4 ms·

And its good because Excel is a horrible tool. Much better spend your time learning a language aimed at data analysis.

by ekianjo 3y ago

rrjjww 3y ago
I’m convinced that the people who constantly parrot this online are just terrible at excel.
- ekianjo 3y ago
  I am terrible at Excel because I find it a complete waste of time to get good at it. And its tied to Windows, yet another abomination of nature.
Vaslo 3y ago
There is a reason for Excels dominance. CFOs and executives down to junior analysts can quickly do tasks same day that would take weeks to build using overly complex programming languages. This is just more empire building by “I’m Smarter than all you dumb people, why can’t you think like me” programmers. I don’t need c++ or Python to build a net present value analysis that I could show the CFO. I can build it in excel in an afternoon, and I can hand it to the least technical salesperson, and he can make changes in my model and see the results instantly then send it back to me. And then my boss can. And then someone in marketing can. And the bonus is I don’t get a bunch of useless errors that no one understands like “Stack Overflow Error” or “Cannot process because of error 1994505 super duper data transformation canooter value fluctuation.” The worst error I get is that it crashes and I lose 10 mins of work. Face it - Excel is a phenomenal achievement and it burns arrogant programmers like you because we can do all this and completely shut you out. That’s why it’s a “horrible tool” lol.
- adrian_b 3y ago
  It is better said that the "spreadsheet" concept is a phenomenal achievement (and it is true that currently Excel is the best spreadsheet program). Already Lotus 1-2-3 for MS-DOS, more than 30 years ago, matched or exceeded almost all features provided today by MS Excel (and it had an optimized keyboard-based user interface that enabled experienced users to perform most tasks faster than in modern spreadsheet programs like Excel or LibreOffice Calc). Excel has just provided the same features, with nothing original of any importance, but only with trivial improvements enabled later by better computers. The fact that Excel has succeeded to replace Lotus 1-2-3 has not been caused by any technical superiority but by the integration in MS Office and by the fact that Microsoft has cheated, by ensuring that nobody else could keep up with the changes in the public Windows API and with the undocumented parts of the API.
  - namdnay 3y ago
    > Already Lotus 1-2-3 for MS-DOS, more than 30 years ago, matched or exceeded almost all features provided today by MS Excel Only someone who isn’t a heavy user of Excel could say this.. the improvements really aren’t trivial.
    - adrian_b 3y ago
      Actually I have been both a heavy user of Lotus 1-2-3 in Antiquity and a heavy user of Excel in recent times. It is obvious that a program which may have a size of many megabytes is able to include a lot of improvements over a program whose size was limited to a few hundred kilobytes. Nevertheless, all improvements provided by Excel are quantitative not qualitative, they do not enable any essentially new application. For instance, there is no doubt that it is much easier to write a maintainable Excel script in Basic, than in the awkward macro language of Lotus 1-2-3. Even so, using just the macro language of Lotus 1-2-3, it was possible to write amazingly complex applications, e.g. for the automation of the tracking in real time and of the generation of reports about the flow of partially processed products through the steps of complex technological processes in some factories (with thousands of different products and with more than one hundred manufacturing process steps through which they might have to pass, depending on the part number), also of the printing of all the documents that accompanied the manufacturing batches, and where the data was stored in a database embedded in the spreadsheets. Today, writing such a program by using a modern database, a modern programming language and a modern computer would be very easy, but doing the same within 640 kB and with a 33 MHz CPU was not a little accomplishment. At that time, Lotus 1-2-3 was one of the most useful applications for most businesses, especially when computers and programs were much less affordable and many would not have been able to buy any other program, except perhaps some word processor. Excel has taken all the features of Lotus 1-2-3, except the user interface, which was no longer suitable in Windows, but everything added later were just enhancements that were obvious when faster CPUs and more memory became available.
  - Vaslo 3y ago
    You are right, I gave Microsoft the credit when it really is the spreadsheet concept.
- ekianjo 3y ago
