Posts in Category "Uncategorized"

Who You Gonna Trust?

I recently wrote about Adobe’s new Flash Player 9 and Flex 2 solutions, and mentioned a forthcoming Adobe eBook solution that leverages these technologies. Some folks took issue based on a perception of security problems with Adobe software, citing security advisories such as this. The criticism even descended into satire, warning readers of “Stephen King level horrors” ahead. Well, the bottom line here is (to paraphrase GhostBusters): “Who You Gonna Trust”?
Satire aside, security is a serious topic and merits serious consideration. Before specifying or redistributing *any* reading system client software publishers should consider the security implications, including the track record / capabilities of the proposed vendor (or themselves, if contemplating supporting a OSS/home-grown solution).
While Flash Player and Adobe Reader have not been entirely free of security issues, I believe Adobe/Macromedia’s track record is quite good and compares favorable to other major SW vendors – including browsers. I’m not focused in this area but we’ve pretty much had researchers discovering hypothetical exposures, vs. users experiencing actual malware attacks – in fact the exposure above was discovered by our own dedicated security team. Adobe requires security audits before releasing software, and we treat hypothetical security exposures as critical, issuing patches frequently. We also follow a practice of giving users control over their security and privacy settings. Adobe has distributed far more client software (non-OS) than any one else in the world. When you have 100s of millions of installations are you going to have some security issues? Absolutely. But those who adopt Flash Player or Reader can have confidence in Adobe to address these issues.
Some links:
Adobe security advisories main page
Flash Player security
Detailed white paper on Adobe Flash security
One positive factor here is size. Flash Player is relatively small and that means that the code has proportionally few nooks and crannies (in security geek speak it “presents a smaller threat surface”) – many times smaller than (for example) a J2SE Java VM.
Again, I’m not saying Flash Player is perfect – nor that I personally like all applications of Flash, particularly not in-my-face ads – just that as a basis for a Rich Internet Application (RIA) that goes beyond HTML’s capabilities FP’s security footing and track record is a plus, not a drawback, at vs. any alternative that I’m aware of (certainly compared to trusting an arbitrary native-code Windows app or ActiveX control). If someone tries to sell you another eBook reading system – ask them about their implementation architecture’s sandbox model, their dedicated security team, their track record in issuing patches, their financial wherewithal, and their demonstrated ability to manage large-scale client deployments.
A separate issue I got feedback on is using SWF for eBook content – and I may have confused some people about our intentions on this front. Actually Adobe already has a solution that does this – FlashPaper 2. While FlashPaper delivers some ease-of-use advantages vs. (say) Adobe Reader 7, it did not take the world by storm – if it had I might be a Macromedia employee now ;-). Content publishers want interoperability, transportability, and archivability, and solutions that directly consume open standard formats like PDF and XHTML deliver these benefits. Whereas turning documents into SWF, Mobipocket .PRC, PalmDoc, or BBEB is a one-way trip (although if I had to “compile” my content to one of the above, the format that’s supported on 98% of Internet PCs would arguably be the best choice, and that’s exactly what a number of digital magazine vendors are doing).
That being said, at the point where documents blur with applications – for example learning systems – there’s clearly a role for programmatic interactivity on the client. And for handling interactivity, I’d certainly trust SWF over a native Windows EXE.

More on “Future Libraries”

Walt Crawford was kind enough to point out to me that he has recently published a 10-year retrospective of his own on Digital Libraries: Dreams, Madness & Reality. Unsurprisingly, his view of how FLDMR has aged is more specific (and more insightful) than mine but I think they align reasonably well.
With regard to the strong statement in FLDMR that “Electronic methods are best for ‘housekeeping’ and for giving access to data and small, discrete packets of textual, numeric, and visual information (such as those found in many reference works)… print-on-paper is and will be the preeminent medium for the communication of cumulative knowledge.” Crawford now hedges:

I’ll stand by the last sentence, but the digital/analog split has become fuzzy over time. Specifically, the digital realm as just-in-time distribution method for medium-length narrative in the form of journal and magazine articles has proven far more important to libraries than we could have guessed in 1994. Some would claim that most readers read those journal articles on screen. I believe many students skim through articles in electronic form to find chunks to cut and paste, and that they may glean reasonably good understanding of the sense of the articles. For all I know, maybe KTD really are different and do gain full comprehension from the screen while multitasking up a storm, although I’m still not convinced.

