What's Not Working
This page was established early in the foundation to solicit feedback. It is now primarily of historical interest and not actively updated.
The main part of OSGeo that I think is not working as well as it could is communication. We have many mailing lists, which somewhat hide the real activity that is going on if you aren't following them every hour. There are committee conference calls as well. We also have a very active IRC channel which allows for awesome 1-on-1, real- time discussion, but it is also veiled somewhat from the public. Of course you can review IRC logs, follow many mailing lists and watch the wiki for changes, but we need some way to distill or amplify these great discussions in a manner that allows everyone to know, at a glance, what is going on. I have no real solution for this, but I'm sure we can hack something together.
Another aspect of communication are regular reports from committees, the board and other working groups. I'd like to see us get monthly reporting working - hopefully it can help us keep up with some of the key things that are going on. I also think that we could use a newsletter. It could be fed by news items/summaries from the committees and board, etc. And also highlight what folks in the community are up to with OSS in general.
One other negative point is the membership management. By this I mean the tools (or lack thereof) that allow us to find each, see who is signed up to or interested in certain projects. When I tell someone to join OSGeo, I don't have any meaningful way for them to make a donation or for me to add their name to an official list (other than the wiki). As part of this I'd like to set up a way of making shirts and business cards available for those who want them and get a payment process in place for receiving donations or orders.
- It is still early, and we don't have our software stack in place, but
I think our end goal should be to provide value to the end user. An end user will want to easilly deploy a quality Geospatial Software Stack. We are addressing quality through our incubation process. However, I'm not aware of anyone within OSGeo addressing packaging of an integrated Geospatial stack.
- I have not seen a large take up of the Collabnet platform by projects
and think that we should be considering alternatives. I'm hopeful that this is already in the pipeline.
- incubation process is a bottle neck - developer and user community benefit (so far has been work for no visible gain, I have hopes FOSS4G will change this, and OSGEO is a long term play for us so I am not surprised) - feedback on several issues (documentation license, site look and feel etc...) - viscomm: coordination on promotional materials (only found geotools handout by accident - and name is GeoTools not Geotools, and project logo not present ) - procedures, could not quickly find out how to announce a recent release (I am sure it is my fault but still) - committee communication with projects (ie community is not in the habit of checking out the flurry of email lists and committees have not contacted project representatives as needed) Tom Kralidis
I think one thing that would really be useful is for OSGEO to publish RSS (well, GeoRSS, really :) ). We'd have to figure out how to break these down (if we need more than one, etc.), but having syndication would be really useful.
I have some issues with the internal mediating structure that organises OSGeo and the communications which shape and reflect that. There at least 6 different internal committees all doing things which affect the "outside world"'s relationship with the org in some way. Does OSGeo benefit from such internal overcommittification?
So often when something interesting happens it needs to go to at least 2 or 3 of the different committee mailing lists - VisCom and FunCom, or WebCom and SAC and Geodata, or... so i get 3 copies of each email, with recursively more ridiculous cross-posting headers. Often a lot of the conversation is just configuration-irrelevant to me. I proceed to forget that others aren't also on the same ten $#?@!$ OSGeo lists (and spend time searching through different sets of list archives for references - i want them all to be integrated.)
This isn't just me bitching about tools that don't work right, though. For one, we get a chance to fix a lot of that!  Maybe the fact that people and topics in the committees overlap so much reflects that the committees are not an optimal structure. There's work to do, and people to do it. I understand that committees exist because of governance process constraints about decision making powers.  But i wonder about the extent to which they should be driving activity and do drive activity. I would definitely rather go to One Big IRC Meeting of 2 or 3 hours with definite internal schedule once every 2 or 3 weeks, than the weekly trickle of different topics i care more or less about but might be interested in learning from or helping with.
I think all the cross-posting on the committee lists also reflects a kind of self-selecting meritocratic cabal, the core of which also talks on IRC a lot. I think of jgarnett's complaint of something 'going missing' to public view when it heads off onto the viscom list. But the discuss list is too big and general-interest to get into the minutiae of how many t-shirts to order for people at a conference.
More technological fixes... start an osgeo-awareness RSS feed and moderated list... have people write more down about what they are doing. Try and keep stuff *off* the lists, omg, and not start too many more. The board list, btw is open posting and some non-board-members subscribe. Make clearer about wiki etiquette - it really is okay to tear others' stuff apart - to alter public opinion - that's what wiki is for.
- Site navigation is non-intuitive, at least to me :( . The way the community wiki and the official site are linked is confusing since they use different navigation menus that get me lost quickly. It's not clear to me what the role is of the official web site compared to the WIKI; much of the relevant information is developed on the WIKI. Will the distinction remain? A clear lay-out and navigation structure would help a lot. E.g. through a CMS that provides both WIKI and other document types, I guess that's covered by Drupal as proposed.
This point should not be taken as a hard "this is not working" as I know people work on it and its a tedious and involved process:
- The supporting infrastructure needs a lot of development for new projects or migrating projects in order to keep project administration simple (avoid the need for five different service providers for the tools used by a project).