Hacker News new | past | comments | ask | show | jobs | submit login
India Launches Beta Version of Its Data Site (data.gov.in)
125 points by vikram360 on Sept 3, 2012 | hide | past | web | favorite | 60 comments

For anyone's brain food - the app is opensource and the sourcecode is available at https://github.com/opengovplatform/opengovplatform/

[EDIT:] It looks like it was developed on Drupal - (https://github.com/opengovplatform/opengovplatform/blob/mast...)

According to info, this is a Govt of US-India collaboration.

[EDIT-2:] I hope their data dump is not from the production servers (https://github.com/opengovplatform/opengovplatform/blob/668b...) Because I found this on grepping - https://gist.github.com/3608089

"Its", not "it's", is the possessive form of "it". The current headline reads "India launches beta version of it is data site."

This bugs me too, can someone please change it?

One of the core values of any government should be to make available all of it's data freely to all citizens available via API.

Everything being publicly auctioned and the free data available publicly will stop many of the scams of the nature India is going through right now.

I am not sure if any one would, but if someone comes up with a bill that holds the powers-that-be responsible under the "Duty to Report" bill, I would support them.

> "make available all of it's data freely" Not "all" the data. Many datasets should not be shared. Good reasons include: a)security reasons b)ecology reasons (don't say where endangered species are) c)personal data d) current trial

IMO the rule should be: release everything unless there are strong arguments, to not release specific data sets. These arguments should be transparent and publicly discussed. Just saying "it affects public security" is not a strong argument. It should be at least said, why exactly it affects the security. However, there are good reasons, to not publish some data sets.

Make all the information accessible but restrict access based on User privileges. If some piece of information like endangered species location is to be known, user must identify himself/herself and then proceed to use the data. Anything happens to the endangered species, you know who all got access to that information. Similarly for other data.

India currently has the so called "right to information" bill. That Bill outlines what information can be withheld. Now using the same bill government should ideally come up with "Duty to report" bill.

> One of the core values of any government should be to make available all of it's data ... [emphasis added]

On reading both this post and the title of this thread, one might be misled into thinking that "it's" doesn't mean "it is".

It's one of my own pet peeves to rightly use its spelling, having been a GrammarNazi myself for a long time.

But I seem to have grown out of that: http://www.youtube.com/watch?v=J7E-aoXLZGY

Notice the use of the phrase "allopathic hospitals" in a database of number of beds available! It always cracks me up how modern medicine is referred to as just another form of medicine, like Ayurvedic or homeopathic. The Indian government is clearly of the opinion that forms of quackery that are homeopathy and Ayurveda actually work and are worthy of tax payer money.

Several medical doctors I know (including surgeons) recommend Ayurvedic treatments on occasions and I've found them to generally work well (though they generally are skeptical of homeopathy).

Do you have reason to believe that herbal remedies don't work?

Remedies that have been subjected to scientific experimentation to determine if they are effective are by definition "evidence-based", or "modern" medicine. Any medicine that is not evidence-based should be treated with suspicion and skepticism, and frankly should not be called medicine at all in my opinion.

That's reasonable. I agree that medicines that have not been subjected to scientific tests should be treated with skepticism.

Please qualify your claims with sources and/or evidence. I agree with your statement regarding Homeopathy being quackery, but I see no reason to paint Ayurvedic Medicine with the same brush. Unless, of course, our definitions of Ayurveda differ on a fundamental level.

The burden of proof is on you if you are making the outrageous claim that ayurveda works.

One can only shake one's head at yet another individual fully brainwashed by western Pharmacies and mainstream doctors. Folks like you are always to point out any alternative medicine system as quackery while fully ignoring that mainstream medicine has had a very long tradition of copying and patenting any active ingredient that has been proven beneficial by their sister professions.

Ayurvedic research into herbal ingredients follow rigorous testing in labs, peer-reviewed research and case studies before it gets the govt of India stamp, quite similar to how mainstream medicine works, but of course you didn't care to educate yourself on that.

"are worthy of tax payer money": As far as I know, most of these ayurvedic/homeopathy institutes (both for education as well as treatment) do not receive public funding.

