Module Proposal: Mozilla Communities Web Services
I have been working with Tom Farrow (tad) and some other community members to put together a proposal for better delivering Web Services to Communities.This would restructure the delivery of services like hosting, domains and emails which are currently provided to communities. Creating a module would provide clear authority around these services, as well as allowing communities to be equal stakeholders in what services are delivered, and how. The module also creates accountability on the part of communities which will allow for better communication between resource owners and the groups that provide them (eg Community Ops), faster problem solving (eg downtime recovery), and impact (better supported sites are better quality sites). This document is what we think is realistic plan for implementing the module. https://docs.google.com/a/mozilla-community.org/document/d/1zXuNp8dwyLOW-UqZ4FqcRVxhszHWYqgjClDKZ5kYrTQ/edit?usp=sharing This document is a our vision of what the module could become once fully implemented: https://docs.google.com/document/d/1YdNWOy7QNc1xXcnHGN5IW4HhRgRJCjOdKMu_gHFjUM4 We are proposing that the module be called Mozilla Communities Web Services, that Tom Farrow be the module owner, and that myself and Michael Kohler be peers. Suggestions for additional peers are welcome, but the intention for now is to add people as they take on leadership roles. Questions and suggestions are of course welcomed. ___ governance mailing list governance@lists.mozilla.org https://lists.mozilla.org/listinfo/governance
Re: Module Proposal: Mozilla Communities Web Services
I'd encourage that a peer be from the Mozilla InfoSec Team to audit and vet the accesses to user data that volunteers are being given access too. If a security issue arose on any of these community hosted platforms Mozilla would take a PR hit for not having vetted the processes and access to user data. On Tue, Feb 16, 2016 at 8:49 AM Majken Connor wrote: > I have been working with Tom Farrow (tad) and some other community members > to put together a proposal for better delivering Web Services to > Communities.This would restructure the delivery of services like hosting, > domains and emails which are currently provided to communities. Creating a > module would provide clear authority around these services, as well as > allowing communities to be equal stakeholders in what services are > delivered, and how. > > The module also creates accountability on the part of communities which > will allow for better communication between resource owners and the groups > that provide them (eg Community Ops), faster problem solving (eg downtime > recovery), and impact (better supported sites are better quality sites). > > This document is what we think is realistic plan for implementing the > module. > > https://docs.google.com/a/mozilla-community.org/document/d/1zXuNp8dwyLOW-UqZ4FqcRVxhszHWYqgjClDKZ5kYrTQ/edit?usp=sharing > > This document is a our vision of what the module could become once fully > implemented: > > https://docs.google.com/document/d/1YdNWOy7QNc1xXcnHGN5IW4HhRgRJCjOdKMu_gHFjUM4 > > We are proposing that the module be called Mozilla Communities Web > Services, that Tom Farrow be the module owner, and that myself and Michael > Kohler be peers. Suggestions for additional peers are welcome, but the > intention for now is to add people as they take on leadership roles. > > Questions and suggestions are of course welcomed. > ___ > governance mailing list > governance@lists.mozilla.org > https://lists.mozilla.org/listinfo/governance > ___ governance mailing list governance@lists.mozilla.org https://lists.mozilla.org/listinfo/governance
Re: Module Proposal: Mozilla Communities Web Services
Good to see a proposal. I think it'd be good to articulate a bit more of the plan and its purpose, before diving into the details of module structure, which feel a bit premature to me. In particular, I'd like to understand: - what kinds of websites does Mozilla offer / underwrite, and for what purpose? Same for domains and email services. - how does someone ask for such services, what are the criteria we use, etc? - why does it make sense for Mozilla to provide these services on hardware we control (virtualized or not) as opposed to underwriting/managing hosted services run by people who do this at scale? - to bkerensa's point, how do we mitigate the various risks associated with hosting content on behalf of someone else - how does this relate with other community-facing and community-managing activities across Mozilla, especially those that are focused on local and regional -- Reps, Mozilla Clubs, campus campaigns, etc. I'd also like to understand a bit of a roadmap -- what's the state of affairs now, where are we trying to go, by when, etc. --david ___ governance mailing list governance@lists.mozilla.org https://lists.mozilla.org/listinfo/governance
Re: Module Proposal: Mozilla Communities Web Services
I'll take a first go at answering these questions, but Pierros or Tom might have important info to add: On Tue, Feb 16, 2016 at 12:25 PM, David Ascher wrote: > Good to see a proposal. > > I think it'd be good to articulate a bit more of the plan and its purpose, > before diving into the details of module structure, which feel a bit > premature to me. > > In particular, I'd like to understand: > There is some context that applies to many of these questions in that several years Mozilla decided to support community sites with resources, both financial and technical. The group providing the technical resources is almost entirely volunteers (Community IT which became Community Ops). The idea was not to obtain any *control* over the sites, though centralizing some resources, like domains, creates a much better experience. We've been providing these services under the umbrella of Community Ops, even though they aren't all Ops tasks. That is one of the issues that would be solved through the creation of a module. > > - what