an open university blog (not official!)

Interesting article on the guardian.co.uk Technology section by Darrel Ince (a professor in the OU computing department) about programs used during scientific work and how in order to allow full checking of the data and method the programs should in his opinion be published as part of the research. From the article:

Many climate scientists have refused to publish their computer programs. I suggest is that this is both unscientific behaviour and, equally importantly, ignores a major problem: that scientific software has got a poor reputation for error.

[...]

By contrast [with commercial development methods and the quality assurance and testing functions carried out there] scientific software developed in our universities and research institutes is often produced by scientists with no training in software engineering and with no quality mechanisms in place and so, no doubt, the occurrence of errors will be even higher.

Historically many scientific research teams have refused to release the code involved in their projects and consequently the ’scientific method’ cannot be fully applied since it’s then impossible for others to check, validate and replicate the code.

Professor Ince concludes that:

if you are publishing research articles that use computer programs, if you want to claim that you are engaging in science, the programs are in your possession and you will not release them then I would not regard you as a scientist; I would also regard any papers based on the software as null and void.

The article doesn’t really mention the open source movement specifically but it’s a similar thing in that part of the advantage behind open source as well as allowing modification (subject to relevant licence) of the code in order to produce additional versions of the software it also allows the code to be examined and checked, tested, debugged etc by a much wider audience of people than the teams originally involved in writing the code. It’s also that ‘programming’ historically the province of the computer science geeks is now of course increasingly used in many disciplines and as with many times technology has been adopted from a core group to general acceptance often not all of the stringency around quality and appropriate use comes through. (Statistics is another one that comes to mind..)

Due to the amount of project related work I’ll have this year (T307 has a significant project component and M450 is a project course) I thought I’d try and take a more organised approach to project management, keeping track of ideas and information etc – normally I’ve done this either by keeping lots of stuff in state in my head (usually) or scribbling notes on paper.

The first TMA for M450 is to outline the project topic (including the usual ‘problem definition’ bit such as: what is the problem? why is it a problem? what is the suggested approach to the solution? etc), discuss any practical work carried out so far and a couple of other bits. The blurb for the TMA has a certain structure it’s ‘recommended’ for this to be laid out in which means a bit of step-by-step and lists of information to compile etc so I guess by ‘recommended’ it means that’s the format it should be in, pretty much.

For Linux I’ve discovered the View Your Mind software for creating mind maps (or as I’ve always known them ’spider diagrams’ but apparently most people don’t know this terminology?!) which is also available through the Ubuntu repositories and probably those of other distributions as well.

I’ve actually found it most useful to come up with a ‘rough’ mind map on paper as and when I think of it – which is usually on a train or something similar – and put the mind map into the laptop later. As I think trying to structure information (and mind maps do have structure and hierarchy of course, even though they aren’t a sequential list) too early on is counter productive and very likely to miss connections between ideas because structure was imposed on it too soon. Generally I will throw out ideas in “any” order onto a sheet of paper and then organise them later and this is the way I’ve been using the mind map software.

The first mind map is for the parts I’ll need to work out for the TMA (obviously I haven’t included any detail here such as actual wording from the TMA, specifics of problem or implementation, etc!). I represented it with the “core” M450 TMA 01 on the right and the hierarchy structure leading up to it from the left as it is meant to be that all of the sub-sub-topics etc need detailing first before they make up each group and the groups make up the TMA. Mind maps for actual ideas for the project (e.g. functionality/requirements for the user interface) would look quite different to this.

Ultimately though the limitation I’ve found with most of the mind map methods is that they do have this strict hierarchy (e.g. topics 1, 2 and 3 relate to the central topic, and 1a/1b/1c come from topic 1.. etc) and don’t allow for linking between ‘nodes’ that aren’t part of the same subtree very easily. I don’t think mind maps/spider diagrams will ever become my real thought process for brainstorming ideas or working through things in a structured way but are at least a useful way to represent ‘exploratory’ information in the TMAs and eventual project report! The emphasis seems strongly on process and method as much as the actual results, with a lot of reflective practice and ‘what parts did you find easy/hard’ etc. The project for T211 was a bit like this with a design log having to be submitted and a lot of writing about the process we went through (some of which I suspect was ‘reconstructed’ after the event for a lot of people..) so the process is not new to me but I still find the reflective practice bit quite difficult as steps that I go through intuitively or naturally need to be made explicit and detailed out step by step!

I had been putting this off for a while but this weekend I did finally manage to knuckle down and start on some initial work towards the first TMA (topic selection and associated preliminary work for M450 – the computing project). As is probably normal with these things, I had a number of ideas but many were either too vague and couldn’t be pinned down in the detail the project requires, too ‘large’ or ’small’ or otherwise unsuitable e.g. not being relevant enough to the core course I’ve selected which is M366 – Natural and Artificial Intelligence.

