Thank you, thank you everyone! Your advice has been really helpful. The pep talk will help me get through this period that feels like the equivalent of practicing your scales (ugh) when you're first learning music. I want to play "real" music, write my own songs, but I still have more drills and technical exercises to master. I have a tendency to dream up projects at the moment that are beyond my skills and it can make me feel a bit lost, but I guess you don't grow unless you get in over your head occasionally.
Another analogy is that sometimes it seems like learning programming (et al) is similar to learning any sort of language. You learn vocabulary, grammar and sentence structure, but you're also trying to write a creative essay (or heaven forbid, a novel) at the same time. You don't know the depth of everything you don't know, but you have to plow forward despite your naivete.
<off topic rambling>
Since y'all have given me help, I want to give something back. It's nothing to do with code, but I hope y'all enjoy it anyway. It's music by the Japanese composer, performer Susumu Hirasawa who builds all his music with samples he pre-records and uses a computer to cue while on stage. He overlays the his vocals, guitar, keyboard and some other samples while on stage. He pegs the geek meter for me.
>>> On 5/8/2011 at 12:49 PM, in message <[log in to unmask]>, "Riley, Jenn" <[log in to unmask]> wrote:
From what you've said here (already done some OAI-PMH harvesting and
implementing a personal project as JSP), it sounds to me as if you're
light years ahead of most people in your situation. So my first bit of
advice to you is not to sell yourself short.
In addition to all of the excellent ideas raised here, I'll suggest some
additional strategies for implementing those on-your-own projects that
you use to build skills and concrete outputs to show off. You might
consider picking some existing OS software to implement, and then learning
it, well and deeply. Then add features or otherwise customize it. For
example, get Omeka (or some tool written in a language you have basic
familiarity with or want to learn) up and running. Examine the code to
figure out how it's put together - what's easily customizable and what's
deeply baked into the current implementation, where does it rely on
existing libraries and where does it start on its own. Add in a Google Map
and/or a Simile timeline (if it doesn't do those things already - sorry I
haven't been following that closely). Write some code to parse and load
data from various sources (Amazon API, Freebase, any of the music
services, MARC via library catalog, DC or other XML format via OAI-PMH)
into the system. Revisit it again a year later to see how your
implementation decisions have held up in light of changes in underlying
There's a goal to these sorts of activities that goes beyond the obvious
"learn about the mechanics of this programming language." They give you
experience with implementing various tasks, not so that you can do exactly
that again, but so that you can do it better the next time. You'll learn
from these experiments strengths and weaknesses of various approaches to
solving particular technical problems, and the ability to evaluate
different ways in which you might solve a problem in order to pick the one
that best fits your situation. With some practice doing this evaluation in
relation to the code and requirements at hand, over time you can extend
this analysis to wider technical and organizational infrastructures, and
make good decisions about technical implementations given surrounding
organizational realities. Doing some work inside a pre-existing software
application I believe will help you work on these sorts of larger issues
in addition to the mechanics of writing the code.
Now I'm all inspired to drop what I need to be working on today and play
with Omeka. I'm not a coder, myself, so I'm sorry to say my advice here
does not come from personal experience taking this approach. It does come
from spending lots and lots of time working with developers and driving
digital library initiatives, and seeing where development initiatives go
well and where they don't. The best developers I work with are the ones
that know it's not just about the specific technical task at hand, but
rather can talk intelligently (and not just to other developers) about the
implementation decisions they've made and evaluate their effectiveness.
Best of luck. It sounds to me like you're ready to take the next
opportunity by storm.
Head, Carolina Digital Library and Archives
The University of North Carolina at Chapel Hill
[log in to unmask]
On 5/6/11 4:06 PM, "Ceci Land" <[log in to unmask]> wrote:
>I like this. Maybe it's because it's what I was already thinking about
>doing. I have 3 project ideas twirling around in my head at the moment.
>I can't do them at work, but perhaps the systems department could give me
>a dataset to play around with in my spare time. I already have a good
>dataset for one of the projects that I harvested via OAI-PMH.
>Do these spare-time projects get any respect from the "real world" when
>it comes time to apply for a job? ....particularly if you focus on really
>making it as polished as possible (within the limitations of a non-work
>environment)? I remember building my own darkroom as a teenager and
>doing B&W and color slide and print processing. (yes, I still love the
>smell of D76 and stop bath. I can bring up the smell purely from memory
>:) ). I did manage to work for a while in photography because of my
>original personal investment of time and energy into it as a hobby. I'm
>just concerned that the things may not work that way any more. Life was
>not only slower paced back then, but having an exact skill match wasn't
>required to get a foot in the door. Plus, I'm no Mozart so it's not
>likely that I'll come up with something uber creative or so nifty that
>it's used by a community at large. But I do good technical work. I
>tinker...I make things "go".
>Thanks for the advice. I'm going to start playing with the projects I
>have in mind. One is already done as a JSP, but I think I'll convert it
>to something else and "clean up" the compromises I had to make to get it
>done in a limited time.