A changelog that rivals FeedDemon releases

One of the funniest things about most FeedDemon releases is the changelog.  Nick always includes so many fixes that his changelogs are some of the longest I’ve ever seen.

Well Ephox just one-upped that with our latest minor update release, EditLive! 6.1.  We wound up skipping our February patch release to concentrate on some bigger features, and then in the last few weeks the engineers were given free reign to attack our bug list and knock over as many as possible.

Our open bug count is now the lowest it’s been since July 2002 (and we’ve only been tracking since May 2002)!

I’m really happy with this release, some of the fixes have come out of an integration we’ve been working on (more details on that later) and these will flow through to anyone else attempting to integrate with the editor.  We haven’t been dogfooding our integration APIs as much as we have the core editor, and it showed with a number of bugs cropping up and features we were missing.

My link blog has arrived

One of the features I’ve really been looking forward to in the new version of FeedDemon is Synchronised News Bins, which can be turned into link blogs.  So now that I’m running the beta, I’ve created Andy’s Link Blog 😀

I add to it pretty regularly, like most bloggers there is a ton of stuff that I could post to my main blog but isn’t really important enough to warrant spamming you with it.  Having run this for a week already, I have posts in there dating back 3 weeks and it’s a mix of amusement, tech news and game news.  I hope you enjoy it 🙂

FeedDemon fixes support for flash videos

I installed my copy of FeedDemon 2.5 beta last weekend, and once again Nick has included a ton of the little improvements that I call “quality of life” features. My favourite of these so far is the ability to view flash videos that come through your RSS feeds 😀

For an example, I’m using some old Cute Overload posts that I’ve been saving because I was too lazy to open them in my browser (and I don’t use FeedDemon for browsing).

FD 2.1, note no flash videos

This is how FeedDemon 2.1 looks. Nothing particularly wrong with it, unless you’re lazy like me. I was OK with this, I figured the lack of flash videos was a security thing because that’s something Nick is quite concerned about.

FD 2.5, with flash vid :D

When I loaded up FeedDemon 2.5 however, I was greeted with the sight of videos – here’s the same screenshot in the new version. I’m also pleased with the UI fixes for Vista, Version 2.1 had far too much white (particularly on the menubar and toolbar).

I’m sure anyone using other readers will say “but reader x has supported this forever!” and that’s great, but I don’t use reader x. I use FeedDemon.

Newsgator now tracks feed history

And just like that, NewsGator becomes the best feed synchronisation platform I know of. I’ve said before that the only thing FeedDemon was really missing is the ability to synchronise more than just the current contents of the RSS feed. And now it does 😀

As posted in their forums today, NewsGator has extended their cache so they now keep enough history to satisfy all but the most high traffic feeds. And thanks to forward thinking by both the NG API team and the client developers, this will immediately carry through to the clients without any extra patches. This means that two serious bugs with the synchronisation are gone:

  1. When synchronising after more than a few hours, posts not on the feed (both read and unread) are downloaded
  2. If you download an unread post in two locations, and read it in one location after it’s dropped off the feed, it will now be marked as read in the second location. (This one has really, really bugged me!)

Apart from lacking nested folders, I believe that this was the only major flaw with FeedDemon. I can’t back this up with personal tests, but I’d be willing to bet that using the Newsgator platform (between FeedDemon, NetNewsWire, mobile, online, etc) will finally rival online readers like google and bloglines.

It may not sound like much at first, but don’t underestimate the annoyance of having a post that you read in one location left as unread in other locations. It means that you don’t have an accurate picture of how many unread posts are in your list, and that breeds a lack of trust in both your RSS reader and the synchronisation platform.

FeedDemon 2.1 released

It’s actually been available for a week, but I’ve been waiting for Nick to announce it before I said anything 😉

So now that Nick has not only posted the announcement but uncovered one of the best new features, I wanted to explain why I’m so happy with this release. Much like a few of Nick’s other point releases, it was hailed as being good enough for a new major version during the beta. A quick look at the features Nick has highlighted should make that obvious, the full feature
/ fix list is in the readme.

The two big features for me are the Blue Vista newspaper style and the newspaper pagination. In one fell swoop, Nick has seriously alleviated a lot of the information overload that I often struggle with. FeedDemon has been improving that situation for a while now, but these two features fix a huge problem I have with the often-touted “river of news” paradigm.