kinds of websites does Mozilla offer / underwrite, and for what > purpose? Same for domains and email services. > Mostly regional community sites, though we have also managed project related sites, like the Firefox 10 site and the community Discourse. This proposal is meant to cover the decentralized community sites. Experiments like Discourse fit under the new Participation Software Lab (ParSoL). > - how does someone ask for such services, what are the criteria we use, > etc? > Currently it's a Reps request which is reviewed by at least one member of Reps council. This was because Reps better know the communities and could make sure the request is coming from the right place. The criteria have evolved a bit, but generally they have tried to make sure the resource has a reliable owner (Rep or staff because of official commitment to Mozilla) and an existing project - the community has needed to exist for 6 months. So a single person can't say hey, I want to *start* a community, give me resources. Part of the purpose of this module is that we'd have the mandate to review this process and improve it. Reps Council is too busy. > - why does it make sense for Mozilla to provide these services on hardware > we control (virtualized or not) as opposed to underwriting/managing hosted > services run by people who do this at scale? > Not all communities have the resources to run their own website. Many communities are only running a WordPress blog. Also many communities are much more idealistic and don't trust these other organizations as much as they trust Mozilla. Also, and this is a big deal to us, open can't just be about writing websites. To keep it open, people need to also be able to publish. If we provide these services on resources that we manage, we are providing our volunteers with a place to develop skills necessary to maintaining an open internet. At least two of us have gotten jobs out of our contributions to Community Ops. > - to bkerensa's point, how do we mitigate the various risks associated > with hosting content on behalf of someone else > This is a good question. We're doing it already, I'm not sure what difference a module would make. I don't know if it makes a difference that this is a volunteer managed project. However, supporting this group of people who wish to develop these skills would provide more oversight than exists now. We're Mozillians, we care about privacy and we want to do it right. As you can see from the proposal we do want support from different functional areas to provide expertise so that these sites are effective and kept to a high standard. > - how does this relate with other community-facing and community-managing > activities across Mozilla, especially those that are focused on local and > regional -- Reps, Mozilla Clubs, campus campaigns, etc. > I suppose the same way that Reps relates to Participation Infrastructure. This is specifically about providing the web services to these groups. > > I'd also like to understand a bit of a roadmap -- what's the state of > affairs now, where are we trying to go, by when, etc. > Short version is, again creating a module gives us a mandate to answer these questions. I think the only other alternative is to decide we're not providing these resources at all. However if they're being provided then we need clear authority and accountability to do a good job. Happy to provide a longer version with specifics, but I have a meeting now and want to send this off. > > --david > > ___ governance mailing list governance@lists.mozilla.org https://lists.mozilla.org/listinfo/governance
Re: Module Proposal: Mozilla Communities Web Services
Thanks Kensie -- I wholeheartedly support the general idea of trying to bring some cohesion to the systems that still allow decentralized expression by volunteers of Mozilla in their local context. We can and should make it easier for people to create web presences and collaborative spaces where they can communicate with each other, publish, build software, advocate, etc. Ideally this happens in a way that combines local customization and deep localization while neither requiring that everyone develop all of the skills needed to make world-class websites, and while providing coordination support so that activities in one location are effectively cut-off from activities elsewhere. Ideally this is also a system that makes it easy for there to be multi-way coordination between staff and volunteers, between locations, between people interested in a topic regardless of location or staff status. If this is the kind of thing you're talking about, I'm 100% in support, and I think the scope is broader than just technical website support. I suggest we separate the skills training benefits from the service definition and delivery models. The "what" of the service should be determined by a crisp analysis of what community activities make sense to encourage & facilitate -- it could be websites, it could be other things -- and the skills needed to implement those could range from design frameworks to security auditing, ops or even social media management and marketing. Regardless, I'm sure there will be opportunities for skills development to happen as part of service delivery. With that frame, it feels to me like defining a module isn't the obvious next step. I'd be more keen in an approach that, without needing a priori "authority", gathers a set of stakeholders who can articulate a vision & plan, identify "business needs" (including the needs of the local communities), and deliberately ignores how things are happening today. We can then map those needs to systems that we have and may want to adapt (including the current community websites and community ops, and the systems that IT are currently providing), or systems we need to build anew. I'm happy to help. --david ___ governance mailing list governance@lists.mozilla.org https://lists.mozilla.org/listinfo/governance
Re: Module Proposal: Mozilla Communities Web Services
