What if there was software with the elegance and extensibility of WordPress but all the features you’ve come to expect from social networks like Facebook? Now there is: check out BuddyPress.
BuddyPress is an official sister project of WordPress. The idea behind it was to see what would happen to the web if it was as easy for anyone to create a social network as it is to create a blog today. There’s been an explosion of social activity on the web, it’s probably the most important trend of the past few years, but there’s been a dearth of Open Source tools that enable the social web.
In WordPress we have a robust and extensible base that can scale to many millions of users, and BuddyPress is essentially a set of plugins on top of WordPress that add private messaging, profiles, friends, groups, activity streams, and everything else you’ve come to expect from your favorite social network, like a Facebook-in-a-box.
I don’t think BuddyPress will be something you use instead of your existing social networks, I mean all your friends are already on Myspace, but if you wanted to start something new maybe with more control, friendlier terms of service, or just something customized and tweaked to fit exactly into your existing site, then BuddyPress is a great framework to use. Maybe even someday you’ll be able to connect your BuddyPresses to each other and to the existing monolithic social networks.
This is just a 1.0 release and it’s not for everybody yet, for example it currently requires using MU which is a bit trickier to get set up than regular WordPress, but regardless I’d recommend diving into the community at BuddyPress.org, which is great example of the software in action.
The poll is closed, the votes are counted, and the results are interesting. The table below shows the actual breakdown of the poll votes, of which there were 2,651. As you can see, there were four main contenders: Dean J. Robinson’s Fluency-based submissions (two variations), the existing 2.7 interface, and Matt Thomas’s comp (MT), which exists somewhere between them in terms of style. Note: GB was a late entry, and was posted after over 900 votes had already been collected.
As several people have rightly pointed out, the Fluency-style designs not only took the top spot, but in combination added up to a higher percentage than any other. We’re not focusing solely on that statistic, though, because had other designers submitted multiple versions, the numbers might have looked different. What was most interesting for me was checking in on the votes over the course of the two days the poll was open. The top three (Fluency-dark, Current 2.7, MT) kept beating each other out for the #1 spot as they cycled back and forth through the top three slots, and had the poll closed on time (left it open a little longer in case anyone translated the time zone incorrectly), the order would have been a bit different.
What’s more interesting to me is the overall style that seems to be preferred among voters, as Matt’s comp has some stylistic similarities to Dean’s (see image at left). It also would be interesting to know how many of the votes for the current 2.7 interface were based on thinking it looked the best vs. how many were votes against changing the interface at all so soon after the 2.7 redesign. If you want to comment on what you liked best and/or least about any of the designs, this thread is a good place.
So what happens now? However we look at it, the Fluency-style designs clearly have a lot of fans. Then again, so do the designs of Matt Thomas (he’s behind the current style of 2.7, remember, in addition to the comp labeled MT). To give the interface the attention it is due, and to take seriously some of the interface feedback around usability and accessibility, we’re going to leave the looks alone for 2.8. It’s our guess that a revised style will make into 2.9 early in the development cycle to allow us plenty of time for user testing and revision. How close it winds up being to the comps submitted in this design tweaks challenge will depend, but in the meantime:
Comps for the header/nav design tweaks are in, and the results are mixed. Some people just moved a few things around, while others proposed a new style altogether. We won’t make any major changes to style in 2.8, but if the vote leans toward a submission that proposes it, we’ll do some user testing and make a decision for early 2.9 (which, now that we think of it, is probably the right thing to do anyway. :))
Below are the links to the screenshots that were submitted. Please review each one (I’d open them all in tabs so I could look back and forth while they are all large size, because the voting poll just uses thumbnails), then choose the one you think looks the best/is the most usable.
This poll was supposed to close at 8pm NY time on Tuesday (today), but we’re going to leave it open for an extra day. The voting poll will now be closed at 8pm NY time on Wednesday (that’s 2am Thursday, UTC). If you want to discuss the entries’ pros/cons, this thread would be a good place.
