Deepening the GNU/Linux community
This post would be about things needed to be done to get together to get GNU/Linux pervasive in our community. Some of the things would be about Ubuntu in particular while others would be more about GNU/Linux in general.
I would first describe the issues which the Testing community specifically the Ubuntu community finds itself and then go on to General issues .
1. Less time :- After a release happens, the testing repositories are opened a week later. There is heavy breakage at that time. Some of the advanced/power users usually enter the testing cycle around the 4th week. I usually also do it around that time.
Now the things to test are the kernel releases, the various pieces of the GNOME ecosystem , Pulseaudio , Openoffice.org , Pidgin and whole lot of tools, libraries and services. One of the issues is that new features keep coming in till the 4th and sometimes the 5th cycle, so testing before the release is limited to 1-2 months before the release happens. It has been argued many times that is too short a cycle . The other one is something like what Debian does which has come down to a year or little more bit of release cycle which perhaps is a bit more upto things. The short release cycle usually means apart from the major packages (which are taken from the popcon stats gets major love but other tools perhaps not so much. Most of it is community-based and the community needs to get active on that front.
2. Hardware Database and Compatibility :- The hardware database User-interface on Ubuntu needs to be more user-friendly. There has been a community initiative by the name of Ubuntu HCL which is a good start but again fails on many accounts. One of the better ones is smolt by fedora . We, the users have been too patient in not asking for a vendor-neutral GNU/Linux stats page where test-cases and user’s perspective give an idea of what works and what doesn’t.
Also having a vendor-neutral site could also help closing holes/issues when some hardware some doesn’t work across distributions. This has also been discussed in the past with nothing happening till this time.
3. Device drivers :- Related to these are device drivers issues. While branded hardware gets device drivers and are supported, knock-offs from chinese or elsewhere are troublesome. They may work or not. I know of couple of projects which tried to address this issue but there was no community involvement. One of them had one of the Indian fedora developers putting quite some time to get it started circa 2006.
4. Bug-reporting, Triaging and Resolution :- The sad fact there is still too much turnaround time between a bug is reported and its resolution and I don’t really mean the complex ones. Most of the time if I want the Ubuntu developers to have a look at them, I have to be up at their time and show them the bugs and talk it out. What is needed is Indian developers to take some of the heat away from them. The Users/developers should have triaging as well as priority rights so the issue can be triaged and be closer to resolution. The main issue is taking ownership of some part of the ecosystem.
5. Education :- A related issue is education. Most of the Engineering colleges in our country are mired in mono-culture. Apart from issues outlined in the wikipedia stub, there are cultural issues as well.
From an engineering perspective, they never really learn to troubleshoot and work in a distributed society, work on real problems which are faced by people. Most of the students have a ‘ I can’t do it mentality ‘ and its far safer to make a project for some company where the work may be used or not rather than doing some real work. There are some exceptions to this rule, but those are exceptions.
6. Internationalization :- One of the other things is localization. Majority of the distributions are Net-centric which means English is the preferred medium. While we have had IDN for quite sometime now people are not aware of them. People are still amazed to find websites and e-mails written in an regional language. I know of people who write in their regional language and people ask them how did they do it. Add to this the whole confusion or understanding of dialects and you have a big tower of Babel.
What it also means is that there is still a subtle top-down approach where the english-speaking population have a way which they think is the best approach either to the User-interfaces or any number of things. If more joe’s were involved while designing interfaces or any decisions taken perhaps they would be more intuitive.
7. Culture :- Most of the people I know have ‘I/We can’t do this’ lack of self-believe. Also most of us have to have a thick skin when we are chatting with native English speakers. I’m not being racist or anything but its a fact, you have to show you are twice as better as them to get anywhere. This is also a reason why many people give the struggle half the way.
Does this mean that all is lost. No, not at all. This simply means we need to co-ordinate our efforts and do things right. The real challenge is how we get there.