Category Archives: MTG Southwest

MTGSouthwest.co.uk Launch Procedures

This is here for my benefit.

1. Migrate the database
1.1 Backup existing database to an SQL script
1.2 Create a new database server
1.3 Create a new database user
1.4 Create a new database
1.5 Restore data to the new database from the SQL script

2. Move the site into production
2.1 Ensure there is a folder on the web server for mtgsouthwest.co.uk
2.2 Ensure the folder is empty
2.3 Copy the files from the development folder to the production folder
2.4 Change the base URL settings for the site in wp-config.php
2.5 Run searchreplacedb2.php to update BuddyPress links

3 Make sure it all works
3.1 Test all the page links
3.2 Test users can login
3.3 Test user can register
3.4 Test user profile links
3.5 Test forum links

4 Publish the site
4.1 Announce the site’s publication
4.2 Encourage users to register and login

5 Run a Promotion
5.1 Set list of tasks for new users
5.2 Decide on the prize for the prize draw
5.3 Decide on an end point for the promotion
5.4 Monitor activity to…
5.4.1 Ensure users are responding in appropriate numbers
5.4.2 Ensure that the site is working as expected
5.5 Ensure the winner receives their prize

Yeah, this web development. Easy, isn’t it. I have know idea why it takes so long. Mind you, this is just to make sure it’s published without a hitch.


MTG Southwest Website – a Further Update

So I ditched BuddyPress and started investigating Mingle as the software to power the new MTG Southwest website. Mingle seemed perfect – the theming was intuitive, the navigation of the site was fine and so on. Then I came to test the social aspect of the social network site – you know what I mean, sending messages, adding friends, looking at profiles, that kind of thing.

To do this I created a second user account, fired up a second browser, logged in with both accounts at the same time and sent a message from my original, admin account to the new one. Turns out you can only send messages to people on your friends list. Bit inconvenient, and something I’d like to configure before the site goes live. So I navigated to the other account’s profile and added them as a friend. Except, I couldn’t find a way to do that.

So here I was, with a social networking site, with two accounts both logged in and both completely unable to interact with the other. Fucking brilliant.

And back to BuddyPress I went. For all the failings of the theming engine, at least users of a social networking site could actually use the social networking site for social networking.

So that’s where we are. Actually, we’re quite a lot further than that. There’s forums and groups. The groups allow some forums to be kept private if that’s ever needed. Users can search these to find relevant information and if they can’t find it they can create a group to discuss their interests. There are user profile pages it’s possible to post comments to, rather like the wall of another site you might have heard of. There’s private messaging for, well, messaging in private. And there’s even an events section – a calendar of events, colour-coded by town or city, visible a month at a time with links to information about each event.

There’s a chap called Tom Palmer. Now, Tom is brilliant. He’s only gone an designed a logo and a masthead for the site. I’m really pleased with his work and how the site looks as well as how it functions.

At the moment I’ve thrown it to the lions’ den and a group of about 10 people are using it, trying to break it, making sure that everything works as they expect it to and that the tasks they needs to perform are performed how they expect. This has thrown up a couple of things. First, email notifications were happening for every single action that takes place on the site that affects your account – for Tom this was generating about 200,431 emails a day. Fortunately there are options in user account settings for such things. Phew! Also, another user identified that the way I’d structured the forum did necessarily make sense in all circumstances – although I’ll defend my decision, I can see how having two separate groups for Modern and Legacy format discussion might be an obstacle if you want to discuss prize support for Modern and Legacy events you plan to run.

But that’s all part of the process and all those detail will get sorted out.

So thank you to Tom and thank you also to the testing people.


MTG Southwest Website – an Update

There is a project currently underway to unite the various geographic-based groups in the Southwest that play Magic into one uber-group. This is no small task and at the moment is being rolled out through Devon then is planned to incorporate the rest of the Westcountry.

My part in this is building and managing the website for the group. And this is an update for the people who have an interest.

There’s a common phrase in software development: “It’s always good to throw one away.” If that’s so, then I’ve been very good indeed. It’s now just less than a week into the development of the website and I’ve tried and discarded various bits and bobs. If Magic was software and creatures were PHP and mana was HTML then I’d have tested an entire format.

For various reasons, I’m disinclined to build an entire site for MTG Southwest from the ground up. It’s too time intensive, it’s not worth the money, testing for such a project is more than even I have the stomach for, let alone my volunteer testing team. Not to mention the design work involved with making it look pretty and usable. Instead I’m using a freely-available content management system (CMS) called WordPress. Well, strictly speaking WordPress is a blogging system but it has features that lends it to working as a CMS. In fact, being a blogging system makes WordPress even more suited to the needs of the community than a more business oriented CMS. WordPress is made by people who know what they’re doing and as a widely used system it’s been tested in various stages for years by thousands of people including commercial developers, business end-users and people who keep a blog about their knitting.

I like WordPress. It has some sweet extensions, or plugins, that offer all kinds of functionality to a site as well as themes to make it look pretty. And that’s what I’ve been buggering about with for the last week.

There’s a plugin called BuddyPress. What this does is, after a bit of configuring, it turns a website in a social network site. It supports groups and profiles and all that gubbins and with some further extensions designed to work with BuddyPress it offers forums and event management. There’s one problem with this – BuddyPress only plays nicely with themes designed specifically to work with it. If you don’t use one of these themes then all kinds of funny things happen – menus don’t display, pages fail to render, the functionality that BuddyPress offers doesn’t exist, that kind of thing. This is all well and good. But.

But this site has a few specific requirements that couldn’t be met by themes written for BuddyPress. For example, we decided we needed a “feature box” to appear at the top of the home page, just beneath the banner and above the first article teaser, that provides links to calendars of events run by each of the various groups. Whilst this is possible to achieve in BuddyPress, it’s only achievable through the themes and I discovered that the themes are, well, rather incestuous. And when you think you need to be working in one theme – the theme you’re using on your site – you actually need to be making changes in another theme. And after digging through code to try and find the right script to edit, that script points at a different script and… well…

So I ran out of patience. I ditched it.

I went back to the drawing board – which looks rather a lot like a popular search engine – and I looked for an alternative to BuddyPress. Something that provides user-to-user messaging, forums, customisable user profiles.

And I found a system called Mingle. Mingle isn’t precious about the themes it shares sites with. Mingle is easy to configure. Once you’ve customised one WordPress theme you know exactly what needs to be done to customise a theme for your Mingle-based site.

So that’s what we have. We’ve got got one collection of websites, screwed up into balls and thrown into the corner. And we’ve got one social network site that is powered by a Mingle-enabled WordPress blogging system. It’s got custom features built by my own fair hands, it’s got an event calendar that can be used by the different groups that run events, it’s got forums, it’s got user profiles and the best thing is it’s taken a lot less time to get to this point than if I’d done the whole thing myself from the ground up.

What it doesn’t have is the right look. And a few of the features are in the wrong places. But that’s fine, that’s fixable. Those things are simple jobs to be done next.

And I’m really excited at the prospect of doing them.