RSS technologies: around the web

ReadWriteWeb has printed a top ten guide to RSS and Syndication Technologies for the year, including mobile and web-based RSS tools  (the article has been republished on NYTimes.com but the original has a better comments section).

A surprise omission is LazyFeed which is a great way to monitor specific topics in real time and now includes more sharing features. They’re also working on more filtering options to be rolled out.

Not sure I quite agree that Yahoo Pipes has been usurped by YSQL (do they mean  YQL?) as I’m still hoping that the two tools can co-exist but there are lots of other great RSS tools to check out.

Advertisements

Crowdsourcing answers with Quora (q&a series)

There are quite a few Question and Answer websites aiming to fill a gap left by search engine algorithms by taking a crowd sourcing approach to references services and over the coming weeks we will be reviewing a few of them.

Q&A sites are based on the premise that people answer questions better than search engines do and, I’m sure this is often true. But, as Yahoo! Answers regularly demonstrates, the wisdom of crowds is no sure thing. How do you ensure the quality of answers, or even the questions themselves?

We will be looking at how these services approach issues such as these, starting with Quora.

organize quora
Quora ontology

There are a number of Question and Answer websites aiming to fill a gap left by search engine algorithms by taking a crowdsourcing approach to references services and over the coming weeks we will be reviewing a few of them.

Q&A sites are based on the premise that people answer questions better than search engines do and I’m sure this is often true. But, as Yahoo! Answers regularly demonstrates, the wisdom of crowds is not guaranteed. How do you ensure the quality of answers, or even the questions themselves?

We will be looking at how these services approach issues such as these, starting with Quora. Continue reading “Crowdsourcing answers with Quora (q&a series)”

Google Instant: initial impressions of predictive search

Google Instant is the new  Google search feature, described as ‘search before you type’. It is a prediction-based search giving real-time results. And while the technology behind it sounds impressive (new caching systems and optimisation of page-rendering JavaScript – details not forthcoming), I’m not really sure I see the benefit of Google trying to predict my search query. Coupled with the search suggestions already provided, it risks making what was a nice clean search environment a bit too busy.

It currently works with the following browsers:  Chrome v5/6, Firefox v3, Safari v5 for Mac and Internet Explorer v8. The main benefit seems to be speed in accessing results but there haven’t been any claims of a positive impact on accuracy. Saving between two and five seconds per search really wasn’t top of my list of priorities. Having said that, I’m curious to try it out on Google Scholar and am already imagining how predictive search might be realised in library catalogues.

You do have the option to turn the feature off from your preferences, though Google Instant is only currently available from the Google.com domain and if you are signed into a Google Account. Personally, I tend to use my browser’s search box rather than navigating to the search page anyway so the impact on my search habits won’t be much.

The mobile version is expected soon, which is an environment where speed can have a big impact and I’m curious as to hear more about the reasoning behind this development and where it will go from here. As a first impression though, it made me wish Google adopted more of the mantra,  if it ain’t broke..