Imagine an accountant coming on here and telling you that you don't need vi, emacs, sublime text or VSCode. You certainly don't need your IDEs. After all it's perfectly possible to code in Notepad.
You also don't need your languages. BASIC was perfectly good.
The killer feature of Excel for financial modelling, over 'proper' software, databases etc is portability and auditability.
Everything I do at work is subject to external audit. Every audit firm in the world has Excel. The tax advisors have Excel too, as to the tax authorities. Apprentices are trained in Excel. The people I hire may not have used our ERP, but they have all used Excel. It is the one constant in our world. The actual accounting records are in an ERP, but all of the outputs are in Excel. I have worked at several multinationals and several SMEs. Excel was everywhere.
If I build a financial model, the 60 year old CFO needs to be able to understand it and agree with it.
One other aspect is all of the shortcuts and hot keys. If you switched to a different software you’d have to relearn shortcuts that you have been using for several years.
As a 59-year-old, I am extremely disappointed in your insinuation that people our age are unfamiliar with modern technology. We built the foundations for all the cool stuff you whippersnappers take for granted, including Excel.
This! I love Cursor’s IDE but, boy, does it really sh*t me off everytime I update it and they remove the keybinding for CMD + SHFT + L.
Like, seriously, you couldn’t chose a different combination for your added feature’s shortcut!?
It’s not like the ‘find all instances of’ shortcut is some hidden, unused gem, which you have to tickle the Gruffalo in order to access! It’s one of the most important shortcuts!
Can you give an example? I am unaware of such power.
If that's the case, why do studies show that so many Excel spreadsheets have major errors in them?
https://theconversation.com/spreadsheet-errors-can-have-disa...
https://phys.org/news/2024-08-business-spreadsheets-critical...
https://duckduckgo.com/?q=study+excel+spreadsheets+contain+e...
In my old job, we had to switch away from our working but ancient Information System to SAP for similar reasons.
Now think of 100 pieces of software that you are confident are 100% error free?
Your point, which is completely true, is not incompatible with GP's statement.
Excel is ubiquitous in the corporate world and most spreasheet in the corporate world are completely full of errors everywhere.
Not just errors but also broken assumptions.
I've been tasked with porting a spreadsheet to a dedicated app because people noticed numbers were off and, oh the clusterfuck. Never again.
The world started making much more sense to me from the moment I realized most business decisions were taken based on powerpoint presentation containing numbers coming from broken Excel spreadsheets.
I work in accountancy and while its possible to audit the output of the spreadsheet, (as I'm sure it's possible to audit the output in other domains) as an erstwhile Software Engineer I can't help but think it would be better if I could better verify the integrity of the spreadsheet itself. It's far too easy to introduce unnoticed errors.
Other people have different views on this but I love the idea of spreadsheets but I don't like any spreadsheet software. I think there's massive room for improvement. But if you were to ask me how to improve it specifically, I couldn't really pinpoint anything.
I had a go at doing something in this space to scratch a very specific itch but of course, it doesn't address any of my complaints here. Rather, it tried to tackle another of my complaints - the lack of good support for different number bases https://github.com/JetSetIlly/Ivycel/
Sheets was a piece of junk compared to excel. Anyone who did any real work rather than just looking at a spreadsheet or using a template got Excel back in about a week.
For Spreadsheets-are-all-you-need (GPT2 Small implemented in Spreadsheet, ~1.25GB Excel file, 124M parameters), I have only been able to get the complete model working in Excel.
In Google Sheets, I've only been able to get it to implement a single layer out of the 12 in the model. And there's no way it can handle the embedding matrix. And it randomly quit during tokenization.
That being said, I am a fan of both Excel and Google Sheets. In Google Sheets the formulas are easier to read, and for teaching my class, it's great because students and I can be in the same sheet at the same time during a lesson.
I also tried LibreOffice briefly. While it could open the Excel file, it was unable to save it (it crashed during the save process).
