Yogesh Dhimate

Notes to Myself

Dec 24, 2020 - 2 minute read - Comments - Books

Made to Stick and the Curse of Knowledge

I recently read “Made to Stick” by Chip Heath and Dan Heath. It was a fun and quick read. While I liked their framework of SUCCESs (Simple, Unexpected, Concrete, Credible, Emotional, Stories) to get the idea to stick, I thoroughly enjoyed the explanation of the Curse of Knowledge. The curse of knowledge occurs when a knowledgeable individual/expert assumes that the others have the background to understand a specific concept.

According to Chip and Dan, In general, getting your message across has two stages - the Answer Stage, and the Telling Others Stage. In the Answer Stage, you use your expertise to arrive at the idea that you want to share. This involves studying the material and becoming capable of answering. However, the same factors that worked to your advantage in the Answer Stage will backfire on you during the Telling Others stage. To get the answer, you need expertise, but you can’t dissociate expertise from the Curse of Knowledge. You know things that others don’t know, and you can’t remember what it was like not to know those things. So when you start sharing the answer, you’ll tend to communicate as if your audience were you.

This reminds me of another news I read a long time ago about an audio illusion. Once our brains know what to expect to hear, they do, even if, in reality, it is impossible.

Dec 21, 2020 - 2 minute read - Comments - Programming

MuleSoft Capacity Explanation for Noobs and CXOs

I was talking to a C-level executive the other day and he asked me to explain how does MuleSoft’s capacity/sizing across various deployment models work. Given I always wear my technical hat I jumped right into the concept of cores and virtual cores and containers (and Kubernetes!) and did a pretty bad job. Fortunately, that executive was very accomodating and somewhat understood what I was talking about. Nevertheless, I should’ve done better.

I was thinking about this conversation over the next couple of days and decided to explain it to my wife - who is very non-technical.

Think of MuleSoft as a specialty cookie store. Or even better, think of Salesforce as a big giant grocery store, with a bakery section called ‘MuleSoft’. MuleSoft specializes in cookies. They offer 3 types of options to buy their fantastic product.

  • Cloudhub: Pre-baked cookies, these come in boxes. Very light and crispy. Rightfully branded as ‘Cloudhub’. You can buy as many boxes as you want. Each box contains 10 cookies. This means with one box you can feed 10 guests. If you have more guests coming over, or some guests are unusually hungry you can buy more boxes. There is also an allergen/gluten free version of pre-baked cookies for government guests.
  • Hybrid: A cookie dough to bake your cookies in your oven. The cookie dough is branded as ‘On-premise / Hybrid’. You can decide how big (or small) each cookie can be. You can make as many cookies as you want with the dough. (No more 10 cookies per box restriction!). Of course, some guests will find your ridiculously small sized cookies unappetizing (or funny).
  • Runtime Fabric: After some guests complained that pre-baked cookies were too big, and ‘make your own cookies’ unappetizing. MuleSoft started offering ‘Runtime Fabric’ branded cookie dough. You still have to bake your cookies, but the sizes of cookies are standardized. They won’t look funny.

Oct 24, 2017 - 2 minute read - Comments

Google Spreadsheets and ArrayFormula for XIRR Calculation

I was looking for a simple way to calculate performance of my investments over time. Popular portfolio trackers like Moneycontrol and Value Research did not meet my needs. So I decided to try out Google Spreadsheets. Nice thing with Google Spreadsheets is its support for import and queries. e.g. AMFI India publishes daily NAV’s on their website http://portal.amfiindia.com/spages/NAV1.txt You can import the data programmatically and run a query on it to find desired NAV. I can use following simple query to check the latest NAV of Birla Sunlife Frontline Equity fund, whose symbol is INF209K01YY7

"SELECT * Where Col1 like '%INF209K01YY7%'"),";"), "Select Col5")

Google Spreadsheets obviously supports data from Google Finance with its built in function GOOGLEFINANCE


This will return latest price for State Bank of India. You can also use Stock Exchange assigned ID’s to query the data from Google Finance


This will return latest price for Larsen and Tubro. However the most killer feature of Google Spreadsheets is its support for ArrayFormula in conjunction with XIRR. Let me explain. Consider following cash flow.


We have to update the XIRR formula to include the new transaction we just added in our cash flow


This is very cumbersome to update XIRR formula, if you have a diversified portfolio with 10+ symbols and investing on a regular basis.

Havells =XIRR(D2:D4,C2:C4)*100
SBI =XIRR(D5:D7,C5:C7)*100

You can see for two symbols we have to keep close track of cell numbers etc to accurately calculate XIRR. ArrayFormula provides nice workaround to simplify and automate this process of using XIRR formulas for a growing number of transactions


Enables the display of values returned from an array formula into multiple rows and/or columns and the use of non-array functions with arrays.

XIRR can be easily combined with ArrayFormula like below


As you can see that except the symbol name both the formulas are identical. We can remove the hard coded symbol name and refer to the cell which has symbol name. Using ArrayFormula in the portfolio tracker has simplified maintenance of the spreadsheet and saved quite a few hours for me.