20 ms·

Microsoft Excel. Manipulating tens or hundreds of thousands of rows, including cross-referencing across tables, is just so satisfying and much faster than doing

by towawy 4y ago

Microsoft Excel. Manipulating tens or hundreds of thousands of rows, including cross-referencing across tables, is just so satisfying and much faster than doing it with a DB or code. Alternative spreadsheet software do some things better, but they don't come close as a complete package, I don't regret paying for it myself for the first time in my life. It such a life saver in a pinch. Obligatory Spolsky intro to Excel: https://www.youtube.com/watch?v=0nbkaYsR94c https://www.youtube.com/watch?v=0nbkaYsR94c

manv1 4y ago
Excel gets my vote for one of the top 10 software programs ever. It's crazy how useful it is.
- hermitcrab 4y ago
  Excel, the second best program for any task. ;0)
  - grecy 4y ago
    The really impressive part is that it’s Actually the second best program for every task.
huijzer 4y ago
As a Linux user, using Excel via the web interface or via a virtual machine is not so satisfying nor much faster than alternatives.
- npteljes 4y ago
  Yeah, as someone who doesn't pay for MS, I'm basically using a combination of Calc and Gnumeric, because G can open larger files (handled larger files than Excel in fact), but Calc has more features, and it's more similar to Excel which I was used to. ONLYOFFICE's sheets can also be handy, if MS Excel compatibility is wanted.
  - acapybara 4y ago
    Try: visidata on a sqlite file.
jerzyt 4y ago
100% in agreement on Excel. Even when coding in Python I frequently save an intermediate file as xlsx to explore/debug, or even load into Tableau for viz.
- aquafox 4y ago
  You should learn R and dplyr ;)
  - dotancohen 4y ago
    Tell us more - in the context of them being replacements for Excel for his use case.
    - aquafox 4y ago
      "Manipulating hundreds of thousands of rows" is exactly where R, dplyr and data.table are great at. I do that on a daily basis.
    - steine65 4y ago
      I love R but can't use it at work :( With just the tidyverse library (which includes dplyr), R can be very useful in a data analysis pipeline. It is great for data cleaning and aggregation, especially when a process needs to be done multiple times. It is much faster than excel/power query. I am an accountant in SaaS and spend a lot of my day waiting for excel/powerbi automations to refresh. Similar solutions in R/sql/python would be nearly instant. Also excel/powerbi automations are a bitch to troubleshoot, and are unnecessarily complex. When following tidy principles, a framework designed by the tidyverse dev Hadley Wickham, R code can be very easy to interpret, similar to SQL. Additionally the R community has made libraries for everything, and I consider R a great general purpose language as well.
      - hermitcrab 4y ago
        Note that different flavours of R have very different performance. 'Base R' is quite slow. But R + data.table is blinding fast. Power Query perforamnce is awful, even compared to base R. Some benchmarks of these plus other data wrangling software (including my own product) at: https://www.easydatatransform.com/data_wrangling_etl_tools.html https://www.easydatatransform.com/data_wrangling_etl_tools.h...
jen729w 4y ago
Yeah, just yesterday. Hacky output from `nslookup`, not ideal but it’s all we had. Tens of thousands of rows. No IDE with regex capabilities available. Excel to the rescue! Took me about five minutes to extract exactly what I needed.
