Finally, there is a book available on MODx – it’s about time! MODx is an open source PHP CMS with an awful lot going for it. While gaining in popularity, it’s nowhere near as well known as the likes of Joomla or Drupal. Hopefully, this book is the first of many on the subject and will help raise its profile and encourage more people to give it a go.

I first came across MODx over 3 years ago, when online documentation was somewhat sparse. I could see great potential in it back then, but after using it for a couple of sites, my work changed and I didn’t have a need for it. More recently, I’ve been working on more complex projects where solutions like MODx come into their own. So this book could not have come at a better time for me. With a somewhat hazy recollection of MODx, I wanted to get up to speed on the fundamentals quickly and this book helped.

This book is most suited to those new to MODx and is also appropriate for those new to content management systems in general. The pace is fairly gentle for the most part. The first three chapters (forty pages) just cover a little background and setting up a local web server (XAMPP) and getting MODx up and running.

Next comes 28 pages devoted to templates. This chapter conveys MODx’s ease of templating, one of its major strengths. You have a basic blog up and running by the end of this chapter, without the content getting technical. The book takes the approach of showing you how to get common functionality up and running, without getting bogged down in explaining what the small pieces of code needed mean. The technical details are saved for a few more advanced chapters near the end of the book. This approach frustrated me a little, since as a coder I like to know exactly what everything is doing. However, I can see the merit of this approach for a less technical audience, which is the main market for this book.

The next three chapters (56 pages) cover authentication, content aggregation (the Ditto snippet) and creating lists and navigation (the Wayfinder snippet) respectively. These are the bread and butter of most websites, and the book provides a good grounding in these techniques. Chapter 5 on Authentication and Authorisation is available as a free download from the Packt website.

As an experienced web developer, chapter 8 on snippets was where things started to get a bit more interesting. This is where some of the technical gaps left in previous chapters start to be filled in. The content is still quite accessible, focusing on explaining how to install and use snippets.

Chapter 9 introduces Place Holders Extended or PHx, which was not only new to me, but also my colleague who has built a few MODx sites recently. So, while much of the book caters for newbies, there is the odd nugget in there covering lesser known features or more advanced material.

Chapter 10 brings the focus right back to the practical, covering popular modules for adding common functionality to websites. It looks at the SMF module (for integrating MODx with a SMF forum), the MaxiGallery image gallery snippet, the built in eForm snippet for creating email forms, the WebLoginPE snippet for implementing user profiles and, finally, how to show similar articles using the old reliable Ditto snippet.

Creating Snippets is covered briefly in Chapter 11, and Chapter 13 looks at Plugins and Modules, including how to create a very simple Plugin. These chapters are both just 16 pages long and I would have preferred some more coverage on these areas. This would not be expected from a beginner book, but since it delves into more advanced topics in places, it would have been nice to see a bit more meat to the coverage.

I was delighted to see a chapter devoted to SEO, deployment and security (ch 12); all vital real world areas, but so often neglected, particularly in introductory books. Alas, aside from information on migrating your site to a live server, there is little of note here. Several general SEO guidelines are included, but I feel that the SEO advantages of using MODx relative to other content management systems, which can often cause problems with search engines, would have been more beneficial.

Another gripe is the amount of pages taken up with large excerpts of HTML code, often only differing from previous excerpts by a line or two. Surely this is overkill, even for newbies. The writing also appears somewhat formulaic or clunky in places, but in technical books, that’s less of a failing. I did notice some typos along the way too, which is more unfortunate, but most were fairly obvious.

There is a very strong MODx community, and some excellent tutorials available online. The Coding Pad has a good list of these. While many find it easy to get up and running using these resources, for others, like me, there is no substitute for a good book. Although I don’t get much time to read these days, I have a large collection of computer books.

In this context, I would describe this book as “OK”. It’s not one of those dry, complicated tomes which I never got into. It’s quite an easy read, and gives you confidence about building real websites with MODx, and perhaps even creating your own snippets if needed. On the other hand, it’s not a bible which I will refer to again and again, but that’s not its aim. Overall, while a little clumsy in places, it’s a good introduction to MODx for new users or developers looking for a refresher course.

