This is a real 'Stallman moment'. The issue there is less about updates, than about non-Free software (the thing the FSF talk about), and data silos (the thing the IndieWeb folks talk about).

You committed to a non-Free software solution, the maintainers of which held all the power over the software, denying you the freedom to have the software do as you wish. When they make a decision, it's to maximise their benefit, not yours. (They might also do so incompetently.) When they make a decision against your interests, unfortunately you have little recourse.

The standard story in the computer folklore was Movable Type [0], a blogging platform. In the early 2000s when blogosphere was at its height, Movable Type was used by numerous authors. It was freeware, source code available, and had a big collection of plugins and extensions. It was not Free and Open Source, with restrictions on use and redistribution, but most thought it was good enough.

As time went by, its shortcomings and limitations became obvious and a new version was needed. But the development has already stagnated because the publisher's interests shifted. And finally when the new version was released, it contained numerous bugs, and many bloggers suddenly found they needed to pay several hundreds dollars of licensing fee under the new license.

Matthew Thomas, a blogosphere personality, wrote an outline for anyone who wants to create an ultimate blogging platform [2] that can stand the test of time and prevent similar incidents from happening again. Requirements included free database backend, open archive formats, and free software.

His article was also the source of a famous remark,

> In the long run, the utility of all non-Free software approaches zero.

---

[0] https://www.gwern.net/docs/technology/2004-pilgrim-freedom0....

[1] http://web.archive.org/web/20051107103030/http://mpt.phrasew...

[2] The Movable Type Incident was a major event responsible for the rise of WordPress, which was a new project at that time. Even the name "WordPress" refers to the printing press, in analogue to Movable Type.

"In the long run, the utility of all non-Free software approaches zero."

But to be fair, that is allmost true of any software, with changing archtectures programming languages etc. but a thousands times yes! With free software you have at least the option of doing the work of migrating.

Well, depends on what you mean by "long term".

BSD-licensed low-level unix programs are still quite useful today and are likely to continue to be very useful in coming decades.

And I am sure there are still comercial modules around, from the 70s or earlier, as well..

If they're BSD-licensed, that means they're Free Software.

Nobody said it isn't. You misunderstood the discussion here.

I quoted "In the long run, the utility of all non-Free software approaches zero."

hutzlibu said the same problem affects all software, not just non-Free software. In a constantly changing computing environment, even free software is affected. Nevertheless, hutzlibu was in agreement with the "spirit" of the quote, because at least the user still controls the software, and it's possible to migrate.

einpoklum disagreed, using BSD utilities as an example - showing how free software can be unaffected.

Disclaimer: I didn't downvote the parent post - Why is my parent post downvoted? There's really no disagreement here.

Right, gotcha.

> Why is my parent post downvoted? There's really no disagreement here.

I appreciate it, but even if I'd disagreed, that's no reason to downvote.

Honestly, a blog should just generate itself from plaintext, like Markdown, so that you can always keep the data, and even put it in version control and diff it easily.

Having committed to free software to solve a few problems, I'd say it's not all that rosy either.

I wouldn't say I have much more control over the open source software I use. I spent more time fighting desktop Linux than I did fighting Mac OS. I spent hours setting up self-hosted software that's more often than not just passable. Now I control the data, but I'm also responsible for it. I have to worry about updates, backups, migrations and so on.

If I measure how much effort I put into maintaining open source systems vs how many times I've been screwed by commercial software, I'd say both sides are worth considering.

You have ultimate control over free software, but the costs of maintenance become yours. Take some of the money you're saving from not purchasing proprietary software licenses and pay whoever you want to manage the complexity for you.

I will concede that in the context of personal-use software this is less applicable than, say, in a business context, but it's still applicable.

To preface this, I'm writing this in Firefox on my Ubuntu laptop with Emacs running in the background. I am onboard for free software.

