January 2006 Archives

Spent the last week trying to dig out from having been gone for two weeks - I'm glad to not be traveling for work again until March.

I'm not much of a football fan, but we enjoyed watching the Seahawks win the NFC championship yesterday - a Seattle team has never gone to the Superbowl before, so it's all very exciting (though not exciting enough to skip a Sunday of skiing with my son in two weeks).

We saw the Apple Intel cleanroom TV ad during the game - the folks watching at my house, who are not following the whole technology story of this switch, enjoyed the ad a lot.

I think Ted leung has the sanest take on this that I've seen so far:

It looks like Apple did the most expedient thing that it could, which is to take an Intel 945PM chipset and stick it into a PowerBook case, and add a small number bells and whistles (like the built in iSight and remote control). That explains the ExpressCard slot, and the FW400. If I didn't have to measure the MacBook Pro against something like the Lenovo T60 (see preview), which has 5 hours of battery life with a 2.16GHz Core Duo T2600 in a 4.8lb package, I might be happy. But this is hardly the top to bottom revamp of the pro notebook line that you'd expect for the Intel transition. And let's not even discuss the name.

That's what it looks like to me too - that Apple did what they could to get some Intel-based product on the street as quickly as possible. Craig Wood's got some performance testing data that shows that the new MacBook Pro is quite a bit quicker than the G4 Powerbooks, though nowhere near the four times as fast as Apple is touting.

My primary uses of a laptop don't include a lot of CPU-intensive tasks - for me it's a web-browsing, blog-writing, emailing, presentation-making, IMing machine. So I'm not jumping on the upgrade bandwagon just yet (though the built-in ISight camera and Front Row remote software are a nice touch). I'm still hoping for something lighter, cooler, and with longer battery life from Apple.

Technorati Tags: apple

Chuck Norris from EVDB is talking about how to store detailed location information in calendaring objects.

iCalendar specifies just a single text field that contains location - that's not specific enough for much event calendaring.

They are proposing an extension to iCalendar to handle more structured location data. The concept is that there is a location ID that is referred to in the Location field of an iCalendar file, then there is a Location block that has contains all the detailed data on that specific location.

There's some discussion about why not to just embed Vcard into iCalendar - apparently there are some syntax differences that would make iCalendar parsers choke on vCard data. Another reason is that vCard is really oriented towards people, not location (though it contains some location data).

Technorati Tags: Calconnect, Calendaring

I was sad to read Paul Andrews' final column for the Seattle Times on Monday. Paul has been a consistently sane voice for the knowledgeable users of information technology in a world that has gotten progressively less sane.

Paul's joining the folks at Green For Good - I wish him all the best in his new endeavors!

Technorati Tags: journalism

I don't see any reference to this on the US Nikon site, but in the UK:

Following the success of our digital line-up over the last seven years, which has resulted in more than 95% of Nikon’s UK business being within the digital area, Nikon Corporation has made the decision to focus management resources on digital cameras in place of film cameras. This decision will allow Nikon to continue to develop products that match the demands of an increasingly competitive market place.
...
As a result of the new strategy Nikon will discontinue production of all lenses for large format cameras and enlarging lenses with sales of these products ceasing as soon as they run out of stock. This also applies to most of our film camera bodies, interchangeable manual focus lenses and related accessories. Although Nikon anticipates that the products will still be in retail distribution up to Summer 2006.

Mike Lambert from the Open Group is talking this morning about a challenge for achieving federated free/busy. The Open Group's Messaging Forum is the descendant of the Electronic Mail Association.

The Vendor Challenge approach is a method for bringing vendors to work on real problems presented by users. The problem is defined in terms of a Business Scenario (Use Case) which defines the problem, the business and technology environment, measures of success for the solution, and the constraints. Vendors of relevant products are invited to accept the challenge. On completion, vondors demonstrate the solution.

Afterwards they work to ensure that the solution is deployed in commercially available products in a consistent form. If the activity has identified the need for standardization work (which often happens) they try to make sure that happens in the proper forum. They also have a certification program to recognize and promote products that embody the solution.

Wen Feng from Boeing is presenting the Federated Free/Busy challenge. Why bring this question at this time? Boeing's aircraft programs are increasingly collaborative between Boeing and other partners. The 787 program is being designed in a global collaborative environment - brings up issues of how to exchange data securely, how to work collaboratively. This program has around 300 first tier partner companies. The problem of getting people together in this virtual collaborative environment brings up the issue of scheduling.

Even within Boeing they have multiple calendaring systems, despite concerted efforts to standardize.

