Seriously, how should a community based on short two- to three-paragraph answers react to question after question like this:

I am new to python. I would like to write a program which can collect information from multiple excel and pdf documents to output that in one single excel document to show similarities and differences between the documents . Is this possible ? If so, how and where would I start writing such a programme in python? Thanks

I haven’t tried anything yet

I mean, I’m glad that someone looks at that problem and thinks “programming could do this”, because it could, but it’s kind of a big task and getting someone from “I haven’t tried anything and am brand new to python” to that is beyond any question-and-answer forum. Welcome to programming, you may be able to get there, but it’s going to be a bit of a hike.

This hasn’t been my experience at all, but I’m old and have been using SO since it was new.

I have stopped visiting it to answer questions because the questions aren’t interesting anymore. They’re either “how to do this incredibly obscure thing in SOMELIBRARY” (where I’ve never heard of that library) or “why does my function exit early at the first return statement instead of continuing on” (basic “you misunderstand programming so fundamentally a single answer is unlikely to help” kind of questions)

As far as I can tell, the range of “I’ve tried this, and partially gotten it working, but this thing does FOO when it should do BAR” questions don’t show up, or at least it doesn’t show up when I open the site.

Answering basic questions again and again and again isn’t fun. It’s something I could be paid to do, I suppose, but I’m not paid for that.

You can always have a CI step that runs mypy on your code.

For your home office, see this article and then pass it along to friends. It just makes everything easier: https://www.theverge.com/23642073/best-printer-2023-brother-laser-wi-fi-its-fine

When the poster says “I often just use Photoscan on my phone” what they mean is “instead of dealing with the physical scanner at all I often use an app on my phone that takes a picture of a document with the phone’s camera, flattens out the image in software, and leaves me with a pdf the same as if the scanner were doing its job”.

That is, the scanner doesn’t “work on the phone”, but these days phone cameras and automatic image processing programs are so good actual scanners can often be bypassed.

Does any place still use HP laserjet II-series printers that gave that error? Those things were workhorses but it’s been 25-30 years.

Besides, this error isn’t that bad: it means “I’ve been asked to print something on letter-size paper, but I don’t have letter-sized paper. Please load some.” Either give it some paper of the size it wants, or check that the paper tray it has is properly set to “this is letter-size paper” and not to “this is A4-size paper”.

That’s the real cause of why this error was so common: US letter paper and A4 paper are almost—but not exactly—the same size, and there was a switch on the paper tray you had to set to tell it what it was loaded with, and the printer refused to print on the wrong-sized paper. In other words, PC LOAD LETTER is just another entry in the list of things that are frustrating and annoying because the US never adopted metric.

You want something annoying try PAPER JAM IN AREA 3, where you have to open the thing up carefully and hope that the paper doesn’t rip as you remove it also be very careful not to touch that one spot that is super hot and will burn you. This was usually caused by someone somewhere deciding to save some money and order everyone the cheap, crappy paper that did this all the time instead of the good stuff.

UPDATE: finally sat down and did the upside-down copy of Picasso’s portrait of Igor Stravinsky and the end result did indeed surprise me, although somehow in my version the chin is just gone and as I was doing it I had several moments of “wait these lines from over here are supposed to meet up with these lines from over there and they’re nowhere near each other—why did I follow the instructions and not do the outline first?”. But the end result looks much more like the original than I could have hoped.

Next the book is talking about what I drew as a kid, and I absolutely hated drawing and coloring as a kid so that’s going to be fun. (I would get classwork sent home from first grade to finish at home where what I’d been unable to finish in class was color in regions of a picture. Do the little math problems to find out what color to use in each region? Done instantly. Color? Noooo…)

I’m old (a few years shy of 50), and a second generation professional programmer. E.g.: when I was in kindergarten, my father’s main job was maintaining the COBOL compiler on a particular series of Sperry-Univac mainframes. I grew up in a house where the scratch paper for grocery lists was punchcards because my dad brought home reams of unused ones when they were being thrown out in the early 80s. (Fun fact: with a sharp pencil, it’s totally possible to fit a full D&D character sheet on the back of an unused punchcard)

So for “how did I get started”, I was born into it; I was of the age when you’d get magazines in the mail with code to type in (later, the magazines came with audio cassettes with programs on them). So BASIC initially, then in high school my dad got us a copy of Turbo Pascal and set me loose on that. (Plus tiny TSRs in x86 assembly)

I had a few mid- and upper-level programming classes in undergrad., but was a pure math major, not CS. (so didn’t get any CS theory classes, though I did have a job working for campus networking people) After grad school, I got a job writing code in java and perl for a company you’ve never heard of unless you were in a particular corner of the finance world in the late 90s/early 2000s. I’m now on my third or fourth employer, depending on whether you count a buyout that kept the team intact but moved offices as a change in employer.

My day-to-day coding these days is primarily in python and C++, but in the past six months it’s also included work haskell and go, not to mention sh scripts and the weird groovy dialect used in Jenkins.

Oddly enough, my hobbyist stuff these days has all been HTML+javascript because it just makes simple GUI demos so easy. It’s kind of wild coming from the mid-90s when I was heavily involved in early web stuff at my undergrad school to this new world where javascript mostly works and is a basically sensible language. Recent-ish projects have included a solver for the NYTimes “digits” game, a Mandlebrot set viewer and the “come back for more meeting” timer at breakmessage.com.