All posts by Sankarshan Mukhopadhyay

Anecdotal IM converations

A long time friend suddenly pops up on IM with pearls of wisdom that begged to be noted down. Speaking about “knowing when it is time for a change” it as opined that the following are good indicators:

  • A feeling of “Oh boy it is office day again” when Monday morning arrives
  • Restleness or boredom
  • A recurring feeling of repitition in the work
  • Finding non-work areas of life immensely interesting
  • Inability to visualise a future which would be on the career growth path
  • Envy of what others are doing for work

This was in the context of a large subset of mutual acquaintances changing jobs this season and the thing we came up with was the “One hand Test” ie a hand has 5 digits – the as the number of questions that if asked might tend to avoid the feeling of “darn why do I stick around in this job”. Here they are:

  • What are the kind(s) of work that appeal to you ?
  • Which acitivities/tasks result in more satisfaction for you ?
  • What kind of people do you like to work with ?
  • What is your kind of work environment ?
  • Which areas/abilities do you think you can develop upon ?

Interesting exercise 🙂

Moving towards an Open Audit voting system for the GNOME Foundation Board of Directors elections

During the last meeting of the Membership Committee, we had Ben Adida talking about using cryptographic methods for the elections. Mind you, the current method of holding the elections sure needs an overhaul and as proposed in the minutes, Ben has written out his proposal to the Foundation List. He’s also put up a wiki. So make sure that you put in your feedback. The next stage of course is the implementation of the theory.

Long discussions and then a post…

Here’s the post subsequent to the long discussions on #linux-india. That someone referred in the post is actually the same person who wrote this post. Now, to move forward on this thread I thought it was a good idea to clear up some misunderstandings (the caveat in my earlier post was part of the same reason). A rockstar is someone who fits the following bill: “The girls want to be with him, the guys want to be like him” – the definition was provided by t3rmin4t0r when we began discussing the theory.

There is a good reason why I am opposed to the “We need rockstars” line of argument. The usage of the word as a metaphor tends to load too much unnecessary context into it. The preloading thus morphs the true form of the word into something that’s undesirable unwarranted. However, as the discussion progressed, I realised that the “rockstar” was more of an allegory than a metaphor and thus was a rhetorical flourish. A metaphorical usage of the word normally tends to propagate the urban legend of “the last rockstar standing” who delivers, commands and leads a band of followers to the chosen paradise. The trouble I have with that notion is that the concept of “community” that replaces the band of followers is not something that is all prevalent in the country right now. In current times what also goes missing is the concept of “commons” or the space for “protest”. Protest in the truest sense of the word does not mean a stance against current, but it leads to a position of an alternative theory. The current social mesh does not bear witness to too much of alternative theories.

Does it have something to do with the system of education ? I tend to believe so. The system puts higher stress on individual excellence and this translates into reduced sharing or collaboration. The “harvesting” of knowledge is not for common good but for individual gain. Given that consumption and creation of FOSS (code and non-code FOSS) is yet to see more rural penetration it does provide an insight into what an icon like rockstar might do. Icons, figurines, deities or titular heads are good to hold on to as yardsticks. The idea that a rockstar might encourage enough enthusiasm to attract a merry band is the precise nightmare. A rockstar is an icon and save a breaking down of the icon, what will emerge is monoculture or more or the same – an emulation. Emulations are a constricting feature – the emulatee (if there is such a word) tends to constantly question the peer group perception and thus curb creativity.

What would be of more use is figuring out what needs to be done to get more FOSS (code and non-code) to a larger base of people. The greater the number of users consumers, the fairer would be the chance of getting more contributors. Do we need evangelists right now ? No, but we need mentors and they would only come out of the folks contributing. Some would make good mentors, some not so good, but we need them all the same. We don’t need rockstars – they tend to complicate things.

