Hi all,
I am a fairly new charter member, so maybe the two following comments
here will be irrelevant.
From my perspective, having Orfeo ToolBox as an incubating project
definitively helped us to move in the right direction. I am not saying
that it would not have occured without OSGeo, but the organization gives
the momentum and defines the standards to reach. As such, it is useful
and somehow efficient. The fact that the process is long is mostly on
the project side in our case.
I think that the Github move is hazardous. Sure, it is easy, free for
open-source projects, and really really cool. Granted, it helps a lot in
getting fluid contributions to open-source projects. But ... in two
years, they may start shipping sponsors links at the end of the Readme
files, and in a moments notice you have to watch 20 seconds ads before
cloning. At this point, you will want to bail out, only to find out that
in fact you can not, because you can not delete the project anymore, or
the issue tracker database can not be exported ...
My point is, OSGeo should care about long-term protection of GIS
open-source, and if this goal aligns for now with services that Github
provides, it may no longer be the case in the future .Of course we need
to be on Github: it is a public place to be, like twitter & co. But
completely giving up code hosting and developers exchanges to a private
company is the opposite of what I think the organization should do.
I know proper hosting services requires time and money, I do not have
the solution to that, but for me OSGeo should provide a sustainable
alternative, up-to-date and tailored for its purpose.
My 2 cents,
Regards,
Julien
Le 25/09/2015 21:57, Darrell Fuhriman a écrit :
The recent discussion on the board list
<https://lists.osgeo.org/pipermail/board/2015-September/013172.html>that
came out of the question of the 2014 videos has got me thinking about
a few things again, and I want to try to get them out there.
Grab a mug of your favorite liquid and hunker down, because I put some
time and effort into this, and your own well considered reply is
appreciated.
Keep in mind that all of these comments are coming from my personal
perspective, which, like everyone’s, is an incomplete picture of the
whole. Much of what I’m going to say has been rolling around my head
for a while, so I’m just going to put it out there.
I will start with a provocative thesis:
OSGeo lacks visionary unified leadership and without it will become
irrelevant.
Of course, making such a claim requires support. So let me break down
the statement.
“Visionary leadership” is really two things, “vision” and
“leadership.” I will address each in turn.
OSGeo lacks vision
I looked at the list of “Goals” for OSGeo
<http://www.osgeo.org/content/foundation/about.html>. I wonder: when
was the last time these goals were evaluated for both success and
relevancy?
Here is my own opinion of success of some of these goals. (In the
interest of brevity, I haven’t tried to tackle everything. That’s left
as an exercise to the reader.)
Example 1
To provide resources for foundation projects - eg. infrastructure,
funding, legal.
Allow me to break each of those examples down.
Infrastructure
It’s true that OSGeo provides some infrastructure, such as Trac
instance, Mailman, SVN repos. If the budget is to be believed, we pay
some $3,500/yr to OSUOSL for said infrastructure. I wonder if such a
service is necessary, however. Issue tracking and source control are
much better provided by Github, which is free for organization such as
ours.
I say this because a) that’s money that could be better spent
elsewhere and b) supporting these services burns precious volunteer
time (more on that below).
There are clear cost savings available, which are not taken advantage
of. For example, OSGeo could be hosting FOSS4G infrastructure:
conference websites and registration, a central location for
conference videos (regardless of platform/provider). This neglect is
especially galling given that FOSS4G is OSGeo’s sole source of income.
Funding
OSGeo does not fund projects. It has provided some funds to pay for
Code Sprints — $15k in 2014 according to the budget
<http://wiki.osgeo.org/wiki/OSGeo_Budget_2014>.
Legal
I see nothing that has been done on this front recently. Please feel
free to correct me.
Conclusion
OSGeo, where it actually does what it claims, has not adapted in ways
that could save money.
My grade: D
Example 2
To promote freely available geodata - free software is useless without
data.
The geodata working group is dead. As near as I can tell by perusing
the mailing list archives, and the wiki, there has been no meaningful
activity in the past two years (maybe more).
My grade: F
Example 3
To promote the use of open source software in the geospatial industry
(not just foundation software) - eg. PR, training, outreach.
The Board of Directors
<http://wiki.osgeo.org/wiki/Board_of_Directors#Packaging_and_Marketing>page
says:
Packaging and Marketing
OSGeo’s marketing effort has primarily been focused around the
packaging and documentation efforts of OSGeo-Live, and to a lesser
extend[sic], osgeo4w. […] It has been entirely driven by volunteer
labour, with 140 OSGeo-Live volunteers, and printing costs have been
covered by local events or sponsors. In the last couple of years,
OSGeo has covered local chapter expenses required to purchase
non-consumable items for conference booths (such as a retractable
banner). In moving forward, OSGeo hope to extend marketing reach by
providing co-contributions toward printing costs of consumable items
at conferences, such as toward OSGeo-Live DVDs.
Local Chapters
Much of OSGeo’s marketing initiates are applied at the local level. In
many cases, this is best supported through as little as an email list
and wiki page. OSGeo also supports local chapters by offering to pay
for an Exhibition starter pack for local chapters. Local chapters are
also usually the coordinators of conferences and related events, as
mentioned above.
Exhibition starter packs almost never happen; OSGeo-Live
explicitlygets no support; and OSGeo struggles to staff a booth at its
own conferenceto say nothing of any other conferences.
Note: Local chapters certainly do do marketing and outreach, but these
efforts are essentially unsupported by the OSGeo Foundation. In fact,
this goal and the Board of Directors webpage seem to be explicitly
contradictory.
My grade: F.
