Now that D7 has been released, we have to make some fundamental project decisions.
- Our own code base or wait for Drupal Commons to port to D7.
- Which contributed modules to support?
- Which unique modules to design and build?
- How does the project get supported? (More below.)
- How do we handle themes? Harvard's system in Open Scholar is elegant.
-
How do we upgrade installations? Features or the new kid on the block, Site Update?
-
Our own code base or Drupal Commons?
Benefits of Drupal Commons
a. We benefit from Acquia's work plus from the community they are building
b. Acquia has deeper resources than we do and I sense that it is Jay's (CTO of Acquia) personal project and love.
The drawbacks are:
a. There is no D7 version yet and there won't be for six more months, at least (I asked at BADCamp)
b. Installation won't be just one-click; users have two steps and potentially two upgrades to perform.
c. We create a dependency on Drupal Common's and their release schedule
The drawbacks seem to outweigh the benefits, to me at least. So despite the fact that we are likely going to duplicate a lot of the work in Drupal Commons, I think we should get going. If we make sure nothing we do is tied to our core, there may be a way to leverage Drupal Commons in the future if that makes sense.
- Which contributed modules?
There are about 800 modules ready but some big ones still aren't (like panels). Still, let's run through the list we've created, check against the Drupal Commons list and put together what is available now. We might find quite a bit is ready to use and possibly even to deploy.
- Which unique modules to design and build?
Ah, the fun stuff. I'm ok working on the infrastructure for now to create a solid base for the project. So I'm going to leave this for a bit. But if other people have something they'd like to work on they should start planning it out...we will need a base design for them to work on i.e. which profile fields will a person have, etc.
That means the information architecture team has to deliver their product...I emailed Jacob but haven't heard back. Jacob, if your email has changed please email me!
- How does the project get supported?
I think we should go ahead and make something but before we release it I think we should have a paid support person in place. I was at a presentation at the Bay Area Drupal Camp (BADCamp) a few months ago and one of the speakers made an impression on me. They said, "maintaining a distribution is a lot of work. Seriously, it's a lot of work." That got me to thinking that we should aim to operate like http://civicrm.org — IIRC they have four or perhaps even six full time people supporting their distribution.
I'm thinking along these lines because I'm not unemployed yet (I keep turning away Drupal work) so it wouldn't work for time-senstive problems to land on my lap — I couldn't turn them around with any time guarantee. I presume it's the same for other people involved. Thus, let's discuss getting foundation money or similar. Thoughts?
- How do we handle themes?
Check out Open Scholar, another presentation I went to at BAD Camp. Very well done. At the time they didn't have plans to port it to D7 but time has passed. We should ping them and ask again.
- How do we upgrade installations? Features or Site Update?
I'm just getting into Site Update and I find it very elegant. I have already started porting it to D7 with the original module creator (Dave Cohen). He and I are going to work on it together. Here is Dave Cohen's description and I recommend that every developer read it.
The Drupal Staging Problem
http://www.dave-cohen.com/node/3932
Making the correct decision with updates is going to be a pivotal decision for the project. From my discussions with users of Features Site Update seems to present a lower bar. Dave's presentation to the 80 people at the San Francisco Drupal User's Group meeting late last year was a hit. Many people thought it was a very good method and we anxious to try it out. I'll provide more feedback as the port progresses.
That's it for now. I'm beginning work on a D7 project for the University of California tomorrow so I will be knee deep in D7. Plus I've already ported some modules from D6 to D7 and have a good sense of the API changes already.
I also played with the distribution system, something I hadn't looked at yet. We could have a distro relatively soon after we figure out how we perform upgrades but it seems to me that we need:
a) an information architecture
b) one or more D7 themes that support the IA
c) a plan for support.
Please chime in!
-André
Comments
Fantastic
Hi André,
I was wondering when we'd be getting back onto this. I had a quick play with D7 the other day and watched the Lullabots video intro and it looks sensational.
Drupal Commons looks awesome too - such a shame it's not available on D7 yet but at least we can use it as a very successful model. I particularly like the fixed-width home page and the very well implemented fluid layout on the internal pages. Very clever.
Jacob and the other IA team never did get round to meeting to discuss IA at different levels so perhaps that should be a new priority now?
It's exciting to be back on this now.
Theming
Hi again, André,
After reading your comments again, I'm wondering more precisely what you meant when you referred us to look at OpenScholar?
Were you citing it as an example of good visual styling, content layout or something in the admin interface or even code that you specifically liked the look of?
As for Jacob, he's still to be found on Twitter so I'll ask him if he's still interested in the project.
What do you think about focusing exclusively on just a single "transition initiative" for now, rather than trying to also think about wider networks such as city hubs and larger? Having taken the Transition training course about a month ago, I became aware that so many different network structures within and between initiatives are growing simultaneously in different parts of the world that it might be rather a complex task to research them, let alone map them to an IA.
Perhaps the TN group might have some insight into this which would be valuable here?
I was also a bit puzzled on
I was also a bit puzzled on the OpenScholar stuff, so some more info on that would be great :)
Agree on focussing on the needs of a single inititiative for a first release. There is a lot of complexity in that and I think we're going to find lots of customisation requests for things like taxonomy and themes, so figuring out a useful baseline will be a challenge.
Fabulous UI
For OpenScholar I was referring to the very excellent way they handled themes in the UI. I don't know if there is a video tutorial out there but the demo I saw was very impressive. I haven't looked under the hood but there might be some interesting things there, too. It's easy to install to take a look or perhaps they have a demo on their site somewhere. (I installed it locally.)