The Indian government shamelessly squanders tax money on both Ayurveda and homeopathy. http://www.indianexpress.com/news/no-extra-fund-for-ayurvedi... check this link. Some quacks in the Indian state of gujrat were denied additional funding. On a side note, gujrat is also the state where state sponsored ethnic cleansing happened not so long ago! To be fair, I'd rather they use tax payer money to fund quackery than kill some of the tax payers!

In India, you can even get a degree in the field: http://en.wikipedia.org/wiki/Bachelor_of_Ayurveda,_Medicine_...

It's pretty crazy that this sort of crap still gets attention in this day and age.

Having known people who benefitted from ayurvedic treatment, I tell you that the regimen (medication and food restrictions- you are not assigned one without the other) is one of the toughest I have seen. the people who successfully get through it, swear by ayurvedic medicine. I have been to some of the facilities and met the doctors and they are above average ( by Indian standards). This is why, this mode of treatment in still popular among many in india. needless to say many treatment centers are based in cities and visited by what you might call a more educated and worldly crowd (I mean no offense here, but could not find a better way to say it :) I am willing to forgo my personal observation as an one off, but please qualify your comment.

> Having known people who benefitted from ayurvedic treatment

Anecdotal evidence.

> needless to say many treatment centers are based in cities and visited by what you might call a more educated and worldly crowd

Appeal to authority.

How about some actual scientific evidence? I'm no expert on the matter, but from what I can find on Wikipedia[0], the benefits of Ayurvedic medicne were inconclusive at best.

As someone who has spent time in medical research, I will admit my opinion is skewed. I've worked with plenty of researchers from India, and none of them have given any credence to Ayurvedic medicine. These are people who (like any researchers) will use any and all means available to them to come up with scientifically viable theories that they can use to get grants. If Ayurveda were as great as you claim it is, then they would have definitely given it a shot.

0: http://en.wikipedia.org/wiki/Ayurvedic_medicine#Scientific_a...

is it anecdotal even if it is one of your best friends? I am not so sure. ok, here is some more information. A study comparing ayurveda with allopathic treatment for Rheumatoid Arthritis. http://journals.lww.com/jclinrheum/Abstract/2011/06000/Doubl... "Conclusions: In this first-ever, double-blind, randomized, placebo-controlled pilot study comparing Ayurveda, MTX, and their combination, all 3 treatments were approximately equivalent in efficacy, within the limits of a pilot study. Adverse events were numerically fewer in the Ayurveda-only group" Ayurveda is recognized by nih as alternative and complimentary medicine. http://nccam.nih.gov/health/ayurveda/introduction.htm

I think at some level, you need to believe not all doctors wear white coats. modern medicine is a few hundered years old , but people have been getting sick and have been treated since the dawn of time. When a country sets up regulatory agencies ( http://www.ccimindia.org ) , qualifies practitioners and monitors the practice ( albeit poorly) , to dismiss the structure based on wikipedia articles is incorrect to say the least. if only conclusive evidence , agreed upon by everyone is acceptable we should have stopped taking aspirin a long time ago:) http://chronicle.uchicago.edu/950817/aspirin.shtml

Definition of anecdota:


"Best friend" has nothing to do with quality of data, in fact it might taint it. Also, causation must still be established, i.e. uncovering the mechanics of the treatment.

Also, your last few sentences are difficult to understand.

I like how you start by pointing out logical fallacies in someone's argument and then go on to say things like "I'm sure they would have definitely given it a shot if it's as great as you claimed".

Like I said elsewhere here, I know several medical doctors (well qualified ones as well) who recommend ayurvedic treatments in certain situations (not the commercialized medicines you get in shops, but more basic 'take these things and grind them and eat it before lunch' medicines and they work reasonably well.

I think I'll take my anecdotal evidence over your link to a Wikipedia article stating inconclusive evidence.

You can believe whatever you would like to believe, but the fact of the matter is that there is no scientific evidence to suggest that ayurvedic treatments are effective whatsoever.

It's scientifically proved on various occasions.

