eIFL (Electronic Information for Libraries) is an interesting organization of developing countries.
I have been in Vilnius Lithuania for the last 5 days working with the members of a great library group called eIFL. Art Rhyno and I gave an all-day workshop on open source (Art did all the work) and we then participated in a couple of days of conferencing. It was a very good set of sessions but more importantly, it is a great group of people. In fact, it is the only other event I have attended that had a feel similar to Access. A close-knit group with a passion for the issues they are meeting to discuss with an opportunity to enjoy themselves. The group has representation from about 50 member countries, mostly from Eastern Europe, the Middle East and Africa. Much of the sessions were concerned with things open, so I really felt at home. What I particularly noticed was that the group is very keen to make use of open source and open content and is very savvy about the technology and issues - more, I suspect, than many people in North America would think.
2005-Feb-05 eIFL Open Access project
I think we have to be careful between what the vendors want SOA to be, and what we as organizations want from SOA. They are using a number of approaches:
With that in mind, I will mention some SOA stuff from Sonic Software.
They have been doing SOA Fora in Europe. It's not clear how much is seminar and how much is product pitch; this is always a risk whenever a vendor presents anything. They describe them as follows
In this interactive technical workshop, you will learn how service-oriented architectures (SOA), enabled by the enterprise service bus (ESB), help solve the integration challenges faced by most organizations.
David A. Chappell of Sonic writes about the seminars in Week 2 of the European SOA Architect Forum series.
This week’s round of SOA Architect Forums began in London with over 100 attendees! We then traveled to Milan, and ended the week in Utrecht, Netherlands. I’m currently sitting in a coffee shop in Amsterdam reflecting on the events of the week.
The attendance at these forums has been averaging between 80 and 100. The quality of the audience at these events continues to be extremely high, with most people being Sr. level architects who are responsible for defining and delivering a SOA strategy for their organization. Its interesting to note that a year ago, when I asked the audience for a show of hands on who is actively involved in a SOA related project for teir company, I would get about 10% of the audience. Now when I ask the same question I get almost everyone in the room raising their hands. The intensity of the questions and depth of their inquiries indicates that they are keenly interested in what is being presented and discussed at the forum.
InfoWorld did a cover story a month or so ago (September 12, 2005): Building SOA your way.
It discusses some of the concerns that vendors may be hijacking the SOA momentum with the huge and complex WS-* "standards".
A fault line runs beneath the groundswell that began a few years ago with XML Web services and continues today as SOA (service-oriented architecture). True, nearly everyone agrees that XML messaging is the right way to implement low-level, platform-agnostic services that can be composed into higher-level services that support enterprises business functions. Yet, here’s also a sense that the standards process has run amok.
It continues on with a survey of how several organizations are using SOA, and concludes:
Of course, toolkits and frameworks are double-edged swords. Even when wire protocols are standard and open, you can get locked in to proprietary abstractions layered on top of those protocols. That’s why pragmatic architects and developers who don’t yet need advanced WS-* features tend to focus on the basics: SOAP and WSDL. “If you need some kind of envelope, why wouldn’t you use SOAP?” Subramaniam asks. “And if you need to describe your interfaces precisely, why wouldn’t you use WSDL?” Frank Grossman, co-founder of Mindreef ..., says that most of the customers who use his company’s SOAPscope diagnostic suite have adopted this strategy, which he adroitly labels “WS-JustRight."
For Grossman and others, WS-JustRight means using SOAP and WSDL to strike a balance between formal contracts and agile interoperability, while laying a foundation for future use of more advanced SOA features. PGP’s Brodbeck agrees that WSDL is the key enabler of reusable business transactions and processes. He also extends the definition of WS-JustRight, however, to include enterprise-enabled RSS as the key enabler of reusable content.
Jon Udell, author of the above article, is also soliciting input for the next SOA Executive Forum in New York November 7-8, 2005 (there was a previous event in May).
SOA Executive Forum Day One: Proposed Topics
SOA Executive Forum Day Two: Topics
To me I guess there are a number of challenges to SOA. You can certainly communicate the concept of business agility to management, and get their buy-in, but still encounter resistance when you have to radically re-structure projects in order to align with SOA objectives. As well, in your development groups, you will need to bring in more agile methods of development, including streamlined requirements-gathering.
People get caught up in the mantra of "it's just XML, SOAP and WSDL" but you can design siloed applications using these technologies, just as you can implement an SOA without all the Web Services bits. To me it's really about creating an agile organization. That is a deep structural change, it's not like taking out C++ and replacing it with Java. It touches all levels of the organization and its processes.
