Grow Food Party Crew - Ventura
Join us in creating edible abundance all over town!
The Grow Food Party Crew is a network of people who create edible abundance by throwing parties in neighborhoods all over town. We are committed to everyone having gardens that provide for their nourishment, and building a resilient and friendly culture in which we take care of each other. ...learn more
GROUP DETAILS
Created: Jun 06, 2008
Updated: Aug 10, 2009
Membership: Open
Semi-Private
"Ubergeek Week"
Access Privileges
Or: A cooperative route to some technical recommendations for the Transition Web Project
A focussed time-bound session ending on at 5pm GMT May 13 with the goal of developing and clarifying some technical recommendations for the work that will enable this project to move forward in a robust, resilient and sustainable manner.
(A note about the name:
On the subject of the name “Ubergeek Week” this is not just for the technically gifted code-monkeys among us, dubbed 'ubergeeks' (by Ben I think - yes, blame him not me). It is open to all, but we are focusing on people with a deeper level of web tech know-how cos we think they may have more informed contributions to offer. Therefore I make no apologies for the jargon on this page. Where appropriate I may even offer an explanation.)
This is about hosting, about web applications, site development frameworks, open source, etc.
We’ve got lots of useful, valuable and interesting input already from discussions on this site, the previous Ning site (http://transitionwebproject.ning.com/), the Transition Network forums (http://transitiontowns.org/forum/forum.php?id=30 and http://transitiontowns.org/forum/forum.php?id=23), and so on.
The information on this wiki page aims to outline and act as a guide to what we have been talking about, boiling the hours of discussion into some focused and handleable information to enable you to constructively criticise and work cooperatively with others to jointly forge this page into a sparkling gem setting out some clear and elegant recommendations that we can all be proud to stand behind. To spur this on I’ll aim to include some stuff designed to stir some debate and get some thinking going and thereby encourage a bit of innovation. I don't pretend to know all about all of this stuff. I certainly don't know all the answers, which is why we are looking for your contributions.
How to Contribute
Edit this wiki page. We will get to where we want to be if people are bold and brave and edit the wiki page. Don’t worry about over-writing older stuff. It all gets kept in the revisions history (I hope) so we can always back-pedal. Please edit and add to the wiki page. Please. If you don’t edit the wiki page I’ll have to do it all myself, which won’t be half as much fun, and we won’t advance as quickly or as far as we could do.
Comments: Using the comment/discussion thread below the page is only useful if you want to discuss the stuff on the page rather than build it. If you are unsure, edit the page. I can always edit it right back again.
While this page will remain up on the site after the closing date for submissions, I can’t guarantee that anything added after that time will make it into the final recommendations. So please don’t wait until the last minute.
I will be aiming to add to and edit this page on a daily (at least) basis during this process, so please come back frequently to pitch in with your ideas.
So, here goes:
First off, some underpinning thoughts, guidelines and principles:
What this project is not:
- This is not a plan to build some monolithic all-encompassing mega-portal that aims to be all things to all people, and be the global transition website. We know that this way lies madness, and most likely costly failure.
- This is not attempt to force all transitioners to adopt the web as their primary means of communicating. We know that what goes on in the real world of meetings, conferences gardens, allotments, through phone calls and via noticeboards is about a million times more important than what can be achieved online. However we do think that online can add some really useful stuff to make the real world doings even more effective.
- This not an attempt to impose any particular paradigm or pressure transition initiatives to adopt particular ways of doing things online.
What this project is:
- This is a recognition and a celebration of the diversity of skills, tools, platforms and approaches already in use by transition initiatives across the movement.
- This is an attempt to add value to what is already going on through the development and use of aggregation/syndication technologies, semantic search tools, widgets, modules, plug-ins, and the exploitation and use of various third party services that can do neat and useful things.
- This is about providing some useful web infrastructure to support and facilitate networking at various geographical levels, a la Transition East (http://transitioneast.net), and offering opportunities for local initiatives to get online quickly, easily and cheaply, using a supportive and relevant framework.
- This is about offering a service that aims to facilitate knowledge transfer, partly via a training exchange tool matching training providers with people and groups looking for training, partly through the envisioned "Sharing Engine", partly through a project database type thingy, and partly by embedding lots of social tools.
