The Future of Internet Search?

Mac Funamizu has a very interesting posting which you need to see, as the verbal description is not as compelling as his graphic mockup. Hat tip to Brantley, who emailed me the link.

The first example he gives, of a search on a skyscraper vista, is completely believable. Google has been beavering away with its rich cityscapes. You may have noticed that they added Little Rock, Anchorage and 11 other Metro areas last week. City-scape recognition will be a great way of delivering mobile search.

But I found Mac_Fun's second example even more intriguing. He is right, the device that does city-scape recognition will also do user-generated document scanning. That puts the world of digital documents in a completely different ball-game. The thesaurus example he gives, is a little bit lame......there will be vastly richer linkage patterns available. "Show me who else has read this!" For one. Privacy issues come crowding around mobile search.

PersonaNonData had 5 questions

Which we attempted to answer.

I was wondering who else Michael Cairns has been throwing his questions at: Redroom.com, BookNet Canada, Harlequin, Shatzkin on DADs, Rosetta Solutions, Bondi Digital, Shared Book and Lonely Planet. Interesting company.

True Detective

• Jesse James -- in case you missed the film
• If you subscribe you get access to a year of the back issues, which includes records of all UK and Irish 20th Century executions.
• Ireland used to censor this magazine.
  

Murder Most Foul

• Free trial issue
  
• Hungary's Jekyll and Hyde
• Readers' views
  

iPhone and your lifestyle

Or should that be iPhone and myLifestyle?

While I tend to get excited by the iPhone as a new reading device, it is going to be a lot more than that. I bet you did not know that it could do all this. Apple are filing patents for the iPhone to become your fitness trainer, nutritionist etc ...... (hat tip to Unwiredview.com). Here it is with a lot of sensors:

If our preferred reading device is also our omnipresent health monitor, we will at least find out whether or not our reading is good for us.

Conservatism and Change

Many years ago (really a long time ago) at the behest of Michael Lesk I sat for a few sessions on an advisory panel for the Commission on Preservation and Access. I see this is now folded into the CLIR. The CLIR is a very worthy body which supports important library projects (such as the Digital Library Federation which is now directed by Peter Brantley). Although I found the committee discussions fascinating, there were only five or six members from diverse backgrounds, I did not last long. The preservation issues seemed to be a long way from my interests at the time (scientific software) and it was absurd to be flying to Washington for committee meetings. This was before the web, before we knew about global warming. But even in the early 90s flying the Atlantic for a 3 hour meeting seemed pretty pointless.

With hindsight, I think I should have stayed more in tune with the technologies of preservation. Access has always seemed to me a crucial part of the library mission, but preservation is also a fascinating topic.

This thought was prompted by reflection on the way that the Exact Editions platform appears to be heading in a deeply conservative direction. The deeper we go, the more we seem to be finding reasons for tracing and respecting the formalisms and the structure of print. Here are some examples:

1. Our quick view browse mode for rapidly scanning a publication, uses 16 pages. It may be a bit of an accident that this is a good number to show thumbnails of print pages on a web page, but if you have a print background 16 is a really important number. 8, 16, 32 pages this is the arithmetic of the traditional printed sections that go right back to Gutenberg. It seemed to me such a happy accident that this is the way our cookie crumbled.
2. When our technical director proposed to incorporate navigational links in our automated content management system (derived from contents pages and lists of illustrations), it did not occur to me that Indexes would also be included. It did not occur to me either that the indices would be so very useful. But in curious way Indices and Complex Tables are even more useful when a book or magazine has been digitised. Of course the search function of the printed index is not what matters in the digital edition; but indices are still great ways to browse the content of the book.
3. The deepest conservatism in our approach is over the centrality of the page. We are not alone in this. The conventional wisdom five or six years ago was that pages would not matter in digital publishing. It took Google, with its Book Search, to turn around conventional wisdom on that point (and was that entirely down to Larry Page who seems to have been particularly keen on the library project, or did Sergey Brin also play a part?). In fact, Exact Editions is perhaps even more conservative than Google Book Search, or Adobe or the Open Content Alliance in the way that we offer books as Verso and Recto two-page views. Whereas Google et al use a scrolling system. I am still trying to figure out whether this Gutenberg-conservatism of ours is a key strength or a mere cosmetic difference between our style of representation and the rest of them with their Ptolemaic scrolls.

