The Mechanical Turk

After a long gestation period, we’re now unveiling the new look—and engine—for the Mechanical Turk.

The look

The basis for the new look is of course our corporate site with a few tweaks. The masthead is based upon the back of our business cards—we are very fond of our bonsai logo.

We also took the opportunity to widen the display from our corporate look—there’s a lot of evidence that shows 1,024 x 768 pixels is now the most common resolution. In due time our corporate site will also reflect this wider presentation—which might be good, because we know the site is a bit text-heavy. Perhaps the change will make this appear less of a problem!

The engine

Previously, we had used the Typo engine for our blog. It’s lightweight & has some nice features, but despite using the most recent release, we found it to be very unstable. For instance, if I discovered, after posting, that I had made a typo (ha ha) & I corrected it, Typo’s lighty processes would eat up all the resources on the server, grinding everything to a halt. After restarting lighty processes four times in a day, we just decided that enough was enough & that we needed to switch.

But to what? If you want to run a blog on Ruby on Rails, it appears that there are three contenders: the aforementioned Typo, SimpleLog & Mephisto. While SimpleLog looks really nice, it appears that it does not have a big developer community. Mephisto does, even though there are a few worries about its future development. Apparently there is now a core team for Mephisto, so perhaps it’s got a new lease on life. Time will tell. In the meantime, Ian has tinkered under Mephisto’s hood & when he has a moment (ha ha, aren’t we funny?), perhaps he’ll write a bit about what he’s done. We are considering releasing some Mephisto plugins as well—again, in time.

The name?

As noted in the comments to my second post on the blog, we are considering a new name. We’re coming around to a name revolving around “falling leaves”—suggested by the image of the bonsai on the masthead. It’s also somewhat evocative of the nature of blog posts—posts fall when they’re ready and… OK, I’ll shut up already.

Finally, I’d like to publicly thank Ian for taking time out of his horrendously busy schedule to help get the new blog engine running so smoothly.

We have a client that is part of a large institution (both shall remain nameless to protect them). Like many in such cases, they’ve had a CMS thrust upon them. I won’t name the CMS, but you might be able to figure it out from this screenshot:

Behold in all its horror!

Look at this from a user’s perspective. OK, we’ve got a nested system of some sort. What are the radio buttons for? We can also delete “publication details” (notice the wastebasket at the right). Below the wastebasket is a tickbox. What does it do? And do you dare click it? How about the “temporary save” button—why is this not automatic, as in online apps such as GMail & Basecamp?

Let’s look at another example:

“Once more”? “Switch”? WTF?

This second screenshot comes from further down the page, from a list of publications. Clicking the “Switch” button lets you move the publication up or down the list; when you do this, the page reloads. It would reload every time you moved the position of the item (average page refresh time: 30 seconds).

Now imagine that you’re re-organising a list of, say, 20 publications so that the most recent publications come first in the list—a not uncommon task. Imagine that, previously, the publications were listed in order of publications (i.e., that the oldest one was first). To just move the latest publication, you’d have the page refresh 20 times. After each refresh, you’d have to scroll down the page to find the publication again so you could move it up one more position. Imagine that you have to re-order several pages’ worth of publications: now do it a million times!

After doing this for a couple of hours, finally imagine saying “Fuck it, it’s not worth the effort”. You have now reached the position of my client—and, indeed, the position of many people who have had CMSs thrust upon them.

A lot of things have gone wrong here. At my bitchiest I’d say that these screenshots showed the unholy marriage of Windows 1.0-style software design with all the flair of a Linux user interface. How did it come to this? Why was this software chosen? How do you get out of this situation?

I don’t have an answer to that last question, unfortunately—large institutions make their decisions on behalf of all departments & it takes years to move such a big ship. But I know how others can avoid this.

Let’s gain some perspective: CMSs became a big thing before the dot-com bubble burst in the early 2000s. They were designed for the browsers of the day—Internet Explorer 4 & 5, perhaps even Netscape 4. Any web developer will tell you that these browsers are pretty limited—and that support of legacy browsers quickly reduces the options you have to design an intuitive interface, progressive enhancement techniques notwithstanding. So, CMSs were designed in an era when you couldn’t have asynchronously-updatable data on a page—yes, we’re talking AJAX here. In other words, many CMSs come from a less technologically-advanced world.

Mixed in with this is the glacier-like pace of large organisations in their technological adoption. Years ago I worked for a university & saw first-hand why the technological decision-making process took years: often for legitimate reasons. For example, if you are running a corporation with 10,000 computers, it takes a long time to produce all the tests necessary to ensure that you’re not going to bring down the entire corporate system in your enthusiasm to adopt new technology. Accordingly, many corporations start looking at tech products & it can take years for them to reach their decisions. (NB: many corporations are a lot better at making swifter decisions now, but it’s a rare academic institution that has matched this quicker pace.)

So let’s recap: many CMSs were originally built in a time when dinosaurs walked the earth & this is when large corporations might have begun eyeballing their products. Years go by, the corporation makes its (now dodgy) choice & then the corporations’ employees are stuck with the result for years to come—perhaps simply to justify a decision that cost millions of pounds even when that decision clearly was a bad one.

