Advice to graduates just starting out

A few weeks ago the Sunday New York Times had a collection of wisdom from college commencement speeches. Here's one from John Patrick Shanley, playwright, speaking to the class of 2009 at College of Mount St. Vincent:

Not to bring up something upsetting, but when you leave here today, you may go through a period of unemployment. My suggestion is this: Enjoy the unemployment. Have a second cup of coffee. Go to the park. Read Walt Whitman. Walt Whitman loved being unemployed. I don’t believe he ever did a day’s work in his life. As you may know, he was a poet. If a lot of time goes by and you continue to be unemployed, you may want to consider announcing to all appropriate parties that you have become a poet.

Another dumb idea

Students, don't waste your time with ideas like this: Corrupted-File.com. What a lame way to get around writing a paper.

(Basically this is a web site that promises to send you a corrupted Word, Excel, or Powerpoint file so that you can send the corrupted file to your insructor and buy yourself a few extra days writing a paper.)

First of all, anyone who needs a few extra days to make a Powerpoint.... ?!?!? I don't know. Second of all, I can't wait to get one of those files sent to me :) I will love to look inside it. Thirdly, I wonder how much they make at $3.95 a pop? Also knowing that there's no recourse if they send you a file that doesn't work, or if you get caught.... geez. What a racket.

Finally, whoops, by publishing this, I guess I'm annoying the owners of the site who say "Keep this site a secret!"

Mechanical Turk

I have to tell you I feel absolutely fascinated by Mechanical Turk (the original Amazon version and the nascent generic versions), from the perspective of micro-tasks, from a labor perspective, from a computer science perspective. Here's the wrap-up of a recent Mechanical Turk meetup about these topics.

How to be happy in your work

How to be happy in your work: Memorize this Venn diagram: What we do well - What we can get paid to do - What we want to do

Open Source Conference, Day One

I'll add to this post as I go through the day.

Where am I? I'm blogging from the 5th International Conference on Open Source Systems in Skovde (pronounced HURVdeh), Sweden. I've been to the previous 3 conferences as well, but missed the first year.

Who else is here? After a show of hands, we determined that more people are here from business than government (of course most people are here from academia), and more than half of the people are here for the first time.

What am I doing here? I am presenting a paper called "Using FLOSS project metadata in the undergraduate classroom" that I wrote with my colleague Shannon Duvall at Elon.

Day One:
Women in OSS breakfast 7:30-9:00. We were greeted with a lovely breakfast for everyone, and then there was a separate short meeting with some women interested in OSS. We had a speaker Eliza Roszkowska Öberg, a member of Swedish Parliament, briefly present about what Sweden is doing with e-Government and how open source is involved there.

Keynote speaker: Stormy Peters from Gnome Fndtn is talking about "Open source is changing the way work gets done". Her main points: Collaboration, across companies, history of open source, how to make money on open source a bit.

Session 1: Commercial open source
Session 2: Mining open source
Session 3: Communities of open source

Some notes on talk 3A1: Brooks' Law & open source.
--issues of coordination and early productivity.
--creating graphs of community cohesion based on shared artifacts as the links between people on a project. Take people and artifacts from svn repository (cohesion is average shortest path between people); edges between artifacts are weighted (e.g. people have worked on 8 artifacts together, this is weight); Use Floyd-Warshall to measure shortest paths between nodes; use Kamada-Kawai algorithm to place high edge weights together for viz purposes.

Social event & dinner: informal dress. Not sure where it is yet, but they showed a picture of water, so...?

Note about students working this summer

E-net posted a little story about some the funding I'll be using this summer to have two students work for me on my open source data collection effort.

office hours moved today

I'll be having office hours from 2-3 today (Monday) instead of from 10:45-11:45.

free software foundation INTERNSHIPS

The Free Software Foundation is offering internships for summer, fall, or spring.

This looks like a great opportunity. It's not just programming either, so if you have interest in any of the related areas, please apply.

Impressions of the Kindle, part two

First, a report that (and I quote) Old People Love the Kindle. Well, maybe I am getting older. And I wouldn't say "love" as much as "am fascinated by".

