Happy Stories

From Dev8D

Jump to: navigation, search

Dev8D will be a chaotic event, and it's easy for us organisers to fail to hear about some of the innovative things which happen.

While Dev8D is a lot of fun, there's a really important point to it, which is to benefit the developer community and the communities they support. We know that given 220+ peers, coffee, beer and working wifi a developer can't help but create useful new things, but we want to know the details!

There's some good solid sensible reasons for us asking this question, but there's also the fact that we want to know because it's cool, and we're excited to hear what you people come up with!

Contents

Dev8D 2010

If possible, give URLS, names and say how the idea/code/gadget could be of benefit to others, especially the education sector. We may well get in touch to write a blog entry or similar about your contribution!

Code

Developer Bingo

There's a game we'll be running on the first evening. I'll release the code used to generate the game sheets after the game's over. Cgutteridge

Programme & Wiki

The RDF programme is a first attempt to provide useful data to support conference delegates, as oppose to the current ontologies which only really tell you what happened, and are not so concerned with what happens before and during. (planning and navigation of delegates!). Davechallis developed the twitter emedding and logging-in extensions especially for the Dev8D wiki, although he's planning to release the code. Cgutteridge

dev8Dfeed twitter account

Not code, but may be useful for other events. The dev8Dfeed account tweets blog, video and photos which mention dev8d. see this brief blog post for more

  1. First collect a few rss/atom feeds: Google Blog Search, Google Video Search, Flickr tag results for dev8d. It seems sites like Flickr and vimeo don't provide feeds for search results (i.e. they provide feeds for a user, and maybe a tag, but not search), which is annoying.
  2. Create twitter account.
  3. Use twitterfeed.com, add each feed one at a time and point the output to the twitter account
  4.  ???
  5. Profit!

Curses animation of Richard Jones' antmaze

Media:CursesAntmaze.txt

Takes the output mazes/ants generated in the Genetic Algorithms and Emergence session, and animates it in a terminal using python's curses library. Dave Challis

Wayback/Memento Animation

Watch JISC evolve

  • type=wb|rss|mem
  • dir=f|b
  • delay=n (seconds)

Source: http://lemur.ecs.soton.ac.uk/~cjg/Graphite/examples/anim.php?url=http://www.jisc.ac.uk/&type=wb

Author: Cgutteridge 09:48, 28 February 2010 (UTC)

  • I've added an Atom feed and set the wiki to allow (almost) unlimited export of the history as ATOM:

TechTales

Created as an entry to the Internet Archive API/Data Challenge.

We used the Internet Archive API to find out the dates it has archived versions of a site for, and then downloaded each archived version of that site.

We then run a series of plugins against each version of a site downloaded, measuring things such as tag usage, colour usage, accessibility, number of javascript/css files, etc.

This data was then visualised using Google's Chart Tools, to view things like:

  • How has site A's accessibility changed over time?
  • How has site A's usage of images compared to site B's over time?
  • How has site A's usage of div tags compared to table tags changed over time?

Dave Challis and Mike Jewell

PDF Structure and Syntactic Analysis for Metadata Extraction and Tagging

Originally created as an entry for an ePrints bounty for metadata extraction, this mini-project consists of Python code for analysing structure and syntax of PDF contents in order to derive information about the document e.g.:

  • headings
    • title
    • author
    • chapter / section headings
  • references
    • title
    • volume
    • page numbers
    • cited publications and URLs
  • suggested social tags

Headings are identified by looking for text that deviates from the norm in terms of size, colour, or weight (bold).

References are identified by looking for patterns in the "References" section. Elements making up each reference are also identified by pattern matching.

Titles, Authors, Headings, References and the component elements are tagged with quick and dirty XML tags.

Facial Recognition and OpenCV walkthrough and code Ben O'Steen

Blog post on how to do facial recognition (detection not discrimination) in 60 lines of python: http://benosteen.wordpress.com/2010/03/03/face-recognition-much-easier-than-expected/

Github repository of code: http://github.com/benosteen/FaceRecognition

Turning a Blog into a Book Ben O'Steen

Blog post walkthrough of all the steps needed to turn a blog into a printed book: http://benosteen.wordpress.com/2010/03/02/blog-to-book-experiment/

Ideas

You heard it here, first, folks!

