+1
iWork lost a lot of features in 2013 when Apple rewrote the apps to make their features equal on Mac/iPad/iPhone. I don’t know if all the missing features were ever replaced. It was 9 years before Mail merge returned.
1 Like
A new update is out. I have been having issues with syncing between iPhone and iPad. I hope it legitimately fixes those issues and how slow and messed up (showing the same episodes a bunch of times before fixing itself after a delay) the Recent Episodes playlist has been.
There is if there is a complete requirements document developed prior to the rewrite, specifying what is, and is not, in scope. Of course, there is then the matter of implementation. But at least there is a definition of what is expected.
I also wonder who was doing the testing? There is a reason to have testing done by someone other (or in addition to) the programmer. How many of these issues were known prior to release?
Good release documentation, describing what is and is not yet included in the release, along with a summary of known issues, would have helped.
Finally, rushing code out to meet an arbitrary deadline is rarely a wise course of action.
(The above is of course a simplification, and the development process can be much messier.)
This is true, although I was trying to speak to how the early reviewers missed things. From the end-user perspective there’s the assumption that all previous features will be there until they experience that they’re not.
1 Like
This has been fixed in the latest version. If you had play from the top of the playlist selected after finishing an episode selected, it stopped as you describe.
1 Like
Marco is a developer who famously doesn’t write or perform any form of tests post development.
1 Like
Right. I wouldn’t expect Marco to have caught it. I was trying to speak to why his friends/reviewers/etc. wouldn’t have caught the glitches.
He said on ATP (I think, it could have been Under the Radar) that the Beta groups was kept very small, which is a curious thing to do for a major rewrite
1 Like
Marco also said (prior to release) that he had received almost no feedback from the small beta group, which he interpreted as meaning there were no/few issues. He explained a few different times that he was using the new app as his primary podcast player for 6 months before release. It would seem that that only tested it the way he uses it. He later admitted that some of the biggest bugs only occur with the various config options set in a combination that he never uses personally. Perhaps the silence from his beta testers was because the small group of his closest friends are all inclined to use the app in similar ways. If so, a larger test group definitely would have been beneficial.
The issue with deleting listened podcasts is still unresolved in the latest release. I’m fine with Marco removing streaming as an option, having listened to both podcasts where he explained the reasoning behind it, but I feel the management of downloads should then have been adequately thought out before releasing this.
Now, it’s basically either automatic or nothing, with no controls on what automatic means. Deleting old podcasts is apparently not working, and there’s no storage status screen anymore—quite a few regressions! Basically, he wants Overcast to have carte blanche for how it manages storage.
@waylan Apparently, he also has no intention of processing all the feedback he received via email and on Mastodon (source: Under the Radar podcast). I get it that it’s a flood of feedback now and that he’s a solo developer, but given how he’s released this update in a half-working/half-broken state with limited beta-testing, I would think there is some valuable information there that either he or someone else has to process (and not ‘sample just the 1%’ as suggested on the podcast).
1 Like
It is resolved on my end, but there is an asterisk attached to that notion:
And this is exactly what I am experiencing: those episodes I have listened to since installing the update were deleted instantly when completed. Those episodes I have listened to previous to the update’s installation have to be deleted manually.
Apparently the deletion is being triggered by setting something like a label in the moment when an episode is completed - either immediately or with a 24 hour timestamp or manually, depending on the setting.
I have no idea,
- why there is no way to select all or multiple episodes of any given view and then to choose an action like delete or mark as played; this is a feature that has been missing in Overcast forever (bigger picture),
and
- why he did not just run an SQLite query like
SELECT * FROM whatever_tablename_is_being_used_in_Overcast WHERE status='completed' AND show_deletion_setting='when_completed'followed by the deletion of any file that is showing up as result of the query (cleaning up the mess of last week’s update) - done!
1 Like
Especially as Marco’s included an announcement feature in the app for the first time now.
I can understand why Marco wouldn’t want to delete any podcast episodes as a one off. It’s a destructive action with the potential to go wrong, and people likely won’t have listened to too many podcasts in the week this has been available.
As for multi select to delete, I can’t imagine a scenario where this would likely be needed by many people often. For myself the only reason I can think I’d want to delete multiple episodes are twofold:
- I want to delete a whole podcast. Deleting the podcast would achieve this rather than particular episodes
- there’s some form of server issue and suddenly a lot of old podcasts are downloaded accidentally.
4 Likes
As a sole proprietor of a small business, I’m willing to cut Marco a fair amount of slack for a buggy release. Based on the ATP episode:
- He felt like he might have to quit development entirely and needed to prove he could still do it by learning a new framework.