  Learning how to script a problem out of Excel brings many advantages. Its beyond the pure practicality. It helps you to think in terms of data structures. People who spend their lives on Excel have no idea how to best format data for processing and over and over repeat the same inefficient patterns.
  - criddell 3y ago
    Efficiency is all a matter of what you value.
  - Vaslo 3y ago
    A large majority of the people I work with have been working 20+ years. A lot of them make 200k-300k and have never written a line of code. They are doing great without it and my company stock value keeps increasing without coding in Finance.
    - Given_47 3y ago
      There’s no need to hide behind the salary figure. If you and your colleagues r fine with a tool that’s “good enough” and would rather not undertake the mental burden of learning a new thing that’s fine. But don’t toss out all these irrelevant reasons that try to prop excel up as some truth
      - Vaslo 3y ago
        I didn't say I was making that salary but our executives are. The "mental burden" for us is thinking deeply about our business and how our products affect our profitability, not wasting understanding the difference between a list and dictionary. While not perfect, Excel is a fantastic tool and the only relevance I need to prove my point is the fact that it dominates the industry because it gets things done. I'd be willing to bet you also probably look down on people who use Windows instead of Linux desktop. Or why they don't "burden" themselves to learn the oh so easy command line. If this is you, I get exactly the person you are and it defends my point even more.
epistasis 3y ago
I came to excel far after learning lots of programming languages, including those aimed at data analysis. The only thing that comes even close for my current Excel spreadsheets use cases would be Rshiny. Interactive responsive analysis of various scenarios, with traceable computation for a very wide audience, is far better via Excel than even Rshiny, which hides all the code. Starting with Excel is probably a great way to get to other tools. Especially when computations are straightforward and don't require the advanced stats libraries available in something like R.
grigri907 3y ago
IF your job is largely dedicated to data analysis and IF you don't need to communicate or hand off the raw work to a non-technical audience, and IF you've got the time and support to learn those skills, then you're probably right. But most of us don't live in that world.
- ekianjo 3y ago
  Nothing prevents you from exporting the result of a analysis to excel at the end. This way you give people what they like while you dont have to use it for all the intermediate work.
makeitdouble 3y ago
Excel and spreadsheets in general are horrible in many ways, but being proficient in them is a basic life skill at this point and should be taught at school if it isn't already. It's not just analysis; data representation, manipulation and collaboration are essential parts as well (especially the last one). Being able to quickly import numbers, format and surface the important part and show it to someone to discuss and tweak in real time makes a world of difference.
- ekianjo 3y ago
  in python and R you can do a very complex graph in one line of code while it would require 30 clicks at least to do the same thing in Excel when its even possible. the power of abstraction.
  - Vaslo 3y ago
    In excel I click on a block of data and click on the graph I want and it appears. I use Python (Pandas) and Matplotlib pretty proficiently and I haven’t found a need to sub those for an excel graph. No picky data types. No installing and teaching someone Jupiter or Pandas or an IDE. The anti tech sales guy can just change the data in the cell, or pull the little box down. Or double click in the graph where he wants to insert a title. You’re fighting a losing battle here. I have need for Python in what I do, but none of my Finance partners do. Coding has to get much much simpler before it will ever replace excel. I’d bet money you are also the person who wonders why the world uses Windows over Linux for many tasks. When you can answer why regular people choose Excel over code and why regular people choose Windows over Linux, you’ll get it, and you can make a positive change in code to bring Linux and Code more mainstream.
    - Given_47 3y ago
      Tbf, MPL can be an absolute pain to work with. Generally tho idk if it’s a good idea to optimize for usability from sales dude
  - _dain_ 3y ago
    >very complex graph in one line of code in matplotlib? get outta here
refurb 3y ago
"Everyone except me is doing it wrong"
- ekianjo 3y ago
  nice strawman
  - refurb 3y ago
    It’s not. It’s pretty much sums up the comment.

