Next Steps for CCIF / Cloud Forum
afalcon
computing, than we need to formalize quickly. I suggest the
following.
1) Establish a 501(c)(6) non-profit (membership association)
2) Create an "office" to manage the association's assets (domains,
groups, etc.) and membership, and to identify facilitators for the
process.
To do this, we will need to identify an initial board of directors (at
least 5), contract with an individual or firm to provide the office
resources, and secure funding.
The funding is best raised through annual membership dues. Based on
experience, we would need $200K - $250K for the first year of
operation of the office. Not difficult with enough vendors and dues
in the $5K to $10K range.
The person or firm hosting the office should not be a vendor with a
direct interest in the resulting standards, and no member of the firm
should serve as a director.
Creating a charter sufficient to obtain 501(c)(6) status is not
difficult. We can then develop a more elaborate charter and set of by-
laws.
I know firms that could do this for us, and as a point of disclosure,
am involved with one such firm that might be willing to do it at cost.
Is this of interest? Will the CCIF's current sponsors be willing to
provide initial funding? Do we have volunteers willing to serve on
the initial board of directors?
Allen
Gary Mazz
First we need to determine the purpose of ccif; advocacy, standards,
marketing or discussion.
-gary
Gary Mazz
-g
Agreed..
First we need to determine the purpose of ccif; advocacy, standards,
marketing or discussion.
-gary
afalcon wrote:
Mark A. Carlson
standards when there are plenty out there already that some
of us already pay dues to?
Why not just keep this as an advocacy group that advances
standards and interoperability without the need to create
their own standards?
Some things you can do without an IP policy and membership
dues:
Gather, document and promote requirements
Write white papers, positioning papers
Create models for taxonomy, roles, APIs, etc.
Document common terms and definitions
At the OMG cloud day last week there was a good discussion
about having a central point of coordination that could help
document a landscape of cloud APIs and standards that could
both leverage and coordinate the various cloud computing and
storage standards. It can be as simple as a wiki site with
a common way to describe each standard with pointers off to
the actual standards and SDOs.
-- mark
eprpa...@gmail.com
front money to a group that hasn't produced anything? Why not just
create the document, publish it and then form the organization?
Chuck Wegrzyn
eprpa...@gmail.com
a non-partisan open environment. At that point we have a track record,
and it would be easier to sell the idea of an organization.
Chuck Wegrzyn
> Sr. Architect
>
> *Systems Group*
Reuven Cohen
Reuven
Sam Johnston
Why create yet another "pay to play" organization to craft
standards when there are plenty out there already that some
of us already pay dues to?
Agreed - there's already the Open Cloud Consortium (and others) for those who feel the need to part with their hard earned. Budgets for this type of "investment" are tight to non-existent in this economic climate anyway and such a barrier would, for better or worse, exclude most of the people who have contributed so far, myself included (it's one thing to donate time and energy but another altogether to have to pay for the privilege).
Why not just keep this as an advocacy group that advances
standards and interoperability without the need to create
their own standards?
Some things you can do without an IP policy and membership
dues:
Gather, document and promote requirements
Write white papers, positioning papers
Create models for taxonomy, roles, APIs, etc.
Document common terms and definitions
Exactly. I've already done a huge amount of work on models et al and while I'm chuffed to see them being picked up by the likes of IBM (Architectural manifesto: An introduction to the possibilities (and risks) of cloud computing) I would rather be working in a workgroup than essentially alone on this; there are some very complex models and overly simple ones with little in between.
There's a lot of content for bootstrapping at http://wiki.cloudcommunity.org/ including the stack (x2), a glossary, the start of some interoperability guidelines, the standards roadmap, the cloud platform reference architecture (needing a FOSS reference implementation), occi (temporarily), the bill of rights and manifesto, the incidents database, the open cloud principles, some graphics for other to use and even some community metadata and a privacy policy.
This is a heck of a lot of work for one person but it would be no problem for a community.
At the OMG cloud day last week there was a good discussion
about having a central point of coordination that could help
document a landscape of cloud APIs and standards that could
both leverage and coordinate the various cloud computing and
storage standards. It can be as simple as a wiki site with
a common way to describe each standard with pointers off to
the actual standards and SDOs.