Disclaimer: Packt Publishing kindly gave me a free copy of this book to review.

Both scripts allow you to add impressive functionality to websites with very little effort. While the built in behaviour of Galleria is often ideal, I had a requirement to integrate lightbox functionality into it recently. I would consider myself a jQuery newbie, so this may not be the most elegant approach, but it seems to do the job on my browsers at least.

Galleria allows you to show a list of thumbnails, along with a larger version of one thumbnail. Clicking a different thumbnail enlarges that image instead. I needed to extend this functionality to give 3 image sizes for some of the gallery images: Clicking a thumbnail would give you a “medium” version, and clicking the medium version would show a larger version again in a lightbox pop-up.

I achieved this as follows:

1. Firstly, I made sure that I had included all the files needed by both Galleria and Slimbox in the head of my page. (The basic usage instructions on both script sites can fill you in on this.)
2. I gave any images in the gallery that needed a large lightbox version a longdesc attribute, setting it to the URL of the large version. (Not ideal semantically I know, but this attribute does expect a URL, and in this case the detailed information about the image is just a bigger, more detailed image, so I can live with that.)
3. Next, I made sure that the clickNext option in my Galleria call was set to false. The clickNext option makes each main gallery image click through to the next to provide an alternative gallery navigation to clicking each thumbnail. However, I need to have some of my main images linking to lightbox pop-ups instead, so I disable this behaviour:
   
$('#mygallery').galleria({
history : false,
clickNext : false
});

4. Here’s the key step. I used the onImage option in my Galleria call, to specify a function to add the lightbox behaviour:
   
$('#mygallery').galleria({
history : false,
clickNext : false,
onImage: function(image,caption,thumb){
//#check if thumbnail indicates that a large version of this pic is available
var large = thumb.attr('longdesc');
if(large)
{
//#if so, link to it and show it in a lightbox
image.wrap('<a href="' + large + '"></a>');
image.parent('a').slimbox();
}
}
});

   A sample onImage function is provided in the Galleria advanced demo, which gave me my function outline. The parameters give you access to the thumbnail image, the main image and the image caption, giving you great freedom to extend the functionality. Here, all I want to do is check whether the thumbnail of this main image has a longdesc specified. If so, that means it needs the lightbox effect. So, I wrap a link around the main image, linking to the large version. Finally, I make a call to the Slimbox API to add lightbox functionality to the link I just added. (Note that I first tried just putting rel=”lightbox” in the anchor tag, but this seemed to have no effect. The rel attribute may have some other purpose in the Galleria script which overrides this.)

A few short steps and you’re done. This is the kind of thing that I thought would be a nightmare. For so many years, JavaScript gave me the shudders, but jQuery makes light work of so much. I’m using it increasingly, but still only now and then, with renewed awe each time. Warning: jQuery has been known to cause giddiness and excessive drooling in web developers.

Firstly, the current version (1.0) does not work instantly with jQuery 1.3. You get error messages like “Syntax error, unrecognized expression: [@rel=…”. This is easily fixed. You just need to open the Galleria file and replace @rel with rel, since @rel has been deprecated (there are only a few occurrences).

Once I was up and running, I realised that I needed to do something a little bit different with this implementation. On clicking a thumbnail, I wanted to replace the content of an existing div on the page with the large version of that image.

Galleria automatically shows the large main image above your list of thumbnails. Alternatively, you can specify an element to place the large image in by specifying the _insert option:


$('.galleria ul').galleria(
{
insert: '#banner'
}


The only problem was that the div I specified (#banner in the above example) already contained some content. I wanted to replace the content with the large image, but Galleria sensibly appends the large image after any existing HTML in the element.

