I am assuming here that part of the assumption behind the Open Source
model is that the maintainer of the code base can also maintain
profitability and survive at the same time as providing a community with
the source code. It's not just about getting the job done.
Prior to JT's post I had spoken to several contributors to Asterisk who
had informed me that the core components of the project and core future
rewrites are normally completed by a dedicated team of developers
(normally internal on payroll at Digium). In some instances the
developer team is outsourced under contract from Digium. Under either
method almost the entire price of the dev work hits Digium's books.
A lot of the development is just too critical to leave to a 'best
efforts' community. I believe that this is just reality. The community
is useful in testing and bug reporting on the product, suggesting
patches and work around on the periphery (not often in core architecture
issues). Even where a significant component is contributed by an
external consultant (eg fax) Digium have little choice but to bring one
of their internal teams up to speed on the code so that it can be
considered in future release versions.
I therefore believe that the price of developing Asterisk to Digium is
high. Therefore researching into funding options is a very logical
outcome.
Now, if you look at the type of contributor who is a) asking for fixes
and/or features and b) willing to significant hard cash for the work,
you start to see some common themes. These contributors have created a
business of varying size that is using Asterisk, quite often as a core
component of some commercial product (Fonality, Enswitch,
Tesco/Freshtel, etc). These businesses have the cash and are willing to
part with funds for desired features. The current implementation of
Open Source provides little avenue for relieving these companies of
their cash (I regard this as a system failure).
Also, if we now turn our attention to the way Digium has managed its
relationships with the likes of Fonality etc I believe you would see
something of a cold shoulder approach to these businesses. In my opinion
this is somewhat of a failure to make the best of a significant
opportunity.
And now, JT is suggesting that there could be a means of accessing
funding from the likes of these businesses to improve the development
process. Is it too late to reverse previous decisions and change a
fundamental management approach towards those entities that are the
prime candidates for contributing significant funding. I believe it is.
But who knows, however what I know is that thus far Digium's
implementation of its Open Source model has some significant
shortcomings.
That's just my 2c.
Thanks
Craig Lawrence
-----Original Message-----
From: asterisk-biz-bounces@lists.digium.com
[mailto:asterisk-biz-bounces@lists.digium.com] On Behalf Of C.
Savinovich
Sent: Saturday, 1 November 2008 2:20 AM
To: 'Commercial and Business-Oriented Asterisk Discussion'
Subject: Re: [asterisk-biz] General development funding: discussionand
survey
Craig,
May I ask in what sense do you believe the open source project is
not
working for asterisk? My opinion is that it is, but I am open to hear
yours, since it is the educated opinion of someone in the industry.
And I gather, based on your response, that that's how you see it.
Best regards
CS
-----Original Message-----
From: asterisk-biz-bounces@lists.digium.com
[mailto:asterisk-biz-bounces@lists.digium.com] On Behalf Of Craig
Lawrence
Sent: Friday, October 31, 2008 1:09 AM
To: Commercial and Business-Oriented Asterisk Discussion; Asterisk Users
Mailing List - Non-Commercial Discussion
Subject: Re: [asterisk-biz] General development funding: discussion and
survey
John
You are inferring that in this instance (with respect to Asterisk) the
Open Source Model isn't working.
As a consequence, you are researching alternative means of funding the
existing Digium Open Source model.
I would respectfully suggest that you look further into the reasons for
the failure of Digium's implementation of the Open Source Model before
you canvas the community with requests for funding.
Regards
Craig Lawrence
-----Original Message-----
From: asterisk-biz-bounces@lists.digium.com
[mailto:asterisk-biz-bounces@lists.digium.com] On Behalf Of John Todd
Sent: Friday, 31 October 2008 6:12 AM
To: Asterisk Users Mailing List - Non-Commercial Discussion;
asterisk-biz@lists.digium.com
Subject: [asterisk-biz] General development funding: discussion and
survey
[sending to -users and -biz in a slightly different format to broaden
participation]
Summary:
Would you help fund different Open-Source Asterisk enhancements,
bugfixes, or documentation if there was a way to collectively
contribute money towards the effort without a profit margin
incorporated into the price? If so, jump to the bottom of this
message and fill out the form on the URL provided. There is no
obligation to anything by filling out the form - think of this as a
"market survey".
