Monday, 5 October 2009

A new life...

Graduating from college, throwing away the shite I collected over 22 years of my live, moving to another country, finding a place to live, getting my first job and where do I find myself? Back in the throes of Open Source.

I've always found it a struggle to find where my life, work and my passion for open source can all live together without stepping on each other's toes. Just now that I'm starting to try and get used to the norms of office life and working for a living ( sigh ) I'm finding myself trying to define where my time should be spent. We are all familiar with working 9 -5 , so that one is simple enough. Most of us know about "time well wasted" as I like to call it, where you come home from work and spend way too long making a cuppa tea to ever be considered an efficient tea maker or you sit in a chair looking out the window at that cat that has made it onto your car and seems to be making some kind of nest. Time well wasted is always time well spent, it helps to keep us sane! Some of the really lucky ones know about "Hobby Time" and are little known Ultimate Frisbee champions or have an almost complete collection of sea shells that resemble pop stars. But in all this mishmash of social norms where does the term "Open Source Development" come in?

I do not have the good fortune of having a 9-5 that satisfies my Open Source itch. I am certain that I have too few brain cells working during any of my "Time well wasted" sessions to even know what a computer is! So then is my Open Source Time doomed to share a room with random collectables and fringe sports? Do I have to class my Open Source Development as a hobby and stop refering to it as "my calling"? And more importantly how do I fit Open Source into my busy schedule of TV, Drinks with the guys, dancing in sweaty nightclubs and counting the amount of peas I have left in my freezer!? These are very pressing questions that hopefully I will soon have an answer to.

A Long time ago I was crap at time management. I'm still crap now but I like to think I'm getting better ;) One of the things that I found on my journey of self enlightenment was David Allen's Getting Things Done® and I found it FANTASTIC! I was a mess of lists and to-do items and forgetting stuff before David Allen and decided to follow his teaching like they were gospel! .... that lasted about two weeks, maybe 3 if you count having the lists right and not following them. I think the problem was that it didn't fit into my lifestyle, I was a student who was interested in a whole load of things and needed to balance research for 8 college modules with what a wanted to know in the world of open source! If I were a high flying banker or an office goer then I'm sure the system would have worked perfectly but it was lacking something that I needed. It took me a fair deal of time to figure out what that something was.

And now to the point of this long post. The Getting Things Done® method was wrong for me because the "Things" that I wanted to do were not defined. These "things" that David was trying to sort in such a fantastic way were things that lived in 9-5 time. The "things" that I wanted to sort out were usually crazy Ideas that began their life while I was counting peas: "Wouldn't it be great to have an Open Source Project that counted all the peas in my freezer and geolocated all the farms that the peas came from". These crazy thoughts were not definable, could not be put into lists and could not be thought of as "things". Instead they are stuff. Stuff is more fuzzy, less defined and does not conform to to-do lists and schedules, until now!

This long winded blog post is a way for me to announce a new project. "Getting Stuff Done". This name may spark some lawsuits but for now I'm going to leave it as a codename, so that it gets across the idea behind this project. Its a simple tool that aims to organise lists in a non-standard way, so that they are more human and much more malleable. I have the first screenshot here:

The more well read of you will recognise this as the simple Kontact "Hello World" example. This does mean that the project is in its most infant stage, but each long journey begins with only a single step.

Comments are very welcome ;)

Tuesday, 21 July 2009

And i give you .....

OSM rendering in Marble!

Oh yes, that's right! We now have full OSM rendering. It has been quite a struggle to get to this point but the hard work has really paid off. The results are just beautiful! I'm not just braging about my own code, because as it happens there is a lot of work to do to get the OSM rendering in a pretty way, but instead I'm talking about the experience of loading an OSM file. It looks marvellous! See the video below and notice how precise the positioning of the lines and nodes are with respect to the underlying OSM tiles ( I know this is obvious but it just looks so synchronised! ).