- He had learned a lot in 10 years and just wanted a fresh start. (I don’t begrudge that.)
- He’s always viewed his app as something he made mostly for his own taste, and I get the impression that he’s glad other people share that taste, but he’s still making it for himself.
- From my perspective as a designer/developer with a music side gig, the guy views part of what he does as artistic, and he’s not wrong. It’s craft. He thinks he’s gotten better at it.
So yeah, it’s probably unlikely he wins an ADA for this. But what are his friends supposed to say? Publicly, on a podcast, they should berate him and say “this new app has unforgivable bugs and you suck as a designer and you’ll never win an ADA”? That doesn’t make any sense. John has a legendary reputation for being frank, but he’s not an a**hole.
If a friend is in Marco’s mental state — one where he feels like he lacks confidence, his life is in shambles with a huge transition (moving is real hard, guys, and he’s been living in a half-finished house for like half a year, which, if you’ve never done it, takes a real toll on your mental health), and he hasn’t developed any serious updates for his main project in something like 4 years.
If I were in any of his friends’ places, I’d be like “Yeah man, awesome! You shipped something!” Which is what they are all saying.
I get the impression Marco knows there are bugs. It sounds like John told him prior to him shipping. I get the impression from Under the Radar that David told him about bugs prior to shipping. But sometimes, you need to ship so it feels like you’ve accomplished something. In my own life, I’ve watched as client projects have ballooned in scope and size. Work that used to take 3-6 months now takes, through nobody’s fault, 6 months - 4 years to complete. That’s just how it goes on big projects.
And when they’re done, everybody says, “This isn’t perfect, but this shippable, and for the sake of our own morale, we need to ship this.”
And that’s even more true when you’re one person working in your house with no other working professional around to tell you it’ll be okay when you are in your deepest, darkest thoughts. Eventually, all you see is what you haven’t done, and what you haven’t done becomes what you can’t do.
So I’m glad he shipped it. He’s good at his job; he’ll fix the bugs.
I’m not saying the app is perfect; there are a lot of UX things I dislike and several bugs that have affected me as well. But it’s a podcast app. It isn’t the critical infrastructure powering 911 lines. It’ll be fine.
11 Likes
I agree in some ways, and I don’t in others.
I develop back-end code for websites, and yes, at some point you need to ship. It’s great that Marco got something out the door.
The issue for me is how that was done. The new Overcast release was basically a complete surprise to the users, and it replaced software that was completely functional with software that was broken in a number of very notable ways. There was no large, public Testflight until AFTER he pushed a broken app, at which point he Testflight-ed the fixes - which exacerbated the problems for end-users by prolonging the issues.
I agree that it’s not running 911 lines. The risk isn’t that of a catastrophic failure. The risk is that people like me get annoyed enough that they stop subscribing or switch to a different app.
For somebody that considers myself a podcast “power user,” who has been using Overcast since day 1, and a subscriber since it was offered, things were disruptive enough for me that I’m actually using a different app now - and seriously considering whether I should cancel my Overcast subscription.
7 Likes
Fortunately, I have yet to upgrade. And I’ll give Marco some time to sort it out.
But I find it [astonishing | disappointing | saddening | arrogant] (not sure of the right word here) that programmers think it is ok to use their paying customers as testers. Not cool. Not cool at all.
Either learn how and test your code yourself or hire a professional tester to test for you. And no, just testing the happy path is not enough.
I understand, and know people, with mental health issues. But good friends are supportive and honest. Do you think all the negative feedback is good for his wellbeing?
And while this is a first world problem, software development best practices are known. This is a solved problem. I’m not accepting any of the excuses given.
3 Likes
I draw a distinction between two kinds of testing.
There’s the kind Marco was (I guess until a week-ish ago) famous for not doing - unit testing. I personally don’t care if somebody writes unit tests. It’s extra code, and for me that’s a personal choice.
And then there’s the kind that’s affecting the public perception of this release - offering the software for testing to other users. For small devs, a Testflight might be overkill if they’re slightly adjusting the position of some UI elements or something. But for a complete rewrite, where there are likely major bugs, this is how you catch the vast majority of the day-one bugs so they can be fixed before a general release.
But for me, it’s not just that. That, by itself, probably wouldn’t have done it in for me. Let me back up for a second.
I listen to podcasts off and on all day. I rely on them for some of my professional education, daily inspiration, news, etc. I literally begin my day with a long dog walk listening to about 8 short-ish morning episodes. Some of this is actual podcasts (audio Bible, news, etc.), some is chunked content that I construct feeds for on Backblaze B2 using drip content from my Mac (audiobook segments, etc.). But it’s a critical part of my morning, and getting my brain going for the day. And that all is part of my developed mental health regimen. Routines and consistency are pretty important, and throwing that for a loop really, really, really sucks for me.
