This post was originally entitled “Handbrake doesn’t care about black people… I mean end-users.” After reading through some of the comments made by the Handbrake developers at the bottom of this page, it seems that they do care about end-users… they’re just frustrated by bad end-users. Thanks for the dialog and clarification, Handbrake team!

First off, I need to clarify that the title of this post is a joke in reference to idiot rapper Kanye West’s claim that “George Bush doesn’t care about black people.”

Now that we have that settled, I came across a posting on Handbrake’s support forums that irritated me to no end, and I felt I need to share. If you’ve never heard of Handbrake, the team describes their project as “…an open-source, GPL-licensed, multiplatform, multithreaded DVD to MPEG-4 ripper/converter, available for MacOS X, Linux and Windows.” It’s a cool piece of software for exercising your fair use rights that the DMCA has attempted to take away from us.

If you haven’t read the above linked post, please read it before moving on. I’ll wait.

My first response when I read this was “No, no, no, no, no!” Now, the posting begins with a few bullet points that I generally agree with. Yes, software innovation is good. Yes, code inspection by other peers is good. No, it’s not cool when people come and begin demanding features immediately. No, it’s not cool to hassle or insult the developers. But from there, the author completely misses the point of writing software for other people to use.

Now, maybe the author has reached that point because the Handbrake community has had a number of “bad egg” users who hassle the developers like they owe them something. I feel very fortunate to be part of the SimplePie community which rarely ever has bad eggs or raving lunatics demanding features be built right this second. But even then, who are they building the software for?

Communities are very important to software projects. Bug reports, patches, feature requests, and all sorts of other things are important parts of the growth of a product. No, we don’t always have time to answer support forums in the timeliest of manners, and no we can’t implement every single feature that gets asked for. But we certainly listen, and we certainly try to work in the more relevant and useful features, and we’re always looking for ways to make things simpler for people either by a change in SimplePie or a tutorial and/or code sample of some sort. Why? Because this project would never have become as good as it has without our users and their feedback.

Handbrake’s team seems to think that they are just fine without end-user support. They seem to believe that they already know what is best for Handbrake, and that there’s no suggestion their users can make that they haven’t already thought of. GIVE. ME. A. BREAK. SimplePie started out as one guy scratching an itch and opening up the source for code review and suggestions. It was built around a series of functions, only supported one feed per page, and had maybe 5 users. Nowadays, SimplePie is being used by thousands and thousands of people all over the world, it’s fully object-oriented, supports all sorts of specifications, supports several feeds per page, and a whole variety of other cool features. Why? Because we already know what’s best and don’t want your input? No, no, no. SimplePie is successful because we listen to our users. I never envisioned SimplePie would be able to do the things it can now. We know that we’ve got good people in our community who have all sorts of ideas about how SimplePie could be better — and it shows!

So, best of luck to Handbrake. They seem like they know how successful they want their project to be. As for myself, I’ve deleted my copy of Handbrake and will look elsewhere for my video transcoding needs.

SimplePie is getting ever closer to releasing 1.0 final! We expect to have the first release candidate available in the next few weeks, and we’ll be asking the community to report bugs so that we can fix them for when 1.0 goes gold! Here is a list of features, changes, and improvements that are already in the latest trunk builds:

• A completely re-written parsing core which is faster and supports the standards better than ever before.
• Access to ALL of the available XML elements (just like MagpieRSS does).
• Multifeeds support integrated into the core, and has access to all standard SimplePie methods (Multifeeds 1 didn’t).
• Support for embedding Flash Video (.flv) enclosures.
• Faster-loading Flash-based MP3 playback.
• Better compliance and support for feed auto-discovery.
• Full support of the date/time parts of ISO-8601.
• More accurate support for XML mime types (RFC 3023).
• Improved xml:base and XHTML support.
• Literally hundreds of other bug fixes and other minor improvements.

We’ve also got a few “blockers” on the list, meaning that we won’t launch 1.0 until we have support for these:

• Support for iTunes and MediaRSS elements.
• Complete support for GeoRSS elements.
• Image and Favicon caching.
• A few other minor bugs and feature requests.

This will definitely be our most exciting release so far, and we hope that you guys will be happy with it as well! If you want to keep up with what’s on the list, what’s not on the list, and what’s already been done, you can keep track of our progress and test your feeds in the latest builds on the Development page.

The upcoming SimplePie Lite service will be based on 1.0 code, as well as updated versions of the Plugins for WordPress, Textpattern and Mediawiki.

I just realized how long it’s been since we’ve had a real blog posting here. Work on 1.0 is coming along, albeit a bit slower than with previous releases. But that’s mostly been an issue of time available, rather than any particular technical challenges.

Geoffrey has been hard at work with re-writing SimplePie’s core engine, and that’s been available in the trunk builds for several weeks now. Among other things, it includes the MagpieRSS-like ability to parse out ALL elements within the feed into a huge data array. Because of that, we now have access to things like geo:, media: and itunes: data, although we haven’t yet built any easy-to-use API functions for them yet.

Geoffrey has also been working hard on improved standards-compliance within SimplePie. In addition to going over more bug reports, RFCs and specification documents than I care to count, he’s also been working on our automated testing system, which significantly helps with things like regression testing and proper support for third-party testing suites (like Mark Pligrim’s Universal Feed Parser unit tests). This system has already been immensely helpful, and will really help us move forward into awesomeness in 2007.

I (Ryan) have been working on a number of consumer-oriented functions, such as one that will allow us to shorten descriptions by character or word while maintaining existing HTML. I’ve also been working on code for ‘Multifeeds 2’, which will allow feed mashing in a cleaner and simpler manner than the current Multifeeds code. I’m also in the process of refreshing and realigning the SimplePie website. After going through the statistics provided from Mint and Google Analytics, we’ve decided to make some adjustments to several areas of the site in an effort to improve the overall usefulness. Expect to see those changes with the release of our 1.0 Release Candidate.

Lastly, we’re getting ready to release a preview of what we’re calling SimplePie Lite. SimplePie Lite is a web service providing a JavaScript interface for SimplePie. Here are a few features and advantages:

• It is a hosted service. There is nothing to install, and no futzing with installing missing PHP extensions. Everything runs on Dreamhost servers, which have everything installed.
• When passed a feed, SimplePie Lite returns a gzip-compressed JSON object, which is perfect for integrating into your AJAX apps. On average, the gzipped data is anywhere from 70–90% smaller than the uncompressed data. Couple that with SimplePie’s caching and ‘HTTP conditional get’ support, and you’re looking at some very small, very fast downloads.
• ALL of the relevant SimplePie PHP functions have been made available in SimplePie Lite, albeit JavaScriptized. Where SimplePie has $item->get_permalink(), SimplePie Lite has item.getPermalink(). All of the relevant configuration options are also available, the only exceptions being those which need to be static for optimal efficiency for the server.
• After the guts are finished, we’ll be building a user interface which allows people to enter a few bits of information, click a few checkboxes, and generate code that can be copy-pasted into webpage or profile where JavaScript is allowed.
• This also means that SimplePie is no longer solely available to those building with PHP. SimplePie Lite can be used with Ruby (on Rails), Java, ASP/ASP.Net, Coldfusion, and virtually any other web programming language.

The preview will be available in the coming weeks, and will be formally launched sometime after SimplePie 1.0 goes gold.