excel with python + chatgpt with python is about to give us an entirely new generation of "AI powered" start ups that make even less sense than before and I cannot wait
@@julianomoura3631 that's going to completely depend on what features of Python Microsoft allows to be executed in the cloud. I'd imagine any native functions or packages involved in data transfer (i.e. database connections, email, etc) are going to be excluded or disabled in the runtime. Not to say hackers won't find a workaround (they figured out how to do that with Power Query, a read-only ETL language in Excel), but having the actual execution take place in the cloud is going to be a very different challenge for hackers.
I have love/hate relationship with Microsoft as a software developer AND media designer. I love .NET and C# and it's been and still is my strongest field in software development (since I began with VB and .NET as a kid), and Microsoft nailed it with .NET Core and their ideas, but some fields are just plainly fucked as if the devs had no idea on how the framework is supposed to be (see Blazor and the .NET 8 changes, introducing a completely new structure out of no where)
@@386enhanced I remember in the early 2015 when Nadella had just started the ramp up to change Microsoft from the old "Lets buy things to kill them" to the current dev-friendly approach. They announced Visual Studio Code and I was just floored. "What do you mean when you says its a free editor without 40% of screen estate taken by an obnoxious Ribbon UI? AND it supports Linux and Mac? Ha! Thats def. a late april fools joke". Followed by them actually realeasing it and being left dumbfounded a second time going "What have you done with Microsoft?!" And now I just thought back at the fact that Github was bought by Microsoft nearly FIVE years ago... and yet it still looks and feels like Github.
Oh no, why do that? Just come up with your own extensions to python that slowly become more and more incompatible with the original language and all other packages.
This is probably the worst thing to happen since the inception of excel. The number of businesses running off the back of an unstable excel spreadsheet is horrifying.
Seriously, i imagine all the incredibly convoluted macro powered excel files , and now i imagine the python powered equivalent files that will come out of this.
Once they realize that everything they do in Excel can be done better in Python+NumPy+Jupyter+Pandas+Matplotlib+all the rest of it, then Excel just becomes the shell you start to get to the other stuff. Microsoft will likely make some money off the crowd who are too dumb to realize that they can dump the Excel part and use the rest of it ... for free.
Building a backend with excel is probably a shitty idea tho, for which advantage ? It's probably slower to figure out what to do and come up with "hacks" than learning some basic SQL which takes a day at most
@@heroe1486 if the customer is using Excel as a backend AND frontend and it hasn't scaled out of control, then he doesn't either need SQL or your services for that matter. The minute the excel sheet gets corrupted or the data grows beyond the 65k lines, then you'll get the call
I think a Jupyter notebook would be far more flexible: each cell can hold as much or as little code as you like, and can produce multiple associated items of rich output - graphics, sound, video, maths, formatted HTML, plus interactive widgets to control same - interspersed with cells containing narrative text so your notebook can be self-documenting. And you can include code in other languages besides Python.
Not quite. It runs in an Azure cloud, meaning that whatever packages and versions will be available to you are only going to be the ones Microsoft approves. And if it's anything like the ol' MRAN R language repository (that Microsoft just shut down), you'll be two years behind the latest releases.
@@adamestrada7610Yeah, conda-main tends to run a bit behind conda-forge, which itself can be sometimes be a point release behind PyPI. But it's usually not quite "apt repository" level of staleness.
@@GSBarlev Totally, but if I know Microsoft as I do, they'll probably be running package versions at least six months behind because they'll need to ensure that those packages don't conflict with core Excel functionality. This is similar to Power BI where only Python 3.7.7 is currently supported, which was released all the way back in 2020
Being a Google Apps Script developer for more than a decade, knowing the incredible potential it has, and how it is mostly unknown to the masses, it was a happy surprise to see it mentioned in such good terms, thanks Jeff!
My job has the option to use Apps Script and I was planning on giving it a go, but I didn't know there were people working only in it :0 I thought it was a minor thing for internal tools
Man they really had one of the biggest ideas for a Microsoft product ever - Python in excel - and implemented it in the worst, most crippling way possible - it has to run in the cloud. Incredible
Exactly. Wtf. I was so excited for this, to pitch python to my non-programmer friend that manually processes a lot of data in excel sheets at work. But now i find out about this cloud shit .....😤
@@thisaintmyrealname1 same for me. Would’ve loved to tell some people I know about this. What’s the point of a tool that can’t be used locally. Cloud is supposed to provide infrastructure and management benefits, not act as a paywall to real features. Dumb.
This is a great feature that will be very beneficial to those Excel users that are comfortable with a little scripting. However, I shudder at the thought of pre-existing, _already over-bloated Excel files_ gaining hundreds of lines of Python code strewn about various cells across multiple sheets.
Quadratic is a web-based spreadsheet application that runs in the browser and as a native app (via Electron) [2]. It is an infinite canvas spreadsheet with built-in Python, SQL, and Formulas[1][5]. Quadratic is designed to enable users to pull data from its source (SaaS, Database, CSV, API, etc) and then work with that data using the most popular data science tools today (Python, Pandas, SQL, JS, Excel Formulas, etc) [2]. Quadratic has several features that make it unique, including: - Multi-line formulas: In Quadratic, you can expand your formulas to as many lines as you need[1]. - Python library support: Quadratic has Python library support built-in, allowing users to bring the latest open-source tools directly to their spreadsheet[1]. - Powerful AI integration: With GPT 4.0 powered autocomplete for data, formulas, and code, users have a powerful spreadsheet companion[1]. - Infinite canvas: Unlike other spreadsheets, Quadratic has an infinite canvas (like Figma), allowing users to pinch and zoom to navigate large data sets, and everything renders smoothly at 60fps[3]. Here's a summary of some of the key features of Quadratic: - Infinite data grid with Python, JavaScript, and SQL built-in[2]. - Data connectors to pull in data from various sources[2]. - Multi-line formulas and Python library support[1]. - Powerful AI integration with GPT 4.0 powered autocomplete[1]. - Infinite canvas for easy navigation of large data sets[3]. Quadratic is available for download from its website[1], as well as from GitHub[2] and the App Store[6]. Citations: [1] www.quadratichq.com [2] github.com/quadratichq/quadratic [3] news.ycombinator.com/item?id=35456509 [4] sourceforge.net/projects/quadratic.mirror/ [5] www.producthunt.com/products/quadratic [6] apps.apple.com/us/app/quadratic-master/id351899924
@@Caellyan I believe we shouldn't blame the tool if we don't know where to use it and where not to use it. I started my career with Excel, after few years, I realised that it can't handle some of my increased work load so I shifted that work to Python. It's not really hard to understand where to use a tool if you know its strengths and limitations.
@@lawrencedoliveiro9104 lol they don't even offer Excel for free. And I highly doubt that unless you are on an enterprise plan, those cloud-based calls to a Python runtime are going to be throttled hard.
not as good as macros though, the implementation is crap. E.g I cant access the file system to copy data from multiple spreadsheets, and create an automated data feed from hundreds of spreadsheets into one.
The fact that it all runs in the cloud makes it a complete non-starter at my company. Everything has to be on the company network to comply with data protection laws.