Jane: It is a thorn in my side that the blog name header is above the “dashboard” nav section in the admin, since in MU installations and with plugins (like stats), things in the Dashboard section span multiple blogs. Makes more sense for the header to head only the per-blog content area.
Mark: I agree about the header. “This is the menu, this is the content.”
All: Yep.
Five minutes later…
Mark: What do you guys this of this quick mockup I just did, playing with the admin header?
Jane: I like it that the nav is not under the header. Might need some styling help. I was also thinking maybe the favorites menu should drop down into the white h2 area by screen options/help tabs.
Ryan: Menu color to the top with blog title pushed over and favorites next to screen options sounds quite nice.
Jane: I’ll ask Matt Thomas if he could style it [ed. note: Matt Thomas created the visual style for 2.7], and we can see what people think, maybe post on wpdevel for feedback.
Ryan: If it’s quick, maybe we could even get it into 2.8.
ACT II
Matt T: Here are some comps based on what you told me.
Jane: Cool, but where are Screen Options and Help tabs?
Matt T: Still working on that.
Jane: Hm. Wonder if there’s time to open this up to community designers? I know we’re in freeze, and it’s no notice, but you didn’t get any notice either when we dropped this styling request on your lap a few hours ago. That’s the way open software development works: sometimes the best ideas come at the last minute!
Matt T: I’m all for letting the community take a stab at it. Especially if they come up with something brilliant to do with the Screen Options and Help tabs.
Jane: I’ll ask Ryan about release date and see if there’s time. I know they wanted your style recommendations today.
Act III
Ryan: Tuesday is probably doable, no later than that for final delivery of style and any gradient graphics, etc.
Jane: Awesome! People will hate me for the short notice after the has-patch marathon, but since it’s a small project and over the weekend, and wasn’t even something anyone was planning until a few hours ago, I’m *really hoping* people will take this for what it is, an attempt to give more people input into an upcoming visual change in the interface, even if it’s not a huge one.
Ryan: Would have the benefit of warning people that header and menu will be changed a bit.
Jane: And we can have a vote. If I can get all the materials together and post in the morning, that would give 2 days of design time for submissions on Monday, and if we do a day of voting Tuesday, that’s 3 days notice for the vote. I’ll make sure to post to all the lists, etc.
Ryan: Will we announce with comps from Matt T as examples of what we’re thinking?
Jane: I’ll write up the UX reasons for considering the change, and Matt T can provide some style guidelines and his original comps so no one will have to waste time mocking up the basic screen layout.
Ryan: That would help set the scope. We just want tweaks here and there, given the timing.
Jane: Woot!
On Your Mark, Get Set…
Okay, so here’s the deal. Modifying the nav/header to be a little nicer is was a last-minute design idea, and if it can’t be worked out in the time we have left before 2.8 (which is very little), we’ll just wait until 2.9 to work on it. But! If someone comes up with something the community really likes and it doesn’t break any of the design guidelines for the rest of WordPress, we could sneak it in.
UX and design guidelines for this mini-project are posted here (so as not to clog up anyone’s feed reader with big graphics). Read through the UX stuff, check out the comps Matt Thomas mocked up last night (with absolutely no notice, for the record). Use the .psd as your base, and when it’s time to submit your ideas, make a .jpg or .png and post a link to it in the comments on this post. (Note: Only comments containing a link to a design submission using this format will be approved. For general discussion about this design challenge or any of the submissions, please head into the #wordpress-dev IRC channel.)
Submit the link to your comps by 1am Tuesday, April 28 UTC (7pm Monday, April 27, New York time). If you have questions or want early feedback, we’ll be in and out of the #wordpress-dev IRC channel between now and then.
Once we’ve received the submissions, we’ll post a voting survey (much simpler than the icon survey; this one will be more of poll, just choose the one you like best) as soon as possible, and will post the link to it here as soon as it’s online. We’ll only keep voting open for one day because of the 2.8 deadline, so put it on your calendar if you think you’ll forget. Voting will close at 2am Wednesday, April 29 UTC (8pm Tuesday, April 28, New York time). Results will be announced the following day.
* Chats above are a conglomeration of actual chats.
