EclipseCon 2010 Program In The Bag

If you have been tracking the EclipseCon submission system, or tweets from @eclipsecon, or blog posts from Donald Smith, you will have noticed that the Program Committee have finalized the program for 2010.

We’ve rejected far, far more submissions than we accepted, and probably upset some people. Sorry about that. Lucky for me, I’m getting the positive mail and Donald is getting the grief. Just in case you wanted to know, the decision process went a bit like this:

• the big list of submissions was reviewed – we set up small teams from the Program Committee to review based on tags assigned to the submissions to do this
• each tag team did a full review of the abstracts in their tag area and came back to the group with the “must-haves”
• we gathered all the “must-haves” into lists for each talk type and spent over eight hours arguing about which were the most appropriate to choose for the program

All during this period, talks were being shrunk in size, coalesced, moved around and generally acting like quicksand.

I need to offer many thanks to guys on the Program Committee for the big pile of work they put in over the weekly conference calls and the big push in the endgame. They are:

• Bernd Kolb — he rejected your talk,
• Chris Aniszczyk — he rejected your talk,
• Doug Schaefer — he rejected your talk,
• Dave Carver — he rejected your talk,
• David Williams — he rejected your talk,
• Donald Smith — he was the voice of reason,
• Ed Merks — he rejected your other talk,
• Ketan Padegaonkar — he rejected your talk,
• Peter Kriens – he and his team rejected your OSGi DevCon talk,
• Scott Rosenbaum — he rejected your talk,
• Wayne Beaton — he rejected your talk, but in a nice way.

When these guys failed to reject your talk, I stepped in and did so. I’d also like to recognize the excellent contributions of two others who for one reason or another needed to withdraw from the committee:

• Bjorn Freeman-Benson — he kicked the whole thing off,
• Doug Gaff – he didn’t reject any of your talks.

Gabe O’Brien provided great help and timely support in the extension and massaging of the various recondite figaries of the submission system.

If your talk has been accepted – good stuff, start making your slides and, more importantly, your demos. Resist the urge to crow about it on twitter, if you can. If your talk has not been accepted – we will have outside-program space as part of the Unconference in the evenings. If you are coming along to the conference, you will get a chance to pitch your project or work. We’ll roll out plans and guidelines on how that will come together over the next couple of weeks.

Keep working on your talks. You should know your own stats at this stage. It takes me a total of about twenty hours to prepare from scratch a mediocre forty-minute talk about something I know pretty well. To do a shorter and better-quality talk takes considerably longer.

See you in Santa Clara in March.

Blog Administrivia for 2010

New year, new blog theme, new content. This year I’m going to add more non-career-related material to the blog, primarily to keep me interested in doing more writing. You might like it too.

Welcome to 2010!

I predict this will be the year of the Most Number of Disappointed EclipseCon Submitters Ever.

Myself and the Program Committee are trawling through the many, many submissions for EclipseCon 2010, and having a rough time finding space for so many potentially interesting talks. It’s even rougher than usual this year.

If you have submitted a proposal, keep an eye on your email for communications. We’re trying hard to produce a balanced program that caters to everyone – but Eclipse has a large ecosystem and while we are trying to be as diverse as possible in our choices, there will be good material we will have to leave out from the official program.  We may ask you to restructure your talk, or make it shorter, or merge it with another talk. If you have submitted multiple talks on the same topic, or with a core plus tangential topics, well, I guess you can see where this is going

Ok, so much for the gloomy part — the upside is that we will have a really neat collection of talks this year that accurately (as much as we can) reflect the popularity of the various topics. So far it seems to fit the demographic we’ve seen before, with Modeling, Runtime, Java and UI/RCP being the most popular areas. The new tag Build and Continuous Integration has entered the charts pretty high up too, not unexpected with the multiple approaches now becoming available to build Eclipse products and plug-ins.

If your talk is rejected, don’t forget that we will also run an Unconference event in the evenings – you might still get a chance to say your piece. You might want to think of it as a just-off-Broadway show alternative.

EclipseCon Submissions – pendant le déluge

This time of the year is holiday season around where I live and it can get pretty quiet during the day, as people dream of days off rapidly approaching and answer IMs from boomerangs of the diaspora, back for a couple of weeks. It makes me wish I had followed through on my idea to make a machine that goes ping whenever an EclipseCon submission came through, because the constant ringing would be like totally festive. Jingles bells and all.

Sadly, though, the jingling will be coming to a halt TODAY.  It is bah humbug time as the Program Committee close the sack and wander off to their various lairs to peruse its contents. So, if you have a submission that you were thinking about, and you’ve been putting it on the long finger, now is the time to get that finger out and submit your abstract.

