Category Archives: Free/Libre/OpenSource Software

My jottings on FLOSS

Where is Indic L10n nowadays ?

Indic L10n is not moving. Neither forward, nor laterally. After the initial burst of energy and explosive shock of languages nothing is happening that will keep it exciting. True there has been sporadic news of community projects in languages being formed, but nothing in terms of concrete roadmap has been coming forward from these.

In a way it is a sad reflection of the state of affairs. The Indic L10n meet at HBCSE had put across a roadmap in terms of formation of a body as well as encouraging and incubating other languages. Not much has happened since then. Surprisingly enough, the existing community projects also seem to be afflicted with some sort of lethargy. The next step towards localisation of the desktop would be selecting applications to localise. Applications matter. And especially in the current scenario, applications which address the various demands of e-Governance workflow would be something that will drive forward OpenSource adoption.

Applications could include software in the domains of Content Management Systems, Workflow and Groupware suites, Collaboration Software, Management and monitoring tools. The community projects could begin to look towards such existing tools which are mature and are i18n ready to take the effort forward. Additionally, there has to be some means to get in grants for the projects. This is primarily related to developing a richer range of OpenSource fonts, dictionary and spellchecking programs. A method is also required to be found out to interact with the various government bodies which have such research projects in progress. Sharing the fruits of the labor would lead to a rapid release of the final product. And perhaps a much stronger Free/Libre OpenSource Toolkit with which to address Government initiatives.

Sometime back I had written about creation of the L10n Framework as part of the L10n roadmap for Indic languages. The need for the Framework is felt more intensely now. Creation of the Framework would allow a greater number of end-users to play with the toolsets. Such user participation would lead to innovation and feedback which will help the L10n process in the long run.

Usage of access keys in Indic L10n

One of the aspects of Localisation (L10n) which is not being addressed very aggressively in case of Indic L10n is the issue of accelerator keys and keyboard shortcuts.

On the face of it there are two ways of addressing it:

[1] Using access keys in the local language with standard patterns for the menus
[2] Maintaining the same access keys as the English ones

Both approaches have their takers. The first one is generally adopted by nearly all the Indic L10n teams while localising the desktop environments. However, in the long run the second approach might just be something that one would require to look at seriously.

The first approach is already in place for Spanish software where Ctrl+U is the key combination for New (under the File menu) rather than Ctrl+N. This link puts forward the argument wherein the need to keep in sync with the MS-Office shortcuts are put across.

Putting in the English shortcut keys are sometimes an aberration. In the normal case, access keys are supposed to follow the natural flow of the local language. Thus, it could be easier to pick a character in the translation that is easy to remember (or perhaps following the Microsoft or MacOS guidelines). The logic being that consistency is on any day a good thing as it helps people to migrate from one platform to another. Trying to use the same access keys (as the original) makes access keys unnatural for the users of the localised distribution to remember.

In case of Indic L10n, where there is a non-latin script using latin characters as access keys would be a better option as native characters are (sometimes) more difficult to input. The way out in this case would perhaps be using the English access key in parenthesis along side the translation. This would also ensure that a consistent pattern of access keys are used by the users while they are switching between Indic locales and to English and back again.

Should we be looking towards a Compliance Lab in India ?

FSF has a GPL Software Certification Program and Compliance Lab. To quote from the link –

The certification program provides corporations with the assurances they require when building products upon a Free Software infrastructure. When you purchase certification, FSF undertakes a comprehensive engineering and legal review of your software release to ensure that your work has been done in compliance with GPL and related Free Software licenses.

. The question is that should we begin to look at a similar function in India ?

This is an important issue when considering the various structural components of Indic L10n. More importantly, when considering the important issues of fonts and converters. My dear friend Venkatesh Hariharan often laments about the availability of good quality of fonts in the various Indian languages. Even those that are available are having unclear licensing regimes. Perhaps this is one of the reasons which prompted Soumyadip to make this post.

The Free Software Foundation has a strong presence in India, and given the stellar role it has been playing in terms of influencing policy decisions related to use of Free software in e-Governance, Education and the like, perhaps it will not be out of bounds for it to take it on. Being a proud member of the body, I am well aware as to the shortage of manpower and the sheer lack of resources. Perhaps this is the best time to address these issues.

When discussing the idea amongst a few friends, the one refrain that I heard was that this might lead to a duplication of the efforts. I think not. Prima facie it does. But if we look deeper into the matter, the members of FSF-India are very much aware and involved in the Indic L10n process. In fact, the Indic L10n community is small and tightly integrated. Thus, a preliminary validation of the licenses could be easily carried out.

An immediate benefit of such an activity might be the prevention of case where unwittingly GPL/GPL-like license violation takes place.

Umm…work to do

This week the residence was full of people from the Mumbai office. Training was scheduled for the Delhi and Kolkata teams across Saturday and Sunday and thus we had Satish, SandeepK, Palashendu and Sukanta staying over.

BDC has been here in Delhi for sometime now and one of the good things of such a thing has been that we spend some amount of time talking and walking. Have not walked in a long long time now…it is a good thing. Best part is that I get to know a lot of stuff (mostly undocumented) on the issues of the Operating System with databases, shared storage and the like. Good knowledge and a very good store of documentation.

Have a lot of books pending to be read…have to get down to the job. Additionally, the Linux Device Drivers 3rd Edition is waiting to be reviewed.

Playing around with stuff

For sometime I have been playing around with information retrieval tools with the basic requirement that creating my own APIs through extension would be easy. Another requirement is that the existing documentation should be near complete. Take a look at Xapian for those who want the same thing. Here is a sample code of the indexer. Mind you that this is highly simplified. Sometime later would start playing around with Lucene.

Current reading list includes stuff on this.

Here and here are 2 nice news pieces