Ranking Items Based on User Action

veerMe (my major) finds locations in a database that are within a certain proximity of any given route and then suggests the ‘best’ of these to the user. The trouble is that if, for example, ten points of interest are close to a journey route then how do you know which of these the user might actually want to visit, and which of these are shit and unlikely to be of much use?

The answer, in my mind, is to order all of the POI (points of interest) by a rank and then iterate through these items checking for proximity starting at the highest ranked until you have found a reasonable amount to suggest (for example’s sake, 6) or until every POI has been checked – whichever comes first. This can prevent having to go through every single POI unnecessarily (which is a welcome bonus, I’m a bit worried the strain this might put on a browser as number of POI increases). It also means that you can just suggest the POI to the user in the order you find them and its likely to be the best.

This then creates another problem that I am now attempting to solve – how to rank these POI to reflect their popularity. I do not want to burden, or indeed have to rely on, the user having to rate locations because it adds another hurdle for the user and might not actually create an accurate way to rank the POI. If you think a place is awesome but don’t actually want to add it to your journey then chances are other people might not either.

veerMe gives the user four options when presented with a POI:

• Add this to the route and finish
• Add this, show me another
• Don’t Add, show another
• Don’t add, finish

This gives the user sufficient control over their journey preparation whilst also giving me feedback that can be used to help determine a POI’s route. (If a POI is recommended and the user clicks the 3rd option is might show that the POI is not suitable, whereas if they click the 4th option maybe they just don’t want to add a diversion to their route).

That’s all well and good but it still doesn’t help me actually rank the points, to do this I am going to need to use maths. Initially the idea was to simply start with a number (10) and increase the by an arbitary number (0.1) if a suggested POI is used, and decrease by 0.1 if it is not. The thinking behind this is that the higher the number the better the POI is perceived to be. A draw back to this, however, is that if a point simply hasn’t been suggested because it is hasn’t yet been close enough to a route it will remain on a lower number regardless of how good it might be and thus never really get recommended when in competition with POI that are more likely to be within the proximity of popular routes.

The alternative to this is to work on a percentage.  The number of times it is used divided by the number of times it has been suggested – giving you a number between 0 and 1 that seems to work a bit better and be a bit more proportional. Again in this case, the higher the number, the more likely it will be suggested.

What do you think, any suggestions are more than welcome! I am currently thinking the latter option is how I will do it but I can still think of a few drawbacks. The number of times it is used will have to start at 1 instead of 0 to stop them having a rating of 0 and never being suggested, for example.

Is The Digital Redefining Death?

I’ve been reading quite a lot of McLuhan for my dissertation. I’m specifically focusing on the idea that technology can be an extension of our central nervous system outside of the body.

I’ve also been reading a little bit about Rhizome theory which seems very interesting but everytime I think I have got to a point in my critical thinking/argument in the dissertation to introduce the concept and say how it is relevant I find that it doesn’t quite fit. I do think it has potential to become a key concept in pedagogy  when educational institutions realise that for every new generation of students who have grown up with the internet and other digital technologies the current classroom environment and teaching methods get less and less practical/efficient.

Anyay, I’m currently reading my way through a blog post written by William Gibson (author of Neuromancer) about cyborgs and one sentence struck me (just to contextualise, he was saying how change was happening in the 50s and before because of television) about what is enabled people to do. They are “viewing things at a distance, viewing things that had happened in the past, watching dead men talk and hearing their words”.

Dead.

Hearing their words.

Then a thought struck me. In extending our central nervous systems across such vast amount of media are we redefining the notion of “being dead”? Did the Gutenberg Press also impact on this idea, for in an oral society a persons ideas can only live on in the idea (thought/memory) of another person, in a literate age a person’s actual thought could be communicated without being recycled by another persons thought, even in death.

If our Facebook profiles, Twitters, Youtubes, Delicious’, blogs all survive as extensions of ourselves after our death then does this mean they are independent (or mostly independent, they’re not exactly autonomous or self-generating) of other parts of the CNS? It could be argued this spread of our CNS is non-hierarchical; instead horizontal, like a rhizome.

Even before the explosion of the internet age we could leave behind moments in time as photographs, anything we might have written down, us on video, etc etc. Text messages. Answerphone. Library record, etc etc.

By the time we die, have we extended so much of ourselves into other media that death is no longer finality?

And where could the future go? I wonder how unpredictable we really are. In death could the surviving parts of our nervous system one day be analysed for behavioural patterns so that our behaviour could be predicted accurately in spite of our death?

I remember someone mentioning to me that the brain is essentially a collection of millions and millions of 1s ands 0s that dictate our action and thought -  way beyond the current processing power of today’s computer, but if Moore’s Law works forever then its only a matter of time until a computer as capable as the human brain is invented. As far as I can gather, consciousness is a whole different matter and something much, much complicated (and until a machine can ponder that autonomously, we need not worry )

This is all pretty much late night ramblings so if anyone actually reads this and has some input, I’d love to hear it.