What I'm most interested in is agile modelling methods for SOA, and best practices in identifying component granularity and interfaces. (This of course reflects my position as an Enterprise Architect.)
The 2005 Canadian Access and Privacy Association Annual Conference is taking place November 22 at the Civic Centre, Landsdowne Park, on Bank Street, in Ottawa.
from Library Boy
2005-Jan-05: (another different) security and privacy conf in Victoria
What is possible with conference coverage seems to change with every year as more and more people become comfortable with web tools. There seem to be just a few basic elements that succeed:
UPDATE 2005-Nov-03: It's also worth mentioning that if you're going to encourage use of wireless, that means supporting laptops, which means you should try to provide lots of access to power outlets and run power strips everywhere. Until we have wireless power, an outlet is a precious resource to a laptop user.
Anyway, those are my observations based on the conference coverage I have done and the different approaches I have seen.
The conference wiki this year is at http://internetlibrarian.pbwiki.com/
If you've never seen Tim Berners-Lee use his original web browser, it's worth digging up some video. He developed it on the NeXT, which was a strongly object-oriented operating system. Some of those object-oriented concepts and philosophy live on in the MacOS X development environment (which essentially is the next-NeXT, brought to Apple by Steve Jobs), where you can build applications by combining objects.
Anyway, the original browser had the concept of links basically as objects.
What is an object? An object is data plus methods. This is a key concept. Data does not stand alone. Data alone is useless. Data alone you have to wrap in massive hacks to discover it, parse it, and manipulate it in some useful ways.
If you watch Berners-Lee use his browser, here's a bit of magic: you never see URIs. Why should you ever need to know the URI of a page? What you want to do is link to pages. In his browser, you find an interesting page, you just mark it, and then attach that mark to a page of your own. Yes, live local web page editing, in the browser. Here's the unfortunately somewhat complex explanation (it's much easier when you see it being done):
The "Link" menu you can see. "Mark all" would remember the URI of where you were. "MArk selection" would make an anchor (link target) for the selected text, give it an ID, and remember the URI of that fragment. "Link to Marked" would make a link from the current selection to whatever URI you had last marked. So making a link involved browsing to somewhere interesting, hitting Command/M, going to the document you were writing and selecting some text, and hitting Command/L. "Link to new" would create a new window, prompt for a URI (ugh - it should have made one up!) and make a link from the selection to the new document. You never saw the URIs - you could of course always find documents by following the link to them.
From Tim Berners-Lee: WorldWideWeb, the first Web client.
Unfortunately, while the Mosaic and Netscape browsers were a revolution, they lacked this concept of a smooth interaction between web page browsing and web page editing, between link discovery and link creation. So we got all tangled up in the world of manually typing <A HREF and in the address bar.
We need to objectize our data. We need to hide the HTML and Web Services and Ajax and everything. That means we should be able to apply useful methods to data items. (Which of course means we need to identify data within web pages.)
Lorcan Dempsey has talked about a similar idea, with his concept of making data work harder, although he's talking a lot about getting more value from existing databases, rather than individual data items.
Like a web page URL. Standing alone, as a chunk of data, it is useless. You should be able to click on a URL and discover all sorts of useful things, by applying methods to it: what pages link here? how high does this URL rank? who has bookmarked this URL? put this URL in Furl or delicious. etc. (And of course you shouldn't really care about the URL at all, it's "this page" that you're interested in.)
Similarly, an ISBN sitting on a page is useless. It needs to have methods: find this in my library. find related ISBNs. show me a book cover. etc.
We need to work with objects, not with data. That's the next level of web interaction. If you think about it, a lot of the Firefox extensions and toolkits we are seeing are about discovering useful pieces of data on a web page, and providing integrators with methods they can apply (e.g. "run this through my organization's OpenURL resolver").
We have been I think limited by our tools - the web browsers themselves. The focus was on the web browser as purely an accurate rendering engine, so we got all concerned about HTML and CSS and such. Whereas really we shouldn't care about that layer at all. We should be working with the information on web pages, not worrying how they are coded. We should be able to manipulate web page objects in useful ways.
For example, if you click and drag a web link in Safari on the Mac, you'll see that it turns into a bubble that is the link - it knows it is an object you want to work with. Similarly in Flock, we're beginning to see more sophisticated ways of interacting with images and bookmarks.