And its good because Excel is a horrible tool. Much better spend your time learning a language aimed at data analysis.

rrjjww 3y ago

I’m convinced that the people who constantly parrot this online are just terrible at excel.

ekianjo 3y ago
I am terrible at Excel because I find it a complete waste of time to get good at it. And its tied to Windows, yet another abomination of nature.

Vaslo 3y ago

There is a reason for Excels dominance. CFOs and executives down to junior analysts can quickly do tasks same day that would take weeks to build using overly complex programming languages. This is just more empire building by “I’m Smarter than all you dumb people, why can’t you think like me” programmers. I don’t need c++ or Python to build a net present value analysis that I could show the CFO. I can build it in excel in an afternoon, and I can hand it to the least technical salesperson, and he can make changes in my model and see the results instantly then send it back to me. And then my boss can. And then someone in marketing can. And the bonus is I don’t get a bunch of useless errors that no one understands like “Stack Overflow Error” or “Cannot process because of error 1994505 super duper data transformation canooter value fluctuation.” The worst error I get is that it crashes and I lose 10 mins of work. Face it - Excel is a phenomenal achievement and it burns arrogant programmers like you because we can do all this and completely shut you out. That’s why it’s a “horrible tool” lol.

adrian_b 3y ago
It is better said that the "spreadsheet" concept is a phenomenal achievement (and it is true that currently Excel is the best spreadsheet program). Already Lotus 1-2-3 for MS-DOS, more than 30 years ago, matched or exceeded almost all features provided today by MS Excel (and it had an optimized keyboard-based user interface that enabled experienced users to perform most tasks faster than in modern spreadsheet programs like Excel or LibreOffice Calc). Excel has just provided the same features, with nothing original of any importance, but only with trivial improvements enabled later by better computers. The fact that Excel has succeeded to replace Lotus 1-2-3 has not been caused by any technical superiority but by the integration in MS Office and by the fact that Microsoft has cheated, by ensuring that nobody else could keep up with the changes in the public Windows API and with the undocumented parts of the API.
- namdnay 3y ago
  > Already Lotus 1-2-3 for MS-DOS, more than 30 years ago, matched or exceeded almost all features provided today by MS Excel Only someone who isn’t a heavy user of Excel could say this.. the improvements really aren’t trivial.
  - adrian_b 3y ago
    Actually I have been both a heavy user of Lotus 1-2-3 in Antiquity and a heavy user of Excel in recent times. It is obvious that a program which may have a size of many megabytes is able to include a lot of improvements over a program whose size was limited to a few hundred kilobytes. Nevertheless, all improvements provided by Excel are quantitative not qualitative, they do not enable any essentially new application. For instance, there is no doubt that it is much easier to write a maintainable Excel script in Basic, than in the awkward macro language of Lotus 1-2-3. Even so, using just the macro language of Lotus 1-2-3, it was possible to write amazingly complex applications, e.g. for the automation of the tracking in real time and of the generation of reports about the flow of partially processed products through the steps of complex technological processes in some factories (with thousands of different products and with more than one hundred manufacturing process steps through which they might have to pass, depending on the part number), also of the printing of all the documents that accompanied the manufacturing batches, and where the data was stored in a database embedded in the spreadsheets. Today, writing such a program by using a modern database, a modern programming language and a modern computer would be very easy, but doing the same within 640 kB and with a 33 MHz CPU was not a little accomplishment. At that time, Lotus 1-2-3 was one of the most useful applications for most businesses, especially when computers and programs were much less affordable and many would not have been able to buy any other program, except perhaps some word processor. Excel has taken all the features of Lotus 1-2-3, except the user interface, which was no longer suitable in Windows, but everything added later were just enhancements that were obvious when faster CPUs and more memory became available.
- Vaslo 3y ago
  You are right, I gave Microsoft the credit when it really is the spreadsheet concept.
ekianjo 3y ago
Learning how to script a problem out of Excel brings many advantages. Its beyond the pure practicality. It helps you to think in terms of data structures. People who spend their lives on Excel have no idea how to best format data for processing and over and over repeat the same inefficient patterns.
- criddell 3y ago
  Efficiency is all a matter of what you value.
- Vaslo 3y ago
  A large majority of the people I work with have been working 20+ years. A lot of them make 200k-300k and have never written a line of code. They are doing great without it and my company stock value keeps increasing without coding in Finance.
  - Given_47 3y ago
    There’s no need to hide behind the salary figure. If you and your colleagues r fine with a tool that’s “good enough” and would rather not undertake the mental burden of learning a new thing that’s fine. But don’t toss out all these irrelevant reasons that try to prop excel up as some truth
    - Vaslo 3y ago
      I didn't say I was making that salary but our executives are. The "mental burden" for us is thinking deeply about our business and how our products affect our profitability, not wasting understanding the difference between a list and dictionary. While not perfect, Excel is a fantastic tool and the only relevance I need to prove my point is the fact that it dominates the industry because it gets things done. I'd be willing to bet you also probably look down on people who use Windows instead of Linux desktop. Or why they don't "burden" themselves to learn the oh so easy command line. If this is you, I get exactly the person you are and it defends my point even more.