Future Libraries: What Will The Reality Be?

I recently read Future Libraries: Dreams, Madness & Reality, co-written in 1995 by the estimable Walt Crawford and Michael Gorman. I’m still coming up to speed on the implications of digital publishing for libraries, so it was extremely helpful to read this work, focusing on the intersection of digital technologies and the manifold missions of libraries. The general thrust of the book – caution against a “TechnoLust” view of technology as a panacea and the hasty conclusion that physical libraries are soon going away – was clearly prescient in 1995, predating by several years the boom and subsequent bust of “eBooks 1.0”. Paper is clearly still the preferred reading medium in 2006. Our family is lucky to live a few blocks from a fantastic public library and it’s very much part of our world. Staff, space, and collections.
Yet at the same time so much has changed in the last 10 years that many of the specific points in this book now appear quaintly conservative. There’ s a lot of “no forseeable…” things that have in fact happened in a short few years – and the thesis that “in the unimaginable world of 2094, the library might be recognizable to a library user of today” is, to me, unimaginable.
I think it’s interesting look at what’s happened in the last decade as some guide to how much further things could change over the next. By 2015, will we arrive at, if not an “all-electronic future”, a world in which electronic distribution and consumption of texts is broadly adopted? Will the displacement of paper, for immersive reading as well as information retrieval, be well under way? I wonder what Messrs. Crawford and Gorman think now?
Some of the 20-20 hindsight points in this book are:
– No mention of HTML, HTTP or the Web. A year after Netscape’s founding, they are still talking about WAIS and Gopher, and calling the Internet the “I-Way”. A decade hence will unimaginable progress cause current discourses on “Web 2.0” and AJAX to sound as off-key?
– Skepticism about text string searching. “it is entirely possible that there is no universal solution… that allows the user to treat the entire collection of material as a single searchable entry and hope to make effective use of the results.” Google was still a few years off but AltaVista launched the year this was published, and was soon being used by millions of people.
– Elitism of PC/Internet access. Clearly there is still a Digital Divide. Yet, on a global basis, access to Internet-connected PCs and especially Internet-enabled phones is far more prevalent in 2006 than access to “mid-sized” libraries of 50,000 volumes, or good bookstores. We don’t have the $100 laptop yet but we will have the $50 laptop this decade – and it will open the portal to reading millions of works, for the price of a handful of physical books.
– Electronic journals viewed as a “murky” niche of questionable imprimatur. Today major academic journal publishers make the majority of their revenue from “e-only” subscriptions, and dual electronic/print subscriptions are most of the rest. Not only have respected, refereed journals gone electronic, but with “open access” and prepublication articles today’s world of research is a very different world in which physical distribution of printed copies is fast becoming secondary.
– Scanning of magazines impractical. Their numbers of 22MB per page after compression were based on a compression estimate of “25 percent of the original size” – they call this an “optimistic” figure but with the rise of JPEG2000, JBG2 and other more aggressive compression technologies of course this was really a very conservative figure. And, we are now poised to have terabyte+ storage on desktop PCs, and outrageous numbers like “168 gigabytes per year” are now measured in tens or hundreds of dollars. In fact their example of the New Yorker magazine to illustrate the impracticality of digital replicas of magazines is ironic given that in fact that 80 years of the New Yorker is available to consumers as 8 data DVDs for $63.
– Videodiscs a “trail of failures”. Hint: LaserVision was not the answer.
– CRTs entrenched and “the gap between CRTs and thin-screen devices seems to be growing”.
OK it’s perhaps unfair to pick on specific points like these. But I do think they point out just how rapidly things change. I am already e-reading for pleasure quite happily on a QVGA (320×240) PDA/phone and a Sony Reader makes e-reading for me just about equivalent to paper (better, in some respects). 10 years from now when the average smartphone has a full wallet-size XVGA-resolution display (if not a heads-up display), and electronic paper displays are $5 a sheet, will paper still represent 99% of the global reading consumption? Personally, I think it will be more like 70% and falling.
One key point is the understimation of true cost of physical books, especially in the global context. Messrs. Crawford and Gorman make the common mistake of considering just the direct costs of physical printing and distribution and coming with a low estimate for the % of the price of a book these costs represent. But the real point is that these costs and the nature of the physical book artifact multiply by several times the ultimate price that consumers and libraries pay for books. The storage and transportation of physical books and the attendant extra layers of distribution channel (wholesalers, etc.) not only make books and libraries expensive for consumers, but prohibitively expensive to those in the developing world that lack the necessary infrastructure. Not to mention the environmental costs. And one thing is for sure: while a lot of things have gotten cheaper since 1995, the retail price of books is not one of them.

