In an earlier post on the Scrum development process, I outlined why I felt a lightweight agile process was a good choice for the IngentaConnect engineering team. This set me thinking that the way we express our architecture is similarly minimal, with many of the same desirable features, such as ease of adoption and consideration of needs outside the engineering team. However, while Scrum is very much in the public eye and receives ample press, the architectural model we use, Kruchten's 4+1 View, doesn't seem to have had its fair share of the limelight.

The 4+1 view was designed by Philippe Kruchten, and like Scrum one of its guiding principles was to address the needs not just of one specific group, in this case software architects, but all the people with a vested interest in the work. Krutchen describes these folk as stakeholders. These are the developers who will write software, the integrators who will manage servers, the project managers and the product clients. The concepts in 4+1 aren't all new, but they are a unique blend. The UML offers a smorgasbord of diagrams and documents to express an architecture, Krutchen's model distils the multiple representations into just four core models, or views, reinforced by a suite of worked examples or user scenarios.

The logical view is concerned with user features, services and functional requirements. The view illustrates key concepts, or abstractions; these become the software building blocks that will be needed to deliver features.
The development view, is focused on implementation, and the organization of the software. This includes grouping of code into easily managed units, determining functional layers and the interfaces between them.
The physical view illustrates where software components are actually deployed (in simple terms which machines they run on) and considers communications between software services. It is intended for system engineers and integrators, and is likely to present a number of variations tailored to different needs, for example test systems and production systems.
The process view takes abstractions from the logical view and considers how they will behave when executed, including issues like synchronisation and concurrency. This is achieved by breaking each process into tasks whose interactions may be more readily examined.

I've mentioned the four views, what about the +1? The plus one element is a set of user scenarios, or use cases. Although as vital as the views, they are denoted by '+1' because the model is already complete without them, there should be no new architectural information contained therein. The plus one element does however play a crucial role. It illustrates and communicates the architecture, concrete examples reinforce the modelling concepts in the views, and for those not familiar with modelling diagrams they provide an idea of what the architecture can achieve. User scenarios also serve to test and verify the model; working through a realistic set of interactions can expose new concepts, and generate confidence in those already identified.

So that’s 4+1 in a (rather small) nutshell, but what does is it bring to Ingenta’s technical products such as pub2web? The value of software architecture has parallels to the value of building architecture. It provides a vision of what is being built, defining boundaries, layout and interactions. The product shape is described even though the nuts and bolts aren’t. Despite this, the architecture provides enough detail to drive component design and implementation. The discipline of 4+1 ensures that non functional requirements such as scalability and fault tolerance are considered from the outset. In short; good architecture answers questions the architects did and, crucially didn’t, expect to be asked.

Labels: 4+1, architecture, ingenta, ingentaconnect, pub2web, publishing technology, software

Just a quick posting to let you know that the latest issue of our publisher newsletter, eyetoeye, went live yesterday. It's a pretty packed issue! I've already posted some of its stories on the blog, but others are new for the newsletter. Here's its table of contents:

Contents

Welcome

News

Introducing pub2web, a next generation publishing platform
In November 2007, Ingenta announced the launch of its next generation publications platform, pub2web. In this article we explain the benefits of this feature-rich publishing system. [More...]

New faces at Ingenta: Rebecca Lenzini and Jeff Downing join our team
Our publisher and library new business groups in North America will be greatly strengthened by two recent appointments. [More...]

Subscription renewals: the simple way to keep library customers happy
Ensure your subscribers have seamless access to your content by following our simple step-by-step guide to renewals. [More...]

New publishers joining Ingenta for 2008
We highlight the latest publishers and titles to take advantage of our services. [More...]

InSider

Publishing Technology Trends: a look at the hot topics for 2008
In December we held a successful UK forum on the rapidly changing trends in Publishing online. Learn more about the topics discussed in this informative article. [More...]

Low investment, low commitment opportunity to explore new revenues
IngentaConnect's advertising programme continues to grow in strength with the announcement of an exciting new partnership. [More...]

2007 in review: 2008 insight
Last year saw many exciting developments in Ingenta. Catch up on the latest innovations and get an insight into what we have planned for the coming year. [More...]

2008 events calendar
This year is set to be another whirlwind of Conferences and shows. Find out which events we'll be attending during 2008. [More...]