Reminder: Only comments containing a link to a design submission will be published here. All other comments will be deleted.
If you want to leave a public comment about this contest, the design, etc., I’ve created a thread in the forums that you can use. Please discuss these things there. If you leave a regular comment here on this blog, no one will be able to reply to you, because only actual links to design submissions will be posted in the comments here.
Google has announced the successful applicants for the 2009 Google Summer of Code, and WordPress is lucky enough to have eight students allotted to our open source project. It was a tough choice, since we had almost 60 applications to choose from. We’d like to thank all the students who applied, and we’re sorry we couldn’t take on more of you.
Developers, if you see these bright young things in the dev channel, please be your usual friendly, helpful selves. 🙂 Everyone else, wish our students luck with their projects this summer, which promise to be challenging but awesome. Without further ado, I’m pleased to introduce the GSoC projects (in no particular order) and the students tackling them.
Justin Shreve, Extended WordPress Search Engine. Justin will be mentored by Andy Skelton. One of the complaints I hear over and over again is about the search engine, so this could have great benefit to WordPress core.
Rudolf Cheuk Sang Lai, Adding Photo Grouping by Album Functionality. This project will wind up being a piece of a larger media redux project for 2.9/3.0. Mark Jaquith is mentoring, and Noel Jackson will be a backup mentor.
Daryl Koopersmith, WYSIWYG theme editor/generator. This will allow users to create and edit themes without touching any code. Beau Lebens is the mentor on this project.
Michael Benedict Arul will be working on a similar project. Michael will be mentored by Andrew Ozz, since this project will be using jQuery. It’s our hope that having two students working on this idea separately will foster competition and allow us to compare approaches.
Daniel Larkin, Modified Preorder Tree Traversal (MPTT). Lead Developer Ryan Boren will be his mentor. This is Daniel’s second GSoC working on WordPress.
Diego Caro, a student from Chile, will also work on an MPTT project. Diego will be mentored by Thorsten Ott.
César Rodas, social and text processing algorithms for BuddyPress and MU as related to recommendation engines. Alex Shiels and Andy Peatling will co-mentor this project.
Anthony Cole, Event management with WordPress. Co-organizer of WordCamp Australia and New Zealand, Anthony will be working on a suite of plugins (or maybe just one or two out of a planned set, scope TBD) for event management/attendee networking that will be built on BuddyPress/MU/bbPress. We’ll use wordcamp.org as a test case, and release the final product to the community. Jake Spurlock will be mentoring, with Andy Peatling as backup.
Congratulations, guys*!
*Seriously, we didn’t get more than a couple of applications from female student developers. Where are all the geek girls?
As promised, here are the results of the 24-hour has-patch marathon that was announced, begun and completed over the course of a few days last week (more on timing after the results). Results include activity from 8am Pacific time on Thursday, April 16, 2009 to 9am Pacific time on Friday, April 17, 2009.
Total number of patches committed to core: 44
Contributors whose old patches were committed: 9
Marathon contributors whose patches were committed: 13
Tickets closed: 102 (breakdown below)
Fixed – 45
Dupe – 16
Wontfix – 10
Invalid – 19
Worksforme – 12
Tickets created: 20 [I guess not everyone got the memo that we were trying to close tickets. 🙂 ]
Tickets reopened: 4
Number of testers who left comments in ticket threads: 10
Number of testing-specific comments: 18
These numbers are based on opening each ticket that registered activity during the marathon hours and counting the actual comments that indicated some testing of a patch. Contributions to philosophical discussions without a patch, while important, weren’t counted for this purpose. Nor were Trac notices that simply noted a ticket was being closed because it was a dupe, invalid, etc.
