Sunday, 30 October 2016

The Forever War: Forever War Book 1

Joe Haldeman
ISBN-13: 978-0575094147

I don’t remember who it was, but someone suggested I read The Forever War when I got into the Space Operas of Alastair Reynolds. I couldn’t find it on the kindle for ages, so a hard copy of the omnibus edition sat untouched on my shelves for quite a while. Then it became available on the kindle.

The Forever War Book 1 is good, I enjoyed it. I liked the characters and the first person style works really well. I am sure in the 60s it was revolutionary maybe even shocking, but, as with a lot of old sci-fi, it is of its time and feels antiquated next to the technology we have today.

In terms of scope, it’s nowhere near as vast as anything by Alastair Reynolds and I think this was my biggest disappointment. I was interested to read that a lot of the inspiration for the story came from the author's experience in the Vietnam war. If it hadn’t been mentioned I certain wouldn’t have known. I did spend a good deal of my reading time trying to see how the futuristic battles and enemy might relate to the Vietnam war.

The two sequels were written much later. I’m hoping they’ll be better.

Sunday, 16 October 2016

Norfolk Developers: How email works... & Fabric

James Taylor

What: How email works... & Fabric

When: Wednesday, 9th November 2016, 6:30pm to 9:00pm

Where: The Union Building, 51-59 Rose Lane, Norwich, NR1 1BY


How email works... and why you should care. 
Steve Engledow (@stilvoid)

A brief history of some of the aged protocols that underpin the internet and email in particular, a look at the security implications of the way email works, and some steps you can take to improve your safety.

Fabric: Fast one line consistent deployments across multiple servers 
James Taylor (@jmons)

For small organizations, or even hobbyists, deploying modern systems can become annoyingly complex. There are several solutions which we’ll look at briefly, before doing a deep dive and demonstration of how I use “Fabric” to do controlled continual deployments with no stress and maximum fun. In this session we will be doing a live-code demonstration, as well as taking continual questions and heckling is encouraged. Even if you can’t code, but manage sysops or coders, hopefully you’ll get to see what low stress deployments can look like.

James Taylor

James works at Proxama by day inventing new and interesting methods of payments, and works on various startups at night (@getimperium, tether). He despises all programming languages, but some more than others, and is currently the NorDevCon "Just a Minute : Technical Edition" reigning champion (mainly due to a rant about PHP). (This profile contains an annoying mismatched bracket.

Saturday, 15 October 2016

Good Code: Kevlin Henney Full Day Workshop

What: Good Code: Kevlin Henney Full Day Workshop

When: Wednesday, 23rd November 2016, 10am

Where: The King's Centre, King Street, Norwich, NR1 1PH, Norwich

How much: £35.00


Good Code 
Kevlin Henney (@kevlinhenney)

We often talk about good code — that we would like to write it, that there isn't enough of it, that it should not be considered an optional attribute of a codebase. We often talk about it but, when it comes to being precise, we don't always agree what constitutes good code, nor do we necessarily share a common view on its value.

This one-day tutorial and workshop explores what properties we want from a codebase and, therefore, what we can deduce to be good. These conclusions can sometimes be surprising and counter-intuitive! This session will explore some common guidelines on what is considered good, from expression to subsystem, from naming to tests, from fluent to SOLID. We will look at the consequences of good and not-so-good code from point of view of economics, day-to-day work, people and runtime performance and reliability.

Kevlin Henney

Kevlin is an independent consultant and trainer based in the UK. His development interests are in patterns, programming, practice and process. He has been a columnist for various magazines and web sites, including Better Software, The Register Application Development Advisor, Java Report and the C/C++ Users Journal. Kevlin is co-author of A Pattern Language for Distributed Computing and On Patterns and Pattern Languages, two volumes in the Pattern-Oriented Software Architecture series. He is also editor of the 97 Things Every Programmer Should Know book.

Saturday, 8 October 2016

Naked Element’s software for Fountain reduces processing time by 95%

Fountain Partnership Limited are a digital marketing company based in Norwich, established in 2008. With a team of experts, Fountain are able to identify a company’s growth opportunities and build a customised strategy for the best results and command market share. Their strategies drive browsers to client’s websites, increasing opportunities to convert browsers into customers and then increase customer value by testing and measuring. In order to do this, they specialise in the optimization of search engines and conversion rates, as well as pay per click advertising. In order to do this, Fountain uses a combination of pay-per-click advertising, search engine optimisation and conversion rate optimisation.

The problem Fountain were faced with was a time consuming one. When advertisements needing changing or updating it was a painstaking process. They had to upload a new set of adverts from a spreadsheet, going through each one manually to pause all of the old adverts in order to make room for the new ones. it would take days in some cases.