Currently they've built an internal system that allows authenticated users to use a web page to retrieve a table of free/busy time for anyone with a Boeing email address from their Exchange servers. But they need to broaden it outside the company - the Boeing free/busy is probably only 30% of the problem in the collaborative environment.

Current calendaring systems do not all contain sufficient information to solve the problem - for instance, if I'm usually in Pacific timezone, how does the free/busy time reflect the fact that for a week I'm travelling in Asia and am on a different time zone?

Free time is not always reliable - just because you're not busy doesn't mean you're available.

Few organizations have corporate policies on updating of calendar information (e.g. you must keep your calendar updated).

The challenge:

By the end of Q2 2006 there should be a real-time mechanism that is able:
- to extract and c ollate/display free/busy information
- from at least 3 major groupware packages
- using open standard protocols
- for a constrained list of named attendees
- and a constrained list of times.

"as large corporations, we do not use proprietary protocols."

Technorati Tags: Calconnect, Calendaring, scheduling

We've just seen a demo of interoperability in calendaring between the Mozilla Lightning calendar in Thunderbird, the Oracle Calendar client, interchanging events across the RPI CalDAV server. Is that cool or what?

Technorati Tags: Calconnect, Calendaring

This week I'm at the Calconnect roundtable, which is being hosted by Novell in Provo, Utah.

We're starting off the roundtable with a report from the technical committee on time zones. That group is proposing the establishment of a formal timezone registry (for reasons why this is necessary, see my post from October 30).

The whole group here think that's a great idea, and votes in favor of it.

Technorati Tags: Calendaring, Calconnect

Yesterday afternoon the Westwood Advisory Council, the CSG group that works closely with OSAF on development of Chandler for higher education, met.

Mitch Kapor and Katie Parlante laid out the current status of Chandler work and the plans for moving forward. Katie's slides are here.

One of the important themes was the emergence of the OSAF work as being about a whole ecology of related projects, including Chandler (the desktop PIM client), Cosmo (the *DAV server for PIM data), Scooby (the web PIM client), and others, including work on clients for mobile devices.

OSAF is very interested in fostering community work on Chandler and related projects, and we talked a bit about how to encourage people within higher ed institutions to participate. If anyone is interested in finding out more on how to take part, drop me a note.

Technorati Tags: Calendaring, CSG-Winter-2006, PIM, OSAF

Roger Hanson from Wisconsin is talking about their progress on integrated approaches to PIMs.

User Services came to them saying "there's a stampede for Exchange!" They started looking at Exchange. Why are people using Exchange? They went to talk to over 30 places that are using it - most were using as a simple mail system - not using calendar or public folders. They looked at what it would take to do Exchange on a campus basis - the more they looked at it the less they wanted to do it, because they didn't have a whole Windows infrastructure to support it.

They looked at other options - Notes, Groupwise, Zimbra, Scalix, etc.

But what is it that people really want? Can they achieve the kind of collaboration people want without Exchange?

They already had widespread and growing use of campus email (imap) and growing acceptance of the campus calendar system (Oracle).

How could they achieve the kind of integrated collaboration with tools they have now? What do you see on the desktop in Exchange? You see Outlook - can they achieve that integration? Answer should be yes - with the Oracle Outlook Connector and exposing Xythos public folders in Outlook.

Year ago August did initial deployment of Xythos - now have over 28,000 accounts - rapid growth in both student and fac/staff.

The are looking at how people use personal information - students are typically the biggest users of web clients. Students messaging paradigm is heavily influenced by IM - they want to start a message process by clicking on a person, rather than by clicking "new message" then entering an addressee.

They found that gmail is not people's favorite web mail client - Yahoo! is more liked.

Mobile devices is rapidly becoming more important - not "high profile" devices like Blackberries and Treos, but web-enabled cell phones with simple messaging clients.

Now Tim Gleason from Harvard is sharing some of their experiences.

Tim notes that timeliness of data is increasingly important. New data types are being introduced (SMS, IM, Audio and Video messaging).

Devices have improved, users are pressing for more and better access to data, real-time access is important, connectivity methods have improved (but have introduced managability challenges)

Data synchronization - is generally limited to proprietary approaches, ranging from cradle sync to over-the-air approaches.

At Harvard the Med school implemented Exchange in 2002. Word started getting out about the level of functionality it provided. At around the same time Harvard realized that they were at the end of the road with Meeting Maker, and were considering refreshing their imap servers (see Fall 2004 CSG materials for details).

They embarked on an Exchange implementation project, after an 18 month evaluation phase. "It's not as simple as you think". Implementation for 4000 administrative staff. They want to support a broader range of devices than just Blackberries.

