Final (likely) version of JES released, 18 years after first release
March 2, 2020 at 7:00 am 3 comments
JES 6.0 is now available at https://github.com/gatech-csl/jes/releases/tag/6.0. JES is the Jython Environment for Students — it’s a Python IDE implemented in Java and with support for Media Computation built in. It was a lot of work for a bunch of people. Here are the notes from the release as a summary and acknowledgement for all the effort that brought this version fruition.
This is likely the final version of JES, unless a Jython 3.0 is developed.
This version was brought to completion by Nigel Charleston, based on the beta work of Veronica Day and Audrey Zhang (see discussion at this blog post https://computinged.wordpress.com/2019/07/22/beta-release-of-new-jes-jython-environment-for-students-now-available-media-computation-for-python-ide/). Many thanks to R. Benjamin Shapiro for helping us with many technical questions.
JES 6.0 updates Jython to 2.7beta, uses the latest version of JMusic (from https://jythonmusic.me/), fixes many bugs, will run with Java 8, and creates a new facility to generate pictures from a collection of pixels and sounds from a collection of samples.
The Mac version is a little more complicated to run than usual. You will need to have Java 8 installed to run JES. Thanks to Brian Howard and Michael Stewart for helping to figure this out.
The rest of the Mac version installation instructions can be found at the release page.
JES was originally written by a team of Georgia Tech undergraduates taking Senior Design in Summer 2002. It’s been in use and (sporadic) development for almost 18 years now. The previous version of JES was downloaded over 71K times (see counts here). I would not have predicted in 2002 that JES would still be used in 2020, with little maintenance and no additional funding. Software has to be continually maintained, right? I claim no great genius behind the design. How did it happen that it’s still working and being used?
An even more interesting example is our Squeak-based Wikis (Swikis) which were first developed in 1997. Jeff Rick created the version that we used in classes, and wrote about the process in what I think is the first ACM publication on wikis in 2000. Even after he graduated in 2007, they just kept going. The server http://coweb.cc.gatech.edu/ is still running today — I can find at least one Swiki there dating from 2002. I’ve patched the Swiki software only once or twice since Jeff graduated. Jeff did a great job designing Swiki, but I suspect that even he’d be surprised at how long they’ve run with essentially no maintenance.
What are the characteristics of educational technology that remains viable and usable (i.e., useful and actively used) with very little maintenance for well over a decade? Schools are under-resourced, as I talked about in the Thorndike vs Dewey blog post. It’s great to have educational software that just keeps going without maintenance. Maybe that it’s a certain class of software that works like this. Is it that JES and Swiki do so little, such that they’re really just frameworks on which to hang others’ content? Maybe that’s why they’ve been able to keep going for so long?
Your thoughts would be welcome.
Entry filed under: Uncategorized. Tags: computing education, JES, Media Computation, MediaComp.
1.
orcmid | March 2, 2020 at 12:44 pm
This, https://github.com/gatech-csl/jes/releases/tag/6.0 without the ending period, works better :).
I’ve encountered many “niche” software products that manage to stay alive, even despite the kinds of obsolescence that happens among major software producers as attention moves to the new shiny. I run a few of those.
In many cases, it seems that some software is amenable to the attention of hobbyists who are thereby in a position to maintain and support the products they love. It is fortunate that there is enough preservation of backward-compatibility to sustain the platform on which long-lasting code operates (e.g., Swiki and Squeak). This degree of stability and backwards compatibility allows adherents to continue playing their favorite near-ancient computer graphical games too.
I would add something about the MediaComp idea, not knowing whether it is emphasized or simply presented with students left to see beyond the tacit to the explicit or not. I am thinking about the means and varieties of digital representation that makes modern computers so powerful. Media manipulation and rendering is a great demonstration and something to experience.
For someone who looks under the covers, the stacking of abstractions via OS to Java to Python is another way that is evident in the case of JES and the view of stored-program power that is involved.
PS: It is fascinating to see the prices of the MediaComp book at this point.
2.
gmh | March 3, 2020 at 3:08 pm
I think one reason that Logo is still available and still in use has to do with the number of teachers that know it and the number of books and websites that provide interesting problems/programs. When I volunteered to teach high school students introductory programming at BFOIT (first two years with Brian Harvey) we used Logo. I ended up writing a Logo interpreter in Java that ran as an applet on the lesson web pages – read about a concept, experiment with it. But then the browsers dropped support for Java due to security issues.
In any case, the interpreter runs as an application and needed to be downloaded for medium and large programs. When BFOIT ran out of funding I got and maintain the website, still do a little bit. So I
look at the access and download statistics occasionally. I think Logo is a good language for transitioning from Scratch to textual programming.
3.
Media Computation today: Runestone, Snap!, Python 3, and a Teaspoon Language | Computing Education Research Blog | September 6, 2021 at 7:01 am
[…] it today. I did get a new version of JES (Jython Environment for Students) released in March 2020 (blog post here), but have rarely even started JES since […]