Warm Coffee – Reflections of a Web Developer

I’ve just read an interesting piece in the Guardian, concerning post-publication editing of articles, and how the Web has divided opinion on whether print versions of articles should match their online equivalents. Obviously, the nature of the Web allows for minor corrections (typo’s, spelling mistakes, etc) but should journalists have the power to change, or add to their articles once they have entered the public domain, and if so, should any ‘mending’ be made obvious? Does this affect trust and transparency issues?

This piece is obviously angling towards journalism, but I can see parallels with community edited documents, or websites (such as Wikipedia) where savvy users can work out what has changed, but most won’t, and taking a quote from an online source which changes, or at worst disappears can result in issues of integrity in all fields of work (from school homework to journalism and politics). How does this work with the Freedom of Information act, when an article or page on the Web could be changed on a whim, and no record of the original kept? Is this tantamount to shredding inaccurate documents and replacing them? In this current era of the ephemeral and ubiquitous Web, who owns content, and what they can do with it is an interesting question.

Tags: content publishing ownership online

No Comments »

Day 3

This morning’s hangover was brought to you by Smithwicks.

A short day today, with the conference finishing at around 1pm. Before the end, though, a couple of talks to go to before the final keynote.

Firstly, was a session on ‘Data portability with SIOC and FOAF‘ from Uldis Bojãrs of DERI, Galway. This talk described the need and requirements for data portability, and continued with a demonstration of SIOC applications which allow SIOC and FOAF data to be produced from existing online community sites (blogs, forums, etc.). Uldis also described how this information can be ported to other sites with tools such as the WordPress SIOC Import plugin. He also demonstrated the Semantic Radar Firefox plugin which detects the presence of SIOC, FOAF and DOAP data in a web page.

Next up, was Andrew Walkingshaw of the University of Cambridge, who was presenting his work on ‘Representing, indexing and mining scientific data using XML and RDF: Golem and CrystalEye’. I’d met Andrew over lunch the day before, and we had a good chat about many things, and I was really looking forward to this talk. The essence of the presentation was discussing the creation of tools to enable scientists to filter, mine and search both their own data and that produced by other researchers, aggregating supplementary data, and adding value to this data. Andrew described the CrystalEye repository and his Golem ontology language and associated tools, which provides the ability to extract the data from the repository, and then build new interfaces to it – making the data easier to find, analyse and reuse. A cracking talk, delivered in a very engaging way, and I will read his paper as I admit some of the talk did pass over my head!

Finally, after some much needed coffee, was the final keynote. Sean McGrath of Propylon gave a highly entertaining, and engaging talk on the nature of the web as we know it, taking us on a “Celtic-tinted safari of the Web featuring mythical creatures, tenuous analogies and curious interconnections.” No, really, he did. Via a lot of digression (and indeed, orangutans) Sean concluded that today’s Web is: “The Web is URIs + HTTP (and nothing else – no ‘pure’ content)”. And I tend to agree. Definitely recommend scanning through the slides for this talk – it might not make sense without the words, but there’s enough to go on to get a flavour of this great end to XTech 2008.

Conclusions

So we’d made it to day 3 of this excellent conference. As usual, much of the really interesting discussion was held in the restaurants and pubs around Dublin after the main day’s events were over, going on well in to the small hours in some cases.

My general conclusions and reflections, are:

• Much of the really cool and exciting stuff covered in this conference will never see the light of day in the projects I work on (with the possible exception of RDFa, but convincing clients they should do this will be hard, as will coming up with any kind of useful ontology). This may be a defeatist attitude, and perhaps there will be opportunities to embed Fire Eagle functionality into some sites in the future, or make more of FOAF and SIOC to make our sites more amenable to community building – who knows? But at least now I know these things exist, and we can start to look at OAuth for CMS integration with, say, Flickr or other such technical liaisons.
• I don’t know anywhere enough about this stuff, both technically or indeed that it exists, and despite it not being of immediate use to me in my day-to-day work, I should spend more time finding out about it and playing – maybe in my own time if work commitments don’t allow the time.
• What I really liked about this conference was the mix of attendees and presenters, both from academia, and the commercial world both large and small. It made it feel much more valid, and it really felt like everyone was there for the right reasons – not trying to sell anything, but out of a genuinely altruistic wish to make the web better.
• I definitely would want to go to XTech again, and would encourage others to go too.
• Dublin pubs sell some great beer.