They're still thinking about cost models - it looks like Exchange costs will come to between $8-9 per month per user. Over-the-air sync is expensive, due to cell phone data time charges.

There are lots of challenges in supporting multiple devices, and that support is expensive.

Technorati Tags: CSG-Winter-2006, PIM

Mimi talked a lot about focus, and integration of PIM data allowing you to focus - answering the question "what should I be doing right now?" instead of "where should I look for things I should be doing now?"

Mimi notes that task lists are themselves artifacts of a pre-digital era. For information workers, the item itself (the draft of an email, a meeting that needs scheduling, the document you need to work on) should be on the list - just like we leave the video to be returned in the hallway, instead of writing "return video" on a list. That's what the Chandler "stamping" concept is about.

Chandler has "bi-directional" references, where each item that belongs to a collection contains data indicating the collection(s) it belongs to, and the collection knows what items it contains.

Three categories: Tagging, Capturing, and Monitoring. Just as items have a life cycle, so do collections - you don't realize the first time you see a topic, you don't realize that it will be a project (or collection), so maybe you tag it. But later, when more items come in on the same topic, you want to change that tag into a collection without having to go back and find all those items and create a folder and drag them into it.

In response to a question Mimi brings up the idea of collaborative triage of a shared mail list - an intriguing concept.

- Update - Mimi's slides are online here.

Technorati Tags: CSG-Winter-2006, PIM

This morning's workshop is on Personal Information Management. I'm a participant in the workshop and am still working on putting my presentation together, so won't be blogging it in detail... Mimi Yin from OSAF is talking about the methodology used in designing Chandler.

Bernie Gulachek is talking about an initiative to reach alignment on governance and funding at the U of Minnseota. Slides are here. He says that the model should encourage the use of core technologies by faculty, staff, and students, and not ration them out as in current recharge models - there's a slide of what those core technologies include. They're trying to establish the notion that central technology services might be better at providing utility-type IT services that are not those that offer an individual unit competitive advantage.

It's cental admin's goal to make it clear to deans what they're paying for, and to try to eliminate duplication of technology resources. They are recommending headcount-based allocation model. They base it on an annual headcount in time of faculty and staff per unit and come up with a rate per head, allocating it to colleges and admin units. Today's model is opt-in for central services - tomorrow's is a rigorous opt-out model.

Greg Jackson, from Chicago is talking about how his unit is organized. Greg notes that in the formal budget process he sees his budget request and other IT requests from the campus going in, but not other budget requests. And all he sees on output is his budget. This has led him to be a huge charge of recharge operations. For data networking they were considering using a charge rate that combines headcount with square footage. Then they realized that they should do that for phone lines too, so that's what they're now considering.

Phil Long from Yale is talking about their budget models. They used to charge by network port and email account. As of October 2004 they now charge by FTE. They estimated that 60% of the variance of costs of networks were determined by the number of drops, which roughly equated to people. They're charging $10 per month for base 10/100/wireless connection, anti-virus at less than $1 per month, imap mail services at $3.50 per month. They need to figure out how to put the help desk support into this. So in '07 they think the total cost will be about $15 per month for the bundle.

The big thing they didn't get in there was institutional security - they'll head there if they can find a way to do it.

Annie Stunden is now talking about IT service and governance at U Wisconsin (Madison). They've established (since 2001) service teams that have to define service quality indicators, assess customer needs, and develop a method to track and improve customer satisfaction. The service teams get to develop operating plans and create budgets for their service. The service team makes presentations to the budget group - the budgets are in the IT units, so it's a matrix setup. They have to track revenue to expense balancing. They have to hold quarterly reviews of the service.

Who's on the teams? There's a sponsor who's a director (there are six directors), there's a service team leader, a member of business support (to help them with financial issues like budget, unit costs, revenue calculations, etc), team members from relevant organizations (operations, developers, help desk, etc). Implementations have gone much smoother, especially as a result of including the help desk staff.

Some of the teams - calendaring and scheduling; back-up services; directory services (that's hard, it's evolving); e-learning systems; enterprise storage services; myUW-Madison; mail; etc

It's not an overhead free issue, but they've found it better.

Technorati Tags: CSG-Winter-2006, IT-Governance

I was looking at Chris Anderson's post on how many blockbuster releases there have been in the last few years (the bottom line - hardly any compared to earlier years), and a blurb from Forbes that notes that Standard & Poors initiated coverage of Warners Music Group stock with a "strong sell" opinion - the research firm said it was negative on the stock because of the health of the music business.