Top five contributors (committed patches): Denis-de-Bernardy, filosofo, nbachiyski, scohoust, simonwheatley
Top five testing feedback providers: shanef, Nicholas91, Denis-de-Bernardy, sivel, williamsba, mrmist (tie)
Given the short notice/last-minute nature of the marathon, I think we did pretty well. Granted, there were people who complained that two days wasn’t enough notice to clear their schedules, but let’s be honest, the 24-hour has-patch marathon was more of a rallying cry to help clean out Trac than a deadline based on anything specific. Patches are always welcome/encouraged, and now that the big features for 2.8 are mostly done, the lead devs will be able to spend more time reviewing Trac tickets and patches. Still, not too many people tested existing patches (or if they did, they failed to leave the requisite comment in the ticket threads). Testing patches is one of the easiest things you can do to help further development, since patches won’t be committed or rejected until they’ve been tested by several people.
As we get closer to the 2.8 release, jump into Trac any time and test a few patches (don’t forget to leave the feedback!) if you have time. If there’s a ticket you’re sick of seeing there, write a patch and ask your fellow contributors to test it and comment on the ticket thread. We’ll announce an official bug hunt soon (and yes, there will be more than two days’ notice), but the fact remains that addressing new bugs is easier if Trac isn’t clogged with old tickets. If you spot duplicate tickets, mark it a dupe, note the other ticket number in the comments and close the ticket. If you see one that is no longer relevant because the current code base fixes a problem reported several versions ago, mark it invalid, leave a comment and close the ticket. These simple housekeeping tasks may not seem like much, but they do help. Special props to Denis-de-Bernardy, who in addition to writing a couple of patches during the marathon and testing a few others, did a bunch of ticket maintenance like this, and cleared out a number of tickets.
Thank you to everyone who participated, and until the next marathon, happy patching and testing!
The 24-hour has-patch marathon has just officially begun! For the next 24 hours (until Friday, 4pm UTC), the core developers will be evaluating patches that have been tested and committing those that are good. When they run out of those, they’ll start testing patches that have been submitted but not tested. This takes longer, so help us keep the momentum going by testing patches today.
Grab yourself a copy of the nightly build to make sure you’re using the right version, then head over to Trac and start looking at the has-patch* tickets. Pick a ticket, download the diff, test it out on the browsers/platforms you have available, and write a comment about the results in the ticket’s comment thread. Move on to the next ticket. Do as many as you can over the next 24 hours.
And if you’ve got the mad skills to contribute bug fixes and code patches for enhancements and other tickets, now is also a great time to contribute a patch. Why wait? One of the complaints I hear in the IRC #wordpress-dev channel is that it can be frustrating to write patches because sometimes it takes a long time for them to be reviewed and committed. For those people (and everyone else), this is the perfect time to patch, because we’ll be looking at every has-patch ticket in Trac for 2.8 over the next 24 hours. If you have a patch that’s been submitted but it hasn’t been tested, consider doing a little PR for yourself. Hop in the dev channel, post on your blog, mention on Twitter that you need people to test your patch *today* so that it can move forward in the process. This will speed things along. You can also add the keyword needs-testing in addition to has-patch.
At the end of the 24-hour period, you don’t have to put your pencils down, but the core devs will be going to sleep and then returning to the regular dev cycle, so it’s worth it to contribute today. If you miss the deadline, it’s okay; you can contribute or test patches as usual, you just won’t get the more immediate gratification the marathon promises.
We’ll post the results of the marathon here on Monday, after everyone’s gotten some rest and we can go through the Trac logs to see how many people got involved. This is one reason it’s so important to leave a comment when testing patches…it’s how we’ll be counting the number of marathon testers. Top patch contributors and testers will be recognized in the results post, so if that sort of thing motivates you, let the link love lead the way.
Of note: we are now in feature freeze for 2.8!
* – For those who don’t know why I keep using the term has-patch… When someone submits a patch by uploading it to the Trac ticket, they add “has-patch” to the keywords field, so that the core devs will know there’s a patch attached. Not the most elegant system, true, and you’d think maybe Trac could just recognize that a certain file type had been uploaded, but there you have it.
Waiting patiently for a bug hunt to be announced before you get involved as a WordPress development contributor? Pshaw! Don’t be shy!
2.8 currently has about 500 active tickets that need to be resolved. The core devs are largely working on the bigger feature additions, such as the embedded theme browser/installer, the new widgets management, improving performance, etc. so a lot of tickets that are of lesser priority are still just sitting there. Aren’t they calling to you, just like puppies at the pound? “Adopt me! Please!”