- This is about building a new home online for Transition Network Limited (http://transitiontowns.org) and building into that some useful additional services regarding Transition news and events, a support network, a projects database to avoid wheel reinvention, etc. We recognise that this new site may in some ways come to represent a reference site that some transition initiatives may want to use themselves. This is fine, but it is expressly not our aim to deliver an “officially approved transition platform”.
We think this lot will, if we can implement it all, be pretty useful, and help to join the dots a bit, facilitate networking and knowledge sharing, and thereby help local initiatives to achieve their goals more quickly and more efficiently.
Some principles that we think are useful in guiding our approach in general:
- Don’t put all your eggs in one basket
- Transparent; open to suggestion, questions, respect for opinions and giving an honest answer
- To quote Bob the Builder “reduce, re-use, recycle”
- Zero GreenWash policy; ref: Greenwash Guide here (2.8 MB)
- Respect imperfections ; people should be encouraged towards the right pathway/ not beaten with a stick
The big picture
It would be nice to try to avoid the "jelly wobbliness" that characterizes so much of Transition projects and materials, and go for something where the leaves don't hide the tree, or the tree doesn't hide the forest.
Instead of talking about specific platforms and getting lost in the technical details, let's look at it from the point of view of the normal non-technical person that will try to use it. In my experience as a tech support girl for NGOs, what most people want is:
1) Something that works, all the time
2) Something easy to use and understand
3) Something easily scalable, that doesn't become a nightmare when it grows
This page, for example, doesn't fulfill the criteria.
Often, geeks assume that the secret is in the platform or the code. Very often, it isn't. The code for Wikipedia is free for anybody to use, but while Wikipedia is an excellent example that hits all the points above, I've seen many Wikis that use that same code and are an absolute mess. The secret for any information repository isn't in the design of the shelves, but on the good training of the librarians.
What we actually need is:
1) A couple of geeks that understand how the platform works, can introduce changes if necessary, and can ensure that it works all the time.
2) A dedicated bunch of editors-moderators-whatever-you-call-them, non-technical people, that organize the content according to some simple set of rules agreed by rough consensus (somewhat similar to the rules of Wikipedia), with the aims of improving understanding and usability. People should be able to self-select to be included in this group, but there must also be the possibility of excluding somebody if the majority feels that they are sabotaging the efforts of all the others.
3) An easy-to-use platform, robust enough to grow, and able to grow organically
What examples do people know of online communities that have all three? Let's copy them!
Issues:
Hosting:
As set out on the Hosting page, we think the underlying guidance on choice of hosting provider/s is built on:
- An open source platform. The LAMP stack (Linux, Apache, MySQL, PHP) or part of it - Zope, etc.)
- Renewable energy used to power servers, plus a provider with a appropriate outlook/values base/strategy regarding environmental issues across their business. Nice to have would be a provider that shares a transition outlook.
- Avoid over-specifying the server hardware, especially processors. Simpler processors use less energy and can often serve pages just as quickly as a state of the art multicore chip. Virtualization may offer even better energy efficiency.
- Spreading our hosting base geographically, across multiple timezones and datacentres for optimal resilience. The "eggs in one basket" argument. Taking it further, should we be using multiple providers?
- Quality of service. Real uptime, response times to issues and outages, email/phone support, etc. What do you think should be the standard we want/need?
- Shared hosting (cheaper but riskier), VPS (a fairly robust middle ground option), dedicated (flexible, more costly), or cloud? Let us have your thoughts on this. My own experience suggests that VPS would be the minimum we would want to go for, maybe a dedicated machine for core services. I have no experience of using cloud based hosting services. Got any experience of working with cloud hosting? Pros and cons?
- Back up facilities. As the services we aim to develop become increasingly mission critical, we need robust back up strategies and continuity plans to ensure minimal disruption to service. What do you think we should aiming for here as sensible minimum and optimal settings?
- Transition Network energy security - you've heard of Desert island discs, now imagine a case of Desert website communities. How might the Transition Network and Initiatives begin to operate an independent and reliable communication stream?
Of course this will all need to be specified and delivered within the cost constraints that pertain at the time. It is easy to spend a lot on this stuff, so we need to careful that our approach is proportionate, but be aware of the likelihood that over time, and assuming our plans are effective, this stuff will become more relied upon, and therefore we need to plan in that additional reslilience (and the likely additional cost of that) as we go forwards.
A 'Transition Server', run in partnership with a 'green hosting company', as discussed on our Hosting page. It could host various Transition Network sites coming out of this project, including the proposed new Transition Network website, the 'sharing engine', regional networks. We could suggest to local initiatives that have their own websites that they use this too, to provide a common focus and mutual support for the people who look after them.
