Spreadsheets Are Hot–and Cranking Out Complex Code

I&#x27;ve learned products designed to replace spreadsheets have a huge hurdle because the people who use spreadsheets treat operating the sheet as their job. Replacing them removes their autonomy and control over an information process, and subsumes the value they bring to their employers - so they will resist products that threaten that. Excel is a complete management subculture.<p>The other advice I give is if you are generating analytics, have a PowerBI connector of some kind because the people who make decisions (managers, etc) make them based on PowerBI, and not from an interface their staff is a peer at using, and likely has control over. In enterprise, they want data in metrics their staff can&#x27;t see, hence a separate tool.<p>Spreadsheets will always be with us I think. The opportunity may be in creating one that is has sufficient work-alike features with legacy ones, with new power features (python, etc) where there is a connector between the high power open development environment, and the familiar Excel ones managers use. Key thing being not asking managers or sr. employees to change.

Warning: I&#x27;m a founder working in the spreadsheet space, so take the rest of my this comment with a (large) grain of salt. I&#x27;ve written before [1] (HN and elsewhere) about how I think spreadsheets are the most popular programming paradigm ever, we just don&#x27;t talk about it much. As this article mentions, there are many ways we can push this forward.<p>I personally think the most powerful low-code spreadsheet tools we can build are those that allow spreadsheet users to easily transition to full programming languages, if they want to. So rather than locking users into limited and proprietary product number #115 (some of them are mentioned in this article), IMO it&#x27;s better if users can transition to a full programming language (like Python) very naturally. Som I&#x27;ve spent the past 2 years building Mito [2].<p>Mito is a spreadsheet extension to your JupyterLab environment. You can display any Pandas dataframe as a spreadsheet, and edit it in a very similar way to Excel. For each edit you make, it generates the corresponding Python code below for those edits. Practically, you can think about Mito as recording a macro, but instead of generating scummy-crummy VBA code, it generates Python.<p>We&#x27;re open core [3]. Feedback greatly appreciated!<p>[1] <a href="https:&#x2F;&#x2F;naterush.io&#x2F;blog&#x2F;Spreadsheets-are-the-Ultimate-Progr" rel="nofollow">https:&#x2F;&#x2F;naterush.io&#x2F;blog&#x2F;Spreadsheets-are-the-Ultimate-Progr</a>... [2] <a href="https:&#x2F;&#x2F;trymito.io" rel="nofollow">https:&#x2F;&#x2F;trymito.io</a> [3] <a href="https:&#x2F;&#x2F;github.com&#x2F;mito-ds&#x2F;monorepo" rel="nofollow">https:&#x2F;&#x2F;github.com&#x2F;mito-ds&#x2F;monorepo</a>

Every day I think about how Microsoft Access allowed otherwise not-so-tech-savvy users to, with just a little training and practice, build a complete relational database for their entire business, supported by a relatively sane GUI and a way to build forms and reports with very little (if any) &#x27;code&#x27;. I have no idea what small companies are using nowadays, but I think there has to be some untapped middle ground somewhere between Microsoft Access and a dumb spreadsheet.

Founder of excel collaboration and versioning startup [1] here so I am a believer.<p>There are only about 30 million programmers. There are over 1 billion Excel users. Excel is Turing complete. Excel is <i>by far</i> the most used programming language on the planet. It is easily 20 times more popular than the next contender.<p>The value of Excel is that it is presenting the data, with a set of formulae that let you keep derived data up-to-date. This inferred data provides sums and computations, sometimes simple, but sometimes exquisitely complex. And through this whole range of complexity, with a billion users, virtually nobody treats Excel seriously like a programming language.<p>We have a programming language which is essentially acting as a declarative database, and yet we don&#x27;t do unit tests, we don&#x27;t keep track of changes, we collaborate with Excel by sending it to our colleagues in the mail and god-forbid we should doing any serious linting of what is in the thing.<p>Anyone who has used Excel in anger realizes why it is so brilliant. Show me another declarative constraint based, data driven inference language that I can teach to my grandmother.<p>The problem isn&#x27;t Excel. The problem is that we are treating Excel like its a word processor, and not what it is: a programming language.<p>[1] <a href="https:&#x2F;&#x2F;versionxl.com&#x2F;" rel="nofollow">https:&#x2F;&#x2F;versionxl.com&#x2F;</a>

We are a small B2B, and during the pandemic our primary source of income disappeared. Our president built out a new service, one that is VERY analytics heavy. He built it in weeks using Excel, rather than the months it would have taken using &quot;proper development&quot;. It&#x27;s a beautiful monstrosity, and eventually needs to be ported to our codebase, but it saved the company.