+1 - almost all of my Cloud Computing Community intiatives are based around a single MediaWiki installation and that's all you need to create an incredibly utile resource (e.g. Wikipedia).
Incorporating a legal entity to hold assets like domain names, trademarks and a bank account as well as to issue tax-exempt receipts is trivial and inexpensive. We need not elect a board as such - the existing committe will do fine for that too. Most "open source" style organisations like CAcert tend to work like this and it's both efficient and effective. It's not urgent either as in the mean time individuals and supporting organisations like Enomaly can hold assets in trust and process transactions as they have done thus far.
Is there anybody who thinks we need more than this on the legal entity side of things, and if so, why?
Sam
Jesse L Silver
Is there anybody who thinks we need more than this on the legal entity side of things, and if so, why?
1. I tend to think that the CCIF will probably find itself in situations (similar to last week) going forward where we'll need to be able to officially state our positions, and/or vote on other initiatives. Speaking for myself, the "committee" does not want to find itself in a position similar to that of last week again, and so I think we need a way to fairly decide our position on platforms or make decisions.
2. Reuven, Jesse et. al. may not be around forever, and I want the CCIF to be an institution that persists for much time to come. Cloud Computing here here to stay, IMO, and I want the CCIF to be here to stay as well. Therefore, a governmental structure that is stable and predictable enough to deal with future changes in leadership is best.
--
Jesse Silver
c: 310-766-2006
http://www.jesselsilver.com
twitter.com/silverguru
David Bernstein (daberns)
Guys
in order for companies to collaborate on anything, be it architecture, open source, proposed standards, or even marketing events, the concerns around IP rights, legalities around various DOJ type concerns, and public company SEC types of issues. This is often not on the radar by individuals or smaller companies but is a prerequisite to properly handle for large companies. Thus is why a trade or industry association is formed in many industries, so all these activities can occur. This association can incubate ideas, hand work over to standards bodies, coordinate/promote umbrella architectures, even be aneutral party to host interconnectivity and educational events. My company as well as other companies our size contribute to efforts like this where formal associations exist, but in cloud we do not hace such a formal association. I am one advocate of forming one to allow all companies large and small, and individuals to participate, this costs money to do, however usually memberships are skewed greatly towards large companies footing the bill although nomunal memberships in other categories are reasonable I think
Thus we can't just "write a document" without this kind of structure in place
Respecrfully,
David Bernstein
Cisco
-----Original Message-----
From: eprpa...@gmail.com [mailto:eprpa...@gmail.com]
Sent: Monday, March 30, 2009 08:33 AM Pacific Standard Time
To: cloud...@googlegroups.com
Subject: Re: Next Steps for CCIF / Cloud Forum
I also think we are jumping the gun here. Why would any organization
front money to a group that hasn't produced anything? Why not just
create the document, publish it and then form the organization?
Chuck Wegrzyn
Gary Mazz wrote:
> Sorry, forgot one; interoperability validation.
>
> -g
>
>
> Agreed..
>
> First we need to determine the purpose of ccif; advocacy, standards,
> marketing or discussion.
>
> -gary
>
Alexis Richardson
Why can't people license any doc under { GPLv3, IETF, W3C, CC-* } ?
alexis
Jesse L Silver
Though, my thought is that the CCIF could stay more community oriented than "trade organization" oriented, so the CCIF should probably count more on volunteer work and small financial contributions.
This leaves room for another, larger trade organization with a larger budget than CCIF should handle, that the CCIF partners with in some way, forming the fully open community arm of the industry.
David Bernstein (daberns)
Alexis
Its not the licensing post facto, its the development, creation, construction of the work. Any times big companies get together to agree stuff there are all kinds of laws that come into play. For example are we colluding to lock out competitors or are we tryimg to lower the barrier of entry for all? If we do the work with no umbrella strycture this can be interpreted as the former. If we do this under a trade assoc with a charter and open access it is the latter. Likewise if one company wants to contribute an idea that might already be patented by his company maybe unbeknownst to the contributor, another company does not want to be contaminated by even hearing the idea, they wat air cover that any ideas discussed while developping the specs are inherently licensed
Does rhis abswer you question?