What do I conclude from this? It is possible that print will survive without paper and prosper in a digital world by adhering strictly to the formal structures and conventions of print. Preservation and access will work hand in hand in the digital world. We will model digital books after printed books because that is the easiest way to use them and to preserve the practice of learning, scholarship and reading which they embody.

Opinionated, Syndicated and Speculated -- a Magazine for the Kindle?

Peter Brantley linked me to Joe Wikert's blog with the story that there is a new magazine, Opinionated, launched through Amazon, specifically for the Kindle -- Amazon's new eBook reading device which is in short supply.

The new magazine is culled (kindled?) from the syndicated columnists of the Chicago Tribune group. The Kindle will not support colour reproduction, so there is no need to worry about paying photographers or designers for wonderful layouts. It would be interesting to know the way in which the funding for this operation goes. I am guessing that the 'syndication' is all for free and that the Amazon royalties might pay to Tribune Media Services are peanuts. I suspect that TMS and Amazon both see this as a promotional exercise in which neither party has to invest anything at all; and that is the long and the short of it. If there really is a market for this magazine, there can be no reason at all to limit its distribution to the Kindle.

So if it really works, Opinionated will escape the confines of the Kindle.......

Territorial Rights?

Publishers and broadcasters in their dealings usually define market rights in terms of 'territories'. It will matter to a publisher whether or not their 'co-publication' rights include an exclusive right to sell copies of the English-language edition of a book in Australia, New Zealand and/or the whole of the European Union. Lots of stuff on the BBC websites is only available to users who have cookies or IP addresses which suggest that they are based in the UK.

Globalisation and internet technologies will render these rights increasingly indefensible. BitTorrent or similar peer-to-peer technologies will undermine the discreteness of territorial restrictions. Myka make a small set-top box device that hooks up to BitTorrent and will allow users to download any TV programme, anywhere. BitTorrent can also be used for downloading books, sometimes perfectly legally. According to the economist Gerry Faulhaber (as reported by Teleread) 'copyright is dead'. That sounds a bit extreme, but I reckon that exclusive territorial rights may be in the intensive care ward.

All you can Eat Music

Apple through its iTunes and Nokia through its "Comes with music deal" are preparing to offer unlimited access to the major music companies catalogues through monthly subscription plans. See today's report in the Financial Times. Such services would be pitched at the $7-8 per month level, with Apple apparently needing a large slice of the revenue (this we understand is Apple's style).

I wonder if this will work for music? I wonder if it would work for those of us with non-mainstream interests? Surely premium music would elude this framework? One can al least think about such a scheme in the case of music since the 4 majors control a large part of the recorded music pie. Book publishing is incredibly much more fragmented (also by language), so it is scarcely conceivable that a technology platform could negotiate a global rights pie with umpteen different major print publishers.

Mind you its an idea which might be attractive to some. Would such an over-arching subscription scheme be one way for Google to negotiate a settlement with the publishers and author's societies which are opposing its Google Book Search project? Google would then need to start charging subscriptions for full access to the in-copyright resources in its database.

If this is to be the Google digital books charging model, I would guess that the other players in the market have some years in which to test alternative approaches. Which is what we are doing with the Open Searching/Subscription Content Reading offering that we this week launch for Berkshire Publishing.

Google as an aggregator of Book Services

A friend asked me the other day whether we were concerned that Google may simply come to dominate the digital book publishing space. Well in one way there clearly is a concern. If Google were to become a monopolist in the digital books/subscription services area, many companies would find it hard to compete.

Suppose that Google were to start selling access to digital book collections would that be a concern? These thoughts were in part stimulated by the announcement of a Google API for viewing Google Books (see our comment here). But they were further encouraged by our own announcement, today, of a platform from which publishers can licence and sell access to their own books.