InSight

What's happening in the information industry: KBART, Transfer, COUNTER, MESUR
We bring you up to speed on the various industry-wide initiatives we're involved in. [More...]

Labels: "online advertising", "publishing technology trends", counter, eyetoeye, ingenta, kbart, mesur, news, pub2web, transfer

(This posting reviews a presentation entitled "Key issues in the development of publishing: the age of information services and solutions" given by David Worlock, Outsell's Chief Research Fellow, at the Publishing Technology Trends event held in London in December 2007).

The history of the web can be viewed as three ages:

• “web transfer” – the early 1990s, when early adopters hastily and perhaps clumsily posted everything we could online
• “web synthesis” – the late 90s and early 00s; we tried to build on the web’s capabilities by making new things work
• “web catalyst” – the age we are now entering; a networked society reinventing the online experience from the bottom up

Larger publishers have been slow to come on board with the catalyst age because they are generally too distant from their users. Nonetheless, during the last year, a lot of activity has been stimulated by:

• workflow and the desire to enhance productivity for our users
• compliance issues as the industry becomes more standardised
• growing familiarity with users as we monitor what they do before and after visiting our site
• building brands in the networked environment
• helping perplexed users to filter the wealth of content with which they are inundated.

2008, David suggests, may be the year in which we:

• break the schackles of the network operators; Asia and Europe are driving change in this area, though the US is still attached to inappropriate and old-fashioned standards
• embrace vertical search for the authority it can bestow upon subject-specific content
• use tag clouds to monitor rapidly changing user requirements
• create competitive advantage by positioning organisations within social and enterprise networks to support growth of sector-based communities
• capitalise on consumer power with innovative marketing and product development
• recognise the value of bundling and syndicating content to grow revenues

David's slides list the companies that he thinks will perform best (or most interestingly) in 2008, along with his top ten "issues that matter". He closed with "the agility imperative", which is spot-on I shall reiterate it here:

• insist on agile mindset, organisation and processes
• be clear on value proposition
• adopt new channels
• experiment with new revenue models
• look for cost reduction through reinvention
• assess readiness
• drive speedier time to market
• watch valuations and drive cash flow.
  

Labels: "publishing technology", compliance, ingenta, publishing, trends, users, workflow

(This posting reviews a presentation entitled "Keeping pace with online publishing challenges: why and how" given by Randy Petway, VP Publishing Technology, at the Publishing Technology Trends event held in London in December 2007).


Even now, the online revolution remains the biggest challenge facing publishers today; many have not yet scratched the surface of the digital potential of their content, while those who have made some progress are nonetheless challenged by the speed of technological evolution, which increasingly disconnects key sources of competitive advantage from their traditional core competences.

The digital challenge can be broken down into a number of elements:

• user expectations, particularly as users become more familiar with the web and other software applications
  
• technology shifts - and knowing which ones you need to keep up with
• rate of change, which is much quicker in the online world than in the print world
• how to monetise content; even mission-based publishers should be seeking to maximise the revenue potential of their content.

Continuously-evolving user expectations
Online shopping (Amazon) and social web (Facebook, MySpace, orkut, YouTube, reddit, StumbleUpon...) sites not only encourage more users to spend more time on the web, they shape those users' expectations of how websites look and, more importantly, how they function. And users of these sites are more in our target demographic than we expect, with 53% of MySpace users and 42% of Facebook users in the 35+ category. Furthermore, a strong brand or compelling content will not protect providers from the need to innovate. Fodors, an authoritative and well-known brand founded 70 years ago, with a 10-year-old online presence and 700 global correspondents, has lost ground to Trip Advisor, a 7-year-old newcomer which can boast more traffic than its older rival. Trip Advisor is interactive and incorporates user-generated material along with practical functions (e.g. trip planning) which attract and engage users more than plain old-fashioned reviews.

Technology shifts and rate of change
The adoption curve for new technologies is increasingly steep. Three years ago, only non-adopters had heard of AJAX, but 62% of CTOs surveyed during 2007 were expecting to have it in use by the end of the year. Analysing adoption of Facebook applications tells a similar story - Funwall, introduced in Aug 2007, had 4 million users by December - and 1.5 million of those had signed up in the last 7 days alone. The converse of such quick adoption, of course, is quick abandonment when things go wrong - a minor bug in Facebook app Superpoke saw it lose hundreds of thousands of users in a few days; they migrated to a substitute application that did not have the bug.