Is this a good idea? Who should we partner with? (several have come up in the discussion) How much would it cost? It might be great from a cost and ease of management point of view, while at the same time delivering a high level of reliance on a single system, such that when it falls over (as it surely will), all the sites and services on that server will become unavailable at the same time. The mutual support network for site managers/admins could be delivered via other means, and if we take a distributed approach we could ensure that all our servers were identically configured, if possible.
Web Energy
Computers, plug in devices and the internet use vast amounts of energy, one way of reducing the energy load is to take a serious look at the details - think along the lines of efficiency first
Design & efficiency
- Keep the file size of graphics, photos and documents to a minimum
- Keep a close eye on the bandwidth and target areas where the energy might be saved
- Run the network in a 'low energy' mode by default
- Use ecofont for printable documents, eg, Transition Primer, Newsletters, etc
Advice and keeping up to date
- Web page highlighting the pinnacle of 'greener' web things
- Constantly update the above from reviews and suggestions by Transition People & Ubergeeks
Website frameworks:
There has been a lot of debate on this. It's a bit like the Mac vs PC arguments that roll on endlessly across the net. We don't actually care too much, as there are several very high quality CMS tools and web application frameworks out there that could probably all do a fairly good job at supporting what we need.
The approach we are taking is we believe a pragmatic one, driven by a number of factors:
- What knowledge and skills do we have readily available to us?
- Which tools offer the best opportunities to leverage those skills and that knowledge?
- Which frameworks look suitably robust and flexible to carry us forward for the forseeable future and then some.
- What languages are used? what's the size of the developer community? Are there tools/components/plug-ins already out there that could save us time and cost.?
- What do we already have in place that we can build on?
- The new Transition Network site along with its additional functionality would be built on Drupal. Why? Simply because it ticks all the right boxes. It is a mature and proven platform, PHP/MySQL based, that offers lots of the required functionality out of the box. In addition it offers access to some modules and developments in the pipeline that could deliver some major benefits in terms of: a trading platform using alternative currencies; "Web 3.0" semantic content tools (see the Sharing Engine page for more on this), multi-level group support, the list goes on. Essentially we can't think of anything we want to do that Drupal won't help with.
- The regional support network system piloted at Transition East is built on Plone/Zope, a system built with an alternate (and some would say superior language called Python). Why go with this? cos it already exists, it works, and it is being made available to the project. As noted by several commentators, going with Plone is not without risks, the primary issues seeming to be the scarcity of skilled developers and system administrators. I'm guessing we should make a direct approach to the Plone community to establish what might be achievable in terms of putting together a team that could effectively service and support a basket of Plone sites delivering regional hub services. Could be costly.
What Are We Aiming to Put in Place?
Here follows a useful diagram describing the web project:
Much more information about all this is on the What is the web project? page, so I won't bang on about it here, other than to say that the various bits of its are all eminently do-able, and fairly straightforwardly so, with the probable exception of the sharing engine stuff, but even that will likely be pretty straightforward by the time we more fully understand the detailed issues involved.
Common IDs and logins: Everyone hates logging in to websites. We seem to spend large chunks of our time keying in passwords, filling in forms, requesting replacement passwords. As someone that looks after a number of sites I see that probably 70-80% of all support queries are about problems with user logins. Having said that, there is currently no real alternative if we are to have any degree of security online, and be able to manage our own data on the net.
We've considered what's possible in terms of trying to simplify and streamline the whole burden of logins, and on balance we think that Open ID is the way to go. This from the Open ID website says it fairly succinctly:
"OpenID eliminates the need for multiple usernames across different websites, simplifying your online experience.
You get to choose the OpenID Provider that best meets your needs and most importantly that you trust. At the same time, your OpenID can stay with you, no matter which Provider you move to. And best of all, the OpenID technology is not proprietary and is completely free.
For businesses, this means a lower cost of password and account management, while increasing site visitor registration conversion rates. OpenID lowers user frustration by letting users have control of their login.
For geeks, OpenID is an open, decentralized, free framework for user-centric digital identity. OpenID takes advantage of already existing internet technology (URI, HTTP, SSL, Diffie-Hellman) and realizes that people are already creating identities for themselves whether it be at their blog, photostream, profile page, etc. With OpenID you can easily transform one of these existing URIs into an account which can be used at sites which support OpenID logins.
