This topic is somewhat more nuanced than "one-time always"; while buying something you own forever is almost always better, there are cases where SaaS can make sense.

One specific example: when you are in "early lifetime" of a business or venture, and you need to spin up services without a lot of overhead. For example, running and maintaining a corporate email system is complex and expensive. Yes, you will probably need to do this eventually, but when you are small, outsourcing it to a cloud vendor is cheaper and more expedient. This is the case for lots of high-overhead business systems, and can be a valid option even for larger organizations (eg: cloud system resiliency).

However, for most things, if you can buy it once, it's probably better to do so. As noted elsewhere in this thread, SaaS products have an incentive to ship buggy versions, remove functionality, nickle and dime over time, etc. This is also somewhat true for any product which features mandatory "updates" (see: Tesla removing some autopilot capability, or Microsoft adding adds to the start menu in Windows 10). If you can buy something in a manner which does not allow the manufacturer to later "alter the deal" without your consent, its probably better to do so, even if you ultimately pay more.

Honestly, it's more interesting to discuss why this topic keeps getting brought up than the actual topic itself. Re the original topic, it's pretty much a dead horse at this point: Linux is objectively bad for people who want something which just works (especially for the last 10% cases). That's the main issue, it's always been the main issue, it hasn't gotten any better over the last 15 years, it's unlikely to ever get better, end of story.

So why does this horse keep getting dug up and beaten again every few years? Are the hardcore Linux supporters simply delusional? Is there some marketing push to get more people onto Linux? Is it just a Slashdot thing (ie: keeping the dream alive, even though it's been dead for decades)? Is there any new reason to think the status quo will ever change? Why is this "news"?

Politics in the US is a mess; divisiveness is up, discourse is down, and partisan fighting takes priority over any improvement. Swapping one side for the other won't fix this, and people are too focused on the symptoms to address the underlying problem. There are plenty of people in the country with plenty of reasonable ideas for improvement in government, but no practical way to affect any actual improvement.

If we want to fix the underlying problem, we have to solve for the meta-problem: how to get better quality people in office, preferably not politicians, and certainly not just people on "the other side". This is a solvable problem, and possibly the most important problem for modern society, yet we're making minimal progress on it. Hopefully sometime soon we can start trying to solve the actual problem.

This argument works just as well for other professions; allow me to demonstrate.

People have been building airplanes for roughly twice as long as their have been computers, but yet we are still paying hundreds of millions of dollars for top-end fighter jets. Why is that? Why have we not yet advanced to the point where building fighter jets is commoditized, and can be done with minimum wage workers with high school educations? Why do we still have to pay exorbitant salaries for so-called "experts", in this nerd-driven culture of exclusivity? #RocketScienceForAll

See, the answer is pretty simple, when you are enough of a "rock star nerd" to apply some simple logic. Making complex computer programs is "hard", and thus requires people with "intelligence" and "experience". It's the same in pretty much every specialized profession; programming is sorta the outlier because for some reason (media, outreach, or otherwise), people seem to have the misguided notion that if we were better at dumbing it down, we could make it accessible to everyone. The reality, though, is that for the same reasons that we don't have many high school only educated doctors, or lawyers, or physicists, or rocket scientists, etc., we also don't have (and it's not really possible to have) a plethora of minimally educated/experienced good software developers.

TL;DR: Don't be an idiot, good science/engineering is hard, and that's why "normal" people cannot magically be good engineers. Try to remember that good engineering is hard, even if ignorant people say programming should be easy.

Not sure if this should even be "news", per se. Google's total lack of care for qualify and usability of their software is sorta ubiquitously understood at this point.

I mean, this is the company which produces a phone operating system which unlocks the display for a device (in your pocket, for example) when answering via Bluetooth, a significant usability flaw which has been reported, acknowledged, and complained about for 6+ YEARS. Based on experience, there's no reason to suspect that ANYONE at Google cares AT ALL about usability of their software, and/or fixing bugs, and/or customer experiences.

But to be fair to Google, this has been the case since day one, and this is not new or news. If you want something you can tinker with, Google has [plenty of] products for you; if you want something which works, try elsewhere (eg: Apple, Microsoft, etc.).

I don't miss anything about Windows Mobile in particular (and I detest how it was allowed to influence the desktop OS in some weird, genetic abortion of design failure), but I do miss the time before it was released, when the ecosystem was full of promise. I miss the idea of a mobile OS which wasn't a walled garden, but also wasn't a cluster-f of unpatched, vendor "optimized" garbage, with tons of bloatware and more persistent bugs and usability issues than anyone would have thought possible. I miss the promise of something better than iOS, like the Microsoft of old, taking the solid foundation that Apple has built (in contrast to the garbage heap that is the Android ecosystem), copying it, and extending it to generic hardware, so the price would go down, and people would no longer have to choose between affordability and usability.

But then, it was not that Microsoft... it was the new Microsoft, the one which wasn't even intelligent enough to grasp why people on the desktop didn't want a phone UI, much less what might constitute a "better" mobile experience. So much promise, so much failure.

People who can code all aspects of an application exist. They are typically older, more experienced, more rounded, and the good ones can get paid a considerable amount of money, precisely because they can understand and effectively code all aspects of the "stack".

