Dear petrinet-owner, dear petrinet list members,
I had hoped that the list moderators would step into the discussion, to
clarify which changes are possible within the current framework, and
which would have to involve unwelcome departures. This message is
copied explicitly to petrinet-owner, with a summary of the changes that
have been mentioned recently, and some oddities that I noticed while
browsing the "Petri Nets World" pages:
http://www.informatik.uni-hamburg.de/TGI/PetriNets/
Please see this as attempts at constructive criticism, listing changes
that I hope would improve the accessibility of Petri nets, widening
access to more users who could profit from them. Since there seems
to be some agreement among list readers, I'd be interested to hear
the opinions of the Petri Nets World maintainers and PetriNets list
moderators.
1. Several list members have contacted me about that forum idea, so
I should clarify:
- I would much rather see this mailing list revived for discussions
than see a separate general forum opened, with overlapping topics
- announcements and discussions should go to separate lists, if only
to make archival search realistic. To my surprise, the mailing list
FAQ does mention two such lists already, but I can't seem to find
the PetriNets-discuss list (nor does Hamburg's mailman know
about it):
http://www.informatik.uni-hamburg.de/TGI/PetriNets/mailing-lists/faq.html
According to the FAQ, the PetriNets-discuss list is exactly the
list I would like to see in addition to PetriNets. The discussion
list should be free of announcements, and white-listed, so that
only the very first posting by a list member gets delayed waiting
for moderator approval, providing it isn't excessive in length, such
as this one (allowing to keep out spam without slowing discussions
to a halt).
Could PetriNets-discuss please be revived, or could someone point
me to it if it still exists, or the reasons why it doesn't?
2. While I wouldn't want to be forced to use a forum for long-term
interests, forums do have their place for short-term interests,
such as school teachers looking for contacts in the few weeks
of the school year when they might look into Petri nets, or new
Petri net users looking for introductory material and tool
recommendations in the few weeks before they switch to a
specific Petri net tool's user mailing list. What makes them
attractive for short-term visitors is their low barrier to entry, as
well as the ability to get an overview of current hot topics without
having to register anywhere.
However, that can only work if
(a) something is done to compensate for the short-term memory
of forums, eg, by adding a wiki on which useful assets and
insights can be stored, by the users, for the long term
(b) the short-term forums are connected somehow to where
the Petri netters with longer-term interests hang out, eg, the
general discussion list - I'm not sure how best to facilitate this
Probably, both the wiki and any forums should be located on the
Petri Net World pages, or at least linked prominently from there.
Each forum should ideally be moderated by a member, eg. a school
teacher or teacher trainer for a Petri-nets-in-schools forum, a systems
biologist for a Petri-nets-in-systems-biology forum, a workflow designer
for a Petri-nets-as-workflow-systems forum, etc.
3. Something is wrong with the searchability of the Petri Nets World,
e.g. searching for mailing list postings, introductions, or tutorials, via
Google, seems to give very few hits on that site.
Try, for instance, the "Search recent discussions on Petri nets" link at the
bottom of http://www.informatik.uni-hamburg.de/TGI/PetriNets/ , for
rather dire results. It appears that Google doesn't know about this list,
nor is does it seem to be registered with any of the usual email-to-news
gateways, such as gmane.org: http://gmane.org/find.php?list=petri
Registering Petri net mailing lists with Gmane would make them more
widely accessible, and in a variety of formats (news, web interface,
rss feeds, ..): http://gmane.org/about.php
4. The organisation of the Petri Nets World appears mostly unchanged
from when I last looked at it (then at Aarhus), which is not necessarily
a good thing. PetriNets-discuss doesn't seem to exist anymore, and the
mailing list archive for PetriNets ought to point at the mailman list info
page https://mailhost.informatik.uni-hamburg.de/mailman/listinfo/petrinet
where the archives are rather more uptodate than in the search-based
archive
http://www.informatik.uni-hamburg.de/TGI/PetriNets/mailing-lists/archive.html
Overall, the pages appear to be directed at academic researchers,
with events/bibliographies/publications playing a prominent role. This
might be somewhat intimidating/off-putting to potential Petri net users
happening on those pages. They would be looking for introductions
(but online, not in the form of expensive/out-of-print books; also, the
interactive tutorials could be emphasized more?), tool advice
(but up-to-date, with helpful introductions relating tools to application
areas,
rather than a plain database listing net classes), and discussion groups
(what happened to discussion here, anyway?).
