India’s Flipkart to shut down website, thanks to popularity of mobile | Hacker News Reader

India’s Flipkart to shut down website, thanks to popularity of mobile

geekwire.com

91 pointsrohanaurora75 comments

I am an regular Flipkart User, and I do not like this behaviour, Flipkart is forcing me to move to an mobile app, just to have better control and tracking.

They have already stopped supporting the mobile website, and it simply instructs you to download the app to continue.

This is the moment I have completely stopped using flipkart and moved to amazon.in

I like website more than app, both gets the job done, where as In an mobile app I have to give various permission which is not required for a simple E-Commerce website (which suggest the are just doing it for the data)

See the list of permission used by the app below.

  Device & app history
  retrieve running apps
  read sensitive log data
  Identity
  find accounts on the device
  Location
  approximate location (network-based)
  precise location (GPS and network-based)
  Photos/Media/Files
  read the contents of your USB storage
  modify or delete the contents of your USB storage
  Camera
  take pictures and videos
  Wi-Fi connection information
  view Wi-Fi connections
  Device ID & call information
  read phone status and identity
  Other
  receive data from Internet
  use accounts on the device
  view network connections
  full network access
  control flashlight
  prevent device from sleeping
  read Google service configuration

I've never used Flipkart as I'm not in India, but just browsing to their website seems to work on Firefox on Android. I just check the "Request Desktop Site". It's a bit janky but doesn't seem unusable.

Though I agree, taking business elsewhere is a better response to this kind of privacy-invasion. (And Google should shoulder a ton of the blame here, as they not only allow but encourage this kind of shit.)

What the freaking hell. Android permissions are jacked up.

Oh, don't worry, it gets worse:

* Android does not let you selectively grant permissions (unlike iOS)

* Android expects you to give an app all permissions at once at install time, carte blanche (unlike iOS)

* Doing certain very basic things often requires a broad permission - for example, checking if music is playing so the app can mute itself? "Read phone state" - which includes the phone number you're calling

* Google are actively making this worse by making Android permissions broader on Google Play - rather than accept something very specific, you now accept an entire category!

Potentially even worse, Google had an excellent per application permissions system they were working on in Android 4.2 & 4.3 called App Ops that worked similarly to iOS, but that project appears to have now been cancelled: http://techcrunch.com/2013/07/26/android-app-ops/

Do these problems exists on rooted devices also? Is there any work around or some tweak?

Cyanogenmod still includes the app ops.

Thanks for creeping me out, Google. https://news.ycombinator.com/item?id=9430839

I'm really interested in expanding my mobile development into more Android but stuff like this sounds awful. I hate the idea of offering say, GPS or datastorage to a customer but not letting them opt-in to it.

The above concerns would never bother (coz they are not aware) average population , which is more than 95% I guess. Rest is something Fipkart may be willing to trade off.

I guess it will be damn interesting to see how this pans out for Flipkart. FWIS, Flipkart is very unhappy that it wasn't allowed to control (yes control!) the market over web with its Airtel Zero plans. Read more about Net Neutrality [1] events in India lately.

Amazon India is certainly a serious threat to them. On a closed platform (like that of Android which is major in India) they probably have a lot more control and tracking (as someone mentioned it above). I guess it would make a lot of sense for FK to tap the naivete of Indian market and pull them all away from the web.

It's all politics written all over it!

[1] http://techcrunch.com/2015/04/13/flipkart-airtelzero/

just deleted flipkart app

Regarding the list of permission, my reaction was no wonder. Just couple of days ago, I got this news that Flipkart is going to start something in advertisement domain. They( and almost every other internet company?) want as much data as they can gather.

"See the list of permission used by the app below"

The permissions demanded by android apps have moved from alarming to out and out clownish.

Imagine consenting to this:

"modify or delete the contents of your USB storage"

they are asking you to backdoor your own device.

It's already past time that Android had (out-of-the-box) a "don't grant access but make the application think I did"-checkbox.