I litterally used python in libreoffice 2 days ago and thought: "It's so cool! Why doesn't Microsoft have this feature?" Honestly python is so good for handling data imagine having to use the basic language
@@ajarivas72 vba macros are getting switched off all over the place recently because of the security risks associated with them. I love VBA, but yeah you can destroy someones computer with a macro.
also.. Quadratic is a web-based spreadsheet application that runs in the browser and as a native app (via Electron) [2]. It is an infinite canvas spreadsheet with built-in Python, SQL, and Formulas[1][5]. Quadratic is designed to enable users to pull data from its source (SaaS, Database, CSV, API, etc) and then work with that data using the most popular data science tools today (Python, Pandas, SQL, JS, Excel Formulas, etc) [2]. Quadratic has several features that make it unique, including: - Multi-line formulas: In Quadratic, you can expand your formulas to as many lines as you need[1]. - Python library support: Quadratic has Python library support built-in, allowing users to bring the latest open-source tools directly to their spreadsheet[1]. - Powerful AI integration: With GPT 4.0 powered autocomplete for data, formulas, and code, users have a powerful spreadsheet companion[1]. - Infinite canvas: Unlike other spreadsheets, Quadratic has an infinite canvas (like Figma), allowing users to pinch and zoom to navigate large data sets, and everything renders smoothly at 60fps[3]. Here's a summary of some of the key features of Quadratic: - Infinite data grid with Python, JavaScript, and SQL built-in[2]. - Data connectors to pull in data from various sources[2]. - Multi-line formulas and Python library support[1]. - Powerful AI integration with GPT 4.0 powered autocomplete[1]. - Infinite canvas for easy navigation of large data sets[3]. Quadratic is available for download from its website[1], as well as from GitHub[2] and the App Store[6]. Citations: [1] www.quadratichq.com [2] github.com/quadratichq/quadratic [3] news.ycombinator.com/item?id=35456509 [4] sourceforge.net/projects/quadratic.mirror/ [5] www.producthunt.com/products/quadratic [6] apps.apple.com/us/app/quadratic-master/id351899924
Grist and Quadratic are both web-based spreadsheet applications that offer unique features. Here are some key differences between the two: Grist: - Offers a unique "visual data modeling" approach that allows users to create custom views of their data[1]. - Has a powerful formula editor that supports advanced functions like array formulas and nested functions[1]. - Offers a wide range of templates for different use cases, such as project management, inventory tracking, and budgeting[1]. - Has a user-friendly interface that makes it easy to get started with[1]. Quadratic: - Offers built-in Python, SQL, and JavaScript support, as well as powerful AI integration with GPT 4.0 powered autocomplete[1]. - Has an infinite canvas that allows users to navigate large data sets with ease. - Offers data connectors to pull in data from various sources. - Has multi-line formulas and Python library support[1]. In terms of alternatives, both Grist and Quadratic have competitors such as Google Sheets, Microsoft Excel, and other spreadsheet applications[1]. However, each of these applications offers unique features that set them apart from one another. Ultimately, the choice between Grist and Quadratic (or any other spreadsheet application) will depend on the specific needs and preferences of the user. Citations: [1] stackshare.io/getgrist/alternatives
I find this incredibly awesome. Excel becoming more powerful and sophisticated is something with which I am wholeheartedly pleased to see. Love it or hate it, you can't deny the significant influence and usefulness of Excel.
The fact that I had to learn the Microsoft VB Excel language just to make my own "function" like 3 weeks ago, to only find out this has happened. Gosh, I hate this year.
Quadratic looks very good, haven't tried yet though... Quadratic is a web-based spreadsheet application that runs in the browser and as a native app (via Electron) [2]. It is an infinite canvas spreadsheet with built-in Python, SQL, and Formulas[1][5]. Quadratic is designed to enable users to pull data from its source (SaaS, Database, CSV, API, etc) and then work with that data using the most popular data science tools today (Python, Pandas, SQL, JS, Excel Formulas, etc) [2]. Quadratic has several features that make it unique, including: - Multi-line formulas: In Quadratic, you can expand your formulas to as many lines as you need[1]. - Python library support: Quadratic has Python library support built-in, allowing users to bring the latest open-source tools directly to their spreadsheet[1]. - Powerful AI integration: With GPT 4.0 powered autocomplete for data, formulas, and code, users have a powerful spreadsheet companion[1]. - Infinite canvas: Unlike other spreadsheets, Quadratic has an infinite canvas (like Figma), allowing users to pinch and zoom to navigate large data sets, and everything renders smoothly at 60fps[3]. Here's a summary of some of the key features of Quadratic: - Infinite data grid with Python, JavaScript, and SQL built-in[2]. - Data connectors to pull in data from various sources[2]. - Multi-line formulas and Python library support[1]. - Powerful AI integration with GPT 4.0 powered autocomplete[1]. - Infinite canvas for easy navigation of large data sets[3]. Quadratic is available for download from its website[1], as well as from GitHub[2] and the App Store[6]. Citations: [1] www.quadratichq.com [2] github.com/quadratichq/quadratic [3] news.ycombinator.com/item?id=35456509 [4] sourceforge.net/projects/quadratic.mirror/ [5] www.producthunt.com/products/quadratic [6] apps.apple.com/us/app/quadratic-master/id351899924
@@randomuser5237 Adding onto this, you also can't manipulate the components of Excel object tree or develop custom UI forms with Python, only data manipulation in data frames and visualization output. As much as I'd love a full replacement for VBA, that language is like a cockroach: ugly, unwelcome, and will probably survive a nuclear apocalypse.
Javascript would be the sensible alternative to vba if you think of the pieces that ms already has. JS is already designed to be a document manipulating language. You would just have to add more host objects to it to do what vba does. On that note, I'm not sure why people hate vb so much, but whatever.
I cannot stress to you how validated I feel hearing you praise Apps Script. It’s my daily rider at work and has advanced my career in very serious ways. I’d love to hear you talk about it more tbh.
Using app scripts I automated my company HR onboarding process with no costs (besides initial time on development). It takes Google form, fills a Google docs template and saves it into pdf then sends it as an email. 😀 it generated possibly thousands and thousands of documents, it's very reliable and doesn't cost anything. App Scripts ftw
I think this video should have mentioned that JavaScript is already supported by Excel and is to my understanding executed locally (either within the browser when using the web version or webview when using the native app). The feature (called Office Add-Ins or Office.js) has additional benefits compared to the Python announcement: - Available on Windows, Mac and Web (Python is only available on Windows thus far) - Allows you to package and deploy your code as an add-in The Python use-case really seems to be to do something in only one workbook, whereas the Office.js stuff is for implementing something you would like to distribute for people to use across workbooks.
If your company allows it. My companies group policy and sharepoint auth disables it in desktop and browser. I'll be interested in seeing how they will handle the python function execution. They may black list the api call in our proxy, or it may just go unnoticed. Making my life significantly easier since they blacklist pandas because of c++ build tools requirement. They don't want people compiling code on their machines.... It is pretty cool when you have it
@@batmanatkinson1188They cannot block networking modules like urllib and requests because data analysis often has remote dependencies. Instead, they would be running this inside an isolated Kubernetes cluster and set application level permissions for the running process to prevent direct communication with its container. And even if one container gets compromised, there's usually security in place to prevent access to other containers in the cluster. (Or the devs are underpaid and didn't bother putting up all this, so I'll try to get the service account tokens 😂)
@@lawrencedoliveiro9104No, conda is an alternative to pip (really poetry) that is used for creating virtual environments and installing packages. Unlike venv, it's not just limited to Python libraries and can install binaries for other languages and even alternative versions of Python. The package solver was historically very slow (as opposed to pip's solver, which was historically VERY BAD) but it's gotten better lately. There's also a drop-in replacement called mamba that is super-fast.
