The US Presidential election was covered at length on the Web, and there’s not much that I can add to it. One thing did catch my attention though: a variety of ways of visualising the results at state and county levels. It is very impressive.
The people at LibraryThing - management and users - are becoming a little perturbed at an announcement from OCLC that cataloguing downloaded from their WorldCat service will have certain restrictions put upon it. One of these restrictions may be that for-profit organisations (like LibraryThing) won’t be able to use this metadata freely. You can see why they’d be concerned. I’ll wait until the dust settles a bit, but at this stage I can see OCLC’s point. Much as I like LT, it is a competitor to the boyos from Dublin OH.
Update: While I can see OCLC’s point, I definitely don’t agree with it, and neither do most of the commentators discussing the issue. One of the better responses is this one from Rob Styles (an employee of Talis, but that doesn’t at all lessen the value of his contribution).
You may not have seen it (I certainly don’t recall anything of the sort) but Google does us all a favour by flagging any site that they find containing malware, with a note that “This site may harm your computer”. In a post to their official blog they point out how this might come about and what you as a webmaster can do about it, if you’re unlucky enough to be compromised. They also discuss whether or not they should inform the webmaster before they flag the site in their search results, and make a convincing case that the user’s rights should prevail. I still marvel at the processing power that they have at their disposal to do this sort of detection.
There’s a service called Mosio, where users send SMS messages to ask questions of the community (before you rush to try it, the bad news is that it’s not available here yet). Looking at their website, some of the questions are pretty basic: what is veal; how many cups in a pint; is infidel a word? You probably don’t need a particularly wise crowd to manage those. But for slightly more difficult questions, they’ve introduced a new service called Text a Librarian, which takes a little time to understand but seems to involve your library paying for the service that your patrons can use to submit questions to reference staff. I’ll be interested to see how it takes off.
As an old ABS hand, I’ve always believed that you can’t have too many statistics (a view that appears not to be held by the current government). That’s why I was so impressed by the report recently released by Opera, and not only because it reinforced a few of my prejudices - although that’s not really a bad thing. Some of the highlights are: around 4% of pages had valid HTML, around 33% had Flash content (this did vary considerably between countries) and 80% used CSS. Further on the HTML front: only 50% of those displaying “valid HTML” badges were actually entitled to that claim.
I was at a meeting of the AGLS Working Group this week; it’s a jolly little gathering that happens twice a year, run very effectively by Adrian Cunningham from the National Archives, and he pointed out that the AGLS metadata standard was just ten years old. Yes, back in September 1998, it came into being after a busy year of development. It started off being based closely on the Dublin Core elements, but pretty much from the start there was a degree of divergence. At first, the AGLS meant Australian Government Locator Service (a passing reference to the GILS system from the US Government) and it was in the Commonwealth Government (with some interest in the states and territories) that it began to have an influence.
Within a couple of years though, the Working Group realised that there would be benefit in a wider adoption of the product, and negotiations began with Standards Australia for AGLS (now known only by the initials) to become a standard for all Australians, whether in Government or not, and so we saw AS 5044 emerge in 2002. And there it starts to get interesting, because I think it’s reasonable to want to know if it’s been all worthwhile. Is anyone outside government using it? Standards Australia have been asked about the number of copies of the Standard that have been sold, but apparently that’s a closely guarded secret.
It could well be that no copies of this document have been sold at all - possibly because it’s always been available for download from the AGLS part of the NAA website. But I guess we’ll never know.
But to return to the subject of how much it’s used, we really need to know. In the beginning, there was a general expectation that use of the standard would result in a golden age of findability. Well it happened - but not in the way that we expected. The search engines (which in 1998 were not as pervasive or as powerful as they are now) soon lost interest in using META tags as a source of indexing material, when unscrupulous types began to put all sorts of irrelevant keywords in the META tags. And of course, only the zealots actually took the time to put quality metadata on their sites. But at the same time, the search engines kept improving their indexing and search capabilities so that we can now get the sort of results we hoped for, just from the page content itself.
Metadata is still being used, but mostly in more specialised applications like service portals and places like Healthinsite that can afford professional cataloguers. I doubt if it will ever make a mark outside those specific applications. So it goes.
Andy Clarke in his blog raises a good point about the differences between browsers. Web developers are familar with the problem of trying to code so that different browsers (and different versions of browsers) will display pages exactly the same: it’s often very time-consuming. What Andy’s asking is “does it really matter?” And the answer is that if the client thinks it matters, then it does. So there. But maybe it’s just a matter of educating the client about the extra work involved and the little extra benefit of exact duplication of the appearance. It’s certainly worth trying.
The National Library, along with several other major institutions around the world, is working on an ambitious but very worthy project to define the business requirements for “a thoroughly re-examined model of library operations”. The aim is to have at the end of the project a design for a next-generation library system using Service Oriented Architecture concepts. Just as important as this goal is the development of a committed community of interest that will have contributed to the specification. Have a look at the OLE Project site.
I suppose that someone far smarter than I can do a proper analysis of the whole blog phenomenon and predict its rise and fall (and rise), a bit like the famous Gartner hype cycle, but at present I’m just watching it with quiet amusement. Some of the luminaries in Web Issues are saying that they won’t accept any more comments on their posts, because it’s too hard to manage them, and that’s not what the blog is for, and so on and on. Fortunately, that not the case with many of the most useful blogs around (like Matt Cutts’) but it just shows me how much our little endeavours are still in their infancy.
A couple of weeks ago, the Disability Discrimination Commissioner said that it was time to start naming and shaming government websites that did not provide material in accessible formats. Today he did just that, identifying that the draft report from the Garnaut Climate Change Review had been published only in PDF (and not even tagged PDF, I noticed). They must have responded pretty quickly, because by the time I checked there was a note on accessibility that apologised for the PDF and offered to provide upon request the document in an alternative format.