  • I'm going to try and write some data.gov.uk demos using Graphite
  • John Harrison had an idea for some semantic analysis of PDF docs for an ePrints bounty. This idea has evolved into code!
  • Duncan Davidson Trying out few thoughts with Yahoo Pipes:
    • To do repeatable Data Validation/Cleansing/Completion using TechCrunch's CrunchBase API and Companies House XML Gateway.
    • Using the Unlock Places API to do simple/lite GIS reporting for embedding in webpages.
  • a quick hack to find JISC events, real ale pubs or beer festivals that are on at the same time - Owen Smith
  • Repurpose the code from the nanoprojector people, but make a macroprojector version which uses a laser pointer for a mouse-- just use a filter so a calibrated cam can just see the laser point and move a mouse based on it.
  • Ross McFarlane I'm working on a Memento/Web Archive project that will retrieve historical versions of a current page from a time that means something to you, like "The day Michael Jackson died" or "Charles and Camilla's Wedding".
  • Mark MacGillivray and Richard Jones: after discovering Memento from Rob Sanderson, we got to thinking how this might be useful for us and the way that we use the web. We read a number of sites regularly, so we've started working on Pulse, a bookmarking service which analyses data change over time, and recommends when to check back on your favourite bookmarks.
  • I've had an idea about exposing a RESTful API for MS-Zentity using ASP.Net MVC which lead to providing programmatical access to the scholarly works data from any platform. The API could be easily extended to accommodate any content namesspace that is plugged into a Zentity instance. - good times! Martin Evans
  • Michael Aherne has been invited to the University of Kent to talk to Steve Coppin and his colleagues regarding their move from WebCT to Moodle last summer.
  • Steve Coppin shared his Agile blog (http://blogs.kent.ac.uk/agile), which he promises to start positing to again, with a few people and started planning an AgileFest at the University of Kent with Ben Charlton
  • chris_s_yates - We've been discussing the idea at the OU of being able 'Tweet this item' in the repository (thanks @smithcolin). Tagging the repository and item so we can track what people are saying about our research. I like the idea of being able to use Twapperkeeper (thanks to @jobrieniii for this great service) to save the tweets and pull the data back into individual items maybe to give an idea of impact/potential collaboration? All thoughts welcome!
  • Stephen Vickers has completed the implementation of a Basic LTI connector written at dev8D for WebPA (a JISC-funded project and recipient of a Bronze 'Learning Impact' Award from IMS in 2008). Anyone with a Basic LTI tool consumer wishing to connect to a test WebPA tool producer, or just wanting further information should contact Stephen at Stephen.Vickers@ed.ac.uk. Stephen will be contacting the authors of WebPA to discuss refinements to this connector and to have the functionality included in a future release of the software.

Eco Game

  • External wiki set up http://ecogameos.eco-sapien.co.uk/wiki/ (new temporary location!)very bare at the moment. Please start to contribute and return frequently for updates. Thanks for the updates so far.
  • Two lightning talks to push the concept by the end of the conference, Friday at 15:30 and Saturday at 14:00
    • these are supposed to be interactive to an extent - and on the fly presentations of what has just happened in the last few days.
  • pictures uploaded to picasa: pictures of the flip charts from the bar

update 02-03-10

Note the new location for the wiki (above). lots of interest behind the scenes, really looking forward to more full understanding this project and conveying to you all why you'll be excited to participate... more soon here and on the game development space. Please contact me about this using my name at gmail' mailing facility, .com in case you didn't know.

thank you, Audaye

3D Scanning + RepRap 3D Printer = 3D Photocopier

John Harrison

  • Backup your own door keys
  • Immortalize yourself, family and friends in a life-size accurate bust without paying an expensive professional sculptor!