Essentially, showing fake data to the app if you pick this (eg: no photos, fake contacts, etc).

AFAIK, there are some android apps (and cyanogen) that add this features, but it really needs to be worked on upstream.

This is doubly useful for google, since they can always keep their own apps as "superprivilidged" on google-sponsored phones.

I lost my hope when they removed the "access to the Internet" permission from the list of privileges. Now you take for granted that all apps will try to do that.

I use AFWall+ and CyanogenMod's Privacy Guard to control everything, but it's sad that the data of most people I know may be stolen without their knowledge after they install yet another Subway Surfers-type game or "vintage photo effects" app.

CyanogenMOD's PrivacyGuard is pretty close to this, I don't use stock android anymore due to this.

Xiaomi's MiUI has also had this feature for quite some time. They call it the Permission Manager - http://www.androidbeat.com/2014/08/use-permissions-manager-m....

That would be horrible for normal developers.

People would use it all the time, and then complain that the app doesn't work.

The actual problem is the "all or nothing" security model that needs to be replaced with contextual permissions (like "This app wants to access your photos right now. Allow?", see iOS).

The "all or nothing" model defeats its own purpose by effectively leaving no choice for the user.

It'll be better if there are checkboxes next to each permission, and if the user doesn't grant a certain permission, the app can know.

Older apps targeting older API levels will receive fake data, whereas newer apps will be able to know whether they were actually granted the permission or not.

This would prevent apps (and app developers) from being left in the dark, and would also kill the "all or nothing" permissions model currently in place.

"Sorry, but please reenable GPS so SuperFlashlight can function."

Nah.

Android starting from 4.4 does have a privacy feature that allows you to control what data out of permissions a particular app can access, on my phone I have turned everything off for everything. P.S. I don't have any google app

* It's already past time that Android had (out-of-the-box) a "don't grant access but make the application think I did"-checkbox.*

The weird thing is, it does. Or, rather, it's hidden-in-the-box. There's an app called App Ops that can be enabled if you have root access that limits permissions to installed apps - I'm actually not sure if it sends fake data instead but I'm yet to have an app crash from it. But without root it's not usable. Sigh.

Unreal. Do they really expect it to cost more to maintain their web presence in addition to their mobile app than their web presence will make them? If you already have a back-end for servicing all of those mobile apps, is having an HTML front-end really that difficult?

I vastly prefer doing almost anything I can think of on a real desktop with big displays and a good keyboard over a mobile app. Am I that out of touch? Why would anyone, that has a choice in the matter, prefer a mobile app for shopping, when it's much easier to open and manage dozens of tabs of product listings and reviews across lots of different sites for ease of price comparison and product comparison?

... wait, I think I realized why they're doing this. They want to create a walled marketplace where you're locked in to buying things from them, and can't easily compare products or prices elsewhere.

What about Gopher? FTP? Usenet? AOL? WAP? HTTP/HTML is just another way of consuming the content, but the justification for using it over other mediums is that usage is high enough. If it falls below a certain threshold, then it makes sense to throw it out, just like the other low-performing mediums. Although walled-garden reasoning may also play a factor, a simple market-driven analysis would be sufficient in many cases to warrant the decision to drop HTML.

There are some big social media companies whose mobile traffic so far outstrips their web-usage, that they consider from time to time dropping the html frontend. It seems inconceivable to us, but it makes a certain amount of sense.

Not quite. Absolute traffic levels don't matter. The only time it makes sense to throw it out, as the parent comment suggested, is if the maintenance costs exceed the benefits. Even if the usage is low compared to mobile, it only makes sense to stop if it's a long-term net negative.

Personally, this makes me suspect bad architecture. If they've got well-factored APIs, the additional cost of maintaining a modest web presence should be pretty low. From what they describe, mobile is now somewhere between 40 and 60% of their traffic. Even if they were pretty sure that the trend to mobile would continue, why wouldn't they just hedge their bets by investing less in the HTML front end until the web traffic is really negligible? Few people use faxes anymore, but a lot of businesses still publish fax numbers just in case.