eBook pioneer Jim Baen, RIP

Jim Baen, one of the first traditional book publishers to really embrace eBooks passed away Wednesday night. I enjoy science fiction and a Baen Books best-seller 1632 was the first eBook that I read as immersively as if it were a paper book – a credit to the content, much more than to the e-Reading experience.
Groundbreaking efforts by Jim Baen to market digital and paper forms of work in combination include the Baen Free Library, the Webscription program , Advance Reader Copies. While perhaps not well known outside Baen’s particular segment, his programs have paved the way for other publisher programs such as O’Reilly Rough Cuts.
And, most of all, he demonstrated that draconian DRM is not a sine qua non for effective monetization of digital works. Both selling and freely distributing unprotected content, Baen Books increased, rather than decreased, their revenue and profits. 1632’s author Eric Flint makes an impassioned case against DRM – and Jim Baen took a big risk and proved a good part of this case.
Yet, IMO there’s still a reasonable role for digital rights management in protection of eBooks and other digital content. Publishers will make different decisions about how openly to allow access to works, and Adobe is committed to offering the tools to support content protection and to working to increase the interoperability of DRM solutions. But we are also committed to supporting openly distributed, unencrypted content that can be deep-linked, mixed-in, and mashed-up. And I personally welcome publishers who, like Jim Baen, bravely experiment in a partnership of trust with their readers. Meantime I also want to soften the rought edges of DRM – when publishers deem it necessary at all – so it becomes more of a gentle reminder, a way to help “keep honest people honest”. Hackers will be able to crack any DRM, so in reality that’s the most that a rights holder can expect. iTunes DRM forexample, is a joke cryptographically. But as a user I have to intentionally decide to defeat it. If publishers and eBook sellers fairly price their wares I believe that most users will respect their licenses and that we can all increase our customer base and revenue.
Jim, I never had the pleasure of meeting you, but I thank you.

Flash Player 9 & Flex 2 Released

I don’t typically plump for Adobe products in this blog. But the official release tonight of Flash Player 9.0 and Flex 2.0 is a special case, and IMO is also significant for the future of digital publishing.
The extended Web/browser ecosystem largely is based around three client-tier file formats: HTML, PDF, and Flash SWF. People are finding that Flash is not just a great solution for interactive animations and video, but also a great cross-OS platform for rich internet applications (RIAs) that break the page-at-a-time barrier of typical Web applications. The new Yahoo! Maps , the new Google Finance, Zillow, Redfin, and many others are Flash-powered. But despite growth in adoption of Flash for interactivity and applications, AJAX web apps are getting all the attention. One reason is that Flash has been challenging to program, with a timeline-based authoring metaphor and a scripting language (ActionScript) that, while bytecode-based, offered no real advantages in either language capabilities or performance over browser JavaScript.
Well, now Flash app development is even more compelling and a lot easier. Flash Player 9’s new ActionScript 3, compatible with proposed next-level of ECMAScript, is a powerful language that supports direct XML processing (ECMAScript for XML, aka E4X), strong types, inheritance and mix-in interfaces, exceptions, and other robust software-engineering capabilities. It’s also got a JIT compiler which means a lot closer to native-code performance. Adobe folks don’t like it when I call AS3 “Java Lite” but that’s not far off the mark (and not a bad thing either IMO, given the API bloat of J2SE). And, unlike Java, AS3 is built on an optimized graphics and media engine which enables a far richer user experience than typical Windows desktop apps, much less HTML-based web apps.
And Flex 2.0 makes developing these rich interactive apps much easier – goodby timeline-centric designer-focused UI, hello IDE and XML UI description language for MVC applications and structured data binding to visual controls. Adobe is also easing in to the open source community by building the new development tooling for Flex, Flex Builder, on top of the leading open-source IDE, Eclipse. I’m a useless manager, and even I have been able to whip up cool things with Flex Builder. With 10x less code than, say, a Swing or SWT Java application, and 100x less time than it would take me to debug an unmanaged-memory C++ app or the JavaScript spaghetti of an AJAX web app.
I’m truly impressed with what the Flex and Flash Player teams have pulled off here, and we are eating our/their dogfood. Our team’s forthcoming eBook solution (codename “Twist”) leverages the Flash platform, with a Flex-based UI layer.
And I believe the potential importance of the Flash platform to digital publishing goes a lot further than how app UIs will be put together. As eBooks evolve, they increasingly will not be just digital replicas of paper. They will have interactive multimedia and the line between “document” and “application” will further blur, especially in publishing segments like K-12/higher-ed textbooks and STM. I don’t believe that HTML, even with AJAX and SVG, is going to cut it for representing this interactivity and richness in digital books. AJAX apps are tough enough to make work on the browsers du jour, much less to be expected to be usable on browsers of the future. PDF’s great but is really optimized for representing electronic paper, not interactive applications or free-form rich media. Microsoft’s WPF has a great “spec sheet” of capabilties but a Windows-lock-as the industry-wide cross-platform solution is repugnant. Java is way too heavy and not content-centric enough. Thus Flash SWF seems like an obvious choice to deliver the “surround sound” richness for the books of the future.
OK, sure, SWF and MXML can be viewed as “proprietary” – at least for now. But ActionScript 3 is fast becoming the future of the ECMAScript standard, and there are an increasing number of open source Flash solutions. And Flex is a great way to consume standards-based XML data. It certainly seems possible that some or all of SWF could, like PDF/A, follow the road from vendor-developed innovation to ISO standard. Meantime – trying to be neutral about it – if I were a publisher and had to choose between a cross-OS solution from a vendor like Adobe and a Windows-centric solution from Microsoft for interactivity and rich media in my digiital publishing strategy… or sticking with static PDF or XML documents that don’t deliver the richness that users increasingly expect… the choice seems pretty easy.
Anyway it should be an interesting next twelve months. I congratulate “Macro-Adobians” on FP9 & Flex 2. They rock.