This still leaves a few ideas though (I’m a person of a lot of ideas and less action!) and I think I’ve now decided on one to pursue in a bit more depth. As it’s a real piece of academic work and thus subject to all the normal rules (plagiarism etc) I won’t be able to put actual parts of the project and code etc on here but think I can write in general terms as other students have done in previous years (and why is it that most of the OU student blogs are from people studying computing, IT or technology courses?).

The project (if it’s agreed with the tutor) will be based around genetic algorithms and related ideas (GAs are explicitly studied in M366 in some depth but there are a number of other AI techniques that are similar which I will investigate as well) and will relate to music and music generation. As a bit of a music enthusiast myself it’s a good chance to tie together different interests and because it is actually of interest to me rather than a purely academic exercise (though it is academic of course) there’s a much better chance I’ll be motivated to work on it…

Subject to the requirements of any project course (M450 or another) I would always suggest to choose a topic you are personally interested in and know something about, rather than something ‘easy’ – whilst staying objective as much as possible.

Most of the students on M450 are probably new to independent project work and doing literature reviews etc so there is a focus on that element of the procedure in the early TMAs – luckily or unluckily having already done the Masters I’ve been through the literature review and research project bit before (in child development and psychological research methods, not computing, but most of the same principles apply in terms of technique) so hopefully this will be less of a learning curve, which is useful as I’m starting to think I’ve bitten off more than I can chew so to speak with this GA idea!

Semi regularly I do a search on ‘next course’ from StudentHome to find out if there are any shiny new courses being launched (that I haven’t otherwise heard about e.g. through the Alumni e-newsletter) and one of the courses being launched in this time’s ‘oeuvre’ is this one on Linux: An Introduction (T155) which is a Level 1 short course, as might be expected from the title, being a introduction to the use of Linux (seemingly primarily in the desktop area but also as a server etc).

Starts in May 2010 and then October (then presumably May and October presentations – “twice a year”)

From the page:

The myth of Linux is based on the misconception that it’s a difficult operating system to understand, hard to use and has many issues surrounding its quality. As one of the most extensively used operating systems worldwide, the chances are that the website you last visited was running on a Linux-supported system. In this ten-week course – designed for absolute beginners – you’ll examine the many similarities that exist between Linux and operating systems such as Microsoft Windows. You’ll also explore the diverse technology available in the Linux community. You’ll be provided with tools to access free versions of Linux and software to install this on your computer.

[...]

You will spend about half of your overall study time doing practical activities. The course uses the current long-term stable release of Ubuntu that you can access via the course website as well as directly from the Canonical website. The computer specification listed below is a guide; Linux will run on many older systems with considerable success.

You will have the opportunity to create your own bespoke version of Linux based on a common core operating system in the end-of-course assessment.

This is good news I think in terms of introducing Linux to more people (as you may have read from some of my other posts I am a primary Linux user at home and only regress boot to Windows for the odd thing e.g. TomTom sat nav software..) though I would debate whether a 10 point introductory course is needed, Ubuntu (and many of the other distributions such as Mandriva) is actually very user friendly, has good help and a decent community available and the usability is such that carrying out many tasks is ‘obvious’ at least to me as an experienced computer user. Still, for people newer to computers, who are less confident or who prefer to learn in a more ’structured’ way this course seems a good idea..

..now, if only the situation with official support of ‘minority’ operating systems such as Mac (though Mac is getting better) and the various Unix/Linux variants could be improved!

I am not quite sure who this course is aimed at, whether it’s students new to the OU who want to take this specific course to learn about using Linux in a structured learning environment or to existing students who want to learn this as an additional skill? Would be great if this leads to more acceptance of Linux at the OU more broadly but I don’t know what the plan is with this.

The changes to the OU/Tesco clubcard voucher scheme (briefly: as of June the clubcard deals vouchers will only be able to be used towards Level 1 courses whereas currently they can apply to any undergraduate level course) have now been detailed on the student page and reflected in the ‘terms and conditions’ part – also this article on BBC news online picks up the story and their understanding is that this has come from the OU rather than Tesco and is due to budget constraints and the current economic climate.

All of this makes sense of course but understandably some students (including myself) are a little annoyed about this having changed shopping habits etc but not much that can be done (unless you count joining a campaign group on Facebook, apparently one has been started!)

As mentioned in this previous post the OU offers some audio and video tracks and podcasts from the various courses through iTunes U. For most people (willing to use iTunes) this is fine, however as a proprietary format and given that iTunes cannot be used under Linux (unless you do a virtual installation of Windows or dual boot etc – it won’t run under WINE!) it would exclude Linux users from being able to view/download content from here.

I have recently discovered the OU Podcasts site which offers the audio and video in a more friendly format as it can be accessed through the browser – each podcast can be subscribed to through RSS or just played on the site (or subscribed to in iTunes…!) allowing viewing of updates etc. Not all courses have media available but out of my courses T307 does so I spent some time already going through the videos there. Audio tracks of many of the videos can also be downloaded.

Being unable to use iTunes U on Linux I haven’t been able to check that the content is absolutely the same, but since the Podcast pages allow viewing in iTunes U it seems a fair assumption that they are the same.