-----Original Message-----
From: Alexis Richardson [mailto:alexis.r...@gmail.com]
Sent: Monday, March 30, 2009 10:41 AM Pacific Standard Time
To: cloud...@googlegroups.com
Subject: Re: Next Steps for CCIF / Cloud Forum
David,
Why can't people license any doc under { GPLv3, IETF, W3C, CC-* } ?
alexis
On Mon, Mar 30, 2009 at 6:29 PM, David Bernstein (daberns)
Mark A. Carlson
specific documents and software that get implemented and
adopted. You don't want someone to inject their IP into something
you end up shipping, then shaking you down for license fees or
even an injunction to stop shipping your infringing product/service.
This is why, when creating a new group, participating companies
spend an inordinate number of lawyer cycles ($$$), in addition to the
membership fees, working out an IP policy to everyone's satisfaction.
For an existing SDO, this has already been worked out. Groups like
OMG, OGF, DMTF and SNIA have nascent cloud activities starting up
right now. Shouldn't we wait to see where there are gaps that need
to be filled before saying there is a need for yet another standards
organization?
-- mark
 |
| Mark A. Carlson Sr. Architect Systems Group |
|
Phone x69559 / 303-223-6139 Email Mark.C...@Sun.COM |
 |
Jesse L Silver
Alexis Richardson
I don't see the need for a need org for this. How does the IETF work?
What about W3C? The latter has a membership structure that includes
a patent covenant. It is also perfectly possible to license on a
per-document basis using a license with patent covenants, eg GPLv3.
This elegantly delivers a decentralised model.
alexis
On Mon, Mar 30, 2009 at 7:16 PM, David Bernstein (daberns)
Alexis Richardson
eprpa...@gmail.com
structure and openness to make it all possible. I am just not sure we
need another organization given the number that already exist.
Chuck Wegrzyn
eprpa...@gmail.com
legislating no patented IP but instead make a declaration at the
beginning that all patented IP will be available for reasonable
licensing fees.
Chuck Wegrzyn
Mark A. Carlson wrote:
David Bernstein (daberns)
Mark, Alexis
Good point, actual standard should be developed inside existing standards orgs whenever possible
But overll how does cloid interoperability work? And what abot the educational and marketing and plug fest type of activities?
To cite a specific example, DMTF will assert that cloud interoperability is at the IaaS or Voicemail level, if we want PaaS interoperability, who is tackling that? Some of us think XMPP is important for the latter case, somaybe some extensions to IETF XMPP is required. But is and RDF/OWL ontology also needed? If so,who does that? W3C handles semantic web technology but not the cloud use case as far as I know.
Maybe we want to get them to do so, in which case first we have to work on a proposal together outside then take it to them
Thus we need the incubation vehicle
My thoughts ...
David
David Bernstein (daberns)
Guys
I am not pitching to overlap any standards body work. I am suggesting a trade assoc that can incubate, hand over, and also, do things outside of an individual standard, like overall architecture, plug fests, marketing/training, etc
David
-----Original Message-----
From: eprpa...@gmail.com [mailto:eprpa...@gmail.com]
Sent: Monday, March 30, 2009 11:31 AM Pacific Standard Time
To: cloud...@googlegroups.com
Subject: Re: Next Steps for CCIF / Cloud Forum
That is why the IETF works so well! Minimal overhead and enough
structure and openness to make it all possible. I am just not sure we
need another organization given the number that already exist.
Chuck Wegrzyn
David Bernstein (daberns) wrote:
> Alexis
>
> Its not the licensing post facto, its the development, creation,
> construction of the work. Any times big companies get together to agree
> stuff there are all kinds of laws that come into play. For example are
> we colluding to lock out competitors or are we tryimg to lower the
> barrier of entry for all? If we do the work with no umbrella strycture
> this can be interpreted as the former. If we do this under a trade assoc
> with a charter and open access it is the latter. Likewise if one company
> wants to contribute an idea that might already be patented by his
> company maybe unbeknownst to the contributor, another company does not
> want to be contaminated by even hearing the idea, they wat air cover
> that any ideas discussed while developping the specs are inherently licensed
>
> Does rhis abswer you question?