Tags: xtech2008 conference report

1 Comment »

Day 2

After learning the lesson from yesterday’s ‘worst Full Irish Breakfast I’ve ever had in Ireland’, I headed off to the conference via Starbucks – this time my route problems of the previous day were banished, and I actually arrived early. Not early enough for the 8am sponsor presentation, though, so I hope somebody made it in for that.

The first two presentations of the day were by Stephen Dunn and colleagues of the Guardian, and Brendan Quinn from the BBC. Both presentations concentrated on their recent ‘updates’, both approaching this upscaling and development in different ways – the Guardian talk was more ‘what we did’ whereas the BBC is more at the stage of ‘what we are doing’. The Guardian talk concentrated on: using the web for services to avoid internal complexity in their software; implementing web 2.0 ideas at an enterprise scale without the enterprise baggage; using tagging to connect content and services; How we prepared our archive and ongoing information architecture for the social web. There were several interesting concepts discussed including the use of third-parties to provide data and also cut down on cacheing issues, and also an automated tagging system based (in part) on folder structure. It was interesting to note that these large scale projects rely on many, many techies, editors and writers, and it was ‘fun’ to contrast that with how we work in ILRT (albeit on sites several orders of magnitude smaller). The BBC talk was even closer to home, with resources being their prime issue – having to cojole and argue to many people the virtues of introducing a new web delivery platform if the BBC wanted their site to continue, whereas from the outside all appeared to be going smoothly – the swan metaphor was used, I’m sure. The mutterings of surprise around the room when it was announced that the BBC site still ran on SSIs and Perl scripts was oddly noticeable.

Coffee, chat, and then on to listen to Kellan Elliot-McCrea from Flickr talk about OAuth. This was interesting, as I knew nothing about it beforehand (which was probably a bad thing, as this talk relied on knowing what OAuth was!) but hey, I can pick these things up fairly quickly. Anyway, so OAuth turns out to be an emerging standard for authorisation, not authentication (ie OpenID). Much like it’s authentication equivalent, OAuth works by the exchange of tokens, but also adds the ability to customise attributes for additional privileges. I must admit that this started to go over my head a bit, so more reading is required, but this definitely does look something worth more investigation. One of the sites making use of OAuth is Fire Eagle, and the next talk was by Evan Henshaw-Plath of Yahoo! Brickhouse, on Fire Eagle. This site is interesting in that it’s been designed primarily as an API, and several demo’s of how it has already been integrated with other web sites and tools were given. I now have an account on Fire Eagle, so will be trying some interactions with it soon, and hopefully coming up with some use for it!

After lunch and some sun outside the hotel, it was back to the dark halls to listen to Jeni Tennison of The Stationery Office talk about SemWebbing the London Gazette. I found this really interesting, and was my first real introduction to RDFa. Jeni also covered the underlying problems, in that documents need proper URIs, and how they redefined the URI space for the London Gazette. This talk opened up many possibilities in terms of making data better available via RDFa, without relying on dodgy HTML scraping scripts. Defintely a paper worth reading, and food for thought.

Next up was Ralph Meijer, talking about getting Social Networks to talk to each other. He described how, using Jabber/XMPP technologies besides HTTP, you can enable two-way communication between third-party clients and services (XMPP as an API), and have services exchange (events on) social objects and people. In near real-time, with built-in authorization and authentication. Very cool stuff, and worth keeping an eye on – there seems to be quite a movement to free up information held on individual social networks to get out of the closed-silo mentality of recent times.

