Vote. It’s simple.

Posted on March 4th, 2010 by Nicholas Jackson

This week, for those of you who didn’t know, is SU Elections Week. Go vote now if you haven’t done so already.

Why you should vote should be obvious – if you don’t vote then you have no right to complain about the SU next year. You may feel that your vote doesn’t count, but in the past people have won places with six votes. No, that’s not a typo. With six votes, not by. Vote, dammit!

On a related note, why is it that for two weeks every year the MAB turns into a bedding store? You can get enormous posters printed (suitable for outdoors) which will have far more impact than a bedsheet and spraypaint for not much money at all. Please, make the effort.

Now I’m really not going to fill out your questionnaire.

Posted on February 17th, 2010 by Nicholas Jackson

It happens every single year. A final year student, needing questionnaire responses for their dissertation, will decide it’s a jolly good idea to email the entire student body. I received one today in fact, from a student who managed to do the whole emailing thing catastrophically badly.

First of all, learn how to send emails to lots of people. Your mail client will have three options for addressing emails: To, CC and BCC. Use To if you’re sending to a couple of people. Use CC if there’s someone you want to copy in. Both these options let everybody see everybody else’s email address, which is fine if the email is going to anywhere up to around 6 or 7 people. If you’re sending to 10,667 people as this student decided to, for the love of God and all things holy use BCC. That means that each person receives the email as though it was addressed just to them, meaning I don’t have to scroll through 10,667 names before getting to the content of your message. It’s simple courtesy and means that my client doesn’t have to download 284kB of headers.

Next, tell me what the hell you’re writing to me about. A title of “Dissertation questionnaire. Regarding student support services.” is pretty vague, has incorrect capitalisation, is grammatically dodgy, has no call to action and generally makes me go “meh”. Even more important, the body of your email will do well from actually telling me what your dissertation is about, what your questionnaire is going to do and why I should fill it out. The wrong way to do the email body is this:

This will only take 10 minutes of your time.

Please help me with my dissertation research. It is very important!

Thank you.

Come on, signing your name isn’t too challenging, is it? Why is your research important? Why is it more important than my 10 minutes?

Still, assuming I have downloaded the stupidly long email header, dragged my way through the thousands of names in the To list and haven’t been put off by the terrible email body I might still want to complete your questionnaire…

Or not. It’s a Word document. I’m on a Mac. Some people are on Linux machines. Even those with Windows machines may not be bothered kicking open Word for you. Once people have completed your questionnaire, how do they get it back to you? You’ve never mentioned emailing it back. Once you’ve got the replies, how are you collating it? Manually? Why not make your life infinitely easier and use something like SurveyMonkey?

Finally, if you’re going to email 10,667 students, at least learn to spell “course”. “cource” is not acceptable. Also, we’re the University of Lincoln, not Lincoln University.

Institutional URI Shortening

Posted on February 16th, 2010 by Nicholas Jackson

Today I’ve been bulking out my references in my dissertation, finally adding items from my ever-growing references database to my text. One thing I discovered to my annoyance is that the University’s presentation regulations make displaying long URIs incredibly awkward. The margins required mean that (using LaTeX to typeset my work, and a monospaced font for displaying URIs) you can only fit 68 characters across a standard A4 page.

Now, for some URIs this isn’t a problem. However, when you’re referencing papers online you start to find longer and longer URIs, including the institution, department, authors, the entire title of the paper in the address. Even this blog’s URI is a whopping 73 characters, and some things I’m citing are in excess of 150 characters.

You may at this point be thinking “just use bit.ly, j.mp, tinyurl, ow.ly, tr.im or one of the other hundreds of URI minifiers”. I’m tempted, but there’s no guarantee that the minifier will still be there in a year. I accept that a lot of URIs will simply go AWOL in a few years anyway (although persistent URI practices are improving), but I’d like something more robust I can use in terms of address minifying.

At this point, enter the institutional minifier. Quite apart from the benefits inherent in being able to replace https://portal.lincoln.ac.uk/C14/C2/UniversityRegulations/Regulations/University%20Regulations%202009_10.doc (no, really, that’s a genuine URI which Lincoln puts in some material when telling students to look at the regulations) with something like http://go.lincoln.ac.uk/regs0910 (sorry, not a real URI. I’m working on it) the same functionality will be available to people trying to fit 150 character URIs into their dissertations. It could even be a different subdomain, just CNAMEd across (http://cite.lincoln.ac.uk) to make dissertations a bit more… professional.

Anyway, kudos to the people in L&LR who are beginning to use nn.nf to minify URIs in their print works, but wouldn’t an institutional one be so much nicer, as well as more likely to still be here in 5 years? It’ll take all of a day to build and implement.

Where has my lecture gone?

Posted on February 9th, 2010 by Nicholas Jackson

Today, I woke up at 8.15am (as is my new custom) and yomped into campus for a 9am lecture. I arrived at the room, only to find that it was disturbingly empty.

“Huh”, I thought. “It must have either moved or been cancelled. Never mind, Blackboard will tell me where my lecture is!”

