A keynote, a panel and FOSS
This would be a sort of report of two events -Minidebconf at IIT Mumbai (January) and GNUnify 2015 which happened at SICSR, Pune. #IIT Minidebconf Mumbai 2015, #GNUnify 2015
I would like to start the report with Venkatesh Hariharan’s speech whose key message and takeaway was that it’s no longer the issue that whether FOSS works, the question is now do if we have enough people to scale up the work. There were lot of other things Venkatesh (or Venky as he likes to be called) shared as well some of which I had been guessing and some things I knew but it was still refreshing to hear it from him. I wish somebody had recorded it so I could review the keynote speech couple of times as there were quite a few questions I had for him but later forgot. Anyways, the key takeaway was we need more people.
While, from his address it seemed that he was talking about sys-admins and developers, the fact is we need a whole lot of people. We just don’t need sys-admins and developers but also need software auditors (who are a cross between being a developer and security specialist),Devops geeks, security people,font,theme designers and everything in-between.
The point which was raised there as well as raised in the NEN panel discussion at GNUnify was the same that quite a bit of Government money is used to develop software and that software should be FOSS but it isn’t and basically boils down to nexus between the Government of India and various private parties. If such corruption or improprietry (even if it is just of values) has to be checked then FOSS has to play an important role.
Which brings me to the NEN panel discussion which was headed by Mr. Lalith Kapadia, Director SICSR (Moderator), Danese Cooper (formerly of Sun,Intel,Google and now Paypal), Mr. Prashant Pansare from Inteliment Technologies who’s mainly into data analytics (Looking through raw data to find interesting patterns using some algorithims) and Mr. Anupam Saraf (only term CEO of Pune City). They tackled the same question in a slightly different manner.
The question was more of how, especially in big organizations the same work is done n number of times. People took name of few organizations where they were either working or had worked and for the same client or a different client, the same code was written 3 times by 3 different teams. In most big organizations it is the same, and the reason being the insecurity because the old idea specifically for software development is/was SLOC (Source Lines of Code). The ‘old idea’ being if you write more lines of code it is better and have more job security, whereas in free software/open-source the idea is to refactor your code and make it as small as possible, once if its reaches maturity if possible (and many times it is) make it into a library which can be re-used by one and all. Also the idea is not to monetize everything but keep something on the table for the client which is how and what FOSS is. The idea to cut down the management levels, open-source the code and work-culture within the organization. This is nothing new for us who are or have been into FOSS for a long time but for lot of organizations there is still lot of push-back as they would be more accountable for their time and a newer way of working for them.
What was also interesting during the GNUnify that there were a handful of companies who were recruiting people who had FOSS backgrounds to work on FOSS technologies.
One more interesting question/query which was raised is that why in software development projects the rate comes out higher than MS-Windows quotes. The answer which was given by Mr. Saraf seemed to be that most of these Development contract managers or their underlings are somehow influenced by MS-Windows people, so more often than not, when such tenders are called and people having experience working on FOSS submit the tenders, the quotation is shared with people from MS-Windows background so that the price could be under-cut and then once the contract is given, then the prices are raised under one pretence or the other. While in my limited experience, that might have been one of the factors in quite a few places, it would be nice if others can also share if the same has been true at their end as well.
What was also interesting were statements from Mr. Pansare regarding innovation and market share and how one doesn’t necessarily lead to other. Mr. Pansare took the e.g. of whatsapp vis-a-vis telegram. Now both are real-time instant individual and group messaging. From what little time I had been able to use/see both, I have found telegram is the real innovator while whatsapp has been riding on the popularity wave. I have been able to use both, telegram as well as whatsapp on the desktop. While whatsapp as a protocol is proprietory, IMO telegram should also be looked as ‘half-sourced’ as only the client part is open-sourced, the server part is still closed-sourced. The best app. from both the open-source as well as security aspect would probably be the apps. from Guardian project, namely Chatsecure but that’s going outside the scope of the discussion which ensued. The problem with FOSS is still one of mind share because the target for FOSS also keeps changing. For instance, it’s no longer that we are ok with platforms like Facebook and Twitter which are cloud-based and are constantly looking to liven up with Diaspora. Now what Diaspora has come up with the federation technology, an extension of an idea known as Roaming User profiles. The original idea being great in concept, it never got picked up due to MS Windows inherent poor security culture and people in general. The idea is that you can do that fully with all your posts and re-shares to any other service provider if you are not happy with the current provider is a cool thing.
In fact, Praveen A. at minidebconf, IIT Mumbai made a fantastic presentation with having multiple versions of Ruby in Debian and the resulting issue of not being able to have an apt-gettable version of Diaspora. There is an alpha installer but would probably need some of funding to let many people test it out on real servers, report issues in the implementation and at the end of the day have an apt-get installer, and this will also just be a start as there would need to be a stable and a development channel as it is with most FOSS so sysadmins can have the stable server software running and having real users, with a dev. build running which has few pro users who can checkout and report issues (either in UI, connectivity, security or whatever.)
Behind the scenes at GNUnify there is possibility of a font or two getting released in public domain if some funding and technical issues can be sorted out. While I would not be able to give any more info. about that at the present, hopefully if things pan out then in the future would probably talk and even test out and share some of the more interesting things about fonts, glyphs and things in-between.
P.S. – I am sorry for the absence of the photos, while I have a couple of Praveen’s pictures, GNUnify pictures (specifically the panel pictures) are simply unavailable atm.