“What was taking 20 hours of work is only now taking an hour.”

Naked Element were chosen to build a script which would allow Fountain to manage one of their largest clients in Google AdWords, saving time and ultimately money. This particular client currently had 75 individual accounts. What the Fountain team were doing to update these was logging into an account, creating the new adverts, then log out of that account, then log into the next one 75 times over!

So what was Naked Element’s solution? In simple terms a script was created that allowed the user to specify AdWords accounts, campaigns and ad groups and then to enter a search, replacing each with a phrase or word. When running, the script would look through all of the ad groups in the campaigns specified and would copy the ads found with the search phrase, update the field with the updated phrase and pause the previous ads.

For example: A client has a campaign called ‘Car Sales’ with three ad groups – blue, red and black. Within those groups there are ads with the description “newest model”. The search phrase “newest model” is entered, followed by the update phrase “drive away today” and the script is run. All of the previous ads with “newest model” will be paused and replaced with the new ads containing the phrase “drive away today”.

“By developing this software to improve process efficiency, Naked Element has saved us four weeks worth of work per year!” said Laura Jennings, Strategic Director at Fountain. “The script they built us is saving up to 95% of our processing time. What was taking 20 hours of work is now only taking an hour – a big saving. ” When software produces such immediate benefits, the advantages are clear.

“The process of working with Naked Element was really straightforward, they tried hard to extract the correct information from us. Lewis, the developer, came out to see us several times. Documentation was also really easy to understand.” The only issue Fountain had was during the creation of the script, as it didn’t always work quite as expected. “However Lewis was quick to identify and resolve the problems. It was never too much trouble to sort out an issue and assistance was always available via telephone.”

Sunday, 2 October 2016

Ever wondered what an algorithm is?

Algorithms are everywhere. Our children are even taught about them at school, but have you ever found yourself wondering what an algorithm actually is? Maybe you’ve thought they’re something used by computers and created by computer programmers, but don’t really know what they are?

The dictionary defines  an algorithm as:

“A set of rules for solving a problem in a finite number of steps.”

Algorithms come in all shapes and sizes. They can be extremely complicated, but they can also be very simple and easy to understand.

Examples of more complex algorithms include those used to price financial products in a bank or to determine the best route between two points in a satellite navigation system. Simpler algorithms include those used to sort lists of numbers, such as Bubble Sort.

Bubble Sort

Bubble Sort is one of the easiest algorithms to understand. As its name suggests, it’s an algorithm used for sorting. Often the easiest list of things to sort are numbers. Bubble Sort works by comparing each number in the list to the number next to it and swapping them with each other if the numbers are in the wrong order. This process is performed again and again until a pass over the list requires no swaps. At this point the list is sorted. Knowing when to stop sorting the list is just as important as knowing how to sort the list. As we know when to stop (when a pass has no swaps), Bubble Sort can be used for lists of any size.

The easiest way to demonstrate Bubble Sort is with a simple example. Take the list of numbers:

3, 2, 1

We can use Bubble Sort to reverse the list. The first time we pass over the list the first two numbers are 3 and 2. 3 is greater than 2 so we swap them over:

2, 3, 1

Next we compare 3 and 1. 3 is greater than 1, so we swap them over:

2, 1, 3

There are no more pairs of numbers to compare on this pass and there were two swaps (3 & 2 and 3 & 1) so we pass over the list again. The first two numbers on the second pass are 2 and 1. 2 is greater than 1, so we swap them over:

1, 2, 3

Even though we have successfully reversed the list, we’re not finished. Next we compare 2 and 3. 2 is not greater than 3 so we don’t swap them. There are no more pairs of numbers to compare on the second pass and there was a single swap (1 & 3) so we pass over the list again.

1, 2, 3

The first two numbers to compare are 1 and 2. 1 is not greater than 2, so we don’t swap them. Then we compare 2 and 3. 2 is not greater than 3 so we don’t swap them. There are no more pairs of numbers to compare on the third pass and there were no swaps so we’ve finished and successfully reversed the list.

Sorting Algorithms in the Real World

Bubble Sort is taught to trainee software engineers as it’s easy to understand and implement. However, it’s rarely actually used in the real world as it’s inefficient and there are other more efficient sorting algorithms, such as Quicksort, which are only a little more difficult to understand implement.

Sorting occurs in all sorts of systems in the real world. One example is in the software used to sort mail into the order a postman will deliver it as he walks along his round. Postcodes and other address elements are read into the system and sophisticated algorithms used to sort the mail into the correct order.

The next time someone mentions algorithms to you, remember it’s a set of rules for solving a problem in a finite number of steps.