I don't know about you guys but seeing that map rendering, where the data has been collected and modified with a lot of hard work from a lot of people, really gives me a sense of grandeur. Fair play to all the people in Open Street Map, my hat is really off to you guys ;)

Link for those reading this on the planet.

Friday, 17 July 2009

Life Hacks Really Do Work!

Hey all!

As my friend John Layt suggest I'll give you a little recap on what I'm doing. My GSoC project is to implement a nice Open Street Map annotation framework to make the lives of all you mappers a little bit easier! More info can be found on my proposal blog post. The thing that has really been a challenge with this project was the fact that I'm adding support for a QGraphicsItem like API to marble. Little bit hairy at times but its good fun too! This has really come quite far in the last week and you can see quite an exciting example of this in the video below ( planet viewers have to come to my blog because the planet doesn't embed embedded youtube for some reason ).

Starting Monday I brought out a little "life hack" that I used quite a lot in my last year of college. Quit simple, very loosely based on the "Getting Things Done" idea and it can be really effective when you have something that doesn't have a strict system of "This must be done by this time" or "This must be done first". Maybe I might blog about it sometime ;)

With this new life hack in hand I have fixed two bugs, closed one bug, identified the cause of another bug, cleaned up most of my code, made the Marble plugin system more flexible, started implementing default base GeoGraphicsItem classes ( quite similar to the idea of QGrahpicsLineItem being a default implementation of QGraphicsItem ), fixed the problem with the GeoXmlParser that was preventing me from loading OSM files and made my little text editor a damn sight more pretty! Wow that was a mouthful!

Now with that all under my belt I announce this little clip of the results. As you all know a lot of the stuff I was doing is hidden from view, and also a lot of it is very much visible but just being done a different way but I think it looks pretty cool too ;)

(Link for people reading on the planet)

There are some very exciting things planned for the coming week ( and even just today ) so watch this space. I hope you all like my little updates, if any of you would like me to go into any more detail about anything just drop me a line ( comment or email ) and I'm sure I can accommodate you in some way!

Happy Coding!

Tuesday, 14 July 2009

Morning Blog on GeoGraphicsView

Hello again everyone!

I too am one of the lucky people that got the chance to go to Akademy ( or whatever we are calling it these days ) and it was a truly great experience. Apart from being fantastic craic ( its an Irish thing ;) ) I got a lot of work done, or at least a lot of work discussed, and I'm ready for the next phase of my project. The phase of awesomeness!! ... maybe I need to work on the name...

The main problem that I have been faced with is a pretty big one. We are currently implementing a Marble API that is as close to the QGraphicsView as possible ( see my last post for info ). I mentioned that a lot of the stuff has been done before but there were two problems with the current direction of things : 1) there was no real interaction framework or hit tests and 2) we were working with implicit sharing... which is a big NO for a QGraphicsView re-implementation.

Before a week ago I didn't even see this as a problem, but now that I've taken the role of a Guinea Pig implementer of this framework it has been shown in a new light. The problem was that somewhere during the implicit sharing and copying values from references there was an explicit downcast of all classes. This may not be a bad thing all of the time because we adopted a QVariant style system but for plugin developers this made things get very hairy. To make a long story short, if a plugin designer wanted to make a new weird and wacky GeoGraphicsItem they would have to directly edit the sources of the Marble Library to make this happen. Not Good.

So that's what I'm working on right now. If anyone feels like watching marble crash for the fun of it just check out the svn and enable my stuff ( the osmannotation plugin ) and let the fun begin! The joys of CMake and plugins have given me a great sandbox or more importantly kept my crazy code from all you nice people. Hopefully after all this all you third party developers will be able to make a really cool marble plugin with absolutely no effort, so watch this space!

Saturday, 20 June 2009

Update and Exciting Features

Right so I thought it was about time to give a bit of an update on what I have been working on so far for my GSoC project. I have done some really interesting stuff with a lot of ground yet to cover so I'll try to keep it brief ( and I hope to update you all more frequently ).

