[TOS] Need advice/opinions about Spring 2013 semester project

Fri Nov 30 21:48:06 UTC 2012

Hi Karl,

Good to hear from you, and that you're considering a FOSS project for the spring.  I've done a few of these projects with real clients in recent years.  They're a lot of work, but the payoff can be tremendous -- both for the client and for the students.  My website http://myopensoftware.org/content/new-projects-2012 summarizes the most recent of the projects that I have done with Bowdoin students.  

Here are some ideas that hopefully address your questions:

First, it is important to set realistic milestones for the project so that it fits within one semester and the client and the students get a real sense of success when they are done.  A sample set of milestones for recent client-centered projects that I've done is at  http://http://myopensoftware.org/content/extended-course-syllabus.  

Second, it is important that either you or Worcester State assumes responsibility for maintenance of the project after the semester is finished.  The client needs to have some assurance that the software will be supported in the future, and that it's not just another academic exercise.  For my projects, I have been doing this, but I still need to find a more viable solution to this problem long-term.

Third, a quick look at the application you describe leaves the impression that it is far more than a 1-semester project.  I'm also guessing that finding a non-proprietary code base that your students could adapt to fill the needs of the WAM will be difficult at best… maybe others on this list will prove me wrong (I hope so!).  If there is no established software that can be adapted for this app, you have the advantage of picking the licensing arrangement, platform, software architecture, programming language, etc -- and also the advantage of not having to work through somebody else's code :-)

Fourth, if you want students to start by doing a requirements analysis, that will eat up a lot of early time in the semester.  My projects all start with a requirements document in hand, so that students begin the semester doing development almost from day 1.  But at the end of the semester I expect them to have a working implementation.  Of course, that puts the requirements job on you and students don't learn that part of the software process.  So it's a trade-off.

Fifth, I use the GPL for licensing and the students and I share the copyright to the finished software -- Bowdoin has no interest in getting into the software licensing fray.  This allows the software to be freely replicated and refitted for other clients elsewhere, which has in fact happened for some of my projects.  (All my projects were built from scratch, so there was no problem with inherited licensing restrictions.)

Remember, there's a difference between license and copyright -- the license is about ownership and the copyright is about authorship.  So you can have a FOSS license (like the GPL) which allows the software to be freely shared and modified and still claim authorship of the original work (through the copyright notice in the software).  For an example of such a copyright notice, look at the one at the bottom of the page at http://rmhportland.org/volunteers/homebase.

Finally, you might want to talk with Bonnie MacKellar and/or Mihaela Sabin about your ideas.  They have recently completed FOSS projects with their students that used a similar approach to the one I outlined above.  In fact, the three of us just submitted a paper that addresses many of your questions.  Small world!

Anyway, I hope these comments "from the trenches" are helpful.  Let me know if you want to talk more about this project -- it looks very interesting and I'd be happy to help as needed.

All the best,
Allen

cc Bonnie, Mihaela

On Nov 30, 2012, at 12:18 PM, Karl R. Wurst <karl.wurst at worcester.edu> wrote:

> The last few years, I have had the students in my Software Development Process course work in an existing FOSS project. I was planning that again, but have been approached about another project and am trying to decide whether to do it, and how to handle it.
> 
> Worcester State University has entered into a partnership with the Worcester Art Museum[1], with a faculty member on our campus named to the position of Presidential Fellow for Arts, Education and Community. This person has contacted me about a request from the museum to have our CS students develop an app or apps to support a major re-installation of their extensive collection of Old Masters works. They are planning a salon-style installation similar to the one at the Museum of Fine Arts, Boston[2]. The impression that I get from our discussions so far is that they want an app that lets patrons view catalog information about the exhibit, and comment/post about their own impressions/experiences of the works and participate in extended asynchronous conversation about the exhibit. I am intrigued by this project and am seriously considering it as the semester project for my Software Development Process course.
> 
> I have not met with the WAM staff yet, but I have been thinking about the project, and written a list of questions to ask them when we do meet. Some of these are questions that I think I need more information on before we meet. I am hoping that TOS will have some answers, or at least opinions and/or advice.
> 
> 1. I am still hoping to incorporate FOSS into this project. The first idea would be to find an existing FOSS project that does something similar to what they are looking for, and use/adapt that software and contribute our work back to the project. If you know of any project that sounds remotely simliar to the (somewhat vague) description I've given, I'd love to hear about it and investigate it.
> 
> 2. If we have to develop it from scratch, it would be nice to release this as a FOSS project that other museums could use/adapt for their own exhibits.
> 
> 3. I am concerned about being stuck maintaining the project, once the semester has ended and the students have graduated. Adapting existing software may help with this problem, as there would (hopefully) be a community support it. Starting our own project for this is not going to solve this problem, at least in the short term.
> 
> 4. How to handle the Intellectual Property?
>    1. Obviously, the static content (catalog material) created by WAM staff will be owned by WAM.
>    2. Patron-generated content (comments, posts, etc.) probably becomes property of WAM as well, but there will have to be an appropriate legal statement to that effect, that the user will have to agree to before using the app.
>    3. The program code itself is a bigger question.
>        1. I don't believe that WAM will own it - since the students and I are not being paid/under contract to produce this, it is not "work for hire."
>        2. If WAM wants to own the code - there would have to be a contract that all the students (and I) would sign, that would transfer ownership as a donation to the museum.
>        3. I know that any portions of the code that I produce are owned by me - this is specified in our faculty contract. I can, however, license or transfer it to someone else at my discretion.
>        4. I believe that any portions of the code that the students produce are owned by them, but they could license or transfer it.
>            - This may be complicated by the use of University resources in developing the code. That may give WSU some claim. I need to investigate this.
>        5. My preferred solution would be to license the code under an open source license. But who would own it?
>            - I guess WAM could, but I'm not sure if they would want the liability.
>            - I guess WSU could. I can investigate that.
>            - How do most projects handle this? Do they create some sort of legal entity to own the code, and have all contributors license their contribution to that entity?
> 
> Lots of questions. If you have any opinions or advice, I would appreciate it.
> 
> Thanks.
> 
> [1] http://worcesterart.org/
> [2] http://www.mfa.org/collections/featured-galleries/european-painting-1550-1700-and-hanoverian-silver
> 
> -- 
> 
> __________________________________________________
> Karl R. Wurst Ph.D., Professor and Chair
> Computer Science, Worcester State University
> 486 Chandler Street, Worcester, MA, USA 01602-2597
> Email: Karl.Wurst at worcester.edu
> Web: http://sharepoint.worcester.edu/faculty/kwurst
> Phone:  +1-508-929-8728
> Fax: +1-508-929-8156
> 
> _______________________________________________
> tos mailing list
> tos at teachingopensource.org
> http://lists.teachingopensource.org/mailman/listinfo/tos