Archive for ‘Software Development’

7 results.

Chess Timer

by admin on December 12, 2011 at 10:27 pm

What I would like, when starting a software project in a sizable organization, is a chess timer.

Chess Timer

Every time I ceased to be able to progress with the project, because a detail needs to be clarified, because some decision needs to be made, or because it is waiting on a sign off, I would click off the chess clock. And relax…

I am a hopeless optimist, and fiercely competitive. This results in one thing when it comes to software development- I cannot predict how long things will take, within even 50%, maybe worse. When I start a project, it’s generally “about a week” from completion for the entire 6 month duration. And as far as I’m concerned that is absolutely fine.

What I don’t need is plenty of time to get things done, what I *want* is a ticking clock that I can see all day measuring my performance. The thing is I don’t want to be judged based on, is other people’s priority lists. Just because Bill in accounts is too busy to sign my project off, well, I don’t care.

If developers are to be treated as delivery agents or micro project managers- by looking at the absolute time to project completion, then you are losing all their efficiency. They stop worrying about how lambda functions can help and best practices they could be learning and that git command you only ever use once in your life, and start worrying about why Mike is chatting to Kat rather than emailing them back. Give them that clock, and ‘click’, they can switch to some other project safe in the knowledge that they are not being measured.

According to modern software management opinions, putting developers under pressure does not make them more efficient, and I agree with this. In my experience the negative pressure which is experienced is usually due to events outside the control of the developer- eternal delays, changed specs, pointless meetings, rather than because I screwed up and took too long doing something. If that is the case, my pride as an engineer trying to create a quality product will mean I work twice as hard to finish on time. As a manager I think you need a way to get your developers to feel competitive about their own performance, let them put their effort into creating quality, and let you worry about these other events and pressures. That to me is the role of a manager, and the chess timer fits this perfectly- you are giving the developer the chance to be responsible solely for the things within their control.

The other thing the timer provides, is an accurate way of tracking how long you spend blocked on projects. I don’t think it matters too much how long developers take to write a quality piece software (within reason), because you are getting some increased quality or feature for every hour they work (provided of course that what they are working on is well managed!). However for time projects are blocked, you get nothing at all.

One could argue that if you don’t care about time to completion, and developers can effectively switch between projects then there is no problem here, since your efficiency is still optimal. However we all know that business *does* care about time to completion- it’s just the developers who shouldn’t and/or don’t. The point here isn’t to minimize blocked time, although that is obviously a goal, but that it is properly tracked by the chess timer, so we can see exactly what proportion of a task or project’s total duration it took up. This isn’t something I have seen in my (limited) experience, but am interested to know if other companies do track.

1 Comment

Poker Hand Rank (Haskell)

by admin on December 2, 2010 at 12:37 am

I’ve spent the last couple of days trying to get my head around Haskell beyond the cursory level which my university course demanded. The first project I’m attempting is to write a little poker game, probably not with any AI, but requiring a bunch of different techniques which I need to work on. Tonight, happily, I’ve managed to implement the hand ranking algorithm. It’s not *fully* tested yet but it seems to work.

No licence with this code (use as you like) and no warranty obviously!
I would appreciate any pointers on improving the code if you see fit…


import Control.Monad
import Data.List
import List

data Suit = Hearts | Clubs | Spades | Diamonds
  deriving (Read, Show, Eq, Ord, Enum)

data Rank = Two | Three | Four | Five | Six | Seven | Eight
  | Nine | Ten | Jack | Queen | King | Ace
  deriving (Read, Show, Eq, Ord, Enum, Bounded)

data Card = Card {
  rank :: Rank,
  suit :: Suit
  }
  deriving (Eq)

instance Ord Card where
  compare (Card x1 _) (Card x2 _) = compare x1 x2

-- Pretty card display print out
instance Show Card
  where show x = show(rank x) ++ " of " ++ show(suit x)

data HandType = StraightFlush Rank | FourOfAKind Rank Rank |
  FullHouse Rank Rank |  Flush [Rank] |  Straight Rank |
  ThreeOfAKind Rank [Rank] | TwoPair Rank Rank Rank |
  Pair Rank [Rank] | HighCard [Rank]
  deriving (Show)