Excel is great and versatile, yet lacks some critical features.<p>- It&#x27;s easy to load data from database or API without VBA, yet impossible to write updates back without VBA. With VBA it&#x27;s still messy string concatenation of SQL queries.<p>- VBA is security nightmare<p>- Version control is bad. Distributing spreadsheets in emails and collecting changes back is nightmare. Sharing a file on network drive lacks fine-grained permission control.<p>- It&#x27;s hard to maintain proper normalized relational data model. It&#x27;s impossible to abstract the normalized model from the user (i.e. show labels in selectboxes but store ID&#x27;s)<p>- It&#x27;s locale dependent. Date formatting, column separator (comma&#x2F;semicolon separated), even function names. Unusable in international data exchange.<p>- It mixes formatting, logic and data. Impossible to make reusable blocks. New lambda functions help somewhat.<p>And so much more.

&gt;Suddenly, the field has begun to bloom. A small cluster of startups have in the past year released spreadsheet products<p>All of which are subscription based SAAS. Maybe I am showing my VisiCalc oldster roots, but I want my document related tools to be pay once &#x2F; run locally.<p>I long for a computing world where the free tier is run-in-a-browser-on-someone-else&#x27;s-computer and the paid tier is run local and native.

I’ve been doing this for customers for years now using Microsoft Lists (sometimes I use Excel) as the back end and PowerApps as the front end gui&#x2F;data entry. Now the language from PowerApps (which is based on Excel) has been open sourced and Microsoft want to make it available in other apps too. That has questionable value but my solutions continue to be popular because the people I make them or can easily edit and make simple changes.<p>Some of the solutions we make have been widely adopted too, with hundreds and in some cases thousands of users.

I just learned how to use spreadsheets recently, and <i>I love them.</i> As a programmer, spreadsheets are <i>exactly</i> the kind of tech I want to use every day. User-friendly point and click, IDE features, complex functions mapping data across rows&#x2F;columns&#x2F;tables. All I want now is to hook it up to Postgres, and then maybe a new app to automatically design simple web pages that pull out and display data with the same functions. This would save me months of time per year versus manually coding the same things.

Excel users tend to do a far better job of data storage, analysis, and forecasting than most of overpaid SaaS-reliant &quot;modern data team&quot; (Data Engineers, Data Analysts, Analytics Engineers, Data Scientists).

As much as I hate to admit it, it is going to be a <i>long</i> time before we get rid of spreadsheets in de facto production use. And for that, I <i>really</i> would like to see one thing:<p>I&#x27;d like to have a separate worksheet type for &quot;datasheets&quot;. Looks and behaves just like an ordinary worksheet, but:<p>- plugging in a formula applies the formula on every cell of the column. No exceptions. - you can not have different data types in cells within one column. That is, if you have dates in a column, you can&#x27;t have a string in one cell of thecolumn etc.<p>Yes, I know about powerthings in excel. No, I do not want to mess with those. Just normal spreadsheet formulas and sheets.<p>(Okay, a couple of other things: Get rid of vba and bring in the full python ecosystem instead. And if not yet available, version control. It is luckily a while I have worked with excel, so these may be outdated comments)

People always want to reinvent the spreadsheet with their latest whizzbang solution without a problem. But Excel solves 80% of the common problems so nobody really cares.

Excel, and spreadsheets, need a re-branding campaign. They are cloud-based, low-code, ELT tools; these terms are typically associated with buzzy enterprise tech companies, not technology developed in the 1980’s.

