[glenstein]

Pocket still saves articles offline. It just saves them in article view. Which is the same as what Instaaper does. I just checked my phone and it has 1.07 gigabytes of articles saved for offline reading.

I don't recall it ever saving web pages in web view (though maybe you used to be able to see them offline when you used the premium plan, but now the premium plan doesn't save the WebView offline? I'm not interested enough in paying for it to find out so all I have is guestimation.).

So they do save, but they save as article view and not the webpage itself. Which is the same as the other services.

[AshamedCaptain]

You definitely were able to save the web view at some point during the "Pocket" timeframe. I explicitly remember even the "notification with progress" that showed up when it was doing that (since I would have to wait for it to finish before turning off the radios). Heck, the Mozilla docs still mention Pocket as downloading the web view, e.g. https://support.mozilla.org/en-US/kb/getting-started-pocket-...

> Offline Downloading: Pocket will decide the best view to download by default. If you want to specify whether to download Article View, Web View, or both, uncheck Download Best View.

The article view was a nice novelty 10 years ago, but as it is completely based on heuristics (readerJS-like, in fact I think it _is_ readerJS ) and they have not really bothered upgrading it, it is no longer a reliable way to capture web sites for offline reading .

Frankly, I want this https://en.wikipedia.org/wiki/Plucker https://web.archive.org/web/20080903165121/http://plkr.org/ back :)

[glenstein]

I stand corrected! I will note that at least from my use case a saved copy of web view is a rather idiosyncratic user need not catered to by I think any of the other apps. As you do get article view and article view is what I actually want.

But you have reminded me, because I do recall that phrasing that you quoted about pocket deciding whether to choose Article or Web View.

I don't it's a particularly huge loss, and the exasperation from the commenter I was replying to seemed to suggest that they thought saving of any articles had been removed, which would be an appropriate cause for exasperation. But I don't think that reaction is warranted at all, just from not saving web view.

I also think it's not quite right to suggest that saved article view "doesn't work" or is a mere novelty. I just went through a bunch of my articles and randomly spot-checked them, and all of the ones saved to article view are without problems. Although I do recognize there's an occasional issue, but those are the exception rather than the rule. It remains a critical antidote to the unviewable overloaded with ads viewing experience that is the default experience for most people with most web pages. There's also an open secret that it bypasses login screens and pay walls, fir which it continues to be one of the best solutions.

But alas, I was not correct to suggest this had never previously been available.

[mananaysiempre]

Here are the ten most recent pages on my list, with [A] for article view working correctly, [B] for article view being broken, [W] for article view (and thus offline access) being completely unavailable:

- [B] https://spotbugs.readthedocs.io/en/latest/running.html (just completely mangled, starts with “-sortByClass=filepath:”, that is, skipping around half of the actual page)

- [W] https://web.itu.edu.tr/~dalyanda/mssecrets/hotmail.html

- [A] https://web.itu.edu.tr/~dalyanda/mssecrets/

- [B] https://learn.microsoft.com/en-us/windows/win32/fileio/types... (the “Note” inserts with important caveats are gone)

- [B] https://jimbojones.livejournal.com/23143.html (table at the end is gone so the sarcasm in the last sentence does not come through)

- [B] https://advgamer.blogspot.com/2018/07/kings-quest-vi-cliffs-... (screenshot after “At one point in this section it says...” is gone, leaving only the joke in the caption)

- [B] https://mtnphil.wordpress.com/2016/04/09/decompiling-sci-byt... (“Here are links to the rest [of the blogpost series]: [empty space]”.)

- [W] https://old.reddit.com/r/gaming/comments/ucdigm/taken_me_yea... (I guess this one’s fair, I only saved it so I could later copy it to my quotes stash)

- [B] https://josephmate.github.io/2025-02-26-3200p-cpu-util/ (“and it looked like this: [code snippet missing]”)

- [B] https://docs.rs/crossbeam/0.8.0/crossbeam/epoch/index.html (the actual reference, “Structs” / “Traits” / “Functions”, is cut off)

[glenstein]

The same test on my saved articles would show [A] for just about everything.

I don't think Pocket ever built itself with intended use cases being such things as looking at tables of server logs showing IP addresses (jimbojones) or Microsoft support pages, and I believe those are probably exactly the instances where switching to a web view exists to account for that.

If your deal-breaking use case is that you need an offline only archived web view of Microsoft support for a note about how "OpLock" means the same as "opportunistic lock" I think they're well within their rights to say that that's outside of their intended use cases, and that's what Web View (built into pocket) is for.

Certainly room for improvement, but they're just very idiosyncratic use cases. Do you get more success with those same pages on InstaPaper?

[mananaysiempre]

> a note about how "OpLock" means the same as "opportunistic lock"

This one is a bit silly, I admit, but consider this one:

> Note: Your application should not perform any file system operations on the file between [calling CreateFile2] and [requesting an oplock]. Doing so may cause deadlocks.