So I logged in to Blackboard from my trusty phone, only to find a total lack of information about a moved or cancelled lecture. No messages, no alteration to the learning materials, nothing.

“Huh”, I thought. “Never mind, Timetabling will tell me where my lecture is!”.

So I clicked the timetable link and had a look at my timetable, proudly displaying “MB1001″. I double checked that this matched the number on display on the door – it did.

I honestly don’t mind lectures being moved around. I’m only slightly less annoyed by them being cancelled (since my tuition fees work out at about £11 an hour for my lectures and seminars). I do, however, appreciate being told. Our timetable page says, in a fetching shade of ‘pay attention to me’ red:

Check your timetable weekly to ensure you keep up to date with any changes that may have occurred.

It’s no bloody use if you don’t update the timetable, is it?

A Post About Nothing

Posted on February 6th, 2010 by Nicholas Jackson

Not much interesting or unique to report from the student life of Nick. Assessments continue to appear, be submitted, and disappear (incidentally my dreadful hand-in passed – just). Dissertation progresses more or less according to plan. I need to get hold of FERET to start doing some proper recognition and training stuff so I can find out where the whopping huge holes in the system are.

My never-ending role as Student Rep is more or less the same as before – nobody is complaining about anything particularly urgent, so we’re just waiting until the next subject committee. It’s in a couple of weeks, and I won’t be going so if you have any problems make sure to let us know in the moaning corner.

Finally, if you haven’t done so already, fill in your National Student Survey! If you don’t then you’ll be bugged by endless post, phone calls, and people knocking on your door at 3am. Well, maybe not the knocking on the door.

B is for…

Posted on January 26th, 2010 by Nicholas Jackson

Semester B kicks off, and so far I’ve forgotten to change my alarm on Tuesdays to reflect the joyous new 9am start. Fortunately I was with it enough to leap from bed and sprint to my lecture so I’ve not missed anything.

Distributed Computing involves Java, one of my least favourite of all languages. Apparently it’s good because it (probably) works on most platforms, but as far as I can see that’s a bit like saying anal sex is good because it works on all genders. I want to know when we get a unit on LOLCODE.

In other news, project continues to irritate. I’m really tempted to abandon C++ altogether and use SharperCV to muck about in C#. Sadly, SharperCV is no longer developed and I lose out on all the OpenCV 2.0 goodness. Looks like C++ is here to stay.

Communicate, Dammit!

Posted on January 14th, 2010 by Nicholas Jackson

Today, two things happened.

Firstly, I submitted possibly the single worst assessment I have ever done in my whole life, ever. Entirely my fault underestimating the time required (or the inner complexities of the network configuration), but to put it simply my report sucked and my code, although functional and commented, is flaky around the edges and would fall over if asked to do anything moderately tasking.

Secondly, I attended a mock job interview as part of Frontier Technologies, a unit described by some as talks nobody wants to attend by people we don’t know talking about things which aren’t relevant. The interview itself went well, and I received some handy pointers about technique and form of my CV and covering letter. So it turns out that after all this time I know how to apply for a job, woohoo!

However – and the reason for the title of this post – deadline for formal submission of my cover letter and CV was made yesterday, not today. This is the first time since I started University when a submission deadline hasn’t been a Thursday at 4pm, and it threw quite a lot of people and led to mild panic as people hurriedly scrambled to submit on Wednesday.

Communications wise, this was a disaster. We received one email in the afternoon of the day of submission saying it was a “reminder” about the change in date, but a quick skim through Blackboard revealed that it was in fact the first notice anybody had received. A few lucky souls had checked the submission schedule at the start of the year and spotted the alteration, but for most people that was it.

So, my main point is simple: Tell us about things! If a submission date has changed from what was originally published then everybody taking that assessment should be automatically informed. Not “Oh, well you should check the schedule weekly”. If alterations are made from what was originally published it’s your job to tell us (especially given that when I checked the schedule shortly before Christmas there was no change, making this relatively short notice) and not our job to check for revisions to your documents.

2010…

Posted on January 6th, 2010 by Nicholas Jackson

It’s been a while since I’ve bothered to blog here, so here goes with a huge update about everything that’s going on.

First of all, happy 2010! This is the year I graduate from Lincoln and have to enter the scary world outside of education, but so far it seems to have been quite successful. The snow was fun as well.

Student Reps LogoNext, my ongoing role as a Student Rep. So far we’ve succeeded in tweaking the odd little thing around the place, but we need to know what you (yes, you!) have to say. You can now spot student reps by our distinctive black and green hoodies with a gigantic reps logo (as shown to the right) on them. In the near future you’ll also be seeing new and easy ways of getting in touch with us, including clearer information on who your reps are and what we can do made available both electronically and physically.

Mobile Computing is the big unit I need to get submitted at the start of this year, and I’m using a brief respite from hammering out C# code to bring you this blog update. I can’t divulge too many details because it may affect the grading of my work (plagiarism is a bugger), but as a broad overview I’m working on an encrypted, autodiscovering chat client capable of traversing firewalled networks using a node/supernode architecture with persistent user identity for use in networks where a connection to an external server may not be available, or may be monitored. Hopefully I’ll get a decent grade, I’ve got enough supporting documentation to destroy a small forest should it be printed.