So now we’re at “what’s reasonable to expect here.”
Marco’s commitment to customers is, “I’m still a one-person operation, with no funding or external ownership, serving only my customers.” Even though he famously rarely replies to feedback, I believe he’s delivered a fantastic app up until this previous week-ish. But the initial release of this redesign, IMHO, broke that commitment. He’s on ATP talking about trying to win Apple Design Awards and anniversaries and celebrating, while the app that he rushed out to meet an arbitrary anniversary date is broken in major ways.
And here’s where his famous lack of communication came back to bite him. How many of the problems in the new release were design choices, and how many were bugs? I may have missed it on ATP, but I definitely wasn’t seeing any “mea culpa” on Mastodon or his blog. There were things that were completely unworkable in the version I had, and there were rumors of a Testflight that might fix some of them - but nothing official from Marco.
That’s the point where I started looking at whether it might be worth re-examining the field of podcast apps. Which sucks, but Marco is big on things like data portability, so I can always export my data. Except…I can’t. As the cherry on top, OPML export in the app was missing on day one, and as far as I can tell it’s still nonexistent. So you can’t even get your data out of the broken app.
That’s the point where I not only was looking at other apps, but I began seriously considering not coming back. Were it not for the Overcast web interface - which I know he kind of treats as a relic - I wouldn’t have even been able to export my data. And for me, that means switching apps would have involved re-entering about 80 podcasts.
Fortunately, I figured out how to get my OPML out using his web interface, dumped it into Pocket Casts, and I’m playing with the features over there.
Now…first-world problems? Yes, absolutely. Will things be fine in the long term? Very likely. But as @MevetS basically says, it was all reasonably preventable.
So I’m not really massively upset. I’m more a combo of frustrated and disappointed. I feel like trust has been broken, but I don’t think Marco sees it that way - so I’m not holding out hopes of that being fixed.
Either way, after half an hour of goofing around with settings I think I’m mostly “back to normal” with PocketCasts. Whether I come back to Overcast is an open question at this point, but it’s looking less and less likely.
10 Likes
I’ve spent quite a lot of time looking at alternatives to Overcast: podcasts are important to me.
It has been more than irritating when Marco has broken the app (this is not the first time), not kept up with development, not explained anything about the changes, so you have to go and figure it all out again for yourself, finding bugs within seconds of trying any new release (even the latest) and always having “reasons” why things are not where they should be - for a long time he seemed to blame Apple for all this, but is doing that much less these days.
But, and it’s a big but, the alternatives are so much worse. Whole podcast ecosystems and apps that are designed to monetise your attention and convey even more advertising while selling your data. Apps that just sound bad. Apps that aren’t developed for years. Apps that disappear suddenly. Apps that move from owner to owner at random. Apps you can’t trust. Above all, so very few apps that reliably sync across devices or that do what you need them to do quietly and immediately.
I used the latest Overcast while walking for hours yesterday, with a podcast list and playlists I’ve more or less had for years. I didn’t have to think about or look at it once: no adjusting so I could hear the voices, a touch on my watch to skip the ads or boring bits, and I get this for peanuts. I wish Marco wasn’t like Marco in some ways, but I am very glad that he is in others. I’ll wait for improvements and keep hoping he hires someone else to do his customer relations and communications.
6 Likes
Only bummer is that I haven’t been sure if new app behaviors are features or bugs, lol.
3 Likes
I’ve looked at alternatives, as I’m not in the mood to beta-test Overcast for months until the issues are ironed out (if that ever happens, which – given Marco’s attitude, it probably won’t). Pocket Casts is now owned by Automattic, one of the few companies with a demonstrably good track record in developing the open web and insisting on interoperability.
1 Like
Respectfully disagree. I prefer Pocket Casts and Castro over Overcast. I also think that Marco has developed this semi-famous dev personality that people in the Mac communities seem drawn to for some reason. I mean, I listen to ATP and respect his work, but I’m not downloading an app just because he’s behind it. I’ve used Overcast quite a bit and I grant you, it sounds a little better than even the best sounding alternatives, but the workflow, design – it just doesn’t do it for me, and I’m not downloading it just because “oooh, it’s Marco” 
or because half the podcast hosts I listen to are friends with the guy.
Not saying that’s why people on here have downloaded it or you specifically @chrisecurtis, but that’s definitely a thing.
3 Likes