- dgacmu 4y ago
  If you can use dig output, the next time you need to do something like this you might find 'jc' helpful: https://github.com/kellyjonbrazil/jc https://github.com/kellyjonbrazil/jc
  - jen729w 4y ago
    Government laptop. No can use anything that isn’t whitelisted. This is the beauty of Excel. It’s a powerful text manipulation tool that everyone has.
- melagonster 4y ago
  where can I input regex to excel? I really need that, last time I copied table to nopad++ then copy paste it back.
  - jen729w 4y ago
    I didn’t. I pasted the output and munged it with Excel’s trusty text functions.
    - melagonster 4y ago
      so in this time, Excel isn't so good...
- jodrellblank 4y ago
  > "No IDE with regex capabilities available." If it's Windows there should be PowerShell ISE installed.
  - 4y ago
eitland 4y ago
Crazy useful. So useful I like it despite how it always goes out of its way to waste my time by trying to misrepresent a number of different types of strings and numbers as American dates :-|
- zemoose 4y ago
  Dear god, I would use Excel so much more if I could get it to stop mutating my data.
  - alenrozac 4y ago
    January, February, Maruary
systemvoltage 4y ago
Pssh... adults use JMP ;-): https://www.jmp.com/en_us/software/data-analysis-software.html https://www.jmp.com/en_us/software/data-analysis-software.ht... Trivia: JMP stands for "John's Macintosh Project". Also, the entire semiconductor industry depends on JMP just like the entire pharmaceutical industry depends on Minitab.
- BeetleB 4y ago
  And we go full circle. I learned pandas in 2012 so I don't have to deal with Excel or JMP scripting. Still use Excel for quick stuff but otherwise it's pandas for me.
- hermitcrab 4y ago
  The JMP buy page doesn't even give you a price. How expensive is it?
- analog31 4y ago
  I work at a shop where a few people are JMP users. I think it's a problem when a proprietary app is only available to a handful of users at a site -- they tend to get a lot of work dumped on them, and nobody can use their stuff without going back to them. It's a different experience when a site has a site license. This isn't specific to JMP of course. A good thing about Excel is that businesses are OK with just paying for "everybody" to have it. Also, people use Excel for more than analysis. It's also a crude database and platform for creating small "apps" that do things, that are easy to share with others.
leokennis 4y ago
Often useful CLI tools are linked on HN (like in this thread). And probably if you take the time to learn them or if you integrate them into your automation they are 100% useful. But, given the problem “I have some unfortunately formatted data which I need to analyze”, there is no better solution than sanitizing it in VS Code and then analyzing it in Excel.
yawnxyz 4y ago
I don't have a Windows machine, and I'm having mixed feeling about Excel for Mac — is it as powerful as the Windows version? Are they exactly the same or are there big difference in functionality? The web / Office360 version doesn't seem much better, either though.
- cyberge99 4y ago
  Native Excel on macos isn’t bad. I’m on an M1 and it’s slow to start initially, but once loaded it’s fine. Better than any web version.
SuperSandro2000 4y ago
And for me excel wastes 100 hours a month.

