Sunday, November 28, 2010

#Wave from #Google to #Apache ?

When Google ended its development of Wave as a product, they promised to fulfil their promise and make Wave open source. As Wave is very much thinking outside the box of dated technology, it is encouraging that Apache is considering to adopt Wave as an Apache incubator project.

With Wave as a truly open source project, new  opportunities appear. There are developers who just come along and start contributing to the code and with Wave as an Apache project it benefits from an organisational support that is credible.

With Wave open source, there is one other aspect that is vital for it to become a success; internationalisation and localisation. Google told us that they were ready to provide localisation support for the core product. Now that it will run at Apache, it becomes possible to internationalise and localise robots as well.

We live in interesting times and with Wave becoming "Wave in a box" the promise of a new paradigm is not lost.

Thursday, August 5, 2010

Wave goodbye

Sheer brilliance was not enough, it is curtains for Wave. Bits and bobs are likely to appear in Google products but the whole notion of Wave as the next generation, game changing product is not going to be realised.

For us Wave proved itself quickly; it was easy to produce a WYSWYG interface for MediaWiki. This generated a lot of interest in many places, it had us generate a lot of air miles and we had something to show and tell.

For us there were a few issues that were not resolved to our satisfaction. Our aim was to provide a WYSWYG interface for MediaWiki and this meant that our complete solution had to be open source. This took just too long, there were understandable reasons but it killed interest in the Wikimedia Foundation. As the Wave interface was to be used for Wikipedia, it was just not acceptable that there was no way to internationalise and localise Wave and its extensions.

I am sad that it is over, here is finally a video with Erik Schmidt talking about the demise of Wave..The sound quality is poor..


Tuesday, March 2, 2010

Can we use Wave to edit MediaWiki on a mobile ?

I spend quite a lot of time helping the Mobile Wikipedia become more prominent. I learned a lot from doing this so I consider it time well spend. The number of languages with sufficient localisation has gone up from 20 to 52 or 34 depending on how you count. We created some concept mobile main pages and we requested the implementation of two main pages for two Wikipedias.

The software used is build in Ruby on Rails and it is developed separate from MediaWiki. It is currently useful to read Wikipedia articles, you cannot edit and you can not sign on or set personal preferences.

The MediaWikiWave project proved the concept that it is possible to create a WYSIWYG interface for MediaWiki. We have a functional editor for MediaWiki, the question is will it also work on a mobile phone, the question is will it also work with other languages and scripts.

I expect that for Wave to succeed, it will support both mobile phones and other languages and scripts. We have been told that internationalisation is part of the core MediaWiki functionality, and as I argued earlier, internationalisation and localisation need proper support for bots and gadgets as well. The one thing I am not sure about is mobile telephone support. What I do expect is, if not great then rapidly improving, support in Chrome and on Android and Chrome OS devices.

Thursday, February 11, 2010

Internationalisation is an architecture

The #wave internationalisation and localisation is said to be planned for the future. I believe that this will be true for the core product. However, this does not give me the confidence I need for the robots and gadgets that are build by all the people outside Google.

When you look at the Etherpad software, you will find that it has not been internationalised and consequently it is estimated to be some 3 to 6 weeks of work to introduce internationalisation to the code. I expect that a similar situation exists for the robots and gadgets but the problem is more complicated because of the many developers involved.

Internationalisation is an architecture and consequently it should be part of the design not only for the core code but also for everything else. A good example of how this should be done can be found at where software like MediaWiki and StatusNet are localised. Both these applications have extensions and for MediaWiki almost all extensions are localised as a matter of cause. The first extensions for StatusNet are appearing at and, it will prove a lot of work to get them internationalised and ready for localisation.

It would be great if a road plan for the total internationalisation for both Wave and its gadgets and robots would be published.. It may be even an idea to consider, it boasts a community that aims to support over 250 languages...

Why people do not get Wave ...

With the release of Google buzz, many publications write that people do not get Wave. They write that it is not widely adopted and consequently a "failure".

My answer to this is: "Wave is not ready yet!!". The idea of Wave is that it is a federated system defined by an open protocol, using open standards and published as open source.

Wave is alpha software, the protocol is not complete and the source is neither complete nor published. Wave is functional because it can be used. To some extend you can develop add-on functionality but realistically only for the published parts.

When Wave is functionally complete, when both the complete protocol and the source code are published it will be the time when developers and companies can start using Wave for real and make it their own.

Sunday, January 17, 2010

#Wave does not work with a bare IE

I am at my sister's home today without my laptop. I wanted to finish a blogpost .. I tried to start Wave and I got the following screen. As I never use Internet Explorer, I do not really know it to be new. In a time where the problems with IE have had the German government advise on its use, I think it is significant..