Open Microscopy Environment

Dear All

this month our OME related software development project got off the ground.

The development team is spread all over Switzerland and as the project lead I certainly could use some advice from you guys on what you found to be useful and what you found to be disruptive in keeping the team working together smoothly, in a coordinated fashion and without getting on everybody's nerves.

In other words I could certainly profit from some SOFTWARE DEVELOPMENT PROJECT COORDINATION advice from you guys, as you have done this for a while and it certainly appears like you have the machine working quite smoothly.

In particular I would like to know what you do to take full advantage of the available WIKI, concurrent versioning and ONLINE collaboration tools available.

Thank you for your help

Caterina

Caterina Strambio De Castillia
Dottore in Scienze Biologiche, Ph.D.
OMEGA project coordinator

Biologiste 2 - Senior Research Scientist
Laboratory of viral replication, pathogenesis, and immunity
http://www.mimo.unige.ch/luban_lab/

Department of Microbiology and Molecular Medicine
University of Geneva
1 Rue Michel Servet
CH-1211 Geneva 4
Switzerland

+41 78 7367537
_________________
Visiting Fellow
Laboratory of Cellular and Structural Biology
http://www.rockefeller.edu/labheads/rou ... ject_3.php
The Rockefeller University

Hi Caterina-

Very, very sorry for the slow reply. Alot going on this month.

Sooooo, project management. I guess I would say that we do many things, and we are learning as we are going. We sometimes debate how we work as much as we debate exactly what we will work on. As you will doubtless know by now, there are lots of 'solutions' for this, and everyone in the field has strong opinions. Probably the on universal truth is your team can certainly benefit from all the hard work many others have done, but there is very likely no single canned solution, ready "off-the-shelf". You'll need to tailor some aspects of this to your own needs, and the people on your team (more on that topic later).

The tools we use represent a fairly long evolution, and the scale that we are now operating on. We have developers at 5 different sites across 7 timezones. Maybe the summary of everything we do is that anything that gets people talking to each other more about what they are thinking and doing is a good thing, as long as it is no oppressive or forced. There's a fine line there.... anyway, without further ado. We use:

Trac-- http://trac.edgewall.org/ for all ticket tracking, milestones, etc. After a very long (many months!) discussion, we bit the bullet and accepted alot of the Agile methods, and put agilo on top of our trac systems (http://www.agile42.com). As always, the melding of these two frameworks was simple in principle, but a real piece of work in practice. Part of this was technical, but much involved us learning how this process worked, and agreeing what the different terms meant (no, we didn't hire a 'coach' or consultant). Together, these tools provide the foundation for all our requirements, ticketing, and most of our process management. You can see these at http://trac.openmicroscopy.org.uk/omero (note that some of the agilo facilities-- whiteboarding-- require login.

Jabber conferencing-- we maintain an open Jabber conference for all OME devs. Generally, we are in this whenever they are on-line and working.

Continuous integration-- we use Hudson (http://hudson-ci.org/) and this resource has become vital as we have grown. See http://hudson.openmicroscopy.org.uk for our use of this tool.

Regular meetings-- OK, that sounds not very techie, but after lots of playing with this, there is just no substitute for talking together, peppered with face-to-face meetings. For VoIP, we use Teamspeak (http://www.teamspeak.com/), as we don't run into silly limits on number of attendees, and it seems to handle high latency connections much better than other, more well-known solutions. We have two meetings a week, Tuesday and Thursdays, that are minuted, with actions etc. (http://www.openmicroscopy.org/site/comm ... ence-calls). The format of these meeting varies, but we try to have one meeting/week where one person talks about what they've been doing, shows progress, etc (more like a normal research group mtg), and one where we go through tickets, status of sprint, etc.

Interactions with user community is certainly critical. We run production installations of our own software, and let our own users critically evaluate what we've done. We also have been fortunate enough to obtain grant funding to run our annual OME Users Meeting (http://www.openmicroscopy.org/site/comm ... rs-meeting) and run a both at the annual ASCB meeting. While not strictly project management, we try to time our releases around these meetings, to get as much advantage as possible from user feedback.

Hope that helps.

Cheers,

Jason