Before we have a bug hunt and see the addition of dozens of new tickets, we need to clear out some of these old ones. Not being the kind of shelter that euthanizes its bugs after a certain amount of time (seriously, there’s one ticket that goes all the way back to version 1.5), we are hoping people will step up and bring home a bug today.
To keep things moving, we’re announcing a new kind of event, related to bug hunts, but with a different slant. We need a sprint to clear out these tickets. Thursday is the day (and Friday for those over the date line). Core devs will spend 24 hours going through all the tickets tagged with has-patch, and committing those that have been tested and work. So how can you get in on the Super-Awesome WordPress 24-Hour Has-Patch Marathon?
Write a patch. There are dozens of tickets for discrete little pieces of correction (change … to actual ellipses in admin interface, change the ‘go back’ link to a ‘view page’ link, etc.), dozens that are browser-specific bugs, dozens that might be more challenging. Pick the one you want to work on, add a comment to the thread so other marathon contributors know someone is working on it, and get the patch submitted before the marathon ends. If you start coding now, your patch could be in by the weekend!
Test a patch. There are, as of right now, 177 tickets marked with has-patch. Patches can’t be committed until they’ve been thoroughly tested. If you’re already running the nightly build start testing out these patches in as many operating system/browser combinations as you have. Only have one? Hey, it’s probably more than has been tested already! If you’re not already running the nightly build, you can download it here to set up a test blog. Don’t forget to add what you found to the comment thread for each ticket. If it doesn’t work, be specific about what is not working so that others can jump in and fix it.
24 hours of patching, 24 hours of testing, 24 hours of committing. Don’t miss the excitement; get started now!
The Super-Awesome WordPress 24-Hour Has-Patch Marathon begins Thursday, April 16 at 8am Pacific time (that’s Thursday, 4pm UTC) and will end, as you might have guessed, 24 hours later. No reason to wait, though… start early and get patching/testing. The more patches that have been tested by Thursday morning, the more that will be committed during the marathon.
As mentioned previously, the icon design contest was such a success that we realized we needed to create more ways for graphic designers to contribute to the WordPress open source project. Our community is chock full of talent, and this talent is just itching to get involved. So, we’re trying out a few ideas.
First and most immediate is the creation of a new “component” in Trac for graphic design. We’ll use this component label to create tickets for things like making graphic buttons, such as making a new version of the favorites menu graphic or WordPress mark that looks better with the blue admin theme. In some cases graphic design tasks will overlap with CSS tasks, so designers interested in contributing can search for open tickets with either component label. In cases where a base PSD file is needed as a starting point, we will attach it to the ticket.
In this vein, if you notice a graphic that needs fixing (like the aforementioned favorites menu button needing a blue version), please use the graphic design component label to report it in Trac. Please don’t create tickets for graphic you just don’t like… keep it to things that look broken or overlooked.
Graphic design tasks will follow the same protocol as development tasks in that volunteer work will be overseen/curated by the core team. Both Matt Thomas (who provided the art direction for 2.7) and I will work with the core development team to identify design tasks and review submissions (i.e., we’ll make sure things look awesome before they’re committed). We’ve been trying this process with an early volunteer, and it seems to be working well. In addition to creating individual Trac tickets, if a project arises that requires many graphic elements to be created, we will post about it here on the dev blog to give potential contributors a heads up.
The second opportunity for graphic design contributions will be less about discrete tasks and more about contributing to the evolving visual design of WordPress. When there are larger design tasks in the works, such as creating an admin theme or designing the new media upload process, there will be opportunities for screen design. In these situations, potential contributors will be given access to materials such as wireframes or specifications and will be able to submit comps of design approaches. In the event that we receive multiple good submissions for a screen design, we will use a community vote to influence the final selection, as we did with the icon design contest.
Since 2.8 is wrapping up currently, there aren’t many design tasks waiting to be completed right now, but when we roll into 2.9 development, there will be more opportunities for graphic designers to get involved.
Recent Comments