On to the Steven Pemberton experience. I always enjoy seeing Stephen present, and this was no exception. His talk examined how Web 2.0 partitions the Web into a number of topical sub-Webs, and locks you in, thereby reducing the value of the network as a whole (the value of the Web is that there is only one). He went on to explain that user contributed content is not (always) a Bad Thing but it is the method of delivery and storage that is wrong. The future lies in better aggregators. This (somewhat) controversial approach suggested that we should hold all our personal data (photos, personal profile, contacts list, etc) on our own personal web servers, which can then be uploaded or shared with third party sites (Facebook, Flickr) etc, so I can have my information on any social network site I like, but only need to update/manage it in one place – my personal site. Still undecided on this approach, but then I’ve still to find a need for Facebook, and I do host my own photo archive, but as most people were trying to say, I’ve yet to find a better photo site than Flickr.

Finally today, was Ian Davis from Talis, presenting a paper on how to manage your API to keep everyone happy. Ian covered the design of API URLs, versioning strategies and techniques for preserving backwards compatibility – most of which seemed quite obvious, but how many of us have gone ahead and designed or updated APIs with little or no thought for who it might effect when we go from v2 to v3.

Lightning Talks

Later in the evening, everyone gathered once again for the highly anticipated Lightning Talks session. These were generally very good, and subjects covered included HTML5, Microsoft Popfly, and a break from the normal 20:20 format by Ian Forrester from the BBC, who presented a mammoth 72 slides in 6 minutes… it wasn’t wholly successful in getting across anything tangible, but it was funny! Another talk focussed on work for the new virtual, personalised tour of Amsterdam’s Rijkmuseum, which was a great talk, and very well presented. I really like lightning talks – you can learn so much from them, and in some cases I feel they are far superior to the usual 45 minute presentation.

Talks I wish I’d been able to go to:

Most of them, to be honest – it was a really good day, with some excellent talks.

Tags: xtech2008 conference report

1 Comment »

Intro

This is a brief write up of Day 1 of the 2008 XTech conference held in Dublin, Ireland from 6-9 May 2008. Days 2 and 3 to follow. The proceedings and presentations from the conference are available for further reading.

The theme for this year’s conference was “The Web on the Move”. Not being able to put the intro any better, this is the intro from the parent website and the conference programme:

“For years we have been developing and promoting open data standards, enabling data portability. Recent developments have led to web-wide programming APIs and virtualization. It’s no longer just our data on the move, it’s our applications and even our servers too.

What impact will this era of unprecedented portability have on us? How should we change the way we build for the web? XTech 2008 will examine the technology, war stories and practical concerns of developing for today’s web.”

Day 1

The conference opening keynote was already started when I arrived (being slightly late, hot and flustered after taking the long way round to get to the venue due to a Google map print-out of restricted detail!) so I missed the introduction, but it turned out to be a presentation on the role of “open” in the process of innovation to commoditisation. I must admit to missing the point slightly, but this talk seemed to summarise (in many, many, slides) that the ‘open meme’ will be a major driving force in behind the changes in the market in terms of getting from innovation to commoditisation, by changing the market and the services that will need to be provided. This talk was followed by David Recordon of Six-Apart, who presented a very interesting journey through the technologies behind Open Platforms (concentrating on social applications), and their common requirements, ie: ways to share abstract information, ways to communicate, ways to know who someone is, ways to know who someone knows, ways to know what someone is doing. This involved discussions of OAuth, XMPP, OpenID, XFN Microformats (XHTML Friends Network), FOAF, Twitter, and others. Sadly, many of these I was oblivious to, but no more – I have, as a result now got accounts on Twitter and Fire Eagle, and am experimenting further with Google Talk and its XMPP client.

After a welcome coffee break, next I sat through an enlightening talk from Douglas Crockford, currently of Yahoo! on “Javascript: The Good Parts” where he attempted to show that Javascript is “a beautiful, highly expressive language that is buried under a steaming pile of good intentions and blunders”. He also demonstrated JSLint a Javascript verifier, which should (will?) save you from all the pitfalls and nightmares of badly coded Javascript. Coincidentally, Douglas has a book being published by O’Reilly on this very subject coming out shortly.