But I also cannot let perfect be the enemy of good. There are closed source products I use simply because nothing else does the job and the alternative is the job not getting done. I do not know the solution to this problem, but I want there to be an answer to anti-consumer practices in software design (such as not allowing users to downgrade). I realize this creates all sorts of other issues: security, competitive disadvantage for existing products, difficulty complying with new regulation, etc. so perhaps open source is the only true option, but that locks out a lot of users.

But the economics don't seem to work out for a free-as-in-freedom model for consumer apps. Chiding people for using proprietary software when it's very, very often the only viable option feels very out of touch. I'd like to have a free accounting solution that covers all my needs, is easy to use, well-maintained, available on my mobile device, up-to-date, fully internationalized, but there doesn't seem to be one, but there are numerous affordable proprietary options. Which I totally get – I wouldn't want to build high-quality, well-maintained, fully localized, etc. accounting apps for free, either.

The modern problem of having your data locked up is a much bigger one than the old arguments about Free Software. Data portability would be a good solution to a lot of the problems with user-hostile technologies.

The trouble is, given the vast range of things we do with technology, it may be unreasonable or impractical to create a universal obligation for software developers to provide export facilities that allow transfer of the data in a standard format. For example, what if there is no standard format, perhaps because your software genuinely does something new or in a different way?

Going after hardware developers who try to contain data within their walled gardens isn't a bad idea, though. For example, the GDPR imposes certain requirements on data controllers to provide for data portability, and imposing similar measures on the makers of devices could both solve immediate problems for some users and provide for much more competition in the marketplace.

Is it so hard to just give the user a sql dump of all of their data? (or a json blob if it’s some nosql thing.) If it’s not in a db then just give them the files. None of that seems complicated.

Surely this will depend on what your software does and the data it works with.

Huge amounts of software, particularly embedded software, doesn't have any sort of single structured database or a simple set of data files that you could just dump. Even where you do have these things, the dump won't be worth much unless the DB schema or file structure can be successfully interpreted in order to reconstruct the meaning of the underlying data within some other system.

That might or might not be easy, depending on how complicated the original data formats were. If it's something common and widely understood, perhaps some text a user wrote or a photo they saved, that's one thing. But what about something more specialised? Suppose you're moving to a new home. How do you export the training data from the AI in your old smart front door lock, which now recognises your friends and family's faces, so you can set up the smart locks on your new home to recognise the same people?

Unless there are industry standards for representing whatever data you're working with, providing portability between applications or devices that offer similar functionality from different manufacturers/developers is always going to be a problem. Encouraging standardisation is certainly beneficial here, but imposing some sort of blanket obligation on the manufacturers/developers to provide import/export functionality in the absence of any relevant format probably isn't.

Yeah, for embedded devices I think most people would be happy with an eeprom/flash image.

I'd imagine it could be hard if a company folds. No money to pay to keep servers running. No money to pay employees who know SQL to keep working.

Possibly, you could mandate that all software include a working self service data retrieval feature and fine the trust if they shut it off before a certain point in bankruptcy, but that seems heavy- handed and still won't work if there's no money in the accounts.

This is why the export mechanism should be available before hand.

Navionics had a variety of bathymetric charts in the app store that underwent a similar fate when they were bought by Garmin. It didn't matter that I'd paid $20-$30 (forget the exact value) for a lifetime license on a piece of software; they felt the need to shove a crippling update onto the version I'd paid for before releasing a new product line.

On the bright side, in the years since I bought their app a free, better competitor was released, but I still don't think that absolves them of their responsibility.

Off-topic, but what is the Navionics free, better competitor? Thanks.

Polaris Navigation GPS.

It was hyperbolic of me to say that they're "better" in every way, but they cover all my personal marine use cases and several others, so I was excited when I found their app.

> Then sadly the author died, and it languished for a while until it was sold to someone else. The new owner seemed to try to keep it going, but then finally gave up last year and crammed an absolutely awful “update” that was actually a totally different software via the automatic update process. Everyone’s financial data got locked up in this new version that is almost useless. ~20 years of financial tracking just gone. To make it worse, they did it in December, just before people would need to pull reports for tax season.