What do are you using spreadsheets for privately? I never dig deep into their capabilities, so I mostly use them to track expenses within some particular scope, e.g. healthcare. When I recently wanted to compare several loans and estimate our financial situation several years in the future, I wrote some Python code and used a Jupyter notebook to enter parameters and make plots. Has any of you done something similar using spreadsheets?<p>On a side note: I didn’t find a Python library for time series generation (not analysis). Something where you can build some models (e.g. loan, income, expenses) which depend on a common parameter (time) and then evaluate all your models for different values of the common parameter. Right now, I generate pandas series&#x2F;dataframes and combine them afterwards, which also took some massaging of pandas (which I also usually don‘t use a lot).

I think &#x27;products replacing spreadsheets&#x27; is not the right angle. Spreadsheets are great but not for everything.<p>I think there are a lot of mis-uses of spreadsheets. When users record list type data, long-form content, relational data, etc within spreadsheets, they are probably better off using an app - they, or their IT team can build.<p>I once seen an enterprise org process where a health and safety check was completed on a clipboard. That user would go back to their desk, push that data into excel, another user would write a script to push that excel data into MySQL.<p>There are many use cases where spreadsheets are perfect - building financial models.<p>For transparency - I am the cofounder of a no&#x2F;low-code tool called Budibase. I am also a happy spreadsheet user. <a href="https:&#x2F;&#x2F;github.com&#x2F;Budibase&#x2F;budibase" rel="nofollow">https:&#x2F;&#x2F;github.com&#x2F;Budibase&#x2F;budibase</a>

All this and not one mention of DabbleDB !!<p><a href="https:&#x2F;&#x2F;www.youtube.com&#x2F;watch?v=MCVj5RZOqwY" rel="nofollow">https:&#x2F;&#x2F;www.youtube.com&#x2F;watch?v=MCVj5RZOqwY</a><p><a href="https:&#x2F;&#x2F;news.ycombinator.com&#x2F;item?id=102" rel="nofollow">https:&#x2F;&#x2F;news.ycombinator.com&#x2F;item?id=102</a>

Check out Budibase.<p>It&#x27;s the leading open source low code platform and perfect for building UIs on top of spreadsheets<p><a href="https:&#x2F;&#x2F;github.com&#x2F;Budibase&#x2F;budibase" rel="nofollow">https:&#x2F;&#x2F;github.com&#x2F;Budibase&#x2F;budibase</a>

Excel is by far the most successful programming language and IDE. People love to hate it (and the people using it), which is somewhat misguided: there is simply no way to change people (and they keep making new ones), so telling them that what works for them is somehow wrong is both wrong <i>and</i> doesn&#x27;t work.<p>Instead, the spreadsheet paradigm has the promise of being far more powerful. Jupyter notebooks are one example of adapting it to a different realm, and it <i>also</i> ended up being used everywhere and looked down upon by the snobs.

Been thinking about this problem for a long time. I use python&#x2F;jupyterhub notebooks for our data analysis flows at work. I’ve become an expert at it and am a go-to person in our org… but even still when I have to try something new on “small” amounts of data, I _still_ go to excel first. It’s barrier to get the results you want os extremely low - thus-far unparalleled by any other software I’ve used in the last 20 years… and I have a hard time seeing anything that would replace that (though maybe I am just set in my ways…)

At uni, one professor used to say: &quot;spreadsheets have all of the problems of software engineering, but none of the solutions&quot;

&gt; Spreadsheets.com, for example, lets users dump almost anything into a cell. Drop a photo or a PDF into a cell and the product will immediately create a thumbnail, which you can then expand, as if the spreadsheet were some sort of blog content-management system.<p>Yes please! Can I get this for Google Sheets? :&#x2F;

Of course you can do many stuff, except access control and server side validation. Spreadsheets rather shine with quick planning &#x2F; analytics.

This is a ridiculous piece of native advertising.

<a href="http:&#x2F;&#x2F;www.paulgraham.com&#x2F;submarine.html" rel="nofollow">http:&#x2F;&#x2F;www.paulgraham.com&#x2F;submarine.html</a>