>
>
>
> -----Original Message-----
> From: Alexis Richardson [mailto:alexis.r...@gmail.com]
> Sent: Monday, March 30, 2009 10:41 AM Pacific Standard Time
> To: cloud...@googlegroups.com
> Subject: Re: Next Steps for CCIF / Cloud Forum
>
>
> David,
>
> Why can't people license any doc under { GPLv3, IETF, W3C, CC-* } ?
>
> alexis
>
>
> On Mon, Mar 30, 2009 at 6:29 PM, David Bernstein (daberns)
David Bernstein (daberns)
My PDA turned "v m" into voicemail sorry bout that
-----Original Message-----
From: David Bernstein (daberns)
Sent: Monday, March 30, 2009 11:36 AM Pacific Standard Time
To: cloud...@googlegroups.com; cloud...@googlegroups.com
Subject: Re: Next Steps for CCIF / Cloud Forum
Mark, Alexis
Good point, actual standard should be developed inside existing standards orgs whenever possible
But overll how does cloid interoperability work? And what abot the educational and marketing and plug fest type of activities?
To cite a specific example, DMTF will assert that cloud interoperability is at the IaaS or Voicemail level, if we want PaaS interoperability, who is tackling that? Some of us think XMPP is important for the latter case, somaybe some extensions to IETF XMPP is required. But is and RDF/OWL ontology also needed? If so,who does that? W3C handles semantic web technology but not the cloud use case as far as I know.
Maybe we want to get them to do so, in which case first we have to work on a proposal together outside then take it to them
Thus we need the incubation vehicle
My thoughts ...
David
-----Original Message-----
From: Mark A. Carlson [mailto:Mark.C...@Sun.COM]
Sent: Monday, March 30, 2009 11:18 AM Pacific Standard Time
To: cloud...@googlegroups.com
Subject: Re: Next Steps for CCIF / Cloud Forum
Jason Meiers
agree, if its not affordable then it doesnt make sense to suggest,
complement and support a process.
@david
it would also help to have a process for CCIF to add/suggest new
protocol supported. How can a group claim to be open when XMPP came
out of the blue and when there is no acceptance for other protocols?
On Mar 30, 8:40 pm, "David Bernstein (daberns)" <dabe...@cisco.com>
wrote:
>
> - Show quoted text -
eprpa...@gmail.com
framework staying away from specific technology or implementations. That
gives us something to discuss and a common language.
After that is done, then we can be more specific asking outside
organizations to participate and provide some substance. This will get
us the buy-in we need from others. Once you have momentum you will get
the likes of IBM, MS and others to participate; they won't want to be
left out.
Chuck Wegrzyn
David Bernstein (daberns) wrote:
eprpa...@gmail.com
protocol let alone anything at this point. I tend to think CCIF will be
difficult to describe let alone document without confusing the matters
with too many details.
Remember that sometimes you need to see the forest first before
describing the trees in it.
Chuck Wegrzyn
Alexis Richardson
releasing work for approval.
What is needed is a PURPOSE. That must come first. I propose "CCIF
is the voice of the user".
alexis
Matthew Zito
I don't think there was any real consensus _for_ XMPP, though some people had some good arguments as to why it was a good item to consider. I think it was more that you were pushing SEMP without any clearly reasoned argument as to why we should consider it.
Thanks,
Matt
Jesse L Silver
We do have enough of a purpose to form a government. I think we generally agree on these things (as laid out before by Reuven and others):
1. Advocacy
2. Promote outside standards efforts
3. Thinktank/incubator of ideas
eprpa...@gmail.com
Does that mean vendors which write the software and companies that host
the services will not be allowed? If so, the proposal will die. We have
to be more inclusive; every one, individual or company, that has an
interest in CCIF should be allowed to participate. Our goal should be to
make the best proposal that fits the goals of the many (remember the
80-20 rule). It should be broad enough to allow enlargement but specific
enough to work.
Chuck Wegrzyn
eprpa...@gmail.com
done so for many years.
Chuck Wegrzyn
>
> >
Alexis Richardson
could rub shoulders with vendors outside of a direct sales context.
I think "forum" is an appropriate concept.
Jesse L Silver
Good point. By 'the voice of the user', I meant a place where users
could rub shoulders with vendors outside of a direct sales context.
I think "forum" is an appropriate concept.
eprpa...@gmail.com
and take a vote. That also means that we throw the doors open for a
while to discuss the issue, and at a fixed time we take a vote. The
outcome of the vote is what we go with.