OpenID is growing quickly and becoming more popular. Today, it is estimated that there are over one billion OpenID enabled user accounts with over 40,000 websites supporting OpenID logins."
So we'll be looking to implement support for this across everything we build, and do what we can to enable and encourage transition sites around the world to do the same.
Comments (1 - 20 of 30)
|
|
@jbaach - your point on hosting is very valid - I know of several incredibly high traffic sites that run on a single server - websites are often highly overspecced in this dept - running a cluster of servers for a relatively low traffic site is hugely wasteful in terms of money and energy. Even running a second redundant server is wasteful if it is never needed - it would be better to fire up a VPS (or several) on demand in ther unlikely event that one server isn't enough, rather than running redundant hardware "just in case"
|
|
|
For a couple of days I have been following the discussion and I must say you guys are doing a wonderful job. Whatever the outcome may be, it is going to be great. In this post I'll try to throw in some topics I may have missed here but you may have discussed elsewhere or before. When to open shop? I've been running the dutch *.transitiontowns.nl web hosting facility on my personal VPS for about 8 months. It's a set of wordpress sites all based on an empty skeleton /theme site, which I can copy quickly whenever a new town requires one. This approach is so succesful (..) that people wonder if there can't be any 'control' from 'central-dutch' to what's going on. Well of course I can change anything but I feel as if I should not. I really feelthat the towns have their own responsibility and when there's a problem, they'll call for help. It works. The larger towns have set up their own stuff but the smaller ones can do with it. So let's call this topic 'responsibility / power'. It comes in when we have e.g. smalltown.nl.ttmovement.org and .nl an smalltown is quite new and only consists of a small steering group. Dutch political question: at what point does smalltown qualify for the site / support? Control panel I've seen numerous posts about CMS and platforms. Sure. Running on a plain linux system, I often have that problem where I do not like to hack apache configs manually etc etc. On shared hosting servers you all have seen CPanel, Plesk or DirectAdmin. But these are proprietary systems. Maybe you'll love openpanel.com just like I do? It has been wonderful to program shell scripts to create vhosts / email / ftp / etc within the blink of an eye. Openpanel should be there for all flavours but actually it tends to suit best for debian / ubuntu. Two systems I agree with the vision that we should have a set of external sites based on a CMS system plus a system (or set of systems?) for the internal information. Looking at http://transitiontowns.nl/ it contains too much information to potential participants to become really attractive from a marketing point of view. Looking at the provided information for established participants, it's still not enough. So we definately need two things for each audience. You're on the right track, as far as I can tell. |
|
Drupal, I'm pretty sure, supports a variety of SSO logins: twitter, facebook, gmail, etc. Might as well take 'em all! As for Wiserearth -- I think it lacks some of the features a TT site would like. |
|
How about using Facebook Connect as a way to logon to the CMS?
|
|
|
Quick question: it seems from reading the thread that there is a lot of talk about creating a customized CMS based system for sharing knowledge through a largely autonomous network of local TT sites (a variety of platforms such as WiserEarth, Ning, Drupal, Joomla, Google etc). I was wondering if anyone has given thought to WiserEarth being your knowledge repository (and possibly place for key cross-network groups to collaborate) and using the soon to be launched WiserEarth API as a way to create widgets that distribute information across the automous network? I suspect this would be a lot cheaper in the long run ... and get TT content across to a wider community.
