We started from Bangalore at around 5 in the morning. The route to Horanadu was mostly fine except for the last stretch. We took the following route: Bangalore (NH4) -> Nelamangala on to NH48 -> Kunigal -> Yediyur -> Channarayapatna -> Hassan -> continue on SH57 to Belur -> Chikmagalur -> Aldur -> Balehonnur -> Magundi -> Balehole -> Horanadu. The road from Magundi to Horanadu was not very good although it is very scenic. It is a stretch of about 40km.

You can very easily make out when you enter Chikmagalur district. The place becomes more green, you can see quite a few hills, lakes, areca nut trees, coffee plantations, and the air is cooler and pure. We clicked a few snaps on the way.

We reached Horanadu at around 1:30pm. We visited the temple and had lunch there. After a few more snaps, we headed to Kalasa. There are quite a few places in Kalasa and a shutterbug can easily spend a couple of days here.

Our first stop was at Kotehole, then to Rudra Teertha and a visit to Amba Teertha. The road to Amba Teertha is not very good and it is recommended that you take a local vehicle. We spent quite some time at Amba Teertha.

We then visited Duggappana Katte. This is on top of a hill and you can see the Kudremukh hill range, Ballarayana Durga etc from here. On the right is a hill which if you trek to the top, you can see Sringeri on the other side. Duggappana Katte is an ideal spot to watch the sunset.

This was followed by a visit to Vasishta Theertha. It was dark by the time we came back from here and we called it a day.

(This post is still not complete.)

We had done the required preparations for the trip – mapping out all the places that we intended to visit and the distances to each of them and the things that we had to carry.

We left Bangalore at 5am in the morning. Our plan was to reach Chitradurga by noon with a visit to Vanivilas Sagar Dam on the way. The road to Chitradurga is very good; it is part of the North South corridor. The driver of our car felt like we had paid the price (toll) to rip on the roads and he was treated to a wonderful road where he was easily reaching 120-150 kmph (the maximum speed of an Indica).

We visited Vanivilas Sagar Dam around 10:15. Vanivilas Sagar Dam, aka Mari Kanive, is the oldest dam in the state. I had been to the dam just a couple of months back and so I knew that there is a way to the top.

We clicked quite a few snaps and experimented with our cameras for about an hour’s time. We then headed to Chitradurga. On the way you can see quite a few wind-mills.

We reached Chitradurga around noon. We had made bookings in Fort Resort.

We wanted to have a quick lunch and then head to the fort, so were asked to check out the Khanavalis just outside the fort. There is one street which is full of them and you can get a full meal for as much as Rs. 25.

Post lunch, we toured the Chitradurga fort. We took a guide because we didn’t want to miss anything. The guide took us to all the important places in the fort – it took more than a couple of hours for him to give us just a brief overview of the history of the fort. The guide started with an explanation of the 7 circles (Elu suttina kote) and said that 2 of the circles were lost to the city expansion. The main entrance of the fort is close to the 3rd door. The doors are positioned in such a way that you could easily lose your way inside; you cannot see one door from another. There are quite a few watch towers (batheris) and the watch towers are positioned such that any information can be passed on from one tower to another. Enemies would have a tough time inside the fort because the people at the watch towers can see the enemies while the enemies are unaware of the fact that someone is watching them. Some of the important batheris in the fort are visible near the 4th and 6th doors.

One of the first things you see after you enter is ‘Yenne Kola’ – a small tank where they used to store oil required to illuminate the fort.

Beyond this is the 4th door. A popular way of attacking forts was to use logs to pierce through or use elephants to barge through the doors. In order to prevent such attacks, the doors were built with a wall in front.

The sixth door has an Islamic style on one side and a Hindu architecture on the other. This was done by Hyder Ali after his victory to show the unity between the Hindus and the Muslims.

The place is filled with temples. One of the first ones you come across is the Meludurga Siddhi Vinayaka Temple. The statue is carved out of a stone which has the resemblance of an elephant in the back.

Chitradurga is named after “Chitra” “Kallina” “Durga” which means “art in rock forms in the fort”. Near this temple, you can see quite a few such artistic formations – the elephant, the rabbit, the frog, the fish, and the turtle.

