Hacker News Reader

Comment by chime | Hacker News Reader

> Has anybody shown him these things? http://www.electronicplus.com/images/products/TN811C.jpg

The article did mention a simple timer. I'm going to guess there is some additional complexity in turning on the water that a multi-mode timer cannot handle. Timer comes on at 2am and shuts off at 3am automatically but what happens when there is a power failure between 2.10am and 2.50am? That happens in India ALL the time. Now you missed 40 minutes of watering. A human might be able to take that into account by turning it off at 3.30am instead remotely.

> For that matter, is it so hard to run a mile of wire across a field?

Not hard, just expensive. The Indian farmers he's aiming to help are the ones making $1-$5/day. They might understand the long-term ROI on such an investment but they might not be able to afford it.

View on HN

If the power fails they'll come on at the wrong time.

Of course the pumps will not be able to run when there is no power but it might be less than ideal to run the pumps at the wrong time.

That said, you could solve that fairly easily by running the timer circuitry off a rechargeable battery and drive the pumps through a relay. When there's power the battery recharges from the grid, when the power has failed for a while the timer hasn't lost track of the time of day.

I wonder how much an SMS message costs an indian farmer (it also requires an extra cell phone), it might very well be a substantial part of the cost of operation. In the end it's a convenience thing (you don't have to walk) which is worth some $ or it would have to save money or create a better yield.

Indian mobile networks are fairly well connected. And sending local SMS messages are, for all practical purposes, free. Cell phones can be had for US$30. Possibly less.

If you want to build technology-related services that are accessible to the largest possible part of the Indian population, using SMS as a protocol is probably the right way to go.

(Edit: But from what I understood, this device involves phoning a number and using a keypad to issue commands. )

> If the power fails they'll come on at the wrong time.

I was assuming the timer was battery-powered and not prone to power failure, while the pumps were. This could cause the timer to turn on the pumps, not detect that the pumps were off for 40 minutes during the cycle because of power failures, and then shut off the pumps after a predetermined duration. So the timer would start/stop a one hour cycle but during those times, the pumps would only work for 20 minutes. That could cause crops to fail because the pumps didn't irrigate sufficient water.

To fix this, you'd need a timer that would also detect the duration the pumps were working and adjust accordingly, increasing the cost of the system. Plus how does the timer deal with the next cycle schedule to start at 4am-5am when it has been pumping 30-40mins extra before the cycle before it took longer? I was just trying to say that setting up a workable schedule on such a device could be a lot more complex/expensive than a human texting "pump 2 on", "pump 2 off".

But a human texting 'pump 2 on' suffers from the exact same problem, compounded by the delays of receiving the SMS (which in my case sometimes only appear hours or days after sending).

So it was meant as being the exact equivalent of what they have now. (I didn't see any evidence of a feedback mechanism in the article but I may have overlooked that).

Cellphones are everywhere in India. Practically everyone already has a cellphone, or is going to get one this week. Text Messages (locally called SMS) are often free, or are dirt cheap. It is common even for low-wage workers (day labourers, porters, gardeners) to have cell-phones in India. This is, in part because the landline infrastructure is very poor and unreliable, and a open economy has meant that cellphone operators have been able to provide good service for really competitive rates.

http://en.wikipedia.org/wiki/Communications_in_India

SMS is cheap in India. You can get plans with upto 500 messages a day for 99 rupees a month.

99 rupees = $2

Someone should let Neelie Smit-Kroes in on that little secret.

why?

Because it's at least an order of magnitude more than what you'd pay in Europe for the same thing which is a good indicator that there is price fixing involved.

According to Wikipedia http://en.wikipedia.org/wiki/SMS

The key idea for SMS was to use this telephony-optimized system, and to transport messages on the signaling paths needed to control the telephony traffic during time periods when no signaling traffic existed. In this way, unused resources in the system could be used to transport messages at minimal cost.

So, in a sense sending SMS doesn't have to be so costly.

Are you sure?

500 messages a day costs < $0.07 total. It's cheaper than that in Europe?

No, rather the opposite.

Reply on news.ycombinator.com