Sam, (and the others who have offered comments)
Thanks for writing and I will try to address each of your comments. And
no, I was not referring to you as the one that flamed me, it was another
individual. I rather enjoyed your response to the article on UGA
because I felt that was precisely the type of dialogue that needs to
accompany such initiatives. I was surprised at the tone of Dr.
Tarleton's response, but I agree that any article sent out from what we
call the "Banana Herald" down here, needs to be read with the
understanding that it is probably written by an underpaid kid who would
miss key details.
I also want to say that Canadian Entities do far more than other
countries at promoting development information.
And now to your specific comments:
> First, XML is being developed by others - not Bellanet - as
> the markeup language that follows HTML.
I never assumed XML to be part of the Bellanet Initiative, but that DML
was clearly the initiative.
> The DML proposal from Bellanet is just to promote dialogue
> around where the development community might want to go
> within the XML initiative which, as I understand it, is
> taking place at the W3 level. There is no secret work and no
> secret agenda. I will make sure the people involved get your
> comments and they will no doubt wish to reply - hopefully to
> clear the air and remove misconceptions.
I have reread the documents on the bellanet site and cannot agree with
this interpretation. Again I will quote from Mr. Davies' report: " the
field is wide open for Bellanet to take the initiative and to spearhead
a standards development activity". I use the word 'secret' because in
the time that I been have subscribed to the Dev-L mailing list, this is
the first inkling about the project I have seen or heard, yet it is
clear that a great deal of effort has gone into this. It is probably my
fault for not visiting the bellanet site more often and seeing the
information there. By the same token, I wonder how else people may have
found out about it. If promotion of dialogues was the case, word should
have been sent around long ago.
> I hope that we can sort these out and deal with what is the
> only real issue on the table.
I am always open to discussion. I responded to the message you sent
rather than the one on the Afrik-IT list because yours had much more
> ...Is there a useful place for an DML initiative within the
> larger XML initiative?...
That is the very first question that really should be asked. Is there a
useful place? I am sure there is because everything has a proper use in
a particular context. My concerns are the chicken and the egg approach:
should we let the standards drive the technology or should the
technology drive the standards. And what is the lowest common
denominator technology in the development community? And how do we not
cut out groups from the South who may not have the skills or equipment
to implement DML (whether it be implemented in Oracle, Notes, or other),
when many groups do not know how to use HTML yet or consistently abuse
But before I get to that, I think I need to be clear that any initial
discussions should focus on "metadata" and information content
standards. The development community needs to put its heads together
and decide just what information should be consistently delivered by
development organizations through the web and other electronic sources.
Is there core data that every organization should provide and everything
else is gravy? Can all parties come to an agreement on that? Will
there be different levels of information needed depending on the type of
organization? What about political considerations? How will the
information be used? What will its shelf life be (do we follow a 20
year rule or some subset of that)?
I will say that no one group or organization (not the SDGateway,
OneWorld, IDN, or IDRC) has all the answers or will be able to come up
with answers that will serve every case. And we all know that
organizations don't always want to share information.
It is my experience in sitting on a number of web standard development
(including the odious metadata) work groups, that parties have their own
biases and that nothing gets resolved unless some "higher power" steps
to the plate and says "this is the way it is going to be". And who's
place would that be? Bellanet/IDRC? The World Bank? CIDA? USAID?
Back to the standards and technology issue...I work with a number of
technologies with varying degrees of success. When I build web sites, I
build to the lowest common denominator, which in some cases is access by
a text-only browser (scary isn't it that a US Entity would still use
this technology?) Many times I have seen standards developed which
absolutely drive the technology to be used (in an overt attempt by
management to get the tools they want). Oracle can be a nightmare, but
that may be the only choice when you are forced to use a unix server
that doesn't support any other extensions such as NT & Front Page or
Cold Fusion (I manage 2 sites that fit this bill).
And then you have the case where you give the kids a new toy (such as
no value, and cause browser crashes. Some examples of development
organizations that do this (some in combination with bad color sense):
the OAS, UNDP, UNECA.
Of course you have the organizations who make decisions that will
absolutely cut out entire classes of people. You know, like the ones
who use frame sets with no alternative available so all visitors with
non-frames capable browsers see is a blank screen. The worst offender:
USAID with their new "Technical Consultant Services Resume Database"
which only be used with Netscape Navigator and Communicator non-beta
versions 4.0 through 4.0.7, Microsoft Internet Explorer non-beta
versions 4.0 and 4.01, and Opera version 3.21. Got 3.0 or 4.5? Forget
about it. And then after the fact (many weeks after posting the site)
do they tell people the tool is only really meant for use of a
particular program office.
So we go out and develop DML and some groups use it, some well, some
badly, some not at all. Then what have we accomplished? What have we
done to those people/still running windows 3.1 on donated computers?
What about groups who serve more than one master (I know of many that do
work for the development community, the environment, and the scientific
community). Are they going to have to know 20 different flavors of XML
(and does this not get away from the notion of "Open Systems")?.
You need to understand that the IDN was formed because of the poor job
most development organizations do telling their stories, sharing their
successes and learning from their failures. I know full well that the
next step in our evolution is moving to a database driven system and I
have explored a number of options, which led me to start to break down
my resistance to the product and strongly endorse a Notes based solution
to development information. Why?
A number of development organizations are using Notes either internally
or externally: CIDA, the World Bank, USAID, ICRC, World Vision, the
SANREM CRSP, the Caribbean Development Bank, and the Education
Development Center, among others.
The tool is strong and provides the following combination of features
that no one else can touch:
Multiple levels of access and security down to the field level.
Full replication between servers
Full support of HTML (and eventually XML)
Two way pumping of data between Notes and relational databases.
It may not be the tool for everybody (and we can all probably agree that
we agree on this), but I really believe that it is the best tool to
reach the objective of common data standards before XML even gets here
and without having to deal with XML. I like to point my cart at what
is, not what MIGHT come to be.
I have a scenario on how to implement this, but I won't go there right
So this is where I am coming from and part of the reason I have very
strong feelings about this initiative.
Lets keep the discussion open.
p.s. to Kerry Miller: I try to avoid high horses, they might keep the
New York brass from clanging:)