by towawy 4y ago

manv1 4y ago

Excel gets my vote for one of the top 10 software programs ever. It's crazy how useful it is.

hermitcrab 4y ago
Excel, the second best program for any task. ;0)
- grecy 4y ago
  The really impressive part is that it’s Actually the second best program for every task.

huijzer 4y ago

As a Linux user, using Excel via the web interface or via a virtual machine is not so satisfying nor much faster than alternatives.

npteljes 4y ago
Yeah, as someone who doesn't pay for MS, I'm basically using a combination of Calc and Gnumeric, because G can open larger files (handled larger files than Excel in fact), but Calc has more features, and it's more similar to Excel which I was used to. ONLYOFFICE's sheets can also be handy, if MS Excel compatibility is wanted.
- acapybara 4y ago
  Try: visidata on a sqlite file.

jerzyt 4y ago

100% in agreement on Excel. Even when coding in Python I frequently save an intermediate file as xlsx to explore/debug, or even load into Tableau for viz.

aquafox 4y ago
You should learn R and dplyr ;)
- dotancohen 4y ago
  Tell us more - in the context of them being replacements for Excel for his use case.
  - aquafox 4y ago
    "Manipulating hundreds of thousands of rows" is exactly where R, dplyr and data.table are great at. I do that on a daily basis.
  - steine65 4y ago
    I love R but can't use it at work :( With just the tidyverse library (which includes dplyr), R can be very useful in a data analysis pipeline. It is great for data cleaning and aggregation, especially when a process needs to be done multiple times. It is much faster than excel/power query. I am an accountant in SaaS and spend a lot of my day waiting for excel/powerbi automations to refresh. Similar solutions in R/sql/python would be nearly instant. Also excel/powerbi automations are a bitch to troubleshoot, and are unnecessarily complex. When following tidy principles, a framework designed by the tidyverse dev Hadley Wickham, R code can be very easy to interpret, similar to SQL. Additionally the R community has made libraries for everything, and I consider R a great general purpose language as well.
    - hermitcrab 4y ago
      Note that different flavours of R have very different performance. 'Base R' is quite slow. But R + data.table is blinding fast. Power Query perforamnce is awful, even compared to base R. Some benchmarks of these plus other data wrangling software (including my own product) at: https://www.easydatatransform.com/data_wrangling_etl_tools.html https://www.easydatatransform.com/data_wrangling_etl_tools.h...

jen729w 4y ago

Yeah, just yesterday. Hacky output from `nslookup`, not ideal but it’s all we had. Tens of thousands of rows. No IDE with regex capabilities available. Excel to the rescue! Took me about five minutes to extract exactly what I needed.

dgacmu 4y ago
If you can use dig output, the next time you need to do something like this you might find 'jc' helpful: https://github.com/kellyjonbrazil/jc https://github.com/kellyjonbrazil/jc
- jen729w 4y ago
  Government laptop. No can use anything that isn’t whitelisted. This is the beauty of Excel. It’s a powerful text manipulation tool that everyone has.
melagonster 4y ago
where can I input regex to excel? I really need that, last time I copied table to nopad++ then copy paste it back.
- jen729w 4y ago
  I didn’t. I pasted the output and munged it with Excel’s trusty text functions.
  - melagonster 4y ago
    so in this time, Excel isn't so good...
jodrellblank 4y ago
> "No IDE with regex capabilities available." If it's Windows there should be PowerShell ISE installed.
- 4y ago

eitland 4y ago

Crazy useful. So useful I like it despite how it always goes out of its way to waste my time by trying to misrepresent a number of different types of strings and numbers as American dates :-|

zemoose 4y ago
Dear god, I would use Excel so much more if I could get it to stop mutating my data.
- alenrozac 4y ago
  January, February, Maruary

systemvoltage 4y ago

Pssh... adults use JMP ;-): https://www.jmp.com/en_us/software/data-analysis-software.html https://www.jmp.com/en_us/software/data-analysis-software.ht... Trivia: JMP stands for "John's Macintosh Project". Also, the entire semiconductor industry depends on JMP just like the entire pharmaceutical industry depends on Minitab.

BeetleB 4y ago
And we go full circle. I learned pandas in 2012 so I don't have to deal with Excel or JMP scripting. Still use Excel for quick stuff but otherwise it's pandas for me.
hermitcrab 4y ago
The JMP buy page doesn't even give you a price. How expensive is it?
analog31 4y ago
I work at a shop where a few people are JMP users. I think it's a problem when a proprietary app is only available to a handful of users at a site -- they tend to get a lot of work dumped on them, and nobody can use their stuff without going back to them. It's a different experience when a site has a site license. This isn't specific to JMP of course. A good thing about Excel is that businesses are OK with just paying for "everybody" to have it. Also, people use Excel for more than analysis. It's also a crude database and platform for creating small "apps" that do things, that are easy to share with others.

leokennis 4y ago

Often useful CLI tools are linked on HN (like in this thread). And probably if you take the time to learn them or if you integrate them into your automation they are 100% useful. But, given the problem “I have some unfortunately formatted data which I need to analyze”, there is no better solution than sanitizing it in VS Code and then analyzing it in Excel.

yawnxyz 4y ago

I don't have a Windows machine, and I'm having mixed feeling about Excel for Mac — is it as powerful as the Windows version? Are they exactly the same or are there big difference in functionality? The web / Office360 version doesn't seem much better, either though.

cyberge99 4y ago
Native Excel on macos isn’t bad. I’m on an M1 and it’s slow to start initially, but once loaded it’s fine. Better than any web version.

SuperSandro2000 4y ago

And for me excel wastes 100 hours a month.