Ben Wheeler :-

  • Repairs:
    • Plumber says he hasn't got the right part and has to order one from Elbonia? Scan the old one, get your heating back on sooner!
    • An end to having to throw away entire appliances/systems because they no longer make parts for it
    • Had a prang in the car? No need to listen to the garage worker suck his teeth and pluck some exorbitant figure from the air. Scan what's left of your bumper and print a part that perfectly fits the gap.
  • Ikea didn't pack enough #52B gromits again...
  • No more envying your neighbour's $item! Anything they own, you can own too! Just fit the scanner with a special Paparazzi lens...
  • Makes design democratic and meritocratic - whatever is well designed will be widely (and freely) copied; poor designs will die out. In other words: natural selection. Caution advised if using Genetic Algorithm-based design software, especially as AI increasingly added to appliances, furniture etc. Heed expert advice.
  • Develop new shapes/designs in soft materials, then scan and manufacture in plastic. Stock up on Play-Doh and Fuzzy Felt!

please feel free to add your own ideas...

Gadgets

..and gizmos. Anything with a hardware component.

Many thanks to Gary Bulmer for his introduction to the Arduino stuff. I haven't had so much fun since my mum told me I was too old for lego!

A great opportunity to get back into hardware - built my first piece of hardware since... 1989??

Davetaz - Been playing with ZWave equipment all week. Home automation kit which now has an open source library (thanks to open-zwave). Got an API working on Sunday before the event, web interface on Wednesday and by Friday, an iPhone app. Thanks to @samscan for running a great iPhone workshop, my first app ever. Thanks @dev8d, can take this new knowledge back into work :)

--Scottbw 10:48, 26 February 2010 (UTC)

chris_s_yates - Thanks to Gary Bulmer for such an informative, interesting introduction to Arduino. Its given me some really interesting thoughts on potential projects, especially for server monitoring. If you've not used Arduino, definitely give it a try!

I didn't book on Dev8D expecting to do hardware, but I totally fell in love with Arduino in Gary's clear and interesting introductory workshop. Glad I decided to carry on into the afternoon, I learnt a lot but most importantly I came away with a lot of confidence that I'll be able to do Cool Stuff with it at home - that's the mark of a good teacher. Quaestor23 18:11, 1 March 2010 (UTC)

Training/Learning

Things which mean you can now do your job better

Drupal 7 and RDF

Presentation about RDF in Drupal, followed by one-to-one chat with Lin Clark, very interesting. Drupal's come on a long way since I last looked at it a few years ago. Am intending to try and do some web site experiments with this during the meeting. This could play verh well with our data curation / annotation ideas.

During the course of Dev8D, I was able to install Drupal 7 and make a start on an experimental web site for Dev8D developer outreach. I was able to spend some time with an experienbced Drupal developer who pointed out the Drupal "mindset" which I wasn't able to easily glean from the documentation. If Drupal lives up to its promise as a platform for bridging between web sites and machine-readable linked data on the web, this alone will have more than justified the time spent at Dev8D.

Developers training developers

Someone just said that Dev8D was about developers training developers -- as an alternative to paying substantial sums to someone who doesn't really understand the technical nuances to read manuals to developers? It works for me!

Met with a developer on the first day who fixed a problem I was having. Now I have a contact any further problems. Thanks Dev8D!

Developers supporting Developers

Looking at a "minor problem" in a JSR-168 Portlet when rendered b y Sakai, I took the chance to talk to Charles Severence and had a superb bug chasing session to apply fixes to sakai which I wasn't to bothered about but tanks for the great opportunity to make a project just a bit better. Thanks Dev8D!

Links to other stories from Dev8D

GNOME & GNOME acessibilty

10/10 happy that #GNOME are coming to #dev8d. Will Walker doing stuff on #a11y, Brian Cameron and Vincent Untz on release management. Watch this space for more news.

Wookie

@rgardler Very happy that @stevealee and @samscam have connected at @dev8d re #happierpipe and #wookie (I almost feel like I'm involved ;-) 7/10 we plan samscam to make the hapiness pipe widget

Basic LTI Bounty

Scott Wilson, Dan Hagon and Mark Johnson Have created some W3C widgets to run on the #wookie server (providing Basic LTI support). One provides a shared multi-user view of molecules (ported from a Google Gadget) and the other is a multi-user rich text editor using TinyMCE.


NewsLine

Adrian Mouat and Tilaye Alemu have created NewsLine, a web service that uses memento and the internet archive to search historical headlines on the BBC news website. This allows users to see how stories develop over time.

