home > plop > 2006 > Papers > Notes > 7-notes.php

PLoP 2006 Shepherded Papers

Notes on Paper 7: ACCEPTED

I want to make sure, I vote in favour of including Philipp Bachmann's
Static and Metaprogramming Patterns and Static Frameworks
paper into a Writer's Workshop at PLoP.

You should already have the latest version.

Yours
Peter.

Prof. Peter Sommerlad
Institut für Software: Bessere Software - Einfach, Schneller!
HSR Hochschule für Technik Rapperswil
Oberseestr 10, Postfach 1475, CH-8640 Rapperswil

http://ifsoftware.ch


Kyle's summary:

I'm not going to be as easy on this one as Peter was. I found the paper to be very focused on C++ (I couldn't even imagine how I'd make these patterns work in Java or Smalltalk) and thus limited in its applicability due to that because of its dependence on C++ semantics. I also found the variations on the GOF patterns to be minimal at best and, once again, very dependent on specific features of C++ to be useful. Given the limited space at PLoP this year I can't recommend this one for inclusion based on these issues.



Paper 019 - Welicki - The Blind Cache Antipattern

Paul's vote:

Hello Kyle,

Even if you haven't been paying any attention to my email exchanges
with Leon, you must have noticed that we have spent a lot of time
discussing his pattern and that he was working very diligently on
subsequent revisions. Let me summarize the shepherding process for
you. I was assigned to his paper late. Initially the paper was an
antipattern that complained about the misuse of the Caching pattern.
After initial discussion, Leon and I isolated two patterns that were
hidden underneath - design for cache data editing and caching of
config parameters. Leon has focused on the second one and we went
through 4 or 5 iterations since. The last iteration is not complete
and we're likely to continue for some time more, but all the main
elements of the pattern are in place.

Although the technical merits of his pattern may not be so great yet
(it still needs work), Leon's paper is significantly better than most
of this year's submissions. I've skimmed all the papers that were
submitted this year, because I'm in charge of the PLoP submission
system.

In summary, I think that Leon has gone a long way to improve his
pattern. He has been receptive to my comments, but he also made many
changes on his own. I believe that both he and his pattern will
benefit greatly from participating in the writers' workshop and I
recommend that his paper be accepted to PLoP 2006.

I will ask Leon to upload a new version of his paper to the PLoP
submission system tomorrow morning.

Paul

Kyle's summary:

This one was rewritten from an antipattern into a pattern, and benefited from it. It's actually a specialization of the caching pattern from the Party of Five book, but it's a good and very common specialization. This is one approach I've seen a lot, and the exposition of it is clear, with simple code that could be implemented in nearly any language. Good references to other work, and good use of the pattern form. I'd recommend we accept this one for workshopping.


Kyle Brown