Another quick change to the Galleria script file seemed to change the behaviour so that the large image container HTML is overwritten instead. In the minified version, just search for _wrapper.empty().append(_img); and add the following line beneath it:


_wrapper.empty().append(_img);
/*add line below to overwrite initial non Galleria markup*/
_wrapper.prevAll().remove();
/*end edit*/
_wrapper.siblings('.caption').text(_thumb.attr('title'));


Mashups are basically about taking data from 2 or more different sources, and making something new from it. Similarly to how you can mash 2 songs together and come up with a surprising result, lots of people take information available from multiple websites and make something interesting. There are lots of great examples at Programmable Web. I have never made a mashup myself, but I think it’s a cool idea and it was always something I meant to find out more about when I had time. So it was a good opportunity to get up to speed yesterday.

IBM spoke about their QED Wiki application and other software that facilitates non-programmers in making business mashups. You can download the IBM mashup starter kit with all the software. Martha Rotter from Microsoft spoke about Popfly which is aimed at the social end of the spectrum. You can easily display your Flickr photos, Twitter updates etc in novel ways and then easily share that info with other Popflyers or on Facebook etc. I find the mashups in between the hardcore business and social extremes to be most interesting – ones that have interest to consumers, but have a purpose beyond eye candy. For example, I love JustRoutes, which maps public transport routes for Dublin. Unfortunately, I missed the talk on OpenKapow, but I’ll defnitely be checking that out. While more and more sites are developing APIs, screen scraping is still the only way to get data from so many more. Scraping can be a major pain, and OpenKapow makes that much easier – good stuff!

The key point I learned from Mashup Camp was that there are lots of free user-friendly tools around now for making mashups quickly and easily. From the little I knew about Mashups before, I thought that it would take specialist knowledge, and often substantial time, to get a worthwhile mashup up and running. The opposite is true. I heard the term “situational apps” again and again yesterday. Whenever a situation arises where it would be helpful to have certain information to hand that is not currently available from one source, you can knock up a mashup in a matter of hours to meet this need. The only time consuming bit is finding the data that meets your need. Once you have the information in a suitable format, assembling a mashup takes minutes.

There are general and business mashup contests on as part of Mashup Camp. Before I went, I never considered entering as it’s an area I knew next to nothing about. But I left thinking that if only I had a decent idea for a mashup, I would give it a lash. Far from being sole the preserve of developers, mashups are open to anyone with an idea.

There are certainly a huge amount of open source content management systems out there. I found that the CMS Matrix was a great help in narrowing down my options. You can specify your requirements and it will return a list of applications that fit the bill. It lets you quickly compare all the suitable content management systems at a glance. Another life saver was Open Source CMS which lets you try out demo installations of a large number of content management systems which saves you the hassle of installing them all yourself.

On my travels I came across all sorts of content managers, ranging from extremely basic tools to the likes of Typo3. That seems to be the king of Open Source CMSs, but naturally its power and flexibility comes at a price: complexity! Their site is excellent though, and it is refreshing that they admit that if you don’t have at least a month to invest in learning the system, you should look elsewhere! I hope to delve into Typo3 a little deeper someday, to at least get an idea of what it is capable of. Then there were the popular portal friendly solutions like Joomla / Mambo and Drupal. These certainly seemed far quicker to get up and running with, but they did not seem to lend themselves to easy to the more corporate sites I was seeking to manage.

At last, I came across something special by chance. MODx is a PHP based CMS that was born out of another CMS called Etomite. It has not been around too long but already has a growing avid community following. I am particularly impressed by the following features:

• Strong support for web standards
• Facilitates creating search engine friendly websites
• Back end navigation tree mirrors front end structure
• Quick edit functionality enabling on the fly updating via browsing the front end
• Clean CSS navigation menus embraced!

What I like most about MODx is that it is relatively easy to get up and running, yet it is an extremely flexible CMS that packs a lot of power into its small core. Unlike many content management systems, there is no such thing as a typical MODx site. It is just as suitable for a corporate site as for a blog. Its functionality is being extended all the time thanks to the ever growing collection of plugins developed by its loyal developer community. This expansion is facilitated by the clean, elegant approach that the development team took in designing the application. It really is a pleasure to work with.

It’s not perfect! It’s young and it’s open source, so it’s not fully there yet. Documentation is lacking fin placings, and some common functionality, notably e-commerce capabilities, is still in the works. Bugs are present, but it is amazing how quickly they are fixed. The developer community is very special. No matter how many questions you have, someone is always happy to help at the forum.