Assuming that this isn't purely internal politics, then my best guess is that they have a lot of terrible web code. Somebody saying, "Mobile is the one and only future! Look, shiny future!" gives people a chance to declare bankruptcy on a lot of technical debt while appearing as forward-thinking leaders, rather than bozos who made a giant mess.

Yes, exactly. Any CEO worth his salt would know to hedge his bets and stay on a platform till its already dead. That is what Stephen Elop taught everyone.

I do think that some amount of bad architecture is involved, but flipkart does have a lot of money, and a lot of good engineers. They even have cool stuff on github [0], which needs good people. They have likely fixed their architecture issues by now.

Flipkart has money, and lots of it (current valuation is $11 billion). There is absolutely no way that they are considering this on a cost-analysis basis.

The Nokia analogy bears approximately no relationship to this situation. Nokia was a platform company. Flipkart is just using other people's platforms, ones that will continue to exist independent of their efforts.

Your theory that people with lots of money and stuff on Github have no architectural problems does not match my experience.

Also, I like you using the web to tell me the web is dead. Do go on.

No, the justification for HTTP/HTML is that the client required to use these platforms covers almost your entire addressable market.

> I vastly prefer doing almost anything I can think of on a real desktop with big displays and a good keyboard over a mobile app. Am I that out of touch?

Yes I'm afraid you are. Worth watching Luke Wroblewski talking about Mobile at Google - https://www.youtube.com/watch?v=Y-FMTPsgy_Y ... some graphs near the start help make this clear.

Interesting. Thanks for the pointer.

>Why would anyone, that has a choice in the matter, prefer a mobile app for shopping.

That's the clincher. No desktop, no choice.

Well, the choice is to use a competing service that offers a (likely superior) desktop experience.

One reason I haven't seen mentioned yet is that countries like India can be overwhelmingly mobile-dominated (as in, few people own desktops).

I never had data on India but in places like the Philippines or Indonesia (with similar GDP/capita), your buyer's only internet access might come from his (very basic) smartphone and I sometimes saw > 90% mobile visits. At a previous company, we saw our Indonesian traffic soar after we fixed a bug that prevented the site from displaying on Blackberries... somehow an incredibly popular brand in the region.

If 90% of your buyers are on mobile, you might as well become a mobile company, if your IT resources are limited.

The gap between mobile and desktop is closing though, as companies like Xiaomi keep releasing ever cheaper, more powerful and larger devices in those markets. A Redmi (1, 1S, 2...) costs a tenth of an iPhone for 80% of the functionality.

The reach of mobile devices never fails to impress me. We went horseriding a few hours drive from Jiuzhaigou, and after a day of climbing the mountains on horseback, sleeping on the sloped floor of a tent (heated by a dried yak dung fire, of course), it was quite surreal to see all our guides, on horseback, playing mobile games and happily texting away on their 5" smartphones - even as they were herding yaks...

The issue here is that Flipkart is not poor. It is one of the largest private companies in India, with valuations at $11 billion. They have already spun off their costlier operations (such as courier and warehousing) into separate companies which they have since sold off.

The only thing flipkart is doing right now is tech and sales. They are definitely not "limited" by money or IT resources.

They might be valued highly, but they aren't having strong revenue streams, they still make 2.4Rs loss per 1Re revenue, hardly profitable.

You'd be amazed how money can be wasted in online retail. I know of a 2,000 employee company that took 1.5 years to produce a mobile version of its site and another year for the app. Another (also thousands of employees and over 500 developers to manage code bloat) hires a "Chief Data Scientist" to build a data science team every 6 months or so - the guy quits after 3 months without even getting access to a database, and a few months later the CEO asks the CTO how that data science effort is going and they put a job ad out again. I think they parallelized the effort and hired three in one go last round to speed up finding one that will stick...

These companies have in common that they hired "cheap and many" rather than "expensive and few" initially to lower their startup costs and "tech risk", and that they are run by former bankers and McKinseys who consider tech "below them", something that is "commoditized" and can just be "bought in".