evaluate :: [Card] -> HandType
evaluate cards = let sorted = sort cards in
    case (flush sorted, straight sorted) of
    (Just f, Just s) -> StraightFlush s
    (Just f, Nothing) -> Flush f
    (Nothing, Just s) -> Straight s
    _ ->
    case nOfAKind (groupBy (\c d -> rank c == rank d) sorted) of
    ((3,c):(2,d):[]) -> FullHouse c d
    ((4,c):(1,d):[]) -> FourOfAKind c d
    ((3,c): xs)    -> ThreeOfAKind c (recombine xs)
    ((2,c):(2,d):(1,e):[]) -> TwoPair c d e
    ((2,c): xs)    -> Pair c (recombine xs)
    xs             -> HighCard (recombine xs)
 where
    recombine xs = snd (unzip xs)
    nOfAKind [] = []
    nOfAKind (x:xs) = (length x, rank (head x)) : (nOfAKind xs)
    flush (c:cards) = case dropWhile (\x -> suit x == suit c) cards of
    [] -> Just (map (\x -> rank x) (c:cards))
    _ -> Nothing
    -- N.B must account for 'Wheel' hand
    straight (x:xs) = if (rank x == Two) && rank (last xs) == Ace
                      then checkNext Two (init xs)
                      else checkNext (rank x) xs
    checkNext r []        = Just r
    checkNext r (x:xs)    = if (r /= maxBound && (succ r) == (rank x))
                            then checkNext (rank x) xs
                            else Nothing

1 Comment

On Web Apps

by admin on November 16, 2010 at 12:19 am

I’m currently ploughing my way through the Ruby on Rails documentation, trying to get used to thinking of applications in the server-client model rather than the far more familiar (to me) desktop model.

I really like the idea of web apps as a portable platform to do nearly anything you would want. I love the idea of being able to do work from any computer. I love the idea that if you are paying, then you are paying for service and/or data, not for the lame idea of owning a piece of software (because you never really own software). And I love the idea of being able to share work with other people with a single click- no need for installed software. Prezi is a an excellent example (you should check out their cheesy demonstration video if you haven’t seen them before).

I don’t have a problem with frameworks like Silverlight or Flash (or Java…)- I think that they can solve problems which would otherwise be very difficult to solve. However, I think that they should be a last resort, and in being a last resort they kind of become redundant, since the ubiquity of installations will drop and thus the idea of not relying on installed software fails if the user needs to download a framework.

Furthermore there are further problems with these frameworks since they are not open standards, it makes it hard for interoperability beyond the manufacturers intention. How long did we have to wait for a 64 bit Linux build of Flash? From my experiments, Moonlight (Silverlight Linux port by the Mono project) still has a way to go, and will always lag behind the latest version of Silverlight. There are rumours going around that Silverlight’s future may be in jeopardy, and we all know what Apple have in mind for Flash, so perhaps it is a bad idea to be investing our time as developers learning to use these frameworks when their control is out of our hands.

On the other hand, from what I have seen, developing web apps is incredibly messy. HTML, CSS and Javascript are the basic tools, and HTTP is the basic method of communication. It seems like anyone writing web applications front ends with just these tools must be crazy. What with browser incompatibilities, AJAX, and the horrific DOM specification, it seems very difficult to achieve anything of any significance using only these tools.

With the rise of MVC frameworks such as Rails, ASP.Net MVC and Cake there is a wealth of support for developing complex server side solutions with data layers bound to presentation layer content, and toolkits such as GWT and jQuery provide functionality to keep the evil aspects of DOM and AJAX at arms length. All this is great and the sheer amount of different frameworks and libraries available is incredible. This can only be good for web development as a whole.

I don’t believe in a silver bullet for web or any other type of development. Design is a hard problem which cannot be solved without being rigorously specified, in code or other formal method. However I do think it is possible to position ourselves as developers in a place where we can see the way in which all of the components in our system are working together, and go down the levels and see how each of them is working. Getting the right level of abstraction is the hard part of development, and that goes for development of frameworks as well.

Given the amount of work required to build really cool stuff using raw HTML + Javascript. I guess the problem for the developer is which framework to choose. No-one wants to spend time learning something only to find it deprecated after a single project. However, it seems that with web development, whether you like it or not, things will move on and you will find yourself constantly challenged to learn new technologies and libraries. You can choose to focus on learning one thing properly like the old school UNIX hackers prided themselves on, or you can keep Google and StackOverflow open at all times, set up unit tests and get to work. This can either scare you, or it can excite you.

In fact I think it can do a little bit of both. It can be hard to know where to start with all of it, but the great thing about the web is that you only need a text editor and a browser to get started.