You also see the famous Jhanda Batheri from this place – a watch tower where they hoisted a flag to commemorate the victory in the battles and wars.

Once you enter the seventh door, you see quite a few temples and monuments. On one of the hills is Tuppada Kola – a place where they used to store a special type of ghee, used to sharpen the swords. Then there is Gaali Gopura, Muruga Math, Hidimbeshwara temple.

You also see a mud fort near the center of the fort – this was to absorb the heat from the otherwise rocky fort. In order to withstand the attack of water, the mud was solidified with egg shells, pieces of pottery, jaggery etc and was mixed together by stamping on it, using horses for this purpose.

Beyond this, on the other side of the fort, you can see Akka Tangiyara Kola – place where there was natural rain water harvesting, Obavvana Kindi etc.

The walls of the fort have been built in such a way that you cannot remove a single rock from the bottom without dismantling it from the top.

One of the attractions in the fort is a person who is popularly called Kothi-raj who climbs the walls and boulders in this fort without a rope. Kothi-raj has an international reputation in climbing and has 9 gold medals to his credit.

We returned to the resort after a long and tiring day, feeling accomplished!

The second day, we started with a visit to Holalkere. Holalkere is about 33 km from Chitradurga city. It is famous for the Ganesha statue which is more than 500 years old.

We then visited Chandravalli caves (or Ankali Math). This is one attraction that you should not miss if you are to visit Chitradurga. This place has 3 things – a trek to a hill closeby, the caves and the huge rocks. The caves were inhabited by the saints of Ankali math who used to meditate here. The caves are pitch dark inside and you need a torch to visit these places. You can get guides to show you around.

This was followed by a visit to Adumalleshwara temple – about 10km from the city. This place has a zoo and is a popular picnic spot for the locals.

Finally, we visited Himavat Kedaya – this is a short trek of about 1km from Adumalleshwara temple. You see a water fall and a nandi. They say, water flows from the Nandi’s mouth 365 days a year.

The only place that we had planned to visit but couldn’t was Jogimatti. You need permission of the forest officials to visit this place and we couldn’t get it.

I have included a map of the route to Mari Kanive, and to Chitradurga from Bangalore, and the route to Holalkere, Jogimatti and Chandravalli caves here:

All in all, it was a good start for the year.

So here are the steps I have been following with full manual photography:

1. Composition – Photography is an art. The most important aspect of a photo is the composition. We have seen how sometimes an image taken even with a point-and-shoot turns out to be better than the best of the DSLR’s. Composition is more important than you may think! So whenever you intend to click something, look at what you want to capture – move around to see what might be the best position from where to capture. Decide on horizontal or vertical orientation of the camera. What is the amount of noise (unwanted things in the image). Can it be reduced? Can the noise be made interesting? Think about what your image should look like before clicking – don’t leave it entirely upto the camera. Sometimes you may actually have to take a couple of snaps to see what it looks like, before you get the right shot but don’t that’s only for minor recorrections.
   The composition determines the focal length – distance from your camera to the subject that you want to capture. We then look at the exposure triangle.
2. Determine where the subject is and bring the subject in focus. Lock the subject and then adjust the composition. You can use the Al-Servo focus mode when you are clicking moving subjects, but otherwise it is recommended NOT to use this mode, so that you don’t introduce motion blur due to focus-ring adjustment. If you are clicking images from the same distance like in parties, you may even want to put the camera lens in manual focus mode so that it doesn’t readjust itself everytime you click a new snap.
3. Aperture – One of the important things that is often got wrong by most beginners is the depth of field – the part of the image that appears sharp in the image. Digital cameras automatically adjust the depth of field when you shoot in one of the auto shooting modes, but a perfect photo is one where you clearly define the planes that you want to capture. Determine the depth of field that you need. There are no specific rules here, but I have seen that an aperture of f/11-f/14 works well for flower macros. One thing that people get confused about is the relationship between distance and aperture. A subject far away from the camera does NOT mean that the aperture is lower. It is the distance between the nearest plane and the farthest plane that determines the aperture that you need. If you have a subject close to the camera and you want the landscape as well, you will need a very low aperture setting, while if you just need the subject OR the landscape far off, you can do with a higher aperture. First determine what planes you want and your distance from the subject to determine the aperture.