Remember that Microsoft is doing this because it has to, not out of the goodness of its heart. The only reason it has to pay attention to Python is because it is dominating the professional-quality data-analytics industry, and leaving the noddy Excel jocks in the dust. This is a way to make those Excel jocks feel they are still somehow relevant, and also pay Microsoft for the privilege.
@@Alticroo yeah ive done it before. looks something like this Public Function ShellRun(sCmd As String) As String 'Run a shell command, returning the output as a string Dim oShell As Object Set oShell = CreateObject("WScript.Shell") 'run command Dim oExec As Object Dim oOutput As Object Set oExec = oShell.Exec(sCmd) Set oOutput = o Exec.StdOut 'handle the results as they are written to and read from the StdOut object Dim s As String Dim sLine As String While Not oOutput.AtEndOfStream sLine = oOutput.ReadLine If sLine "" Then s = s & sLine & vbCrLf Wend ShellRun = s End Function
Or try a Jupyter notebook. Each cell can have any number of lines of code, and produce any number of items of associated rich output-graphics, charts, sound, video, mathematical formulas, even arbitrary HTML. And interactive widgets to let the user control that cell output. You can intersperse code cells with narrative text cells, so the notebook becomes self-documenting. And you can use languages besides Python, if you want. Where is there a spreadsheet that comes close?
@@lawrencedoliveiro9104 yeah jupyternotebook is awesome and useful but thats beyond the scope of this context. we are talking about using excel because excel rocks. jupyternotebook is not better than excel, its useful but not as practical for the average user.
@@lawrencedoliveiro9104 I think both of these options miss the point - which is to have a complete spreadsheet + code environment in one package. Python isn't part of Excel, hence developing a project in both Excel and Python requires that any other contributors have both installed; which isn't always feasible. Being able to tap into python through Excel with just your regular run off the mill corporate Office package is a big deal for a variety of industries; both approaches seem to come from a developer or data scientist / engineer POV where the having Python is basically a guarantee
Not running the code locally is too ridiculous to describe. They could run it in a local container. Also VBscript can in practice do anything the user has permissions for anyway, and a warning pop-up is still good enough for VB-macros. My guess is that they want to add value for azure services, and market the ability to run python code safely as a revolutionary advance in office work.
Problem is VB Macros are no longer allowed to run in most government departments, unless it is on a vm. Personal experience on that one. Redeveloped a whole bunch of VBA Macros for a govt dept, only for their IT dept to switch it off as a security risk without telling anyone, and then I (a third party dev) get blamed for them not working....
Honest question: Why do people still use excel instead of a programming language like Python? Is it only because it is easier to learn and required for many jobs, or am I missing something extra that Excel has?
It is easier, and therefore, widely used. Interacting directly with data and coloring cells is, though useless to a programmer, very useful to a true general public. Because it is so easy to input and edit data and higher level functions are not really needed to a general audience, they don't look for something more like incorporating Python or SQL unless their spreadsheet can't open. And even then, they would just make another file called "database_2" Microsoft did a hell of a good job to make simple programs for an audience that don't know much about computers and that work even if you dont do them properly. And now, they raised the ceeling. Just my opinion based from experience
Because it’s higher level. Excel isn’t really optimized for big data workflows, nor should it be used for those. That’s why programmers that work with big data use other tools. But you want to outline a calendar, a todo list, or even record a simple set of records that stay on disc and you make small edits from time to time, then excel is great.
Excel spreadsheets are notoriously hard to debug. What’s worse is that most of its userbase has no idea about debugging techniques anyway, since that requires programming knowledge, and they use Excel because they hate programming. There have been many embarrassing cases of wrong data being published because of Excel spreadsheet bugs -- even in scientific research papers. When other researchers want to replicate those results, they ask for the original spreadsheets, only to discover that their formula ranges are incorrect, that kind of thing. The most depressing case I heard of is that geneticists are changing the names of some genes, because when they import info into Excel, it wants to interpret those names as dates.
Why would you spend time coding something that can be done easily in Excel in seconds. Unless it's something that you do all the time and needs automating then there is no point. And don't forget that formulas are powerful and can accomplish a lot. Even if you used python you may still need a spreadsheet to present the results. Also big companies may not even allow you to install dev tools for security reasons. When excel really sucks is when you have to handle large datasets, it just can't and Access can't help either because it's garbage.
So you're saying we won't hear from analysts for about 6 months and then we will be flooded with broken python in Excel spreadsheets to replace? Thanks Microsoft!
This is neato, but I don't ever see most companies using it. These days, every company I've worked for only uses excel for simple tools that go to sales, or financial packages for upper management. Visual Basic is plenty for these uses. Anything more complex is just done in SQL. I cannot imagine a scenario where this would be preferred to SQL or even just regular python.
As a German it's curious to read these comments compared to those on German developer news sites. Almost everyone in the comments here ignores the obvious security risk that is "everything is computed in the cloud" while almost no German developer failed so hard at spotting this. What if your building's / company's / country's internet connection is lost? It's also a terrible business decision from a customer's perspective. You needlessly jump head-first into vendor lock-in. You're basically asking for a "subscription" pricing model that will cost you huge multiples of a buy-once product. For what benefit? Sucks to see how oblivious people in other countries are
There is already something similar in excel with Javascript (Excelscript) but obly available to companies and student accounts (my case) if I'm not mistaken, its actually quite similar to apps script which I've also used in the past but limited in that you cant make formulas with it for example, only scripts to run on the sheets, I have one that creates a graph and formats it the way I need to 90% of the time with a trendline and all, saved me hours already but yeah python gonna be awesome, can't wait :)
Right. Its not just that either, what these Tech tubers are describing is a unique excel function that was designed to appear as if it incorporates all of Python's features, when it only involves the utilization of the Pandas library
This is a massive security issue. You could write some quite sophisticated but simple and undetectable data exfiltration code and steal a whole database without anyone noticing.... For instance a malicious user connects database (via ODBC) to Excel locally, throws up a API endpoint using Flask and then steals the data dropping it into open S3 buckets from outside the organisation. They could even create an active gateway where they query the database in real time and have Python parse the SQL queries. There will be a generic hosting fork out soon enough avoiding Azure host and sidestepping security lock downs. Weaponised Excel's will be chucked all over the internet soon enough...... as a VBA master the future I dreamed of in the 90's coming true brings a wry smile to my face
VBA just doesn't have the capabilities or 200k + library extensions so it can't do much I wish it did but MS hobbled it in the late 90's in favour of .NET which really upset me at the time. Lots of tricks you can deploy once you are in like spinning up encrypted VM's, replicating existing windows processes, side loading payloads, intermittent extract, creating protective supervisor. Problem is the typical usage is to create a secure connection to a remote server for remote code exec which is very similar to what a malicious actor will want to do
@@WhiteboardMario198 think of this, the obvious place to go next on this is to turn it into a sideloading, transmorphic worm that refactors it self every few moments. Don't know what you want to steal or destroy? Load a small NLP model in there to read everything it sees and customise action based on what it find. No need for C&C, just find all the account number, credit card numbers etc let the worm decide on the fly tasking / triage and then plop out the goods in a single email attachment while the rest of the worm goes wild tripping every alarms on the network
"Write a plugin" When you say this, you need to be more specific in what you are suggesting. Write a plugin where? In excel? Why would you refer to it as a "plugin" and not a function?
now no one is gonna suspect of a script running through a excel file... i can now jump the security without making a specific C malware to move in the system, now i just open a spreadsheet in excel and start a script inside it, run it and jump over the security *cheff kiss*
1:52 This is probably the most I've ever learned from a meme. I'm currently using some of these, so I took a look at the rest and will def start using some more. This is brilliant :D
I think that adding Python to Macros instead of using VBA would be a much better approach. And running Python formulas in Azure makes no practical sense.