More Progress on Digital Publishing Standards

The IDPF today issued a press release covering a number of positive developments in standardization of digital publishing formats. To me the key underlying messages here are improved cooperation within the digital publishing industry and across standards groups.
Successful industry standards should codify and “bless” established best practices – trying to do invention in committee leads to “castles in the sky” failures. It’s critical to have multiple parties committed to implementing a proposed standard. Appropos is Michi Henning’s just published article in ACM Queue, The Rise and Fall of CORBA. So it’s great to have multiple vendors announcing implementation plans around IDPF standards for reflowable eBooks. While PDF is firmly established as the standard for paginated fixed-format content, we need to get past the “Seven Dwarfs” of competing proprietary approaches for “liquid” reflowable textual content. Coalescing around enhancing OEBPS, the incumbent standard interchange format for reflowable eBooks, supports the general principle for standards success.
Of course the last thing we need is a splinter “fork” of OEBPS but luckily this is shaping up to be a non-starter. Kudos to the OpenReader folks though for keeping the lights on after the eBooks 1.0 bust, for acting as a prod to the industry around the need for open standards, and for contributing to the IDPF standards development process.
Another important principle is appropriate cooperation across standards groups and industry consortia. Unnecessarily duplicative standards fragment the market and block interoperability, which is after all the main goal of open standards. IDPF is well along the way to taking the approach of successful industry-specific groups like the Open Mobile Alliance (OMA). OMA has the means to create industry-specific standards when necessary, but by preference leverages and cooperates with broader industry standards like W3C and OASIS. IDPF joining OASIS and working to align OEBPS Container Packaging with OASIS OpenDocument is a positive development, and should lead to further cooperation. Working more closely with DAISY should lead to open standards that meet requirements for profitable commercial digital publishing as well as accessibility requirements such as NIMAS.

Microsoft and PDF and openness

I’ve received many questions about the recent Microsoft PR campaign about changes to portable document features in Office 2007. My colleague Mike Chambers posted a definitive summary of Adobe’s perspective last night. Recommended reading.
Note that I posted on Microsoft monopolism and PDF (here and here) back in October 2005. Needless to say these were personal opinions of someone not involved in our discussions with Microsoft.

Microsoft “invents” electronic annotations

Software patents are a quagmire, so without further comment here’s a link to a post by Alex Turic at MobileRead. Via David Rothman of TeleRead, who provides additional commentary here.

