Earlier, I wrote about my pains with having many Firefox tabs open for Google Mail, Reader (news), and Calendar. The post included several ideas that could be used to move towards satisfying my pains. However, I feel that out of those ideas was a general cop out.

• User Interface: This will be hard!!!

To fix this glaring issue I spent some time thinking about how I would combine the the interfaces for Google Mail, Reader, and Calendar. I have made a very basic mock up of how I feel it could work.

I’m not posting screenshots of the existing applications, but if you’re interested, and don’t already know, you can preview the Google interfaces @ About Gmail, Google Reader – Tour, Google Calendar Tour.

I said it was very basic.

This only works for 1024×768, and larger desktops. For people using 800×600, I would suggest moving the calendar bar to the top left, and rotating it (i.e. make it horizontal). The agenda could also be collapsible to reduce vertical scroll to access content. Fitting in the AdWords panel for email might be interesting, but there is a lot of open space in the header bar, or it would work into the content panel somehow.

As I said in the previous post, I’m not sure exactly what the Terms of Service and Program Policies actually say about this. But to er on the safe side, I would presume that only Google can roll this sort of Mash Up. In a lot of ways that makes sense. We are not meant to reformat search results, for good reason, they are Googles work, they deserve the recognition. Of course if we kept the Google branding, well, maybe that could work. Similarly Google accounts are extremely generous for a free service, if the policies did not prevent this, any Bob or Alice could simply serve reformated Google accounts to their users, and make a significant profit. (It’s easy to setup postfix as a mail relay, so they could hide the email address, from most users anyway).

I’d like to see something like this come out of Google, I do have it pretty good though, considering what I’ve got for free. It’d be nice, but I can deal without it.

Everyday I have two Firefox tabs open permanently. One for Google Mail (Gmail) and one for Google Reader. Usually I will open another tab for Google Calendar. Occasionally I will open another tab, again, for Google Maps and/or Google Documents. That is 4+ Firefox tabs regularly running a Google Web Application, at least 3 everyday. Three Tabs!!! I would love to have one Web Application that managed my email, news and calendar. From a user interface point of view I accept Google Maps and Google Documents being separate, but the other three, surely a MashUp could be made. Still leaving them individually accessible, but, simultaneously, also delivering a combined interface where preferred.

Another thing? Why use frames? Really? Why use pseudo links that do not involve the anchor (<a>) tag? There are other issues too I would guess. To quote the Code of Conduct: “Our goal is to build products that organize the world’s information and make it accessible to our users”. The key word here is organize. Semantic nonsense is not organization. The information is definitely accessible, maybe not to everyone, but “our users” covers that. Not by much though. Organization implies there is some meta-data, and in this case the meta-data should be a semantically valid, preferably XHTML, document. But I digress.

There are other products that provide the above combined interface. Zimbra and Open-Xchange come to mind, and there would be many others. Also, I could use Greasemonkey to merge Gmail and Google Reader, but that is by no means a complete solution. My main issue with both Zimbra and Open-Xchange comes from the enterprise target. They are suitable for large scale deployment, not just for my personal use. Also, Google manages all of the server backend, Google looks after storage, etc… I don’t need to run a full blown infrastructure to manage and store my data.This is a weak argument, perhaps, but really, all I want is to access my information. Google wants to make information accessible to me, and I’m happy to let them. There is no need for me to compete. However, I would like to decide how to access that information, and I would also like it to be organized in a way that suits me.

While the analogy: Having my cake and eating it too; is apt, I would like to try and redeem myself. Here are some ideas that might move towards combining the three applications.

1. Ditch the frames. Each of the applications have a heavy dependency on JavaScript. Use AJAX/AJAH instead.
2. Make the applications more backward compatible, use anchors and real input buttons. Server side this may mean deciding how much the content needs to be wrapped, but it is possible.
3. Widgetize the interface. Let users reposition elements, similar to a the Personalized Home.
4. Better temporary offline support. If the user loses connectivity while writing an email, cache it locally for sending later. Naturally this will fail if the user closes their browser. Hmm, maybe not then. Make this the lowest priority.
5. User Interface: This will be hard!!!
   • Don’t want internal tabs, that defeats the purpose.
   • Question: How do you fit that much information into one screen?
   • Answer: You can’t! But you can fake it.