Of “rockstars”, contributors and all others in between…

[Small note: In case you think that my blog post disagrees with Gopal’s blog post I’d suggest you read both of them again. Effectively they are the same stuff although I’d like to elaborate a bit about the rockstar in a later post]
There’s been a mighty long discussion about “What’s wrong with the FOSS scene in India”. The actual question might just be something different really – more on the lines of “what’s being contributed where and by whom”.

The myth of the last “Rockstar” is a bit too tempting. Rockstars are “icons” – larger than life and more in the line of what can be called ideology sometimes thinly bordering on hagiography. The other thing that bothers me is the slowly growing urban legend that “code is FOSS and non-code contributions are not FOSS“.

Here’s what I see the situation as.

There are contributors from India (and we are not talking about of Indian origin). For example, there are around an approximation of 11 Indian languages being supported or partially supported in GNOME. In precise terms that means there is at least one person doing commits for his/her language for GNOME releases. Now, take into consideration KDE and XFCE and you have a fair idea of the number of commits that go in. Now think about “maintaining” the language (in all its aspects of correctness and linguistic behaviour) for multiple releases as well as for downstream projects like Fedora and you have a large base of contributors. Contributors and not rockstars since evidently no one seems to talk about them.

Since, we are on the subject of localisation what follows as a due course is internationalisation or i18n and here there’s a fair lot of work that is being done by Mayank (Evolution), Parag (Fonts), Sachin (Qt) among the others and the work is upstream and in bugzillas. But they are contributors and they get the work done. Not rockstars mind you.

And then of course there is Shreyas,Pradeepto, Sayamindu (who is being unfair to himself by doing cool hacks for deployment but not blogging about them) who keep on doing their bits around FOSS. Sayamindu has been decidedly active on Sabayon, Conduit and surprisingly Exaile (he got me using it and filing some enhancement requests too) but he is a contributor not a rockstar.

The legend that things are happening if only there is mass hysteria seems to work for retailing and thus that does in no way hold for contributions. If one does not see contributions, then one should look around more and ask the folks who might have the information. An off-the-cuff remark that things are not happening is not really the way forward. One of the side effects of being entrusted to process the applications for membership to the GNOME Foundation is I sometimes manage to get a ring-side view of what’s happening (at least in GNOME) and trust me – there’s no problem with FOSS in India. What needs to be done to augment the process is create the second rung of contributors by:

  • Creating the infrastructure in which they can contribute
  • Create a process by which more meaningful tie-ups with upstream can occur (the Lord of the Code will require a reworking to do this)
  • Get the largest possible bit of the Government sponsored/funded/proposed FOSS projects doing work transparently

We are here to rock your world…

We are back. In case you did not notice, we had a meeting last week and since then a large chunk of the ActionItems have been worked upon. We have new members and I daresay that the GNOME Foundation would see a more responsive Membership Committee. The old adage of a stitch in time saves nine holds all too true for us, but with a larger number of folks looking into the RT, we expect to have much decent coverage and thus better turnaround times.

Defining standards – need or semantics ?

Venky writes about his observations on the state of affairs at the BIS Summit (incidentally I guess this was the one where Craig Mundie muddled up standards, specifications and intellectual property) and concludes that a lot of ground needs to be covered for getting the concept of Open Standards across. Here’s what the Wikipedia says about Open Standards (italics are mine):

Open standards are publicly available and implementable standards. By allowing anyone to obtain and implement the standard, they can increase compatibility between various hardware and software components, since anyone with the necessary technical know-how and resources can build products that work together with those of the other vendors that base their designs on the standard. Many technical specifications that are sometimes considered standards are proprietary rather than being open, and are only available under restrictive contract terms (if they can be obtained at all) from the organization that owns the copyright for the specification. 

There are examples of such standards here.