Should we be worried if Google were to make a similar announcement (they may have such a development well in hand, as has been foreshadowed by well-informed observers, eg Personanondata)? The short answer is "No, this is not a concern", Google very probably will do this and when it does the position of competitor/complementor companies and technologies will be clearer.

At some stage Google should offer or enable a method through which titles deposited by publishers on the GBS platform can be licensed. But how Google will do this (whether they will be 'trading for their own account' or merely facilitating the creation of digital markets -- eBay-style) is a pretty good question. My own guess is that Google has enough on its plate trying to develop and energize its advertising platforms to spend too much time figuring out methods of selling access to content. But when they do start, we will all be in attendance.

Berkshire Publishing Shop

Berkshire Publishing have launched a shop using the Exact Editions platform and our e-commerce solution. So their customers can now license their major reference titles online, either for individual use or for institutional subscriptions.

The Berkshire 'home page' on our system has a look and feel which reflects the publisher:


















All the text in the reference works is searchable, and to an extent the books are also viewable for free in the shopping environment. The limitation is that the pages can only be seen in the 2-page form, which is reasonable for browsing but not at all comfortable for reading. Here is an entry from The Berkshire Encyclopedia of Human-Computer Interaction on 'Fly-by-wire'.



















One of the advantages of showing the 2-page view is that users can then see and benefit from the links in the books. Here is a page of internal navigation links from that book:
















This innovation marks an important step change for Exact Editions. In working with book publishers we intend to make it easy for them to create a direct-to-customer e-commerce offering for their titles. Berkshire's titles are all searchable for free because the depth and value of these titles is gracefully and thoroughly exposed by searching. They are for sale and at reasonable prices because the publisher knows that they have a real market in schools, colleges and among experts. Karen Christensen, Berkshire's publisher, is keen to push the envelope on ways in which traditional reference can become even more relevant and indispensable in the web eco-system.

There is a surprisingly widespread myth in some web circles that all content on the web will be free and attempts to sell subscriptions will not work. Our experience with consumer magazines and the STM publishers experience in selling periodical subscriptions to institutions shows that this is nonsense. Book publishers need to stop sitting on their versos and start offering their goods through the web medium. The challenge now is to provide scaleable and easy to navigate services for end-users.

The Exact Editions system has been generalised in some crucial ways to support this development. One development has been the new form of searching and full-content browsing which the Berkshire shop is showing to full effect. Another development has been to make our 'e-commerce solution' a highly customisable framework, in ways which also retain the benefits of scale among a diversity of publishers. There is a page where our alternative shopping links can be reviewed. There is scope for many more shopping niches and many more currencies.

Google Book Search API

Google have announced an API to their Book Search service. Some explanation and a demonstration of how this can be used at the Librarything. Publishers may be uneasy about the way this tips the information flow about books in the direction of Mountain View (Eoin already is). Amazon and the major publishers must be thinking about the implications of this: will all commercial transactions go with a Google information flow?

The API could be either a step in the direction of Google becoming the primary source of all books through the web (its an API to the viewability of books, but access rights go with viewing opportunities, and it would not be so difficult to extend the API so that it interacts with Google Checkout). For the sake of a healthy publishing industry one hopes that there will be many alternative and additional sources for access to digital books. Google has to build an API (it certainly should not be criticised for doing so) but its implementation and regulatory focus on Google's dominant position could become a matter of concern -- also to Google.

What do we do best?

It is sometimes quite difficult to understand the core competence of the business that you run. I was musing about this following a conversation with a book publisher who has been looking at our system and likes it. He thinks that it handles his books well, that we appear to be easy to implement, easy to use, little investment is required and we appear to fit in with the strategic direction of the company he is running. He also appreciates that our platform would be customised to match the requirements and style of his company. He went on to say: "But aren't you really a half-way measure whilst we work up to creating our own in-house content management solutions and creating our own in-house e-commerce system and hosting our own books and customers on our own servers?"

I didn't have the perfect answer for him there and then (though I think I did refer to computing in the cloud). I am not sure that I have the perfect answer now. But I am absolutely sure that he will not want to build his own proprietary e-commerce and content delivery system. I guess that even the largest publishers Elsevier, Wiley and Pearson etc will be making big mistakes if they reckon on building their own customer and content delivery systems in the days of cloud computing.