2 Comments

FluidDB Thoughts Part II

by admin on October 14, 2010 at 9:04 pm

I’ve been thinking some more about FluidDB. I mentioned in this post that I did some work adding features to the .Net FluidDB open source project hosted here.

Currently the API has a basic layer for making raw calls to the RESTful FluidDB API, a second layer abstracting the behaviour of FluidDB into CLR objects (FluidObject, FluidTag, FluidUser etc), and a third layer allowing attributes to be used to easily specify a mapping between CLR properties and FluidDB tag values.

For example:

[FluidDBObject]
public class Person
{
/// The tag value will be a json primitive
[FluidDBTag]
public string Name {get; set;}

/// The tag value will be the GUID of the father FluidDB object
[FluidDBObject]
public Person Father {get; set;}

/// The tag value will be the serialized data with the MIME type specified
[FluidDBTagValue("image/jpeg")]
public BitmapSource {get; set;}
}

Person g;
// create a FluidDB object for my CLR object
// tags will be in the namespace guy127917/test
FluidDBGUID id = g.WriteToFluidDB(FluidDB, g, "guy127917/test");
// retrive a Person object with the specified ID in the specified namespace
Guy p = FluidDBAttributeWriter.ReadFromFluidDB
(FluidDB, "guy127917/test", id);

This is a work in progress, some of it sort of works, but don’t count on it yet (if you need it to work, help make it work!).

One of the slight issues I have with it is that if you have a class with a whole bunch of properties it can take quite a while to retrieve each value. The time taken is purely due to making a socket connection for every http call, which occurs every time you query FluidDB. Since primitive values are stored as JSON (strings, numbers, booleans), and JSON has provision for storing “objects” (collections of named values), it would seem that storing all primitive values in a single tag value as a JSON object would solve this problem- all values would be retrieved in one API call to be parsed into the appropriate CLR properties.

The drawback of this approach is that there would be a single FluidDB tag value on an object which means that the data is harder to query for a given tag value, and harder for other people to see what tags are available for them to use. Since data interoperability is the core value of FluidDB this is quite a big deal.

From the perspective of the .Net API, it could easy allow a tag name to be specified as an argument to the FluidDBTag attribute for primitive types, allowing grouping of property values into a single tag value e.g [FluidDBTag("TagGroup")]. This would allow you to decide which values ‘deserve’ their own FluidDB tags. However, what is not particularly important to your app may be the killer for someone else wanting to use the data. For example you may bundle geo-positioning values in with a load of other data because they are not crucial to your app, but someone else may want to search for objects in a given location.

I’m not going to implement this right away. For starters I don’t have an app which requires performance. Secondly I’ve heard unsubstiated rumours that some sort of batch request API may be in the pipeline. Thirdly other more important things in the .Net API need work much more urgently. And finally, I’m kind of convinced by my own interoperability argument, data should be as easy as possible for other people to be use in order to get other people to add value to your application data.

3 Comments

FluidDB

by admin on September 19, 2010 at 6:01 pm

I’ve been doing some work with a new web platform called FluidDB in the last few days, basically adding some functionality to the open source .NET API library. These will be available publicly very soon (which hopefully should take the state of the library from ‘very raw’ to ‘barely useable’.

I will describe my understanding of FluidDB at this point:
FluidDB makes it possible to create and use OOP style objects stored in “the cloud” where you or anyone else can add properties to those object at any time. So if Facebook were to use FluidDB, each person and photo would be a publicly available object, and it would be possible for anyone to attach additional properties to people at any time. So www.hotornot.com would be able to simply attach a property to each photo with a rating out of ten. This would not impact on Facebook at all, but would provide additional functionality for HotOrNot. Of course, the creator of properties attached to an object has full control over access rights, so you can make properties hidden or read only if desired.

What this means in practice is that if web services start to use the system, then it is much easier for new services to come along and use this publicly available data and combine it in new and amazing ways.

I have two major thoughts about the system:
1. Privacy – It is already scary how much data is available online and how easy it is to find. If there was one large backing store for all your personal data and it was all linked together, would that be a good thing?

Actually I don’t think you can treat this as a negative, because ultimately the data is already out there, and it would only mean that you would need to be more careful about how you present your digital persona. Raising awareness of that is probably a positive thing!