On Tue, Feb 16, 2016 at 4:04 PM, David Ascher wrote: > Thanks Kensie -- > > I wholeheartedly support the general idea of trying to bring some cohesion > to the systems that still allow decentralized expression by volunteers of > Mozilla in their local context. We can and should make it easier for > people to create web presences and collaborative spaces where they can > communicate with each other, publish, build software, advocate, etc. > > Ideally this happens in a way that combines local customization and deep > localization while neither requiring that everyone develop all of the > skills needed to make world-class websites, and while providing > coordination support so that activities in one location are effectively > cut-off from activities elsewhere. > > Ideally this is also a system that makes it easy for there to be multi-way > coordination between staff and volunteers, between locations, between > people interested in a topic regardless of location or staff status. > > If this is the kind of thing you're talking about, I'm 100% in support, > and I think the scope is broader than just technical website support. > > I suggest we separate the skills training benefits from the service > definition and delivery models. The "what" of the service should be > determined by a crisp analysis of what community activities make sense to > encourage & facilitate -- it could be websites, it could be other things -- > and the skills needed to implement those could range from design frameworks > to security auditing, ops or even social media management and marketing. > Regardless, I'm sure there will be opportunities for skills development to > happen as part of service delivery. > > With that frame, it feels to me like defining a module isn't the obvious > next step. I'd be more keen in an approach that, without needing a priori > "authority", gathers a set of stakeholders who can articulate a vision & > plan, identify "business needs" (including the needs of the local > communities), and deliberately ignores how things are happening today. We > can then map those needs to systems that we have and may want to adapt > (including the current community websites and community ops, and the > systems that IT are currently providing), or systems we need to build anew. > > I'm happy to help. > > --david > > I have some objections here. First of all is the scope creep. It would be great to do more, but just because it would be great to do more doesn't mean that all of it should be done as a whole. We have real resources that we're providing to real communities that frankly have been provided very poorly for the past year because of the lack of structure and authority. I don't see why this module can't exist, and then be swallowed up by a larger module as such a program expands. As it is the scope here is pretty broad. I'm sure everyone reading this understands that more than webdev goes into a good website. We have included things like security audits, and branding updates in our proposals. Branding tasks are left off of the initial roadmap because consulting with Branding, they're in the midst of some reorganizing and can't at the moment tell us what sorts of standards they would suggest for sites. Also, a priori authority is necessary. Especially with volunteers. A staff member has a place in the org chart and the authority structure is built in, even if it's not explicit for a specific project. This is lacking for volunteers and as I said, has been blocking those of us managing the delivery of these resources from doing the kind of job we'd be proud of. If you have a suggestion for providing us the authority to do this planning work besides a module that would be worth considering. We can redesign what services should happen, but we also can't ignore the ones that already exist. One of the first tasks we plan to undertake is documenting a definitive list of the stakeholders on the communities side, necessary work if we're going to revisit entirely what value decentralized community sites provide, and how they should exist. Basically it feels like this - we've been providing a product, with an entirely volunteer team, and we're asking to officially be made owners of the product, and you're suggesting that before that happens, the product should be entirely revisited and reevaluated. Can we not do that as owners of the product? I would be interested in discussing this more and better understanding what you have in mind, but we'd also want assurances that formulating a new approach will actually progress, and that we'd be given real authority in these discussions, and to improve the services that are currently being provided until such a time as they're being replaced. ___ governance mailing list governance@lists.mozilla.org https://lists.mozilla.org/listinfo/governance
Re: Module Proposal: Mozilla Communities Web Services
I may be missing some background. Can you explain what you would do as module owners that you can't do today, and/or whose code or actions you seek to influence as module owner? On Tue, Feb 16, 2016, 7:44 PM Majken Connor wrote: > > > On Tue, Feb 16, 2016 at 4:04 PM, David Ascher > wrote: > >> Thanks Kensie -- >> >> I wholeheartedly support the general idea of trying to bring some >> cohesion to the systems that still allow decentralized expression by >> volunteers of Mozilla in their local context. We can and should make it >> easier for people to create web presences and collaborative spaces where >> they can communicate with each other, publish, build software, advocate, >> etc. >> >> Ideally this happens in a way that combines local customization and deep >> localization while neither requiring that everyone develop all of the >> skills needed to make world-class websites, and while providing >> coordination support so that activities in one location are effectively >> cut-off from activities elsewhere. >> >> Ideally this is also a system that makes it easy for there to be >> multi-way coordination between staff and