Our fellow Edgeryder @Inge is in Tibilisi and today reported live from the major flood that hit the city. Several people died (8 confirmed atm) and a lot of animals. I can’t help with the clean-up efforts on the ground 
but I’d like to share best practice recommendations for community-driven disaster response. We have observed these over the last 6 weeks of the Future Makers Nepal project, an initiative together with UNDP Nepal which so far researches community-driven disaster response after the major earthquake of 25 April 2015. So, dear @Inge, @Khatuna, @Nino_Japiashvili, @Nick_Davitashvili and whoever else is in Tbilisi right now: I hope it offers some inspirations, and wish you much strength wherever you will contribute to the relief efforts.
- Set up a crisismapping platform. Here in Nepal, an organization called Kathmandu Living Labs had quakemap.org up and running within a few hours. This is possible because they used the open source Ushahidi Crowdmap software for this (and I think it;s the best choice). If you need a server or help setting this up, we can put it on our edgeryders.eu server. **Edit:** In 20117 Ushahidi Crowdmap is no longer the first choice in 2017. I recommend OSM uMap now – [details here](https://edgeryders.eu/t/id/6704).
- Set up a hotline phone number. This is something that was not done here in Nepal, but from our experience with how volunteers coordinate, this would be a great tool. Because volunteers "in the field" can't be bothered to do structured data input or query a database, so other volunteers should take on that task and provide a simple voice interface to the others.
- Just do the work! Zero-overhead action is the beauty of volunteer-driven disaster relief :) It is also the main point how it differs from the long-term approaches by big NGOs and disaster relief professionals. Volunteers save much more by not having meetings and plannings than they waste by doing the wrong work at times. Volunteers also do not wait for the right tools, they just use what they have at hand. Solutions will pop up while the work is ongoing, but probably not before. A great example, found by @Dipti_Sherchan, is something Milan Rai said: "We only had two gardening tools but our aim was to build 100 toilets and none of us had even broken a stick!".
- Volunteer coordination efforts should agree on role separation. In some cases, this happens organically (here in Nepal, there would be one Facebook group per district for example). In some cases, it needs a decision. But don't call a meeting to decide it, just do it in one phone call.
- Pull-based model of information flow. What worked well here is that responsibilities were organized (informally) in such a way that the person wanting to do some new relief work would have to call or otherwise contact other relevant groups working in the same topic or geographical areas, in order to avoid duplication of efforts, and in order to collaborate around logistics etc. where meaningful. Initiatives do not have a responsibility to publish or report back what they do. (The hotline operators however would want to call them and collect that information to publish it online.)
- Needs assessment: don't underestimate it. We had some cases where villages received relief supplies that were "nice to have" but not needed there. This does not exactly harm anyone, but at least in the early days after a disaster there might be high-priority tasks to do instead. Good assessment options for volunteer-driven disaster responses are: (1) asking a trustable local contact person from the affected area, (2) going to the area and doing a quick damage and need assessment (go around, pin down names and what they need … don't start immediately with distributing stuff, it should reach those who need it most).
- A new role for funders! NGOs and international organizations looking for "partners on the ground" to do the relief work have a new, uber-efficient option, now that volunteer-driven disaster response is widespread: contributing to the small-scale crowdfunding campaigns of these local initiatives. If they can show documentation from what they started by themselves, it will often be enough to gauge how much money to entrust them with. Collecting these fundraising campaigns and evaluating such "trustability stages of collaboration" for such initiatives and communicating them to potential funders could be one more very useful volunteer activity that I have not yet seen. Funders for the local initatives can also be "normal" people from the same and other countries who want to help, but want to give directly because they don't like the overhead costs and hard-to-understand scale of the "traditional" disaster responders. We've seen such donations happen a lot in Nepal. Also think of the "diaspora" community as potential funders; Nepal has a big one, Georgia has also one as far as I know. In the case of Nepal for example, the HELP Nepal Network collected >500,000 USD in donations, mostly from the diaspora community. That was however the largest volunteer-driven fundraiser that I know of.