Need fresh review of move to Drupal 6

Project:The Hive project
Component:Miscellaneous
Category:task
Priority:normal
Assigned:Unassigned
Status:active
Description

I have done a couple of prior passes on module status and issues for moving to D6. This seems important now that the world is moving to D7.


Comments

Comment viewing options

Select your preferred way to display the comments and click "Save settings" to activate your changes.

#1

That review could conclude (again) with worries that the Project_issues module and OG (organic groups) module still don't work together in D6. If so then what are the options:

(1) wait some more, hoping that someone will repair the compatibility problem

(2) upgrade to D6 or D7, keep project_issue module, drop OG module

(3) upgrade to D6 or D7, keep OG module, drop project_issue or replace it by some other issuetracker (if exists in Drupal)

I do not have faith in option (1) anymore. I really love the concept of OG but I also believe that project_issue is an absolute must-have (unless there is a different issue-tracker we can convert to). If there is no other way I would choose option (2).
What do you think?


tamhas's picture

#2

Given that Drupal has moved on the D7, I agree that it is questionable whether we will see any further improvement on D6 module interaction. Either we accept things and stay on D5 forever or we figure out how we can best move forward. Unfortunately, I think this is the world of open source tools ... a really cool thing in one version can get left behind in later versions because the maintainer goes away and no one else picks it up.

I agree that the project functionality provides more unique and special value to OE Hive. To be sure, we have a bunch of things that are projects which shouldn't be since they are one person, one product, one version. So, maybe we need some kind of review or approval process going forward. But, where it matters, there is functionality that would be hard to replicate.

OG, on the other hand, seems like a nice idea, but I don't know how much mileage we are getting out of it. The idea of getting mail only on topics of interest sounds good, but I have very little confidence that people really belong to all the groups they should, especially if the group was created after they joined. Plus, we have group abuse ... people who mark a dozen groups for a new node. And, some other funny things. So, we might see if there is some other mechanism for notification which would have fewer issues.


john's picture

#3

I sometimes wonder if it would be easier to start fresh. Start with a clean copy of Drupal 7, and then figure out how to fit the old content and site features into D7. We'd have to find a new way of doing a lot of things, which could be good because a lot of what we have currently is clumsy.

For future upgrades, it would be nice to depend on as few third party modules as possible.

Maybe there are features in D7 which will allow us to drop some of the modules which have been holding us back. I don't know. It would take some pretty serious D7 expertise to figure that out.

I think most of the data from the third party modules we are using is pretty simple MySQL tables. So, maybe it would be easier to migrate the data than figure out all the php programming needed to make the old modules work with new Drupal.

For the question "should we even try to upgrade": For me, the ROI just isn't there. I'd love to see us on the latest Drupal, and many bugs and annoyances fixed. But, the site as it is today does do the basic things I want it to. Even though some of it is clumsy, I really like the features it currently has.


tamhas's picture

#4

We could certainly look at D7, but my guess is that there is a whole lot which is not going to be there, especially the project stuff. Of course, there has been quite a push for people to commit to having their module ready when D7 shipped, so maybe there is more available than I think.

I don't think the goal should be using as few modules as possible. It is the modules which give Drupal its richness and provide the potential for this to be a really standout site. Rather, I think we just need to be conscious about out choices. To be sure, a well maintained module one year can turn into an orphan two years down the line, so one can't protect against everything. But, I also think there are modules that would be nice to have, but that wouldn't kill us if they stopped being maintained at some point. E.g., I really like the module that allows one great big long node to be viewed in pages. it is just a filter on the output, so if worst came to worst and we dropped it, all people would need to do, at most, is to edit the node to drop the [ page ] markers. Low impact if it goes away, but a very nice touch when there. Likewise, I think we could add a lot of usability to the site with a good set of FAQs. Making that obvious and readily available would save us a lot of misuse. And, if it goes away, convert it to regular book nodes. Big deal.