I’m delighted that I found this gem of a content management system and if you are looking for a CMS that is quick and easy to get up and running but won’t hold you back, give MODx a try!

Update August 2006: A revised Movies@ site has now been launched. It’s still far from perfect, but it is a huge improvement on the original eyesore. I wish I had taken some screen shots, but you really had to browse the original version to truly experience the horror!

I have not got around to visiting the new multiplex cinema at the swanky Dundrum shopping centre yet. Having read about Blúire’s recent experience, I may not be missing out on much. I will of course give them the benefit of the doubt and wait until I see a film there myself before passing judgement.

However, I have already experienced the Movies at Dundrum website. The mere mention of it is enough to send a shiver down my spine! I realise that as you sit reading this post on this tragic pink barbie-on-speed website, the words pot, kettle and black may be entering your head! Web site design and usability are extremely subjective areas. Each to their own is my motto. I actually very rarely slag off websites. Others are constantly pointing out so called horrendous websites to me and I’m usually the one saying that it could be a lot worse. I think I will be using the Movies @ Dundrum site in future to illustrate how much worse. Look at the state of it!

I could build a very long list of what is wrong with this site but I don’t like to descend to that level and it would probably take me a while to compose my thoughts! Besides, I think that it really does speak for itself. Apart from the aesthetics and amateur appearance, more seriously, this site shows that having a website can sometimes do more harm than good. I understand that the facility to book online will be available shortly, but as far as the current site goes, I think that even an under construction page would be more beneficial and less damaging to their image. Whatever about “redefining cinemas“, Movies @ Dundrum need to look at rethinking their website.

Several visitors had reached me after searching at Yahoo! for quite generic phrases like freecycle, janine and cross stitch. On clicking through to the Yahoo! results pages, I saw that the searches had been made at Yahoo! UK and Ireland with the search in Ireland filter option selected e.g. search in Ireland at Yahoo! UK and Ireland for janine

If you search at Yahoo! for the same generic terms mentioned above using the search the web option my site is either ranked lower or has no significant ranking at all. Yet, a pages in Ireland search has me very near the top.

I was very surprised to see my site included in a regional search like this. Why? Well, I have a .com domain rather than a .ie one, and I host my site in the USA so my site does not have an Irish IP address. I understand that these are the major factors involved in appearing in pages from Ireland searches at Google.ie. I thought the same criteria was used for all major engines.

Yet, Yahoo! clearly recognises my site as being Irish. The question is how? Is their algorithm simplistic enough to think that if your content mentions the word Ireland enough you must be Irish? Clearly that would make things far too easy for spammers and it seems unlikely that Yahoo!’s process would be so flimsy. Are they doing something far more subtle, recognising certain authoritative hub sites and directories for Irish sites e.g. Planet of the Blogs, IrishBlogs.ie and deeming sites included in these to be Irish?

More likely, there is just a very straightforward reason for this that I have missed. I’d love to find out how Yahoo! decides which sites belong to a particular region, so please let me know if you’re in the know.

Since this site is the playground of the cream of the crop, the new look is being pored over by web monkeys the world over. My 2 cent ain’t worth anything but, like anyone, I know what I like. I have to say, the design is not exactly my cup of tea. It does look a lot classier than the older versions, and they definitely achieved the scholarly authoritative look. However, it’s a bit too staid for my tastes and I think that the home page lacks focus somewhat.

I’m probably just jealous because I will never be on the plain of such web gurus. I am sure that I will grow to love it by next week! In any case, I have already ordered my t-shirt – the very latest in web geek chic.

The service is in Beta at the moment (like most Google services!) and as such has limited support. Basically, there are references and lots of help docs, some sample code and a developer’s forum which is not exactly buzzing. The documentation is easy enough to follow and there is a fair amount of it. However, I found that it may not be complete.

I am only using the API with C# in very straightforward ways to gather reports, but I have already come across 2 occasions where following the instructions has not worked.