I mildly disagree with Venky here. The one singular issue that has stood out at every (in)formal body of standards folks I have attended is the mixing up of standards with specifications and therein move on to the slippery slope of IP (Intellectual Property). Apparently, there isn’t a way one can traverse that ground however given that the standards committee/bodies are normally populated by gentle folks from the world of research the captivating essence of IP is not far. What does not however get discussed is the need to distinguish IP from public knowledge and this I think would put a finer understanding of the concept of Open Standards. The key context of the usage of the word standards is that it could mean a common set of guidelines that facilitate interoperability. Thus, taking an example, the font rule-sets are public knowledge but the glyph shapes (the cursive elegance perhaps) is an intellectual property. I am not a lawyer so I might be way off track in this example but it suffices to convey what I am trying to get at.

The need of the hour is to detach the intellectual property bit from Open Standards so that comprehension of the concept and usage of the same is facilitated. This would be more important in cases where one is talking to the government at the state or central level to adopt and use more of Open Standards. The explanation should also include aspects of how the economics of business is not harmed by adopting Open Standards.

Innovations, platform and v12n

When does innovation occur ? When the tools that drive and facilitate the innovations are available in a state of maturity. What does innovation mean ? It means a lot of things, but most importantly what it means can be summarised as:

  • Increased quality (and utility) of the product
  • Reduction in cost (cost to consumer and cost of production)
  • Scalable products which are easily commoditised
  • Affordable commodities

In fact the above can be wrapped in the catchall term as qualities of innovation. Thus, a lowering of prices and/or complexity of the tools that lead to lowering of the development barriers result in an increase in the entropy mix of innovation. So where does this fit in ? It fits in the scheme of things of the (in)elegantly termed LongTail. The current tools that enable creation, maintenance and deployment of solutions have reached a state of maturity that will enable a larger number of small developer outfits to develop and deploy FOSS based applications. Couple this with v12n (virtualisation) and it is no longer necessary for the application to keep on chasing and being autobuilt for the base operating system. This is perhaps as good a time as any to start thinking in terms of application stacks that bundle along with the operating system and sit as a guest stack on a virtualised platform.

Inflection point for localisation

An “inflection point” can be defined as a point on a curve in which the curvature changes sign [1] ie the curve might change from being concave downwards to concave upwards. The next inflection point when it comes to discussing localisation in Free and Open Source Software [2] is the deployment of localisation in Citizen Centric Services or what is covered by the catch-all term “eGovernance”.

If one is tempted to figure a linear equation as to what stops localisation from being deployed in eGovernance it would be like as:

No project scoping <- Lack of applications <- Lack of specifications for applications <- Lack of standardisation <- Lack of a resource center. However simplistic that appears a large bit of it is accurate. So, to begin with, let's talk about standards [3] and especially Open Standards [4]. These are are publicly available and implementable standards. By allowing anyone to obtain and implement the standard, they can increase compatibility between various hardware and software components, since anyone with the necessary technical know-how and resources can build products that work together with those of the other vendors that base their designs on the standard. The standards that would be relevant are Unicode, W3C, LISA etc that would make the end user experience complete and useful. Add to this standardisation of glyphs. What this ensures that the solutions are interoperable and consistent. This also means that the user experience and user goodwill is enhanced. Accessibility or a11y is also the fallout of using standards in deployment. The next stage is creating Specifications for the applications. Specifications which leverage Open Standards and create solutions in an Open Architecture allows the creation of applications by the developers which are scalable and (re)usable. If the developers have incentives to create applications adhering to specifications, the roadmap for deployment becomes much more easy. Once these are in place, the ideal way forward would be the creation of a resource center. The top level objective of the Resource Center would be a framework of collaboration between Industry and the Research Units involved in Localisation Research. The Center would also include linguists and Special Interest Groups who would have innovative UseCases for Capacity Building. Building of skills would be required in the areas of IT infrastructure operators, IT managers, developers and deployment support personnel. Additionally, the Resource Center would be involved in the tools and solutions required for deployment of FOSS, incubate entrepreneurship. Most importantly it would participate in Policy formulation in the various relevant fields. References: [1] http://en.wikipedia.org/wiki/Inflection_point [2] http://en.wikipedia.org/wiki/FOSS [3] http://en.wikipedia.org/wiki/Standardization [4] http://en.wikipedia.org/wiki/Open_Standards