Monetising content
Content delivery processes used to be binary; publishers delivered their content to consumers. But now we use multiple channels to reach users, and package content with supplementary elements which have begun to be perceived as part of the product (community and mobile features, for example). A consistent platform is required to support this level of functionality - one that allows for modules to be bolted on when new features are required. Topical features include:

• Granularity - of content, of licensing, of reporting, and so on
• Microsites - interlinked but separately-branded sites that support a publisher's house of brands
• Virtual bookshelves through which publishers can participate in the buzz around their product (author blogs and "view inside" snippets, etc.)
• Community features such as user-generated reviews and recommendations
• End user personalisation, allowing users modify their experience according to their preferences (for example, users over 36 tend to be searchers whereas those under 36 tend to be browsers)

Given the futuregazing required to develop appropriate modules, and the complexity of building, implementing and maintaining them, many publishers choose to outsource their platform to a third party that can deliver and develop it in a cost-effective manner. As Home Depot would say, "You can do it. We can help".

Labels: "online publishing", "publishing technology", ingenta, trends

(This posting reviews a presentation given by Paul Goad, Managing Director of Tacoda, at the Publishing Technology Trends event in London in December).


Behavioural targetting is gaining traction in the advertising industry. Traditional online advertising values users based on their context, and favours big sites that have their own sales force and connections to networks; smaller providers simply had no way of breaking into these networks. Behavioural targetting, however, values users for themselves rather than for their context. It's the long tail of advertising; small user groups can be monetised much more successfully, enabling smaller site owners to benefit. An understanding of the user's interests enables the site owner to deliver content based on *them*, rather than on the niche interest reflected by the site they are visiting. Behavioural targetting has thus caused an attitudinal shift among agencies and advertisers; suddenly, it's not about *where* an ad is displayed, but about *who* it is displayed to. Publishers can now be rewarded for having built up a community of users.

Behavioural targetting has also extended the areas that advertisers can tap into, beyond the limitations of contextual channels. Agencies such as Tacoda, which provides software and services that support behavioural targetting, can provide comprehensive statistics that not only enable optimisation of advertising strategies but that further help publishers to understand their users, which in turn can help publishers refine web strategies from traffic management to content segmentation.

Simply, behavioural targetting works by categorising the sites within a site owner's network and noting which categories of site are visited by a user. Profiling is anonymous (using browser cookies) and no personal data is collected or stored. Browsing behaviour is monitored and the data is matched into one of many hundreds of profiles. Once the cookie (and thus, the user) has been matched to a profile, appropriate ads are displayed. Site owners have control over the types of ads displayed and can respond to any sensitivities in their userbase; they also benefit financially from contributing the data that helps to build core profiles.

Labels: "online advertising", "publishing technology", behavioural advertising, behavioural targetting, ingenta, tacoda

Remember December?
We held the first event in our new Publishing Technology Trends seminar series. The venue was Shakespeare's Globe theatre, on the banks of the river Thames in London. The session we held there was designed to communicate the latest developments in information industry technology to selected publishing industry executives. I made copious notes, which I will share with you in a series of postings over the next few days. First up is a review of the session by our own Chief Technology Officer, Leigh Dodds, entitled "Authoritative? What's that? And who says?"


Sated by lunch or fascinated by the topic? For whatever reason, you could have heard a pin drop among our audience as Leigh Dodds reviewed the ways in which we ascribe authority to content, explored the potential for crossover between traditional peer review and emerging Web 2.0 systems, and considered whether we can make processes more visible to end users.

The massive amounts of information available both through conventional publishing channels and on the web make it difficult for users to find reliable information. Particularly disturbing for publishers is that users are, ultimately, more concerned with finding an answer to their question than with issues of authority. Furthermore, users often have a very different understanding of authority to publishers; consider, on the one hand, the widely-accepted Google model wherein subjective measures of popularity and relevance are a proxy for authority, and on the other, publishers’ expectation that authority denotes submission to, and acceptance by, a formal process.

Despite the well-documented cases of abuse in the last year, editorial control – and particularly peer review – remains the most effective way to filter research output ensuring that published content is the most relevant, interesting and authoritative in its field. However, this formal publishing process is subject to pressures including the costs of filtering ultimately unsuitable content (average turnaways are 80%); the time it can take for content to undergo the process; and the constant need for the new material which attracts most usage.