It’s a way to squeeze some more revenue out of those Excel jocks too dumb to realize that they can leave out the Excel part-indeed, the entire Microsoft part-and run a full professional-quality Python-based data-analytics stack for free.
Bloomberg Finance did a presentation at a Python conference a few years ago, where they demonstrated how they create quite amazing-looking custom dashboards for their users inside Jupyter notebooks. They even open-sourced the add-on toolkit they use to do it.
excel with python + chatgpt with python is about to give us an entirely new generation of "AI powered" start ups that make even less sense than before and I cannot wait
you summed it up in the most pythonic way sir
exactly, a shitload of shit is out there awaiting
I can't wait to see "investors" losing money with some shitty services that does nothing, solves nothing, worth nothing, but they are cool and techy
Ya one of those apps which use "AI, ML" to calculate your BMI.
Just leave out the “Excel” part.
This will enable so many new and exiting ways for non programmers to get social engineered into executing untrusted code.
Hahaha ransomware goes brrrr
The code is ran in the cloud in a sandbox. I don't think it can affect your pc
@@shashu1999 but it can get data from the spredsheet and send to a remote address. Spredsheets may have lots of sensitive information about companies.
@@julianomoura3631 that's going to completely depend on what features of Python Microsoft allows to be executed in the cloud. I'd imagine any native functions or packages involved in data transfer (i.e. database connections, email, etc) are going to be excluded or disabled in the runtime. Not to say hackers won't find a workaround (they figured out how to do that with Power Query, a read-only ETL language in Excel), but having the actual execution take place in the cloud is going to be a very different challenge for hackers.
Hahahah
How can Microsoft consistently amuse me in both good and bad ways at the same time
Because there is nothing more critical to the success of a business than consistency.
I have love/hate relationship with Microsoft as a software developer AND media designer. I love .NET and C# and it's been and still is my strongest field in software development (since I began with VB and .NET as a kid), and Microsoft nailed it with .NET Core and their ideas, but some fields are just plainly fucked as if the devs had no idea on how the framework is supposed to be (see Blazor and the .NET 8 changes, introducing a completely new structure out of no where)
@@386enhancedthe aubcription model lol
@@386enhanced I remember in the early 2015 when Nadella had just started the ramp up to change Microsoft from the old "Lets buy things to kill them" to the current dev-friendly approach. They announced Visual Studio Code and I was just floored. "What do you mean when you says its a free editor without 40% of screen estate taken by an obnoxious Ribbon UI? AND it supports Linux and Mac? Ha! Thats def. a late april fools joke". Followed by them actually realeasing it and being left dumbfounded a second time going "What have you done with Microsoft?!"
And now I just thought back at the fact that Github was bought by Microsoft nearly FIVE years ago... and yet it still looks and feels like Github.
same lol
Managers everywhere celebrating that they can continue using a spreadsheet as a flat database instead of paying for a SQL developer: 🎉🎉🎉💃🕺💃🕺
> but it crashes when I open it
Does it crash *every* time you open it?
💃🕺💃🕺🎉🎊
@@nannan3347 'it works on my machine'
Best of both worlds: Unmaintainable imperative excel Formulars with none unit tested untyped python code. Perfection.
LMAO im laughing but crying on the inside.
MS Access ( and its 2GB file size limit it is)
I'm surprised Microsoft didn't just create a new code called "MSPython" that is _remarkably_ similar to... well, Python.
... PSA... PyScript for Applications...
The horror.
@@guyvleugels8507They succeeded with C# which is just a Microsoft version of Java
Or P# to make it clear it's a whole new, legally distinct language.
Oh no, why do that? Just come up with your own extensions to python that slowly become more and more incompatible with the original language and all other packages.
Elastic Python Runtime powered by AWS...
Excel and Python are two of the three "can be used for everything, but absolutely should not" technologies of the world. The third is JavaScript.
Never underestimate Atwood's law.
i would say javascript is the worst out of those 3.
@@vectoralphaSec I Agree.
Why?
I think someone is trying to make a kernel in javascript on a dare.
This is probably the worst thing to happen since the inception of excel. The number of businesses running off the back of an unstable excel spreadsheet is horrifying.
Seriously, i imagine all the incredibly convoluted macro powered excel files , and now i imagine the python powered equivalent files that will come out of this.
@@GeorgeMixalis*Spiderman pointing meme*
Once they realize that everything they do in Excel can be done better in Python+NumPy+Jupyter+Pandas+Matplotlib+all the rest of it, then Excel just becomes the shell you start to get to the other stuff.
Microsoft will likely make some money off the crowd who are too dumb to realize that they can dump the Excel part and use the rest of it ... for free.
As developers, it is our duty to create problems today so that we can get paid to find a solution tomorrow. All hail the Spaghetti Code monster.
Nah.... At least the problem will. Be resolved by a computer instead of a single guy that reads the excel 8 hours per day
“Most developers would never admit it, but that game changing new app they are building probably could just be an Excel spreadsheet”
Real
Building a backend with excel is probably a shitty idea tho, for which advantage ? It's probably slower to figure out what to do and come up with "hacks" than learning some basic SQL which takes a day at most
@@heroe1486 fun fact, there's an ODBC driver for Excel so you can run SQL queries on your spreadsheets...not that you should.
@@heroe1486 if the customer is using Excel as a backend AND frontend and it hasn't scaled out of control, then he doesn't either need SQL or your services for that matter. The minute the excel sheet gets corrupted or the data grows beyond the 65k lines, then you'll get the call
You only need one app for everything. A Spreadsheet.
I think a Jupyter notebook would be far more flexible: each cell can hold as much or as little code as you like, and can produce multiple associated items of rich output - graphics, sound, video, maths, formatted HTML, plus interactive widgets to control same - interspersed with cells containing narrative text so your notebook can be self-documenting. And you can include code in other languages besides Python.
I only understand 10% of this channels content but I'm entertained 100% of the time
Are you not a programmer?
inaccurate content often gain the most traffic
Same lol, I ain't a programmer 😂
Nope .. 3D artist @@LuisSierra42
@@Based-Pharaoh well then tell us what was inaccurate in this video, and do not say everything, I want precise examples
Wow, I can now mess up my pip dependencies directly in Excel! Amazing!
Not quite. It runs in an Azure cloud, meaning that whatever packages and versions will be available to you are only going to be the ones Microsoft approves.
And if it's anything like the ol' MRAN R language repository (that Microsoft just shut down), you'll be two years behind the latest releases.
Not really, and that's half of the point
@@adamestrada7610 Does Microsoft pay for that Azure instances or the customers pay ?
@@adamestrada7610Yeah, conda-main tends to run a bit behind conda-forge, which itself can be sometimes be a point release behind PyPI. But it's usually not quite "apt repository" level of staleness.
