Category Archives: Code

Software Development

Conway’s Law & Distributed Working. Some Comments & Experience

The eye-opener in my personal experience of Conway’s law was this:

A company with an IT department on the 1st floor, and a marketing department on the 2nd floor, where the web servers were managed by the marketing department (really), and the back end by the IT department.

I was a developer in the marketing department. I could discuss and change web tier code in minutes. To get a change made to the back end would take me days of negotiation, explanation and release co-ordination.

Guess where I put most of my code?

Inevitably the architecture of the system became Webtier vs Backend. And inevitably, I put code on the webserver which, had we been organised differently, I would have put in a different place.

This is Conway’s law: That the communication structure – the low cost of working within my department vs the much higher cost of working across a department boundary – constrained my arrangement of code, and hence the structure of the system. The team “just downstairs” was just too far. That distance was composed of gaps & differences in priorities, release schedules, code ownership, and personal acquaintance.

Conway’s Law vs Distributed Working

Mark Seemann has recently argued that successful, globally distributed, OSS projects demonstrate that co-location isn’t all it’s claimed to be. Which set me thinking about communication in OSS projects.

In my example above, I had no ownership (for instance, no commit rights) to back end code and I didn’t know, and hence didn’t communicate with, the people who did. The tools of OSS—a shared visible repository, the ability to ‘see’ who is working on what, public visibility of discussion threads, being able to get in touch, to to raise pull requests—all serve to reduce the cost of communication.

In other words, the technology helps to re-create, at a distance, the benefits enjoyed by co-located workers.

When thinking of communication & co-location, I naturally think of talking. But @ploeh‘s comments have prodded me into thinking that code ownership is just as big a deal as talking. It’s just something that we take for granted in a co-located team. I mean, if your co-located team didn’t have access to each other’s code, what would be the point of co-locating?

Another big deal with co-location is “tacit” knowledge, facilitated by, as Alistair Cockburn put it, osmotic communication. When two of my colleagues discuss something, I can overhear it and be aware of what’s going on without having to be explicitly invited. What’s more, I can quickly filter out what isn’t relevant to me, or I can spontaneously join conversations & decisions that do concern me. Without even trying, everyone is involved when they need to be in a way that someone working in a separate room–even one that’s right next door–can’t achieve.

But a distributed project can achieve this too. By forcing most communication through shared public channels—mailing lists, chatrooms, pull request conversations—a distributed team can achieve better osmotic communication than a team which has two adjacent rooms in a building.

The cost, I guess, is that typing & reading is more expensive (in time) than talking & listening. Then again, the time-cost of talking can be quite high too (though not nearly as a high as the cost of failing to communicate).

I still suspect that twenty people in a room can work faster than twenty people across the globe. But the communication pathways of a distributed team can be less constrained than those same people in one building but separated even by a flimsy partition wall.

References

IIS Express : Run a child web application in a virtual directory under a parent application

Like this: Edit your IIS Express config file at

"%userprofile%\My Documents\IISExpress\config\applicationhost.config"

Create a site which has two applications defined in it, e.g.

<site name="MyTopLevelAndChildWebAppsInOneSite" id="123" >
    <application path="/" applicationPool="Clr4IntegratedAppPool">
        <virtualDirectory path="/" physicalPath="C:\Users\me\Source\TopLevelWebApp" />
    </application>
    <application path="/Child" applicationPool="Clr4IntegratedAppPool">
        <virtualDirectory path="/" physicalPath="C:\Users\me\Source\ChildWebApp" />
    </application>
    <bindings>
        <binding protocol="http" bindingInformation="*:51234:localhost" />
    </bindings>
</site>

And then run the site, matching it on the siteid:

start "Woo!" "C:\Program Files (x86)\IIS Express\iisexpress.exe" /siteid:123

Browse to, and close, your web apps in the usual way from the IIS Express icon in the systray.

Optionally, experience the pain that is web.config inheritance. But try not to.

