Archive for the ‘feeds’ Category

Google Reader redux

Friday, September 29th, 2006

Google’s feed reader has undergone a radical redesign. So radical it now resembles every other feed reader out there. The previous incarnation was loved by a few and disliked by many. It was an interesting experiment but was not succesful, unlike the GMail interface.

At first glance it is a Bloglines copy; 2 panes showing feeds on one side and items on the other. You click a feed or folder and all the unread items are displayed in a river of news down the right. You can organise by folder, you can star items, email them, share (more on that later) and add tags. Nothing exceptional and certainly not stunning (I am worried when people call a minor step forward as “stunning.”)

Google Reader redux

There are some good features though and I imagine Google Reader will follow GMail in being an incrementally improving product.

The share feature lets you mark an item as shared and this is then added to a publically available web-page which has its own Atom feed. This is useful as it lets you broadcast to the world any items you find interesting.

It is limited though in that you can’t branch the sharing out to different groups. Thankfully the tagging feature comes into play here. You can add tags to any item and then make that tag public. You can then share the public tag link with other Google Reader users and they can subscribe to it. You could tag items to target specific groups of friends and co-workers. This is a very useful feature though I’d prefer if it displayed as the sharing feature does as currently the public tags require you to be using the Google Reader.

Another novel feature is that, unlike Bloglines where clicking a feed or folder marks all unread items in it as read, Google Reader will track your scrolling and only mark as read the items you have scrolled past. This makes it very easy to start reading a list of unread items, close the reader and then come back to where you last where.

Starring like in GMail works well, useful for coming back to items later to read. The feed management interface is a bit clunky but works well enough for now.

One thing I have noticed is that Bloglines seems to be faster both in using and in noticing updates in feeds. Hopefully Google Reader catches up in those two regards.

All in all it is Bloglines plus some useful features and mechanisms. I like it but feel it is hardly ground breaking and that there is so much more potential in aggregator applications. I will be swapping to Google Reader not because it is stunning but because it is mildly better.

The Times Reader

Thursday, September 28th, 2006

The Times Reader is an interesting project between the New York Times and Microsoft. For Microsoft it is a technology demonstration, in this case .NET 3.0 (WinFX really) and for the Times it is an exploration of delivering readable content to screens. The web and even most desktop software makes for a bad reading experience. Low DPI, wide blocks of text, poor fonts and, especially on the web, little if any typograhic control.

The Times Reader aims to make reading on the screen easy on the eyes. Using the Microsoft technology it offers the Times’ fonts, builds in readable columns and uses up all the screen real estate it can to present what looks very much like a print newspaper on your screen.

Here are a few screenshots of it in action though I recommend you download it and give it a go to see the layout technology at work (it is a Windows XP/Windows Vista only app, no Mac or Linux versions.)

Untitled-2

When you first load it the app downloads a complete set of articles from the server. It then periodically synchronises allowing you to later use the app and read articles offline. As you can see it mimics a newspaper layout. On the right you will see some articles are just being displayed as headlines. If I made the window bigger the summaries of each would start to be displayed and the layout would change.

Along the top you have the different sections which smoothly flips the view from topic to topic.

Untitled-12

Here you can see the front page after I have resized the window to a smaller area.

Clicking a headline/article takes you to the article view:
Untitled-3

The article is nicely displayed in columns and flows around the image. You can see an advert in this screenshot but not all articles display an advert. If the full article was too long to fit on this one view then the page control (bottom right) would let me flip to the next page. It is a bit fiddly though as you have to click the little arrows rather than a bigger region.

You can also see the text-size control in the bottom left. As you adjust this the view automatically re-flows itself to fit. Honestly though the sizing is in steps and is quite jumpy. I’d have preffered something smoother.

You can go to the next article using the control near the bottom right. The arrows are once again a bit small but the transition to the next article is quite smooth.

Untitled-11

An article has several tools one of which lets you annotate with a pen. Handy if a bit limited (other users can’t see it for instance.)

Untitled-10

The search has three views one of which is this fluid diagram that shows an article with its assigned “topics” radiating out from it. Topics are much like tags.

Clicking on a topic takes you to a view which shows all articles tagged with that topic:
Untitled-9

The other search view shows a grid layout of all stories that match your search, sorted by relevance (as indicated by the litte filled in dots):
Untitled-8

Then onto the News in Pictures view which is very, very simple:
Untitled-4

It simply takes all the photos in the paper and displays them one after the other with captions. A gallery would have been ideal and also links back to the originating articles.

Untitled-6

I am not convinced of the usefulness of the What’s Read but it is an interesting view. It shows you each section with little squares that on roll-over display an article you have read. I would have liked an indicator next to each article title on any page to show that I have read it (much like links on the web change colour.)

All in all the Times Reader is interesting and works quite well. My main problem with it is that it is a separate, desktop app. bound to one OS. We need to get this technology into the browser.

