What's the status of plans for hosting the NORMA project at Neumont? There was some mention of possibly moving to N.U. IT infrastructure; is this still being considered? If the NORMA project and the forums will stay on S.F. for some time, are there things that can be done (within the S.F. restrictions), to add functionality to the venue? having a place where users can upload files (sample models, text documents, PPT slide decks, etc..., would be very helpful - these can be on another site, but would be better under a single umbrella). If this isn't doable here, can extra forums be added - to provide granularity and specificity to thread subject matter? I realize that the current volume of forum activity doesn't warrant it, but it might be a kind of 'chicken and egg' thing, or 'Build it, and they will come' idea. In any event, should the project and forums be relocated at some point, better categorization of the notices, questions and comments here, will make for a better resource - where ever.
Related to this, what's the best way to report errors/omittions/typos found in NORMA and ORM 2 documents? BRN..
A most important subject and one I've been raising with Terry and hopefully, Matthew.
In terms of involving a wider community (for example including former Neumont students),
and to stimulate more discussion, it's essential that a new approach is taken. The forum
approach just doesn't have the immediacy of, say, email lists, for those who want that.
Most decent email list services offer the opportunity to read messages online only, or
in a daily digest, to receive every email, or to track discussions through an RSS feed.
An email list can also be submitted to Gmane.org, which aggregates Usenet as well and
archives every posting forever, so you aren't subject to the expiry of other providers.
No important project I'm aware of is happy using just the SourceForge facilities.
I have personal experience of creating and administering lists over a long period using
both Yahoo and Google, and it's surprisingly easy and resistant to spammers. Both also
offer areas for uploading files and images in addition to other features. Yahoo is more
mature, and is my preference. It identifies members even though they may have multiple
email addresses, which is important.
I will soon have a need for a discussion list for my own work on ORM, independent of
NORMA, and would be happy to share the administration of a combined list, rather than
fragment an already small community of modelers. As a result, I've also put quite a bit
of thought into naming such a list to minimise confusion and yet to cover the subject,
while making it easy for interested folk to recognise the name as relevant in searches.
The candidates I considered were:
* conceptual modeling
* object role modeling
* fact-based modeling
* information modeling
Conceptual modeling has wrong connotations for many people, including the idea of high
level but fluffy thinking. The next two are technical, but only directly meaningful to
those already in the know. Information modeling is most inclusive, though when combined
in a search, turns up information about some girl's modeling career... It's a bit of a
hard decision, but I still favour this last one. It doesn't exclude NIAM and can draw
in ER people.
I searched for existing relevant groups that have activity and seems well-managed, and
there really aren't many. Many groups cover DBA roles and data administration, but not
the design of database systems. So I don't think that joining an existing group is
I have reserved the name information_modeling on Yahoo (i.e. the group now exists). We
might wait for some discussion here before deciding to proceed with memberships however.
I've set the list to allow members to hide their email addresses and post only via the
web, but I haven't yet published it to the Yahoo groups directory. If people approve,
I'd hope that Matthew or someone he nominates at Neumont would be a co-moderator.
Perhaps you also, Brian? It's not an onerous task.
Hello, I am a student here at Neumont and worked on the NORMA project for several quarters updating the Reading Editor. Its great to read the contributions which cjheath and brian27 post to the current list (in its limited form) and your suggestions for a better way to provide information and build the community are completely valid.
I cannot speak for Matthew or Terry but when I entered the project they were just getting the SF repo setup and we did have discussions about beginning community site which could eventually be expanded with full knowledge base, lists, forums, etc.
I Pulled Matt aside just now and spoke with him about the direction he would like to take this in and what developments have been made recently. Matt desires a news-group which can be connected to via a newsreader (such as Outlook express or thnderbird etc) and not just thru the web. Matt also infomred me of a domain, www.ormfoundation.org but I am not privy to what exactly Kens plans are or what resources it will be able to provide at the moment.
Matt expressed his apologies for not being more involved in this list at the moment, due to the current development load on NORMA, but said that he will be more active shortly when they overcome the current changes required on the project.
Thanks taber. I also knew of the new web domain. It will serve well
for promotion, but isn't really a place to host a mailing list.
Not many folk even know what a Usenet group is these days, though
I'm still an avid user after 22 years. The reason I brought up Gmane.org
in my previous message is that they provide a Usenet feed for any ML
they subscribe to, as well as all the non-pron newsgroups, which they
rename into a sensible hierarchy rather than the historical one. If we
get them to carry the Yahoo information_modelling ML, they'll probably
call it comp.information.modeling, which would suit just fine. They also
provide all their groups as RSS feeds.