It was SO. IMPOSSIBLY. SLOW.
Maybe it's different now with M4 macs, but I still remember a friend telling me that response times above 1/10th of a second were not interactive anymore. Made total sense after that.
If your ERP outputs are in Excel, I guess pdf or csv or html could have been chosen as well, since those are also constants in the world?
It's the original agile, functional programming environment. Code iteration and basic testing is done in microseconds.
I remember laughing very hard when someone excitedly told me about REPL in python, like it was a new idea.
Having to wait for Excel to load up every time would be a killer.
I also really like Google Forms, and use it for tracking things like my piano practice time[0]. The page loads instantly and it only takes 6-7 seconds to log a new time.
https://i.imgur.com/xK6tnJe.png
I don't think Excel offers anything like this.
On day 1 I realized my work was to deal with several excel sheets of 10~50MB, with huge amount of VBA scripts, rule-based highlighting, charts... They even wrote an internally used extension for those files. Numbers could not handle and crashed.
I need Excel. I asked if I could borrow a PC laptop from the company, no, that's not for interns. I need to bring my own device. I asked if they would pay for Microsoft Office to get installed on my personal MacBook. The answer was obvious, what are you thinking? I had to manage to get Excel for myself, while the full salary they paid me was not enough to cover a license.
I left consulting after that experience and end up working as a "real" data scientist. Nothing like that happened ever. I collaborate with colleagues using different IDEs/editors everyday. If an accountant joins us and tells me notepad is all you need, I would highly respect his choice and continue with my editor. We don't need to argue, our codebase is not locked in any tools.
You could change ERP, you could change CRM. But there are always something important stored in Excel spreadsheets that no one wants to touch or transfer to something else.
I don’t want to know how many decisions are made based on faulty Excel spreadsheets.
And people are aware of it but they don't know where the faults are and need to go forward anyway.
So they must act as if nothing could be wrong, and if something does come up later on it can be treated as a complete surprise to everyone concerned.
With that kind of acting, how about how far the bar has been lowered on even more critical decisions? :\
Because if you tell me I could get my job done with a different IDE, I would say yes that is correct.
(And they use pretty much the same mess of a formula language. I reject the idea that using not-excel is like using BASIC on top of using notepad.)
At best it’s an arranged marriage, at worst it’s Stockholme syndrome.
In business cases, the risk with home grown spreadsheet 'workflows' is pretty big, and in home usage cases you don't really need more than just a basic spreadsheet. The intersection is perhaps where you have some offline data, cannot write a program as fast as you can use a spreadsheet and there is no risk associated with taking the data and workflow out of a guarded environment.
The same can be said for things like word processing, when you need a book or a paper, you might be in need of typesetting rather than a program that tries to do everything but at a level a novice can use it. You're not going to get a paper or an offset printed book from a text editor. Perhaps if you self-publish a PDF that doesn't need to meet any requirements you can get away with an all-in-one DIY solution. But the time period where that was relevant (around the late 90's) has come and gone.
Let's not stop there, the same can be applied to someone doing some Apps Script, Python notebooks or other solo yolo work, because it felt faster or more productive to them. Surprise! Your cowboy behaviour doesn't actually work at scale, doesn't fit in a shared system and doesn't work in production. But you wanted to be quick and 'get it done'. Instead you waste your own time and everyone else's time, and didn't deliver something that works. (at work; and that includes "but I have done it this way for 1000 years! - doing it wrong a lot doesn't make it less wrong)
But say you want to do some budgeting at home, and all you need is some boxes to type numbers in, then yes, why not use a spreadsheet. But that's not what people celebrate. People celebrate running a company with 100's of jobs on a single spreadsheet, and probably only because it hasn't gone horribly wrong yet. And then there's the real bad scenario, a specialised system (say, an ERP or PIM or CMS) where you pull the data out, do the thing in excel because you didn't want to learn how to use the system and you happen to have written some lists of numbers in excel once and therefore it is now your universal hammer. Good for you, bad for your department or entire business unit because you just broke out of a shared workflow because you thought you knew better which tends to have universally bad consequences.
