Automation in agriculture/gardening - Drip irrigation

Given :
 Water is required by plants(roots only),
When :
 The traditional way of putting water using hose pipes
Then
  Water and energy(electricity) is wasted

Solution:
  Drip irrigate the plant roots like human circulatory system (network of pipes - remember to use T-bends at 90 degrees to avoid lag in flow).
  Reuse existing overhead tank to circulate the water into this system. (Overhead tank uses gravity. Fill this tank when sun shines using a solar powered water pump to make it self sustaining)
     -Also can add a rain water harvesting system
  Connect automatic timer on the tap to irrigate water on certain times of day/month/year
  Sow the seeds and enjoy self sustaining flower and fruit plants.
 
  Scale it to a farm!

Exceptions 
  Cost of building overhead tank (reuse existing tank or water tap)
  Clogging of pipes (needs maintenance)

Practicals
  This i will try and post in detail when in India.
 
Links:
  Drip irrigation timer only on Alibaba or ebay(buy pipes and connectors in local shop to save Rs.) :
 
 Pls leave a comment if you find it useful. That would keep me motivated :)

ॐ सर्वे भवन्तु सुखिनः
सर्वे सन्तु निरामयाः ।
सर्वे भद्राणि पश्यन्तु
मा कश्चिद्दुःखभाग्भवेत् ।
ॐ शान्तिः शान्तिः शान्तिः ॥  meaning

Preparing for a Solar adventure (Solar PV based systems)

"I am bored of software" Its all virtual in the mind, no concrete matter. Hmmn.. I am thinking of building a hardware system(solar PV based farm) :)
I aim to blog here my findings as i venture into the Solar adventure and open source all the knowledge gained in the due course.

With the solar energy based electricity production catching up in Germany and other western countries, I have been thinking for a while the best way to get into this space to contribute and monetize by setting up a self sustainable business as sun light is abundant in India.

Germany seems to be leading in the solar energy space (They call it Energiewende  or energy revolution). This revolution seems to be because of rising energy prices(as they were major importers of electricity), the Fukushima nuclear disaster and government policy to give good feed in rates for energy producers. 


Given
   India has temperature of  around 25-40 degrees around the year
   Solar PV panels cost have reduced drastically (0.4 to 0.7$/w)
   Indian Government has solar feed in tariff with net metering(in Karnataka) at subsidized rates
   Current goverment has backing for solar energy (considering Gujrath's model)
When
   Solar PV based system are setup as farms or on roof tops
Then
   Electricity can be generated and fed into existing grid. With two way net metering we can run the meter backwards. Carbon footprint can be reduced and a small profit made

Exceptions and unknowns
   Power companies like BESCOM might be unwilling to pay the rates and provide net metering(considering corruption in govt offices)
   Risks like fire disaster or grid tie cut off (when main grid is turned off, the unit should stop supplying power so the line man doesnt get electocuted etc ) need to be evaluated
   What if feed in tariff gets reduced drastically by government?
    Profit need to be computed:

Estimated Profits 
  Ingredients
    Solar panel  at say 0.5$/watt = Rs.25/watt a 10kwph system would cost : 2.5 lakhs
    Micro inverters (per panel inverter is better than one single inverter)
    Net meter , wires and other components
    Labour
    Land
    Permission from electricity board
 Profits
   Feed in tariff rate say Rs.12/kw . If a 10kw system runs say 10 hrs/day produces 100KW in a ideal scenario. So earning is around Rs.1200/day on a good day?? Roughly one can make Rs.30,000( $600) a month with 10kw unit (considering the feed in tariff rate is real)?? 10% returns
Note: these are rough calculations and might be wrong. will update once i get the system in place and start earning in real from these power companies.
 
Hmmm, so I need to buy the ingredients. The most important being solar panel. 2 options - Mono or poly crystalline. Mono is expensive and compact than Poly.

A Report on Design Estimation of 1MW Solar PV Plant (Utility Scale) with potential earnings is here 

The Government policies (in India)

• Subsidy : The Ministry of New and Renewable Energy (MNRE) provides up to 30% capital subsidy for roof top systems (off-grid)

•  Tax breaks  A company can claim 80% accelerated depreciation in the first year of installation under section 80 IC of the Indian Income Tax Code, leading to savings on income tax.