   The flower in the center of the image below is in full focus, while the one on the right is partially blurred out, and the one on the left is mostly blurred out giving a sense of depth. (You can read more about my flower macro photography experiences here).

   
4. ISO and Shutter speed – Have the ISO as minimal as possible. The optimal shutter speed is automatically determined by the camera and this is shown as an exposure reading which should be close to 0 (some snaps might require it to be at -1 or +1 or further away but that’s an exception than a norm). With the minimal ISO, see what the camera indicates as the shutter speed. If the speed is slower than 1/(current focal-length) of the camera, then you will need a tripod. If you can’t use a tripod, increase the ISO until the shutter speed is manageable without a tripod.

   The following image had to be taken with a higher ISO because the interiors of the temple was very dark.

   

   Here are some variations to the rule:
   When you want to introduce blur into images, you may have to use both the shutter speed and the aperture to do the math and then control the ISO accordingly.

   When clicking fast moving far-off subjects like birds, you may have to increase the ISO to a much higher number so that you don’t introduce shakes.

   Depending on your camera type, increasing the ISO beyond a point introduces a lot of noise, so you may NOT want to increase the ISO beyond a certain value, so you under-expose the image and then rescue the under-exposed image using tools like GIMP.

   
5. Now determine other things like White balance (depending on the amount of light available). In bright sunlight, I have switched between sunlight, shade and cloudy. Increasing the temperature, gives a warmth to the image. However, when post-processing an image, you may have to reduce the saturation. I sometimes prefer to set the white-balance as auto and then post-process the image to introduce the warmth by playing with color levels and the saturation settings.
6. Determine the metering mode.
7. Click!

Note: The first 4 steps are extremely important. It is extremely tough to rescue an image if these go wrong. A lot of things beyond 4 can be done via post processing using tools like GIMP or Photoshop. Sometimes you may want to deliberately over-expose or under-expose your image and then post-process it using GIMP (or Photoshop) to get what you want as I have shown in these posts on rescuing under-exposed images and adding warmth. Check out the rest of my posts on photography here.

The recession was expected to hit and it was supposed to be a bad time to switch jobs.

It was a roller coaster ride right from the beginning but I enjoyed it thoroughly. The first all hands meet (all hands meets in startups are NOT about going to a fancy hotel and spending the day playing games), we discussed the recession that was about to hit and several alternatives were discussed to cut the costs and extend the runway. Things didn’t work out as expected and the next all hands I saw the team strength reduce to roughly half!

Right from the beginning, I liked the “openness” culture that was part of the company. Engineers got to decide to a large extent the roadmap and product features, and also interact with and learn from the users directly. A lot of what was being built was being determined based on user feedback. We had a whole bunch of success stories on the way.

4 months into my job, I came to know that we were about to be acquired. It came as a surprise to me, because things appeared so drastic. We had only started tasting success. Why this sudden decision? What about my dreams of working in a startup? Will Ugenie continue to function like one? How will things be different? Will our dreams be crushed by this parent company? Will there be a change in culture?

However, things didn’t change a lot – if anything it got better. While we did gain some financial stability, which allowed us to think about and execute longer term projects (measured in weeks instead of days), things didn’t change drastically in terms of work and the responsibilities that people had. Plus, we got to work with an excellent team over at Lulu spearheaded by Bob.

Now when I look back, I feel, making the switch was one of the most important decisions I took in my life. The last 1.5 years has been such an experience that I could not have expected from IBM.

So, after this wonderful journey, here I am, making the next leap in my career. I have accepted an offer from an early stage seed funded startup and am beginning work early next week. Why the decision? Well, it just feels right and I am mentally prepared to take up the responsibilities that this job entails.

During the course, I made some amazing new friends each of whom is unique in their own personal way. I would like to thank every one of you, who I have interacted with directly or indirectly, for the many things that I have learnt from you. It was a pleasure to know you.