Jacob emailed me yesterday after I emailed him again. He lost his computer when I emailed him last time and he thinks my email got caught in that. I've asked him if he's able to pick up the project and I'll let you know his response.
I too am ok with a single initiative given that the IA for the larger structure hasn't occurred yet.
No one is talking yet about how to support the project with funding. Let's spend some time on that, too. We could fairly quickly get a version of a distribution modeled along Drupal Commons ready but who supports the possibly 100 initiatives who might have support questions??
Andre Angelantoni
Founder, PostPeakLiving.com
Customisation
I suspect most of the theme customisation will be similar to the type of changes people make in Ning:
- tweak text and element colours
- change background and header images
- change logo
- edit site name and tag line
- etc
Almost all these can be handled with the standar Drupal core theme tools
But, Paul, what do you imagine people might want to change in taxonomy?
- term translation into local lingo?
- structural changes to the sections available on the site?
Agreed on theming
Agreed on theming customisations, but we need a nice looking theme that allows all that. Does that suggest Fusion is the way to go? (with integration with Skinr, etc)
Say we had a simple theme taxonomy of Food, Energy, Transport, etc, similar to TN.org. I know in Belsize there could easily be requests to expand that to cover more specific groups. I mean here the Drupal definition of taxonomy, not necessarily all the sections of the site (although they might need to be customized too).
We need to document a proposed IA on TD.org and ask for feedback I think...
Themes and funding
I'd like to at least try to have the best benefits of Fusion and Skinr integrated with the new colour functions in D7 and the flexibility of the theme layouts you can see in Drupal Commons.
As for funding, André, could you give us an idea of how much funding ($US) you'd be wanting to obtain and on what basis: a one-off payment, a multi-year agreement or an on-going fund?
Also, if we struggle to obtain funds, I know I've always assumed this would be a free product but perhaps we might ask for some sort of voluntary small donation from each initiative to help support fees etc?
What does everyone else think?
General thoughts
Commons is a great distribution, but I agree with doing our own thing, mainly cos the flexibility is probably quite important, also the installation is simpler. But I wonder how much of Commons could we just use and wrap into our own distribution? Its all open source after all?
For IA, perhaps we do need to just get something started for a basic site, that also accounts for what is possible with D7 now (e.g. there seems to be little action around Location currently, so that might affect some features). And then build over time?
Totally up for funding ideas, not sure what would be the best avenues to approach this. If the goal is support then we will have a more compelling argument when we have the core of a installation profile put together.
For distribution, is it worth outlining the kinds of scenarios we expect? Will it always be our responsibility to roll out upgrades? In many cases it will, is it worth using Aegir to help some groups to just get a site going with little hassle? I now have hosting for unlimited domains so maybe one model is clusters of sites in different regions?
Using site update could certainly have advantages but we need to be clear on what settings are constant across ALL these installations, which ones are custom preferences and which have to change because of certain issues, e.g. timezone. And might Features be needed anyway? If we start creating some advanced or edge features they may not be wanted in all sites, hence could be useful to have split out and configurable?
What are your plans for infrastructure? Are we going to get a project setup on GitHub or the like?
Accelerating the transition movement
Although we're all primarily motivated by custom modules and complex sites, I suspect that what is most needed for the growth of the Transition movement is free hosted solutions for small initiatives. Many initiatives pay over the odds for decent websites to be developed for them, or they get by with inadequate solutions, and this may hinder their progress.
Most start ups would probably be happy with a 100 mb account and a basic, themed, CMS. Ideally the TD solution would get started right now at the small, free and very simple end - and then grow with the initiatives. If we took this approach we could also follow a more agile development?
To date there hasn't been much need for regional transition initiatives - but I think regional hosting - provided by regional-level transition groups - would be a logical way forward (Transition Scotland were talking about this model at last year's Transition Network conference in the UK). The aim of the regional group would be to encourage more registered initiatives in their region with free hosting and future upgrade plans at near-cost price. Technical volunteers in a region could sign up their skills with their regional TI as well their local group, and developers could also sign up at TD.org.
I think before we can seek proper funding for TD development we need to demo the basic product and the scope. If we each host a few sites for small initiatives in our regions, then we can demonstrate the network model and start talking about funding for TD.org to oversee these networks and start pushing the functionality.
In the meantime - a "donations" button at TD.org would be a good idea.
I'm about to set up a web presence for a second initiative - I'd like it to be "TD" from the outset - even if this is basically a vanilla drupal install downloaded from TD.org as version 0!
Any thoughts on a logo for TD? (@dj - graphic design is your area?).
Community Tools
We have created a Drupal based platform with kind of similar aim to enable communities to work better together and share their knowhow.
There are some fix-my-street and idea mapping application built with it. Mapping problems & ideas geographically is one central function.
There are some modules which we have planned but not executed yet: local currencies, radio, stuffarium (for sharing stuff and spaces) and mobile frontend apps. Maybe we can combine something from our efforts with TransitionDrupal.
You can look at the demo site:
http://demo.selts.eu/
The package can be downloaded from:
http://www.communitytools.info
Current live instances can be seen at:
http://www.selts.eu/
Questions welcome ...
Call to arms
Hi!
I posted suggestion about using Commons distribution as a foundation for the real-world community web sites: http://groups.drupal.org/node/219564
Feedback is much appreciated. :)
Sorry for cross-posting, this is just in case somebody is only following this particular thread...
Cheers.