Then it was on to a talk by Michael Smith of the W3C who gave a rapid look at some of the important changes in the browser landscape since XTech 2007, and at what those changes mean for developers. There was a lot of information on how Webkit is becoming more prevalent, and how there has been much development in the mobile browser arena, with Opera still a major player in the absence of any significant challenge from Mozilla (so far – they are now working on a Gecko based mobile browser). Otherwise, it was some CSS3, better SVG and much surprise that the IE8 beta seems to be better than people were expecting.

I then went to a couple of presentations on AJAX, the first by Bob Buffone of Nexaweb Technologies who talked about ways to optimise AJAX applications. Areas covered were: Mozilla’s Rhino JavaScript engine as a complete performance monitoring tool for Ajax code bases; Injecting monitoring code into every function of JavaScript within an application to create a complete performance picture; Locating performance issues through drilldowns of function call counts, total time spent, average time per call, and call stacks; and start time optimization using Dojo, Gzip, and Compression. Most of this was fairly obvious, but a useful summary, and was also introduced to YSlow, a Firefox plugin for measuring page download/rendering times, which I’d not come across before. The second talk was by Clinton Smullen of the University of Tennessee who presented a paper on how they tried to increase the performance of part of their University website by using AJAX. He described the tests they used to determine the best way of delivering the AJAX content (ie partial HTML, XML, JSON, and CSV). Interesting stuff, but I felt fairly limited to that particular application, but definitely worth bearing in mind that there may be better ways to serve your content than the obvious, and that you should look at other methods where possible.

After coffee, I went to listen to Arve Bervendsen of Opera give a talk on giving Web Applications and Widgets access to device and user data. Not knowing really what widgets were in this sense, I thought this would be an interesting learning session for me, but I have to say I’m still none the wiser, and so this probably wasn’t a great talk to go to as it did presuppose some knowledge on the subject. Still, it appears you can do what he said you could, and had some demos. Mostly I just caught up on email.

Finally today, I went to listen to Fabrice Desre of Orange Labs talk about Open Mashups. The Open Mashups editor is a Firefox extension, which aims to allow the creation of mashups which provide: clean separation of functional (what the application is doing) and non-functionnal (e.g look and feel) aspect; independence from devices and execution platforms: no vendor, device or platform locking; and user friendliness : no need to be a coder to use it. This looked very cool, and I do intend to have a play when time allows. Currently reliant on FF3 though.

Talks I wished I’d been able to go to:

Blaine Cook – Building the Real-Time Web
Simon Willison – Unobtrusive JavaScript with jQuery

Tags: xtech2008 conference report

1 Comment »

I recently attended a Web Application Security workshop run by SANS in London.

Dominic Hiles has made extensive notes on this at:

• Day 1
• Day 2

No Comments »

I received an email the other day from a colleague asking the following:

“I was surfing for info on Unique Users and came across this – I’d really appreciate your views on it if you’ve got time to have a look. Do you know anything about this company and would it be worth getting involved with them?”

http://www.thinkmetrics.com/New-methods-of-web-analysis.ph

I read the article, and suggest you do to. Hmmm.. sounds like a miracle cure, doesn’t it?

After four years supporting the WebTrends web log analysis software in the 1990′s, I’m used to the limitations of log and cookie based analysis, and there are several assumptions in the method described in this article that I would take issue with…

“1. If the same cookie is present on multiple visits, it?s the same person.”

Not true – it’s the same browser / computer, but not necessarily the same person. Take a public library computer: hundreds of people could use that – it’s most likely that on a popular site with no authentication, that a cookie is set once, which would then be used by anyone using that computer.

“4. We know some IP addresses cannot be shared by one person. These are the ones that would require a person to move faster than possible. If we have one IP address in New York, then one in Tokyo 60 minutes later, we know it can?t be the same person because you can?t get from New York to Tokyo in one hour. ”

Not true – IP addresses can be set by a remote data centre / proxy depending on your ISP. The user doesn’t physically have to be in those places, it just depends on how their traffic is routed. The ‘AOL problem’ (many IP addresses designated during any one session) applies here too. Whoever wrote this, doesn’t seem to understand network topography, which I would suggest is fairly useful when talking about Internet user measurements.