The first thing I had trouble with was setting the aggregationType for my reports. Although I was specifying:

myReportJob.aggregationType = AggregationType.Daily;

I found that I was stuck with the Summary view!
I only found the solution by chance when I saw a line of code I was missing in a post on the developer’s forum:

myReportJob.aggregationTypeSpecified = true;

I still cannot find this line mentioned in the documentation and it seems weird to me that you need it. Surely you are specifying the aggregation type by setting a value for it?!

I am still looking for a solution for my second problem. To use the AdWords API, you need to have a My Client Center (MCC) account. The idea is that the MCC account links all your AdWords accounts together in a central place. They all still exist separately, but the MCC provides a parent level of access where you can drill down to each account but also perform operations on multiple accounts together. This is useful for search marketing companies who manage AdWords campaigns for multiple clients.

Google recently added reporting functionality to the MCC, so that it is now possible to create just one report in the MCC that includes data for all accounts. Since API usage is limited to a certain quota of operations each month, depending on how many accounts you have and how much you spend, this feature is great. It means that you can just run one report instead of one for each account which is far more efficient. Except I cannot get this to work!

Update: I have seen reports on the developer forum that if you add a line ReportJob.crossClientSpecified = true, then generating a report at MCC level will work.

The relevant Adwords documentation states the following:

If you have registered with My Client Center (MCC) to administer
accounts on behalf of your clients, you can get reports about your
clients’ AdWords accounts. To do this, login without setting the
clientEmail header. When setting up the report job, you need to set the crossClient and clientAccounts fields. Set crossClient to true, and specify a list of the Ids of the accounts to include in the report as the value of the clientAccounts field.

If you set the crossClient field to true and do not supply a value for clientAccounts, the generated reports include all the client accounts

These instructions seemed simple enough, but when I use the following C# code while specifying the MCC account login details, all I get is empty report columns with zero totals:


//#instantiate Report service
ReportService service = new ReportService();

If I specify login details for an individual account instead, I get a complete report which suggests that my code must be OK in the main. Maybe I have missed something in the docs or perhaps there is another magic line of code I’m missing. At the moment, I have to run a report for each account so if you know where I am going wrong, please enlighten me!

A final niggle I have found is that even though I just want to replicate running a Keyword report, I have to run a Custom report to get the info I need. This is because even though you can check a box on the UI to include conversions for a Keyword report, the same option cannot be set in the API, unless the docs forgot to mention that as well. This is a minor gripe, but Custom reports are slightly longer and messier to code as you have to specify all the columns you want etc. It would be nice if the API mirrored the UI. Perhaps when (if?) the service comes out of Beta the 2 will be perfectly in sync.

Overall, it is fantastic that Google have made an API for AdWords freely available and are continually improving and extending its functionality. It is just frustrating when you run into issues and the chances of resolving these quickly is drastically low due to the lack of real time support.

Useful links:

• AdWords API developer forum
• AdWords API support
• AdWords API blog
• AdWords API home page
• Webmaster World AdWords forum
• Digital Point AdWords forum
• SEO CHAT AdWords forum

Adobe, whose products include the industry standard graphics software, Photoshop, and the pdf format, will now own the Flash format and add the standard web page editor Dreamweaver and graphics packages Fireworks and Freehand to its offerings.

While I’ve no gripe with Adobe, this buyout places an awful lot of power in their hands. I think it is vital to have a number of major players in the web development software industry or any other for that matter. It will be interesting to see how this works out.

On a personal level, I’m most concerned about what will become of Macromedia Fireworks. While I am not a designer and have very limited graphic skills, Fireworks is my weapon of choice. For some reason, I just cannot get my head around Photoshop full stop! I realise that a lot of my issues may be psychological, but I know I am not alone in thinking that the program is not the most intuitive. For making simpler web graphics, I believe Photoshop is not the best tool. Fireworks was developed specifically for that purpose and it shows. However, with most of the world now used to only using Photoshop (and its hefty price tag), I fear that Adobe will get rid of Fireworks. Time will tell.

I suppose it could be worse – Microsoft could have bought Macromedia!