Microsoft Max and RSS

Monday, September 11th, 2006

Microsoft Max

Microsoft Max started out as an interesting technology demonstration around photos. It has now strangely added RSS to its bag of tricks.

While it is not meant to become your default feed aggregator it does have some interesting techniques to watch out for. The newspaper layout is nothing new to feed aggregators but this one is done particularly well with what I think is the best flow and layout around. It also makes use of the powerful typographic rendering of WPF (Windows Presentation Foundation.) The feed list down the left is also nicely done and useful for non-power users who only have a few subscriptions. It displays the latest bit of news in the actual subscription list.

Colour your updates

Wednesday, August 30th, 2006

newzie

Newzie has a fascinating and well implemented colour system that denotes the last update time/date of a feed. The colouring is a bit bold for my tastes but the idea is sound and I found it worked well.

Newzie itself is one of the better desktop aggregators out there.

A Unique Problem

Friday, August 18th, 2006

James Holderness writes on how various feed aggregators attempt to judge the uniqueness of items in feeds:

Detecting duplicate items in an RSS feed is something of a black art. How does one uniquely identify an item in a feed while still allowing for that item to be updated? …

I can’t say for sure what algorithms applications are using, but after running 150 tests on more than 20 different aggregators, I think have a fair idea how many of them work.

He summarises some reasonable ways of judging uniqueness and brings up good arguements for and against.

He goes on to say:

I would recommend you also include a unique link element for each item in your feed, to allow for aggregators that don’t handle guids very well. No two items should ever have the same link element,

Unfortuanately the link element is as abused as dates and GUIDs and while this principle is ideal it isn’t how many feeds are constructed. The link element is meant as a permalink to the item itself. Not to what the item is talking about or any other link. But many linkblogs will put the link they are talking about in the link element. In fact a big source of RSS, del.icio.us, gets it wrong and links to the link being mentioned. Now consider that many people on del.icio.us link to the same link and you suddenly have seperate feed items that some aggregators may treat as duplicates.

So ideally yes, GUID and link are good but in practice sadly not. This is the way of much feed parsing as we have found out in the FeedHenry.com project. Feeds need fuzzy logic to make much sense of.

(James’ article is worth reading.)

Bloglines attention

Thursday, August 10th, 2006

A feature I’d like Bloglines to implement is one that tracks the links I click on from feeds. Each time I read a post in a feed and click through to its link I’d like Bloglines to record that.

Why?

So that every now and then I can look at what feeds are generating the most number of actual clicks which is a reasonable, though not infallible, indicator of quality.

Other attention data from Bloglines would be great to help in determining which feeds are of actual value to me. Sometimes I find it hard to remember from where I read an interesting item and so find it hard culling feeds.

Tom’s Feed management service RFP

Friday, August 4th, 2006

Tom Carroll outlines the features a feed management service must have for him to use it. I don’t know of a feed management service that comes close to what he wants. Feedburner certainly doesn’t and that is the best one out there.

The migration to the service must be seamless for my current subscribers (Yeah I know all 3).

Currently his feed is on http://www.tomcarroll.org/?feed=rss. So any feed manager will have to retain that URL which means he will have to install something on his hosting box. A WordPress plugin might work. This also means Feedburner is not suitable.

The service should provide a suite of metrics that allow me better understand who, how often, when, how long, how many times

Feedburner does this and does it quite well. How does this gel with another requirement of Tom’s though?

Integration with analytical packages say like Google analytics

Should Google Analytics handle all metrics or just match feed usage to entry pages?

In the case that I choose to leave your service, I must be able to migrate my feeds transparently and also be able to migrate my usage data.

The first bit shouldn’t be hard but the second bit, usage data, I haven’t seen anything close to a solution yet. Do any analytics packages import and export on common formats?

Configurable support for social network services like Digg or del.icio.us

As simple as “Add to Digg/del.icio.us” or something a bit more involved?

Integration to see where my feed is bookmarked and subscribed (share.opml.org, del.icio.us, Google reader ect…)

Definitley something all the aggregators need to standardise on and emit.

Support for the widest set of RSS extensions

Absolutely need this.

The rest of his list is pretty much already here:

I would like support for one click subscription to the various news readers.
The service must have tight wordpress integration.
feeds must be human readable
I would like it to support leave comment from the feed.
A strong record of availability, reliability and scalability (I might hit 4 subscribers someday)
Should support multiple feeds per site, no limit.

Feeds for forums

Tuesday, July 4th, 2006

In Threaded RSS readers for Yahoo! and Google Groups EirePreneur nails how we can use OPML to make feeds work with discussion forums:

Yahoo! will provide you with a single OPML URL for all your Groups subscriptions - each node at the top level will correspond to a Group. The next level down will be a node corresponding to each topic lead and the leaves will be feed items (individual topic posts), or sub-nodes as necessary.

While EirePreneur goes on to say that this is where feed grazers are strong I don’t see how a traditional 3-pane feed aggregator with a live OPML subscription feature won’t work either.