But I have been thinking about how we are also looking to outsource as much of our business as we can provided that the outsourced service is reliable and cost-effective and scaleable. After all we don't want to run our own boxes, or employ more database managers than we absolutely need. The answer for this mid-sized publisher and for us is that you really only want to hang on to the functions that are essential to your business and which you do better than anybody else. Increasingly IT functions will not be the function that anybody wants to retain in-house. We are looking for better ways to outsource important but non-distinctive functions to Rackspace, PayPal, Amazon S3, OpenID etc....

One function that we would still find it quite hard to outsource is customer service (we can do more to automate it, but we have no plans to outsource it to the sub-continent in the manner of Dell). On the other hand we would be very reluctant to hand this crucial function back to the publishers and they are by and large very happy to leave the responsibility with us. I am not sure that customer support is one of our core competences, but it is a reputation-maker and a reputation-breaker. Whatever you do best, you need to make sure that your customers are happy with it.

Orphan Copyrights

At the recent Tools of Change conference I met up with Peter Brantley, the Executive Director for the Digital Library Federation. We had a very brief conversation about Orphan Copyrights and I suggested to Peter that he produce some thoughts on the subject for the Exact Editions blog. He has produced some specific and intriguing suggestions in an essay which should attract discussion. Since his essay is in part a call-for-action which may involve lots of parties -- the posting is now up on his Shimenawa blog.

I particularly like the features of Brantley's proposals that amount to an Adoption Agency for Orphan-ed copyrights (improved access, digitization, take-down provision, escrow account, and reconciliation of public and private interests). This is a set of important issues which is best tackled by the broadest possible range of interests.

Site Licenses in the Night

We expect to do business 24X7. So customers buy magazine subscriptions at all hours, and it is fun to look at our shop in the morning to see how many Americans, Japanese and Australian customers have done business overnight; but I did not expect that we would be selling institutional licenses at the weekend. Our institutional customers can sign up on-line and complete the purchase by informing us of their IP range.

This last couple of days we sold site licenses on Saturday and Sunday in countries which have our western weekend. From this I conclude that librarians work at weekends. Well, I kind of knew that already, but its nice to have it confirmed.....Is it also important that site licenses carry the kind of price (lowish as institutional prices go -- €300.00) that a librarian can sign off on her own decision on a Sunday?

Site licenses for our consumer magazines are proving to be popular.

This demand from institutions for campus wide access to a digital editions of the whole magazines, is a market which most consumer magazine publishers are simply unaware of. Book publishers also do not appreciate how much they are neglecting a market which wants their books -- as digital resources.

Green Web Services

Although it is reasonably obvious that digital magazines are much less damaging to the environment than printed consumer magazines (Chris Anderson has a weirdly contrary view), it is quite hard to do serious measurement of the ecological cost of digital publishing. Steve Souders has done some analysis of web sites with Yslow in an attempt to put some cash/calory value on the inefficiencies of the Wikipedia home page. This is highly conjectural and guesstimatory but probably important. O'Reilly are publishing his book.

The technicalities of the Yslow plug-in (its a plugin to a Firefox plugin ) are well beyond me; it checks whether the inspected web page could make fewer HTTP requests, uses a CDN, puts CSS at the top, JSS at the bottom, avoids redirects etc. etc. If you understand all that (which I dont) you will want to plugin Yslow to your already installed Firebug.

This kind of measurement is obviously not the whole picture and Yslow is primarily tackling the problem of why some pages are slower than they should be (slow partly because they waste energy). Its only part of the picture because what happens on the web page in a browser is one thing, how much power the databases consume behind the scenes are another ball game (possibly more important in ecological impact). These issues will concern us as we move towards cloud computing. Apparently Google and Microsoft are anxious about the ecological costs of cloud computing, and are aiming at frugal computing.

Utopia

Utopia is now digital in Exact Editions

• ToC
• Another design magazine with fabulous ads.
• A bath with a view.
  

New African

New African is in Exact Editions.