Excel specifically is an example of "they just don't know any better", just like the everything-is-a-nail example (where Excel is the hammer and every problem looks like a spreadsheet). You could replace Excel with something else and it would still be the same problem (i.e. replacing Excel with Word in scenarios where people want to send someone an image that is on their clipboard - they know they can paste it in Word, so that is what they do and you on the receiving end get a grainy compressed image in a word document). It's not that the sending and receiving didn't work at all, or that the software or the people are bad, it's just a really shitty "solution" that shouldn't be glorified and be seen as the failure to educate that it is.
You can enjoy it as much as you want on your own time. That doesn't make it the great universal fit you think it is.
A normal finance team gets constant one off requests for information. Download something from the ERP, lookup against something else, pivot it by department. Then you get a follow up for the same but with x excluded, but they need it in 10 minutes before a meeting. Excel is great for that.
Then there are the dozens of files behind accounting records. Take a Fixed Asset Register (since it is at the top of the balance sheet), a list of your assets and their depreciation schedule. Maybe your ERP doesn't have one, or your company didn't buy that module, or you have a special class off asset that doesn't fit in it. Maybe they coded it for one country, but in your country the rules are different, so it ends up in Excel.
Then maybe you have a bank account and you have cheques that are posted to your ledger but have not cleared the actual bank yet. Then you have items on your real account that you couldn't post yet, maybe someone paid you with no reference. So you keep a little Excel reconciliation to check.
Then inventory, maybe the ERP didn't split it in the way your consolidation needs it, so you transform the data in Excel and reconcile it back
Then you have all the things that belong in this accounting period but haven't been billed yet, so you have a running list of accruals and when they will reverse. Then you have all the things that are posted in this period but relate to later periods, so you keep a prepayments file. Maybe both of these spit out a journal to upload back to the ERP and reconcile the balance.
Then you get a download of the payroll for the month, but you need to rearrange it into net pay, gross pay, taxes paid by the employee, taxes paid by the employer, pension contributions... then you have to split it by cost centre too. This could be coded, but it is different in every country... and the cost centres keep changing, and the analysis head office wants keeps changing... so it goes in Excel.
Then I want to verify that the system posted the correct absorption to inventory, so I paste in an inventory report and a TB and last month's reconciliation updates.
I can code, but I can't maintain 100 pieces of software that change all the time. I also need all the people in the chain to be able to follow it....
And yes, finance people know how to apply the famous 'checks and balances' to keep out most classes of mistakes....and to detect the same mistakes made by the engineered 'proper software' we have to work with.
I hate Excel for many reasons but it is king for a reason.
This is an elitist software engineer's take.
People building on Excel isn't a result of failure to educate, it's humans doing what humans do best—automating their own workload. It's people using general-purpose computers as general-purpose computers. It's the closest to the personal computing dream that we've come, and likely the closest we ever will.
The alternative in the real world isn't "everyone learns Python", it's "we lock normal business people out of computing and keep it in the hands of the trained and very expensive software engineers". That's not going to happen, and it's frankly not something we should want to happen.
I think this kind of Excel denigration comes up so often in software forums because we're usually called in to rescue a business when their Excel workflow gets completely unmanageable. We miss the decades that the company ran very successfully without any software engineers on the payroll and see the giant spaghetti mess that made them finally decide it was worth the cost. But it's important to remember that these same businesses reached the point where they could afford to pay us to build something custom by building a successful business on top of Excel.
The power of ERP is that it is so rigid and repeatable. Which is great for all the things that stay the same. My country just introduced a law on reporting payment performance to suppliers. My US ERP provider didn't think of that 5 years ago...so we did it in Excel. Say we are in 50 countries (we are in more) each with their own changing requirements... think the ERP could keep up?