Its also states that “30-50% of people delete cookies”…. and then proposes a whole system based on persistent cookies. It’s no good using a range if the range drops by 30-50% in the first place, even if you weight this. Again, a flawed concept.

But I think the most telling ‘wouldn’t touch with a bargepole’ comment in the article would be:

“We’ve approached some household names in metrics, online advertising delivery, and major search engines. The response has been zilch. No one wants to know.”

If online ad agencies and other metrics companies aren’t interested, you must be flogging a dead horse.

To be honest, if you read this without focussing on the techie bits it’s a simple sales pitch for a dodgy scheme.

So the summary answer I gave to the emailed query was “I’ve not heard of this company, I’d not heard of this ‘research’, and I would suggest our time could possibly be better spent.”

On a similar note, an article on the BBC site – Web counting tools ‘need change’ – makes interesting reading, although again, offers no real solutions. I’d be very interested to know if anyone has come across any other methods of measuring users on Web sites.

1 Comment »

This is a neat trick which allows you to create ‘active’ bookmarks which provide functionality without having to initially visit the site in question first. By creating a bookmark with an associated keyword, you can run searches on target sites from the Firefox address bar.

Example: The Oxford English Dictionary

To get the relevant query string, go to www.oed.com, and run a search. You will then be presented with the completed query string for running a search on this site, eg:

http://dictionary.oed.com/cgi/findword?query_type=word&queryword=test&find.x=16&find.y=22&find=Find+word

We want to make this dynamic, so need to replace the test string with a string variable. In this case, the relevant form element is ‘queryword’, so we can take this URL with its query string, and create a bookmark from it.

In Firefox, create a new bookmark – call it OED Search, or whatever you like. In the location box, add the following:

http://dictionary.oed.com/cgi/findword?query_type=word&queryword=%s

You can strip the rest of the query string, this is just cruft. Add a keyword, say ‘oed’ (this should be unique, so make it fairly obvious yet distinct). Save your bookmark.

Now, in your Firefox address bar, you can type something like ‘oed floccinaucinihilipilification’ and your browser should then integrate this into your bookmark location string, run the search, and return you a page of results as normal.

Obviously this can be done with any site which runs a search through a query string. Some sites obfuscate the search page in the results, so you will need to take the relevant form values from the initial form, rather than a completed query string, but the principle is the same.

No Comments »

I spent this afternoon attending an introductory workshop on MPEG-21 DIDL, at the Watershed in Bristol, organised by UKOLN. The presenter was Frances Knudson of the Los Alamos National Laboratory Research Library. For the uninitiated, MPEG21 is a standard which “defines an open framework for describing multimedia applications” – although this can really mean anything!

The programme consisted of a whistle-stop introduction and overview of MPEG 21 DIDL, followed by some more hands-on theory where we were asked to create creating DIDs for various datatypes, based on what we had learned in the earlier session.

MPEG21 has 18 components, and this short workshop only really looked at parts 2 and 3, (Digital Item Declaration and Digital Item Identification respectively). Parts 5, 6 and 7 (Rights Expression Language, Rights Data Dictionary, Digital Item Adaptation) were mentioned in passing, but not explained in any depth.

I can see how this stuff would be useful – the model is deliberately abstract, and so can be implemented in a number of ways (XML, RDF, etc). The logic in the packaging of items, components and resources is fairly easy to grasp, but in the true way of metadata, you can pretty much do what you like, which is fine for individual repositories, but this means that true interoperability is still a way off (although the RAMLET project may assist with some of this).

MPEG21 currently has only one real competitor, which is METS, although most of the current major repositories which support one support plugins for the other, so you really do pay your money and take your choice.

A worthwhile session, which was well attended (and turned out to be part of a two-day workshop on content packaging which I now wish I was attending all of!).

NB. Notes and slides are being made available, and I will post these here when received.

1 Comment »