A parting quote:
…, you can’t connect the dots looking forward; you can only connect them looking backwards. So you have to trust that the dots will somehow connect in your future. You have to trust in something — your gut, destiny, life, karma, whatever. This approach has never let me down, and it has made all the difference in my life. –

It’s just not possible for me to visit every feed category every day, so I frequently see some of these categories overflow with posts.

Now I know there are extensive blog posts which describe how to better manage feeds and to cut down on information overload. But as we all know there is no simple solution.

So here I was using Google Reader and just skimming through the posts when I came across this need.

Suppose a feed has about 100 unread posts and I have skimmed through half of them, and read one in between that I thought was interesting, I am now left with quite a few posts on top of my read post, that I am not interested in reading but want to mark them as read so I don’t need to see them again. Would it be possible to mark these as read leaving the rest untouched?

The recent changes to Google Reader provide one option – Mark all entries older than a day, week or month as read. But this does not exactly serve the purpose.

I ended up hacking a Greasemonkey script to do exactly what I wanted.

Here is how the script behaves:

Just press Ctrl+Alt+Y and the script will mark all entries above the current read entry as ‘read’. Ctrl+Alt+I will mark all entries below the current entry as read – for people who read backwards.

Added benefits:

• This also works with search results in Google Reader.
• The script works with entire folders, so you can skim through all posts in a folder marking the ones you have skimmed as read.

How it works:
The script uses the css class names to determine which posts are unread above (or below) the current post. Once it obtains this list, it simulates a click on each of these posts and thereby marks them as read. Simple as that!

This script is part of the Better GReader extension and has featured in Lifehacker.

In order to install the Google Reader – Mark Until Current As Read script, visit this site.

The last time I visited the Lalbagh flower show was in January 2007, which meant I hadn’t been there for 2.5 years!

My initial reaction was that flower macros will appear good when clicked with my Sigma 70-300mm lens and so I had that on. But I quickly realized that it was impossible to maintain the distance required, plus the shutter speed made it really tough to get a steady shot. So I switched over to the 18-55 lens.

I really missed the 55-100mm focal length range. I think this is the right distance that you will need for flower macros. However I had to manage with the 55mm limit, which basically meant I had to get close to the flower.

I got quite a few good snaps. The toughest ones to capture were the low contrast bright colors – like the red, yellow, white colors.

I am increasingly getting comfortable using the full manual mode. I basically kept the focal length at 55mm for most images, and the aperture at 5.6 and the ISO at 400. I only had to adjust the shutter speed depending on the amount of light present and either use manual or auto focus to get a sharp snap.

The final touches were given using GIMP. This mostly included color level editing and some basic work with hue and thresholds for the low contrast bright flowers.

It was good to be back. If you have a passion for macro photography, then this is the place to experiment. The show is on till the next weekend.

More Lalbagh Flower Show snaps here.

1. Make sure wvdial is installed
   sudo apt-get install wvdial
2. Add the device configuration to your /etc/wvdial.conf

   Replace <phone-number> with your 10 digit Reliance connection number.

   
[Dialer Defaults]
Phone =
Username =
Password =
New PPPD = yes


[Modem0]
Modem = /dev/ttyUSB0
Baud = 115200
SetVolume = 0
Dial Command = ATDT
Init1 = ATZ
FlowControl = Hardware (CRTSCTS)


[Dialer cdma]
Username = <phone-number>
Password = <phone-number>
Phone = #777
Stupid Mode = 1
Inherits = Modem0


3. Run wvdial
   
   sudo wvdial cdma

   You will see some output like this:
   
~$ sudo wvdial cdma
--> WvDial: Internet dialer version 1.60
--> Cannot get information for serial port.
--> Initializing modem.
--> Sending: ATZ
ATZ
OK
--> Modem initialized.
--> Sending: ATDT#777
--> Waiting for carrier.
ATDT#777
CONNECT 230400
--> Carrier detected. Starting PPP immediately.
--> Starting pppd at Sat Jul 11 22:56:19 2009
--> Pid of pppd: 4299
--> Using interface ppp0
--> pppd: ????[18][18]m X[19]m
--> pppd: ????[18][18]m X[19]m
--> pppd: ????[18][18]m X[19]m
--> pppd: ????[18][18]m X[19]m
--> local IP address <IP>
--> pppd: ????[18][18]m X[19]m
--> remote IP address <IP>
--> pppd: ????[18][18]m X[19]m
--> primary DNS address <IP>
--> pppd: ????[18][18]m X[19]m
--> secondary DNS address <IP>
--> pppd: ????[18][18]m X[19]m
--> pppd: ????[18][18]m X[19]m
--> pppd: ????[18][18]m X[19]m


That’s it! You must now be able to browse the Internet. In order to disconnect press Ctrl+C.