Frontier Technologies is a more… unusual unit which I’m not ashamed to say I find almost completely useless in every aspect. I need to submit my CV for grading (Yes, I’m being graded on how well I can apply for a job) and then attend a mock interview. In theory I should have been blogging religiously on Blackboard about the weekly talks, but this is an area I strongly suspect I’m going to get a rollocking for not doing properly (or for some weeks at all). I fully intend to mention at the end of unit survey about the inherent uselessness of having a private blog for my thoughts – it would have been vastly more likely to be completed if we had to submit it as a reflective journal, or even if we could have made it public to receive feedback and comments on. At the moment it’s a desolate wasteland containing nothing of worth, little more than an overview of what happened  and thoughts like “It was useful to hear about how processors are now multicored and can do parallel processing” with the unspoken sarcasm of “because I’m a computing student, and have no notion whatsoever of how many cores my processor has and how parallel processing can improve performance but only if the program is written to take advantage of it”.

Finally, the all important third year project is continuing roughly according to plan, with my initial research completed and mostly written up (No, you can’t see) and the rough framework of my code coming together (I’d forgotten how much I dislike C++ as a language). I haven’t updated my public reading list recently because I’m working on a way of automatically transcribing my reference database (BibTeX is the future) online. If I can’t get it to work soon then I’ll just copy and paste the boring old way.

That’s it for now. I’ve added a new recurring task to my ever growing list reminding me to blog about things more often, probably on Thursdays so I can summarise the week’s submissions and so on. Any questions, comments or thoughts – as always – can be left on this blog.

Dropbox Your Files

Posted on December 1st, 2009 by Nicholas Jackson

Dropbox LogoOver the past couple of months I’ve been using a piece of software called Dropbox to keep a folder in sync between various computers, which was amazingly useful. I then installed it on my work machine in the University as well, and was confused to discover it followed me around campus even without additional installation.

After a bit of enterprising prodding I discovered that the Dropbox client actually doesn’t require installation on a per-machine basis at all, and with a bit of careful hacking you can ‘install’ it in your user account and have it follow you around the University. Yes, you can have a single folder kept in sync between home (Windows, OS X or Linux) and any University machine you use which even has a cool web interface for those times you’re somewhere else entirely. Here’s how – all you need is around 30MB free on your H:\ drive for the installation.

  1. Get yourself a Dropbox account. It’s free for up to 2GB of storage space, and this is expandable (to 50GB or 100GB) for a small monthly fee. Use the handy referral link provided to get yourself a complimentary 250MB of storage on top of the free 2GB.
  2. Install the Dropbox client on your home machine. Configure to your liking and make sure you’re syncing properly. If you want you can do this bit after you’ve set up the University side of things, but you need to do it at some point or the whole thing’s a bit pointless.
  3. Head into University and log in to any corporate desktop machine (any of them in the Library, L&LR centres or Computer Lab A).
  4. Download this handy zip file and unzip it to your H:\ drive. It’s a 16MB download, but well worth it.
  5. Click Start, go to Programs, right click on Startup and select Explore.
  6. Move the Dropbox shortcut from your H:\ drive into the Startup folder.
  7. Log out, and log back in.
  8. Dropbox will appear as if by magic, and ask you to log in and set up the PC. Make sure you put your Dropbox folder in a subfolder on your H:\ drive (the default, “H:\My Dropbox”, is fine), or nothing will work properly and you’ll spend ages trying to work out why. You only do this once for the whole University, as your settings follow you around.
  9. You’re done.

See? Easy data transfer. You can also add additional Dropbox clients with no worries, handy if you have a desktop and a laptop or a dual-boot machine. Even better, it features shared folders which automatically sync across multiple users (ideal for group work), has an iPhone client (great for getting to that document without having to get to a desktop) and lets you undelete files for up to 30 days in case you accidentally nuke your dissertation.

The only thing to watch out for is your disk quota on your H:\ drive as if you run out of space Dropbox won’t be able to sync, but other than that this solution is not breaking any rules, falls within the AUP for the University network and is generally really useful.

If you want, check in your Dropbox account to get your own referral code and collect 250MB of free space for everyone you invite. Feel free to send them here (http://j.mp/uldropbox) and tell them to start at step 2 for easy installation guide and the ready-to-go download.

Real-Time 3D Modelling

Posted on November 27th, 2009 by Nicholas Jackson

The guys down at Cambridge have been doing some cool stuff with 3D modelling. More specifically, modelling an object in 3D in real-time using nothing more than a commercial webcam. Take a look at their blog entry, or check out the video:

3D realtime modelling has a lot of uses in just about anything, but particularly for computer vision and situation awareness in robots. At the moment a wide range of sensors are used to map environments in 3D including sonar, laser and 3D cameras. What’s interesting about this particular method is that it interprets 3D in the same way that you do if you close one eye – by using fixed reference points in the image, looking at how they move and then building the 3D model from that.