David:
You may not remember me, but we met when you worked with Stephen Loosli.
I think he will remember me too-- I helped him try to raise venture
capital through Wayne Brown Institute.
I think Jacques has some good points. As a newcomer to the project (but
who has watched it from the outside for several years now), I think the
project currently has a learning curve that is both intimidating to new
users, and perhaps much higher than it needs to be.
Open source projects really only grow and flower when they get a
critical mass of users and developers that can take over the bulk of the
development effort. Compare NetBSD to Linux, for example. One uses the
cathedral model, the other the bazaar. The more users you have, the
more developers you get (although that is never a 1:1 proposition).
Open source is all about "scratching your own itch", so the more users
you have, the more itches (of all varieties) you have that need
scratching, and the faster the project develops, though still perhaps
never in the direction any one person wants.
But you also make some interesting observations, that I would like to
comment on.
On Mon, 2010-01-18 at 14:15 -0600, David E Jones wrote:
I think you're right Jacques, though I also think it's a natural and
maybe not a bad thing.
Agreed-- human behavior is a natural thing, and probably not bad, though
often problematic if we have not properly incented or encouraged the
behavior we want or expect.
Most people in the world aren't investing in much right now, even those
who have time and/or money to invest, because they don't expect much
of a return on the investment. Any successful investor will tell you
that it's best to get into things when it is cheap to invest (ie when
others don't expect a return and aren't investing), and then get out
when the market is great (when others do expect a return and are
investing a lot). However, the same people will often fail to
recognize that such conditions are not only impossible to predict in
the future, but also impossible to recognize in the present, and they
further fail to recognize that the fact that they were able to invest
in a low market and get out when the market picked up was probably
just the luck of timing. Just ask any honest economist (if you can
find one!)... they can't even accurately analyze the present, let
alone predict the future... even if they are asked to every day.
Which means that this is the best time (economically, on the macro
level) to invest, right?
Or perhaps (on the micro level) the reason people aren't investing is
that they are having a hard time seeing the potential payoff in whatever
micro-environment you are looking at, including OFBiz.
The CPI was down 0.4% in 2009, the biggest loss since about 1955 (before
I was born). Yet our business was up over 10%. Best Buy was up over
25%. People are spending money-- just not where they can't see or
expect an immediate return.
Anyway, the point of all of that is that yes there isn't as much
activity in OFBiz right now. I think this is natural given that most
end-user organizations have pulled back their budgets significantly.
That makes sense because the sectors where OFBiz has been strongest
(at least from my experience) are getting hit really hard right now
and many companies are not just cutting back, but totally failing.
End-user organizations have always been about 90% of my work, but are
now down to 0%... and I haven't designed or built anything for one in
almost a year now. Fortunately there are other types of companies out
there, but one way or another the demand for OFBiz is WAY down as the
industries and sizes of businesses it serves are hit so hard.
I was tempted to pull the trigger on OFBiz a year or two ago, but was
quoted $10,000 just to customize it for our company, and it was not at
all clear to me 1) what I would get for that sum, and 2) how long it
would take. That is a hard proposition to sell to your management team.
From the perspective of the project what does this matter? Unlike a
company a non-profit community-driven open source project has no
liabilities and there is no way it can just disappear (unless no one
is using it, ie it falls into irrelevancy). Even if somehow the ASF
failed to get adequate donations of money or server time to keep
hosting the project, it could be hosted elsewhere, or even turn into
some crazy peer managed ecosystem using distributed git repositories
(some people would love that, but I think it would be a nightmare for
both contributors and users).
Irrelevancy is the risk, as you point out. The best software in the
world goes nowhere if no one uses it. As as Microsoft has shown for
years, crappy software rules the world, if everyone uses it.
In any case, the worst case scenario is that the project slows down a
little bit. If we're lucky people won't lose too much hope and will
continue to invest in open source projects like OFBiz, which is really
an investment that developers do in hopes of getting more and better
work in the future. If people don't feel much hope for that then
chances are they won't invest as much, and maybe just the hobbyists
who want to develop neat stuff will be contributing.
Unless the slowdown is an indication of a project getting too heavy to
support its own weight. I hope that isn't the case, but if "free" isn't
a draw in a bad economy, that may be an indicator of a more serious
problem.
Asking developers to contribute in the hopes of future work is putting
the cart before the horse, IMHO. Instead, offer a product that gives an
immediate return without cost, and new users will come in. Once in,
they will need to scratch their own itch, which results in further
development of discrete additions/functions funded by businesses that
need them. They would rather build an integrated solution onto OFBiz,
than scab a proprietary solution onto the outside somewhere and deal
with data import/export.
Sooner or later things will of course pick up, and then we'll be back
to complaints about too much going into the project instead
complaints about not enough. On the other hand, given an infinite
potential universe of things people might want to see in OFBiz, there
will ALWAYS be complaints about things missing that certain people
would like to see, or things organized better for their current
requirements, or wishes for free help instead of peer collaboration.
If the framework develops in a way that unnecessary features and
complexity can be safely ignored by newbies, they will still use it. If
they feel overwhelmed, they won't.
There is a reason that Linux and OpenOffice are widely used, while TeX
and GIMP are not. Easy-to-use distributions, and familiar user
interfaces make all the difference.
