What We're Working On

Gato and I are sitting together drinking beers and hacking on Ignite code. After a hiatus (too long), we’re both back to working on Smack and Openfire weekly. As I’m writing this, Gato is doing some super low level Java debugging to figure out a strange XML parsing error that we’re seeing when running the Smack test cases. Assuming we get to the bottom of the problem, we plan to package up and release a new beta release of Smack. It includes lots of great improvements, but I’ll leave the details for the next blog post. We have several goals around our weekly hackathons:

  1. Jump start software releases – it’s time to get more regular releases of all the projects going again.
  2. Recruit and empower community leaders – there’s already a large number of people in the Ignite community doing some amazing work on the code. Now it’s time to equip them with the tools to be as effective as possible and to let them take on more explicit leadership roles.
  3. Have some fun – hence the beer

We’re looking forward to demonstrating progress and to keeping the Ignite projects at the forefront of the XMPP world.

How about some rapid fixes for Openfire 3.6.0a (especially the nag message to upgrade). How about some Jive members actually attending the open chats. How about fixing the blatant errors in the the open source plugins that were released. How about actually having a presence again. It is beginning to feal like ignite realtime is the bastard stepchild of Jive. In short less beer more involvement.

Hi Matt,

Thanks for the update. I agree that if you’re not having fun then there isn’t much sense in continuing to work on whatever it is you’re doing. I’m probably like some of the other community members in that I would love to be able to contribute more to the community and/or focus exclusively on the various Ignite related projects but other work commitments keep getting in the way, got to pay the bills somehow. :slight_smile:

While I think the community has really stepped up their efforts in helping other members the thing that seems to be missing is that small group of two or three people that provide leadership and direction.

I’m looking forward to continuing to watch the community grow and prosper.

Cheers,

Ryan

Todd – Jive does not currently have any commercial products based on Openfire/Ignite since we Open Sourced Openfire Enterprise. That essentially means that our time commitment is on a volunteer basis as individuals only.

I think your ideas are all good ones and I hope you can take the tone of my blog in the spirit of good fun. With that said, I’m definitely going to ignore the less beer suggestion.

Hi Matt,

Thanks for the posting. Part of involving the community is to not burn them again like what was done with Openfire 3.6.0 . Who at Jive authorized/forced this release out the door? What is being done now to ensure that does not happen again?

I completely agree with what Todd says above.

daryl

Welcome back Matt. We sure have missed your friendly face around here . Its Friday and I am off to the pub for a pint. Cheers everyone!!

I definitely realize that you open sourced your products and released them to essentially fend for themself in the community. But in many people’s eyes your release flawed productes out of the gate, and have done nothing to fix the flaws that were reported (missing graphics, authentication errors under reproducable conditions). To make matters worse there are no clear methods in place (or at least no easy way to find them) for the community to contribute to these projects as we still rely on Jive to integrate any submissions and build and post the new versions. So while you say the products are open source you still maintain ultimate control and have in a sence limited the cummunities ability to actually treat these products as open source.

Additionally, I find it shamefull to think that any company would take the approach that Jive has with regards to their “open source” products. These are still products with your company’s name on them yet you act like you just don’t care about any product that does not give you revenue. I personally have contributed tons of my time to these project with getting a dime. I have also gladly donated funds to many open source projects over the years. I rarely see a Jive member in open chat anymore or responding to posts ( Gaston being the exception). You have one of the best products out there for XMPP chat, which integrates into your paid products, yet you choose to not commit resources to it. That is a shame. Your paid products will suffer for it.

Todd,

If you read my original post again, I’m not seeing any major disagreements between us. But, I’ll re-iterate:

  • Gato and I have been away from the community for awhile. We don’t intend to let there be any sloppy releases going forward.

  • We agree that external community members are not empowered enough and intend to change that.

Regards,

Matt

I am not saying that there is a disagreement on what was said. I was just stating the ongoing pattern from Jive. It started with Spark. Jive has pretty much orphaned Spark starting about a year ago. The Jive developer moved to a different position somewhere in Jive, and the project never got a new lead at Jive. WinSrev has done wonders with it since he has taken to trying to maintain it on his own. Then came the open source announcement. Almost immediately the developers disappeared from the chat. Then came the plugins, which have cosmetic as well as programmatic errors (which were reported). Yet nothing has been done to fix this. Then came openfire 3.6.0 and 3.6.0a. They have issues a plenty including that upgrade nag. Again what has been done other than once again orphaning the project with the dismissal of the lead developer. As an active user and very active community supporter I find these actions disheartening at best. I would hope Jive would see the value in re-committing resource to these projects sooner than later. You have some very large deployments out there, that you have effectively hanging in the wind. We have all heard the empty promises of future fixes and replacement products (Flex Spark) in the past. I would like some substance instead of more rhetoric.