I’ve not seen any reference to OU Podcasts on any of the OU sites so not sure how many people know of this or how well publicised it is, but it is obviously intended to be publicly viewed (though some podcasts are marked as staff/student accessible only) and is not a ‘back door’ or anything like that, just a more friendly way of viewing the content!

The site is organised in a similar way to the OU prospectus with the various subject areas in a navigation bar at the side allowing ‘drilling down’ to particular disciplines (also a search box).

Seems there will be a change to the Tesco Clubcard Deals / OU voucher scheme later this year. From the Sesame website:

From 1st June 2010, Tokens may only be used as full or part payment towards Level One courses.

(see link for full text)

So far I and a lot of other students have been using the vouchers towards Level 2 and 3 courses as well as Level 1 as part of the attraction was that they could be used for ongoing students as well as new recruits (you would probably expect new students to be more likely to start at Level 1 anyway – though I don’t have any figures for this) so this will potentially affect quite a lot of people. Of course nothing is guaranteed with things like this and terms and conditions can always be changed but this may change a few people’s plans for how they will pay.

This does not yet (as of the date of this post) appear to have been changed on the OU’s site and so far the only reference I can find to this is the Sesame site though I daresay the wording may somewhere else ‘out there’…

Therefore if you have any Level 2 or 3 to register for next year, the suggestion on the Sesame site is to do the payment before the end of May (e.g. for courses starting in Feb) and the vouchers can be used, but not after that.

Personally this is a bit of a blow to me as I had started to consolidate shopping at Tesco in order to build up the points and exchange them for the vouchers – other Deals still apply but I couldn’t find that many other things I was interested in (though don’t get me wrong, there is quite a lot of choice!)

As suggested a few months back the OU have now launched a degree (BA or BSc (Hons) depending on choice of courses etc) B61 in Design and Innovation which will use U101 (Design Thinking) as compulsory level 1, also T211 (Design and Designing) and T307 (Innovation) as well as some other specified courses. U101 has only launched this year so is currently in its first presentation – the others have been going for a while.

The optional courses for the degree (other than obviously the Design ones) involve at least 60 points at level 2 and 60 from level 3 subjects such as business, engineering, systems, etc. (The web page has the full list of acceptable courses)

As I am interested in the programming element as well as design/innovation I will be sticking with the other BSc(Hons) in Computing and Design, but this new one seems to offer a more specific title (and set of courses) to those interested in design and innovation specifically. Previously we’ve had the Diploma in Design and Innovation which I will complete this year (all being well) with T307, but nothing further than that – the Computing and Design is itself a relatively recent introduction.

Having now received the materials for T307 as well (it was a few days ago now) I now have the complete set of books etc to start the 3 courses this year! T307 contains the course guide, course DVDs, blocks 1-3 with their associated Guides and a couple of other bits and pieces. So far I’ve not gone into it in too much depth but the course seems to start where T211 left off and the actual design of the books is very similar so hopefully it will seem quite familiar.

Just a case of reading all of them and doing the TMAs now then

A couple of days ago the M256 books arrived via DHL and as usual instead of getting on with more pressing tasks I went straight into whizzing through the box and having a look at the shiny new materials!:

The contents of the box are:

• Course Units 1-14 (see below)
• NetBeans guide
• Course guide
• Letter from course team chair
• Disc with M256 course software
• Disc with NetBeans software
• M256 handbook

(Hooray, they have finally stopped sending a “general software” disc with every package – how many copies of Winzip etc do we really need! If anyone still really wants one of these I expect they can be requested from the OU… No flowers.)

The course is divided into 14 blocks and according to the index details they are:

• Introduction to software development
• Requirements specification
• Creating a conceptual model
• More conceptual models

• Component based software development
• Core system designs
• Design decisions
• Designs in action

• Detailed design
• Coding and testing
• Human-Computer Interaction

• User interface design
• Building user interfaces
• Pursuing software development

I’ve had a scan through a couple of the books including Unit 1 in some depth, and they look generally approachable though quite full of diagrams. UML (or less strict variants) seems to be used quite extensively for giving relationships between different types of data and code etc, also diagrams for the various different software development methods and so on.

The emphasis of the course as detailed in the prospectus is much more on the design and ‘business’ aspect of software development rather than the actual code itself, though some Java is used throughout for illustrating the ideas etc.

Other materials such as the course calendar and TMA booklet are online only and available through the course website once it opens. As there is about a week to go until it opens, I haven’t been able to get a look at the TMAs yet but expect I will look as far ahead as I can (sometimes they are only published on a “just in time” basis rather than all at the start) to see what sort of thing is likely to be requested!

Having looked at this M256 will almost certainly be the easiest and most approachable of the courses this year, though will also help with M450 in terms of development methods etc (part of the M450 planning is to come up with a project approach and decide what sort of development model would be used: waterfall, iterative, etc). With the user interface and design aspects it has links to some of the other courses I’ve done as well so I don’t think there will be that much in the way of totally new ideas here.