OSDC 2009 Day 2 Wrapup

Day 2 of the OSDC conference, one talk delivered today with Paul King on Using Groovy for Testing. The following are my notes from the sessions that I attended.

Keynote: Simplicity

Marty Pauley delivered the keynote, my notes are as follows:

  • good code is easy to read, beautiful (aesthetically pleasing), useful
  • evil code is difficult to read, ugly, but it is still useful (otherwise you would have no code!) because unfortunately it is still in use
  • good code should be fast, concise, advanced, maintainable
  • “Simplicity is the ultimate sophistication” – Leonardo Da Vinci
  • comments are an indicator that you’re code is crap, documentation and comments are not the same (documentation is useful), ironic that some languages put documentation in comments (eg. JavaDoc)
  • always out your scripts in a module, makes it easy to read (the comment was made in relation to Perl) and makes it easier to test (one script that calls one module, that can then call other modules)
  • Google is good example of simple (as compared to Excite and Yahoo! at the time) – search engines started complicated and became simple
  • Example of simple first is that Americans used the Space Pen in space (highly engineered pen that would work on all surfaces and work in space), but when they asked the Russians what they used, it was a pencil
  • “Java was designed for stupid people! – was designed because it was deemed to hard to write code in C”
  • look outside your current toolset, we all have problems in common

How to get Rails Web Applications Accepted in Industry

Harley Mackenzie presented this talk, my notes are as follows:

  • why Ruby? – expressive language, object oriented as everything is an object in Ruby (lends itself to good, readable, maintainable code), dynamic allows to delviver scripts quickly and maintainable
  • hard to find Ruby people? – many recruiters do not understand the roles they are recruiting for, Chuck Jones (of Warner Brothers cartoon fame) looked for artists knowing he could train them to animate, the same goes for quality developers
  • open source – lower initial cost, source will never disappear as will always be around in some way, don’t emphasise the FREE but the FREEDOM, dynamic languages you always have the source (not compiled)
  • efficient – productivity important to industry
  • elegant – write easy, maintainable code
  • reliable – “testing to oblivion approach”, everything has tests, commercial environments do not value testing, all say it is a good idea but will not pay for it
  • expressive – easy to understand what the code is doing
  • why is Rails not adopted? – management are risk averse (don’t want to go outside the Microsoft norm), corporate IT motivated by fear and uncertainity (like to do things the way they are always done), loss of control (don’t understand and don’t like when you know about things they don’t), outsourced IT providers (only in it to make a buck so resistance to change)
  • solutions and adapt (know what you can change and what you can’t)
  • demonstrate – give them a VM (“take the puppy home for the weekend), find the champion
  • cloud solutions – if you can’t deal with anorganisation and their infrastructure, bypass it (use Amazon or similar)
  • draw the line – sometimes you have to walk away than comprimise (figure out where the line is), usually can deal with server operating system, web server and database but not the language it can be written in

Open Source Web Apps in Azure

Jorke Odolphi delivered this session, my notes are as follows:

  • software as a service (SaaS) – multi-tenant, pay as you go
  • platform as a service (PaaS) – applicationn frameworks, languages
  • infrastructure as a service (IaaS) – pay as you go, scale (like Amazon EC2, GoGrid)
  • Windows Azure is an operating system in the cloud, where you run applications, designed to scale
  • lots of servers sitting in shipping containers (2,000 servers per container, 7 hours to get up and running after delivered) with VM’s running Windows (called the Fabric)
  • components – web role (front end facing, static content, ASP.NET 3.5, WCF, Fast CGI applications such as Perl, http(s) inbound), worker role (like a Windows service), storage (blobs, tables and queues)
  • running PHP in Azure – Eclipse tooling available (WindowsAzure4E)
  • MySQL in Azure – run as a worker role (configure ports and storage)
  • Tomcat running in the Azzure Cloud (http://oss.cloudapp.net/)
  • No environment yet in Australia, but Singapore coming soon, pricing

Desktop Applications for Web Developers

Ben Balbo presented this session, my notes are as follows:

  • cloud issues – network outages, working offline, server outage, application failure affects everyone, access/ownership of data, dependence on third party to fix bugs
  • Google Desktop Gadgets, Adobe Air, Windows 7 – just HTML, all the information stored in the gadget or via a web service
  • Mozilla Raindrop – Mozilla’s reaction to Google Wave, pre-alpa, a local web service
  • WordPress + Google Gears
  • iPhone applications
  • XUL – XML User Interface language from Mozilla, load any interface around Firefox (use Firefox as a framework)

Using Groovy for Testing

Presentation I gave with Paul King, and was an interesting experience on how to break down a 3 hour presentation to a 30 minute talk (which we started about 20 minutes prior to the talk commencing)!