This post covers quite a few aspects of GIMP image editing:

• Image decomposition
• Quick mask
• Alpha channel
• Selective editing

Using this, we convert an image that looks like this:

to:

Ok, let’s begin.

Let us first quickly go over the process and then we can see the steps involved.

In the image we are taking as an example, we split the image into 2 layers, with one layer containing the trees and the other containing the creek and then selectively apply the various tools on the 2 layers.

The process involves first splitting the image into layers by identifying the key parameters in the image. Each image is different and you need to use different decomposition techniques to do so. Once the image is decomposed, we apply the threshold tool to divide the decomposed layer into black and white pixels, where one set of the pixels represent one layer and the other set of pixels, the other layer. We then apply this layer on the original image and select these pixels out into a different layer. Finally, we apply the various tools like the Color level tool, the Hue Saturation tool etc, to get the right image we went.

Most of the details on this technique are described in the GIMP documentation for Threshold.

So here are the steps:

• Decompose the image
  We first decompose the image using the HSV option. The Decompose option is available at Colors -> Components -> Decompose
  
• Apply the threshold tool on the ‘Saturation’ layer
  Hide the hue and value layers by clicking on the ‘eye’ icon in the toolbox. Select the saturation layer. Now use the Threshold tool to divide the pixels in the layer into white and black pixels. I have tried to divide the image into 2 layers where one layer contains the trees, while the other contains the creek.
  
• Use the Quick Mask option to create a selection mask
  Now that we have split the layer into 2 sets of pixels, we select the pixels and apply it in the original image to select those pixels from the original image.

  Copy the saturation layer where you applied the threshold tool.

  Name the single layer in the original image as Creek. You can do this by right clicking the layer and editing the channel attributes.

  Click on the Quick Mask button in the lower left corner of the window in the original image.
  

  Paste the selection into the masked image.

  

  Remove the quick mask.
  
  You will see the creek getting selected as shown in the screenshot.

• Move the selection to a new layer
  Now that we have the creek selected, we will move this to a different layer. In order to do this, press Ctrl+X to cut the selected pixels. Now, right click the layer in the Toolbox and click New Layer. Add a new layer of exactly the same size and call it Trees. Make sure this new layer is selected and press Ctrl+V to paste the selected pixels. You will see a floating selection in the layers tab. Right click this selection and click on Anchor Layer.
  
• Adding an alpha channel and removing the white pixels
  This step is useful on its own. Whenever you want to add transparency to an image you add an alpha channel to the layer and make one of the colors (white in this case) transparent.
  In order to add an alpha channel, right click the Creek layer and click ‘Add Alpha Channel’. Now use the ‘Select by Color Tool’ to select the white pixels and press Ctrl+X to remove these from the image.
  
  
• Moving the wrongly mapped pixels
  Ok, we now see that some pixels are in the wrong layer. The Trees layer, contains a few pixels which should actually be in the Creek layer. So let’s move them to the right layer.

  In order to do this, use the Free Select Tool, select the Trees layer and click points to cover the entire creek area as show in the screenshot.

  

  You must start your selection from one point and end at the same point. Once this is done, press the Enter key.

  

  Now cut these pixels from the Trees layer and paste them into the creek layer.

  

• Wow, that was long, but now starts the fun. You now have the images in the 2 different layers. So you can just apply the various tools to get the right image you desire.
  

2qkmxntcye

We left Bangalore on Friday night and reached Mysore early morning. There were some adhoc changes to our plan; we were supposed to go to Mysore by train, but later realized that a car is going to cost the same and so hired a car in the last minute.

We left Mysore on Saturday morning at around 7:30 am. We stopped for breakfast on the way and reached Gopalaswamy betta at around 10 am.