Entered in memento and Internet Archive bounties.

RepRap

@ianibbo - Completely awe-struck by the RepRap pres and demo - Deffo taking back a proposal to sheffield hallam and sheffield uni that they get together with the local hackspace group and we get one of these build - Awesome!

Alternatives to SL for interactive exhibitions

Had a great talk with Iona & co (HalfLife2 for social engagement and oral history) about platforms for interactive installations. I'd tried in the past to wire up various cultural heritage search engines into a SL building to try and generate a dynamic interactive museuem exhibition of search results. Alas SL proved a nightmare and I never made it work. Great to talk to someone who had tried to do something on similar lines and encountered similar problems. Iona left me with a whole list of alternative platforms to investigate that might fit the bill so much better. Thanks to this, a really cool UK cultural heritage visualisation tool might yet see the light of day.

VSM Meet-Up

Fingers Crossed that @scottbw, @samscam and others might get a VSM meetup group going. Be wonderful if we can make this a more regular meetup.

RepRap

Very pleased to have made contact to Adrian after failed emails. Will discuss how to best showcase it at OSS Watch Summit. Adrian does a great talk and it really attracts interest.

When I heard about Reprap for the first time it was partly responsible for renewing my faith in the potential of technology to help us get us out of the mess we're in, rather than just making it worse. Was pleased to see it in the flesh, and meet one of those responsible, ask some lingering questions, and get some reassurance that fine minds are at work on the home plastic recycling plant that it really needs to be sustainable. Quaestor23 18:17, 1 March 2010 (UTC)

list8D LTI integration

We put together the initial dev8D challenge submission in 2 stages. Steve set up a moodle instance and installed the Moodle Basic LTI plugin and spent some time tracking down bugs. Ben took the PHP Basic LTI libraries and wrote a script to extract a simple list out of the list8D JSON API based on the course information provided by LTI. This came second in the LTI challenge, which we're delighted about.

It was very easy to get a working proof of concept, and this has been useful taking the work forward as you can tear down and rebuild properly knowing the basic bits work.

Since returning from dev8D, we've submitted some more changes and bugfixes back to the Basic LTI Moodle plugin. Our Moodle integration going live in mid april 2010 using LTI with some additional changes to allow the direct proxying of content without relying on an iframe.

Learning about LTI has saved us a substantial amount of time - with perhaps 3 days work we've saved 2 weeks. In addition, the nice decoupled nature of the consumer and producer mean that we should save on maintenance time. Upgrading one is unlikely to break the other, and we've gained the ability to embed list8D in other VLEs effectively for free.


Conference & Event Tools Cgutteridge

@samscam & I (Cgutteridge) did quite a lot of fun hacks based on the RDF programme.

  • Programme to KML to show you what's on now (obviously won't work now as the events are all over!)
  • Programme+Twitter to KML to show the last tweet from each room on a map
  • Added geohashes to the programme
  • Idea of using wifi signal to triangulate position indoors at events
  • iPhone tool to view programme and tweets, next on in a augmented reality way
  • we're going to review and revise the RDF ontology for use at more conferences and festivals

Data.gov.uk Cgutteridge

I built a page showing easy examples of how to use Graphite to start playing with the government data: http://lemur.ecs.soton.ac.uk/~cjg/Graphite/gov.php Cgutteridge

Museums data Cgutteridge

@samscam and I designed a mock up of how the data could scatter photos on a map to allow people to get a feel for the collections near to them. Cgutteridge

  • Idea of using the wifi trangulator inside galleries and museums to provide augmented reality there.Cgutteridge
  • Idea of using wiki-like annotation for galleries & museums so that people can create new commentaries and threads around exhibits. Cgutteridge

Arduinos Cgutteridge

I'm totallying buying one or more of these.