But not all of the places in my country are on the same ERP. Many of the foreign places are not (the joys of acquisition!) yet. What do we do in the mean time?
Overall it allowed a segment of the population to continue to remain ignorant of all of the other tools available that were better suited to their needs. It also allowed them to remain ignorant for all the methods to relate data between multiple tools that are adapted to do their jobs better.
So ultimately the only people this has really benefited is Microsoft. the users believe that they're getting a benefit but it is actually a net negative because it discourages growth it discourages learning it discourages understanding the best tool for the job outcomes.
It literally is the axiom of if the only tool you have is a hammer every problem looks like a nail personified to the maximum extent possible.
On the contrary, in many domains in life the right tool for the job is almost always going to be the one that everyone is already using. It's usually not the most satisfying tool or the most elegant, and its use will usually make perfectionists scream, but if you're actually interested in getting anything done you'll default to using the tool that the human beings in the system are already comfortable with unless it would be completely impossible to do so.
As another example: the NEMA connector is objectively bad compared to modern alternatives, but it's not going anywhere. The benefits of standardization usually outweigh the benefits of optimization, and you usually don't benefit from being the first to move to the "better" solution.
(I realize that last part is difficult to swallow for a forum that's focused on startups trying to disrupt the existing, standardized tech.)
Suddenly instead of being comfortable using the tool themselves, they have to request engineer time to update the UI. Instead of being able to write whatever weird excel formulas they have they maybe now need to learn SQL. Maybe they "should" but that is kind of besides the point
For me, I've used it for budgets, to-do lists, etl, reporting, financials. Its data handling utility is pretty much unmatched. It's super easy at this point to see anything with it.
It's well worth any price for sure. It's really the only reason I have a 365 subscription at this point. And like everyone else, I doubt I've used more than 2% of it.
Google Sheets is great, but I'm probably one of the very few people who thinks we need a more robust and feature rich spreadsheet software, if only to compel Microsoft to strengthen their own.
Perhaps you could use an alternative differentiator for your data presentation?
I’m morbidly curious to hear more about this scenario.
Serifs in increasing somberness for the most concrete numbers, sans for everyday estimates, and papyrus whenever I just toss random numbers in.
I can think of a solution: don’t use 9 different fonts.
You have cell background and text foreground color, font size, bold, underline, italics, strike through, cell border styles, conditional formatting, etc. There are plenty of formatting tools in Excel.
What are you not able to accomplish with those that you can accomplish with 9 font faces?
so, what you are calling "fonts" (Arial, Times) are actually called "typefaces"
every pointsize in a typeface is a new font, and bold and italics the same.
so, we don't know if Excel is making the same common/casual error you are, but if their error message is "technically correct (the best kind of correct)", then italics and bold in each pointsize are all separate fonts, and that's what he might have too many of.
It might be OK until 8 specialty fonts need to be reproduced beyond a certain number of cells. If more memory was all it takes to fix, that would be good to know.
Doesn't the fact that you can't imagine a single reason why youd want multiple fonts on a spreadsheet alarm you?
Thanks, you completed my bingo card.
Why do I now doubt your assertion?
Excel is incredible for being so simple to use for even the most basic tasks (Baby names v2.xlsx) to the much more complex data analysis.
As someone who uses Excel for 75% of my workweek, I wouldn't trade it for a 10% comp increase, because doing so would increase my workload at least 25%.
I recently tried moving to Google Sheets and it was unbelievable how slow computationally heavy queries would take compared to Excel, or how painful the lack of shortcuts is.
Even things like how cells are frozen (if you're on B2, are you freezing the first row, or the first and second row?) just feels wrong.
So, is it product quality or deeply built habits? Probably a mix of both. Yes, I grew up with it too, and I wouldn't have it any other way.
FWIW one my frustrations with Excel is how it does freezing of rows. I find Google's appraoch more intuitive. But the issues on computation I agree with (see my other comment on this post).
