Brett,

My work on enhancing compiler error messages in Java has been published by 
*Computer


As a commercial compiler writer in a past life I know that error
messages rarely get much priority.  Any programmer who knows what they
are doing should be able to figure the first message out (close, by
line, cascading messages are often not worth looking at).  Student
problems don't count, unless you happen to be teaching them.

The kind of error messages generated by a compiler are almost as much
to do with compiler internals as the error in the code.  It is possible
to figure out a lot about how a compiler works by the kind of messages
it produces for specific coding errors (at least if you know about
compiler internals).

The only way you are going to get user oriented error messages is by
writing a front end designed for that purpose (which means ignoring
the common "how to write a compiler" advice because this is not
designed to produce meaningful messages).


Science Education. *I believe that as a 'recent article' it is free to
download (for the moment anyway) from
http://www.tandfonline.com/action/showAxaArticles?journalCode=ncse20.

I also have the accepted manuscript available on my personal webpage here:
http://www.brettbecker.com/becker2016effective/, along with a free link to
the paper if it isn't freely available at T&F, but I think as a 'recent'
article it is.

Soon I will have my 2015 thesis online which includes details on the
ground-work. I'll send another email when that is up.

I hope that some of you find this of interest. As I said in my earlier post
the software is at the end of its life in many ways but I have a PhD
student starting now who is going to revive/update/expand the project. I'll
let you know of any interesting developments.

Regards,

Brett

On Monday, 18 April 2016 09:38:16 UTC+1, Brett Becker wrote:


Hi Linda,

I have just completed a masters thesis on enhancing Java compiler error
messages. The thesis in its entirety isn't online yet - I'm waiting for a
few administrative hurdles to be jumped. What I do have that may help at
the moment are the following:


    1. I have attached a pdf containing about 120 references, mainly on
    novice programmer behavior and enhancing error messages, with a slight bias
    towards Java, but there are plenty of references on other languages also.
    Some of these took a long time to track down, and I only discovered some
    later than I would have hoped, so I hope this gives you a jump start. There
    are a number of references on software that enhances error messages and on
    designing good error messages. There are several PhD and MSc theses
    included that aren't always available on google scholar - some of them are
    hiding in fairly dark corners of the net. If you have trouble tracking
    anything in particular down let me know. I should have copies of all of
    these.
    2. I recently presented some of these results at SIGCSE:
    (http://dl.acm.org/citation.cfm?id=2844584&CFID=602145466&CFTOKEN=63845132
    
<http://dl.acm.org/citation.cfm?id=2844584&CFID=602145466&CFTOKEN=63845132>).
    I found that for the groups that participated in my study and using my
    software, enhancing errors did make a positive difference in terms of
    number of overall errors committed, the number of errors per student, for
    some error messages the number of errors per error message, as well as to
    some extent the number of repeated (consecutive) errors.

The software is now several years old and has reached its end of life in
too many ways. I have a new PhD student starting next semester however who
is going to hopefully expand upon this work, in addition to moving in a few
new directions.


I'll send you an email as new information becomes available. I am very
interested in this area, so please feel free to contact me at any time with
questions or updates!


Kind regards,


Brett



--
Derek M. Jones           Software analysis
tel: +44 (0)1252 520667  blog:shape-of-code.coding-guidelines.com

--
You received this message because you are subscribed to the Google Groups "PPIG 
Discuss" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to ppig-discuss+unsubscr...@googlegroups.com.
To post to this group, send an email to ppig-discuss@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

Reply via email to