By media girl on

Please forgive me if this is redundant, but I believe this is a bit different from the Intallation Wizard planned for 4.6, and in my search here I found no obviously related discussion in the past 6 months. Perhaps I missed something, as I do not believe this is incredibly insightful or profound. But, FWIW....

One of the biggest barriers for administrators, I believe, is the amount of database and programming knowledge required for maintaining a Drupal site. Thanks ot phpMyAdmin, module installations have become a breeze for me. But I'm still trying to master the patch thing, and dread the day when I will have to perform an upgrade that involves changes in database structure. But recently I installed a module (forgive me, I've forgotten which one) which had its own installation scripts in it. It was incredibly easy. (Another hint that this topic may have arisen before.)

I suppose you can guess where I'm going with this.

This is simply a humble suggstion to streamline the installation and especially upgrade procedures on Drupal so that:

(a) upgrades can be done seamlessly on the fly, with the site live, if possible, and
(b) it is possible, from the admin menu, to ping the Drupal.org site to see if any upgrades to the core or modules are available.

As it is, it can be easy to miss important fixes or feature enhancements if one is away from this site for a few weeks, as there are no release dates visible on the Downloads page, and those red asterisks are up for only so long. And the process of upgrading/patching can be quite intimidating to an admin who may know css and xhtml but is perhaps a bit out of her depth in php and sql.

Please don't get me wrong. I'm not suggesting that Drupal become Mambo-like -- I tried that, and found their system rather rigid and unhelpful, in a lot of ways. What I am thinking of is something like what we have with Firefox and its extensions.

Anyway, I'm not suggesting that the core or development system needs to be fixed. I'm just speaking as a site admin with two Drupal sites (so far), looking at what would make Drupal really easy to manage over the long haul. Think of it as a way to strengthen the connection between Drupal-powered site admins and the development team here.

I really believe that once Drupal institutes either a generic install/upgrade module or a core standard for contributed modules to install and upgrade themselves automatically, we will see a veritable explosion in Drupal usage around the world.

Thanks for reading.

media girl

Comments

crazybooks’s picture

crazybooks commented

I second this, being at a similar level when it comes to fiddling with the databases. I've not even attempted the patch thing yet, as I would rather not make a newbie mistake and mess up my installation. (yes, I'm making a test server soon.)

I believe this is something that would hugely benefit drupal for the 'middle' admins like myself who have some web application expertise, but not enough to be as effective as some coders.

----------------
I R Admin Monkey. Damn those users. Damn them with Zim like hyperbole

carlmcdade’s picture

carlmcdade commented

I am marking this thread also since essentially it is the same as another in that it calls for an expansion of the development process which is beyond the capabilities of the present system.

I wish I could find those articles I read comparing the commercial development process and open source. It expalains a lot why somethings will not happen to Drupal development for a long long time.
---------------------------
www.hivemindz.com (running PHP5)
www.fireorb.org (documentation and hacks)
__________________________
Carl McDade
Information Technology Consult
Team Macromedia

carlmcdade’s picture

carlmcdade commented

ahh! found one.

http://sern.ucalgary.ca/students/theses/KimJohnson/chapter3.htm#UserDriv...

If you read the last parts of this then you will see that there is a large gap in the open source development of Drupal. This gap is being filled partially by Civicspace. But growing popularity is making it difficult for the core Drupal group to keep up.

The chapter after this one is particularly interesting but not dirrectly applicable to this thread. So I will post it in the other.
---------------------------
www.hivemindz.com (running PHP5)
www.fireorb.org (documentation and hacks)
__________________________
Carl McDade
Information Technology Consult
Team Macromedia

sepeck’s picture

sepeck commented

Not accurate. Many developers and contributers are involved with Civicspace (I know the ID's of many and you will see links to Civicspace stuff in a variety of posts here). There have been statements made by Civicspace folks that these enhancements and modules are planned to be rolled back to Drupal in the same manner in which the work done by Bryght (http://drupal.org/project/sxip) is.

Welcome to Open Source. In order to provide quality code, there are rules and standards (http://drupal.org/node/316) or there is choas and disaster.

Unfortunately, this can sometimes lead to "feature creep." Alan Cox (McMillan, 1999) discusses the Linux kernel: "My main concern is that however hard you try to keep the kernel lean, things gradually creep in. It's very easy to put something in a kernel; it's really hard to get it out. That's the real thing: making sure you keep the gradual growth under control. Getting that wrong could be very bad."

The core team is behaving very responsibly here. What goes into core has to justified as to it's effect on future upgrades/maintenence, hook accessability to modules and that it is a feature that by it's very nature needs to be in core. It seems it is better to be conservative and not have to rip things out later. From my software architecture class, this deliberative planning process where more time is spent planning out enhancments results in time saved with the reduction of bug hunts later.

The disadvantage is that while a technical emphasis benefits feature-set control, it can adversely impact other areas, most notably end-user design. Participants in open-source projects are not really users, but programmer-users. Linus Torvalds (Ghosh, 1998) remarks: "The thing that makes 'real users' so interesting is that they have so [sic] different usage patterns from most developers, which is why a product that is solely targeted to developers tends to lack a certain stability and finish." [23]

Open-source projects are therefore inclined to offer workable imitations of popular commercial interfaces. Features address user activities rather than user behaviour. Innovation that is routinely found in the underlying code is not present in the interface because there is no feedback loop to true end-users, and no imperative to create one (Kuniavsky, 1999).

Open-source software has tended to be more infrastructure-oriented. Traditionally, most projects have focused on operating systems and network services. This is because end-user applications are hard to write. Graphical, windowed environments are relatively complex, and most programmers are not good interface designers. Open-source projects also tend to thrive where incremental change is rewarded, which has meant back-end systems more than front-ends. (Behlendorf, 1999)

And in response, anyone can propose features http://drupal.org/contribute) and suggest methods to solve perceived issues (http://drupal.org/node/15506). For instance this thread started the admin interface redesign (http://drupal.org/node/7862 last year and no I am not going to find the other threads on it) which attracted some interface specialists and resulted in the tabbed interface for admin tasks.

Lots of exciting work being done here. Participation is fun and a lot can be accomplished within the development framework and collaborative process that has been established and documented (http://drupal.org/node/316).

-Steven Peck
---------
Test site, always start with a test site.
Drupal Best Practices Guide

carlmcdade’s picture

carlmcdade commented

By reference to Civicspace I meant the user base. I am being a little judgemental probably when I say that because CivicSpace seems to attract a different type of use than Drupal.
---------------------------
www.hivemindz.com (running PHP5)
www.fireorb.org (documentation and hacks)
__________________________
Carl McDade
Information Technology Consult
Team Macromedia