@@GSBarlev Totally, but if I know Microsoft as I do, they'll probably be running package versions at least six months behind because they'll need to ensure that those packages don't conflict with core Excel functionality. This is similar to Power BI where only Python 3.7.7 is currently supported, which was released all the way back in 2020
Being a Google Apps Script developer for more than a decade, knowing the incredible potential it has, and how it is mostly unknown to the masses, it was a happy surprise to see it mentioned in such good terms, thanks Jeff!
I've been looking for GAS jobs for a few months now, any suggestions where I can find some opportunities?
Good thing the masses don't use it, because like anything google, it can be discontinued at anytime, and eventually will.
My job has the option to use Apps Script and I was planning on giving it a go, but I didn't know there were people working only in it :0 I thought it was a minor thing for internal tools
Thank you, Microsoft. Because of this move, now we have a new video from Fireship
Perfect, just right when I finished my excel data analysis project at work writing thousands of python lines 💀
Same 🥲
Oof
Your code can run locally though.
Man they really had one of the biggest ideas for a Microsoft product ever - Python in excel - and implemented it in the worst, most crippling way possible - it has to run in the cloud.
Incredible
They want to make money from it.
They'll be using that to further train ai.
Exactly. Wtf. I was so excited for this, to pitch python to my non-programmer friend that manually processes a lot of data in excel sheets at work. But now i find out about this cloud shit .....😤
@@thisaintmyrealname1 same for me. Would’ve loved to tell some people I know about this. What’s the point of a tool that can’t be used locally. Cloud is supposed to provide infrastructure and management benefits, not act as a paywall to real features. Dumb.
@@HoneyBadgerLikesYouyeah like wtf
From what I remember, the third step is extinguishing any of the remaining bugs to improve the developer experience.
Well, extinguishing something. Not necessarily bugs, usually the competition.
This is a great feature that will be very beneficial to those Excel users that are comfortable with a little scripting. However, I shudder at the thought of pre-existing, _already over-bloated Excel files_ gaining hundreds of lines of Python code strewn about various cells across multiple sheets.
“I need a new laptop. Mines too slow at opening things…”
if i was still working for my shitty old boss, i would probably be about to automate their whole spreadsheet with python
Thanks for your code reports I always love your videos, been learning a ton from them for past few years #Respect
Obviously third step is "Examine", because after Microsoft's AIs examine the developer, it is easier to extingui- I mean, to empower them 💪
Quadratic also has been redesigning spreadsheets to be python native! And they run in your browser I think.
Quadratic is a web-based spreadsheet application that runs in the browser and as a native app (via Electron) [2]. It is an infinite canvas spreadsheet with built-in Python, SQL, and Formulas[1][5]. Quadratic is designed to enable users to pull data from its source (SaaS, Database, CSV, API, etc) and then work with that data using the most popular data science tools today (Python, Pandas, SQL, JS, Excel Formulas, etc) [2].
Quadratic has several features that make it unique, including:
- Multi-line formulas: In Quadratic, you can expand your formulas to as many lines as you need[1].
- Python library support: Quadratic has Python library support built-in, allowing users to bring the latest open-source tools directly to their spreadsheet[1].
- Powerful AI integration: With GPT 4.0 powered autocomplete for data, formulas, and code, users have a powerful spreadsheet companion[1].
- Infinite canvas: Unlike other spreadsheets, Quadratic has an infinite canvas (like Figma), allowing users to pinch and zoom to navigate large data sets, and everything renders smoothly at 60fps[3].
Here's a summary of some of the key features of Quadratic:
- Infinite data grid with Python, JavaScript, and SQL built-in[2].
- Data connectors to pull in data from various sources[2].
- Multi-line formulas and Python library support[1].
- Powerful AI integration with GPT 4.0 powered autocomplete[1].
- Infinite canvas for easy navigation of large data sets[3].
Quadratic is available for download from its website[1], as well as from GitHub[2] and the App Store[6].
Citations:
[1] www.quadratichq.com
[2] github.com/quadratichq/quadratic
[3] news.ycombinator.com/item?id=35456509
[4] sourceforge.net/projects/quadratic.mirror/
[5] www.producthunt.com/products/quadratic
[6] apps.apple.com/us/app/quadratic-master/id351899924
yes they run using Web Assembly
Never thought I'd see the day that Microsoft and Python would partner up, but just like with ChatGPT anything is possible these days.
They hired Guido Van Rossum a few years back, so this was inevitably going to happen.
Guido currently serve as a distinguished engineer @Microsoft
windows uses some python scripts too
@@oh-yes_10-fps15 Wait really? That sounds intriguing... I used to think it was purely coded in c.
nothing to do with ChatGPT Microsoft brough Guido like 2+ years ago to dev python they dev 3.11 which was a great step
Excel has Javascript support also. It is called Office Script.
And has cool features like all declarations secretly being globals for developer convenience.
@@Caellyan I believe we shouldn't blame the tool if we don't know where to use it and where not to use it.
I started my career with Excel, after few years, I realised that it can't handle some of my increased work load so I shifted that work to Python.
It's not really hard to understand where to use a tool if you know its strengths and limitations.
Remember, Microsoft is not going to offer you Python-in-Excel for free.
@@lawrencedoliveiro9104 lol they don't even offer Excel for free. And I highly doubt that unless you are on an enterprise plan, those cloud-based calls to a Python runtime are going to be throttled hard.
not as good as macros though, the implementation is crap.
E.g I cant access the file system to copy data from multiple spreadsheets, and create an automated data feed from hundreds of spreadsheets into one.
The fact that it all runs in the cloud makes it a complete non-starter at my company. Everything has to be on the company network to comply with data protection laws.
USA #1
I litterally used python in libreoffice 2 days ago and thought: "It's so cool! Why doesn't Microsoft have this feature?" Honestly python is so good for handling data imagine having to use the basic language
Why don’t you the Python code through a VBA macro and then import the results to Excel?
@@ajarivas72 vba macros are getting switched off all over the place recently because of the security risks associated with them. I love VBA, but yeah you can destroy someones computer with a macro.
I love your code reports for three reasons, but I forgot which is the third one
What are the first two?
For those wondering : the third option is Extinguish.
Embrace, Extend, Extinguish. EEE.
Grist is an open source alternative to excel that is also collaborative, and it has python support in the formulas :)
also..
Quadratic is a web-based spreadsheet application that runs in the browser and as a native app (via Electron) [2]. It is an infinite canvas spreadsheet with built-in Python, SQL, and Formulas[1][5]. Quadratic is designed to enable users to pull data from its source (SaaS, Database, CSV, API, etc) and then work with that data using the most popular data science tools today (Python, Pandas, SQL, JS, Excel Formulas, etc) [2].
Quadratic has several features that make it unique, including:
- Multi-line formulas: In Quadratic, you can expand your formulas to as many lines as you need[1].
- Python library support: Quadratic has Python library support built-in, allowing users to bring the latest open-source tools directly to their spreadsheet[1].
- Powerful AI integration: With GPT 4.0 powered autocomplete for data, formulas, and code, users have a powerful spreadsheet companion[1].
- Infinite canvas: Unlike other spreadsheets, Quadratic has an infinite canvas (like Figma), allowing users to pinch and zoom to navigate large data sets, and everything renders smoothly at 60fps[3].
Here's a summary of some of the key features of Quadratic:
- Infinite data grid with Python, JavaScript, and SQL built-in[2].
- Data connectors to pull in data from various sources[2].