The Long Version:
Everyone on this list is presumably an Asterisk user, advocate, or is
in some way benefitting from the project. Your ideas and survey
participation would be welcome on the topic below.
Many coders love coding for Asterisk but often can't find the time to
do it for free when faced with things like buying food, paying
mortgages, and keeping current with their insurance - this is totally
understandable. Many coders have and continue to contribute things to
Asterisk at no cost, but these patches are typically their own
"itches", where they have solved a particular problem of their own.
Rarely do people pick up problems that are not related to anything
they're doing, or pick up unrelated problems that are so large that it
would involve 100% of their time for any significant period. Some
people ("Bless your heart!" as they say here in Huntsville) work on
bugs and enhancements that don't directly benefit them at all - these
are the most valuable contributors we have - you know who you are.
Most of the time, though, there is a directly relevant reason why
people work on code and often that means more obscure bugs or feature
implementations languish, though still worthwhile if someone were to
complete them.
On the other side of the scale there are many people or companies who
perhaps would like to contribute to paying for various features in
Asterisk that would be described as "large enhancements" or even minor
bugs and annoyances, but do not have sufficient funds to pay for an
entire project themselves. There are perhaps also many people who
would like to help out Asterisk in a way that allows them to
contribute funding towards the project, but they're uncomfortable
sending money to a corporation and hoping that it gets eventually
applied to OSS Asterisk (and I'm not only talking about Digium in this
case.) There are coders available for a fee (perhaps much less than
market rate, perhaps not - we'll just say "non-zero cost") who could
do this work and would love to do it if they could justify the time
spent. Open-Source Software doesn't always imply that the code is
"unpaid work", and Digium's contributions towards Asterisk are a case
for the benefits of having an income stream and payment system
(salaries) that supplements OSS development.
So there is a disconnect between two groups of willing consumers and
willing producers - how do we bridge it? The answer some have come up
with is "Let's create an Asterisk fund and collect money and disperse
money to pay for work by community members!" This is a great concept,
but the devil is in the details, and I've found that when money is
involved, the detail devil is much larger and angrier than usual.
The problems with this idea have continually been:
- Escrow of capital. It is not feasible to trust that donors will
be good on their contribution post-release. This may be because it
takes a while for the code and economic situations change, it may be
that internal paperwork processes take forever to get done (Hi, Raj,
sorry about that delay from Tello!), or it may just be that a large
portion of funders are flaky. I'm willing to be convinced this isn't
the case, but personally I certainly wouldn't code a large amount of
hours based on the say-so of people I'd never worked with before.
Perhaps some sort of metric could be created for more reliable payers,
like a rating system of integrity?
- Agreement of project goals. Who defines the project? Who gets what
they want? Based on money? Based on some arbitration? What and who
defines "success"?
- Corporate structure for payments. If there is an agent in between
the coders and the funders, then what kind of agent is that? For-
profit? Not-for-profit? Who pays for the creation of this entity?
It's possibly the case that Digium Inc. is not the best place for this
funding repository, though possibly that would make life a lot easier
from an organizational standpoint. (not sure about taxes, though.)
- How to pay? Obviously, the more the merrier, but credit cards, bank
accounts, PayPal, and other payment instruments are complex and
expensive. Payment to consultants is another problem - taxation may
be a problem again.