Enough of the reminders. This year we picked up Donald Smith by his ankles and shook him and enough money fell out to fund a Program Committee Top Five Early Bird Pick competition, wherein awesome Eclipse schwag would be deployed to a lucky few. The talks were selected to be great representatives of the different topics, styles and directions that the PC is looking for this year.

The ever-on-the-button Don has already blogged about the winners. No doubt he was a little miffed when we couldn’t settle on just five talks and had to shake him again to get enough funding for a sixth!

Reminder! Submit your EclipseCon Talks Now!

For extra-special goodness, and to get all you potential presenters off your collective keister, we have a little bonus going here. If you get your talk in today, 10 December 2009, before about 1800 Pacific Time, we, the ever-hard-working EclipseCon Program Committee, have the opportunity to pick 5 Talks of Awesomeness and Timeliness from the current tally of submissions.

These talks will receive some universally coveted Eclipse schwag in recognition of their their awesomeness and their timeliness. The PC will be up all hours over the weekend haggling over the winners with the intention to announce on 14th December. There are no lobbying guidelines in place!

There’s news on the keynotes too – check the conference page – what can I say here, except if you like Programming, Space or Robots, we are catering to you.

A short post on EclipseCon 2010

[Update - once you read this, go check out this entry in Don's blog for some extra goodies!]

The submission floodgates have been opened on EclipseCon 2010, and you have until December 18 to get your submissions in to attend the usual West Coast extravaganza of all things Eclipse and OSGi. Go for it.

I’ve been seriously quiet about the conference, even though I’m Program Chair and should be running around shouting about it. If you are following me on twitter, or searching on the #eclipsecon hashtag you will have seen a few leaking tweets over the past little while. But here, now, ladies and gentlemen, boys and girls, is a little more information. This marks the point from where you may begin the countdown to the insufferability of /me on this topic.

The Committee of The Caring

The Program Chair does get to innovate a little bit on the approach and set-up of the spirit of the conference and some of the structures that will support same. This person also gets to pick the Program Committee, a shower of dedicated and committed professionals with a overweening fondness for Eclipse and that most qualifying of characteristics — they care. I don’t mean that in a fluffy-cardigan-cuddles-and-tissues kind of way. I mean it in the way that if you try to mess with their relationship with Eclipse, they care enough to see you outside in the car park, toot sweet. You know what I’m talking about.

What’s Different This Year?

Back to the innovation piece. When Bjorn announced the conference way back in July 09, the first little peep of innovation squeaked out. This was the themes. I think it’s important that we create themes that directly address the Eclipse Ecosystem’s three constituencies – consumers, contributors and community.

• Making with Eclipse — you are consuming Eclipse open source software to build your own products, internal or external to your organization. You could even be selling them and making pots of cash, which is good. You want to come to this conference and see what people in the same boat as you are doing, what new innovations have come to light that you can use to speed up your processes and potentially reduce your costs and to show everyone how cool the stuff you’ve made happens to be.
• Making for Eclipse — you are a contributor or a committer, you write code, or docs, or both. Maybe your language skills mean you do translations. In some way you are injecting some of your expertise and time in to enhancing and extending the corpus of Eclipse creativity. You’re here to show people the awesome stuff that you are producing and adding to the Eclipse Ecosystem, as well as to teach people APIs, announce new projects and talk about project directions.
• Making Community — one of the most important things at a conference is the fact that you have many human meteors zinging around the halls bumping into each other and exchanging information. You’re here to meet your forums and IRC buddies face-to-face, or to have a full-duplex discussion with a group telling them why they should take a certain path, or to finally grab a project lead and suss out exactly how such-and-such a weird API works.

But what about the technology, I hear some shriek, won’t someone please think of the technology? High level themes like the above are not enough to navigate a conference the size of EclipseCon. When you go to the submission system, you have a whole passel of tags that will help you mark your talk. If you are looking for something in particular, you can search the talks using the tags too. You think maybe we need more tags? Let me know on a comment here, or address yourself to @oisin in tweetenland.

Yes, we’ve got ski ratings too. If your talk is totally hard-core, you should give it a double-diamond marking to set expectations. People with dashed expectations sometimes get a little heavy-handed with the -1 cards. Don’t worry about marking your talk as being at the easiest level, the Program Committee wants talks at all levels for all comers.

Types of Talks

Here’s where things get a bit more interesting. I made a post at EclipseCon 2009 where I blew the lid off the story that I had been at some boring talks. Yikes! I got a bit of ribbing for that, as you can imagine, so look out for the case of rotten fruit wherever I do Eclipse talks. When I got this Program Chair gig, I had a few chats with a few people, looked at a number of presentations and proposed to the Program Committee that this time around we are going to savagely cut the number of hour-long talks. This will cut waffle. It will cut meandering code walk-throughs. It will cut monster slide-decks. Your talk will be clear, sharp, to the point. You will say all you want to say and you will do it in twenty-five minutes.  There will be applause at the conciseness of your perspicacity, and you will be mobbed by well-wishers in the halls. Seriously.