Web 2 publishing certainly reduces the time-to-market as the majority of processes take place post-publication. User-generated sites such as Wikipedia certainly benefit from the speed and simplicity with which pages can be created, reviewed and edited - but even Wikipedia itself does not describe the content delivered through such "creative anarchy" as authoritative. Its creator attempted to deal with some of its infamous problems in the business model for second-attempt Citizendium - which only allows editing by registered users, incorporates marks of "quality" and is managed by subject editors. The success of sites such as Postgenomic suggests that the "wisdom of crowds" approach is even more effective within a subject silo, while their ability and tendency to include related material (conference programmes and reports, blog postings) brings them closer to traditional publishing.

Nature Publishing Group attempted to combine the traditional and emerging approaches with its open peer review trial. Although ultimately unsuccessful, the project was noted for its transparency - options included making identities of authors and reviewers public, publishing reviewer comments and even allowing end users to contribute. However, this transparency may have contributed to the project's lack of popularity, as academics will naturally be wary of publicly criticising one another's work. The lack of integration with other workflows was another factor preventing this concept from catching on at Nature, but one interesting observation was that posting content online earlier in the publication process did encourage authors to make it more presentable; a transfer of responsibility for some part of the copyediting process from publisher to author.

Leigh proposed taking forward the "open" concept in terms of openly demonstrating to users that content has been reviewed in some way, for example with a kitemark for peer review. As with Creative Commons licences, this would need to combine human-readable logos with machine-readable embedded metadata. Kitemarked content could therefore by searchable (as Creative Commons content currently is). A current example which displays promise is the BPR3 scheme, which enables bloggers to mark postings containing scholarly subject matter (in order to separate them from personal postings); BPR3 is being reviewed by CrossRef, which may carry out prototype work (in which Ingenta would participate).

Leigh has bookmarked further reading material at http://del.icio.us/ldodds/charleston-2007-11; please do share your comments below.

Labels: "editorial control", "nature publishing group", "open peer review", "peer review", "publishing technology", "web 2.0", bpr3, citizendium, crossref, ingenta, postgenomic, wikipedia

I spent last week in Frankfurt, where the publishing industry gathers each October for the annual Book Fair. It is always a huge amount of fun, but exhausting (one of my feet has still not recovered full sensation). The fair is vast (about 40 acres, if my calculations are right - 170,000m²), with nearly 300,000 attendees this year, and more than 7,500 exhibitors from over 100 countries.

This year's fair was particularly exciting for us as it is the first time our three divisions, Ingenta, VISTA and PCG were all together on one stand (pictures to come), and we were launching our new group brand, Publishing Technology. Plus we had some big news to share with the community: at the top level, all three divisions have been involved with, and will continue to contribute to, our recent deal with the BBC to power their BBC Monitoring web service, a subscription-based news service for which we will provide market research, information architecture, web design and maintenance, content conversion and hosting, subscriptions management and billing software, and sales representation to help them sell the service into new markets.

Each of our individual divisions also had major news to broadcast at the Fair; Ingenta announced its new deal with The Charleston Advisor, consumer report for the information community, which will now be hosted on the IngentaConnect platform.

Our VISTA division announced the ongoing success of the information commerce software it acquired from the merger with Ingenta; Dutch publisher Sdu is the latest client for the ICS system, which is already fully live and supporting three websites for launch customer Institute of Physics Publishing.

Meanwhile, Publishers Communication Group (PCG) released its latest research, into the evolution of renewals trends in scholarly journals over the last four years, which shows the increasing migration from print to electronic subscriptions and the growing influence of faculty recommendations on cancellations. We had copies of this study with us at Frankfurt and it has certainly aroused the interest of a large number of both librarians and publishers.

As if that wasn't enough, we also pre-released a white paper at the event, titled "Trends in Digital Commerce", that assesses current strategic trends and future expectations regarding publishers’ digital publishing activities. Keep an eye out for the official release of this white paper in the next month or so. (And for tomorrow's posting of the other news circulating at Frankfurt this year).

Labels: "charleston advisor", "digital commerce", "frankfurt book fair", "information commerce", "publishing technology", bbc, ingenta, pcg, renewals, vista