What about Pandion?

The Spark client looks good from what I can tell. It seems to work well, too. Good work guys.

It does, however, resemble Pandion (http://www.pandion.be) considerably. Have you guys worked with the Pandion people on developing this? What technologies are you using for the UI? It doesn’'t look like Swing so much to me. Is there any possibility these projects could converge and ultimately deliver a far superior cross-platform Jabber client that what is available now?

kzantow,

While we’‘re quite familiar with Pandion, it wasn’'t used as a basis for the UI. However, a goal for the client was definitely that it feel familiar to EIM users. Therefore, it borrows UI themes from many of the non-Jabber clients. Most professional clients seem to have a similar professional look and feel, which is why I think it somewhat resembles Pandion.

The UI is in fact fully Swing. Perhaps Derek can provide some more details about how we achieved it. Unfortunately, the implementation between Spark and Pandion is very different so I don’'t know what (if any) collaboration would be possible.

Regards,

Matt

kzantow,

The client is indeed written just in Swing with a chat event model behind it. Swing is the best One of the nice things about Spark is that most areas are pluggable, and if they are not, well they need to be. So everything from the presence manager, to the different types of rooms can be customized to fit your needs. The idea is to develop the best jabber client out there by listening to and getting help from the community. So the question really is, “What are all clients including Spark missing that we need?”. We get those answers and implement it quickly to get it back out to the community. We are also hoping some members of the community are interested in writting some plugins, but more to come on that.

I just tried Spark today and it seems to work well, but I’'ve noticed that it uses more resources than Pandion.

Any thoughts?

As a Java application, it is inevitable that we will be somewhat higher than a native written application, however there is still much more optimization to be done on our end for future releases.