As much fun as the river is, I often leave feeds unread for days which leaves more posts to look at than I have time. This meant that instead of skimming them all, I had to go through and click on each one as I read it – I couldn’t turn on the “mark as read when changing feeds” option or I’d miss the older stuff when I wanted to stop reading. Pagination gives me the best of both worlds, and vastly improves the speed of each page load to boot.

Blue Vista on the other hand takes the expando newspaper which has been around for a year and shows a summary for each post using FeedDemon’s built-in “show post summaries” option. I never used summaries before, because it meant viewing the full post required fully opening it which interrupted the flow of the page. And I gave up on expando pretty quickly because post titles aren’t usually enough to tell me if I want to read the post or not.

The result of all this is that I can quickly skim 10 post summaries at a time (the default pagination size), expanding the ones that look interesting and then use ctrl+d (or the hidden toolbar button) to mark that chunk as read and move to the next page. You can even do all of this with keyboard shortcuts 🙂

The final thing that I must mention is a new ability to synchronise the “flag for later” state of posts. This should be the most fantastic feature since synchronisation was first introduced, unfortunately the whole sync feature is a bit crippled at the moment because Newsgator will only sync the posts currently in the feed. For low traffic feeds this is fine, but geeks like me subscribe to high-traffic feeds rendering the sync close to useless for half of my subscriptions.

Fortunately Nick has promised me that a fix is in the works for that, which I’m really looking forward to. It’s a serious hole in the feature, but once it’s plugged FeedDemon will become the true competitor to online RSS readers that Nick has been trying to achieve.

FeedDemon introduces a desktop meme tracker

FeedDemon was released a few days ago, and the speed improvements are quite good over what wasn’t a slow product to begin with. But there’s a beta feature hidden in plain sight that isn’t on any of the release notes.

A couple of months ago, Nick posted about how he started coding a feature without doing any design. The feature he was working on is currently called “popular topics” but if you’ve ever visited techmeme you’ll be very familiar with what FeedDemon comes up with. Or at least you’ll think you are.

It turns out that having a meme tracker for the feeds you subscribe to can produce some interesting results. The big issue is that some feeds either rewrite URLs to include a redirect through their server, or strip all HTML and just give you a snippet of the article. This makes it basically impossible to determine if two items link to the same article.

I have a few feeds that suffer from this, so FD doesn’t pick up any match there. I’ve also found some minor bugs, but this is a beta feature so I can understand that some things still need to be ironed out (the date filter is a bit weird, and the initial meme check only covers the last 48hrs which can be confusing).

My biggest problem though is that after suffering from information overload, I scrapped most of my high traffic feeds as well as some slower ones that I wasn’t reading. As a result, the first time I opened popular topics the only match it found was a technorati search that had picked up some posts from another feed I subscribe to!

This obviously isn’t a very good sample to work with, so I headed over to techmeme (this is what it looked like at the time) and picked some sites at random to subscribe to. The result was much better, and listed only the items from techmeme that were in the blogs I had subscribed to. Here’s a screenshot:

From this view, you can see a few things.

  1. Google is big news at the moment.
  2. I have too many unread items (although the “high traffic” section is just because I added a ton of feeds to test this feature).
  3. The style Nick has chosen to present the information – it appeals to me, and all links (except the picture in the middle) point to FD’s cache of the feed rather than the article (making navigation instantaneous).

To be perfectly honest, I’m not sure if I’ll use the feature much. I have this annoying habit of trying to read everything that comes through my feed reader, which means I could easily get lost for hours in the collection of feeds I just added. I can see the use for people who subscribe to a ton of feeds though (Scoble comes to mind) and with a few improvements I could be tempted to keep a folder full of feeds that I never read except via Popular Topics.

My opinion? Right now this is a very cool feature, albeit held back by the inability of some site owners to realise the value of full-text RSS feeds.

Microsoft have changed, seriously

31st May 2007:
Note that this has become my primary source of spam so I’ve disabled comments. Sorry if you felt the need to comment on a 12 month old post, if you really feel the need to comment please contact me at the links above and I’ll re-enable them temporarily 😉

I woke up this morning to find that Nick Bradbury had posted about a particularly nasty WinINet bug that causes the application to hang (or in this case, the update thread sat on 100% cpu).  I’ve worked with this API a bit myself, and it’s essential unless you want to get bogged down in TCP and HTTP details in a windows application.