One of my favorite tabs is taking a high interest loan or credit card and increasing the payment amounts (even a little) to reduce the overall impact of the interest rate, something a lot of people don’t understand.
You will pry Excel from my cold dead hands.
And less tangible: it just feels much faster. I don't have any data for this last claim, it's purely based on personal experience. Although, when I use web Excel, this edge goes away.
And I don't use local files at all with Excel. Everything is either in OneDrive or SharePoint, accessible through web or app on whatever platform.
In reference to your second statement though, I also dont always carry my laptop but it doesn't matter because my spreadhseets are in onedrive too, so I can close my laptop and open my phone and have the same spreadsheet with zero thought.
Looks like I belong to those people that don't understand: What "payment amounts" are you increasing?
Assuming that this relates to checks/cheques, are they going away in the US?
This used to be more important when you wrote paper checks and received a monthly paper statement from your bank. Most people who “balance” their accounts today seem primarily concerned that they are adhering to their personal budget. But the term remains.
The year was 1986, pre-spreadsheets. Writing up an undergrad physics experiment in 1986, I needed to do a hundred or so similar calculations and present the results in a table. Luckily, a computer science acquaintance wrote a small program to do this task for me. Thanks, Dan.
In 1989, before Excel, there was Lotus 1-2-3. Loved that spreadsheet software. My PhD task involved plotting lots of data points, including smoothing some of them. Doable with Lotus 1-2-3, probably not doable otherwise.
More recent times.
Spaghetti Excel. An engineering acquaintance told me his student summer job, at an aluminum refinery, was to check and simplify their Excel spreadsheets. Apparently they had numerous spreadsheets linked to each other. I assume the main purpose was inventory control. I know I didn't envy him his task.
"Please, not just Excel." I took a class of high school students to a uni chem lab and had a small argument with a chemistry tutor who insisted the students use Excel to plot their data. I wanted them to first do it by hand with graph paper. This would have given them a much better feel for their data.
"Rinse-and-repeat Excel". I was tutoring a construction guy trying to learn maths. His job as an assistant on a high-rise construction job involved putting lots of numbers into an Excel spreadsheet. The check for this was to repeat the process and see if he got an identical result. I thank G*d his boss made him do this.
And that's it. Helped other people to use Excel, but I'm thankful that I haven't had to spend my life inputting data into spreadsheets.
There were millions of spreadsheet users by 1986, as VisiCalc was released in 1979[0] and similar programs like SuperCalc[1] were also in use. They were both ported to IBM PC and saw significant use in the corporate world prior to and including 1986.
I started asking how to do stuff on libreoffice on chat gpt and google gemini. And looking at how detailed explanations we have on how to use spreadsheets explained in easy to understand terms for comparably non tech users a lot of stuff became easy to do. As llm had detailed usually correct answers and suggestions. I would have learned my way to doing it anyway but LLM's brought the learning curve down from a few hours to a few min so was willing to just use libre office.
When I set Win 11 or earlier version PCs for others, I always load up Void Tools "Everything" and then swing by FossHUB for the full set of tools.
Libre Office, Inkscape, and many more all go on the machine.
And then I explain what they just got and what it is worth. Almost everyone ends up being able to use the OSS tools.
Doing this saves a ton of money!
The rest, yes I love OSS tools on Windows.
But have fun when their important word docs get garbled
(it's an example, do not come after me with some "uuuh acually word docs open fine now")
Primarily I want searching, sorting, and filtering, and the ability to quickly get sums/means of selected cells would be nice (although not really that necessary). I've been using Modern CSV for a while now (after discovering it on HackerNews last year sometime), and it's mostly very good (good enough that I don't regret the purchase price), but it has some stability issues and the documentation is severely lacking. But the main thing is that it's not quite good enough for me to not have to at least occasionally break out excel.