Business, Law, Open Source

Brendan Scott presented this talk and these are my notes from the session:

  • each decision you make limits your subsequent decisions
  • Starting a business has effect on how easy it be to sell down the track (setting up within a corporate vehicle) – selling of shares make this easy as the legal person transfers. Most businesses are setup where the owner owns all the assets and personally signs up to all the contracts such as phones where their name is on the contract (makes it hard to transfer later on)
  • stay away from partnership (and the use of the word partner)
  • Pov Ray had pre-open source licencing, wanted to sue someone who was bundling it and selling it – needed copyright ownership to sue but had to go and find all of the contributors first
  • code consents in case somebody takes code and uses it in breach – keep records of who, what, when
  • IP Ownership – once you have transferred rights, it is very difficult to recover these rights, you need to get legal advice early
  • dispute resolution – usually because people not speaking to each others, lawyer can help you identify issues, send the nasty letters, courts look favourably if you have tried to sort out the issue
  • negotiations – lawyers familiar with the lines of argument
  • legal arcana – knowing the legal secrets and finding holes in contracts, etc…
  • copyright – owned by employee unless agreed otherwise
  • don’t lie about your products and services (Part V 52 Trade Practices Act)
  • Part V, Dvision 2 Trade Practices Act – don’t exclude warranties, limit recourse to repair or replace, it is illegal to exclude warranties

Lightning Talks

An extended session of lightning talks. here are the notes:

Make my PHP 66 Times Faster

  • slow code read in a really big library everytime it is run
  • idea is to make a socket call and run off the server, loading only once

Something About Cars

  • entertaining comparison of programming languages to cars

The Coming Programming Language Crisis

  • entertaining look at new programming langauges, recruiting for the Australian LOLCODE Developers network! (see http://lolcode.com/)

OSDC – Open Source Developers Club

  • parent organisation for the conference
  • like meetings (found about every third session, people attend a session that is not their usual language)

Hosting Web Sites In The Cloud

  • created Grocery Choice, wanted something that had scale initially but assumed would not need it going forward
  • Amazon EC2 – virtual machine farm, upload your VM to the cloud (takes about 3 hours)
  • using commercial, licenced server so limited to CPU’s that coule be run
  • can move your instance onto a larger virtual machine
  • pay as you go, for what you use

Geography, Databases & Government


Phoebe’s Netbook

  • Phoebe is 4, has an original eePC
  • the distribution is very easy for children
  • TuxPaint and other programs in Linux that are good for kids

GMT (Generic Mapping Tools)

  • set of command line tools
  • pscoast – will show a map, can then map cyclones as they move down the coast

Byte code optimisation using Promise

Open Source Licencing vs Microsoft


  • starting a business with no money
  • do something disruptive – is a potential client trying to solve a problem that your product addresses
  • low-end disruption – easier way of doing something that already exists (eg. MySQL vs Oracle)
  • new-market disruption

Handy Python Functions

  • if building libraries, pleause use doc tags!


  • open source SMS gateway
  • written in Ruby
  • will be released shortly


  • written in Python using pygame
  • open source, GPL
  • team based, strategy game

Google Wave Bots

  • Wikifiy
  • Piratify
  • Flippy

OSDC 2009 Day 1 Wrapup

OSDC 2009 Australia is being held in Brisbane this year at the Bardon Conference Centre, and I was lucky enough to get 3 speaking slots (2 shared with Paul King). Day 1 was full of interesting talks, my notes are summarised below.

