Working Ideas

Tantek Çelik, of Technorati and microformats.org ( and a css / web standards guru :o), posts this microformat proposal. I'd been thinking about similar issues just moments ago, having posted an event-news item in ossite's open-news blog and wondered about sucking the event information out of it for calendars...

The iCalendar standard (RFC2445), has been broadly interoperably implemented (e.g. Apple's "iCal" application built into MacOSX).

In addition, bloggers often discuss events on their blogs -- upcoming events, writeups of past events, etc. With just a tad bit of structure, bloggers can discuss events in their blog(s) in such a way that spiders and other aggregators can retrieve such events, automatically convert them to iCalendar, and use them in any iCalendar application or service.

This specification introduces the hCalendar format, which is a 1:1 representation of the aforementioned iCalendar standard, in semantic XHTML. Bloggers can both embed hCalendar events directly in their web pages, and style them with CSS to make them appear as desired. In addition, hCalendar enables applications to retrieve information about such events directly from web pages without having to reference a separate file.

He notes as well that:

The iCalendar standard (RFC2445) forms the basis of hCalendar.

Note: the editor and authors of this specification are tracking the "iCal-Basic" effort and intend to base the core hCalendar profile on iCal-Basic. See references for a link to the current draft.

The basic format of hCalendar is to use iCalendar object/property names in lower-case for class names, and to map the nesting of iCalendar objects directly into nested XHTML.

While it's fresh in my mind, here's a very quick attempt to summarise the Q & A from presentation I "phoned in" to the Aarhus meeting this morning.

NB - I will try to get some time to reflect-on the and improve the summaries. At least by noting them here they'll be recorded for later reflection.

Apologies if I have mis-represnted anyone's questions. It was very difficult to hear over the telephone line, and I was very tired at the time :o)

(Morning of June 7) - Mike has just had a second bash at editing these. Still nit very well recorded, but I hope this at least gives the gist of the questions and answers.

Dai Griffiths

There is much confusion about exactly what a standard is, what various kinds of specification mena, etc. These need to be clarified.

Mike

Good point. The working group should try to develop and put up some basic definitions and clarifications. (Afterwards it occurs to me that a FAQ would be a useful goal for us to try to develop. For instance, a later question - on reflection - seems to have been assuming that all relevant standards come from the IMS Project).

Dai Griffiths

SAKAI - what do we think? (Dai says he does not have the expertise to understand the techie aspects of this but finds it interesting and would like to know our opinion)

Mike

I have followed these developments with some interestm but not closely enough to have an opinion. My instinct is that the project is intriguing but that I am skeptical, but I am not well-informed enough to be sure of anything. I will look into this more closely and email Dai with a frank answer, and try to put a more carefully considered reply up on the site.

Peter Becker

(hard for Mike to hear) VLEs recommendations - Swedish government? Possible that recommending standards would be a more effective approach than recommending particular VLEs? Possible something about the Swedish government recommending some standards in the autumn?

Mike

Yes - it might be betterto recommend a standards-based assessment rather than endorse particular VLEs. Standards might ease the decision-making by reducing the risk of lock-in to the system chosen. But we should also be wary of recommending that particular standards be prescribed. A wide range of standards - existing, proposed, and as-yet unimagined - will be required for different purposes in software for education.
NB - Please see my notes on Fiona Concannon's question below for more on the dangers of official endorsement of a standard, and how officials can miss the point of standards and do more harm than good by trying to enforce them.

Raymond Elferink

A point about the life cycle of OSS projects: (Mike sort-of butts-in and mind-reads this one as Raymond begins, so might have misunderstood a bit)
...they often start very small and very limited and focused in scope, but as more developers and users come on stream

Mike

...feature bloat starts to happen! (Mike: Have I got this right? - Raymond: yes) Standards work can amelioriate this by letting developers 'plug in' features that are not part of their core mission. We need to educated developers - this should be a high priority since it could have a direct and practical influence on the survival of open source projects.

Alexandra Todt

??? Code of practice for certified assessment by computers?

Mike

I do not understand. This sounds like a quality standards initiative. (Mike tries and fails to find it on google search while on the telephone, so I do not know - Alex can you reply to this entry with the url or project name?

Dai

Sounds like a quality standard to him as well

Dai

We are at risk of making the issues sound as if they are intrinsically technical - about XML - rather than communicating the essence of what these man to end-users - features.