If you can grab Matthew again and get his nod, I'll proceed to publicise
the ML and we can get started without him - I'll grant him moderator
rights as soon as he joins, if he wishes... or someone else.
Last week I passed along a link to Ken Evens, and in a reply he acknowledged that he is swamped with his degree work. I think the website is important to him, and it's just a matter of finding the time to work on it. He mentioned wanting to add a forum and a repository for ORM models (in .pdf format), where users can upload models for review, etc.... Currently, the website is sparse, but there is some material there - once you register. You can always email him (through the website). I get the feeling that he has a pretty firm idea of what he'd like for ORM Foundation, so I suggested waiting until he has time to get the ball rolling before getting involved. Once he gets it on track, I'm sure he'd appreciate help keeping things moving. I've been involved in a few organizations where there were too many cooks. I try to remember that here; ask questions, make suggestions, but let the people that have done the work set the agenda.
BTW, how would you like to be addressed? TL? BRN..
Thanks for concurring on the usefulness of of improving the community infrastructure for NORMA/ORM 2. If you'll give me more info on the combined community support model you have in mind, I'd like to help out in some capacity.
As you pointed out, accommodating (rather than segmenting), the currently small active community is important. My hunch is that finding (or creating), a good umbrella organization for community support is important too. Stability is one concern. Where the infrastructure is dependent on one or a few people, personal circumstances can lead to orphaned initiatives. So, is it better to create a new support structure, or augment an existing one?
It would be helpful to get some feedback on these ideas, especially from T.H. and others at the center of ORM and NORMA initiatives. I think we all agree that a better system of on-topic communication is required. Pragmatically, I'd favor options that can be folded into, or shared by, other organizational structures - to reduce the risk of dead ends, and good input going to waste.
Given what I just said, going forward with some type of newsgroup/mailing list sounds like a safe bet. It's not something I have experience with, so you know better than I what to look for, and look out for.
On the naming issue: it sounds like you have given it a lot of thought, and I don't know what factors make for successful newsgroup/mailing list names. I don't think of ORM as Information Modeling ("Fact based, Conceptual Data Modeling (where objects play roles), to support Information Systems" might be a better description; but it's not likely have the same cache ). "Information Modeling" should be close enough, and broad enough to serve the purpose - I kinda recall a product called InfoModeler; and if marketing concerns won out over semantic exactness in that case, why shouldn't it here? Well done on reserving the name. Do you have any ideas for a good way to redirect those that get there by mistake? (With a broad cover like Information Modeling, that's going to happen - a lot. Yet, you'll want to make it known that people looking for UML, O/R-M, ER, etc..., might find a better alternative, or additional support in ORM!) BRN..
> finding a good umbrella organization for community support is important
Well, a community that provides support is its own organisation, but if
any other organisation were required, it must be the ormfoundation being
established now. However I expect that the NORMA team will support NORMA
users there, and ORM Foundation will make conference announcements, and
a community of practitioners will help each other out with problems. The
only discussion that might be a bit out of place is that among developers
of ORM software - chat about ORM meta-models and the intricacies of the
ER mapping algorithms won't be to everyone's taste. But if that volume
grows too loud, it'll branch off to a second forum. Right now I'd welcome
just one functioning forum with a few more active contributors.
The role of moderator of such a ML is mainly to approve new members' first
postings, or to delete and ban them if they're spammers. A list I run that
has 600 members gets less than one new member's first post a week, and that
is covered by one of four moderators - so it's really not hard. A couple of
browser clicks a month is all.
> Do you have any ideas for a good way to redirect those that get there by mistake
When you take a look at my introductory paragraph I think you'll agree it's
covered: <http://groups.yahoo.com/group/information_modeling>. We can add
to this if needed, for example to indicate that ORM can be used as a valuable
precursor to UML (as well as ER), should your employer require UML :-). But
perhaps we shouldn't do that yet when there's no tool support for it.
Just signed into new Yahoo group and posted brief message there. BTW, on the home page I noticed there were some targeted sponsor links; including this one:
modeling get info - Become a Star today, Join FREE, and get Discovered by an Agent.
Gee, I don't know if I'm ready to be an Object Role Modeling star... just yet.