|
|
|
Quick thoughts
|
|
@blebleu: from what I understand hear:
In my point of view that not all those arguments play nicely together. I see the biggest contradiction between 1 and 2/3. I guess if one would erase 1, and let 3 use 2 to create 4 it should work ;-).Have project leaders take their tools to create an easy usable/ownable system. What I would leave from argument 1 is: "don't rely on only one developer" - even with 3 in place, if you don't have good people to replace leaving members of a team the project is in trouble. The emphasis is "good" - the list of features for the tt project is extremly long. Even after prioritising / cutting away not so important ones you still end up with a rather complex system. You will always need very good people to code and maintain a complex system. Thats why I am so hesitant about the masses of php-coders. Actually, I would rather think the other way around: because php requires more discipline from the coders then python (packaging in php is hard), you need even higher skilled coders for php.Having said that - I know extremly good php/drupal coders (who are busy till 013 it seems) as well as python/plone coders. On the issue of Garys spec: my feeling is that plone fits in quite nicely in there, but I am completely biased - I was in parts responsible for the decision to use plone for Garys projects ;-)
@SteveAtkins: hetzner tell on their eco-blurb page that they use only electricity from renewable sources, and using relatively low power consuming hardware. I agree though that hardware is always an issue, thats why I would recommend to use as little machines as possible. |
|
@jbaach and @rickhurst I do agree that the CMS or the language is
not the most critical issue. You will be able to come up with a decent
platform with any of the CMS proposed. I think that what is important
is to identify which existing solutions correspond the more closely to
Garys Software specs, is the most advanced and requires the less work
to bring to completion. The one that will also be opened to
enhancements by members. I've built online encyclopedias on Lotus Notes which ten years ago was an advanced collaborative web solution, but with a rather basic scripting language, a sort of downgraded version of VisualBasic. You would have been right in telling me it was not the best choice. It did the job. They hosted 30 millions visitors running on a "bas de gamme" server without any failure in 10 years. The critical point is identifying the people who will lead this project, their experience, their vision for this new platform. The other critical point is the "easyness" it will allow people to locally inhabit this platform and make it their own. Here, I don't see Drupal or Plone as necessarily "convivial" solutions (when compared to something like Ning for example). But here I'm probably wrong. I just need to be proven the contrary.
|
|
@jbaach - thanks for the hetzner link, here's what they say... "As best German hosting provider for private clients (eco Award 2007) you can expect a lot. We offer depending on the requirements a number of different equipped root server. Convince yourself by the powerful and environment-friendly server hardware, the unbeatable prices and personal service!"
I'm not knocking the company (but they do need a GreenWash nudge); they mention "environment friendly"...just wondering... How can any piece of hardware be 'environment-friendly'? They could say something like: 'not quite as bad for the environment as other hardware devices' Also, not sure what (eco Award 2007) is all about?
|
|
|
@hosting / server: having a cluster etc. is all fine, but from the green perspective I wonder about the ecological cost of manufacturing all those servers. Also, there is the maintenance side: if you run a cluster you end up maintaining a cluster with all its inherent problems. I'd rather go for one powerful machine (an idling cpu does not consume that much power). I personally can recommend hetzner. Affordable, reliable and green.
@openid: nice system, makes most sense if you have more then one big application. However, authentication (login) is only one aspect. Managing permissions, setting who your friends/groups/people are, who can do is much harder to get right / usuable, and thats where one might want to avoid multiple systems (with multiple interfaces / storage systems).
@php/drupal/python/plone: Its important to first get the specification right (one list of requirements ordered by priority). Once that is known one can start to look for developers and implementation details. Language/framework is implementation detail. TT doesn't want to develop technology, it wants to use it. Hence one would try to get 'experts' as developers - an expert can develop a magnitude faster/better then the average coder. Having millions of average coders around does not substitute having an expert. And getting an expert is the hard thing. So I would focus on the human side, rather then the technology side. (Having said that, I personally now only do work with python, using plone to have a high level starting point, and would consider using dolmen as a very slim but elegant 'start from scratch' starting point)
|
|
|
@bkclements re:"I am not comfortable with PHP because I think there are still potential security problems. Also I strongly believe in separation of content and logic, which just isn't done in PHP." It is perfectly possible to seperate content, presentation and logic with PHP, it just isn't enforced. There are now several mature MVC frameworks for php - CakePHP, Zend Framework and Symphony for example, that make this very easy. As for security, most problems occur as a result of poor php coding, not with the PHP engine itself. |
|
|
Sorry. I tend to bow out quickly in any discussion on CMSs because I'm always on the wrong side of the fence. But if I can share insights or ideas about how I imagine this TN platform... Basic overview of VillesEnTransition.net VeT.net is a juxtaposition of sites edited by initiative appointed editors who have total control on the content of their site (or section if you prefer). There is a main section (/transition) that is devoted to providing "official" documentation (TH translations) on the movement, but technically it is just one site amongst the others.Perhaps the most interesting feature at this time, is the shared content feature which allows editors to produce and share contents (videos, news, articles) across all sites. This as the advantage of helping smaller initiatives to come up quickly with a more frequently updated website. This shared content is an "opt-in" feature, i.e. the editors have absolute control on what appears on their page, what type of shared-content. We have also created spaces for parnters such as Les Amis de la Terre Belgique or L'Agora who provides reflexion on the "transition culture". Search and indexes are local, restricted to the initiative site you are viewing. Features / services available (or to be available) : news/text editing (using FCKEditor), newsletters system (with subscription), blogs, project management, stickies (postboard), events calendar, RSS feeds. A lot of these features should have be implemented at this point, but the French is growing slowly. Which leaves me time to rethink some approaches.