Mike

VERY GOOD POINT! It would be good to communicat this more poerfully - for example, RSS means a simple feature which is available to users from a site - a news feed which they can open in an external reader. It does NOT mean an XML or RDF format to users, it means the ability to track content in an external news reader. From a developer's point of view it means that one's end-users can get content feeds from your content without you having to write the external news-reading applications - all you have to do is deliver a content stream which complies with the very simple specification(s) for feeds.

Dai

Yup. But beware - even RSS sounds very XML-like

Mike

Agreed, sadly :o) ...This sounds like an important job for the WG: advertising and communicating - making the message simple and comprehensible to different groups

Fiona Concannon

(hard for Mike to hear) Something about there being several specifications but only one standard - IEEE LOM

Mike

(Butting in at this point) Absolutely not the case; there are many relevant standards. The ditinction between standards and specifications is not all that important for us.
I REALISE AFTER THE FACT THAT FIONA WAS LIKELY SPEAKING ONLY OF 'IMS ATDNARDS' AND LIKELY MEANT THAT THERE WAS ONLY ONE 'STANDARD' INTIATED BY THE *IMS PROJECT* in which case this is another example of something to clarify: that there are many standards outside the explicitly educational remit of organistions like the IMS an IEEE LTSC which are of importance to OSS for education.

Fiona Concannon

These standards can be barriers to small open source developers. EG IEEE LOM introduces all sort of thorny issues about ontologies much worse than the simple issues of implementing it.

Mike

I take it you have raised 2 issues:

1. Officially-sanctioned standards can be a barrier to open source projects, and
2. IEEE LOM is thorny because real-world uptake is complicated by the requirement that descriptions mean something to the community on the ground in an actual context/country etc.

To answer these separately:

1. Yup eg give example (Dewi Lloyd on australian guy) of IMS Content Packaging being an unnecessary requirement before an educator-developer could allow access to his (independent anf innovative) web application within the institution's blavkboard system. Adding content-packaging support would NIT have added real value to the learner's experience in that case, but DID make a barrier for this small-scale developer. (Institutions seem to have a tendency to fundamentally misunderstand 'standards' and also to be naturally anti-innovation and anti-variety - so there are cases where institutions so fail to 'get' the point of stabndards that they are turned into prescriptive bludgeons to hammer innovation with)
   Beware of prescription - the point of standards is not to say HOW software should do something or WHAT software should do, but HOW 2 pieces of independent software CAN interact together if so desired. Sounds to me like it is IMS-originated standards in particular that you are raising this isue about - but there are many other standards of more-or-less relevance to education, and many more (we hope) to come
2. Yes, IEEE LOM has been a pain for almost everyone who has ever tried to force real-world educators to use it. But IMS and ARIADNE misunderstood in my opinion what the standards issue about describing leanring materials was. It should not have been taken to be 'what is the one universal set of fields to use?' but 'how can we get software components working together to allow people to descirbe things in their own terms but still be interoperable?' In other words - much more like recent social bookmarking systems which are very lightweight in descriptive structuring but very easy to get started with - the term 'folksonomy' as opposed to taxonomy has recently been coined for these emergent, 'folksy' classification structures.

Mike Malloch gave a brief presentation to the SIGOSSEE Project partner's meeting yesterday (speaking by telephone while Graham Attwell clicked through the visual aids). The presentation is available here in two pdf versions for downloading or viewing.

• - this is a print-suitable pdf document on a white background
• - this is a view-suitable pdf document with appearance based on the slides

Please note that the questions and answers for this session hav also been written up (roughly :o)in: this blog entry

... and Mike delivers a rant about the IMS site not 'getting' the small, loose standards that make the web work.

As an aside... I tried to find a permalink for this article on the IMS site, but was utterly shocked to discover there is none - just the 'past news' URL en bloc. Now, in my opinion trackback, RSS and other permalink-based mini-standards have done much more for education than any number of big, specifically education standards (which are what the IMS site is about). And the REST-ful web has done more than anything to make the web a useful experience for learners (REST is an architectural theory about the web which boils down to a few simple principles such as always making sure that URLs make sense and that all content can be reached from a sensible URL). I am tempted to surmise that the IMS site's prominent ignorance of these small, loose but very powerful standards is quite telling :o)

NB - this item also point to the relevant news about an abstract spec - as opposed to JAVA API - forthe Open Service Definitions.