Here’s the nitty-gritty details of the talk sizes

• Lightning Talk — it’s a Blitz. You have twelve point five minutes to do your thing. Get up there and show that crazy mash-up you’ve constructed, tell me what’s new in your project, shoe me the result and give me a page-o-links so I can find out more. I’ll find you and bug you if I have questions.
• Standard Talk — this is the twenty-five minuter. You know the score already.
• Extended Talk — I never said there were going to be no long talks, just that we would be culling their population. An Extended Talk is fifty minutes long. You don’t just get one easy as pie. The Program Committee will be scrutinizing the submissions and you will need to get over a high bar to get an Extended Talk accepted. These guys are professionals and can smell over-stretched talk like a shark smells blood in the water. Top tips — don’t leave your abstract until the last hour before the call for submission ends and then bang it out fast; do link to a paper or document giving more information about your talk; do expand on details in the comments section of the submission; do produce slideware early; do have multiple presenters construct a connected whole from two pieces of cloth; do show some demonstration applications.
• Tutorials — you know these guys. A tutorial can be two or three hours long.
• Panels — you might know these guys from previous EclipseCons or other conferences. Panels are chancy – they can be dull, but there’s nothing better than a good, disputative panel to get people engaged. The Program Committee will be working hard to make sure the Panels won’t have a dull moment in their one hour length.
• Unconference — you’ve heard of these, right? Ever been to a BarCamp? If you haven’t, don’t worry. What we are talking about here is a participant-driven conference space in the evenings – it’s like the base class of a BoF. The idea is that you can get some space to do a short talk or meeting, you put the subject up on a notice board and people turn up, or not. We’re still putting the details together on the logistics for this one, so hang in there for a future blog entry.

Conference Structure

I think this is the most interesting part. An EclipseCon day is a long, long day, especially if you have travelled a distance and it can be tough to keep the concentration levels up, despite liberal caffeination. The best way to stave off that all-conferenced-out feeling is to ensure that there is a mix of things going on during the day. So I’ve applied a bit of innovation to the structure of the conference itself.  We kick off in the morning with either a tutorial, or a keynote followed by a tutorial. Get your learning on. After lunch, it’s talk sessions. Sit back and listen, or ask questions. End of the day, it’s Panels. Ask questions, get engaged in a conversation. Evening time, it’s the Unconference. Did something inspire you during the day? Grab a podium and talk about it! Maybe someone agrees with you, or maybe the opposite. You can go and blog the day later on.

That’s the short description of what’s going to go down this year. I’ll follow up with more detailed articles on submissions, panels, tutorials, unconference and all that. Your comments are welcome. Ask questions, write your own blog entries. While you are doing that, I’ll be over here, watching the submission system.

…then *four* come at once

Arg – I haven’t had a blog entry in a month! I’m sure at this point that the Marketing Dept have opened the Big Book of Slaps and are thumbing their way towards the end of the list, their quill well-charged and dripping with ink the colour of fresh arterial blood. Or something. But now I have three four things to mention.

FUSE Integration Designer 1.2.1 Released!
We’ve pushed out an update to the 1.2 release we did in May. It’s a little point release, primarily about fixing bugs (and the team got through over 100 JIRAs!) but we have managed to sneak in a few improvements, including support in the palette for FTP/SFTP, XSLT and Timer endpoints. Check out the release notes for the full skinny. The download page continues to be in the same place.

One thing to note – if you are updating your existing 1.2 installation from the update site, there’s a small speed bump that turned up at the last second. Make sure that you read the installation notes for some instructions on how to make the update work smoothly for you.

FUSE Forge Open For Business
Yes, we’ve made our own Forge – a chunk of infrastructure to support projects that want to extend or apply the various FUSE technologies in whatever way makes sense. Initial reactions appear positive. I’m hoping I can use it for a bit of crowdsourcing action – I’ve set up a  FUSE Integration Designer Samples project, with the plan to add sample projects which illustrate some of the capabilities of the tool. If you have any examples you would like to share, let me know and we can put them up there!

If you have been watching the development of Apache Felix Karaf, and thought to yourself, you know, that’s just another one of them there OSGi containers with some tinsel, ekcetera, you are right. Stephen Evanchik spotted this too, but he went one step further and started creating Eclipse code to treat A.F. Karaf as an Eclipse Target Platform.  Very recently, he migrated this code to the EPL-licensed EIK Project at FUSE Forge, which is fantastic. I’m looking forward to hacking on this stuff with him.