It's usually better to design your (critical) data around formats that can be shared across a multitude of applications than on the bet than a single app will forever exist. Of course, it's a trade-off between convenience and flexibility.

Is there an open format for bookkeeping data? I suppose you have ledger-cli but that's hardly useful for normal people.

CSV file(s)? It's not like the underlying data is in any way complicated. A lot of software can export and import these.

Those that can't export to CSV have usually another way of getting that data into a spreadsheet application, from where you can export it as CSV...

And yes I know CSV technically sucks (varying encoding, lax standards, etc.), but it's the lowest common denominator for a lot of software.

SqLite is the solution to all our CSV problems.

There is OFX format: https://en.m.wikipedia.org/wiki/Open_Financial_Exchange

Or as someone posted here on HN a while ago - sqlite?

Too far down the stack. It may be the OP's data was already in sqlite but that doesn't help even if the next application uses sqlite because the schema will be different.

Gnu cash

Not even that is risk-free. I was doxxed by a developer after Google's idiotic real-name policy put me in his crosshairs for my one-star review.

I had an extension installed that would pop up a PayPal donation page whenever it updated, (I think it may have been a JSON pretty printer? Don't entirely remember). Whatever, it was a good extension and I can understand that.

One week it updated basically every day, and so every morning at work I was greeted with a PayPal donation page for this extension author. After the 4th update I uninstalled it and left a 1-star review saying an otherwise good extension had now gotten far too spammy.

The developer actually emailed me asking me to take down the review, and tried making the case that it wasn't spam, they just had a reason to update it so many times.

I updated my review to say the extension author is emailing anyone leaving a one star review.

Tip/suggestion: Make a separate Google account with fake name/suername for reviews on Google.

I have a few Google accounts. Only one is my actual name. Especially for games and reviews I use an aliased account (just like I use an alias here). Google can still track me I guess (IP) but a single dev will never find who UnrealName UnrealSurname is. And I do tend to leave brutally honest and detailed feedbacks whether positive or negative.

Let us hope this little trick will not get all of your accounts suspended without recourse.

You can't create a Google account these days without a phone number, and you can't reuse phone numbers. Worse, in my country, you need to register to get a phone number (anonymous /unregistered sims were outlawed a few years ago).

I was just able to create a new google account with no phone number.

Interesting. Must be a geographic restriction.

You don't need a phone number. But it is not obvious how you can get an account without one.

You can get digital disposable phone numbers online

Google/Facebook block those.

This is understandable, though it then creates a new challenge: problems that have always existed with hearsay and rumour are greatly exaggerated by the speed and reach of modern technology, and now we have reviews that come from unverifiable sources, which can cause real damage to a person or business, who in turn may have no way to defend themselves even if the review is malicious or inaccurate.

I'm aware that I will sound condescending but... that's exactly why people "invented" opensource (and/or free software if you prefer). I'm sorry for your problem nonetheless.

You mean opensource has the same problem with updates, but you can always fork (which helps indeed).

Or stick to the same old version. Or know how your data is written and read and create a parser for it to migrate it to another program, even if the original author dies.

Except on iOS devices, you can't really install anything you built without Apple having a say. And I'm not quite sure jailbreak is still a thing.

App stores and SaaS do an end run around open source.

Isn't that a false dichotomy? The package repos from Linux distros are basically open-source app stores (in the full sense, only those that support payment are).

With regards to SaaS, nothing prevents open source SaaS software that can be set up for multiple small to middle-sized-user bases. One could argue that things like phpBB is exactly that. They might even connect to each other to form a larger network... I'm still hoping for a major breakthrough in this area.

You may be right. You can not even usually go back to an older version of an application if you want undo some bad decision by the developer. Source code is spotty if you can get at it. Your data is locked in by them on some server you do not control. In some cases they do not even really control it.

I would add the other anti pattern companies use is opensource vendor lock in. They take what would basically be a 'meh' product and put good stuff ontop of it. But to get that 'good stuff' you have to buy their service contract. Sure you are using some open source. But the key bits are still proprietary.