You have to remember that not everyone will be happy, so don't try. Just
give everyone a chance to be heard.
Chuck Wegrzyn
Jesse L Silver wrote:
>
> >
Reuven Cohen
Jesse L Silver
We agree then. All I'm saying is we need to lay out the specifics.
Geva Perry
On Mar 30, 12:06 pm, eprparad...@gmail.com wrote:
> I don't think we need a government we just need someone to take notes
> and take a vote. That also means that we throw the doors open for a
> while to discuss the issue, and at a fixed time we take a vote. The
> outcome of the vote is what we go with.
>
> You have to remember that not everyone will be happy, so don't try. Just
> give everyone a chance to be heard.
>
> Chuck Wegrzyn
>
> Jesse L Silver wrote:
> > The purpose has been laid out a number of times - the problem is we
> > can't agree on a final mission statement without a government.
>
> > We do have enough of a purpose to form a government. I think we
> > generally agree on these things (as laid out before by Reuven and others):
>
> > 1. Advocacy
> > 2. Promote outside standards efforts
> > 3. Thinktank/incubator of ideas
>
> > On Mon, Mar 30, 2009 at 2:50 PM, Alexis Richardson
> > <alexis.richard...@gmail.com <mailto:alexis.richard...@gmail.com>> wrote:
>
> > I humbly disagree. There are plenty of vehicles (and licenses) for
> > releasing work for approval.
>
> > What is needed is a PURPOSE. That must come first. I propose "CCIF
> > is the voice of the user".
>
> > alexis
>
> > On Mon, Mar 30, 2009 at 7:36 PM, David Bernstein (daberns)
> > <mailto:eprparad...@gmail.com <mailto:eprparad...@gmail.com>>
> > >> [mailto:eprparad...@gmail.com <mailto:eprparad...@gmail.com>]
>
> read more »
Alexis Richardson
Sam Johnston
I am not pitching to overlap any standards body work. I am suggesting a trade assoc that can incubate, hand over, and also, do things outside of an individual standard, like overall architecture, plug fests, marketing/training, etc
If I understand well, all $bigco needs is a contract like this one (dealing with administrative housekeeping such as rights & obligations, use of names, IPR, etc.) and a legal entity to execute it with.
Without an office (and bodies to put in it) the total annual costs for a 503(c)(3) is in the hundreds or low thousands of dollars range - which would be covered by the first few sponsors.
Sam
eprpa...@gmail.com
vote. To participate means you listen to the email threads and when the
time comes you pipe up.
You'd be surprised how well it works. So long as you don't take it upon
yourself to want to keep everyone happy.
Chuck
Sam Johnston
The specifics are a precondition of actual agreement IMHO.
+1
Sam Johnston
> How do you determine who gets a vote?
The rules for the IETF and IRTF is that anyone that participates gets a
vote. To participate means you listen to the email threads and when the
time comes you pipe up.
You'd be surprised how well it works. So long as you don't take it upon
yourself to want to keep everyone happy.
+1
Most decisions will naturally be consensus-based anyway (which elegantly avoids the problem) but those which need a vote can easy enough be limited to participants by sending voting links to each of the participants.
Sam
eprpa...@gmail.com
I think you are right. There is a structure already in place so this
seems like it should be part of that effort.
Chuck Wegrzyn
Sam Johnston wrote:
> On Mon, Mar 30, 2009 at 8:39 PM, David Bernstein (daberns)
> <dab...@cisco.com <mailto:dab...@cisco.com>> wrote:
>
> I am not pitching to overlap any standards body work. I am
> suggesting a trade assoc that can incubate, hand over, and also, do
> things outside of an individual standard, like overall architecture,
> plug fests, marketing/training, etc
>
> How would this differ from the Open Cloud Consortium (OCC)
> member <http://www.opencloudconsortium.org/members.html>?
> If I understand well, all $bigco needs is a contract like this one
eprpa...@gmail.com
only seen it screw up once (that was on the SNMPv2 committee). What
happened that time was we came up with two specs! It took a while but
what happened was both sides built implementations and we figured out
what sucked. V3 was a better product for it.