Well science fiction.

veerMe Holding Page now Up!

I have pretty much just finished editing the holding page for my major project over at http://www.veer.me.uk/. Check it out and let me know what you think. The plan is that any blog updates concerning my major will now be hosted over at that site instead of here.

Lately I seem to be learning alot of cool new stuff, probably because of actually doing more. Which is nice. I messed around with WordPress themes and some PHP to integrate the blog into that page, it is far from perfect at the moment but will do for a holding page. The forms are done using JQuery, and it’s ajax functions. Overall, I’m quite pleased.

I just signed up to AdSense, mostly to get an idea of how it works and stuff, but also because another site of mine www.gamefoe.co.uk – an old webzine I used to run back when I was much younger/more stupid – has a hell of a lot of content on it so I was thinking of trying to build a stack of incoming links (I suppose mostly from related blogs and Wikipedia) and seeing how that might drive traffic to the site, and also see how it might affect my PageRank, and also my search ranking for various terms (and how this might effect traffic from search engines which I imagine would be the biggest driver of traffic to an archived site like Gamefoe).

I think it would be quite interesting to get an idea of how much traffic is needed to generate click throughs, and how it all translates into revenue and if over the course of the year I can make £20 to go towards my hosting costs, even better!

I just need to find the time inbetween my dissertation, finishing my portfolio site redesign (the current design is horrid, too embarassed to link to it!) – which I will hopefully find the time to do this week, and my major project. The good thing is that I know I will inevitably wind up procrastinating time away that should be spent doing my dissertation by doing these other things so instead of feeling like ‘work’ it feels like a distraction, and I’m still being productive!

Interactive Media Showcase

If anyone actually reads this blog (Hi mum!) then they may well know that I am a final year student of Interactive Media Production (BA)Hons at Bournemouth University.

As part of an attempt to flog ourselves to prospective employers and gain meaningful employment we are exhibiting our major project work at a graduation show in both London and Bournemouth. The London show is part of Free Range – Europe’s largest free graduate art and design show – which is pretty exciting and runs for a few days from the 28th of May. The Bournemouth show will be a week or so later, but more info on that when it is confirmed.

We decided on Screengrab 09 for a name and more information will go online at our new website http://screengrab09.com/ so check it out if your interested in the cutting edge of new media etc. Now to big myself up a bit: As it goes, I wrote the code for the nice little Twitter update feed in PHP (didn’t style it though). And I find myself on the grad show’s organisation committee.

If you are Twitter-savvy then follow us here.

Currently…

This week I moved all my various online hosted stuff from my web-mania hosting (which ran out on Friday) over to some brand-spanking new hosting that I bought last weekend on Lunarpages. So far I am enjoying it. Multiple databases mean I can keep various sites seperate and better organised, and allowing multiple domains pointing to various folders within my sites is handy.

Also, have been in Uni more or less Monday – Friday from 10am – 5pm (and later!) for the last few weeks trying to make a dent in my dissertation and other week. It turns out I am shit hot at procrastinating, but finding that sidetracking myself by reading more isn’t really too bad is it? I seem to have the core arguements for my planned chapters sorted and just need to get down to knuckling down and writing the damn thing.

I am keen to get involved with Google Analytics to start familiarising myself with the idea of seeing how my users actually use my site, and where they go, where they come from etc. I don’t really have much practical use for the data at the moment but who knows, eh?

Finally, decided to play around with some CSS frameworks (well actually, just one framework – namely 960). Might be good for quickly banging out some rough designs, and just enjoying learning something new really. Check it out, seems quite useful – http://960.gs/

Hm, one more thing actually – any one able to recommend some good web development software (along the lines of Coda for the Mac) for the PC? At the moment I’m trying out KomodoEdit but although its really handy it isn’t quite the full package as far as I’m concerned. Not really massively keen on staying using something as bloated as Dreamweaver to be honest.

Sayonara, motherfuckers.

Google Maps

So I have been trying to familiarise myself with Google Maps the only way I know how – in small little steps. Mostly using one or two tutorials, the documentation made available by google, and a bit of intuition

So, first of is: Map One – This was just learning how to put the map on the page, playing with adding the controls, centering the map on locations etc. Then I thought I would just copy/paste the code to add ten random markers because, well, it looked pretty cool!

Map Two – For this I used a tutorial – It pulls locations out of a MySQL database (via lat, and long coordinates, as well as a title attribute) using PHP, then creates a simple PHP loop that echoes the code that creates the different markers for as many rows there are in the table.

That was useful but this site suggests that outputting the data into XML to be read into the page will have the following benefits: faster initial page load, easier debugging, and a more flexible map app. So I set about figuring out how to output MySQL data as an XML file: ba-da-bing. There appears to be a variety of methods to be able to do this but I set up a query to my database that echoed the results into an XML format. Maybe I could look into a function called domxml_new_doc() in the future?

Now, map three is using the same information as map two, but it is pulling it from the above-linked XML file instead of connecting directly to my database.

