I have little to add to Jonathan Steele's commentary on the launch of everycasualty and the issues it raises. Here are a few quick notes:
The dedication and courage of Sandra Orlovic and Bekim Blakaj, Deputy Director and Director of the Humanitarian Law Centre in, respectively, Belgrade and Pristina, and their colleagues is magnificent. There is great nobility in projects such as The Kosovo Memory Book 1998. Both Orlovic and Blakaj emphasized the importance, in the face of considerable opposition, of recording and describing in some detail the lives of all who died in the violence, military and civilian on both sides. By way of reminder that this in itself is not enough, Blakaj noted that 12 years after the end of the war there had been only 12 successful prosecutions for war crimes. No justice, however, was possible without an honest account of what actually happened.
Wissam Tarif of INSAN expanded on this last point. Those documenting the identity of individuals murdered or abducted in Syria and elsewhere were sometimes accused of opening tombs and opening wounds. But that was precisely the opposite of what they were doing. Tombs and wounds could never be closed without a full accounting for what actually happened. In his own country, Lebanon, people were not fighting at present but there was no peace, only a ceasefire. This was because the Lebanese had to failed to acknowledge facts, to recognize the humanity of all those who were killed and to face their families.
According to the 2011 World Development Report, around 1.5bn people today live under the shadow of organized violence. Much of this violence is criminal. One of the questions at the launch was: should innocent victims of crime and criminals who were themselves killed also be counted by projects such as everycasualty? One of the challenges in the 21st century, it was argued, is that while war between nations and even 'formal' civil wars are actually less frequent than before, large-scale, inchoate criminalized violence is on a greater scale than ever. This presents a challenge to existing institutional arrangements: agencies such as the Red Cross, for example, cannot act in Mexico even though the scale of the violence (recent small but typical example here) resembles war because the government does not recognize a state of war. 
Dan Smith of International Alert said that by making it possible to know who had died in a conflict and how, the charter had the potential to reduce the traction of wild claims (up or down), which were the meat and drink of propaganda. The charter could help us respect the 'fact of war', a continuing reality which is too often hidden behind cliches and euphemisms.
everycasualty, said Smith, was a civilising idea. Like all great ideas it was obvious once stated, but it also subtly challenged the norm. Also, there was something slightly obsessive, unrealistic about it. In this, it shared much with the ideals of the Red Cross at its foundation -- a 'wildly unrealistic' idea at the time of its inception, which acted on nothing but moral authority.
I think this is right. The everycasualty charter challenges the disturbingly plausible observation, misattributed, perhaps, to Joseph Stalin, that one death is a tragedy but a million is a statistic. It aims to make visible and irrefutable the tragedy of the violent death of each individual, including the deaths of those who are themselves killers.
Wissam Tarif told those present at the launch that the previous evening he had talked by telephone to one the volunteers on his team in Syria. The volunteer had said that in the midst of conflict people are completely focussed on what is happening right now. But others not caught up in the conflict -- such as those gathered together peacefully in London -- had the opportunity to think about the future. This was a tremendous gift.
everycasualty is an idea big enough for a version of the 21st century in which there is hope. It will not of course end tragedy. Ideals are frequently subverted (it is reported, for example, that death squads in Syria are using Red Cross/Red Crescent ambulances to abduct protestors). And even a full accounting need not guarantee reconciliation. But it is a start.
A couple of other points: in June the Oxford Research Group published a working paper on The Legal Obligation to Record Civilian Casualties of Armed Conflict. And, drawing on the model of Iraq Body Count, there is now a Pakistan Body Count.
Note  My language and legal understanding here are shaky.
P.S. 24 Sep: A blog post by Dan Smith, who was on the panel at the launch