If you have a look at my last post you can see the big goals that I have created for myself, which is a definite positive as I believe you can never aim high enough! The first of these goals is what i called the Annotation Rendering Framework, which has proved to be a challenge. The Idea of the rendering framework was to provide similar functionality to the QGraphicsView, QGraphicsScene and QGraphicsItem classes in the Qt Gui Module. And believe me, thats a hard act to follow! The main feature from the Qt classes that I wanted to recreate is the ability to "Click and Drag" items around the interface of marble, which is made difficult by the fact that marble is not a 2D surface and hit tests get a little bit more complicated.

Thankfully I have the opportunity to stand on the shoulders of giants to get all of this working. The Marble developers have implemented a lot of the necessary functionality to get my stuff off the ground an its my job to extend this and integrate it all in a useful way. So without further ado I have a small video to present you with showing a little bit of my work.

The next step is to get nice little pop-up showing the details of each annotation, be it text, video or audio. This hopefully shouldn't take very long as I have already done a few experiments to do with Geo-Located QWidgets so watch this space!

Wednesday, 22 April 2009

Accepted Google Summer of Code Proposal

To follow the trend of my organisation's GSoC mailing list I have decided to make my proposal public, this should help everyone see what tasty treats might be coming in the near future and it is also to allow people to give me some feedback on the ideas.

The post will be , for the most part, word for word the same as the the application. The only bits that will be missing are the contact details or boring information.


OSM Annotation for Marble Desktop Globe with an Optimised interface for smaller screens


Andrew Manson


This project will allow for the annotation of raw GPS data for use in the the mapping process of Open Street Map (OSM). GPS will be annotated with Geo-Location aware text notes, voice recordings, images and videos while tracing new GPS routes. You will also be able to annotate existing OSM data to verify its accuracy. The project will be aimed at a small, light device with GPS capabilities that can run the Marble Desktop Globe such as the EEEPC and the interface will be optimised for this.


During the course of my studies I have been very interested in Geographical Information Systems and have chosen to do as many projects as possible in this area. This is why I was drawn towards the Marble project and I was very fortunate to be given the opportunity to work with the Marble team in the 2007 Google Summer of Code where I implemented the current GPS functionality that can be seen in Marble today. Since then I have been pretty quiet on the development front, because of my college commitments, but have always remained interested in returning to the Marble project to improve the work that I have done in the past and to implement new features.

This project idea was part of the proposed project list last year and since then it has evolved somewhat and I have even tried out some of the techniques that are necessary to implement it as a part of my Final Year Project in University, which I hope to be able to integrate with Marble sometime in the future.

The overall goal of this project is to give Open Street Map contributors an integrated, useful and intuitive platform to collect new raw GPS data while interactively annotating the collected data with Geo-location aware notes ( text, audio and video ) that will be helpful to identify the data while uploading it to the Open Street Map server. During this process we should be able to display the latest Open Street Map data to the user so that they may verify its accuracy. This data can either be streamed live over the internet, if a connection is available, or if not then sections of the data could be downloaded to the local machine.

Although my main goal is to provide this annotation tool there will be a few essential features that must be added to Marble before the tool can be integrated. These will be further discussed in the Implementation section below.

Implementation Details:

The most notable new feature that I would need to add to Marble would be the ability to interact with data that is currently being displayed to the Marble Widget. To clarify exactly what I mean by "interact":

  • The ability to add an information object to the map at a certain Geo-location.

  • The ability to click and drag the item to a different Geo-location within the view.

Currently this behaviour is not available to Marble and once it is implemented it will allow for a more in-depth user experience, offering the potential to edit all data items directly within Marble.

I have been discussing the implementation of the of this functionality with some of the Marble developers and was considering creating a custom QGraphicsView overlay system for the Marble Display Widget. This approach however will not be included in my work for Google Summer of Code because it would be completely experimental and would mean that my work would not be directly applicable to the current SVN trunk. Instead I will integrate the system into the current rendering pipeline and try and keep the API for this section as close to the QGraphicsView API so that future integration with this part of the framework will be an easy process.

