XML/DTD

Revision as of 00:33, 16 April 2003 by 142.177.77.114 (talk) (general person model that doesn't dig too deep into privacy)

The DTDs are @ http://www.consumerium.org/XML/DTD/filename.dtd

I know, we should set up an CVS, this is a mess and not good browsing/editing at the moment. Please bear with us. If you want to contribute new DTDs please email them to [1]

We should propably set up pages for discussing each DTD, it's requirements and what needs it serves, with references to Features.

Functioning


Long-term important, please review

places and times

  • m:spacetime_DTD (This is necessary for automatic relation mapping and borders of )

Spacetime is critical to match suppliers and buyers of non-offensive goods in space and time, without which, buyers may have no choice but to buy offensive goods. It is critical also to identifying borders of natural ecoregions as they exist or as they are degraded over time. This is useful to graphically show the impacts e.g. of w:deforestation.

Ecoregion is critical since an export of exactly the same consumer good from one ecological region may be harmless or even constructive, while an export of the same good from another may be devastating, e.g. exporting water from arid regions versus those with icebergs, e.g. exporting pulp wood from tropical forests versus managed temperate tree farms, e.g. cattle living on former deforested rainforest land versus natural pasture.

persons

This type of control-freakish markup is shunned by many wikipedians for understandable reasons, where as disinfopedians and some NGOs like it.
As for what I think is that this is way overkill for our purposes. Gender, Birthyear, Nationality, Employer, Profession and perhaps some contact info should be quite adequate for employees, campaigners and commentators. Anonymous Cowards can of course be dismissed as nobodys more easily then someone who has GnuPG or commercially signed identities. Consumers naturally can reveal as little or much about themselves as they wish. --Juxo 23:21 Apr 15, 2003 (EEST)
However much or how little you record about a person, it would be useful to make sure that the tag-names don't conflict with more general uses. For instance Nationality can mean Birthplace, it can mean Citizenship, it can mean Ethnicity. So Nationality is a bad tag name. Employer? Some people have many, or none. Profession? Not standard across all countries, better to record degrees and certifications, say as Credential. And all can be multiple. And how to remain Anonymous while still revealing credentials, so called Blind Credential is very important.

Identifying individual people unambiguously is more controversial, a two-edged sword. Some activists like Amnesty and Greenpeace do it often, for what they consider clear moral purposes. Other projects like Wikipedia have less clear purposes and thus less excuse for "outing" or "shaming". This is not an easy issue for consumerium:

On the one hand certain individuals like person X and person Y are engaged in a vast number of unpopular activities, and many would choose to have nothing to do with an enterprise that they profited from. Being sure that the activity is associated with them is impossible without some clear way of identifying individuals and their interests - there is danger that innocents will be targetted if identification is not clear. On the other hand, there is reason to fear "witchhunts" and "dossiers" on ordinary individuals, and automation of person-tracking and identification certainly makes this easier.

However, some means of standardizing references to persons seems to be underway in some projects, e.g. wikipedia, anyway. Consumerium may not be able to igore them. There are also reasons to identify "noble" individuals for merit or recognition, e.g. disinfopedia's naming of reliable sources e.g. James Speth. This can be important to validate information to be used in buying decisions, e.g. as the 'authority' in TIPAESA structures.

coordination with non-XML name spaces

If these DTDs are not standard with large scale data sources like wikipedia and all other projects using the GNU GFDL, they will not be used to maximum extent, and the information available will be limited or flawed. Thus early standardization of these tags is extremely important to consumerium, even if it finds some of them problematic politically.

Also XML tags may not be the only or best place to express some standards, e.g. ecoregion DNS entries or the dictionary also express name-spaces for public use, and may be more appropriate for nouns or verbs useful more generally than consumerium.


Not functioning, under construction

  • XML/DTD/product_group.dtd I've done some work on product_group.dtd, but it's not working. I'm thinking that product_group-nodes could be arranged to multiple trees (production tree, product substitution tree) and directed networks to reflect the different roles of a product_group in different contexts.