Todd – I just want to make sure I’m super clear on one point. Gato and I are doing work on the Ignite projects on a volunteer-only basis. Jive still makes very substantial official contributions to Ignite including hosting and maintaining the website, paying for bandwidth, etc. But, Jive is not officially paying developers to work on the Open Source projects at the moment.

Matt, first off I want to thank you and Jive and all the developers for OpenFire, its a great product and incredibly useful. I do have to agree with Todd though, this project seems to be on its way to irrelevance. 3.6.0 was a disaster, Spark has been the biggest squandered opportunity I think I’ve seen in a while (with Pandion abandoned, the client market was there for the taking), and there just seems to be no direction or leadership on this anymore. I can’t speak for Todd, but I have seen a lot of projects that were “opened sourced” with the purpose of getting rid of them/killing them off. It seems we’re well on the way here too. This project needs leadership, quality releases, and a timely schedule. I am encouraged that you seem to realize this too and I look forward to seeing this project come back to life.

All - First of all we have to acknowledge that Gato and Matt are using their spare time to work on Openfire/Spark and we should be more than happy to see them contributing to Openfire/Spark.

Thank you Matt. Thank you Gato.

The contribution of Jive to OF is huge and abandoning a product usually does not come with open sourcing the nice enterprise functionalities we have seen in 3.6.0. As an community, we have to find a way to start own work on the code and contribute code to the plugins and large products.

IMHO the key success factor for all of us is how we can get a steady stream of releases and who can lead the development. We also have to figure out how to review patches and how to commit them to SVN in a controlled way. I would expect us to be o.k. with testing releases (as we have seen with 2.6.0 beta). In the end consuming a software is so much easier than developing patches (my team is still fighting with the installers for Spark :smiley: ).

I would like to ask Matt and Gato about their willingness to spend some time in reviewing and releasing versions. Matt, Gato, can you comment on that?

Kind regards

Walter

Hey Walter,

I totally agree with your vision. That is what I have in mind too and I’m open to reviewing and releasing new versions. We started now with Smack since very important contributions were made to this library and a new release was needed. Our next goal is to move to Openfire and make it live to your expectations. Code review, a proper beta process and good releases are the way to ensure good quality software. Now…all we need is time.

Take care,

– Gato

and maybe finally fix that apache or what issue which is preventing us to login here on the first time…

I echo Todd’s comments above. My place of employment/eduation is one of those large deployments that are effectively hanging in the wind in regards to Spark and the Enterprise features of Openfire. I agree that it is a shame that Jive chooses to not commit resources to products that integrate into their paid products. Their paid products are sufferring for it.

Gato and I are sitting together drinking beers and hacking on Ignite code.” … this explains a lot why the new code is always buggy (;

Open source products which are not sponsored usually develop slowly or die. You may want to add a “Donate” button so companies can transfer some money to the project. Or look more actively for sponsors for different features.

LG

Hi, i’, m really shocked after reading about this:

http://www.andreas-kurtz.de/advisories/AKADV2008-001-v1.0.txt

http://www.andreas-kurtz.de/archives/63

6 months with a big security vulnerability !!!

Please Jive, don’t let die openfire. Assign resources to it. You’ll see the benefits on the long term…

How is being addressed the vulnerability???

Andres, community is preaparing the patch, some plugins need fixing too http://www.igniterealtime.org/community/message/182660#182660 But i think Jive should do the main thing (incorporate patches, release). Or… there are talks to move that project to some other place.

Check this out: http://www.coresecurity.com/content/openfire-multiple-vulnerabilities

Enjoy the beer

A commitment is a commitment, whether you get paid for it or not; When I volunteered at my local church for a few hours every saturday and sunday for different duties, it had become an obligation; i didn’t say to them one day, hey guys , I’m not paid so let me have a day off today for no reason.

we all here know what volunteering is , at one level or another, and I’m not trying to belittle your effort, but you can make this work, alongside with other know-hows. It has been sad to see spark, then openfire just wilt away; you have a solid product(s) here, and you can’t take advantage of that !!

How about keeping it open source , but getting paid for support ? you don’t think you’ll get money from that !?

I continue to see the momentum for open source fade away with Webhuddle project rotting, dimdim rutting, and Vmukti just not caring; i mean, look at Oracle, the app is free, but the support is a ransom.

Between spark, sparkweb and openfire, you would have had a huge investment, like Lotus Sametime server that is being sold for an arm and a leg. I hope that you will see the light soon before the whole issue goes into the garbage disposal.

Peace.