Another notable feature that will be required for this project is the ability to load and render raw Open Street Map data to the Marble widget so that it can be updated and used as a reference while the user is mapping. Although there is already an OSM view in Marble the implementation I require would be quite different from the current solution. Currently the view is created by downloading flat OSM tiles that have been prepared for Marble, Although they can be used as a visual aid to the user while mapping, they do not contain all of the information that is available on the OSM servers. Once the data is loaded some additional features may be added to the project such as making the system aware when you have ventured into an area that has not yet been mapped or adding simple routing algorithms to Marble using the loaded OSM data. These extra features will be listed in the "If time permits" section at the bottom of the Timeline below.

The other necessary features shouldn't be too difficult to implement considering the implementation of Qt, for instance: rendering text boxes to the map or having small images visible on the map. The difficulty here will be the seamless integration that I am aiming for in the capturing of images, videos, audio and the taking of text notes. Text notes will be easy enough to accomplish as I will be developing specifically for an EEEPC but the integration between Marble and cameras or microphones will be more difficult. Here is a simple image that I hope describes what I want to accomplish from this project:

Although the example annotations in the above picture are unremarkable I'm sure you can understand from my descriptions that getting to this stage will be a great success.

The following is just an itemised list of what I intend to do and is meant as a reference.

Annotation Rendering Framework

  • Be able to add a text, audio or video item to the display

  • Have the annotations linked to a GeoData location aware data structure

  • Allow for annotations to be moved, i.e. interactively alter their Geolocation.

Live Annotation Interface

  • Make use of the live GPS tracking that is available in marble

  • Be able to "drop" an annotation (picture, text ) at a current location

  • Have Geolocation aware Audio and Video

    • Make the annotation item aware of the GPS track that represents where it was recorded.

    • Move the annotation item along its GPS track during playback

Open Street Map Integration

  • Download OSM data live from their servers

  • Download OSM information to a local cache for use when offline

    • Allow users to choose what sections of the map to download

    • Be able to update local cache with newest information from the OSM servers

Interface Intergration

  • Small Screen Optimisations

  • Integrate the Recording devices and commands into the Marble Widget

-If time permits-

  • automatic annotation i.e. adding "Current position not currently on OSM" when in unmapped area

  • moving "camera" along the GPS track to replay where you traveled.

  • sync moving camera to moving Video/Audio annotation item.

Tentative Timeline:

For the most part of the project I will be adopting a spiral development model where I will work on each of the separate parts of this project at the same time, adding a few features at a time. This will make it easier for me to show my supervisor what progress I have made and also allow me to get some feedback from the Open Street Map community ( and of course the Marble community ) at regular intervals throughout the project.

The only full features that will need to be completed first are the ones listed under the heading "Annotation Rendering Framework" as listed above. This is the only essential prerequisite for my project. This section should take me less than a week to implement and after that the main development model can get underway.

By the mid-term evaluation I would like to have:

  • Live annotation, giving the user the ability to add an annotation to their current location with a single click.

  • Preliminary implementation of the Open Street Map display so that users can see without any trouble where still needs to be mapped.

  • Preliminary integration of the Recording devices into the Interface, i.e. having the ability to seamlessly record even if we cannot display currently connected devices.

  • Small Screen Interface patches. This work will be ongoing.

I am confident that the rest of the features that are listed above can all be implemented within the time of Google Summer of Code and I will hopefully be able to adapt the implementation based on user suggestions within the timeframe.

About Me (let us know who you are!):

I'm a final year student of Computer Science in Trinity College Dubin. Since this is my last year of a 4 year degree I've had the opportunity to gain a lot of experience in developing applications and systems to a high standard. When it comes to developing systems as part of a group I consider myself to be a Plant, that is, a person who brings discussion and innovative ideas to a project and tries to think outside the box. This, along with my general aptitude for coding, has allowed me to become the project leader for most of the group projects that i have participated in. Although i feel comfortable and at ease while working as part of a team, I have also excelled in my individual projects, even as early as first year where I won the award for Best Programmer in my class.