2. Data control – The system relies on having a single database for data which raises several concerns. Firstly are practical considerations- can they handle the quantity of traffic, keep spam under control, provide a decent level of security, and support enterprise usage? Secondly is a more complex question- is it a good idea for any individual entity to have control of so much data? How will they deal with copyright and intellectual property infringement, what level of government intervention will they allow when under pressure, what happens if you get banned and become an outsider? This might sound paranoid but I think it is something worth considering (and yes Google scares me).

Still, it is an interesting service in it’s fledgling state, and I would suggest taking a lot at http://www.fluidinfo.com if you are interested.

4 Comments

Analysis

by admin on January 10, 2010 at 9:17 pm

Yesterday I visited an apartment belonging to a friend of a friend for drinks with three guys I know well, and two I had never met before. After a couple of hours we decided to make a move to a bar and so we walked out of the kitchen into the hall, and as I waited for everyone to get their coats, a drunken wrestling match broke out in a bedroom leaving myself and one of my new acquaintances (Jack) in the entrance to the apartment. Seeing legs flying and the creaking of the bed, I couldn’t help but glance into the bedroom to see what exactly was going on. I was suprised to hear the following from Jack-

“I can see you’re not interested in talking to me, you’re thinking ‘I’d much rather be in that bedroom with everyone else’”

I found this very interesting. Firstly, when reading a statement such as that, it is easy to assume that the person saying this has low self-esteem or confidence, taking my lack of interest in conversation as something of an insult. In fact it was delivered in a really flat way as if [i]I[/i] was the one who should be insulted because of my lack of conversation and/or social skills. I say insulted, but of course the comment was light hearted.

Secondly, my reaction was totally defensive, almost as if I was embarrassed to be “found out” of not wanting a conversation. In fact I hadn’t even noticed that we [i]should[/i] have been having a conversation which is why it caught me so off-guard.

Thirdly, once I started thinking that I should be conversing, I could not think of a single thing to say. This isn’t exactly uncommon: at the point of a silence becoming akward (all in the mind of the participant of course) minds go blank, thus creating a self reinforcing cycle.

In conclusion, it’s not what you say, its how you say it.

1 Comment

I want to create something.

by admin on January 8, 2010 at 9:31 pm

I listen to songs everyday which evoke swinging emotions, I read books which contain great stories and great wisdom, I listen to people speaking amazing arguments, and I watch films which create all-encompassing atmouspheres which make me want to just stay, as Aerosmith would say, “lost in this moment forever”.

I don’t really get poetry, I’m trying, but school English lessons have thus far ruined it for me. I’m not greatly turned on by “Art”, either paintings or drawings or sculptures or wanky concepts which have no form at all. I have a passing interest in dance, but it’s not something I am passionate about. I enjoy eating great food, but its not a passion for me (I blame my awful sense of smell for this one). But that is fine, I can see the attraction in these things for other people.

Do you remember the feeling you had after seeing the final scenes of Blade Runner for the first time, or during some of the awesome night forest shots in Avatar, or after discovering a song which totally captured your mood at the time, or maybe when out running in the rain and finding energy from no where to sprint across a field with your arms spread wide? (Please insert one of your own epiphany moments here) If I can give that feeling to someone even once, then I will be pretty happy. The question is, how am I going to do that?

The thing which I really enjoy (creatively) is writing software. It occurs to me that there is no reason why software cannot be totally immersive. In fact, since it is intrinsically more interactive than any of the aforementioned activites it may well be easier. I’m not talking specifically about games, although I think they are by far the best example of this to date. I’m not a big gamer so I can’t really give any up-to-date examples, but when playing the original Half-Life I remember being totally immersed with the whole environment- the visuals, sound effects, music, and control system.

Now if one was to apply the same thinking to an application, giving it a very distinctive, stylized UI, atmospheric music, sound effects, an intuitive control system then what would you have? I suppose it would be a very specialized application, and probably very annoying since if using it for extended periods you would want music to fit your mood rather than the preset music or whatever. The trouble with applications is that they are more general purpose than games, they do things which go beyond the original design of the application, they create things. Games on the other hand give you the illusion of being in control whilst navigating along a relatively linear story.

As you can probably tell, I haven’t had a solid idea about this (otherwise I would be doing it!), I’m just playing about with some thoughts. I haven’t been particularly interested in UX/UI design until recently, but I think that there is a huge amount of scope for improvement over the way in which many current applications operate, and that is something that I think I would like to tackle. Of course with the variety of different platforms running applications from phones to desktops, and the way in which web apps have taken over for most simple tasks, there is more and more scope for interesting design, and more tools with which to implement them.

  Comment