one of our partners we have to integrate with at work sends us reports in ms access format. it’s not fun, especially when everything is running in lambda and there doesn’t seem to be any good libraries for reading ms access files that would easily run in lambda.
I can personally attest to entire universities advocating for student use of powerpoint for all sorts of printshop work to include thesis and capstone presentation posters for conferences :)
For people who don’t want to spend time learning yet another single-purpose application, it works quite well
If it’s just your own little tool and you don’t intend to share it with others: do whatever you want. SQL or NoSQL or JSON, it doesn’t matter. Use your own judgement.
In my experience tho most homegrown JSON-based “databases” tend to load all data into the memory, simply because they are very simplistic (serialize everything into JSON and write to disk, deserialize everything into a struct). If your dataset is too big for that, just go straight for a full-fledged database.
Microsoft spent years and years trying to get people to not use Excel as a database, until they eventually had to give up hope that anyone who doesn’t know the difference would voluntarily use Access, so they started adding database-like functionality to Excel to meet their customer’s demands and try to make the experience at least a little bit less painful.
This is a real-life case of “meet the user where they are” despite the designer’s wishes, because even within Microsoft, there is strong agreement on not using Excel as a DB.
I actually ran this setup for a pretty long while without major issues. YMMV but OneDrive is not a terrible way to store a single user database backend if you don’t have a lot of sequential writes going into it in a short timespan.
Yes, but at the time Excel didn’t support concurrency either ;-)
Anyway, you are correct about the issue with concurrent writes, but that’s only because Access was intended as a single user DB. If you wanted a multi-user DB you should be getting MS SQL server.
Not saying this product strategy worked (it clearly didn’t, otherwise people would not be using Excel), but that’s how they envisioned it to work.
i…isn’t that the entire point of excel? what is it for if not to store data?
similarly i remember a reddit comment that broke my brain, saying no one should be using excel, they should be using a ‘cell matrix organizer’ or similar. we all can name 5 off the top of our heads
Storing data is only one of the parts to the formula of what makes a database. Proper databases require structured storage of the data and some way to query the data constructively. Excel did not have those features until Microsoft gave up trying to convince people to not use it as a DB and added it to Excel.
Excel is this weird mix of storing small amounts of data but so good for visualizing data. If people are saying it shouldn’t be used to store data they mean massive amounts of data as opposed to something like some small scale accounting for a fund raiser.
Excel has a purpose, but storing data long term isn’t it. It’s for calculating data. It shouldn’t be the single source of truth.
One of the things Microsoft did to make it work was extending the row limit from 65k to 1M. Apparently, Economics professors were very excited about that one, which explains a lot.
Honestly, since the introduction of ‘tables’, pivot tables, Power Pivot and Power Query, Excel is way more viable to be used as a database. Tables in particular mean that formulas fill down and the range automically resizes when records are added.
Japanese companies, this isn’t a wish, it’s a fundamental truth of the universe. Like gravity. No matter the scale or importance of them. I promise you your car exists because of an Excel 2003 file on some underpaid engineer’s laptop that they periodically sync with an inventory system.
I once worked for a Japanese company where I had to make a presentation. In Excel, I shit you not. Then we had one of our “shadow” managers make a Japanese translation of the same thing. It took me two days to get the kerning and print layout right, especially with that weird english typeface that is Japanese standard, I hate to think how long the translator took to get their version right.
I used to make a monthly document that was in English and Japanese. I used either Meiryo or Meiryo UI. That looked ok in both scripts but there is another font where the en looks shit. Or maybe I’m thinking of full width characters.
I imagine an alternate 4th panel wherein the genie says “ok you can bring back dead people.” What do yall think? Also I bet we could come up with a themed genie or setting that would punch up the joke too. ♡♡ love it BTW, op.
In my experience it doesn’t work well when you have more than a couple people editing the file. My company had a group of ten modifying the same file in live time; it led to huge desync problems.
I live in Excel hell and even that made me shudder. Just work on separate files and have a master spreadsheet append everything with power query.
I made a similar reply higher up and I fucking hate that that’s a solution but it legitimately would work in this use case. I frequently deal with 1M+ row data sets and our API can only export like 20k rows at a time so I have a script make the pulls into a folder and I just PQ to append the whole fucking folder into one data set. You don’t even have to load the table at that point, you can pull as-is from the data model to BI or make a pivot or whatever else you’re trying to do with that much data.
Parent company doesn’t want ANYONE to have direct read access to the database - only the scant few heavily formatted reports the user-facing software will allow. Data analysis still needs to get done though, so…
Yeah. PQ -> Data Model saves my ass and my co-workers think I’m a wizard.
That, and learning how to quietly exploit minor vulnerabilities in the software to get raw tables I “shouldn’t” have and telling not one soul has been a winning combo!
Add comment