An Asp.Net MVC HtmlHelper.RadioButtonsFor helper

These overloads will do the hopefully-obvious thing with a Model => Model.Property and a List of SelectListItems (or a single SelectListItem)

Use the RadioButtonLabelLayout setting to control whether you nest the radio button inside its label, or lay them out as siblings; and whether you like text before button or vice versa.

    public static class HtmlHelperRadioButtonExtensions
    {
        /// <summary>
        /// Returns radio buttons for the property in the object represented by the specified expression.
        /// A radio button is rendered for each item in <paramref name="listOfValues"/>.
        /// Use <paramref name="labelLayout"/> to control whether each label contains its button, or is a sibling, 
        /// and whether button precedes text or vice versa.
        /// <list type="bullet">
        /// <item>
        ///     <term>Example result for the default labelLayout= RadioButtonLabelLayout.LabelTagContainsButtonThenText:</term>
        ///     <description>
        ///     &lt;label for="Object_Property_Red"&gt;&lt;input id="Object_Property_Red" name="Object.Property" type="radio" value="Red" /&gt; Red&lt;/label&gt; 
        ///     &lt;label for="Object_Property_Blue"&gt;&lt;input id="Object_Property_Blue" name="Object.Property" type="radio" value="Blue" /&gt; Blue&lt;/label&gt; 
        ///     </description>
        /// </item>
        /// <item>
        ///     <term>Example result for labelLayout= RadioButtonLabelLayout.SiblingBeforeButton:</term>
        ///     <description>
        ///     &lt;label for="Object_Property_Red"&gt;Red&lt;/label&gt; &lt;input id="Object_Property_Red" name="Object.Property" type="radio" value="Red" /&gt;
        ///     &lt;label for="Object_Property_Blue"&gt; Blue&lt;/label&gt; &lt;input id="Object_Property_Blue" name="Object.Property" type="radio" value="Blue" /&gt;
        ///     </description>
        /// </item>
        /// </list>
        /// </summary>
        /// <param name="listOfValues">Used to generate the Value and the Label Text for each radio button</param>
        /// <param name="labelLayout">One <see cref="RadioButtonLabelLayout"/> to control the layout of the rendered button and its label.</param>
        /// <returns>
        /// An MvcHtmlString for the required buttons 
        /// 
        /// </returns>
        public static MvcHtmlString 
            RadioButtonsFor<TModel, TProperty>(this HtmlHelper<TModel> htmlHelper,
                                                    Expression<Func<TModel, TProperty>> expression,
                                                    IEnumerable<SelectListItem> listOfValues,
                                                    RadioButtonLabelLayout labelLayout = RadioButtonLabelLayout.LabelTagContainsButtonThenText)
        {
            if (listOfValues == null) { return null; }
            var buttons= listOfValues.Select(item => RadioButtonFor(htmlHelper, expression, item, labelLayout));
            return MvcHtmlString.Create(buttons.Aggregate(new StringBuilder(), (sb, o) => sb.Append(o), sb => sb.ToString()));
        }

        /// <summary> Create an <see cref="IEnumerable{T}"/> list of radio buttons ready for individual processing before rendering</summary>
        public static IEnumerable<MvcHtmlString> RadioButtonListFor<TModel, TProperty>(
                    this HtmlHelper<TModel> htmlHelper,
                    Expression<Func<TModel, TProperty>> expression,
                    IEnumerable<SelectListItem> listOfValues,
                    RadioButtonLabelLayout labelLayout)
        {
            if (listOfValues == null) { return new MvcHtmlString[0]; }
            return listOfValues.Select( item =>  RadioButtonFor(htmlHelper, expression, item, labelLayout) );
        }

        public static MvcHtmlString RadioButtonFor<TModel, TProperty>(
                                HtmlHelper<TModel> htmlHelper, 
                                Expression<Func<TModel, TProperty>> expression, 
                                SelectListItem item, 
                                RadioButtonLabelLayout labelLayout)
        {
            var id = htmlHelper.IdFor(expression) + " " + item.Value;
            var radio = htmlHelper.RadioButtonFor(expression, item.Value, new {id}).ToHtmlString();
            var labelText = HttpUtility.HtmlEncode(item.Text);
            TagBuilder nestingLabel = null;
            switch (labelLayout)
            {
                case RadioButtonLabelLayout.LabelTagContainsTextThenButton:
                    nestingLabel = new TagBuilder("label") {InnerHtml = labelText + " " + radio};
                    return MvcHtmlString.Create(nestingLabel.ToString());
                case RadioButtonLabelLayout.LabelTagContainsButtonThenText:
                    nestingLabel = new TagBuilder("label") {InnerHtml = radio + " " + labelText};
                    return MvcHtmlString.Create(nestingLabel.ToString());
                case RadioButtonLabelLayout.SiblingBeforeButton:
                    return MvcHtmlString.Create(radio + " " + htmlHelper.Label(id, labelText));
                case RadioButtonLabelLayout.SiblingAfterButton:
                    return MvcHtmlString.Create(htmlHelper.Label(id, labelText) + " " + radio);
                default:
                    throw new ArgumentOutOfRangeException("labelLayout", labelLayout, 
                                "This is not a valid RadioButtonLayoutStyle for rendering a radio button");
            }
        }
    }