volunteers, between locations, >> between people interested in a topic regardless of location or staff status. >> >> If this is the kind of thing you're talking about, I'm 100% in support, >> and I think the scope is broader than just technical website support. >> >> I suggest we separate the skills training benefits from the service >> definition and delivery models. The "what" of the service should be >> determined by a crisp analysis of what community activities make sense to >> encourage & facilitate -- it could be websites, it could be other things -- >> and the skills needed to implement those could range from design frameworks >> to security auditing, ops or even social media management and marketing. >> Regardless, I'm sure there will be opportunities for skills development to >> happen as part of service delivery. >> >> With that frame, it feels to me like defining a module isn't the obvious >> next step. I'd be more keen in an approach that, without needing a priori >> "authority", gathers a set of stakeholders who can articulate a vision & >> plan, identify "business needs" (including the needs of the local >> communities), and deliberately ignores how things are happening today. We >> can then map those needs to systems that we have and may want to adapt >> (including the current community websites and community ops, and the >> systems that IT are currently providing), or systems we need to build anew. >> >> I'm happy to help. >> >> --david >> >> > I have some objections here. First of all is the scope creep. It would be > great to do more, but just because it would be great to do more doesn't > mean that all of it should be done as a whole. We have real resources that > we're providing to real communities that frankly have been provided very > poorly for the past year because of the lack of structure and authority. I > don't see why this module can't exist, and then be swallowed up by a larger > module as such a program expands. As it is the scope here is pretty broad. > I'm sure everyone reading this understands that more than webdev goes into > a good website. We have included things like security audits, and branding > updates in our proposals. Branding tasks are left off of the initial > roadmap because consulting with Branding, they're in the midst of some > reorganizing and can't at the moment tell us what sorts of standards they > would suggest for sites. > > Also, a priori authority is necessary. Especially with volunteers. A staff > member has a place in the org chart and the authority structure is built > in, even if it's not explicit for a specific project. This is lacking for > volunteers and as I said, has been blocking those of us managing the > delivery of these resources from doing the kind of job we'd be proud of. If > you have a suggestion for providing us the authority to do this planning > work besides a module that would be worth considering. We can redesign what > services should happen, but we also can't ignore the ones that already > exist. One of the first tasks we plan to undertake is documenting a > definitive list of the stakeholders on the communities side, necessary work > if we're going to revisit entirely what value decentralized community sites > provide, and how they should exist. > > Basically it feels like this - we've been providing a product, with an > entirely volunteer team, and we're asking to officially be made owners of > the product, and you're suggesting that before that happens, the product > should be entirely revisited and reevaluated. Can we not do that as owners > of the product? > > I would be interested in discussing this more and better understanding > what you have in mind, but we'd also want assurances that formulating a new > approach will actually progress, and that we'd be given real authority in > these discussio
Re: Module Proposal: Mozilla Communities Web Services
I think approving a module for Mozilla Communities Web Services would be a great low-cost way to support the community sites that we're already providing resources for. That this is just writing down a process that's already happening is a good sign for it's viability, I think. I also really like the proposal to start with a single module and experiment with what structure works for your purposes. We should totally do this! - Mike Kelly On 2/16/16 8:41 AM, Majken Connor wrote: I have been working with Tom Farrow (tad) and some other community members to put together a proposal for better delivering Web Services to Communities.This would restructure the delivery of services like hosting, domains and emails which are currently provided to communities. Creating a module would provide clear authority around these services, as well as allowing communities to be equal stakeholders in what services are delivered, and how. The module also creates accountability on the part of communities which will allow for better communication between resource owners and the groups that provide them (eg Community Ops), faster problem solving (eg downtime recovery), and impact (better supported sites are better quality sites). This document is what we think is realistic plan for implementing the module. https://docs.google.com/a/mozilla-community.org/document/d/1zXuNp8dwyLOW-UqZ4FqcRVxhszHWYqgjClDKZ5kYrTQ/edit?usp=sharing This document is a our vision of what the module could become once fully implemented: https://docs.google.com/document/d/1YdNWOy7QNc1xXcnHGN5IW4HhRgRJCjOdKMu_gHFjUM4 We are proposing that the module be called Mozilla Communities Web Services, that Tom Farrow be the module owner, and that myself and Michael Kohler be peers. Suggestions for additional peers are welcome, but the intention for now is to add people as they take on leadership roles. Questions and suggestions are of course welcomed. ___ governance mailing list governance@lists.mozilla.org https://lists.mozilla.org/listinfo/governance ___ governance mailing list governance@lists.mozilla.org https://lists.mozilla.org/listinfo/governance