Understanding Volunteers

Karen Pauley delivered the day one keynote. My notes as follows:

  • recommended book by Charles Handy (Understanding Organizations)
  • organisation for mutual support (eg. Tokyo Linux Users Group, Sydney Perl Mongers) – consist entirely of volunteers who have a shared interest, don’t like to be managed, but a group needs some management (it’s like herding cats), anybody can join
  • organisation to provide a service (eg. The Perl Foundation) – requires structure to handle queries, cannot just join if you want you, expectations of communication and responsiveness and regular releases
  • campaigning organisation (eg. Enlightened Perl Organisation) – camapigns aren’t managed they are led, usually fail when the organiser gets bored
  • motivations – in open source we like to reinvent things
  • fun should be self evident, we volunteer to enjoy ourselves
  • lack of respect can stop the fun, need to treat each other well, need to apologize (and not follow it with a but…)
  • anger – “I know you’re all volunteers but…”, feedback can offend volunteers, in many cases people don’t mean to offend, in some cultures you write directly, tone down the rage and pause before replying
  • learnings – want a creative learning environment, learning and using something new is fun
  • as per Maslow, one of our key needs is to belong
  • belong with people who think the same way
  • impressions – decide very quickly our perception of people (4 seconds) – in person we judge looks, on the internet we judge the way we type
  • our first impression of people on the internet could be an email, IRC or a twitter – we make judgements of people based on 140 characters
  • the online life is real, so be yourself
  • everybody has a voice, make your communities fun

Joopal and Drumla: Not Your Usual Mashup

Sam Moffatt presented this brief talk about Joomla! integrated into Drupal (Drumla)

Introducing Django – Calling All Web Developers with Deadlines

Akash Mehta delivered this talk, my notes are as follows:

  • Django is a web framework to build websites quickly
  • used by a lot of newspaper / news / media sites (see DjangoSites.org for a full list)
  • a Python framework – extremely powerful high level scripting language, do a lot with less code
  • a python web application framework – excels at data driven web applications
  • a python full-stack web application framework – if using Django you are using the full system, not portable, mature, been around 5+ years
  • batteries included – functionality you need comes out of the box
  • Django has good support for internationalisation
  • DRY -> DRY (OAE) – Don’t Repeat Yourself (Or Anybody Else)
  • views do not usually have URL’s mapped to them, custom URL mapping takes care of this
  • not a serious business framework (see djangopony!)
  • good documentation – starts with a comprehensive tutorial, documentation then splits into functioanlity / what you want to achieve, stored in Subversion, there is an online version of the book which is also published by APress and available on Safari
  • automatic administration – not a scaffolding, end-to-end solution for managing production data, lots of useful widgets
  • available applications – search (Haystack is the current favourite), blogs (plenty available on github), security and administration built-in
  • views take advantage of decorators in python, such as @login_required
  • templates – template inheritence and blocks
  • URL config – no automatic mapping, create your own URL’s, can use regular expressions, usually mapped to views

Perl6 Now

Scott Penrose delivered this talk, my notes are as follows. Talk will be made available online.

  • when is Perl 6 available – now, language specification
  • looks like Perl
  • biggest problem with Perl is when you have @array you need $array to access it
  • More information at http://rosettacode.org/wiki/Category:Perl_6 and http://github.com/perl6/perl6-examples
  • Pugs – not very active, but a good reference application, written in Haskell
  • SMOP / Mildew – library parsers
  • Sprixel / Elf – implementations
  • Rakudo – run time engine, implementation of Perl 6 that runs on Parrot
  • full version should be available next year
  • performance – slower, Parrot keeps changing, performance increasing month on month
  • CPAN – not currently available but coming
  • 42 modules available now via github, mainly web and database
  • IDE – Vim syntax highlighting and Padre
  • debug – Larry Wall – “why do you need a debugger, just print line”, no interactive debugger yet
  • fun langauage to write in – can write code shorter and faster, execute faster is coming

oFono – Open Source Telephony

Denis Kenzior delivered this talk on oFono, my notes are as follows:

  • many devices and modems to talk to – all talk vendor specific protocols
  • existing projects – FSO, only considering smartphones, assumed AT modems, API not well constructed, implemeted in Python and too slow, does not support multiple modems
  • existing projects – Qtopia, end of life, did not support D-Bus
  • learnt from these projects – GSM is hard, huge and boring, everything exposed in the API but very little used or available
  • GSM is hard – massive amount of specifications (call handling (in, waiting, multi-party) sevices such as call waiting and number exposure, SMS), vendor specifics (enhanced SMS, smart messaging, WAP), storage (number, calls, SIM)
  • GSM is really hard – backwards compatibility (most of the specification still around from 1G), every bit counts, bugs in all of the different vendor hardware that needs to be worked around
  • oFono targets all devices, target all distributions, minimal dependencies, modular, UI easy to customise / replace
  • don’t use SIM for contacts / SMS – limited to number of contacts / messages – simplified the design
  • device detection – udev, modemconf
  • plugins – read / write to SIM, send / process SMS, Call / SMS history, D-Bus interfaces such as carrier logos
  • atom – implement specific functionality eg. dial, hangup, merge, switch, invokes operation on an atom driver
  • currently supports – network registration, make and receive calls, SMS, cell broadcast, GPRS, all supplementary services
  • handful of modems and devices supported
  • Phonesim– simulates most AT modems, make / receive calls, SMS, MWI, cell broadcast, network registration, supplementary services – good for development simulation
  • API – consistent, minimal, easy
  • takes care of magic strings (MMI, those *xxx strings), read / write from SIM, service provider display (MVNO, files on the SIM)
  • oFono plugins for BlueZ – hands free & audio management
  • oFono plugins for ConnMan – 3G network management
  • oFono currently has no CDMA support
  • unique because can support mutiple devices at once

State of Play: PHP in 2009

Akash Mehta delivered this talk, my notes are as follows:

  • 1995 – PHP/FI, 2000 – PHP 4.0 (very successful), 2006 – PHP 5.1.2 , 2009 – PHP 5.1.3 (developed and released at same time PHP 6 being discussed, so many features back ported)
  • language extensions
  • 450 user groups, 1200 members in San Francisco group!, 100 in Brisbane active
  • Microsoft and Zend – awful on IIS, they are working on it apparently
  • used in dynamic applications, social networks (Facebook), business applications (FLOW3)
  • IDE’s – Eclipse – PDT, PHPEclipse, Zend, NetBeans PHP, JetBrains working on PHP editor (Web IDE)
  • SVN still dominant, github growing in popularity following in the footsteps of Rails
  • server – Apache losing ground, growing in popularity lighttpd and nginx
  • phpDocumentor for automatic documentation
  • Phing and Capistrano for build and deployment
  • Project Zero is PHP on JVM
  • Xdebug is popular for debugging, DBG is also reliable
  • CakePHP – Rails for PHP, Mozilla Addons was the big thing but recently dropped for Django
  • CodeIgniter – strong following, low barrier to entry
  • Symfony – popular, strong following, high barrier to entry, hard to use with existing code
  • Zend – full stack approach

Git Me Up

Nigel Rausch delivered this talk on Git, here are my notes:

  • adopted quickly in Rails community
  • distributed version control
  • centralised – CVS, SVN
  • distributed – Bazaar, Git, Mercurial
  • Git created by Linus Torvalds, developed for Linux kernel patch control as replacement for BitKeeper (was one patch every 6 seconds, now 6 patches per second), so needed to be fast
  • effcient in management of files – entire repository for Linux kernel is only 40% bigger than the kernel on its own
  • commit locally to your machine – have full copy of all history – can commit at anytime, can always look back at history
  • simple branching and easy merging
  • support multiple remote servers
  • SVN is simpler to use, single point of truth all the time, central control, until recently better tools
  • git init – creates a hidden directory .git, which has config and history, use to start a repo
  • git clone <repo> – clone an existing repo to start a new repo
  • git add <file> – add to repo
  • git commit -m “message” – commit to local repo
  • git has about 150 commands, plus all their options
  • git branch – master is there by default
  • git checkout – checkout a branch
  • git log – commit log history
  • git merge <branch> – merge two branches and create a new commit
  • git remote add origin <repo> – this happens automatically if you cloned
  • git push origin master – origin is server we are pushing to (by default called origin), master is branch
  • git pull origin master – fetch and merge from remote repository, pull is a fetch and merge, a fetch will just get and leave in a pseudo repo
  • Mac installs with GitK which is ugly. GitX is a better bet, TortoiseGit for Windows
  • git svn clone – will create a bi-directional git repo of your SVN repository, commit locally and push back to SVN, at a later poiint you can change your origin to a Git repo
  • rebase, or work on a branch and merge to master locally so you don’t have all of the little commits
  • initial high learning curve, supports commit often, gaining support, IDE support growing
  • binary files saved and compressed in repo

