Yup

It is a horrible and unusable business model/architecture.

1) Requiring subscription to access when it is not technically required is onerous and very questionable value, especially for tools that I'll use intermittently.

2) Requiring online access instead of local & native apps makes the false assumption that I'm always connected. Often my most productive times are on 6-hour flights (& no, the onboard wifi is not reliable) or away at locations where there is no connectivity.

3) Performance will always suck compared to a fully local app - all the optimizations a developer could do to make it run acceptably could also be applied to a local app to make it really snappy

4) Any SAAS carries significant extra risk that does not exist local native app - that the software /service could disappear any time due to a host of business reasons. Sure, native app companies can also go out of biz or discontinue products, but I still have the software and can upgrade on MY schedule.

5) MOST CRITICALLY, entire classes of very attractive customers are LEGALLY locked out of using your apps. In my industry, there is an entire new class of information called CUI — Confidential Unclassified Information — anything related to DOD projects that isn't quite classified, but there are very strong restrictions on what can be exposed in what way. My wife's industry is legal, and they have almost nothing that could be put on a SAAS, even billing information. And of course same with medical.

Of course there are some apps where the online SAAS is close to essential, like real-time collaboration / conferencing.

However anything else, and it looks strongly like the business model is extractive and rent-seeking, instead of providing value. (and this also applies even to IOT devices - I should be able to access those via the IP, not only through your service...)

We're building a similar tool that's self-hosted only (https://github.com/centerofci/mathesar).

I'm glad there are more tools to make working with data easier, but I don't use any of them because putting my data into a proprietary service I have to keep paying for is a dealbreaker.

Subscription based SaaS doesn't bother me in and of itself. Cloud based supporting multiple collaborators and multiple devices, tech support often built in, no managing upgrades or patches, etc.

The problem is portability and lock-in. Which itself might be due to lack of standardization.

Also that it's pretty hard to make good spreadsheet software I guess ? (Even Excel has bugs sometimes, not to mention various quirks for backward compatibility...)

AFAIK FramaSoft decided to stop working on FramaCalc (based on / hosting EtherCalc, based on SocialCalc), because it was just too hard for a small association like that (which already has to deal with managing lots of other services for free), but they seem to have decided to postpone the shutdown, I guess because of COVID ?

https://framacalc.org/abc/en/

(For instance it supports exporting ODS but not importing it.)

My reactive workspace tool aka spreadsheet replacement (Inflex) is SaaS, but I’ve also considered offering a completely downloadable tool for a one off price. I think if ever get a substantial user base and people ask for that, it’s on the table.

Ironically, O365 is $7/month with max rows = 1m+ and spreadsheet.com is $9month for max rows = 10k.

As a greybeard, I have seen plenty of old ways become new again. It's hard to predict when the next shift will be or where it will go, but it will shift.