As a side note: libraries such as libgmail provide an interface to Gmail and Googles GData API interfaces with Google Calendar. A MashUp should technically be possible. I’m not sure if that violates Googles Terms of Service though. It’s definitely a very fine line.

Today I want to talk about a few things that I’ve been working on lately.

1. A CMS; and
2. A web-based outreach complimented by cheap publishing

First though, I would like to complain about domain names. Part of the web-based outreach includes purchasing a suitable domain name: you know the sort, something catchy, to the point, really draws attention to your website. Sounds simple, there is only one problem. Any domain name consisting of a catchy phrase, or word is usually taken.

Now, don’t get me wrong, some of these domains are taken quite legitimately, either for commercial or individual purposes. Some may not even offer a website. I do not have a problem with these domains. At least they are being used. It is the domains that in an advertisers portfolio, the ones that all look the same, but have slightly different content for keywords in the domain name. Using domains in this manner is very, very, annoying. They’re usually for sale too. At the right price.

Perhaps an example would work better: godsword.com!!! While it would be somewhat amusing, I would not use the domain godsword.com – I need something less automatically associated with “religion”. However, even if preached incorrectly, a Christian message through godsword.com would be a useful, productive, and good use of the domain.

Of course this is nothing new, I’m not the first to complain, and surely I wont be the last.

Web-based Outreach
Web-based advertising is fairly ubiquitous. If you have searched with Google, you’ve seen it in the form of AdWords, and there are many more examples. Recently I have been working on a strategy to extend the web-advertising platform into a successful community awareness outreach. Measuring success is hard, but my chosen metric, at least in this case, is a measure of population who are conscious of the product existing. So this is what I had in mind:

1. Run several simultaneous outreaches.
2. Produce a web portal for each outreach, all through one website.
3. Use AdWords and similar technologies to build Internet traffic.
4. Compliment with paper publishing. People who see 100 eye-catching posters in a day are likely to research the posters origin (sorry no reference, just an opinion).
5. A website branded information request form, allowing interested parties and the outreach coordinators to communicate.

Here are a few of my ideas

• The basic idea would be to explain the lifestyle (and benefits).
• Essentially outlines tangible benefits received by allowing God influence in your life. [Earth courtesy NASA’s Earth Observatory.]
• A very quick, literally 5 minutes, overview of beliefs.

Content Management System
I’ve also been working on a CMS. Yes, I know there are two-thousand-one-hundred-and-seventy-four CMS packages available already, many of which work wonderfully, and many of which are open source. I tried out a few, even tried extending one, in the end I felt writing from scratch would meet my requirements better. So what did I want:

• Manage multiple unique sites (one for each Revival Centre location and then some extras).
• Minimally provide contact details, meeting times and meeting locations for each assembly.
• Provide a news feed on each site, eventually available to news readers (i.e. RSS, etc…).
• Manage unique page content, an event calendar and a set of photo galleries for each site.
• Appropriate user access control: user task filtering for each site, with a single sign on to manage all accessible sites.
• XHTML1.1 Compliance (I have since changed this to mean XHTML1.1 compliance in browsers that support the appropriate mime type. Other browsers receive XHTML1.0 Transitional, however the markup should validate as XHTML1.1 if the xml and doctype declarations are modified.).
• As the above implies: CSS styling, and for kicks, support for multiple themes on the front end.

The theming, access control, and XHTML1.1 compliance were hard enough to find, but to offer that for multiple sites… well as I said, writing from scratch appeared the best solution. This was a daunting task, don’t undertake it lightly, especially not in your free-time. I’m happy to say, that after 8ish weeks of on and off development and a few weeks off over Christmas/New Year, I have a product. This week I am going to undertake the task of porting the existing Newcastle Revival Centre website, fingers crossed there no more major bugs.

After deployment I might write about the steps involved in creating a CMS, this post is already too long. So, here is a list of some of the underlying packages I used to get this done:

• PEAR. Specifically QuickForm (extended), Mail, Calendar, and Sigma templating.
• The excellent famfamfam icon set.
• jQuery.
• Some fancy AJAX/AJAH to make a photo gallery similar to LightBox but without the pop-over effect.
• TinyMCE.

Wow! This is a really long post. I’d really like some feed back on the web-based outreach idea. If you have domain suggestions email them to me, otherwise feel free to comment.

Andrew