    public enum RadioButtonLabelLayout
    {
        LabelTagContainsButtonThenText = 0,
        LabelTagContainsTextThenButton = 1,
        SiblingBeforeButton = 2,
        SiblingAfterButton = 3,
    }

IIS Express GUI

Looking for a admin GUI so I can start/stop IISExpress websites I quickly found iisexpressgui on codeplex/ which is by Matteo Tontini and iis express manager. They both do pretty much the same thing. They also both only pick up only the first binding for a site, so if you’ve got two — for instance one for http and one for https — then you might need to edit your %Documents%\IISExpress\config\applicationhost.config to choose the one you want to use.

The Yes, The No and the Painful: using, and failing to use, estimates for a no-go decision

@AgileKateOneal recently asked for examples of effective estimate use in medium/long-term planning, and making a no-go decision sprang instantly to mind.

Many such decisions are casual and quickly forgotten: the back of an envelope calculation which says that an idea is well beyond what we can afford; and the conversation moves on. An NoEstimator might object that one could profitably try out something rather than nothing, such is sometimes true but creative thinkers in commerce & IT can always generate a hundred more ideas than a team can try out; you can’t try out everything.

But a couple of more specific examples from recent work:

Example 1: Small UK charity looking at CRM options

in November last year I worked with a small UK charity, www.redinternational.org who were badly in need of some kind of CRM software to keep in touch with supporters and project partners. They were running largely on spreadsheets built from downloaded reports from virginmoneygiving.com / mydonate.bt.com etc. They also had an Access database with a fair amount of donor & similar data in it.

Question: Is it better to pay for a CRM solution — typical charity starting price £10,000 going up to easily £100k – or get someone to do enough work on the Access database to make it a usable solution?
My Answer: I first spent some time discovering and documenting their main use-cases (to clarify: their ‘business’ use cases, that is the things the charity had to do whether manually or with IT). I gave that picture to the CRM providers so that they could give us a sensible proposal. And I worked out an estimate for extending/developing the Access database. Based on that, we could see that a CRM consultancy/solution looked like £10-£20k (5 year cost) and the DIY-option about 200-400 developer days.

Even with this level of accuracy it was good enough to see that DIY should be a no-go. I did not expect this. I thought that the charity’s actual requirements were sufficiently small that we could do something useful for a few thousand pounds. But two hours spend going through their use-cases on my estimating spreadsheet showed me that I was wrong. So, I recommended the best value CRM option.

This, I think, is planning 101: a couple of hours working through the detail on paper is a lot cheaper than running the experiment; but can be enough to make a probably-good decision.

Example 2: Provide a system to automate a small team’s manual processes for a capped price

This was for a financial services company in 2013. The team were working on PPI claims for an insolvency practitioner (obliged to pursue potential claims that might bring in some money for their clients’ accounts) and had about ten thousand potential claims with hundreds per month being added. They had been working manually on spreadsheets for over a year.

I spent 4 days on analysis and listed a set of use-cases that covered the processes end-to-end; and I estimated that a suitable system could be done for about 40 days development work. The estimate cost about 3 or 4 hours on top of the analysis.

The contract to provide was capped-price. The customer was not open to a no-estimates approach. And I accepted being bargained down to below my estimate (Doh! I hear you say. Quite so). The actual cost came out close to (but above) my original estimate, but could have used another week’s work to make it user-friendlier.

The better course would have been to use the estimate/budget mismatch to declare a no-go rather than accept a reduced budget. This might have resulted in the client agreeing to go ahead anyway (which might in turn have led to a no-estimates approach to the work). Or it might have led to no contract. Either way would have been less painful and more controlled than over-running the budget.