Chuck Wegrzyn
Sam Johnston wrote:
Geva Perry
there are 30 IBM employees involved, do they each get a vote?
eprpa...@gmail.com
Chuck Wegrzyn
Jason Meiers
>
> read more »- Hide quoted text -
Jason Meiers
multiple email handles and improve the quality of the results of a
vote.
Sam Johnston
a small fee i.e 2 bucks a month or year depending will help clean
multiple email handles and improve the quality of the results of a
vote.
+1 Alexis for "voice of the users".
Sam
Jason Meiers
vote count me in. if not, there is a good chance of an inacurate vote.
On Mar 30, 9:51 pm, Sam Johnston <s...@samj.net> wrote:
eprpa...@gmail.com
Jason Meiers
decide what the consensus agrees upon? that would be inappropritate.
actual numbers helps.
adhi
I think if you use a simple PHP poll module in the web site and a
fixed time slot for voting will automatically eliminate multiple/
duplicate votes by IP ! With so many geeks around it should not be a
problem at all !
cheers
Varadarajan
Eric Windisch
On Mar 30, 2009, at 3:54 PM, Jason Meiers wrote:
>
> If there is another option to keep multiple email handles out of the
> vote count me in. if not, there is a good chance of an inacurate vote.
This might be seen as discriminatory, but... We could create a trust
ring where members could sign keys for other members that attend
meatspace events. While some might argue that locking out those that
cannot travel to these events is discrimination, it would make sure
that those with a vote are real people and have enough of an interest
in cloud standards to actually show up to an event. If someone tries
to defraud the system, they would have to do it in physical space, or
by abusing their own signing privileges (which can be traced). If
members signed for "friends online", instead of those in physical
space, if it was done in any significant numbers, it would probably
come back to bite them. It would be more secure if the signing was
done through officers, but that would bring its own bed of nails.
Meatspace events would be CCIF meetings, CloudCamp, dinner with
friends, etc.
--
Eric Windisch
Eric Windisch
You're talking about "cloud" people. Getting new IP addresses isn't
difficult. Bring up a few hundred nodes, send out your votes, and
bring down the instances. There are plenty of other ways too (public
proxies, private RIR allocations, etc)
--
Eric Windisch
Sam Johnston
This might be seen as discriminatory, but...
On Mar 30, 2009, at 3:54 PM, Jason Meiers wrote:
> If there is another option to keep multiple email handles out of the
> vote count me in. if not, there is a good chance of an inacurate vote.
You're going to say it anyway ;)
We could create a trust
ring where members could sign keys for other members that attend
meatspace events. While some might argue that locking out those that
cannot travel to these events is discrimination, it would make sure
that those with a vote are real people and have enough of an interest
in cloud standards to actually show up to an event.
By your theory *I* don't have "enough of an interest in cloud standards" as I've not been to a single event and don't currently plan to.
Meatspace events would be CCIF meetings, CloudCamp, dinner with
friends, etc.
In reality it's a non-problem (as has already been explained) and besides, there are a million and one ways to skin this particular cat.
Sam
Jason Meiers
since alot of companies go out to the net with a single ip. some small
businesses with 5 employes may only cast one vote by ip.
i think the physical location thing is a good idea only its quite
pricy to travel for alot of folks. just my 2 cents. I am not traveling
to often these days either. but hey if it works i am for it.
are there other options online to authenticate a person online?
biometric fingerprint reader?
eprpa...@gmail.com
matters; it is consensus. That means there is a fixed amount of time to
discuss a topic, usually. The group will almost always come to a
consensus at which point everything is cool.
Chuck Wegrzyn
gary mazzaferro
Jason Meiers
interoperability for clouds.
for example if Azure creates a standard for event management and Sun
another for event management. Outputs of CCIF are advocacy, best
practices, marketing, discussion for interoperability of the two
protocols into cloud computing.
@chuck
how about consensus of two views/groups for each issue within CCIF for
instance from the open group and commercial group. a perspective from
both sides at CCIF will support the validation and credibility of CCIF
and find a home for the majority of the participants. the dual
consensus can include an agreement on an issue, opposing or extention.
at least then all bases are covered.
hope this helps.