From what I see, Transition Network has succeeded well in providing a lot of autonomy to initiatives. While doing so, it could not harnessed the power of the network that was emerging. A network can produce tremendous of and ideas It can also produce impressive data in terms of change, reduced social or ecological footprint. TN must see itself as a federation of independant initiatives, but still a federation with incumbing responsabilities. It should look resolutely at providing tools that gives a lot of of autonomy, yet ask from their members that they accept certain rules, a common vocabulary. You want to come up with global solution that will make it possible to get a general overview of best pratices, solutions, etc. across all sites. This is something at which Ning or Wikies fails completely and why you need a information management tool or CMS like Drupal & co. I've just spent the last year advocating a common platform project for a world-wide federation composed on 135 highly independant, heterogeneous communities. To avoid falling into everlasting discussions with network members, I only see one way out : come up with brighter, keener, more advanced, more economical (and ecological), more simple yet more complete solution than what people can readily put their hands on.
Providing global indicators I would like people to be able to realize from cumulative data gathered across all Transition initiative that their individual action (a few KM less traveled, a few dollars/euros more spent on local food) contributes to a larger, massive action. If you give 10 dollars to a fundraising event to build an hospital, your contribution doesn't mean much in itself, but as part of 1 million dollar collected funds, this is a different story. It was (it's still) one of my main motivation to embark in the desing of VillesEnTransition.net, to create these global impact indicators from cumulative individual actions.
Tools to create change
We think that by limiting the number of key axis which are, of course, not exclusive to TTs, All actions will then be match with one field of action as listed in the TH chapter "Vision 2030" : economy, transportation, etc. The last step will be to check what audience/group is targeted or is actor in this action. This grid will be evidently displayed on VillesEnTransition.net sites and serve as a measure of how local initiative can contribute and create sustainable change.
A technological gift for all communities I do not remember where or why I had this dream that TTs would come up with the next great web application. One that would allow communities to build local portals and social networks reflecting an attitude embodying a real desire for change. One that would supplies the lacks of Ning namely in terms of information management, customization or categorization, and one that would be Open. ((Ning is otherwise a marvelous platform in terms of conviviality and ease-of-use).
|
|
Sorry for dashing in on this discussion late.
I'm having trouble reading the 'napkin' sketch. I've only had time to read whats written above, and the comments below, not other links mentioned on this page.
I've been using Python + Zope in "my day job" since Python 1.5 (Zope 0.9) I guess that's about 10 years. Just about all my work is in Python and yes, its a great language.
Zope 2 is rather antiquated, but it still fits the bill for some tasks.
I used Plone on a few sites when it first came out, before it was mature, but I have not looked at it in more than 5 years.
Please don't read me as a snob when I say I am not a fan of PHP on general principals. I am not comfortable with PHP because I think there are still potential security problems. Also I strongly believe in separation of content and logic, which just isn't done in PHP.
Drupal is great with it's large following and all, but trying to customize it seems unattractive to me because the PHP language feels rather weak compared to other languages.. Lack of namespaces, difficult to modularize, etc.
But if customization is going to be a very small part of the project, maybe this isn't a concern after all.
For the Potsdam and Canton NY transition group (we're still very early in the process), I need a CMS with web forum support (even though I prefer mailing lists), I intend to keep user info in LDAP or use OpenID. I can assure you that none of the members of our group knows what OpenID is, I think OpenID is a good idea. You might have to have several 'flavors' of login to support facebook or google users (where openid has been 'improved' by those companies)
I also need virtualhost support so I can provide CMS functionality to other groups by prefixing their group name on the base domain we're planning to use: transition-initiative.info (just a static page there now)
Anyway, even though I have lots of Zope experience, I actually am taking a very close look at liferay portal.
Being java based, it's a bit of a memory hog and would require a beefy VPS. But I will be hosting the site on one of my office machines which is already running, so that's not as much of a concern right now.
I am interested in hearing what others think of Liferay Portal.
Also I would like to know more about P2P information sharing. Why would various transition servers send 'newsy' stuff to each other, when individual transition group members could just subscribe to various feeds on different servers in the reader of their choice, or we could just aggregate all the feeds into 3 east/central/west feeds? I'm not saying its a bad idea, I would just like to know more about that objective.