That really sounds like a shitty company. I'm sure that while Flipkart does have its fair share of code bloat and overly aggressive management, they still have lots of good tech people, because of what they are offering in the market.

Their salaries are among the best in the Indian tech industry, so I think they do have the talent to keep a website afloat and running.

One issue with phones in emerging markets is that they are very storage constrained -- the cheaper models often sell with as little as 2GB of storage. On Android, you can add more storage with SD cards, but some apps (Facebook & Chrome, iirc?) cannot be moved to the SD Card. People with limited storage often have to be quite brutal in triaging their installed apps list. Maybe Flipkart has data that shows that users often install their app, but I think a few quarters of anemic mobile user growth will help them change their mind.

Another Indian e-commerce site Tradus did this last year, and they failed so badly they went back to stealth mode.

Having said that, Flipkart might be just big enough to actually pull this off. I must note that it is highly irritating. They already shut down their mobile site ages ago with a "Download the app or go die" page. This is extremely irritating especially when you follow links from e-mails, their app fails to catch it and mobile site is giant "Ha Ha".

For this to work they would have to significantly boost the quality of their apps. It is just not that great on either iOS or Android. Flipkart has made apps for insane amount of platforms though, even S40, to probably support this decision. Even my Android Wear has a Flipkart app.

EDIT: FYI, Flipkart didn't really say that they will shut it down. All they said is that their mobile traffic grew 10-fold and in the next year they might become totally mobile. He was obviously referring to traffic, with no word about shutting down website.

Well, Myntra which was acquired by Flipkart is supposed to go mobile-only on 1st May. Flipkart might be testing waters before doing it across their whole business.

A mobile app can have a higher conversion factor than a website. But still this decision seems to be short sighted. A well-designed website allows people the access information from a wide variety of devices - Amazon e-readers, Google Android, iOS, Blackberry, linux terminals. If your device platform is not supported then Flipkart then you will have to select a competitor website.

If you do not want to agree with the many "Allow this app to access X, Y, and Z" then you will have to select a competitor website.

If you want to use a bigger screen and a keyboard then you will have to select a competitor website.

A bug on website can be fixed immediately. If app has a bug and updated fix is stuck in the app review process then you will have to select a competitor website.

A company such as Flipkart can definitely support a web version along with apps. And aren't website and apps supposed to be simple frontends using the same backend APIs?

I think Flipkart is being too much ahead of time by ditching the web. Hopefully, competitors such as Amazon and Snapdeal will fill the gap left by Flipkart.

If they really turn off their website entirely, they'd get no traffic from search engines. I assume this means they'll either keep product pages up for the robots (and then redirect you to the mobile app), or that their 40M registered users simply go directly to Flipkart when they want to buy something.

If there description in the article is correct, I would bet on users going directly to the app to buy something. I bet a lot of discovery is word of mouth and traditional media.

I would bet on leaving the website as a catalog and nothing more. Shouldn't be too hard to dump the database to get product pages.

Google is indexing apps and prompting an install from the SERP when they return in-app results. I doubt that it's as effective from an SEO perspective as having a good mobile website, but it is supported.

They just tweeted out that the decision is not final. https://twitter.com/flipkartsupport/status/59245692561208524...

I used to be a regular Flipkart user, I even bought their "Flipkart First" subscription. That was a waste of money. Nowadays I just shop on amazon or ebay, Prices are usually cheaper for the same items. Plus they don't force me to use the app when on mobile.

I love Amazon Prime Now in NYC (free 2 hour delivery of tens of thousands of items), but am constantly annoyed to have to pull out my iPhone to use it. Of course, I presume Amazon is doing a mobile-first rather than a web-never strategy.

What's so terrible about "pulling out your iPhone"?

If I just read a review on my iMac of something I'm ready to buy, it seems ridiculous to lean over and pull out my iPhone.

Um, I suspect that the reason is much simpler.