Jason
Bechauf, Michael
I'm new here, so just want to introduce myself. My name is Michael Bechauf, I'm responsible for standards at SAP. We have joined in signing the Open Cloud Manifesto. As you may know, we are an applications vendor with our own development platform that is partly based on proprietary technology, but has in the last couple of years also opened up to standards like Java and SOA. We've got our own WS* implementation and Java EE application server.
For the almost 20 years that we have offered non-mainframe based application products to our customers, they have been asking for availability and freedom of choice in terms of databases and operating systems. I tink the jury is still out there in terms of growth projections of Cloud Computing, but I don't think there is hardly anybody who will completely reject the notion of flexible computing power available on demand within either a 'private' environment or the public cloud. So, for lack of simplicity let's just look at Cloud as the future operating environment that application systems depend on.
We are spending a significant amount of money on porting our solutions to the different hardware/operating system/database combinations, which is money that we would rather invest in business process innovations in our applications. Customers want choice, so the only way to make the porting effort more cost effective is through strong standards.
Which brings me - after this long intro - to the point: I have lived through the WS* battles, and was for a while responsible of chairing the Web Services Interoperability Organization (WS-I). The issue is that as soon as big corporate interests become involved, a consensus that is not in line with any particular company's design can cost millions.
I can't claim to be an expert on Cloud standards specifically, but I think I have a fair grasp on the number of standards that may be necessary to develop a profile that gives a reasonable amount of certainty that a piece of application code can run on the customer's cloud of choice. I don't think there need to be new standards; I think similar to WS-I there need to be cloud "profiles".
Now I hear a lot of people saying: No, we don't want WS* all over. But seriously: Do you know why it was/still is so difficult to make progess in WS* ? Because the consensus and even the rules are creatively utilized. For example: Number of votes. Does the number of invidiuals count, or does one company one have one vote ? These may sound like trivial questions, but working groups to change/adapt the rules take forever. In the case of WS-I, a vote requires a supermajority and two participants can block progress.
Somewhere I saw the idea tossed around to develop agreements how companies work together on a Wiki. Now, how is that supposed to work ? As soon as one participant changes a specification published on a Wiki, others need to change their design ? Obviously, it does not work that way, and rules and procedures are required.
The bottom line is that everything changes as soon as commercial interests become involved. If we are all individual developers who do the right thing and aspire to work on the best design on earth, and we quickly come together to hash out details, then it's perfectly ok to synchronize design over a Wiki. Everything else becomes more formal.
In the end, whether a particular group has momentum depends on the number of stakeholders that you can bring together, including users, vendors, independent developers, ISVs etc. That, largely, depends whether a critical mass of participants will agree on a specifc set of IP terms and procedural rules. Then there is money to have meetings and to do some minimal organizational marketing, but that's (almost) secondary.
A final point on the Open Cloud Manifesto: Yes, this could have probably been handled much better, but I still think the principles are right and important. It's easy to shoot at a document by declaring the process was not "open". From a perspective of an individual working without the context of commercial interests I completely understand, and it's not surprising that people were angry and felt let down. But I predict that the truth about those with commercial interests who argue for more openness will be revealed once rules and processes will be discussed. At that point each vendor will insist on specific rules which all of a sudden may feel less "open".
Sorry for this long mail, but I just subscribed to the CCIF list yesterday, and have been getting a stream of emails all morning that I wanted to respond to, but was stuck in meetings.
Best,
Michael
-----Original Message-----
From: cloud...@googlegroups.com [mailto:cloud...@googlegroups.com] On Behalf Of Jason Meiers
Sent: Monday, March 30, 2009 1:05 PM
To: Cloud Computing Interoperability Forum (CCIF)
Subject: Re: Next Steps for CCIF / Cloud Forum
eprpa...@gmail.com
Chuck Wegrzyn
Bechauf, Michael
Jason Meiers
The current issue is what the output of the CCIF is and how it is
generated. The document today didnt represent alot of the participants
from cloud computing interoperability. To prepare a proper concensus
for issues resolved by CCIF it is recommended to be includes two camps
one open camp and one commercial camp. The goals and mission statement
for a CCIF commercial group will be included in a different document
than the one created today. The commercial group mission statement is
still to be generated.
I would like to lead the commercial group and generate a document with
the mission statement for commercial interests. If there are people or
organzation interested in participating and join CCIF commercial I'd
be greatful for the support.
