PostgreSQL -> MySQL?! I just.. I don’t.. oh no.
Uber wrote a pretty good article about why they switched from Postgres to MySQL. https://eng.uber.com/mysql-migration/
However this is a and extremely specific case and overall I tend to agree with you.
This is a service that I do not understand who would actually use it and pay for it. But I like the idea and hope they do well.
The first use-case that jumps to mind for me is something like RRDtool or WhisperDB or something, where this is used for automated report generation. The boxes or services that you’re monitoring POST periodically their data to a spreadsheet URL, and then whenever you want, you go view the spreadsheet and have all your data in a nice, easily-exportable format.
For example, a cron job on my web server posts periodically about requested URLs and the response code (200, 404, etc), or requests or timestamps or something, and that goes in a spreadsheet that my Pointy Haired Boss can immediately use in Office or whatever he’s familiar with.
…honestly, that’s the best use-case I can think of. They should put that in their marketing materials.
EDIT: Except I’m not seeing a URL to hit that just gives you the spreadsheet in CSV format. That’s no good. That eliminates my use case entirely, because I was thinking I could tell my boss to go to “http://example.com/spreadsheet/report/web-usage” and it would pop-up a download for a CSV file that would automatically open in Office, but I don’t think that they do that.
IMHO, they should add that immediately.
Right now, CSV export is only possible with the “export” link on the dashboard. The CSV download through the API is ready in the dev branch and should be shipped around the end of next week.
Hi! Author here.
SpreadsheetDB aims to offer tools to build products that allow non-developers
to query a database and make statistics.
Right now it is a little “raw”, but features will be added gradually. Some
exciting updates are already on the way. :)
Just an observation. Your marketing seems directed at developers (the terminal animation on the main page, for example). However, I wonder if you could have some success with a more semi-technical market.
For example, my SO used to work as a data analyst / statistician. The organization where she worked had a large, Oracle-backed ERP system, but everyone hated using it because they didn’t know how to query it and the GUI tools were terrible. What they really wanted was spreadsheets. So they made spreadsheets and then periodically (sometimes even once a year) synced the spreadsheets with the ERP.
This used to frustrate my SO to no end because she actually learned how to create queries, which then returned outdated data, which then forced her to go figure out who had the spreadsheet with the “real” data she needed.
If they’d had a spreadsheet frontend into the ERP system I bet they would have used it. Your product seems like it could provide something like that, at least for certain use-cases.
So they made spreadsheets and then periodically (sometimes even once a year) synced the spreadsheets with the ERP
I’ve seen pretty much the same thing at a number of organisations. The “real data” lives outside the system and is periodically synchronised (eg, for things like payroll runs after annual bonus calculations).
I’m surprised that the spreadsheet metaphor was never more popular for ERP system access. It’s not uncommon to see with analytics products that slice and dice data, but not really something that ever took off in other areas. And by spreadsheet I don’t just mean a grid with some input fields, I mean the whole shebang - formulae included.
I have a feeling the use case would be more obvious with Excel (or LO Calc) integration.
Yes, I agree with that. I know of some organizations that basically keep all of their company in a multi-GB excel file and it’s terrible. If that were just in a DB, it’d be nice. But at the same time, they need a product with some more safety (backups that they can test, for example)