Lightning Talks

A bunch of 5 minutes lightning talks, here are the brief notes from the session:


  • low level virtual machine
  • risk based assembly language with annotations
  • clang – convert C and C++ to LLVM
  • adopted by Apple


  • GUI programming in Python, wxPython is ugly, PyQt4 meh
  • idea – with gui.x (where x is column, frame) – clean
  • still things to do, code not yet released

Visualising Geo-Data

  • government 2.0 taskforce
  • big csv file about sighting of frogs in South Australia
  • frogspotter – every dot, different species, different colours, then a cluster library
  • the data aggregator – drill down on the data
  • frog census

Connectr Social Network

  • currently Twitter with 650 lines of Python, wish to take on Facebook!
  • current social networks have issues with privacy, incentives for hosts are wrong for users
  • no business model, not hosted on server, runs over xmpp


  • Linux Conf Australasia 2010, to held in Wellington


  • linear programming project, sponsored by GNU (hosted project)
  • applicable to lots of things – scheduling, rostering, optimising sports drafts
  • CPLEX is commercial alternative
  • MathProg is the language that comes with it, describes the program in a math like language
  • active community
  • bindings to scripting languages


  • used for conference books and name tags
  • pdf:toc to generate table of contents from every heading
  • sponsor logos are image tags, use CSS zoom property
  • div tag used for indenting
  • margins in centremetres, told CSS book was A5
  • absolute positioing (OK for print, not for web)
  • scripts will be published
  • similar to Latex, but simple and uses same HTML as web

iPhone Hacking

  • to get SMS out – if not encrypted they are stored on your system by iTunes
  • they are GUID.mddata files which you can access using a database client
  • bambapilla – exports conversation to XML, formats nicely as XSL in browser

Hacking OSX apps with Python

  • objective C is a lie, its just C
  • when you don’t have the source but want to change a behaviour – swizzling

Patents Rock

  • protects inventors and ideas, thoughts and ideas
  • patent voliations – side to side swinging on a swing, licking a postage stamp, inducing cats with a laser light
  • software opatents suck

Haml & Sass

  • Haml – takes HTML and makes it simple and short – implementation for Ruby
  • Sass – takes CSS and makes it simple


Continuous Integration

Earlier in the year, I gave an internal videocast to my colleagues in IT on continuous integration. I finally got around to posting it online and the presentation is now available on Slideshare.

Facilitating a Successful Developer Day

When an executive manager approached me a few weeks ago with the idea of running a developer day with one of his Java development teams, I jumped at the opportunity. The idea was simple: emulate the success of Google 20% time and the Atlassian Fedex day to improve the collaboration and skills of a bunch of developers.


I started out with some research and found the following excellent resources:

The Running Sheet

It was decided (in hindsight, a fantastic decision) to hold the day in a neutral venue away from the usual office in a training room environment. The pre-requisites equipment-wise were at least one machine per two attendees (we ended up with two high-spec developer machines and a bunch of laptops), network access, a projector and the usual agile materials of post-it notes and pens. A supply of snacks and drinks was also available to give it a “start-up” feel.

As we had two days and wanted to also allow for some presentations and discussions, the run of the days ended up as follows:

Day 1

9.00am: Standup Introductions (All), why are we here, team building exercise (paper folding exercise)

Give each member of the group an A4 piece of paper, the facilitator needs one too. Have them close their eyes. The facilitator issues the instructions and follows them as well. No questions are allowed.