Best Regards,
Jason
On Mar 31, 12:10 am, "Bechauf, Michael" <michael.bech...@sap.com>
wrote:
>
>
>
> -----Original Message-----
> From: cloud...@googlegroups.com [mailto:cloud...@googlegroups.com]
>
afalcon
future structure of CCIF. Is it a "Forum", an unstructured or
structured user group, or an "Association"? Should it be minimalist
in structure or more formal?
Will advocacy be only to vendors or to the government as well? If the
latter, how formal will the lobbying efforts be?
Are we publishing white papers and manisfestos, or do we need/want
LCNs and ISBNs?
If left unstructured, how do we select spokespersons? Present
consensus and dissent in an organized manner?
In the last week or so, the CCIF has paid a price for a lack of
structure and facilitation. How do we prevent this moving forward?
I don't see CCIF becoming a standards body, but working with them. If
the leadership is vendor-only, no matter how altruistic, will users
join and participate?
I'm impressed that we're already arguing about company versus
individual votes and how to validate votes. But, what will be voting
on?
Allen
On Mar 30, 10:38 am, afalcon <afal...@horizoninfoservices.com> wrote:
> If we are serious about facilitating open standards for cloud
> computing, than we need to formalize quickly. I suggest the
> following.
>
> 1) Establish a 501(c)(6) non-profit (membership association)
> 2) Create an "office" to manage the association's assets (domains,
> groups, etc.) and membership, and to identify facilitators for the
> process.
>
> To do this, we will need to identify an initial board of directors (at
> least 5), contract with an individual or firm to provide the office
> resources, and secure funding.
>
> The funding is best raised through annual membership dues. Based on
> operation of the office. Not difficult with enough vendors and dues
> in the $5K to $10K range.
>
> The person or firm hosting the office should not be a vendor with a
> direct interest in the resulting standards, and no member of the firm
> should serve as a director.
>
> Creating a charter sufficient to obtain 501(c)(6) status is not
> difficult. We can then develop a more elaborate charter and set of by-
> laws.
>
Bechauf, Michael
-Michael
Gary Mazz
So, advocacy and marketing. Does the user community have a role ?
-gary
Jason Meiers
industry perspective based on commercial requriements for cloud
computing. The CCIF commerce will not support an open source, open or
simular agenda.
An example for this would be to add value to the consensus with
commercial cloud computing requirements.
Gary Mazz
for plug fests, validation suites and an interoperability jig/simulator ?
-g
Jason Meiers
in interoperability processes. The created interfaces must belong to
the author and not required to share or open source the tools,
processes, software, SaaS or PaaS. An example for this is to build an
interoperability component within the enterprise and deliver this as
an actual product to the market.
Jason
Monitoring-as-a-Service(TM)
Michael Richardson
>>>>> "eprparadocs" == eprparadocs <eprpa...@gmail.com> writes:
eprparadocs> That's somewhat true. Personally I think the right
eprparadocs> approach isn't about legislating no patented IP but
eprparadocs> instead make a declaration at the beginning that all
eprparadocs> patented IP will be available for reasonable
eprparadocs> licensing fees.
No. RAND kills open source projects.
Open source requires RF on patents, not RAND.
--
] Y'avait une poule de jammé dans l'muffler!!!!!!!!! | firewalls [
] Michael Richardson, Sandelman Software Works, Ottawa, ON |net architect[
] m...@sandelman.ottawa.on.ca http://www.sandelman.ottawa.on.ca/ |device driver[
] panic("Just another Debian GNU/Linux using, kernel hacking, security guy"); [
Michael Richardson
>>>>> "Eric" == Eric Windisch <er...@grokthis.net> writes:
>> I think if you use a simple PHP poll module in the web site and a
>> fixed time slot for voting will automatically eliminate multiple/
>> duplicate votes by IP ! With so many geeks around it should not be a
>> problem at all !
Eric> You're talking about "cloud" people. Getting new IP addresses
Eric> isn'tdifficult. Bring up a few hundred nodes, send out your
Eric> votes, and
Yes, exactly. Why doesn't this happen to the IETF?
It's simple: it's because such phantom votes do not get counted as an
ordinal, but rather as a movement. 5 well written posts have far more
impact than 100 terse ones that say the same thing over and over again.