There is still a long way to go though. I should be able to visit Flickr and drag images onto my blog and drag links from my blog onto Flickr, and all the underlying infrastructure should just work. Drag and drop and right-click interfaces are powerful ways to manipulate objects that have not been fully utilized. Applications should understand objects and do intelligent things when presented with them, rather than constantly requiring very technical, low-level manipulations by the users. "Cut this, make an A HREF, adjust the stylesheet, FTP, preview" has gotten us pretty far, but it's neither a scalable nor a productive paradigm.
Also, think about tags, and the hacks we use to get them on our pages. One other key aspect of object models is extensibility. You should be able to add and integrate new features. Meanwhile, our web browsers know nothing about tags. You can't drag and drop them, you can't do searches on them, to the web browser they are just dead data on the page. We need to find an extensibility model so that as we add new objects to our pages, they can be manipulated usefully.
Well, not quite one year, but close enough - the initiating event for this blog was Internet Librarian 2004, which I posted about (giving a list of bloggers) on November 20, 2004. The postings had actually been created the week before, but migrated here after I realized it made sense to separate work blogging out into its own space.
I started blogging in general in 2001, but it wasn't until IL2004 that I found I had a bunch of content related to work, and a community of Internet Librarians to share my information with. It has been an interesting experience over the past year, interacting with a very welcoming community of information professionals.
I started out just doing the "blogging as better bookmarking" thing, but then with tools like Furl it seemed to make sense to move that activity off the main blog, and to refocus on more in-depth commentary or specialized information that I can provide. It makes for much more challenging content creation, but it has turned out to be invaluable for the technology exploration and strategizing I need to do for work anyway.
Anyway, thanks to all my readers, commenters and trackbackers for making it an interesting year of weaving a piece of the library web.
I am not at Internet Librarian 2005.
However, there is a hefty list of IL05 Bloggers you can follow.
They have also proclaimed an official tag, which as you have probably already guessed is IL05.
(Last year I unofficially used IL2004, because as a computer science guy I worry about what will happen when the tags collide at Internet Librarian 2105.)
I was going to give a spiel about making a Blogdigger Group but you can't get a jump on the Internet Librarians these days; it has already been done (see above).
I also made one myself for Special Libraries Association 2005: SLA2005.
It can be a bit of a challenge to descry where Canada's national science policy direction is going.
I made a previous attempt in my June 2005 posting "overview of scholarly communication in Canada", which included excerpts from a presentation by Dr. Carty, the science advisor to the Prime Minister. His main two points were
- “Open Access” – Science Meets The Information Commons
- Harnessing The New Power Of Data
In his latest writing "A global information system needs a culture of sharing" in the November 2005 issue of University Affairs, Dr. Carty continues to describe a bold and consistent vision of a very open scholarly communications system. It will be interesting to see how this vision translates into policy.
So, what is Canada’s vision for a 21st-century global system for disseminating and communicating research data? Above all, our goal must be to maximize the impact of research for societies everywhere, not just the developed world. People in developing nations must be able to access and contribute to the vitality of the global research information and communications system. An open-access philosophy is critical to the system’s success: if research findings and knowledge are to be built upon and used by other scientists, then this knowledge must be widely available on the web, not just stored in published journals that are often expensive and not universally available.
From a Canadian perspective, a 21st century research communications system would share certain attributes. It would:
- take full advantage of the enormous potential of new information and communication technologies;
- be capable of handling an unprecedented flow of information in a wide variety of formats;
- bring Canadian research knowledge to the world and bring the world’s research knowledge to Canada;
- be accessible by all Canadians, in all sectors, ensuring that public investment in scientific research leads to wealth creation and improvements in social and cultural well-being.
With this type of system a researcher could access, from any corner of the globe, the full texts of relevant journal articles; a comprehensive set of monographs and theses; research data sets that underlie published outcomes; research reports and non-peer-reviewed research materials from both academia and government; and the electronic tools necessary to manage this volume of material.
Creating a system with these attributes is no longer just a question of developing appropriate technologies; for the most part these already exist. Rather, it’s a matter of building, integrating and improving the technical infrastructure, operational standards, research support systems, regulations and institutional roles and responsibilities. It’s also a matter of nurturing a culture of open access and sharing, beyond what researchers have ever embraced.
via Open Access News
This is great to see: a framework with all the pieces you need to build Firefox extensions for your own libraries.
Ongoing We are currently looking for libraries that are interested in adapting, evaluating and possibly deploying LibX. Librarians: if your library uses Millenium, Horizon, or Voyager, getting your own edition of LibX can take as little as 15 mins, and we are willing to help you set it up. We will add other OPACs as there is interest.