You can import multi GB csvs, we auto infer your format, and land your data in a full-featured spreadsheet that supports filter, sort, ctrl-F, sharing, graphs, the full Excel formula language, native Python, and export to Postgres, Snowflake, and Databricks.
It’s ironic that you cite the one thing that being bad at hasn’t held Excel back. ;)
The interface is customizable. You can remove any icons or group of icons on the ribbon that you don't like and you can reorganize them in any way you want. Just right click on the ribbon and click "customize the ribbon".
Views are so much faster to pick settings in Excel. You can format tables and indent them (nested ledger) You can get higher information density.
It’s always amazing to me to wade through “how to x” in Google and see 100s of people asking for years.
Google docs and numbered headings is one of those.
I used to maintain MIS trackers in Google Sheets once, and when my company made the inevitable shift to MS Suite I saw an immediate improvement. FWIW I'm 25.
Google Sheets today is totally adequate for most people's spreadsheet needs. I'd believe there are cases where Excel is still better, but they are at the extreme end of "power user" use cases.
And, of course, they will use it for AI training, which could be reverse-engineered in the future to extract training data.
I always have at least one machine setup with all local tools and storage.
Any new ideation and certainly anything private gets done on that computer first.
No company used Google Workspace. All where Office 365.
You use what the company wants you to use :)
If you zoomerettes try to use your own google account you will get flagged for data extraction. :)
Also the web version of office is free, did you know? https://www.office.com/
I made an aircraft flight model that made heat maps of line of sight vectors to the ground as a plane banked and flew its looping flight path over a city. I included an overhead view animation tab by plotting the lat/lon over a scatter plot with a map as a background image and walking over each point via the arrow keys. Wrote some VB to output the image frame to files then stitched them together as a gif. Funny enough it worked great on everyone’s machine except the conference room computer attached to the projector. Turns out it was due to it not having a printer driver installed so some screen inch to pixels system variable wasn’t set.
I did all of the problems in the Fundamentals of Astro dynamics book (BMW) via excel by using the cel copy via mouse drag to perform iteration based calculations.
A co-worker made a complete battlefield simulation in excel that was amazing.
Another shrunk the row and columns cells into pixel size and made some pretty cool looking animations by using cell formatting to turn them on/off with color.
Today, the only modern toolset that would come close to being able to made those examples as quickly would be Jupyter Notebooks. Another tool “professional” coders hate. Maybe Matlab with some costly modules but that breaks down once you leave its linear algebra comfort zone.
I truly hate Microsoft but Excel is an amazing piece of software. Which funny enough started on a Mac.
The managers use Excel, as one might expect, although at this point it might be equally accurate to say "spreadsheets" generically. I'm not sure how many of them use advanced features that are unique to Excel. They will use the one with the least friction. With that said, Excel probably still has the most finely tuned UI.
The traditional engineers use Excel. Very few of them program. The ones who can do it well enough to get paid for it, have joined the software industry. The rest are pretty adamant that programming won't get the job done quicker or more reliably than Excel.
The software engineers use Excel. They're good at whatever their role is within the software team, but they use Excel for the same kinds of short-term or one-off problem solving that everybody else uses it for.
Do some spreadsheets get too big and unwieldy for their own good? Sure. But that would also be true if you let the managers and traditional engineers write their own code for solving the same kinds of problems. I know this because I write that kind of code. Nothing will ever bridge the gap between small and large projects, that exists in most mature organizations.
1. Clear Separation between UI/UX and Backend: By separating the user interface from backend processes—perhaps as a module or library—Excel could maintain its clean, familiar interface while supporting more complex calculations and data handling in the background. This concept is somewhat akin to “Microsoft Excel Services.”
2. Multi-Language Formula Support: Similar to how VSCode supports multiple programming languages, Excel could allow users to choose between languages for cell formulas. Many users are comfortable with the current formula syntax, but it feels outdated, and even Google Sheets has (obviously?) largely replicated this old model. Allowing for different languages, while keeping the Directed Acyclic Graph (DAG) structure, could enable more advanced and flexible workflows.