By the same token, cars (and software) are sold by test driving. If it
isn't easy and comfortable to drive, it isn't easy to sell. We all hate
the VCR that is so complicated, we can't have forgotten how to set the
clock. OFBiz looks to me like a hot rod project with a great drive
train, fat tires, and great cornering, but no seats yet (let alone seat
belts, widows, etc). You put a bucket upside down behind the wheel and
say "try it out, you can trim out the interior any way you want, for a
fee". You may get some buyers, but that is definitely the hard way to
sell a car.
Oh well, I guess enough pontificating for now, especially since I'm no
pontiff.
-David
Don't get me wrong, I think OFBiz is a great project. But it tends
toward massive complexity that may be drowning its initial or perceived
utility, at least OOTB.
I think OFBiz has the best data model I have ever seen. The cost of
that is complexity. OFBiz has nearly 1000 tables, OpenERP less than 200
IIRC. That makes it hard to get your head around. Simple things like
importing standard CSR leads seem unnecessarily complex-- parties,
organizations, addresses and phones all in SEPARATE tables. While I
understand the value, that flexibility comes at a cost.
Having said all that, I have just discovered the documentation at:
http://demo.ofbiz.org/cmssite/cms/APACHE_OFBIZ_HTML
which appears promising in organization, but is not written or
translated by native English speakers. I LOVE the fact that it is
incorporated right into the distributed source code, so up to date
(theoretically) with the distribution.
If I could suggest some priorities, they would be as follows:
1) Make good documentation that is easy to find (part of the system is
great). Index it and cross-link it extensively. Most important, start
with a gentle primer on how to kick the tires and start to use the
system.
2) Create a setup process that makes it easy to customize for a new user
to get started, with its own process flow. I'm thinking a turbo-tax
type wizard that steps you through the process. Maybe 1) load demo data
to demo (but not to start real use); 2) configure the system (root app,
ports, etc); 3) load company data from various sources (a spreadsheet
for employees, a text file of customers, a list of GL accounts, etc).
3) Show off some features for an immediate and perceptible return.
Maybe an inventory dashboard or something. Print some invoices, show
outstanding purchase orders, or SOMETHING that makes the user say OK,
now THAT I can use immediately.
But as you correctly point out, suggestions are cheap, it's helping that
makes things happen. I'd love to take you to lunch and see how I can
help. We are a small company, without a huge budget, but we do have
some, and I would like to help make the project better. Perhaps we can
bring some accounting and business management experience to the project.
Drop me a note, I'm in Salt Lake City and would like to see you again if
you are still in the vicinity somewhere.
On Jan 18, 2010, at 11:48 AM, Jacques Le Roux wrote:
And then, those that invested much, slowly invest less...
Jacques
From: "BJ Freeman" <[email protected]>
The missing element is manpower.
Get lots of great Ideas but not many contributors
Axel Van Noten sent the following on 1/18/2010 6:58 AM:
Thanks, I learned a lot.
Still, I'm a little bit disappointed about the overall documentation. Would it
be possible to:
1) have it organised by release,
2) link to it immediatly from the documentation buton (on ofbiz.apache.org), or
the help button in the application,
3) have a clearer structure (per release) like:
Getting Started with OFBiz
OFBiz End User Documentation
OFBiz Reference Manual
OFBiz Technical Guide
OFBiz Tools and Utilities Guide
...
I know its easy to criticise, but I believe the project would gain a lot just
by changing these links and making a different hierarchy.
Kind regards,
Axel
From BJ Freeman <[email protected]>
Subject Re: OFBiz data load
Date Thu, 14 Jan 2010 19:29:38 GMT
http://cwiki.apache.org/confluence/display/OFBENDUSER/OFBiz%27s+Data+File+Tools
here is a good link for a lot of help
http://cwiki.apache.org/confluence/display/OFBIZ/FAQ+-+Tips+-+Tricks+-+Cookbook+-+HowTo
http://cwiki.apache.org/confluence/display/OFBIZ/Handling+of+External+data
Juan Pablo sent the following on 1/14/2010 8:42 AM:
Hi Axel.
1)
http://cwiki.apache.org/confluence/display/OFBENDUSER/OFBiz+End+User+Docs+Home
http://cwiki.apache.org/confluence/display/OFBTECH/Home
2) http://ofbiz.markmail.org/
On Thu, Jan 14, 2010 at 8:30 AM, axel van noten <[email protected]>wrote:
Hi
I've been installing OFBiz and looking for the following information:
1) Where is the primary point can I find the most update documentation?
The documentation buton on the OFBiz website and the help button in the
application are both bringing me to the cwiki.apache site. But there seems
to be no clear structure nor a primary point where I can turn to for manuals
and documentation.
2) is there a possibility to perform a text search on this mailing list?
3) I'm trying to load a product database into OFBiz. What would be the
appropriate way to do so? Is this documented somewhere?
Thanks for your help!
Kind regards
Axel
_________________________________________________________________
Keep your friends updated�even when you�re not signed in.
http://www.microsoft.com/middleeast/windows/windowslive/see-it-in-action/social-network-basics.aspx?ocid=PID23461::T:WLMTAGL:ON:WL:en-xm:SI_SB_5:092010
Windows Live: Make it easier for your friends to see what you’re up to on
Facebook.
_________________________________________________________________
Windows Live: Make it easier for your friends to see what you’re up to on
Facebook.
http://www.microsoft.com/middleeast/windows/windowslive/see-it-in-action/social-network-basics.aspx?ocid=PID23461::T:WLMTAGL:ON:WL:en-xm:SI_SB_2:092009
--
Matt Warnock <[email protected]>
RidgeCrest Herbals, Inc.