  • Alarm clock which only goes off IF I'm in bed.
  • ... and always goes off if I'm in bed an hour before the next event in my google calendar
  • ... and I want to make an android button to turn my central heating on/off/up/down
  • ... and User:Davechallis Dave had the idea of making magnetic lights to stick to our helpdesk whiteboard & show nagios statuses

Linked Data Challenge User:MarkBorkum

I built a Web service that enhances Twitter Search results to include RDF triples - Shredded Tweet. The system uses regular expressions to identify and extract features of interest, including: @mentions, #hashtags and URLs:

http://pegasus.chem.soton.ac.uk/search?q=%23dev8d

http://pegasus.chem.soton.ac.uk/search.json?q=%23dev8d

Within 26 minutes of posting a link on Twitter, User:Cgutteridge built a mashup:

http://lemur.ecs.soton.ac.uk/~cjg/Graphite/examples/tweetdump.php?q=%23dev8d

Thank you dev8D organisers for putting on such a great event!

Miscellaneous

Julietteculver

  • Had looked for a TwapperKeeper API last autumn but failed to find one. Met the developer of TwapperKeeper who is going to send me an early copy of the docs for their new API so I can give feedback before it is released
  • Plan to do a session on getting started with android and iphone dev for our team at work when I get back based on what I learned from Phil Raymond's talk and Sam Easterby-Smith's workshop
  • I didn't enter the Mobile bounty challenge, but thinking about what I might do for it gave me an idea for what to do with mobile versions of our site that I don't think I'd have had if the challenge hadn't existed
  • I've been reading up more about google gadgets, wookie and W3C as a result of the event - I think these might be really relevant to our site long-term. Likewise I'm feeling very inspired to make our data available as linked data - ask me again in a few months time to see if I've done anything with all this!
  • I met one of the Mendeley developers and am hoping to get in touch to look at how we might be able to do some integration with the academic references on our site and Mendeley.
  • Spent time working on a possible website for Developer Contact with Graham Klyne
  • Am hopefully getting an introduction to the Social Innovation Camp folk about doing something similiar for HE
  • If I'd known how much public speaking I'd have ended up doing, I think I might have been too scared to come! You don't learn how to be good at such things overnight, but I think the event has been good at increasing my confidence at doing such things and felt like a very friendly environment for this.
  • Generally feel re-energised and inspired by all the great stuff that I saw!

Blogging...

Have you been blogging about dev8D? Add yours below:

P.S. Can we have transclusion back, please?









(this space deliberately left blank to demark the 2009 section)

Dev8D 2009, Longterm

If you were at the 2009 event, what happened to the things and ideas. What's out there, in the wild, thanks to Dev8D 2009?

http://splashurl.net/

This was created by me (Cgutteridge) and Tony Hirst to solve the issue of getting an audience to quickly type in a really long URL for a live demo. I've since added some cute QR Code related features.

EPrints JSON

I learned about JSON at Dev8D 2009, and within very little time I created a (hacky) export tool for EPrints. One year on and it's been cleaned up and is part of the standard release, and is being used by at least one JISC project - http://microviews.eprints.org/ - Cgutteridge

Regexp Playpen

The "playpen" I wrote for my crash-course in regular expressions is still available at: http://users.ecs.soton.ac.uk/cjg/regexp/ and is a handy tool to support teaching regular expressions in general (ask me if you want the code to hack around with) Cgutteridge

list8D

Readinglist management system born out of our dev8D 2009 developer challenge winning prototype: http://blogs.kent.ac.uk/list8d/ - See Ben Charlton for a demo/chat about this.

MicroViews

This is a really simple Javascript library that was picked up as a JISCRI project following a prototype using Cgutteridge's JSON extension to EPrints. A year later it is being used on ECS web pages, you can see it in action on ECS publication pages[1], or at http://microviews.eprints.org. Marcus Ramsden

Accessibility

Had a great chat with Dan at #dev8D who wants to make his web forms on Kent's public website more accessible. Recommend user testing, Web2Access, TechDisToolbar, NVDA and more. It's great to meet people who want to enhance their stuff.

Firefox

I've learned how to write firefox extensions :) Sam Adams

Wookie BaLTI challenge Moodle widget makes into Blackboard, WebCT and Sakai

And in Minutes! Great collaboration between Dan Hagon, Mark Johnson, Scott Walker, Chuck Severance & Steve Vickers. FTW LTI launch added to Wookie and those widgets work everywhere. Chuck will pass it around LTI etc.

Retrieved from "http://wiki.2010.dev8d.org/w/Happy_Stories"
Dev8D