Progress. Slightly. Hurrah. Next up, geocoding places using markers and saving them to the database for XML use. Then I can get onto the real fun of playing with the routeplanner function of google maps, probably using this as some kind of guide initially.

Halloween

My major project will require me to work with the Googlemaps API and from conducting a bit of research I have deduced that having a slight understanding of Javascript will help enormously. So today I have been playing about with Javascript – it hasn’t really been too challenging so far but then I have only really scratched the very basics of what is possible and having experience in other languages such as Actionscript 2/3 and PHP obviously helps…

Anyway I made this page that uses javascript to tell you whether or not it is Halloween yet – and if it isn’t then it tells you how long it is until All Hallows Eve in months and days.

Check it out here: http://www.gamefoe.co.uk/sandbox/halloween.html

Also, in an effort to stay more organised and not lose notes/references related to my dissertation (and I guess anything else organisation wise) I made a massive pin board for the wall near my desk; two big bits of corrugated cardboard, cheap double sided foam pads from Wilkinsons, and a pack of push pins (also Wilko) to stick stuff up.

Image:

Cardboard Noticeboard

I just have to start using it now, I guess.

Ah, Timing.

I haven’t blogged in ages. Absolutely ages. I tried to address the other day but only managed to write less than a quarter of what I wanted before saving that blog post to my drafts and forgetting about it. I’ll probably get it finished soon.

Such is the joy of laptops I am writing this from in bed with the spectacularly unjoyful and unhelpful case of food poisoning I seem to have managed to acquire. I don’t tend to get ’sick’ very often so tend to assume my immune system is usually quite efficient but I think it may have taken a bit of a battering in the last few weeks and the final straw must have came in the form of some dodgily cooked meat that beat whatever was left of my good health into submission. Nuts.

Whilst this is a combination of bad luck and my own fault (both of which I can’t really curse) I am still pretty fucking gutted at the timing of this – either Wednesday or Thursday is set to be presentation day for this final unit of the year and not being really able to eat/get out of bed/be more than 2 minutes from a loo severely disables me from preparing adequately for it. With any luck (and the aid of Actimel and bananas and sleep and water) I should be OK enough to actually present. Though I wonder how my peer assessment is going to suffer from my not participating in the final vital few days leading up to the presentation.

Gutted, been doing pretty well up to this point, too.

An Almost Group Wide Project Problem?

I logged into my site’s own log in area and if I don’t log out or close the browser before going onto some other students’ websites I am already logged in! Without even registering, brilliant. I imagine maybe it has something to do with the virtual hosting, or the fact we are using the same table names, on the same server, or sessions not ending? Maybe there’s a php command that kills the session and gets rid of the data as you leave the domain or something?

I wonder if anyone has restricted parts of their site, such as an administrative section, dependant on user’s “access” level and if that still carries over.

If you’ve never read a newspaper, and you want to be all you can be. Join the army.

I got the new Gay For Johnny Depp album ‘The Politics of Cruelty’ at the weekend. Absolutely stunning. What’s more there’s a very strong chance I’ll be putting them on in Bournemouth in the new year at iBar. Brilliant!

Wednesday is the birth of a new club night. Robot.Robot.Engage! is the name we’ve given to a new night run by our fledgling promotions ‘thing’ in Bournemouth. Sam from Get Cape. Wear Cape. Fly is DJ’ing and apparently that scared Fat Poppadaddys enough for them to try and poach both Sam and the whole night. Scumbags, how can the Uni actively endorse practices like this whilst simultaneously hindering what we are trying to do by not allowing any promotion of other venues on the Campus or in the student magazine (who have been very supportive to us).

Anyway, I came across a problem in m project the other day. I want to let users upload images to their own cache and then have the browser display them, thus adding greater interactivity and a reason to register to the site (only members can save, or publish, what they upload). Oh, it’s an online comic book.

Although apparently PHP_SELF can be used for sending images to itself it can only display the various attributes of the image and not the image itself (not without saving the image to the site first, which I don’t want to do). At least that’s the impression I get from Matt and The Internet. Javascript doesn’t seem to be able to do it, either. All the research I did seemed to suggest Javascript can be dynamic but not quite as dynamic as I wanted.

I ‘ve been pointed in the direction of something called AJAX:

Asynchronous JavaScript and XML (AJAX) is not a technology in itself, but is a term that describes a “new” approach to using a number of existing technologies together, including: HTML or XHTML, Cascading Style Sheets, JavaScript, The Document Object Model, XML, XSLT, and the XMLHttpRequest object. When these technologies are combined in the AJAX model, web applications are able to make quick, incremental updates to the user interface without reloading the entire browser page. This makes the application faster and more responsive to user actions.

Hopefully I can find what I want to do rather quickly because I don’t want to spend too long on this because although it was initially one of the important functions of the site I realised I could easily survive without it, people will just have to register to play on the create.php page.

Time clearly isn’t on my side.