Regarding 'green' hosting and low-powered machines. To me that seems like a bonus feature right now, but maybe not a requirement at this stage. For this project manpower costs will outweigh energy, networking and hosting costs. Even volunteers 'cost' something.
I may mis-understand this objective but it seems to me that if you have to use a 'far away (high latency)' under-powered VPS to be 'green' at the expense of increasing development or maintenance time, that would not be a good trade at this point.
Though I could try running Liferay on my OLPC. :-)
So, anyone have experience with Liferay? I'm only half-way through the 272 page admin guide and could switch to something else with a strong recommendation.
thanks
|
|
@rickhurst: Yes, I've now read your blog article and find it eminently sensible. For my software spec, I needed to do more than just minor customisation, and with a real Plone expert that has gone very quickly and at a very small cost. It would not have happened otherwise. But, yes, it leaves me very exposed. I need a team of Plone people. Would you be willing to be part of that?
@bleblue: Please don't bow out of this discussion. With your experience you are just the kind of person we need. Yes, Drupal seems to be getting most mention here, and I think for good reason, but we need to learn from you as well. I agree completely that we are developing not just for TTs. We are all learning our way towards a new kind of global collaborative culture. What facilities are you offering the groups on VillesEnTransition.net? Who is writing the content? How are you getting people to contribute? What are you finding are the most important uses? Where do you see it going? |
|
|
@garyalex yes my blog is working: http://www.rickhurst.co.uk/, and the server hasn't had any problems, but I heard that bytemark (the usually very reliable company where the server is hosted) had problems with their data centre over the weekend, so maybe it was something to do with that? Either that or a word wrap issue on the long URL I posted before? @Upperholme Drupal has an aggregator module as standard - it can be configured with no coding to pull in and display content from multiple RSS feeds
In general I think Drupal would make an excellent choice. |
|
|
Before I bow out of this discussion because I don't use any of the CMS that are discussed here, let me just throw in a few suggestions :
1. TT deserves the best. Try to make of these TN tools world-class standard applications. Hire professional designers to create classy, highly functional templates.
2. Try to think for tools not only for TTs but also for their larger, global community. There is tremendous need for community-manage portals tools to give an alternative to commercially publicly owned newspapers or newsletters. VillesEnTransition.net will offer this kind of platform next fall here in my locality. It will host the Cultural center, the Historical Society, the Communatarian Center and a variety of citizen-based groups (readers clubs, etc.). And the local TT initiative will be present amongst all the others, and gain visilibility and built trust as a dynamic local actor.
I think that TN tools should be able fill this gap and by providing this community portal, mark their presence and reach people outside its own circle. This is what I meant in my previous message when I wrote about "tools embodying TTs philosophy" and that will bring a "Transition-oriented CMS for 21st communities" to the world.
Good luck! |
|
We're using hosting by Athaeneum www.ecologicalhosting.com who are UK suppliers of VPS's out of solar-powered California-based Data Centre http://www.aiso.net/. This was the only 100% renewable data centre providing VPS at a reasonable price I could find, but that's probably changed now - I hear there are now 100% renewable Data Centres in UK, although Aiso actually generate their own power.
We are building our site on Drupal as add-on modules are already available for all the functions a Transition site is likely to need: calendar, user and groups privileges, blogs, forums, ability to upload docs and can be formatted to a Wiki-style repository also. It's fairly easily customisable theme-wise and with Organic Groups, sub-groups can have their own themes. For us in TTHackney, it may be that sub-areas of the borough who don't want to pay for and build their own site may want to customise their area and control their look and feel. This makes Drupal a winner, IMHO. Also talking to two Drupal devs we have on our team it seems Drupal is gaining popularity massively and there are more and more people learning it. I disagree it is difficult to bend - it's what you know after all. I'm an Oracle guy so none of it is familiar but it is logical and there's loads of web-based resources, including plenty free video tutorials so the learning curve is steep but fast. Any "Ubergeek" (hmmmm) should find it do-able with some patience.
Drupal offers excellent content categorisation which will be the real challenge for a TT site encompassing multiple distributed local groups. How do you avoid massively repeated content? All content must be categorised on entry and this could provide input to a search page showing similar stuff which the user has to confirm they've seen and are not duplicating before they click the save or upload button.