3. Enhanced Data Types and Representations: Cells or groups of cells could support stronger data types, richer formatting, and custom representations, like embedded charts.
4. Integrated Data Connectivity: Excel could come with built-in tools to connect to external data sources and export data in structured formats, perhaps managed with an external orchestrator. I know you can do this with Excel and some other tools but I would like to just mark a cell and indicating that it can be consumed in a specific endpoint.
5. No artificial limits to the number of columns and rows.
7? Beef up the recommended pivot tables. Not to jump on the AI bandwagon, but any table that has Sales$ and dates is ripe for suggesting sales by qtr, week etc. Let the AI inspect the table and make suggestions.
8. Beef up the programmability. The Record Macro feature emits VBA but the code is very literal (eg it sees you move to row 12345, and records that, not understanding that what you mean is 'move to the last row') and the IDE is primitive. Fold in copilot experience to the IDE, modernize the UI, and allow macros to be integrated with github instead of buried in some personal.xlb file that is all too easy to lose.
For the specific set of problems it is designed to solve, its like shooting the best sniper rifle ever made -- point, calibrate, pull trigger, win.
For any other problems outside of its scope, its like trying to break the world record for Nurburgring while driving a short bus.
The UI/UX is so much better than Excel's and you can export to .xlsx format.
Don't get me wrong--there are some projects that require the functionality that Excel has. Using it gives me 90's software PTSD but sometimes you have no choice.
Numbers is a fine way to keep a shopping list, or a household budget, I'm sure
Don't be fooled by the marketing—Numbers has 90% of Excel's core functions; it even had RegEx before Excel did. IMHO, Numbers UI/UX is dramatically better than Excel's.
Of course, it doesn't have Excel's advanced features or access to plugins or to Python scripts. The users who need that know they need Excel… but you can go surprisingly far with Numbers.
Like I originally said, there are some tasks where you absolutely need Excel.
The thing that stands out to me in these stories is that spreadsheet programs are a (1) the most common site of programmability that features in anyone's lives, even non-programmer; and (2) for many tasks, spreadsheets sit at some kind of local optimum in a tradeoff between simplicity and power. (There's surely something for engineers and designers to learn from the second point.)
It seems like a connection worth celebrating, between people like us and the mostly non-programming public— a little piece of overlap between two worlds.
https://youtu.be/-gYb5GUs0dM?si=vzOGscnTURqhdyDe
Warning - this video shows Clippy, the very irritating paperclip!
You can technically accomplish about any programming task you want in it, albeit with poor ergonomics for anything sufficiently complicated, and a small but real push in that area of work will lead you to use a more accepted form of "programming".
However, due to whatever environment, desired or perceived skillsets (or limitations), other pressures, people stay and remain with that local tooling minimum for things that should have outgrown it. If you ever meet an office's "excel person" you instantly understand this phenomena.
I will completely agree with almost any list of deficiencies that you come up with for R. But I will disagree that almost anything on that list actually matters for an academic lab using it to conduct analyses.
As for it's use outside of academia, I couldn't comment.
I 100% get the appeal of excel for many tasks and realistically when used as a tool it is extremely powerful and removes a lot of things (like how to display the data) out of what I need to care about.
I’ve never done anything crazy heavy with it, but functions and stuff work fine for me I guess.
> What can Excel do that Calc can’t?
For one, a working copy paste. Seriously, I’ve lived with a copy paste bug in LibreOffice Calc (Windows) for years (it wasn’t yet fixed as of a few months ago). It will either paste an older copy or not paste anything even though I just hit Ctrl+C (sometimes multiple times) within Calc. This is one of the most irritating things about Calc.
