Why Companies Don’t Fix Bugs

It’s not just about bugs. It’s about caring for small things that seemingly don’t directly impact the bottom line. From a corporate strategy point of view, there’s not much difference between not fixing a bug in the Music app, deciding the iPhone mini shouldn’t be manufactured because only a few million people are buying it (vs. tens of millions), or not updating QuickTime to support modern video formats.

Ah, a post directly relevant to my pet peeve.

I seem to recall people predicting this was coming even back in the 90s (maybe even before then) when we really started just pumping out MBAs and taking a Gordon Gecko mentality to absolutely every aspect of business.

It's also possible that those who have been around long enough were simply spoiled by a situation that couldn't continue to exist. When the Internet was young the people and companies using it and making software for it were doing it mostly because they loved it and it directly benefited them to help build it.

Now many of those successful companies have become incumbents or absorbed into the corporate machine and the entire situation has flipped.

To a degree, Google and Apple no longer have to care if the users like what they do. What are you gonna do, not use it? Same thing with Microsoft and what they keep doing to Windows. Nobody likes it, everyone hates it, the user hostility has gotten to the point of almost being blatant. But what are you gonna do, not use that Windows software that there is no replacement for?

It's not even about not fixing bugs. Now they are actively introducing hostile features like the strict enforcement of Microsoft accounts in Windows 11, patching out workarounds. It's almost laughable to think that they would fix bugs, their priorities are absolutely at the other end of the spectrum, far more focused on controlling what the users do in order to continue to funnel them into sales.

Not to go on and on but even just today I had to deal with people being forced into "New Outlook" which broke their profile, but Microsoft doesn't actually care that the user's mail works. A popup comes up saying that New Outlook WILL be enabled on next startup, with the choices to either do it now or wait until restart. And since the "New outlook ON/OFF" toggle is still set to OFF, because the user never wanted it on, there's nothing they can do about it.

And so the timebomb goes off and it's New Outlook they get. For now they haven't removed the toggle, but they have gotten more and more aggressive. It replaces the shortcut for the Office 365 version, so even when that version is clicked on, New Outlook launches anyway. Launching it directly from the Program folder works, so they are hijacking the shortcut.

These are the kinds of things Microsoft, just for one specific example, spends its time on now. Fixing bugs, unless they are in the way of management engagement metrics, is simply not a consideration anymore. They don't have time. They have to hurry up and implement the next user hostile trick the psychology department has come up with to get .1% more users on Edge this month.

I build furniture by hand for a hobby, and my first instinct was to point out that I intentionally do not finish the backs of cabinets: I leave roughsawn boards because this is a hallmark of antique, handmade furniture of a certain vintage. Hopefully my notional grandchildren will see them someday and realize they're descended from a madman who resawed 12" wide boards by hand.

The point, though, is the same: I'm trying to craft a bookcase that doesn't sag and lasts a long time, using traditional joinery where I allow for seasonal wood movement, and I make mouldings to cover the end grain and give it a more finished appearance. This is that "back of the cabinet" mentality that Apple used to have, where the end product was carefully fit together, did its job, and looked like it had been designed and crafted for a purpose.

Present-day Apple has a bunch of people who polish their parts of the system, but there's no cohesion or overall thought, and all those shiny buttons look like a polished heap of rabbit dung at the end of the day. I think Jobs could enforce standards within Apple, as he mostly had good taste (except for those damn iCal editable labels) and limited tolerance for usability problems. For 3rd parties, Interface Builder made it easy to get to 90% of the HIG, and users would complain and quote chapter and verse of the HIG until you finished the other 10% ("yes, thank you for pointing out that the label at the bottom of BibDesk's editor windows is off by 2 points").

I'm guessing iPod and iTunes worked pretty well because Jobs loved music and had a collection he'd curated over decades, as I do, on physical media, and wanted access to it all the time. Marketdroids who have have never ripped a CD or used iTunes as a portable analogue of their home music collection will just see it as another place to put advertising, rather than something with intrinsic value on its own.

As an indie developer, one of the best things I did was to make the feedback and support email accessible to users and come directly to my inbox to handle.

Seeing multiple bugs or queries on the same issue is annoying, but also a great motivator to fix those issues – both in order to improve the product quality, but also to reduce the load of having to constantly handle the same questions or issues.

I can pretty much judge the quality of the product by the volume of email. As time has gone on, and issues have been fixed, the volume has dropped dramatically.

I think in larger companies this is where the bug fixing downfall occurs. It's easy to be separate and put up a firewall between the constant and pressing issues that your customers are experiencing and what you prioritise as an organisation to do everyday.

@Bart: it is interesting that they are pushing "New Outlook" like this, as it is obviously way, way worse than the established version. The new one is, as all apps are, just a shell to their web interface; missing essential features (try dragging an attachment out of an email to your desktop for example).

I'm probably missing the situational awareness that is necessary to understand why such decisions and the one for a MS account are being made - it doesn't feel like such changes are in the users interest.

I had a similar experience working more than 2 decades for another large Silicon Valley tech firm. When I started it was like Google. In one of my interviews I was told that if what I did was good for the customer, good for my team or good for the company, in that order, I wouldn't be punished if I failed. That shifted as the company grow and the MBAs moved in. Now it's simply down to juggling the numbers to look good for Wall Street every quarter. They haven't come out with a new product in years and just keep milking their cash cow while laying off left right and centre to goose that bottom line. I got out before they hit the iceberg that is coming.

As to the other thread about developers being sour, when I started at the same company in the 90s their code base was a huge monolithic blob where fixing a bug in one feature sometimes broke something somewhere else. I remember filing the same bug report 3 times because of the fix/break scenario. The final time I ran dtrace, found the offending line in the source and attached it to the bug report with the fix because I (and the customers) were tired of the constant break/fix repeat scenario.

I discussed this with a VP in software development and suggested an approach similar to what I saw with IBM. Rework out code base to make it modular and patchable so that if something broke we could issue a patch for that specific bug without having to upgrade the entire image. His response was that as much as that is an excellent engineering solution it isn't sexy enough to get the required funding because it involves no new features, no new products, nothing other than more stable software. But if I could prove that such an effort would lead to increased revenue he'd try to sell it. That's the mentality in the C-suites, new features = more revenue.

As for Apple, they've been trying to make macos into IOS for years now, something I predicted years ago. Forget that it's a different use case and lock it down. Catalyst anyone? The net result is that if I didn't didn't have a bunch of Mac-only apps I'd have moved on by now. That and unfortunately Linux always promises a great user experience but rarely if ever delivers.

I wonder what one can expect from a company whose first tier support operators are obviously drilled to urge customers (I was about writing "users") to update for whatever issue they're reporting, to follow insane "diagnostic" procedures, to deny the very possibility a huge bug (known to Apple) is at stake. My two calls to Apple Support over last years were both of this kind. One operator was so kind and enthusiast that I hadn't the heart to let him down and, as we were sharing screen, I proceeded to trash right away all launchagents and daemons in main Library (including Little Snitch's, Microsoft Office's etc) that were certainly responsible for my Mac issue. And I screwed everything up of my own free will. I read on the web of a mythical second tier support operators (engineers?) who most likely are at least more proficient at telling lies.

@michael: the “Tim” link seems to be pointing to the wrong page - it’s the same page as the previous link. Can you fix that?

@Vadim It’s a comment on the same blog post.