Then Melissa Rogers pointed out this great Groklaw rant about the DRM on Coldplay's new CD (thanks Melissa!) - more indications of an increasingly unhealthy industry (makes me glad I don't like the entirely insipid Coldplay anyway):

This CD has been manufactured for usage in regular CD players but might not play in the following players:

Except for manufacturing problems, we do not accept exchange, return or refund.

hoo-boy.

Technorati Tags: music, DRM

Bill Clebsch from Stanford is giving this presentation.

He starts with a slide saying "Funding: The Final Governance".

He notes that funding evolved from two different models:

That of Communications, which evolved from a culturally accepted phone billing model and monopoly necessity, which gave universities a major windfall, which got used to fund other programs (e.g. networking and data services, which weren't then necessities). That created within the university a false sense of what IT really costs.

Computing, otoh, had no culturally accepted billing models. The services were perceived as optional. There is a constant influx of new services which require a constant influx of new funding, and those services become necessities - but the funding was based on an optional model.

Admin Systems and Services - automation of tasks cost more than anticipated - e.g. the dispersion of PCs on the desktop was a very difficult transition. It's a problem because we never pick up the savings on the other side.

The current state is a perception that IT is too expensive without seeing hte corresponding value, and there's very little "Budget Dust" left to throw at systems. There's a tenuous link to academic program, and people perceive central IT as administrative computing (including course management systems), and all faculty mistrust everything administrative.

The objective for funding models should be to spread costs equitably across the multiple funding sorces, to link funding with perceived value, to scale IT with university growth and success, and to allow the introduction of new services.

some possible bases of funding could be based on people, facilities, machines, or usage - all of these have to be tempered by how much it costs to bill for it. It was amazing how much the support costs at Stanford went down when they went to flat-rate long distance service. B-School 101: All allocation is arbitrary, but you need one that is consistent and defensible. Several people are recommending the book "Freakonomics : A Rogue Economist Explores the Hidden Side of Everything" (Steven D. Levitt, Stephen J. Dubner)Freakonomics.

Technorati Tags: CSG-Winter-2006, IT-Governance

Richard is recapping the talk that Jonathan Murray from Microsoft gave at the ECAR meeting last month, which I blogged then.

Jonathan's slides are online here.

Alan Usas from Brown is talking about the survey of CSG schools on IT Governance - his slides are available at http://www.stonesoup.org/Meeting.next/governance.pres/usas.htm.

One finding that particularly strikes me is that those schools that think their governance process is effective more regularly include more of their constituencies in decision making processes.

There's also a telling comment that "what works for governance is very dependent on the style of the senior leadership and on the degree of centralization of the institution".

Technorati Tags: CSG-Winter-2006, IT-Governance

Tallman Trask, Executive VP of Duke, and Peter Lange, Provost, are being interviewed by Jack McCredie on IT governance and funding.

In the last plan, Duke set aside $25 million for technology innovation with no prior commitment to exactly what it's for. This came in handy, for example, in being able to quickly move forward on the iPod project. Tallman - "To watch Duke featured for 20 minutes on the CBS Morning News about what a technologically innovative place it is, you couldn't buy that kind of publicity for ten times what the iPods cost."

Jack asks about IT governance and funding issues affecting both the campus and the health system. Tallman replies that they made a very deliberate decision to run common administrative systems where the costs are shared, but that on clinical systems the health care folks are on their own.

In response to a question about how administrative systems are implemented, Tallman states that at Duke central systems are run as an "enlightened monopoly" by the central organization. He went on to note that they made a concerted effort to stamp out shadow systems and were, for the most part, successful.

The whole approach of Duke strikes me as an example of how governance works in an institution where the bulk of decision making resides in the central authorities, rather than in a more distributed manner (which Weill and Ross characterize as "feudal" organization) which seems more common in our institutions.

Technorati Tags: CSG-Winter-2006, IT-Governance

Jack McCredie (outgoing CIO) and Shel Waggener (incoming CIO) from UC Berkeley are talking about their current strategic planning effort.

Six Aspects of IT management - 1. Strategic alignment; 2. Value Delivery; Resource Management; Risk management; performance management; IT governance, which is at the center of it all.

Governance - who is decidiing, how do they decide, who gets input.

Jack asks the great question - What percentage of your acmpus community could describe the IT governance structure on your campus? Most folks answer with under 40% - which speaks to the problem. Kitty Bridges notes that most of the community couldn't describe any of the campus governance structures any better.

Tracy notes that people only ask about governance when they think it's a problem.