  1. Fold the paper in half.
  2. Rip offf a corner
  3. Fold the paper in half
  4. Rip off a corner
  5. Fold the paper in half
  6. Rip off a corner

The group can now open their eyes and find that there are many different shapes of paper. The debrief covers the need for two way communication and that the different perceptions of the people caused the many different designs. If time permits the group can be put in pairs. Have the pairs sit back to back and repeat the exercise using two way communications and find that the patterns come out closer

10.00am: Presentation (Agile Tools & Maturity)

11.00am: Brainstorm / Retrospective: What would make this a better team -> how can we make a difference in 24 (8) hours

The idea was to review the good and not so good of the team and help determine the issues the team would like to tackle.

12.30pm: Iteration 0 / Planning / Lunch

Melbourne GI Developer Day Rules

2.00pm: Iteration 1

Including a quick planning session

3.30pm: Iteration 2

Including a quick retrospective and planning session

Day 2

9.00am: Standup

Overview of highlight of yesterday and what planned to do today.

9.30am: Iteration 3

Including a quick retrospective and planning session

10.30am Iteration 4

Including a quick retrospective and planning session

12.30pm: Developer Practices discussion / lunch

Based on the development practices prescribed on the Donkey open source project.

1.30pm: Iteration 5

Including a quick retrospective and planning session to ensure ready for the 3pm demo

3.00pm: Demo Showcase

Melbourne GI Developer Day Demo

4.00pm: Closing Statements and circle

The Outcome

Well, it turned out better than I could have imagined! Iteration 0 everybody turned a bunch of PC’s in boxes and chairs and tables into 5 distinct teams. Story walls went up, planning started and the software required was downloaded (or people remotely accessed their work machines). Pair programming naturally happened (the limited number of machines helped this) and teams supported other teams when they ran into problems or needed assistance.

Melbourne GI Developer Day

The 5 teams focussed either on stuff they wanted to learn or things they wanted to achieve to help their everyday work. The projects attempted roughly consisted of:

  • a mobile web application using the Play framework
  • investigation into using JMX for application monitoring
  • automating WebSphere environment creation with JACL / Jython scripts
  • scripting legacy mainframe account creation for testing purposes
  • performance testing using JMeter and The Grinder

The morale was high at the end of the two days, everybody clearly learnt something and the developers were asking when the next event was. Success!

Version Control and Git

A couple of weeks ago, Geoff Ford and I gave an internal presentation to some of developer colleagues on version control and (particularly) the use of distributed version control and Git. The presentation is now available on Slideshare.

Dave Thomas on Maximum Software Productivity – Breaking The Rules!

Dave Thomas paid a visit to Brisbane to present his talk Maximum Software Productivity – Breaking The Rules! at the Microsoft office at Waterfront Place. He set the scene by suggesting the last time he gave this talk he was innundated with hate mail from agilista and objectas! Dave promised the slides would be posted, but I have not been able to locate them if they have. Here are my notes from the session:

  • tired of need to get agile and need to get objects – both good, but got nothing from it, where is the value?
  • objects good, agile good, but are we any better off?
  • most software late, bloated, poor to maintain
  • business does not define requirements well, don’t engage or talk to the customers (NEHITO visits – “Nothing Ever Happens In The Office”), insist on the industry norms because that is what everybody else is using
  • storypoints – training wheels for people who don’t know how to estimate
  • IT does not estimate well, do not build continuously or automatically test (need to be prepared to write a big cheque), also fixed in technology
  • if not ready to do TDD, just stop! Scrum will only make you feel good
  • need to change the rules to be competitive
  • living in legacy due to OO mud ball – legacy is code where there is no tests (see Brian Foote – Big Ball Of Mud)
  • objects too hard for normal people, first thing to be voted off the island should be Hibernate, objects don’t work for lots of things (queries, rules, transformations, etc…)
  • 80% of objects are CRUD – no objects except for the junk in the middle – just data, no simulations or data model – taking a solution and making it complicated and slowing it down
  • frameworks – so many to choose from, new versions, latest things, lots of dependencies
  • dependency injection is proof of how much we have screwed up objects
  • object libraries are unstable and languages are complex (attributes, generics, concurrency)
  • very few people use interfaces properly and keep them stable release to release
  • little reuse – promise not a reality
  • serialization carries all of the baggage of the objects contained within
  • objects suck up performance and memory – bulky and computationally expensive
  • objects cumbersome and slow for multi-core processors to run efficiently
  • objects are sequential and parallel
  • Java application will be 4 to 5 times slower than PHP, because objects are slower
  • can we shorten software value chain – shorter, faster, cheaper – Facebook built in PHP but the money rolls in
  • agile shows predictability and productivity but zip about quality
  • agility is good, but in Java it is very difficult to change the code quickly
  • scrum increases morale (everybody feels better), but makes no difference to quality unless doing TDD
  • lean thinking – software waste – if not prepared to spend $2 million to do continuous integration and TDD then they are not serious about their software output and quality
  • don’t go to meetings unless it is increasing the bottom line – will it help ship code?
  • how to make zero defect code? Don’t ship anything!
  • lean – simple – why am I doing this? – do we need a new framework? – NO!
  • staff a team with people who have shipped software (have a track record)
  • fix price your consultants and enforce that they make delivery with acceptance tests – pay when they pass
  • reward people on delivery, not how long they work
  • tangible requirements – story on the front, acceptance criteria on the back, start with acceptance tests, they are more valuable
  • envisioning – what the developer sees is not what the customer wants
  • agile great for prototyping – building small requirements on the fly
  • a backlog filled with lumpy requirements will burn out designers and product owners – original Scrum paper said sprint 0 should be 3 months – envisioning
  • architects – not a job, a role – should be able to code
  • some companies pay senior developers same as VP
  • extreme design – four hours to design software and hardware and cost it – after a few, you can get close and find out quickly what you don’t understand
  • serious engineering needs design
  • API first – design the architecture, should always be able to get architecture from the code (push a button)
  • need API’s versioned in the code
  • want to close gap between needs and solution
  • a picture is 1,000 words, a table 200 and a diagram 50
  • table driven programming – easily understood, easy to refactor, easy to consistency check (look for missing data), easy to version and diff (just data), data driven can be changed live in a running instance
  • integration – talk to old systems use RSS/ATOM feed (almost all old systems will give a feed for each transation) then you can talk it without custom api’s, REST/JSON your services or use ODBC as a simple interface (its not just for databases), use mashup tools to deliver an integrated application view
  • use scripts to save time & dollars – most software gets thrown way even before turned on, C# and Java too heavyweight, Ruby, Python, PHP, Groovy, Clojure, can easily leverage cloud services and existing services
  • productive languages – LINQ and Reactive LINQ (Haskell underneath), Erlang (good for swicthing and moving traffic, highly efficient), F#, Ruby, Scala (a better Java), Kleisli (bioinformatics), Clojure
  • Everybody should read “The Wizard Book”, Structure and Interpretation of Computer Programs by Abelson, Sussman, and Sussman
  • hardware is cheap, cloud is cheaper – all interesting data is in memory, databases are just journals and don’t really exist, Google does everything by brute force search (eg. translation), speed now means complicated stuff can be done easier and cheaper
  • data driven – massive storage means we can store it all very cheaply, run smart algorithms to determine best value customers because we have all of the data, recommendation engines (NetFlix), Net Promoter, complex event processing (open intelligence), real time financials (no end of month financials, know state of company all of the time)
  • query oriented programming (QOP), Greenplum, Q, Aleri (basically extended SQL dialects with functions) and other functional languages
  • array VM’s – better than object VM’s – always boxed, simpler garbage collection, support all data types (array of stuff is an array of stuff), VM can be small (just an interpreter), arrays are column stores already and trivially serialised, take less space and programs concise and compact
  • best practices for functional programming – agile , refactoring tools, FindBugs / Lint
  • challenges for functional programming – get out of math phobia, different way of thinking, need to write literate and understandable code, think in functions
  • yesterday and tomorrow is always wrong – if you get more bang for buck and competitive advantage, why use existing technologies, if you believe IT is strategic you need to do something strategic with it and dare to be different
  • lost of old things work, lots of new things work well, get out of agile and OO box