A major part of the problem is the decision-making process: after a certain cut-off point, no other options may be admitted. This is unfortunate (to say the least) because in the meantime blogging software has emerged not just as a niche player but as a robust solution—which solves many of the problems that CMSs were supposed to tackle. Publication control & template-based “unified” branding cross-site are allied with a significantly better interface. Moreover, many blogs allow for “static” pages (in this context, pages note defined by the date of the post but by their content). And while blogs may not automatically provide all the semantics required for, say, publications, most blogging software allows you to delve into markup to address specific concerns—or have plugins that would address the concern—or (as a last resort), you can tag content according to its semantic meaning. It’s a very good solution that solves 90% of the problems which CMSs were to address.

Now this doesn’t leave our clients in a good place—they still have to use a terrible CMS. My question is: why hasn’t the CMS vendor updated the working of its CMS? At the very least AJAX-based tools to move publications (as in the example way above) would ease so much of the pain. A less charitable person than myself would conclude that they don’t care that their software is so broken, so there is little impetus to fix it. A more charitable conclusion would be that they don’t realise that the software is broken. Perhaps when clients don’t renew their licenses they’ll wake up.

If you've ever subclassed a controller to deal with a particular MIME type, say FBML, you will have had the painful experience of having to copy and paste all of your contoller's actions, adding, or replacing the fbml blocks into the code.

This doesn't feel right - and it isn't. If you have to change some controller logic - you have to remember to change it in all these subclassed controllers.

Enter response_for (rdoc here).

response_for lets you declaratively specify how your actions should respond to particular Mime types. This helps particularly when your action's logic is not simple (like a CRUD action - but it still helps there). Here's an example I just made up:

class ReportController < ApplicationController
  # this puppy gets some models to collaborate to form a report
  def report
    @report = {}
    @report[:invoiced] = Customer.find_invoiced
    @report[:outstanding] = @report[:invoiced].select {|c| c.latest_invoice.outstanding? }
      
    respond_to do |format|
      format.html 
      format.xml { # some funky xml stuff here }
    end
  end
end

Let's say I wanted to make FBML responding version of this controller, normally, I'd have to copy/paste this action and add the fbml responses. With response_for, I do this:

  class FacebookReportController < ReportController
  response_for :report do |format|
    format.fbml { # my funky fbml stuff here, accessing @report}
  end
end

So, down the line, if I change the way my @report object is generated, I don't have to change my subclassed controllers.

Another example: create

You can choose to replace the respond_to block of the action as well. This means that common actions can be written, and the response can be adjusted according to your needs (resources_controller comes to mind).

Here's a standard create action:

class ForumsController < ApplicationController
  def create
    @forum = Forum.new(params[:forum])
    @forum.save
      
    respond_to do |format|
      if @forum.valid?
        flash[:notice] = 'Forum was successfully created.'
        format.html { redirect_to(@forum) }
        format.xml  { render :xml => @forum, :status => :created, :location => @forum }
      else
        format.html { render :action => "new" }
        format.xml  { render :xml => @forum.errors, :status => :unprocessable_entity }
      end
    end
  end
end

Let's say you want to create an FBML only version of this:

class FacebookForumsController < ForumsController
  response_for :create, :replace => true do |format|
    if @forum.valid?
      format.fbml { # funky fbml stuff for valid record }
    else
      format.fbml { # and for an invalid record}
    end
  end
end

One last example

For many actions, you just want to tell your action to respond to particular MIME type and a template does the rest:

class FacebookUsersController < UsersController
  response_for :index, :show, :types => [:fbml]
  end

You don't need to have a respond_to block defined in your actions - the parent controller could look like:

class UsersController < ApplicationController
  def index
    @users = User.find :all
  end
  
  def show
    @user = User.find params[:id]
  end
end

resources_controller + response_for

For those users of resources_controller, response_for gives you an easy way of adding responses on an ad-hoc basis (like if only one or two of your actions need to be RSS feeds or summat)

This is the first release...

and I welcome any bug reports

Steve Ballmer, chair-throwing CEO of Microsoft and

The Creature from Mel Brooks’
“Young Frankenstein”?

Update: Well, this is turning into something of a meme. One of the comments in the Fake Steve Jobs blog has a link to this awesome video:

I am very impressed.

Jason Lee of Big First Name just posted his experiences of using resources_controller on the google group.

He writes:

I've just converted one my existing apps ( http://big.first.name ) to using the resources_controller plugin and thought I'd share some results with the group...
before RC, the output of "rake stats"...

+---------------+-------+-------+---------+---------
| Name          | Lines | LOC   | Classes | Methods 
| Controllers   |  1785 |  1324 |      20 |     112 

after RC conversion (with only minor refactoring)...

+---------------+-------+-------+---------+---------
| Name          | Lines | LOC   | Classes | Methods 
| Controllers   |   877 |   655 |      20 |      45

That's 50% less code and 60% less methods. I've still got some chunky controller code lying around but overall I'm very happy with the result.

That's cool!

I gave a talk about resources_controller at LRUG which was great fun.

Thanks to everyone for listening and for the feedback. Skills Matter are apparently going to post a video of the presentation at some point. In the meantime, the slides are here.