There are quite a few scenic places on the way, including sunflower fields and long road stretches with trees on either sides.

Gopalaswamy hill range seemed endless. If you check this place in Google maps, you will realize that this is a huge range covered with a lot of greenery. There were some elephant and deer sightings.

Here is a Google map of the route to Gopalaswamy betta from Mysore:

We went on a short trek and clicked a lot of photographs on the way.

This was followed by a visit to the temple of Gopalaswamy (Lord Krishna). The temple was crowded but it was worth the wait.

We then left to B. R. Hills. We had miscalculated the distance to be 30 km from Gopalaswamy betta. But the distance turned out to be 95km! You first need to reach Gundlupet and from there go to Santemarahalli -> Yelandur -> B. R. Hills.

However, the road trip to B. R. Hills was amazing. As you enter the forest, you can see a sudden temperature drop and hear a lot of birds/insects chirping giving you a feel of the forest. There were lotus covered lakes, swamps, and a huge variety of colored birds on the way. This was accompanied by a light drizzle so the setup was perfect.

We first visited the temple of Ranganatha (Lord Vishnu). Then, we got to experience the sunset on B. R. Hills. There is a spot in one corner where you can see a steep drop and the stretch of the hills, covered with a lot of trees. This is something that you need to experience and is hard to describe in words.

We came back to Mysore in the evening. The roads back to Mysore was bad, but no one complained after the experience.

The next morning we visited Lingambudhi lake and Karanji lake. Lingambudhi had almost dried up, but had a huge bird density for the size of the lake and the time of the day.

Our next stop was Karanji lake. Karanji lake and the park is like a mini Ranganathittu. There is a walkthrough aviary which is a treat for shutterbugs. There are quite a few birds in the lake, including the painted stork egrets etc. There is also a bird watch tower where you can (photo)shoot birds in flight. And add to it a butterfly park for macro-photographs and you can’t ask for more!

We came back to Bangalore on Sunday evening. Overall it was an amazing experience, and I yearn to go back again.

While the application is working like a charm, a day into the launch, I already see a few issues with the hosting solution that I chose, the Google AppEngine.

• All your data is in the Google BigTable
  This is the biggest problem I see with Google AppEngine. While the concept of BigTable is good for storing data in general, the restrictions it comes with are painful.

  For example, you cannot perform a query that returns all the data for you as there is a limit of 1000 tuples per query, and as this post mentions, you can’t even use offsets more than 1000, in essence limiting your queries to return at most the last 2000 tuples.

  Full text searching although available is quite restrictive. There is the concept of a SearchableModel, but you can’t event start comparing it to Lucene/Solr.

  There is no bulk data export functionality. Worse, there is no bulk data delete functionality either. The ways people do it is to write scripts to delete data which are then called via ‘curl’ and as there are time out limits on the curl requests, you need to split your task in chunks and delete data in batches.

• Cron job support is restrictive
  In general, it is expected that cron jobs take time to execute. However, in Google AppEngine, cron jobs have a 30 second time-out limit, so if your cron task is not completed in 30 seconds it throws an exception. The solution for this is to split your tasks across multiple cron jobs.
• A few things are broken in the development sandbox
  In theory, an application running in your sandbox should work fine in the production environment. Unfortunately, I came across a few bugs which are preventing me to run the application in the sandbox. Google is aware of some of these problems and is fixing it.
• Google AppEngine apps cannot be deleted
  There is a limit of 10 applications per user. Further, the applications cannot be deleted.
• No way to download access logs
  It is not easy for you to download your access logs. The dashboard provides a way to search within your access logs, but that does not act as a substitute for awk/sed/perl churning of the logs does it?
• GQL sucks
  If you have used GQL queries, you will know what I am talking about. It takes a while for you to get used to what is available and what is not.
• No way to restrict the bad bots
  If you see one misbehaving bot sucking up your bandwidth, there is no way to restrict this one IP (or a range of IP’s) from accessing your application.

While Google AppEngine has these restrictions, and many more, it is still free for the most parts, and so acts as a good substitute for Amazon EC2 for trying out your application. But is it good for serious commercial application development?