epistasis 3y ago

I came to excel far after learning lots of programming languages, including those aimed at data analysis. The only thing that comes even close for my current Excel spreadsheets use cases would be Rshiny. Interactive responsive analysis of various scenarios, with traceable computation for a very wide audience, is far better via Excel than even Rshiny, which hides all the code. Starting with Excel is probably a great way to get to other tools. Especially when computations are straightforward and don't require the advanced stats libraries available in something like R.

grigri907 3y ago

IF your job is largely dedicated to data analysis and IF you don't need to communicate or hand off the raw work to a non-technical audience, and IF you've got the time and support to learn those skills, then you're probably right. But most of us don't live in that world.

ekianjo 3y ago
Nothing prevents you from exporting the result of a analysis to excel at the end. This way you give people what they like while you dont have to use it for all the intermediate work.

makeitdouble 3y ago

Excel and spreadsheets in general are horrible in many ways, but being proficient in them is a basic life skill at this point and should be taught at school if it isn't already. It's not just analysis; data representation, manipulation and collaboration are essential parts as well (especially the last one). Being able to quickly import numbers, format and surface the important part and show it to someone to discuss and tweak in real time makes a world of difference.

ekianjo 3y ago
in python and R you can do a very complex graph in one line of code while it would require 30 clicks at least to do the same thing in Excel when its even possible. the power of abstraction.
- Vaslo 3y ago
  In excel I click on a block of data and click on the graph I want and it appears. I use Python (Pandas) and Matplotlib pretty proficiently and I haven’t found a need to sub those for an excel graph. No picky data types. No installing and teaching someone Jupiter or Pandas or an IDE. The anti tech sales guy can just change the data in the cell, or pull the little box down. Or double click in the graph where he wants to insert a title. You’re fighting a losing battle here. I have need for Python in what I do, but none of my Finance partners do. Coding has to get much much simpler before it will ever replace excel. I’d bet money you are also the person who wonders why the world uses Windows over Linux for many tasks. When you can answer why regular people choose Excel over code and why regular people choose Windows over Linux, you’ll get it, and you can make a positive change in code to bring Linux and Code more mainstream.
  - Given_47 3y ago
    Tbf, MPL can be an absolute pain to work with. Generally tho idk if it’s a good idea to optimize for usability from sales dude
- _dain_ 3y ago
  >very complex graph in one line of code in matplotlib? get outta here

refurb 3y ago

"Everyone except me is doing it wrong"

ekianjo 3y ago
nice strawman
- refurb 3y ago
  It’s not. It’s pretty much sums up the comment.