Jack wisely states that people's characterization of funding as a top IT issue is really a symptom of poorly understood governance.

Jack's top IT opportunities and challenges - 2006
- security and privacy
- re-invent central IT organizations
- transform teaching & learning environments
- governance and structure

Berkeley had an audit of their security and privacy effectiveness, which gave them high marks on their policies, but low marks on implementation of those policies across the campus - that's a governance issue.

The Weill & Ross Governance Model (MIT sloan Management Review - Winter 2005) identifies Key Issues for each IT Decision Area

Berkeley tried to identify where they fit along that model, as did Syracuse in a recent ECAR study.

There are some comments that our most common governance model is anarchy. Richard Katz points out the ECAR finding that CIOs who spend more time with deans and faculty leaders report more positive IT outcomes than those that spend their time with central administrators - this speaks to the feudal governance model that in fact rules our campuses, where the middle tier of a hierarchy in fact holds much of the actual power.

Berkeley came up with a good list of principles for IT.

Then they identified six critical campuswide IT issues:
Thre major areas of IT support - Teaching and Learningl Research; student experience

and across-the-board imrpovements in: security, reliability, access, privacy; IT structure, governance, funding; optimization of IT expertise.

These were identified not by IT, but by leaders campuswide.

Shel notes that simplicity, while easy to scoff at, is key to a well-functioning governance process.

Berkeley did a self-study, which came up with the following recommendations:

1. The CIO function needs to be strengthened, defined more clearly, and differentiated from the function of running IS&T.
2. The CIO should be involved in formulating all campus-level IT budget requests.
3. The CIO should be the key link between input/advice from IT stakeholders and formulation of campus-level IT budgets.
4. There should be a clear way for knolegeable faculty to interact with the CIO and for the CIO to receive expert faculty advice and draw on highly-regarded faculty partners to advocate for proposed IT investments.
5. The Berkeley campus needs to reorganize, rationalize and enable technology (and other) investments in classrooms and instructional-technology support systems.
6. The Berkeley campus needs to reorganize, rationalize and enable the provision of the IT resources that faculty, students and staff need to do their work (including responsibility for a minimum standard level of computing capability an desktop support). Shel notes that you need to be very clear what standards you're implementing so others can make their local decisions wisely based on those.
7. The Berkeley campus needs to reorganize and rationalize its approach to hiring and training professional IT staff, to encourage the development of a campus-wide community of IT professionals, and to identify and disseminate best practices.

Shel states that true 100% transparency in IT governance is the key to a successful CIO function.

Technorati Tags: CSG-Winter-2006, IT-Governance

This looks cool -

To all --   

This is a notice that the room block cut-off date has been extended till this Thursday, January 5th 2006, for the jointly hosted Internet2 and New World Symphony workshop.  Be sure to make your travel plans in the next few days!  For registration and hotel details, see:  http://events.internet2.edu/2006/NWS/index2006.html

What:     Performance and Master Class Production Workshop
When:   30 January - 1 February, 2006
Where:  New World Symphony, Miami Beach, FL

Internet2 and the New World Symphony are hosting a hands-on audio/video production workshop on 30 January--1 February 2006 on the campus of the New World Symphony in Miami Beach, FL. The workshop will emphasize real-time experience connecting with remote sites over Internet2 advanced networks, setting up incoming and outgoing audio, creating a multi-camera shoot, placing lights and projectors and operating codecs. This year's workshop will include demonstrations of state of the art codecs for producing events, will explore in depth the elements needed to do everything from simple one-on-one interactions to larger scale stage productions, and will highlight the latest network testing tools developed by Internet2. The workshop is designed for an audience of technologists responsible for setting up distance learning and remote interactive media events. Administrators, deans, CIOs, network and audio engineers, and other hands-on technicians will also find this workshop useful and are encouraged to attend.

Ann Doyle
Arts & Humanities Initiatives
Internet2
(734) 352-7011

Technorati Tags: music, Internet2

I'm in Durham, NC for the Winter meeting of the Common Solutions Group. I'm a little toasty, having left Seattle at 11:30 last night and arrived at Raleigh/Durham at 9 am this morning. But getting in the rental car I found three jazz stations at the bottom end of the dial, so this is obviously a very civilized place.

This afternoon we're meeting with Oracle to discuss the Oracle Collaboration Suite and how it fits into our higher education environment. This meeting is covered by a non-disclosure agreement, so don't expect much blog coverage.

Tomorrow's long workshop will be on Governance and funding of IT functions in research universities, and Thursday's short workshop is on PIMs on Parade.

Technorati Tags: Calendaring, CSG