But wait! There’s more! Lukasz, developer of the ServiceMixIde project had been following Stephen’s progress and decided that he would like to bring his code base to the project, too. This is a better start than I could have imagined. Right now we’re getting bootstrapped, code is going in, build systems need to get made (banging Maven and Eclipse rocks together), web sites, CI and all that good stuff.

FUSE Integration Designer Webinar
I’ll be doing a webinar, as it is known, on Tuesday 21st July to show how you can use the FUSE Integration Designer to make services, wire together mediation routes and poke around at JMS queues. The times are available on the FUSE Webinars page - yes, it’s happening twice in the one day, and both times it will be live and unexpurgated! Who knows what will happen?

Yikes!
I just agreed to be Program Chair for EclipseCon 2010! This should be interesting.

Insert Lightsabre Noise Here

Just a quick announcement this time around – the guys at the FUSE Forge have just started up Project LightSabre, which is a distributed version of the OSGi EventAdmin (example) service, using Apache ActiveMQ as the event distribution mechanism. It’s been tested on Equinox and Apache Felix, but it should work on other OSGi implementations too. It’s licensed with the Apache 2.0 license

Zoom! See the events go!

Read the Getting Started guide for more information on how to get going with this.

Update: ECF has followed on with a similar implementation at Distributed Eventadmin Service using the same technologies with a difference license. Fair play!

Announcing: FUSE Integration Designer 1.2

Phew. It’s out the door. FUSE Integration Designer 1.2 is now available for download.

This release follows on from a couple of preview releases that we ran to gauge our approaches and get feedback. And we got plenty of it, in all flavours.

You might already know what FUSE is about – it’s four popular Apache projects (ActiveMQ, Camel, CXF, ServiceMix), bundled together into a single offering with subscription-based support from Progress Software. These four projects bring together capabilities around messaging, Web services, message mediation and the ESB concept, the idea being to give you a grab-bag of goodies that make sense when you are trying to solve integration problems.

As you might imagine, it’s an interesting task to put together a toolset that lets you blend these technologies in way that can suit every integration issue. In our preview versions, we concentrated on visualization and debugging of Camel routes and creation of Web services. I think the first thing we learned was that users tended to hit the limits of the tool far too quickly – the routes, for example, had a limited set of capabilities that was outstripped by what Camel had to offer. So we concentrated on making sure that the tool could handle any (1.5 or 1.6) route configuration you could throw at it. Let me know if you break it

We also filled in some glaring holes in the preview – you can now deploy your Web services to ServiceMix 3.x and ServiceMix 4.x containers, for example, and we have put in some tools that will let you rummage around inside your ActiveMQ message broker, introspect the queues and topics, inject messages for testing and snoop on messages that are going by. You can save and reload your state too, so it’s possible to set up and share a set of messages and configurations for testing or joint review.

There’s more, of course.

The previews were delivered solely as an Eclipse update site, which could be a bit of a bear to interact with. This time around, you download zip files that have everything – Eclipse core, dependencies, FUSE code, the works. They are big ok, but it means that you get everything in one swoop. There’s online help in there, and some cheatsheets to get you started.

Try it out – download is here, forum is here – and let us know what you think.

What next? The radar is moving on to things like JAX-RS tooling, Camel 2.0 support (runtimes go faster than tools, that’s why they are called runtimes, natch), getting a deep integration with m2eclipse and such like other tasty treats. If you’ve got a hankering for anything particular, let it be known in our tools forum.

By-the-by, since we are all Twittertastic these days, you can get more FUSE news by subscribing to @fusenews or indeed subject yourself to my edgy waffling at @oisin.

Intermittent Bug in Eclipse Community

This bug has been cropping up on and off since Eclipse 3.2. The usual pattern is that it becomes visible early in the year to the Planning Council, who immediately attempt to triage it and limit the potential solution space so that resolution can be made as cheaply as possible. Because Eclipse is an open organization, and is populated by humans, who are in the main scurrilous gossipers and rumour-mongers, awareness of the bug expands in the community as a whole. A Bugzilla is usually created to give the community the opportunity to cosmologically inflate the solution space through the time-honoured approach of getting your oar in.

Eventually, broad interest wanes, and the determination comes back the Planning Council again, who, disappointed that no really clear solution has come out of the community involvement, many eyes == many opinions it appears, just sigh and put a workaround in. Next year the same sequence of events occur.

This year, it’s bug 271054, but the problem is the same.

What do we call the next roll-up release after Galileo?

Comments about bike sheds will be modded up appropriately.