Commentary
I could go on with my own personal evaluations, but I’m not sure
that’s necessary. The only place I see that OSGeo has unquestionably
succeeded in the past few years is the final goal, “To award the Sol
Katz award for service to the OSGeo community”.
So, what’s my point here? It’s simple: there is no longer a coherent
vision for what OSGeo should be. I’ll return to that below, but let me
continue with my other point.
OSGeo lacks leadership
Again quoting the Board of Directors’ page:
The board’s primary responsibility is to efficiently and effectively
make strategic decisions related to the running of OSGeo.
I won’t bore you with the details, but a perusal of the board meeting
minutes would indicate that strategyis rarely, if ever, a part of the
meetings.
The emphasis on consensus-based decision making often leads to no
decisions being made. I can’t count the number of discussions that
have come up on the board list only to devolve into a morass of
nit-picking and eventual lack of action when everyone tires of the
discussion. What action that is taken is often to “delegate” to a
(possibly inactive) sub-committee, then never follow up.
Instead what we have is a great deal of inertia, little interest in
changing things, and no clear indication of what the Board’s
priorities are.
If priorities do exist, they’re lost in a maze of confusing,
incomplete and often contradictory information on the wiki. (Wikis —
like abandonware for documentation.)
On pending irrelevancy
I encourage you to ask some random people in the open source
geospatial community what OSGeo means to them. I would make a bet that
the most common answer is a blank stare.
I would ask the board members to come up with three things, other than
FOSS4G, where the OSGeo membership has shown its importance to the
community as a wholein the last two years. Something where people say,
“Did you hear about[exciting thing]OSGeo is doing on X?” To be clear,
I don’t mean just things that OSGeo has a finger in, but things that
needOSGeo. If OSGeo disappeared tomorrow, would any of these projects
be significantly affected?
I don’t think it can be done. The OSGeo Foundation is sliding into
irrelevancy — and it may already be there.
If anything should be seen as strategic for OSGeo, it’s FOSS4G, the
foundation’s primary (sole?) source of income. Even regarding its
flagship public event, the board is largely absent. Rather than
provide adequate resources and planning, they instead rely on burning
out volunteers, then make post-hoc demands on the way they shouldhave
done it, provide no future support for organizers to heed those
demands, rarely follow up, then go on to repeat the same mistakes the
following year. Honestly, it’s surprising that FOSS4G has failed only
once. (I think this is a reflection of the demand for the conference,
not the blazing competence of OSGeo.)
Michael Gerlek brought this up
<https://lists.osgeo.org/pipermail/discuss/2015-July/014521.html>on
the osgeo-discuss list in July, and probably has a more generous spin
on it. He essentially argues that it’s time to declare mission
accomplished and shut down or rebooted. I agree with his points, and
I’m arguing that OSGeo can have something to offer, but it will
require a major re-think of its mission.
Fixing things
I hinted at this in my recent questions to the board candidates, but I
want to be explicit here: OSGeo needs to evolve or die.
Here’s how I would do it:
1.
The board needs to evaluate all of its goals, as defined on the
About page, to decide if they are still truly goals. Define any
new goals.
2.
Ask the question: “What does it mean to succeed at this goal?”
If the goal is vague, or ongoing, give a timeline: “What does
success look like for this goal one year from now?”
3.
Create measureable objectives for achieving those goals. Ask the
question, “How will we know if we’ve succeeded?”
4.
Prioritize the goals.
5.
Allocate resources to the goals.
Obviously this is a tricky one, but I think we can look at this a
balance between Importance and Effort.
Spend money to reduce to the effort required, more money if the
goal is more important — this might be the hardest cultural shift.
Volunteer time is precious and easily discouraged. Make sure that
you make it as efficient as possible by spending money when you can.
For example, many of the infrastructure services OSGeo provides
can be easily outsourced to more featureful services that are more
responsive and rely less on volunteer labor.
6.
Close the loop on tasks. When a task is delegated to a committee
or individual, track its progress, both to know that it is or
isn’t happening, and to be able to acknowledge and incorporate the
work when it’s done. Failing to acknowledge people’s labor or to
use the results of that labor will virtually guarantee that the
volunteer does not continue to help.
7.
Evaluate success and failure. GOTO 1.
Aside: none if this will happen without a strong executive. Whether
that position is paid or not is up to the board, but it’s clear that
there needs to be someone who can make decisions without endless
rounds of fruitless discussions. The board as currently constituted is
not dysfunctional, but it is mostly afunctional.
I’m will go so far as to suggest this: Fly every board member who is
available to a two or three day retreat. Get everyone in the same
room, a professional facilitator to speed the process, then figure out
what OSGeo is going to be and how to get there. Don’t fret excessively
about the expense — this isn’t about saving money, it’s about saving
OSGeo.
If you ask me, irrelevancy is a fate worse than death. Be bold!It’s
better to try to do something big and new then fail than to simply
fade away and be forgotten.
Though my comments above may sound harsh, they are sent with the very
best of intentions. I want OSGeo to succeed, but OSGeo is never going
to succeed if it doesn’t know what it’s try to succeed at.Without real
reform, I don’t see success happening, just irrelevance. Here’s hoping
this gets the ball rolling.
Darrell
--
Julien MICHEL
CNES - DCT/SI/AP - BPI 1219
18, avenue Edouard Belin
31401 Toulouse Cedex 09 - France
Tel: +33 561 282 894 - Fax: +33 561 283 109
_______________________________________________
Discuss mailing list
[email protected]
http://lists.osgeo.org/mailman/listinfo/discuss