• Renewable Purchase Obligations (RPOs) and Renewable Energy Certificates (RECs) - Under the REC mechanism, developers are eligible to receive one certificate for every 1,000 kWh of renewable electricity fed into the grid 
• Net metering Until May 2014, four Indian states: Gujarat, Andhra Pradesh, Uttarakhand,
  Tamil Nadu and West Bengal have finalized the net metering policies. Another
  four states: Delhi, Kerala, Karnataka and Punjab have net metering policies
  in a draft stage. In Gujrath the roof top FiT price is Rs.11.21/kWh and Rs11.78/kWh

Solar PV companies in India:
  Solar panel supplier list  http://www.enfsolar.com/directory/panel/India
    http://www.evergreensolar.in/index.php  (Tamil nadu based)

Useful links

1. BRIDGE TO INDIA is launching the INDIA SOLAR NAVIGATOR. This online, analytical tool is designed provide you with up-to-date, in-depth business insight on the Indian solar market.
2.  Bridge to India - The India Solar handbook : download pdf

When agile becomes Fragile

Its been a while I blogged, probably i got lazy or too busy with other things after getting married :).

Lately I was involved in a project where the so called scum master/monkey advocated and promoted Agile practices to his bosses overseas probably to get apprised, but wouldn't practice it in his own team. Probably his lack of enthusiasm, insecurity or his ways to get on top of the corporate ladder by pushing others down. God only knows. It did annoy me and can see clearly the team was getting disintegrated and drowned in misery.

There was no tester involved, no business analyst to communicate or document requirements/backlog(we had to produce the backlog and code and test and release and communicate to business and take the risk of getting something wrong as we were doing multiple roles). Just another Dev ops dude Y and me figuring out how to deal with things - digging legacy code/building the new messaging pipeline/ reverse engineer the data model/ liaise with teams and what not. This was not scrum. Just adhoc. The scrum monkey would as well shamelessly admit being a Hippocratic. It was absolute shambles - me and Y (who could spend only some time on this project as he was pulled into another project by the so called scrum monkey). I did mention him several times to get another resource and a tester, but he would simply brush it away. And with his crooked ways would ask for the build result without being involved in the design or build process and would like to take the credit when the boss would make an occasional visit? Also on the other side there was a tech ops/support team trying to coax me into dealing with their stuff applying Social engineering skills/Behavioral science/Jingoism. Well one needs to learn moral science  first and do their duty instead of troubling their colleague who is doing his job.

This was so unnatural and demotivating. Obviously good developers would run out of choice and  move on at this stage, so am I. I can see others following. One dirty fish seems to spoil the pond.

So as usual i asked Google god (our digital god who answers our prayers with search results), Why?

1. Basically we were ending up doing Cowboy coding with the boss only interested in the results - doesn't matter how you do it. And he would pull the resource to do other tasks than the main goal.  An undisciplined scrum master who doesn't care for his team but only the outcome by simply not pulling his weight.

Refer http://www.agilebuddha.com/agile/agile-is-not-ad-hoc/

2. Happiness Metric is something a scrum team should always consider along with retrospectives followed by its implementation. After all a motivated healthy and happy team will always produce good result. This does seem like the natural way of doing things.
Refer http://scrummethodology.com/happiness-metrics/

Digression
My miseries and rant apart. Let me think what was going wrong here.

Ongoing maintenance and build over legacy application is a challenging task especially with so many technologies involved built by several different minds over time. The tech ops team struggle to understand whats going inside when something breaks while the dev team struggle to build over legacy apps as there are no regression tests.
Both team need to co-exist to be an ideal Eco-system.

1. The tech stack builds up with many new technologies(Java and its loads of relatives, Mule, ESB, Oracle, Linux) and functionality while the older technologies(C, VB, websphere, messaging, Sybase, Solaris) and functionality haunt the developers. It is like renovating an old house, and the builders waste get expensive to be cleared and affects the new building blocks.

2. The old process and the new process need to match up. The bosses expect the migration to be done somehow.

3. The programming mindset - Stored procedural programming vs Object oriented programming mindset is hard to change IMHO. After years of PL/SQL programming ones brain gets hard wired to its style. As industry moves to OOP technologies like Java there needs a shift of ones thinking which has a hard learning curve involved.

4. As developers move out of team for grazing into greener pastures - there is a huge loss to the team. The new comers are left with no choice but to reverse engineer the code to understand the functionality as the docs might not exist or tend to get obsolete. Moreover the flat documentation can never be as complete as the relational application code. The code is the only source for latest functionality.

5. Working with the existing complex data model can get harder with time if no good process is followed.Also applications cross referencing database tables make them more tightly coupled

But there is no real process for that defined in Agile afaik for a moving target (for porting legacy apps along with ongoing maintenance ). From the bosses point of view Agile might mean getting things done fast with less resource so one can secure a bonus or promotion fast.

Hmmn thats how corporates work isn't it or am I being too pessimistic?

A similar pattern i could see in the past with another media company and could see stressed out developers and tech ops. Someone was not pulling his weight or there was no one to see the big picture and steer/guide the team(perhaps an Architects role) or was there a lack of process for this moving target?



More later...