Much like other commenters to the post, including one who blogged about the bug in February 2005, I held little hope that MS would ever fix it.  The bug has existed since 1997, and… well this is Microsoft we’re talking about here.  They’re famous for not fixing bugs.

At least, until now.  This is the first sign in my corner of the development world that Microsoft are picking themselves up and actually responding quickly to user feedback.

A mere 7 hours after Nick’s post, a MS dev (Eric Law) finds the post and leaves a comment asking for URLs to reproduce this with.  Same thing I would’ve done.  Nick of course instantly responds.

35 minutes later, another MSFT comment – this time it’s David Powell, Product unit manager, Windows networking developer platform.  This was a WTF moment for me.  The bug has been reopened, and they posted 3hrs after that saying the bug has been reproduced so it’s under investigation.  I’m drooling at this point 😉

So what’s changed since February 2005?  Things like technorati.  This proves without a doubt that somewhere in Microsoft, they’re subscribed to searches such as the one AJ found me on so they can pick up people blogging about the product.


David mentioned in his comment the best place to report bugs in WNDP components, and that well-triaged bugs are noticed – http://connect.microsoft.com/wndp.  A few google searches later, and it turns out the IE team has a site on Microsoft Connect as well.  Woot!  Between the WNDP response and the availability of a good public bug tracker, I’ve now been convinced to report any bugs I see while working with IE code.  Previously I’ve just avoided the bug and found a better way.

A better understanding of taking FeedDemon to Newsgator

When the news first broke about purchasing Bradsoft, Nick gave us somewhat of a brief overview about why he took the offer.

I didn’t have a problem with it then, as I trusted Nick’s judgement; but I now have a much better understanding of why Nick trusts Greg Reinacker so much. He’s a believer in open information flow to customers, which is one of the things I love about Nick.

Yesterday afternoon I noticed, as did anybody else using Newsgator, that the sites were down, and FD popped up with this annoying error message every 5-10 minutes. When it all started working this morning, the first thing I did was check the forums – and Greg had linked to his most recent blog entry as an explanation.

I highly recommend reading his post. It’s full of gory technical detail about database failures and I love it. Open information flow is probably my favourite thing about the web recently; whether it be from bands, podcasters or company management, getting this sort of information out to the public who would have previously been given “we had a server failure, I’m sorry” is so much better and to me, a whole lot of fun to experience as it becomes the norm.

Props to Greg for handling this so well, sounds like it was a pretty nasty issue.

FeedDemon 2.0 is here!

After months of hard work, beta testing and feedback, FeedDemon 2.0 is finally done and available for download. Nick announced the gold release on his blog yesterday, and I couldn’t be happier.

It’s been a pleasure to once again aid Nick through the beta release of a new FeedDemon version, he’s one of the best developers around when it comes to feedback response and discussion. Congratulations on a fantastic milestone Nick, I can’t want to find out what your plans for the future are! 😀

FeedDemon’s hidden solution to Information Overload

Way, way back in the FD1.6 release notes there was mention of an easter egg. I spent a bit of time making foolish key combinations on the about box (based on memories of the IE4 easter egg) but I got tired of that pretty quickly. However during the FD2.0 alpha I found out what the secret is (it’s very sneaky) 😉

You may not believe it, but 5 weeks ago when Nick left a comment about suggested screenshots I immediately went out and took them. And they’ve been sitting on my desktop ever since, reminding me occasionally that I really should post them.

Nick has already spoiled the details but I thought I’d post the screenshots anyway – look at his post for details on how to access the easter egg.

When you first load the game, you crack up at the images of Dave, Scoble and Steve Rubel (at least I think that’s who it is, the bad editing makes his ears far too wide and it’s hard to be sure :D)

Once you start playing, it’s very easy – once you stop laughing at the demon hand catching unread posts.

The game doesn’t take long to pick up the pace though!

And the coolest feature? If your boss comes over, alt-tab to your real work and FD kindly pauses the game for you.

It’s been a while since I saw an easter egg like this that wasn’t in a game, and this one is very cool as well as superbly hidden. Nicely done Nick!