- Multi-line formulas and Python library support[1].
- Powerful AI integration with GPT 4.0 powered autocomplete[1].
- Infinite canvas for easy navigation of large data sets[3].
Quadratic is available for download from its website[1], as well as from GitHub[2] and the App Store[6].
Citations:
[1] www.quadratichq.com
[2] github.com/quadratichq/quadratic
[3] news.ycombinator.com/item?id=35456509
[4] sourceforge.net/projects/quadratic.mirror/
[5] www.producthunt.com/products/quadratic
[6] apps.apple.com/us/app/quadratic-master/id351899924
Grist and Quadratic are both web-based spreadsheet applications that offer unique features. Here are some key differences between the two:
Grist:
- Offers a unique "visual data modeling" approach that allows users to create custom views of their data[1].
- Has a powerful formula editor that supports advanced functions like array formulas and nested functions[1].
- Offers a wide range of templates for different use cases, such as project management, inventory tracking, and budgeting[1].
- Has a user-friendly interface that makes it easy to get started with[1].
Quadratic:
- Offers built-in Python, SQL, and JavaScript support, as well as powerful AI integration with GPT 4.0 powered autocomplete[1].
- Has an infinite canvas that allows users to navigate large data sets with ease.
- Offers data connectors to pull in data from various sources.
- Has multi-line formulas and Python library support[1].
In terms of alternatives, both Grist and Quadratic have competitors such as Google Sheets, Microsoft Excel, and other spreadsheet applications[1]. However, each of these applications offers unique features that set them apart from one another. Ultimately, the choice between Grist and Quadratic (or any other spreadsheet application) will depend on the specific needs and preferences of the user.
Citations:
[1] stackshare.io/getgrist/alternatives
man this is such a huge win - I had to buy plugins back in the day to do this kind of work (7-8ish years ago), so its so nice to not need to anymore!
Now you can give money to Microsoft directly 🎉
@ I mean yeah - but if company paying/u already have office then nbd 😂
also see it as a win because ppl with excel might now get into coding more
While useful, this functionality involves pandas, not python
If you can't run it locally, it's useless
I find this incredibly awesome. Excel becoming more powerful and sophisticated is something with which
I am wholeheartedly pleased to see. Love it or hate it, you can't deny the significant influence and usefulness of Excel.
I was about to say, this would be a great replacement for my sql tables but nah, if it's not local then there's no point
I just love Excel and it's always fun to see what new is coming
Damn, that looks like all I ever wanted from a spreadsheet program.
Think of this as Python+Jupyter+Pandas+etc embracing and extending Excel. Guess what happens next ...
@@lawrencedoliveiro9104what?
@@Don-fw3nvyou heard the man, you have to guess
@@cinderwolf32 what if that was his guess? 'what?'
0:35 Guido Van Rossum, which is totally awesome. What a rhyming!
As a Python dev who uses excel regularly, this is Excel-lent news!
The fact that I had to learn the Microsoft VB Excel language just to make my own "function" like 3 weeks ago, to only find out this has happened. Gosh, I hate this year.
Quadratic looks very good, haven't tried yet though...
Quadratic is a web-based spreadsheet application that runs in the browser and as a native app (via Electron) [2]. It is an infinite canvas spreadsheet with built-in Python, SQL, and Formulas[1][5]. Quadratic is designed to enable users to pull data from its source (SaaS, Database, CSV, API, etc) and then work with that data using the most popular data science tools today (Python, Pandas, SQL, JS, Excel Formulas, etc) [2].
Quadratic has several features that make it unique, including:
- Multi-line formulas: In Quadratic, you can expand your formulas to as many lines as you need[1].
- Python library support: Quadratic has Python library support built-in, allowing users to bring the latest open-source tools directly to their spreadsheet[1].
- Powerful AI integration: With GPT 4.0 powered autocomplete for data, formulas, and code, users have a powerful spreadsheet companion[1].
- Infinite canvas: Unlike other spreadsheets, Quadratic has an infinite canvas (like Figma), allowing users to pinch and zoom to navigate large data sets, and everything renders smoothly at 60fps[3].
Here's a summary of some of the key features of Quadratic:
- Infinite data grid with Python, JavaScript, and SQL built-in[2].
- Data connectors to pull in data from various sources[2].
- Multi-line formulas and Python library support[1].
- Powerful AI integration with GPT 4.0 powered autocomplete[1].
- Infinite canvas for easy navigation of large data sets[3].
Quadratic is available for download from its website[1], as well as from GitHub[2] and the App Store[6].
Citations:
[1] www.quadratichq.com
[2] github.com/quadratichq/quadratic
[3] news.ycombinator.com/item?id=35456509
[4] sourceforge.net/projects/quadratic.mirror/
[5] www.producthunt.com/products/quadratic
[6] apps.apple.com/us/app/quadratic-master/id351899924
At first I thought you were copy pasting Wikipedia. But then... 👏🏻👏🏻👏🏻
@@MarcCastellsBallesta perplexityAI
Doesn't support JavaScript so it's kinda cringe
@@HUEHUEUHEPony yeah but coming soon
They gave us python in excel before native python support in Windows.
Azure runs Linux, don't be surprised.
They know where the revenue opportunity lies. And that is with -Office- 365, not Windows.
Excel now has a alternative to Visual Basic, That is good news.
That's not an alternative. VB can run locally, this has to run on cloud. Absolutely horrible idea.
You can also run C# in excel, you have been able to forever. And C# is a fine language
@@randomuser5237 Adding onto this, you also can't manipulate the components of Excel object tree or develop custom UI forms with Python, only data manipulation in data frames and visualization output.
As much as I'd love a full replacement for VBA, that language is like a cockroach: ugly, unwelcome, and will probably survive a nuclear apocalypse.
@@adamestrada7610lmao, bookmark comment later
Javascript would be the sensible alternative to vba if you think of the pieces that ms already has. JS is already designed to be a document manipulating language. You would just have to add more host objects to it to do what vba does. On that note, I'm not sure why people hate vb so much, but whatever.
Title just about gave me a heart attack... thought we had another Oracle/Sun moment. But hey, that engagement hittin good
It means they know which ones they should crack open first
Python OpenPyxl is also brilliant working with excel. A true saviour.
Awesome, now I can do machine learning while building my spreadsheet
I got scared as hell😂😂😂
Me too. 😅
Same 😅
I read "Microsoft acquired Python" 😢😂
He changed the title 😂😂😂😂
It goes Embrace - Extend - Excel.
noice
So much info in a 3 minutes video. Amazing bro 👏🏽👏🏽👏🏽
My biggest takeaway from this video has been that LibreOffice supports Python.
1:24. Of course. The recently deleted their delete and insert functions. For local use
I would have never thought that MS would make me interested in Office ever again.
I still think that.
Man, your videos are next level! Fantastic work!
I cannot stress to you how validated I feel hearing you praise Apps Script. It’s my daily rider at work and has advanced my career in very serious ways. I’d love to hear you talk about it more tbh.
It's very useful when you can't avoid dealing with spreadsheets, which is something that usually happens at work
Thanks, your comment clarified for me that he said Apps Script and not AMPscript. I was like 'what'
Using app scripts I automated my company HR onboarding process with no costs (besides initial time on development). It takes Google form, fills a Google docs template and saves it into pdf then sends it as an email. 😀 it generated possibly thousands and thousands of documents, it's very reliable and doesn't cost anything.