In the past year i have developed an interest in Embedded Systems and electronics, giving me another outlet for my inventions and for me to develop some more of my interesting ideas.

If I were to be accepted to the programme, the first thing that I would do would be to register for the Akademy conference, I really enjoyed it two years ago and found it to be very informative, so much so that I am extremely eager to go back. Google Summer of Code is sufficiently flexible to allow me to go for the entirety of the conference, and naturally GSoC and Akademy would be mutually beneficial.

Wednesday, 15 April 2009

KDE 4.3 Release Schedule

After Finding out there was no central ICal for The KDE Release Schedule I decided to make my own and share it with the world, this is Open Source after all ;) if you want to follow this calendar yourself in you usual PIM suite then use this link.

Friday, 20 March 2009

Returning to the nest

You will probably notice that the last post that I've actually put on this blog was 2 years ago, 2 long and glorious years that have been a great chance to improve my experience and expertise with software and coding. I'm still in college at the moment but I reckoned that it was about time i returned to the project that started my life as an open source contributor, Marble.

Marble is a nice 3D(ish) globe program that is part of the KDE project. It is mostly built on Qt so it is easily portable to any platform. What i set out yesterday to do was the relatively simple task, or so i thought, of understanding the XML parsing system of marble with the intention to implement a simple test implementation of a Open Street Map parser. When i started yesterday i was full of vigor and looking forward to the challenge, but by the end of the day i was sent home with my tail between my legs.

So you might be wondering what can scare a young programmer so much... Its all just a matter of understanding. The little bit of programming that i have done for marble 2 years ago consisted of a simple (naive) XML parser that read in GPX information and rendered it to the screen, but since then the XML story of marble has become a lot crazier. Not long after my time in marble a unknown programmer ( unknown to me, please post their name in the comments if you know ) came on board with a solution, and because of his work on the Webkit engine you can bet that he knows what he was talking about. The problem with his solution was simple, it was "Right".

Yes that's right, you heard me! I'm complaining that a solution is right and i have good reason! My goal yesterday was to understand the XML implementation in marble, i understood the concept in about 5 mins but the implementation was littered with so many "Best Practices" and "Time saving Tips" that i challenge anyone who hasn't spent some time doing production level XML parsing to try and understand it and implement a parser in less than a week! This kind of a system is so intimidating to any programmer that would consider having a pet project using Marble's Parser would be instantly scared off. I'm not saying that we should sacrifice all performance for "Ease of use" but i really believe in a balance, one that will be inviting to new programmers but still encourage them to learn some of the "Best" ways of doing something.

I'm not going to get into the details of what i really didn't like about the parser but i will list a few points.

  • Macros! god i hate macros, and i would put that down to being thought in my first year of programming "Don't ever use Macros, they are Evil". I wouldn't necessarily say that they are evil but using them to save time really feels like a bad idea to me, a more structured architectural model may result in less use of macros? who knows!
  • Element dictionary! Ok this sounds like a good idea in theory, and i can see where this can be a very good idea in practice too, but i really don't think that a standard parser that uses an element dictionary is the "Best" way to account for the many varying formats that could represent data that a marble module or plug-in might be interested in.
  • Parser only able to return GeoData! one problem with using this approach with OSM is the xml tags do not completely enclose the relevant data in their children or parents. To actually implement an OSM parser you would need to modify the parser itself, which doesn't really make it a generic "One size fits all" parser.
Its not all bad. the system is very optimized but i really don't think that every Programmer that would want to implement a parser should have to ever look at something so complicated as "The way Marble does it". I believe that any Plugin or module that needs a parser should implement their own using Qt's excellent parsing capabilities.

I have proposed a solution for a middle ground to the head developer already, we have discussed it at length and i will be doing an experimental solution based on what we discussed to see what it's performance is like.

I might be convinced, at some later stage, that the parser model in marble is in fact the best solution for Everything! but just not right now!

Till next time,
Over and out