this post was submitted on 29 Aug 2024
305 points (93.4% liked)
Memes
45597 readers
1260 users here now
Rules:
- Be civil and nice.
- Try not to excessively repost, as a rule of thumb, wait at least 2 months to do it if you have to.
founded 5 years ago
MODERATORS
you are viewing a single comment's thread
view the rest of the comments
view the rest of the comments
You're entitled to your opinion but I would say Excel is one of the best, if not THE best spreadsheet application ever produced. It's one thing that Microsoft actually got mostly right and one of the only reasons I still pay for an Office 365 subscription.
If you're just creating simple spreadsheets, there's plenty of other options out there.
But, if you're a power user doing a lot of complex data analytics, Excel is still the king.
My main gripe is that I still have to use VBA for a lot of stuff behind the scenes. Yuck.
I also love word, publisher, and visio.
I grudgingly accept outlook.
I despise PowerPoint
Only if you refuse to learn SQL and do everything in a fraction of the time with way more functionality.
Jupyter gang, any other data scientists here?
That works when you have access to a SQL database instead of a bunch of massive CSV files.
Not saying there's any reason to switch, but I believe you can load CSV's into sqlite.
Datasette would be something that I would try for CSV's as well, that seems like an interesting piece of technology I haven't had reason to use yet.
Finally there's always Jupiter Notebook and any respectable DataFrame-solution.
Not to knock spreadsheet-solutions too much - I certainly see their value and use them frequently - but if I had to do something that warranted writing VBA, I'd probably reach for a tool I could combine with some form of VCS like Git at least.
Importing CSV into SQL is trivial and gives you far more control than Excel can.
cracks knuckles
Now you’re speaking my language!
Curious what you get out of value from the subscription that you wouldn't get from a slightly older non subscription version of excel?
Has XLOOKUP made its way into the non-subscription version? I've found it a lot easier and flexible to use than VLOOKUP.
Index/match crew checking in.
Excel is probably the most powerful application ever created. Especially now that it can be connected to SQL and python natively
There's a tool for a given job. And for some, Excel may be decent. But for many, MS Access used to be far superior to Excel. Not sure if it is still maintained well, as I no longer use MicroShit
If you're going to do complex data analysis, isn't it a pain to use ANY spreadsheet software, no matter how good? I do mine as a Jupyter notebook. The spreadsheet is just for looking at the numbers, maybe sorting some things.
Stop making me feel like my Excel knowledge is useless😔
Excel is definitely not useless! Learning a little Python (especially the pandas package) can go a long way in making data analysis easier though
I’m not a power user, so I’m often frustrated by Excel trying to do things I don’t want it to and by its abundance of features that I’ll never use.
And at least at my workplace, a lot of work processes use poorly-designed Excel spreadsheets for critical tasks, because it’s such a simple way to manipulate data.
I also find that when I need to do more complicated data analysis, Excel starts to become limited, and I find Python to be a more powerful and flexible tool.
Capability is a double edged sword. Any tool that is capable of doing something is going to be used by someone to do that thing, regardless of whether it should be. Excel gets abused and used for things that it shouldn't be frequently in corporate environments because of its capabilities. I can understand being frustrated by that.
I use Excel for reporting and analytics because it makes manipulating and visualizing data very easy. Especially if you know what you're doing. No need to write a UI or worry about portability between workstations, etc. At the end of the day it's a tool. A very capable one. Like any tool, it's not the right one for every job.
I use excel because its stupidly easy to output a shitload of objects with properties (computers/hosts in my case) to a CSV via powershell and sort through the data.
ohh I see you can also use some ACL type of application. Excel is amazing but can't handle databasis, it has a very small limit
Dude just ask any advanced LLM how to do what you need to do in excel. The info is out there and they've digested in a million times.
Or, don't. And let your inabilities to do things consume you emotionally. Idgaf.
Edit: Lemmy's hate-boner for AI is dumbfounding. I get that it's a hype train and shoved down your throats but if you ignore that it will get you results way faster than googling ever did you're basically "old". And I say this pushing 40.
Downvote away, but I challenge anyone to give me a db and a desired subset that I can't produce by simply querying a good LLM.
Lol. Don't bother asking Chat GPT for help. You will get so many completely wrong answers. At least the answers will be formatted nicely. Complete bullshit. But easily readable bullshit.
Every time I’ve asked ChatGPT for help coding, I’ve wound up needing to rewrite it all for myself. LLMs make baffling design decisions (because they are just paraphrasing Stack Overflow, not making actual decisions).
I have found them helpful for turning error messages into more legible explanations of what went wrong, but AI-generated code has not been effective, in my experience