Anyway, second impressions of the Kindle are as follows. Not all books are welcome in the Kindle format. I see two main problems:

1. PAGE NUMBER CONCERNS

First, any book in which you need to know the page numbers will not work on Kindle 2 (or, the Kindle as I have it today). So, think about the following scenarios:

a. Assigning a textbook where half the class (and the instructor) has a Kindle and half the class has a paper copy. No more "ok, class, I need you to read pages 38-46, but you can skip page 40." Pages don't exist on the Kindle, so how to navigate this situation? What about even a book club? ("I loved that line on page 56 where the guy says..." you can't say that anymore unless everyone has the same version, either book or Kindle. Even if a book club is reading different versions of the same material, the page numbers are more likely to be close to each other. You can at least flip around to the right area of the book.) The Kindle concept of a "location" won't translate to paper, and Kindle search won't translate to paper either. This 'location' and 'search' is a definite strong suit of Kindle, I get that, but how to mediate between multiple groups!? Has someone built a web site to translate Kindle locations to paper pages yet? What about the other way 'round?

b. Citations become difficult (for example, giving the page number in a bibliography) unless there is a new APA/IEEE/MLA/whatever citation style that includes Kindle locations. I am not aware of how to do this. I should look this up to see if electronic books have been explained yet, probably have, and how the Kindle is affected.

2. FORMATTING CONCERNS

Any book that relies heavily on non-linear, tabular, or graphical objects is automatically not optimized for Kindle 2. Examples: programming books with large chunks of code to be studied, rock climbing books that show maps of routes, fitness books that show tables of exercises or foods, or sports books that show different pictures of people doing exercises. In the case of rock climbing, for instance, the different symbols used on the maps are critically important, and if the pictures aren't able to be "blown up" (enlarged) in the Kindle, then what's the point really?

I had one book that was so badly formatted that it permanently locked up my Kindle and I had to send it back and get a new one. It took 3 days to track down the problem. Amazon finally tracked the problem to a badly formatted Appendix in a book called The Portable Personal Trainer, that included a table on the glycemic indexes of common foods.

In addition, any book that relies on "as you can see from the discussion previously on page 138..." well there is no page 138. It's up to the publisher before publishing the Kindle version to insert URLs (hyperlinks) between locations, so the Kindle user can move the pointer to the hyperlink that says "page 138" and click to go to that place on the Kindle version.

Which brings me to my real concern here. The Kindle 2 has a lot of potential for things like this, URL hyperlinking and enlarged detailed images. BUT. The publishers aren't using the features. They aren't inserting these features into the "formatted" books. They're doing the bare minimum of formatting, basically running some software to turn whatever format their books are in into an .azw file, making a quick table of contents, and collecting their $9.99.

The real challenge in choosing an e-book therefore seems to be with being able to discern whether a publisher has formatted the book well before buying. Also, if the book is in multiple versions, such as public domain versions, then the user has to be able to choose the BEST formatted version. Amazon currently allows the user to download the first chapter or a sample chapter of many books available for purchase. I've found that this is helpful with choosing between public domain versions (should I buy the $0.99 version or the $0.00 version? What am I getting for my $0.99?), but it's less helpful with new books or with books that have these tricky graphical features that will be spread throughout the book, not just in the first chapter.

Amazon reviews are fairly worthless for discerning the well-formatted from the poorly-formatted. Amazon reviews allow everything from people whining about not receiving their used copies from some 3rd party seller ("One star! I never got my book in the mail!") to plot reviews ("Five stars! I love vampires!"), to opinion pieces about the cover art or the paper content of the book ("One star! Buy the other edition, this one has cheap paper!")

What we need is a comprehensive review site for discussing the formatting of e-books, or at the very least, we need the ability to search and group Amazon reviews by whether they explicitly discuss the Kindle/e-book formatting issues. I understand that Amazon has tagging, but it's not used well and it not scaling to the number of reviews they have posted.

If anyone has any ideas about how to handle these concerns, I'd be interested in hearing.

More O'Reilly books now available for Kindle

The number of O'Reilly books available on the Kindle was painfully small, but here is an improvement with 160 books now available. This posting explains ther decision to release these books and why they have not released more. Interesting.