For instance an year back for patient with chronic trigeminal neurological pain, we have given cyberknife treatment. Most advanced radiation thearphy for this kind of pains. Initially everything was ok and the pain bounced back in 3 months. We kept the patient continuously in pain killer for so many months, before they found a very old Ayurveda practitioner somewhere deep inside southern part of India. He assured them they can take the ayurvedic medicines along with our allopathic medicines and can discontinue all the allopathic and pain killers gradually.

We have accepted to that as honestly, we left with no other option as this is an very rare case and we recorded the patient status on day to day basis. After 2 months of consumption, to our surprise, the patient has shown gradual improvements and discontinued the pain killers in the 3rd month. After an year, she discontinued ayurvedic too and living a normal life.

We still have the entire medical history of the patient, and using ayurvedic as a lost resort for trigeminal neurologia issues. But I agree, this is a rare case, and it has to be thoroughly experimented for general practice.

"Various occasions" do not prove anything. Sample sizes need to be much larger, and even then, the mechanics of the treatment must be uncovered.

Also, there is no reason to use "scientific" as a qualifier for proof. There is either sufficient evidence to constitute proof, or not, but there is no distinction between "scientific" proof and other proof.

You think ayurveda is crap? What makes you think that?

Brave move to open source the project - and handy, as it seems they could use some friendly advice on Drupal security and best practices, e.g. https://github.com/opengovplatform/opengovplatform/blob/mast... vs. http://ogpl.gov.in/superadmin.php ;)

Why is it that contires such as india seems to be behind in the way of website design, the design is just too busy for me and the styles/colours remind me of early 2000 sites.

Yeah, when I saw the scrolling text I wanted to gouge my eyes out. It's straight out of 1999.

And for a country where mobile based access will presumably be quite common (as only 7% of the population has home internet), why isn't there a mobile site at launch?

About design and usability: Most of the people accessing the web from India don't focus on design as much as they do on content and or the purpose. This is changing though as more and more consumers and developers are paying attention to UX and other such aspects.

About mobile based access: I don't have the specific numbers at this point, but most of the mobile access is through 'dumbphones'. So not having a mobile site is not exactly an issue.

Almost everyone has a mobile in India. But not many use Internet on their mobile devices. Situation is changing slowly with the arrival of really low - end smartphones, though.

Most of the government IT projects won't have any designers (or professional designers) unless they are related to designs as such. The designs are mostly created by the programmers over there and most of the times it will be to the liking of the bureaucracy or who ever is in charge there. Appearance is last priority.

I really think that's a gross generalization of a billion (and change[1]) people!

[1]: The "change" is about 77% of U.S. population.

Have you looked at sites by computer scientists? Like some of the Microsoft Research related sites? They serve the purpose. Don't expect them to be pleasant on your eyes.

How about comparing apples to apples? Take a look at the equivalent US site[0]. It's pretty obvious the Indian equivalent blatantly ripped off the design - same sort of bar at the top, then a header with a logo, then a horizontal set of links to diferent places on the site, then a set of rotating images, then links with in-depth info in 3 columns.

Some of the UI differences include the addition of scrolling text and a crappier color scheme.

0: http://www.data.gov/

Did you miss the point that this is a India-US collaboration.

From: http://ogpl.gov.in/about_us

Open Government Platform (OGPL) is being developed by nodal agencies from both the countries (National Informatics Centre, Department of Electronics & IT, Government of India and Office of Citizen Services & Innovative Technologies, General Services Administration, U.S. Government). To know more about the participating organizations visit following links

National Informatics Centre (http://www.nic.in)

General Services Administration (http://data.gov)

I don't think I've ever seen so many separate JavaScript and CSS-files used on a page before. Surely most, if not all, could be reduced into one?

In Drupal any module can inject its own separate stylesheet into the page. It's standard practice to enable Drupal's CSS (and javascript) compression feature once you move the site to production. That they haven't done it suggests the developers aren't very familiar with Drupal, or easily overlook details.

Drupal 7 does this out of the box, with an option to aggregate and compress JS and CSS. I have not developed on Drupal 6 but would be incredibly surprised if there wasn't at least a community developed module to do this.

If I remember correctly this has been available out of the box at least since Drupal 5 up. Have worked on Drupal 5 and 6, don't remember using any custom module to do aggregation.