Fonts and formatting is another thing that Calc miserably fails at in relation to copy paste. Multiple cells will look very different even though the font, size, etc., will appear the same in the dropdowns below the menus (when these cells are in focus). LibreOffice Calc, in many ways, portrays the worst of what one would expect from an open source project (I say this as a supporter/donor of Document Foundation).
The other part about what Excel can do that Calc can’t is that Microsoft (at least with the MS365 subscription version) has been adding more functions to improve older functions and features. I don’t remember everything, but SEQUENCE and XLOOKUP are two functions that come to mind on this aspect. LibreOffice Calc lags behind Excel on functions by several years.
LibreOffice Calc, though it’s a lot newer/younger than Excel, also has ~weird~ different keyboard shortcuts compared to Excel (e.g. adding a new line within a cell is Ctrl+Enter whereas it’s Alt+Enter in Excel; there are many more like this).
Nevertheless, I use Calc because I can’t bother to pay Microsoft and since Excel might be the only application that could be of value to me in its Office offerings (MS Word is total crap, PowerPoint isn’t as keyboard friendly as I’d like it to be, and Outlook is another total piece of crap).
Excel does have more functions but 90% of people don't need those. If you do, you do.
Weird. I've used Calc extensively on Windows and Linux and have never come across this
Calc is ok if you want to work with small spreadsheets and don’t care about the occasional bug.
A few important concrete examples would be much more interesting. In fairness, you did somewhat, but I would be interested in some more details.
of course custom code is better at pretty much anything. I've been a unix user since the bsdi days and so to me spreadsheets are a cruel toy running in a blue tinted mental prison, but when you see excel as a different and sophisticated UX metaphor for a fully featured machine that the user uses for raw arithmetic computation, and as an alternative to a command line or a browser, it's much easier to respect. that's absolutely essential for understanding it and its users.
the thing about excel is if you don't use it i guarantee you work for someone who does as the tabular metaphor is the system of the world. even if reality is more complex and needs more complex objects to model it, human beings ultimately organize around the contents of spreadsheets. hand wavy, but I don't know how else to tell developers and product people they need to take excel seriously.
I started to really loath the spreadsheet data model. "it's a big bag of cells" was not really doing anything for me. I started wanting better data structures and programing environments to work on that data.
For the data storage and basic calculations on that stored data role, I mainly use a relational database, most of what I wanted in a spreadsheet was row level security, that is, for my rows to stick together(one to many bad sorts I guess), and relational databases provided this in spades.
I wish there were better tools to help excel users migrate to more formal coding. Something that allows the immediate visibility and accessibility of Excel code, but avoids some of the problems of updating a formula in one place, but missing another, allowing better testing, and type safety for data.
This is actually the root problem of trying to improve or extend Excel--it has a truly ancient, horribly broken sharing/ipc/embedding model that is integrated into everything and can't be easily fixed or worked around.
It's just a pity the editor is literally worse than Notepad. And the implementation wants to reload a file twenty times. And the security stuff doesn't really work, so you're constantly turning that off. And and and.
Even today, complex logic in Excel is mostly done through VBA, JavaScript, Python and the like.
One thing I’ve had success with lately is chat-gpt 4o to manipulate large sheets. Sometimes it is really dumb and ruins my data, so always keep the last iteration handy… Sometimes it does some really great things, especially comparing multiple excel files.
Another fun facts, on average people spent 10% of their working life using Excel.
Some companies dont want to learn a new tool
Or pay for a new tool like itglue/bitwarden/keeper
Yes, some things could be done better with a separate application or custom software, but that requires approval and contracts etc.
With Excel you can just build that crap yourself and share it or not.
Gratitude and an upvote to the person who can attach a name and citation to the rule.
The IT industry overall remains similarly entrenched. Enterprise software vendors insist on Windows dependency, staying tethered to outdated standards; upholding an inefficient status quo that serves only MSFT, developers, and technicians—not users nor progress.
Not saying you're doing this but it's not uncommon for someone to say something is slow when it's a version from 10 years ago.