App Scripts ftw
@@x55554 that's pretty interesting, i too was looking for onboarding automation idea, can you share more detail on the flow?
The environment thing is extremely important. Its what makes it work.
My day when your video drops out. ❤🎉
The third step is Extinguish XD. Great refrence!
maleware developers gonna love this
I've been able to plug in Python commands to Excel before. It's great that MS have included it natively
I think this video should have mentioned that JavaScript is already supported by Excel and is to my understanding executed locally (either within the browser when using the web version or webview when using the native app).
The feature (called Office Add-Ins or Office.js) has additional benefits compared to the Python announcement:
- Available on Windows, Mac and Web (Python is only available on Windows thus far)
- Allows you to package and deploy your code as an add-in
The Python use-case really seems to be to do something in only one workbook, whereas the Office.js stuff is for implementing something you would like to distribute for people to use across workbooks.
If your company allows it. My companies group policy and sharepoint auth disables it in desktop and browser. I'll be interested in seeing how they will handle the python function execution. They may black list the api call in our proxy, or it may just go unnoticed. Making my life significantly easier since they blacklist pandas because of c++ build tools requirement.
They don't want people compiling code on their machines....
It is pretty cool when you have it
You can do programming with excel with PowerShell, not only visual basic.
The Excel championships are gonna be fire next time they happen 🔥🔥
I don't understand most of this stuff, but I still love your videos.
I was excited until I heard about the cloud thing...
Edit: But great video as always! ❤
Finally! I've been asking for this for years.
Python framework called Excellent coming soon.
If the runtime was on the host machine, I would imagine all the reverse shells from malicious python scripts.
Thats a solid point.
For that, they could have implemented a subset of Python that does not allow system commands with os, pathlib or shutil, or network functions
@@batmanatkinson1188 That could be the case, in case the users gave up functionallity.
@@batmanatkinson1188They cannot block networking modules like urllib and requests because data analysis often has remote dependencies.
Instead, they would be running this inside an isolated Kubernetes cluster and set application level permissions for the running process to prevent direct communication with its container.
And even if one container gets compromised, there's usually security in place to prevent access to other containers in the cluster.
(Or the devs are underpaid and didn't bother putting up all this, so I'll try to get the service account tokens 😂)
They deserve eachother.
This makes sense, seeing as Anaconda's speed when "solving environment" was inspired by Windows Update (and possibly Greenpeace).
I don’t use Anaconda. Is that a Windows thing?
Hence why mamba exists.
@@lawrencedoliveiro9104No, conda is an alternative to pip (really poetry) that is used for creating virtual environments and installing packages. Unlike venv, it's not just limited to Python libraries and can install binaries for other languages and even alternative versions of Python. The package solver was historically very slow (as opposed to pip's solver, which was historically VERY BAD) but it's gotten better lately. There's also a drop-in replacement called mamba that is super-fast.
@@lawrencedoliveiro9104No, it's a python bundle.
@@lawrencedoliveiro9104no, it’s a more robust alternative to pip.
Award-winning ending for those of us who happen to know the third step of Microsoft's typical strategy.
Is it buying out Python? Or making the Python usage subscription based?
@@BrainPermaDeD Extinguish.
Remember that Microsoft is doing this because it has to, not out of the goodness of its heart. The only reason it has to pay attention to Python is because it is dominating the professional-quality data-analytics industry, and leaving the noddy Excel jocks in the dust. This is a way to make those Excel jocks feel they are still somehow relevant, and also pay Microsoft for the privilege.
0:35 that was a good rhyme.
Your videos are great ❤
Thanks 🙏
I hope it works on the Android version!
(yes my workouts are so needlessly complex I need Python in Excel at the gym)
Welcome back! I hope you had a great summer ;)
just write a command shell wrapper to interact with python repl in visual basic and forgo this cloud trap
has someone done this? feels like something that someone might have done
@@Alticroo yeah ive done it before.
looks something like this
Public Function ShellRun(sCmd As String) As String
'Run a shell command, returning the output as a string
Dim oShell As Object
Set oShell = CreateObject("WScript.Shell") 'run command
Dim oExec As Object
Dim oOutput As Object
Set oExec = oShell.Exec(sCmd)
Set oOutput = o
Exec.StdOut
'handle the results as they are written to and read from the StdOut object Dim s As String Dim sLine As String While Not oOutput.AtEndOfStream sLine = oOutput.ReadLine If sLine "" Then s = s & sLine & vbCrLf Wend
ShellRun = s
End Function
Or try a Jupyter notebook. Each cell can have any number of lines of code, and produce any number of items of associated rich output-graphics, charts, sound, video, mathematical formulas, even arbitrary HTML. And interactive widgets to let the user control that cell output. You can intersperse code cells with narrative text cells, so the notebook becomes self-documenting.
And you can use languages besides Python, if you want.
Where is there a spreadsheet that comes close?
@@lawrencedoliveiro9104 yeah jupyternotebook is awesome and useful but thats beyond the scope of this context. we are talking about using excel because excel rocks. jupyternotebook is not better than excel, its useful but not as practical for the average user.
@@lawrencedoliveiro9104 I think both of these options miss the point - which is to have a complete spreadsheet + code environment in one package. Python isn't part of Excel, hence developing a project in both Excel and Python requires that any other contributors have both installed; which isn't always feasible. Being able to tap into python through Excel with just your regular run off the mill corporate Office package is a big deal for a variety of industries;
both approaches seem to come from a developer or data scientist / engineer POV where the having Python is basically a guarantee
Thank you Fireship for programming Python.
Not running the code locally is too ridiculous to describe. They could run it in a local container. Also VBscript can in practice do anything the user has permissions for anyway, and a warning pop-up is still good enough for VB-macros.
My guess is that they want to add value for azure services, and market the ability to run python code safely as a revolutionary advance in office work.
Problem is VB Macros are no longer allowed to run in most government departments, unless it is on a vm.
Personal experience on that one. Redeveloped a whole bunch of VBA Macros for a govt dept, only for their IT dept to switch it off as a security risk without telling anyone, and then I (a third party dev) get blamed for them not working....
I still have nightmares about debugging and maintaining other (non-programmers') excel VBA code.
Honest question: Why do people still use excel instead of a programming language like Python? Is it only because it is easier to learn and required for many jobs, or am I missing something extra that Excel has?
It is easier, and therefore, widely used.
Interacting directly with data and coloring cells is, though useless to a programmer, very useful to a true general public.
Because it is so easy to input and edit data and higher level functions are not really needed to a general audience, they don't look for something more like incorporating Python or SQL unless their spreadsheet can't open. And even then, they would just make another file called "database_2"
Microsoft did a hell of a good job to make simple programs for an audience that don't know much about computers and that work even if you dont do them properly. And now, they raised the ceeling.
Just my opinion based from experience
Because it’s higher level. Excel isn’t really optimized for big data workflows, nor should it be used for those. That’s why programmers that work with big data use other tools. But you want to outline a calendar, a todo list, or even record a simple set of records that stay on disc and you make small edits from time to time, then excel is great.
Ease of use, if you explain to people how they could use python to do the things they do in excel then they'll think you're cooked
Excel spreadsheets are notoriously hard to debug. What’s worse is that most of its userbase has no idea about debugging techniques anyway, since that requires programming knowledge, and they use Excel because they hate programming.