An app can authorize a purchase through the phone provider. A website requires a credit card.

This is interesting. India the primary way for in app purchase is via the provider. Credit card penetration is low.

Can you clarify what "provider" means? Is it Flipkart or the network (Airtel, etc)?

He means mobile service provider (airtel, vodafone, at&t, etc)

Thanks!

Oh, I thought he meant the App or the Play store.. Isnt that how in app purchases work?

In some countries there are agreements for carrier billing instead of credit cards and it deducts from a users' prepaid balances.

Even in cases where you have a Credit Card, there are many hoops to fly through before you can make an in-app purchase:

- It must be an international credit card - It must never have 3d-secure enabled

International Credit Cards are costlier, and it just takes one payment on an Indian gateway to enable 3dsecure accidentally. And then, you can't use your card again on any international purchase because no international gateway supports 3dsecure.

Are you trying to say that Indian credit cards that are 3D-secure enabled don't work for IAP? This is absolutely untrue. I have both AMEX and Visa cards issued in India, and simultaneously use them for domestic and international online purchases, IAPs, and in-store. Never had a problem. Frankly, I've never heard this.

So, if you make a purchase on an Indian website, you get 3dsecure, but you don't get one on an International purchase? It used to work fine for me till a while back, but the newer RBI guidelines have specifically mentioned 2 factor auth against international transactions[0]. Since its just a guideline, it is not enforced as such equally by all banks.

[0]: http://www.medianama.com/2013/03/223-rbi-credit-card-money-t...

This is crazy. It used to be that companies who couldn't do both mobile and desktop just made an app wrap their website. Maybe we need the reverse of a WebView. A website that does nothing more than emulate an app.

It actually already exists: https://www.manymo.com/

I just used it to access cmyip.com, and it shows their IP without identifying it as a proxy... I tried to contact them to say that people can abuse it, but there's no way to contact them without creating an account. :( Welp.

Strategy hypothesis --

They believe they have more knowledge of how to execute in the mobile space than amazon or other outside incumbents and so they think they can move faster and better by shutting off the power alley's of the incumbents (web) by driving the warzone to mobile.

It's a strategy... and just a hypothesis as to some of their thinking... but maybe? :-)

There was a very good blog post about why shutting down the website makes a lot of sense in India (and many parts of the world).

http://www.eugenewei.com/blog/2015/4/6/tech-miscellany

I imagine they will still have a website, just not one that has the functionality. It makes perfect sense, if a particular interface is only driving single digit usage you drop it. No need to waste dollars updating something no one uses.

To be precise, 85-90% of their customers are now coming through mobile app. Going forward, it doesn't make sense for them to support so many platforms. People using their website are now a minority. Keeping the website operational is just too much effort in the long run for them.

You can compare this with, how many developers worry about Blackberry when they launch? Hardly any! The reasoning is simple: it's a platform whose usage is sinking(was?).

Note: They are not closing the flipkart online portal, but they are closing the myntra website which they acquired an year ago. And in the press release they must have mentioned that they are moving completely to mobile version.

They bought myntra because of tough competition and have to close it anyway. They are doing an experiment by keeping the mobile version.

Slow death for Myntra! PR for Flipkart mobile app!

I think they are making a mistake, there is a reason why the internet was created, and forcefully migrating all users to the mobile app is wrong, it defeats the purpose of having a website which anyone can handle, the news might be that flipkart is shutting down its mobile website, shutting down the main website is like shooting yourself with a pistol :D

While they're at it, there's this new wave of services with no interface. They believe in the mantra "The Best Interface is No Interface" eg: Operator, Magic etc. Interesting times!

I hope google penalizes such kind of behavior in their search results

I think there is a mistake with this article. They were and already have shut down their mobile website. There is no official statement the they are going to shut down their website.

Can this be ran in an emulator on a PC?

As an upside they get a real choice of programming languages (unlike "Open" Web).

ಗೂಬೆಗಳು!

Reply on news.ycombinator.com