SaaS does an end run around GPL in particular. Unless it is Affero GPL.

And to be frank, a lot of it at the application level is mediocre.

Yes and no. At least in this case the OP could have installed a different version of the app (older or customized), and kept the 20 years of data usable.

The app I used was "gas cubby" recommended by a friend to keep track of my car. You could record your fillups and compute gas mileage and lots of other statistics. You could keep track of the vin, the insurance policy number, part numbers and service intervals.

I slowly put all my data into it and...

The app was sold, and one day they changed it and uploaded all my data to the cloud.

I deleted the app (and apparently all my data). what jerks.

There are very convoluted ways to backup and restore apps and data on iOS, even to go back to an older version of an app, but they require a lot of fiddling (literally, with fiddler), and if the app has changed its data format you need to have a backup of the old data in a format that can be restored. It would be nice if apple added time machine to iOS and iPadOS and let you restore any previous version of an app and its data.

This example demonstrates that the walled garden is not as safe as people think it is. Granted, reviewing every update in detail is a monumental task. Then again, this is why apple takes the 30%. I don't believe apple can catch all bad updates before they get sent out. An easy ability to revert to an older version of an app and its data would be very useful. Given how an iCloud-powered time machine for iOS would slurp up iCloud storage and push people to higher paid tiers I'm amazed they haven't done it before.

Not only does Apple not care, they themselves have pushed updates to iTunes to remove this functionality…

If it's an option for you, it is possible to downgrade most apps if you jailbreak your phone and install a downgrade tweak.

Some apps are growing wise to holdout users, and force you to update to continue using services. That's ultimately been the death knell in every jailbreak effort I try, some essential app will eventually force me off the jailbroken iOS version.

That’s not necessarily out of malice, though. Quite possibly the older version is using some server endpoint or format that the developer has deprecated and they want to reduce their maintenance burden.

> That’s not necessarily out of malice

It is not malice, it's laziness and stupidity. Most apps are so trivial it is absolutely not necessary to break compatibility when you release a new version.

Depending on the app, it's just sound security. On a jailbroken device, you cannot assume any of the OS safeguards are still in place. On a stock iOS device, there's pretty much no relevant malware that could be on that device. On a jailbroken device, all bets are off.

I find myself in kind of the same case as you. I use a personal finance app on Android that I chose because it wasn't connected, so my data stays on the phone if I don't manually save it elsewhere.

2 years ago I looked at the SqLite db with the project to build a desktop app that would use this db, but I didn't commit to it.

Last week I wanted to finally build the app but I discovered that the developer had encrypted the db with AES256 in one of the previous updates.

I plan on contacting him to see if I can convince him to give me a way to read my data but I'm afraid that if he won't do that or I would potentially have the key to all his users db.

If it ain't broken, don't fix it... Updates on phones: I had been blessed (?) with having a few iPhones in my life. After getting burned from a crappy update I changed my tactics. I have switched off the "automatic updates". I manually updated on iphone5 and if the update was good I'd proceed to update

When I switched to android I do the same with a phone and a tablet. I only update 'blindly' my e-banking apps. Everything else (e.g. I frequently used CamScanner) does not get updated. Ever

For the PC, I use WFC and I block all apps that don't meet internet connection to work. VLC? Excel? Your use may be different of course. You may need the autonatic translation features of Word. I don't. Outlook only speaks to my email server, not MS. winRAR? Why? V5 is good enough. And so on..

I too have been burned by Devs removing features from the "current" version and pushing them to a paid one, or removing them altogether because of maintenance,shifting them to other products they run, and many many other reasons.

Steve Gibson says: if you didn't go looking for it, so not install it. My XYZ app works just find on its current version. Do I want an update? No thank you.

A tax software could be different though, tax code changes every now and then and you can't escape the updates.

I'll check if I am able to extract the data.

It’s in an SQLite file, and I just happen to be the type of person who made backups of the actual files every few months. I’m sure I could probably figure it out, but I’m just annoyed that something that just worked is now a big project.