Report: Academic Library eBook Usage Surging

Interesting post from The Distant Librarian (Paul Pival) summarizing a talk by Ellan Safley of Univ. Texas, Dallas. Some sound-bites:

Usage is exploding–Similar to the e-journal usage 5 years ago.
… the top-circulating ebook was ‘checked out’ 47 times in a month, and the top print book was 1.4 times.
…while UTD librarians reported that lots of people didn’t like NetLibrary, it has shown a steady increase in usage since 1999, and Ellen suggested that librarian perceptions were muddying the picture. Since they often dealt with the password and printing problems with NetLibrary, they assumed the product was no good and nobody liked it, but the statistics suggest otherwise.
…Similarly, ebarary statistics showed a 129% increase from year 1 to year 2.

Obviously e-journals have “crossed the chasm”; Elsevier says that almost all their subscriptions include online and that “e-only” digital distribution sans print edition is now more than 50% of their revenue. So if eBook adoption in academia is really headed in a similar direction, that would represent a very promising sign for the industry as a whole.
And while adoption in a single university doesn’t necessarily indicate a broad trend, we’re talking about Dallas, Texas, not some progressive coastal enclave of iPod-toting hipsters. While it may be UTD’s cross-town rival that’s favored to host the next Presidential Library, if Dallas students are shifting from “look it up in your gut” to “look it up in an eBook”, it’s hard to imagine that the “factinistas” of other institutions will be far behind.

Destroying Books In Order To Save Them: Nicholson Baker’s “Double Fold”

I finally read Nicholson Baker’s Double Fold: Libraries and the Assault on Paper (Random House, 2001). Wow! To me the compelling cautionary point of this book is that the siren song of “beyond physical books” has been playing to libraries for over 50 years. There were microfilm and micro-fiche fads around replacing printed works long before eBooks; fads pushed by meretricious vendors and starry-eyed techno-enthusiasts, often at the expense of the mission and function of libraries and library users.
Having worked on the Paperless Office (sic) since 1985, I certainly appreciate that paper has staying power. While it’s nevertheless obvious (at least to me) that digital representations are going to eventually displace most paper texts, we should be clear about the plusses of the incumbent solution for representing human knowledge, and realistic about the particular situations in which digital alternatives will be preferable in the near-term. Above all we should avoid over-stating the case for digital publishing and the digital library of the future. Adobe is painfully aware of the inflated expectations that accompanied the eBook hype wave of the late 1990’s, but my humility is certainly increased by learning about the microfilm hype wave that began back in the 1940’s. The thought of human knowledge being irretrievably lost due to bumbled conversion is also sobering.
Baker seemingly stretches in depicting a vast conspiracy of librarians and vendors destroying physical books. Paper does deteriorate, even if not literally into dust. And I have heard repeatedly from librarians that they simply do not have enough room for their physical holdings, which tend naturally to expand to fill the space available. As Baker quotes a librarian “there is never enough space”. He does not dispute a study showing libraries spend $1.50 per book per year on physical storage (in 1985 dollars). While as an avid reader I would be happy if library budgets were such that constrained space was a non-issue, what I hear from librarians is more along the lines of “for every volume we acquire, we have to get rid of another”. No wonder that the promise of miniaturization was seized upon by many in the library community.
My challenge to us all is to focus on the user experience. What Baker only obliquely refers to is the reality that microfilm,’s major failings are that it’s difficult to access and much less satisfying to read than paper. Despite his quibbles, it appears that microfilm did deliver the essential benefits sought (saving space, lowering storage costs, and reducing the risk of losing information). The attendant loss of usability and accessibility was what really made microfilm a pseudo-solution. I believe we have to work very hard to make sure that digital books are usable and accessible – ideally to deliver improvement in these areas. Paper’s great but the book is no more the zenith of usability than the traditional photographic print is the zenith of depicting an image. Digital technology brings new capabilities, and we should aim to foster creation of new kinds of compelling content, not just “shovelware” that reduces library storage costs and increases their effective holdings (though these more prosaic needs are real, and are beginning to stimulate eBook adoption).
Anyway, despite its paranoiac tendencies, Double Fold is stupendously good writing. Baker’s obsession with details makes for interesting but somewhat bizarre fiction, but transforms what could have been a dry read on library book (non-)preservation into engaging entertainment. Reading in the Preface that this was an expansion of a New Yorker essay I admit I considered looking up the magazine piece rather than diving in to an over-inflated 350 pages. Well, Nicholson Baker could certainly have made his points with far less prose, but so could Paul Prudhomme serve up smaller portions with less sauce. Recommended.