It's rarely beneficial to compress to one single file, but I completely agree that the way they are doing it it very inefficient, especially spamming the @import notation.

Not sure what you mean by "rarely beneficial to compress to one single file".

But in multiple situations I have seen significant web load performance improvements when the number of discrete files being loaded is reduced - merge all JS into one file, all CSS into another and sprite images. This is due to HTTP 2-connection rule.

And slow-start certainly contributes.

This site is absolute trash, pretty much like anything else the NIC creates. I've tried to register using a dozen possible methods and there are all sorts of errors. Seriously, how hard is it to allow someone to create an account, beta or no beta? They have this ridiculous list of ways to sign up including Foursquare and PayPal and almost none of them work. Yes Paypal, really.

So after about 20 attempts I now have an account, but if I try to download a data set, it takes me to my profile page.

For those who are interested, there are a bunch of us discussing open Govt. datasets (for India) at the DataMeet Google Group - https://groups.google.com/forum/?fromgroups=#!forum/datameet

Oh gods, the source code please no please kill it. 20+imports in css, 20+ script tags please no

For some reason the link to 'Dataset' doesn't work for me. However, this link works: http://data.gov.in/search/apachesolr_search/?filters=type%3A...

Has anyone tried to actually download a data set? It seems you have to register first. Is there any good reason why an open government initiative would need to require registration to access the actual data?


Am I the only one who finds it strange there are no buttons to change to languages other than English (or if there are, they weren't obvious enough for me to find)?

What's up with the dark color scheme?

It is a very encouraging step from GoI, looking at the recent slur about rampant corruption in India [http://news.ycombinator.com/item?id=4447605]. That said, I do feel that the task of opening the Government doesn't end at launching a portal (yet another CMS) or giving out yet another data-set.

Looking at how much impact these platforms have induced even in developed nations there seems little incentive [some advantage definitely] to pursue such an initiative in a country like India. Millions die of basic malnutrition there, and perhaps their focus should be to make food available to consume, rather than data.

Of course the data will help but aren't we missing the big picture?

It is relevant to note that incentive for active development on Gov data-sets like these (believed to improve the existing establishment) is probably not as attractive an idea as it is to disrupt the legacy itself. That's how the relationship has been between the establishment and disruptive thinkers for centuries.

I mean all this seems like a complicated play of a ton of variables - democracy, education, impact, profit, consumption, need etc. Yet no doubt the launch is certainly a positive one from the largest democracy in the world.

Also, it is relevant to note that incentive for active development on Gov data-sets like these (believed to improve the existing establishment) is probably not as attractive an idea as it is to disrupt the legacy. That's how the relationship has been for centuries.

Actually, it is part of the GoI's "Data Sharing and Accessibility Policy" that was made law. http://www.india.gov.in/allimpfrms/alldocs/16473.pdf

Looking at how much impact these platforms have induced even in developed nations there seems little advantage [some advantage definitely] of pursuing such an initiative in a country like India. Millions die of basic malnutrition there and perhaps the focus should be to make food available to consume, rather than data.

That is an extremely uninformed statement - one of the biggest challenges for efficiency in governance is accountability. Think of the govt. as a large legacy codebase, with the QA being done only by the parliamentarians/senators. Release of data allows QA by the ordinary citizens. India has a judicial process called Public Interest Litigation [1]. Combined with public datasets, it makes for powerful governance.

[1] http://en.wikipedia.org/wiki/Public-interest_litigation_%28I...

Perhaps you might want to educate me on it with some examples where there has been successful impact on the Government.

"...one of the biggest challenges for efficiency in governance is accountability" is the problem to be solved, not a proof of success of any such initiative. Your argument doesn't seem to explain how we're going to improve the legacy system or even convince the developers to use these data-sets in the first place.

To crave for improvement in the Government (and its positive ramifications) is one thing but to get job done is an altogether different ball-game.

More information always helps - especially in a democracy. This data is not for developers to make use of. It's just data available for everyone to use.

Guidelines | FAQ | Support | API | Security | Lists | Bookmarklet | Legal | Apply to YC | Contact