When you're ready, I suggest posting an Open Discussion Forum announcement message here, to let others know about the group/mailing list. Also, a note to K.E. about exchanging links, would be in order. I'd also suggest adding a link for the group to some Wikipedia entries.
When you have a chance, fill us in on what the file exchange policies for the group will be (maybe make that part of the announcement).
If you want me to screen some spam, fine. It will be good to have a few people in various timezones taking care of this chore. BRN..
P.S. SourceForge was down for about an hour, delaying posts to these forums.
I've written to both Ken and Terry today. Matthew must be very busy, or
he would have commented already. I don't have his personal email address
to send him a hurry-up. Hopefully we can get agreement to make a joint
In my experience with other groups, there's not much spamming activity
in relation to uploaded files or photos, though the posted links do get
abused. I've disabled the links section except for moderators, but I
just re-enabled file upload by members. We'll still have to keep an eye
out for uploaded spam. I also disabled the database and polls features.
Disabling polls was necessary to allow folk to hide their email address.
It's not possible to allow files but disallow photos, or I would do that.
I guess an image of an ORM diagram is a photo though, so the feature will
get used. It's preferable if each person who posts stuff creates their
own folder for it, otherwise it gets pretty disorganized.
The list does allow message attachments, so that's another way of posting
There's no way we can avoid dumb computing machinery associating us with
fashion models, but it is humorous. So much for the semantic web ;-).
Took a look at Yahoo Group policies. 100 MB total, and 5 MB max for individual size files. I used a typical NORMA/VS ORM diagram page, and it takes less than 20 KB, as a PNG image (Looks like the best format: clear yet small file size). There is the camera icon option to copy an ORM page image from the tool (I guess it's just one page, haven't tried it with a multi-page ORM diagram yet), but I don't know if there's a way to auto save the image from VS - needed to paste it into an image editing app, and save as .png file. [BTW, do you know if PNG files can be used as vectors for malware, the way JPEG files are?] At 20 KB/page, that would be a max of 250 page ORM diagram in a single upload. Lord help us if that's not enough! NORMA/VS .orm files are about as small, I didn't see restrictions posted in the Yahoo policies, so I guess it would allow posting these. I don't know if they'd balk at uploading .sql scripts, but the XML based supporting files shouldn't be a problem. It's probably what you think would be most useful. You'll certainly want a caveat for downloading files (it's common sense, but better to be make it explicit). I'd at least want to know who posted the files.
Good luck with the I dotting and T crossing. BRN..
As many of you probably know, there is now a Yahoo group for discussion
of information modeling in general. You need to register with Yahoo (create
a Yahoo ID), and associate it with a valid email address (though you can keep
that completely private), and then you can receive and contribute messages to
the group via email or online. It's preferable if NORMA-related stuff stays here
where the team will read it, of course (though some are members there also).
The new group is at
As the traffic grows, I'll request it to be carried also by gmane.org, which will
provide an NNTP feed and caching for the list. Yahoo already provides an RSS
feed if you use that.
Get involved! Lets get some discussion going about some of the new things
we'd like ORM2 to do: aspects, security modeling, automatic injection of
surrogate keys, etc, are on my list... What about UML class diagram generation?
You can make more suggestions I'm sure!
Folk, as promised, the Yahoo information_modeling mailing list is now also
available as a newsgroup through the two-way gmane gateway. This gateway
allows anyone to contribute once their email address has been verified by
gmane (the email address will be obfuscated in posts to defeat spam address
If you regularly use a newsreader (NNTP) or an NNTP-capable mailer, for
example Thunderbird, you can easily set up the news server "news.gmane.org"
as an NNTP service, and subscribe to the newsgroup called
I've just posted an update on progress with my ActiveFacts project there, if
anyone's interested. It's not yet doing anything useful, but you might find
some of the information interesting and have comments on my intended
In particular, I'd like to see any ideas or suggestions anyone has about how
best to structure a textual language for ORM2. As I see it, ORM doesn't lend
itself to nested structures as many modeling and programming languages
do, rather ORM2 is a collection isolated statements that together form a model.
The difficulty that raises with a textual language is that there is no implicit
block context, so all objects must be uniquely named and referenced, which
can be quite onerous and awkward. For example, often an objectified fact has
the same name as the fact; how can the textual language then tell them apart?
I've posted this last paragraph and more as the head of a new discussion topic
in the Yahoo list - please comment there!