LibX was created by Annette Bailey and Godmar Back.
I think this is a great development.
Date: Saturday, November 26, 2005
A meeting in the vein of previous Foo and Bar Camps, for developers and interested parties to discuss Web 2.0 stuff and such.
The Librarian List: Librarian blogs listed by PubSub LinkRank.
I'm currently ranked #6, which is quite ridiculous in my opinion. #600 is probably more accurate.
It's based on flawed intake: it's showing 3 inlinks from the October 13th, but two of them are old links
UPDATE 2005-Nov-07: Librarian List now officially released
The NRC's annual report for 2004-2005 is out, with associated blurbage about my organization.
This bit is kind of interesting:
The on-line versions of NRC Research Press journals are available free to all Canadian readers. In 2004-2005, use of the Press' electronic journals increased dramatically as a result of indexing by the internet search engine Google. Readers downloaded almost 1,000,000 articles, representing an increase of 138% over the previous year's 420,000.
11/09/2005 (26 days)
Ottawa - Ontario
This is a continuing position.
Under the direction of the Head, Monographs or the Head, Serials Cataloguing, analyzes, catalogues and classifies according to international and local standards a variety of publications in various formats and languages in scientific, technical and medical subject areas for CISTI and the NRC Information Centres (NICs). ...
Master's degree in library science or in library and information science, preferably preceded by a degree in science or engineering. ...
Bilingual imperative CBC/CBC
The video iPod was foreseeable, here's what I wrote (elsewhere) in March 2005:
iTunes Video Store
I wondered why Apple put trailers inside of iTunes.
But I have figured it out.
Apple is building the platform for selling video downloads the same way they sell music downloads.
Why do they host trailers on their site at all?
Hmm, sort of a daily infrastructure test...
Why video inside iTunes? iTunes is crossplatform... sales platform...
Steve has said that portable video (the "iPod Video") doesn't make sense. (There is already portable Mac video, it's called an iBook or PowerBook.) But downloading video to your desktop... or say the Mac Mini... Send the video over wireless to your TV like they do with the AirPort Express audio thing...
Notice how every time Steve does a talk he goes on and on about some new video thing the Mac can do? I thought he was wearing his Pixar hat, but he's actually building the Mac video infrastructure.
The man's a genius.
I still think video on the iPod is really the least important bit of the whole thing.
There's still the device challenge which they haven't overcome: we want devices to be as small as possible, but we want displays to be as big as possible, and interfaces to be a reasonable size.
I had thought they might at least do a 16:9 wide-angle display by turning the display sideways on the iPod and having a much smaller control area.
Plus which, mobile video seems a bit dubious to me. The only time I use it is when I watch stuff on my laptop on airplanes.
This is the most interesting video part to me, which I haven't seen widely mentioned:
purchase ad-free episodes of your favorite ABC or Disney television shows and watch them on the go.
Here's what it says on the iTunes Videos page
Easy As It Looks
The virtual shelves of the iTunes Music Store now feature a video section. Music videos, Pixar shorts and select ABC and Disney television shows live right alongside songs, podcasts and audiobooks. Browse featured listings or search the archive to find just what you want, then click to buy. Once you do, you get stutter-free, ad-free video delivered directly to your desktop. From there, the sky’s the limit, because you own purchased video forever. Watch as many times as you choose, share between five computers, burn to data CDs or sync to the new iPod. Instant gratification never looked better. ...
In addition to music videos, the iTunes Music Store also features select Disney and ABC television shows, ready to download. For $1.99, you can own the latest episode just one day after it airs. With full seasons of “Lost,” “Desperate Housewives” and “That’s So Raven,” plus episodes of new shows like “Night Stalker” and “The Suite Life of Zack & Cody” at your fingertips, you’ll never miss your favorites. Plus, you can sync shows to the new iPod and watch them on the go.
I strongly agree with Steve's "people want to own, not rent" model.
However, it's worth pointing out... this is the opposite model from the library model.
In any case, while the video is big news, there are also other developments that are just as important.
They are making iTunes into a community:
Send songs, albums, playlists, even music videos and TV episodes to anyone with an email address. ...
Get personalized recommendations, courtesy of the iTunes user community and a host of in-house music experts. Hear what's in store for you. ...
Post your own customer reviews and ratings, and read what over 10 million iTunes users worldwide have to say about what's playing on iTunes.
from iTunes - Download
Building communities is where the web action is. You have to give people a reason to come to your site.
(No TV downloads available from the Canadian store yet. The videos are all 320x240. This posting is a bit of a cross-post.)