"...I just need my OS to keep working for me. This bug could hit me without even owning a scanner at all – someone sending me a PDF that I then unknowingly break before archiving it. That’s the part I’m mad about."
And secondly, that hypothetical could well be using broken software too. We really don’t know how bad the PDF’s ABBY produces are, but as someone who owned this scanner and the software, it’s fairly obvious that the software is barely maintained.
It really isn’t Apple’s fault if someone else is producing bad files that they happened to previously tolerate, especially if that somebody isn’t maintaining their software.
But by this logic, no software other than those that Apple sells you directly (their own or through the MAS) is "supported software". If Apple makes a change and all of them break rendering it useless of anyone using software not bought from them, would you still be making this same apology?
ABBY hasn’t updated their software for Big Sur, i.e. ABBY themselves say it hasn’t been updated yet and is not supported by them on Big Sur.
That’s what unsupported means. It has nothing to do with whether Apple supports it.
I open it with preview. It opens fine. I close it and open it again. It opens fine.
I open it a third time and make a small arrow pointing to something and save it (as I would with ANY OTHER PDF).
It breaks.
I'm not using ABBYY. I've never heard of it. It's not on my system. It's just a PDF file that I got sent.
Now what?
Nothing excuses ABBYY if their PDFs are corrupt.
The PDF works fine before saving in Preview. As in, Preview itself will render the file perfectly fine with OCR. By all accounts, the PDF is completely valid and uncorrupted at this point. Making a change to the file and saving it in Preview is what causes the corruption.
So how the hell could you possibly excuse Preview and call this a ABBYY issue when Preview is the one that causes the issue?
That's absurd to the highest level.
Is it? I don’t see any accounts showing this to be a fact.
I think the disagreement here is that there's no evidence of this. Preview's error handling could very well be interpreting bad data and allowing the file to be opened. The question becomes, then, should Preview continue to propagate that bad data on save or should it try to correct it with the possibility that it corrupts just that data. If the PDF was not in-spec prior to Preview touching it but it is in-spec after Preview saves it, is it a good thing that Preview "fixed" the PDF file and made it "proper" or is it bad because it technically lost/corrupted data?
In other words, what is the "right" thing for a software to do in this case? Keep bad data and leave the file as "bad" or fix the issue to make a valid PDF and, as a side effect, remove the "bad" data?
They shake it up a lot
They hand me said bottle of coke
I open it, and get sprayed by fizz
THAT IS CLEARLY THE COCA-COLA COMPANIES FAULT!
I think it is. Preview used to correctly handle the thing, and now it doesn't. Users cannot be expected to inspect the raw PS code of their PDFs to determine standards compliance. What they can do is assume that something that works yesterday, and appears to work today, is in fact working.
The PDFs didn't change, the application did. It's the application's fault.
If you have compatible behavior that you remove, it's on you to alert the user of the reduced functionality.
But, there is no evidence yet that they removed anything.
It’s quite possible (I’d say likely) that ABBY’s PDFs were corrupt all along.
Preview may have tolerated them not because of a special feature which was then removed, but because it happened not to depend on correctness before.
In this case, Apple wouldn’t have been in a position to even know about the problem.
You know who would? ABBY, who have had free access to Big Sur for months. You’d think that even cursory testing at their end would have shown this problem.
They have had months to fix it, months to work with Apple on the issue, any months to publicize it it really is Apple’s fault.
Even if ABBY does have a bug and fixes it then people will still have a bunch of older documents they will want to read.
This is in principle impossible.
If it turns out that there is a widespread problem with many kinds of document that no longer work properly, I’ll be more inclined to say Apple is at fault.
But as it stands, it just seems like ABBY has been producing bad PDFs.
The PDF that FineReader made worked properly. The one that Preview made didn't.
It's not clear if the issue is caused by the PDF files produced by ABBYY, or if the issue is also present in other OCR'ed PDF files.
In either case, Preview shouldn't silently corrupt the file. It could display an error saying that it was unable to save the file, or a warning that some data couldn't be saved, so that the user can check if the file still works for them after saving.
It’s not clear what’s causing the problem, but verifying correctness of programs is not a solved computer science problem.
PostScript was a stack machine with included operators which implemented a set of vector graphics commands. PDF has a similar rendering pipeline, but it renders a set of primitives described directly in the document -- there is no virtual machine involved.
Having said that, PDF is vastly more complex than it was originally. It can contain forms, JavaScript, audio, video, accessibility metadata, etc.
It’s an extremely complex format.
Without further information this is more likely to be an Apple bug because they are doing the harder job.
Plus some further information:
ABBY has had months to support Big Sur but still haven’t.
That makes it seem like they either don’t care about supporting MacOS users, or they have a very low engineering budget.
Either way, that makes it seem more likely to me that they have a bug, than Apple.
But, regardless of the balance of probabilities, the point is that the original piece blames Apple entirely.
And yet nobody here really seems to dispute the idea that ABBY, a professional PDF generator could be producing a buggy file.