There have been many embarrassing cases of wrong data being published because of Excel spreadsheet bugs -- even in scientific research papers. When other researchers want to replicate those results, they ask for the original spreadsheets, only to discover that their formula ranges are incorrect, that kind of thing.
The most depressing case I heard of is that geneticists are changing the names of some genes, because when they import info into Excel, it wants to interpret those names as dates.
Why would you spend time coding something that can be done easily in Excel in seconds. Unless it's something that you do all the time and needs automating then there is no point. And don't forget that formulas are powerful and can accomplish a lot. Even if you used python you may still need a spreadsheet to present the results. Also big companies may not even allow you to install dev tools for security reasons. When excel really sucks is when you have to handle large datasets, it just can't and Access can't help either because it's garbage.
EEE was a secret code found during 2001 (?) Microsoft monopoly trail.
EEE stands for “embrace extend extinguish”
So you're saying we won't hear from analysts for about 6 months and then we will be flooded with broken python in Excel spreadsheets to replace? Thanks Microsoft!
This is neato, but I don't ever see most companies using it.
These days, every company I've worked for only uses excel for simple tools that go to sales, or financial packages for upper management. Visual Basic is plenty for these uses. Anything more complex is just done in SQL.
I cannot imagine a scenario where this would be preferred to SQL or even just regular python.
Where the hell have you been man, we thought AI have replaced you again, Fireship 2.0 😂😂
Spoiler:
He was replaced by AI a long time ago
@@LuisSierra42 The anticipation leading to unrolling this comment did not pay off.
In google drive you can code with apps script, a javascript clone that can interact with every type of goole drive file
What if you do a loop inside a loop though ? Do Microsoft servers crash ?
As a German it's curious to read these comments compared to those on German developer news sites. Almost everyone in the comments here ignores the obvious security risk that is "everything is computed in the cloud" while almost no German developer failed so hard at spotting this. What if your building's / company's / country's internet connection is lost? It's also a terrible business decision from a customer's perspective. You needlessly jump head-first into vendor lock-in. You're basically asking for a "subscription" pricing model that will cost you huge multiples of a buy-once product. For what benefit? Sucks to see how oblivious people in other countries are
I think also GDPR discourages a lot of 🇪🇺 developers and businesses from jumping wholesale into 🇺🇸-based cloud services.
TFW the two top data nerd tools combine forces 😮💨🙌🏼
You can even use the graphing feature for airfoil design and stress analysis.
Yes and you can calculate the angle of the dangle.
@@TheBendixSA
And run FEA
There is already something similar in excel with Javascript (Excelscript) but obly available to companies and student accounts (my case) if I'm not mistaken, its actually quite similar to apps script which I've also used in the past but limited in that you cant make formulas with it for example, only scripts to run on the sheets, I have one that creates a graph and formats it the way I need to 90% of the time with a trendline and all, saved me hours already but yeah python gonna be awesome, can't wait :)
Man at this point just use python, using it on excel just makes it more complicated
Right. Its not just that either, what these Tech tubers are describing is a unique excel function that was designed to appear as if it incorporates all of Python's features, when it only involves the utilization of the Pandas library
It’s for those Excel jocks who are too dumb to realize you can access all this stuff directly, outside Microsoft’s walled garden, for free.
Can’t believe I’m finally going to have to learn how to use excel.
This is a massive security issue.
You could write some quite sophisticated but simple and undetectable data exfiltration code and steal a whole database without anyone noticing....
For instance a malicious user connects database (via ODBC) to Excel locally, throws up a API endpoint using Flask and then steals the data dropping it into open S3 buckets from outside the organisation.
They could even create an active gateway where they query the database in real time and have Python parse the SQL queries.
There will be a generic hosting fork out soon enough avoiding Azure host and sidestepping security lock downs.
Weaponised Excel's will be chucked all over the internet soon enough...... as a VBA master the future I dreamed of in the 90's coming true brings a wry smile to my face
You could all ready make malicious wb with VBA. I think you will get the same warning when opening a wb with py to enable scripting
VBA just doesn't have the capabilities or 200k + library extensions so it can't do much
I wish it did but MS hobbled it in the late 90's in favour of .NET which really upset me at the time.
Lots of tricks you can deploy once you are in like spinning up encrypted VM's, replicating existing windows processes, side loading payloads, intermittent extract, creating protective supervisor.
Problem is the typical usage is to create a secure connection to a remote server for remote code exec which is very similar to what a malicious actor will want to do
I'm just gonna say this; weaponized excel sounds like the "Bioweapon" of coding
@@WhiteboardMario198 think of this, the obvious place to go next on this is to turn it into a sideloading, transmorphic worm that refactors it self every few moments.
Don't know what you want to steal or destroy?
Load a small NLP model in there to read everything it sees and customise action based on what it find.
No need for C&C, just find all the account number, credit card numbers etc let the worm decide on the fly tasking / triage and then plop out the goods in a single email attachment while the rest of the worm goes wild tripping every alarms on the network
@@nonomnismoriar9601 I'm having a hard time conceiving of how these same ends can't be accomplished using VBA and the libraries available.
awesome....now Excel will be fun to work with
We should be able to write a plugin so that python integration can be run locally offline or fully self hosted?
Right?!
MS: Hahaha, keep dreaming!
It is called XlWings, and has been around for a while.
"Write a plugin" When you say this, you need to be more specific in what you are suggesting. Write a plugin where? In excel? Why would you refer to it as a "plugin" and not a function?
Or, better still, get rid of the Microsoft part and run the Python+NumPy+Jupyter+Pandas+Matplotlib+etc part directly. For free.
step 3, developers volunteer to train msft's ml models with their data.
What we have learned: Python is one of the best programming language for AI and is evolving day by day!
this is going to be a game changer
Did not have that on my bingo card
now no one is gonna suspect of a script running through a excel file... i can now jump the security without making a specific C malware to move in the system, now i just open a spreadsheet in excel and start a script inside it, run it and jump over the security *cheff kiss*
As an amateur data analyst, this is the best day of my life and the worst day of my life because i spent so much in learning R stuff
1:52 This is probably the most I've ever learned from a meme. I'm currently using some of these, so I took a look at the rest and will def start using some more. This is brilliant :D
fun fact microsoft was having issues integrating co-pilot into excel. Hence, python. Co-pilot(GPT) will be integrated into Excel soon.
R in 100 seconds, || Python for haters || Python vs R for AI
panda, RStudio, tinn-r, ess/emacs, statet/eclipse ; microsoft is finally catching up.
waiting...
me too excited
Thanks for showing the @google way !
I think that adding Python to Macros instead of using VBA would be a much better approach. And running Python formulas in Azure makes no practical sense.
It’s a way to squeeze some more revenue out of those Excel jocks too dumb to realize that they can leave out the Excel part-indeed, the entire Microsoft part-and run a full professional-quality Python-based data-analytics stack for free.
That is gonna be so awesome! Imma gonna go nuts.
spreadsheets were the killer apps of the 80s. history repeating itself.
It never stopped being the killer app.
Not really as Excel have a GUI that makes it easier to use, unlinke Python which is just a bunch of code
@@jeffjiang5272 Well, if you are doing more complicated things with excel formulas or even VBS/VBA it can get quite complicated too, like python
Bloomberg Finance did a presentation at a Python conference a few years ago, where they demonstrated how they create quite amazing-looking custom dashboards for their users inside Jupyter notebooks. They even open-sourced the add-on toolkit they use to do it.