actually, these ARE dates–but I still didn’t win that fight.  I was trying to find workarounds when I realized that the graphs GDS was showing me showed one subset of my data when I sorted them one way, and a different subset when I sorted them another.  Is it really too much to ask that it show me all seven days of a week, all at once?  I couldn’t find a way to make the graphs display all the data at once, the help pages were not helpful, and at that point I decided that Google Data Studio is not for me.

So I started playing with SandDance, because someone linked me to it ages ago and I remember thinking it was cool.  I think the interface had a bit more of a learning curve than Google Data Studio did–it took me a while to figure out how to include/exclude parts of my data.  I did get the hang of it after a while, but I was stumped by the fact that I couldn’t find a way to zoom in and look more closely at the graphs when I had a lot of bins going on.  My labels weren’t visible, which meant that I couldn’t tell what I was looking at, which meant it was fairly useless.

At that point I started searching for recommended alternatives, and a result I kept seeing was Tableau.  And, oh yeah, I *have* heard of Tableau. But ye gods, $999 for a personal copy of the desktop version, or $500 to use it online for a year?  The library is great about buying me software, but I would have to know for a fact that we were going to get some serious mileage out of this before I could justify that kind of money.

Except it turns out that there is Tableau Public.  If you’re willing to make everything you use Tableau for public, you can use it free.  So that’s cool.  I mean, I work at a state college so I’m pretty sure all our data has to be public anyw–oh hey, what’s this?  “Tableau Desktop is free for students and instructors around the world“?  Um, yes, please!

You actually have to get approved for the free-for-instructors thing.  In the mean time I’ve been playing with Tableau Public and all I can say is, why was I wasting my time with Google Data Studio and SandDance?  I can tell that there’s a lot of functionality I don’t know how to use yet in Tableau, but it hasn’t gotten in the way of me doing some basic stuff right off the bat.  And I can always read the labels on my visualizations!  And I can export them as PDFs that I could hand to co-workers even if they aren’t at all tech savvy!  And data doesn’t disappear and reappear depending on how I sort things!  And I can *manually* sort things, so that I don’t have to choose between having the days of the week show up alphabetically or in descending order of how many questions got asked!

In short, I think I may be in love.  This is so much more fun than trying to remember how to select a cell in a different column in Visual Basic.