One way to change that in a gradual fashion would be to augment the
existing framework with a wiki, on which users could add the resources
they find useful. Perhaps a user-directed and -moderated forum where
topical groups could arise on demand, and with a lower barrier to entry
than this list.
Another useful item would be separate RSS feeds for the bibliography,
events, jobs, tools, wiki, etc. Currently, there seems to be one well-hidden
RSS feed for the whole, and it has devolved to a mere events/jobs feed.
There are also entries for Petri nets in wikipedia and scholarpedia,
which manage to convey more of the graphical nature of Petri nets
without having to resort to Flash or Java (although they are too
theory-biased to be of immediate use to practitioners looking for
introductory information on which to base decisions, they could be
linked more prominently from the Petri Nets World pages):
http://www.scholarpedia.org/article/Petri_net
http://en.wikipedia.org/wiki/Petri_net
5. *Unpublications* unavailable online, and out-of-print books
A plea to all Petri net paper/tutorial authors: please do not hide your
contributions in expensive paper archives! All major publishers, last
time I checked (including Springer LNCS and ACM), allow authors
to host a copy of their papers on their home page, for timely dissemination.
Yet, how many entries in the Petri Net bibliography contain links
to online versions? If you don't make use of this option, your papers
are simply taken out of public view!
Please do not assume that because, say, German university libraries
have ordered LNCS as a matter of course, that this lucky situation
applies to everyone who might want to use Petri nets. In two UK
university libraries I sampled, there was hardly any Petri net material
to be found (Springer is expensive, books are bought to match current
interests), and not every company/school can afford to build up an
extensive research library, just to get someone up to speed.
Btw, is there any chance that the copyright of out-of-print books
like Reisig's standard introductions might revert back to the author,
who might then make them available online?
6. Other good suggestions have been made by Paula Mangas:
2. Online interactive tools to help understanding concepts;
Some of the tools are there, eg
http://www.informatik.uni-hamburg.de/TGI/PetriNets/introductions/aalst/
http://www.informatik.uni-hamburg.de/TGI/PetriNets/tools/java/
but they would need to be part of an integrated presentation,
where explanations of concepts are side-by-side with tools that
allow to play with those concepts.
3. Challenges and Contests, with apropriate material to self-development;
4. Good e-learning classes, with good pratical examples;
5. Games, where players must know what they are doing (using the theory).
7. And amplified by Michal Zarnay:
|From my experience with the few tens of students that I had on this
|subject so far I can say that they often prefer:
|
|* studying individually - not using available lectures and practical
|lessons in their full potential
|
|* reading easy-to-understand texts - not scientific papers or theoretical
|books with definitions and proofs (since we are application oriented)
|
|* using online resources to paper books - even if I offer them a paper
|book in maternal language, they rather go to simpler articles on Wikipedia
|in English.
|
|In this context the suggestion of Paula seems to be relevant - for our
|university students at least. Maybe such resources exist out there on the
|web of many universities. Is there a central place able to inform about
|them?
All of these suggestions demonstrate a pragmatic learner-oriented view
(someone has to learn enough about Petri nets, tools and practical aspects
to decide whether or not nets will be of use in their application area - how
do they do that, and what resources do they need?).
Again, if the Petri Nets World had a wiki, we could just ask readers
here to contribute their teaching material, something that is unlikely
to happen continuosly with a centrally administered site.
Claus
PS. As a reference point, have a look at how the online world of
Go (the ancient board game) makes use of graphics, wikis,
interactive introductions, online references, etc.
Start on wikipedia http://en.wikipedia.org/wiki/Go_(game)
then look at "Sensei's library" and "The Interactive Way To Go"
at the end. Shouldn't Petri nets deserve at least as good an
online presentation?-) Oh, and if kids can understand and play
Go, they can certainly understand and use Petri nets, even if
formal evidence is lacking at the moment.
----
[[ Petri Nets World: ]]
[[ http://www.informatik.uni-hamburg.de/TGI/PetriNets/ ]]
[[ Mailing list FAQ: ]]
[[ http://www.informatik.uni-hamburg.de/TGI/PetriNets/pnml/faq.html ]]
[[ Post messages/summary of replies: ]]
[[ [email protected] ]]