• ToC
• Kenya What Really Went Wrong.
• Why Africa?
  

Flash on the iPhone?

Rumours have it that the iPhone SDK due to be launched today will not support Flash. Steve Jobs is being quite rude about Flash. But perhaps this is all part of some complex 3 way face-off between Adobe with its well entrenched position on PDF, Acrobat and Flash; Microsoft with its new and promising Silverlight; and Apple which has its own good reasons for not being beholden to anyone for a plug-in.

Who knows? Well it matters to us, perhaps indirectly benefiting us, because quite a number of the semi-plausible way of rendering documents in browsers use Flash (most of the plausible ways of rendering video in browsers also use Flash). But the Exact Editions system doesn't work that way. We only have a modest requirement for Flash in our 'Clipper'. So these problems for Flash are mostly problems for someone else.

On the other hand the urge to use Flash (0r PDF) is the result of a deeper 'missed turn'. Browser plug-ins, however nifty, are seductive short cuts which may encourage content owners/creators to miss the potential of web development. If content has to go into proprietary wrappers that may be a headache for us all. The content may be lifeless rather than fully engaged.

Microsoft's Book Search

Microsoft has a very nice implementation for its Book Search.

See here for some examples 'C Elegans and Mutations', or 'Girard and Scapegoats', and 'Quine and "semantic ascent"'

When you mouse over the individual titles in the search results the second panel shows information specific to the title: including a rule which shows the distribution of the search results within the title (the rule is imperial rather than metric, or maybe its binary: it has 32 segments). That must all come up on the fly. It is impressive what computing clouds can now do for us. Lorcan Dempsey, who notes this strong feature of the Microsoft presentation, calls this immediacy of information flow glanceability.

Tools of Change

I love the quotation attributed to Zhou en Lai -- on being asked what he thought of the French Revolution, he replied "Its too early to say". I love it because its so cautious, so careful and so definitively correct, but also engaged ("Mr Zhou: You mean the French Revolution is still happening?"). Its always going to be too early to say.

I feel like reaching the same conclusion about the extent to which publishing is now about building communities, or whether book and magazine publishers should really be engendering conversations. That seemed to be one big theme at this years Tools of Change Conference in NYC (you can reference presentations by Stephen Abram, Douglas Rushkoff 'Whose Story is this Anyway, When Readers become Writers', Gavin Bell and Ben Vershbow 'Books as Conversations'). Excellent presentations they were, but all a bit too much under the sway of the 'social graph' which has become the analytical mode of the moment.

Publishers still need to concentrate on presenting content in the right way. Priority number one. The 'right way' includes with the appropriate access networks; and the right meta-data. Maybe that is all -- and communities may be another story. Michael Bhaskar at thedigtalist is hitting the right notes.

Dumbo feather

Our first Australian magazine, on its own in our Australian shop.

• ToC
  
• Could you wear this?
  
• Gardening.
  

Opera and Google Documents

Opera, the browser, continues to impress. As does Google Docs. But there are one or two issues to think about before you chuck out Firefox and ceremonially burn your .doc documents.

First off, Google Docs does not support Opera. On the Google Docs systems requirements page they say "Google Docs is not supported, and probably won't run on (Opera)." Well it sort of runs on Opera on my Mac but only in a 'view' mode. I guess this may change now that Google is (probably) paying money to Opera for prime position on the Opera mobile platform.

Before you get too committed to Google Docs you may want to think about just how good it is as a collaborative environment. Its really, really good at this; much more straightforward and intuitive than MS Office. Last week we were having a good discussion with some potential partners about the comparative merits of our platform. They had fielded an excellent set of questions for us and forwarded them as a link to a Google spreadsheets document, which we were sharing online in a Skype conference call. Before the Skype call, we tracked back through the revisions to the document and found some benchmarks that they had established with a competitor. The revisions had not locked out the previous column of tests and I am (fairly) sure that we were not meant to see the comparisons that were unearthed. Collaborative tools may be more collaborative than intended. I like a partner who is prepared to share their evaluation with us, especially when they appear to have looked at the opposition, but "open-ness" has its limits.