Drupal is written in PHP and so modules can be built to provide additional functionality.
As for resilience and/or making it a distributed system, obviously this is not something Drupal does out of the box. The first approach is to cluster separate front and back end servers, removing the single point of failure of one server . Then you have a hot DR site in case of comms failure to your live site, with frequent replication of the live servers to a stand-by somewhere far away. Someone flicks the switch when the live site goes down. This is usually an extremely expensive option, involving lots of redundant hardware. Also there's issues with DNS so you usually need a load balancer etc etc.
Multiple distributed servers with different hosts and different URL's might be quite an attractive option and Drupal runs off a MySQL DB which should be a fairly easy platform from which to set up replication to other servers. I am not a MySQL DBA sadly, but from my DB experience this should be possible even if we have to build scripts for it. It would still take a lot of research and work to implement, I am not sure if MySQL can really handle prioper distributed transaction processing..
I like (love, in fact) the idea of a more peer-to-peer network approach but I think this is an evolution of the internet which is still to come (probably to be spurred on by rising energy costs and the introduction of pay-per-gig and two-tier pricing models as internet TV really kicks off) and I wouldn't imagine TT has the resources to build a platform from scratch for this.
The bottom line, I personally think is this: any TT system will initially rely on the chosen hosting company's record on uptime and reliablilty. 99.999% is not uncommon and I think TT can live with that. One day once the basic system is up and running we can all start looking to nice sexy peer-to-peer server replication and managed data-sharing but right now...
Finally, we are learning that as a transition group we need a presence on ALL of the various social networking and and eco / environmental sites other groups are using, to pull in people who don't notice signs in shops so much as Facebook groups - sign of the times. So this means having a presence on Facebook, MySpace, Ning, ProjectDirt, WiserEarth etc BUT none of these singly provide all the features people in our group are requesting, and are not as customisable. So nice as they are they remain signposts to our own site, IMHO.
I am an Oracle DB, App Server and Linux / Solaris administrator.
|
|
hmmm...I am just becoming oriented here. there seems to be a lot of talk about IT platforms. can I assume there exists a requirements document already? if yes, where can it be found? I'll respond to some of the comments I see here. First, I would go with drupal or joomla and not plone/zope simply because there is a much larger development and support community out there for enablement. Telecomputing is a failure because it has not achieved appliance-like behavior. IT was supposed to be about augmentation and facilitation of human endeavor, not the black hole it has evolved into. Second, I would like to see a peer-to-peer tools application come out of all this that is not as vulnerable to disruption/interference as the highly centralized, vulnerable, Big Brother and corporate-controlled client-server-internet model. Specifically I envision a hard and soft resources sharing clearinghouse using the peer-to-peer model where communities can have ready access to inventories, hard resources in neighboring communities like tools/hardware, soft resources like people skills and knowledge sets. Third, why are you talking about Open ID...an LDAP-like technology? For what purpose? Perhaps your requirements document explains the need for this.
what I have to offer. many years of telecomputing and organizational experience, drupal applications development and hosting, and lastly, the Center for Community Alternatives, a non-profit organization. |
|
I agree with Upperholme Drupal as an excellent choice. It is a community focused CMS which is ideal for transition sites. The first transition site I built was with wordpress which is easy to put together but we found that it ran out of power and the points of interest were groups and events. To build on this I used drupal to allow users to sign up and contribute their own groups and events, join groups, flag whether they are attending events and post news stories.
Using CCK and views, customisation is straight forward.
Drupal is open source with a huge support community making sure it is up to date and secure. There is a huge repository of modules proven on many production sites. For a project of the transition networks size we need an established solution that is highly customisable enough to suit each communities needs. I can't think of a better solution than drupal.
Here is a list of the main modules that i used;
CCK and Views
Organic Groups
Date, Calendar and Flag (for events)
captcha
i will use a newsletter module until our subscription goes over 200
images
image cache
The Zen theme is good if you are new to Drupal but it is heavily bloated.
Another advantage to drupal is that there are a lot of screencasts on the web related to Drupal. We could easily build a training library of links that related directly to customisations common to Transition sites |
Are there any plans to make standardised email addresses such as
contact@mytown.nl.transtion.xx
- standard addressess per subdomain
- webmail, imap / pop
- or just forward email adresses to a remote (gmail) account
For this to works, we will need to make adequate DNS entries for eacht host/subdomain. Creation of email addresses and DNS can be done with openpanel.