- Serious interest. This has been a topic of conversation for the
last 6 years that I'm aware of, and none of the concepts or problems
I'm bringing up here are new. However, it is discussed but no action
is taken. Perhaps now is the time to serious look at this concept
since Asterisk is reaching such a large audience. Traditionally, the
number of people or organizations that would provide "seed" funding
for something like this is low; possibly only a single organization
(Digium) would have the focused interest and capital to create such a
financial/organizational entity as a non-profit or other unrelated
instrument. But who would use it, really? At what level of actual
contribution? To convince Digium (and/or hopefully other founding
members of some as-yet undefined organization) to put their money and
effort towards such a fund/foundation, there would have to be
significant interest beyond idle discussion. The bounty concept on
voip-info.org has been around for a while, but saw only marginal
uptake. I've been a part of three or four (or more) paid projects,
but only two (the sounds-extras recordings, and the SIP session
timers) have actually seen multiple contributors, and the rest were
straight-up consulting. What companies or individuals would actually
put money into such a fund, and would it be enough to make it
worthwhile or self-sustaining on an ongoing basis? Those of you who
re-package Asterisk for commercial purposes but are not active
contributors of patches or enhancements: I'm looking at you,
specifically.
Before these fairly large problems get discussed, I think the first
goal in this investigation would really be to see if there is actually
an interest in such an entity, so that last point is the problem I
think we can try to solve here first. To try to get some metrics on
this that are something other than "around-the-dinner-table"
discussions, I've put together a form to collect some data. If you
have an interest in putting money towards some general fund for
Asterisk development, we're at a stage in the cycle where your input
counts. Please take a moment to put your data in the form below.
Note that all replies (except for email address) are public, given the
spirit of this whole concept being a community effort.
All of the points above are open for discussion. Your comments on
this thread are welcome, since this is truly a "community" concept.
Since this topic has come up on many occasions, but rarely in a wide
forum, this perhaps is a good time to determine if there is truly
interest or if the methods that we have currently are sufficient. If
may be the case that there is no interest - that's OK. It may be the
case that there is interest, but there is no viable way to create such
an entity. (Digium, my employer, is not putting itself on the hook
for anything in this message - I'm trying to gauge interest ONLY.)
You're not obligating yourself to actually paying anyone by filling
out these forms, as this is just a quick survey of how large the
audience really is.
PLEASE keep discussion of policy, structure, and technical details
here on the list. Use the form below ONLY for cataloging your
willingness to pay for Asterisk work via some idealized mechanism.
If you're a coder, a survey about that may follow later which captures
your willingness to work for money, but I suspect that's a pretty well-
known truism that needs little confirmation.
Feel free to forward this note to others (your boss, your co-workers,
other less-involved Asterisk coders) who may have an interest in this
type of funding approach. I figure a month would be a reasonable time
for letting this form collect data, but that's open for discussion as
well.
Form for your input:
http://spreadsheets.google.com/viewform?key=pmllU1ebPNlhUl_3QHUqvAg
Results:
http://spreadsheets.google.com/pub?key=pmllU1ebPNlgCMhLQ2FaZRA
JT
---
John Todd
jtodd@digium.com +1-256-428-6083
Asterisk Open Source Community Director
_______________________________________________
--Bandwidth and Colocation Provided by http://www.api-digital.com--
asterisk-biz mailing list
To UNSUBSCRIBE or update options visit:
http://lists.digium.com/mailman/listinfo/asterisk-biz
_______________________________________________
--Bandwidth and Colocation Provided by http://www.api-digital.com--
asterisk-biz mailing list
To UNSUBSCRIBE or update options visit:
http://lists.digium.com/mailman/listinfo/asterisk-biz
_______________________________________________
--Bandwidth and Colocation Provided by http://www.api-digital.com--
asterisk-biz mailing list
To UNSUBSCRIBE or update options visit:
http://lists.digium.com/mailman/listinfo/asterisk-biz
_______________________________________________
--Bandwidth and Colocation Provided by http://www.api-digital.com--
asterisk-biz mailing list
To UNSUBSCRIBE or update options visit:
http://lists.digium.com/mailman/listinfo/asterisk-biz
Posts
No comments:
Post a Comment