What most companies are looking for are more unicorns: people who are young (ie: less external life, so they can/will work more), can code for all aspects of the stack well, single-handedly take projects to completion and/or coordinate between different groups, and don't know their actual market worth. Those people also exist, but they are much harder to find (and hard to retain, if/when they figure out their actual value).

What companies want doesn't always match what companies will take, if they cannot find their unicorns. Success is about making it work with what you have, not lamenting your inability to find the exact right candidate you think is out there.

Funny story:

In my old company (smaller), the company modified the IT policy to make it easier for employees to access email and company data on personal devices. This made it more likely that I would check email off hours, and possibly respond if necessary (which was not uncommon).

In my new (bigger) company, the IT policy is more rigid, and you cannot access company info (including email) without jumping through several hoops (corporate device, multi-factor auth, etc.). So I no longer check email/phone off hours, or feel any obligation to answer anything (if they wanted off hours interaction, they are either idiots, or should not have erected so many barriers).

The morale, I suppose, is that if you want to encourage good employee work/life balance, you should implement more security policies. Or not, I guess, depending on your corporate goals.

In addition to the totally obvious and "duh" ideas (which would already be done if we actually had any ability to improve the election procedure, such as eliminating gerrymandering, ranked voting with instant runoffs, open source software with cryptographic security and a verified paper trail, etc.), I have another thought.

It would be cool if you could arbitrarily proxy your vote to someone else (and/or multiple people). That is, to paraphrase a hypothetical, "I don't want to learn everything about this contest in order to make an informed decision, but I know someone who does, and that person thinks like I do; just count my vote toward his/her choice(s)." I don't think we're ever going to be able to get the average person to vote in an intelligent and informed manner for the best qualified candidate (see: Trump/Oprah), but we _might_ be able to get voters to proxy their vote to people whom they judge to be intelligent, well informed on political topic, and like-minded (eg: Jon Stewart, Bill O'Reilly, etc.). In turn, that just _might_ be able to get the country away from electing the least hated of bad options.

My 2c.

... one could implore the software vendors to make the update process less arduous, cumbersome, error prone, and OBNOXIOUS AS ALL HOLY HELL.

As someone who has, on multiple occasions/systems, got frustrated enough with Windows Update to disable the service (hint: that's the ONLY way to prevent it from randomly rebooting your system when you are trying to use it, whether you like it or not), I can say with some certainty that I would have no issue with leaving updates enabled, if the process wasn't so GODDAMN TERRIBLE. Suggestion to vendors and prognosticators: the vendors are as much, if not more, to blame as the users who respond to the INFURIATING behavior of their devices. Instead of blaming the users, I'd suggest perhaps it might be more productive to blame the vendors for the poor quality software which drives the users to disable it.

Not that I wouldn't necessarily want it, but since I have the Windows Update service permanently disabled because it's so incredibly and ridiculously obnoxious and "poorly designed" (for which that phrase alone in this context gives the idea of software design a bad name), I don't think I'll ever get prompted for it.

I'm still holding out for the day when MS manages to extract their metaphorical head from their ass for just long enough to comprehend that being as obtrusive as humanly possible with pushing updates is a MONUMENTALLY STUPID business decision, if you want people to actually take updates. Gotta stop drooling on the floor before you can walk, gotta walk before you can run, gotta run before you can pitch an "upgrade" as an actual upgrade, etc.

It's sorta off-topic, admittedly, but...

It's interesting that the editor chose to call out the assumption of the continued existence of the closed-source software businesses, without calling out similar precepts (eg: the continued existence of money, or countries). I mean, if money ceases to exist, then doesn't the question of pricing become moot? What about an asteroid wiping out life on the planet: that would also, presumably, substantially alter the economic dynamics of software pricing.

If you're going to call out exceedingly low probability future events to exclude from consideration, why stop with just one? Alternatively, why call those out at all?

As has already been stated, you generally want to prefer to use a third-party library over a custom implementation, for most security-related code. This is doubly true for well-defined algorithms, which are implemented in well-tested (and preferably open source) libraries.

However... there's an inherent danger in adopting third-party libraries based on uninformed assumptions about quality, as I'm personally well acquainted with. If you have a manager who is prone to making baseless assumptions, and downloading random packages off the internet which purport to be semi-related to the current problem development is experiencing, and insisting they be integrated as the "easy" solution for that problem, you're going to end up with bad quality software (or worse).

As the saying goes: garbage in, garbage out. If you're doing software integrations based on garbage processes, you're still going to get garbage out, no matter what the quality of each third-party module.

I was going to emphasize this too.

I can't speak for all of "tech" as an industrial area, but in software development at least, there are also substantial indirect affects from the quality of work, some of which can be difficult to measure (without someone knowledgeable auditing work). Just because something compiles and produces the expected output, does not mean it handles corner cases well, or works every time, or doesn't have undesirable side-effects, or is easy to maintain, or that the design scales, or is forward-thinking in terms of technology choices, etc., etc. Getting all of those latter things might not be important in a few specific cases (eg: creating strictly throw-away demo-ware for marketing purposes), but in most business cases, each of them has a monetary value attached, and you could certainly be justified in paying more to get them.

Also, the point about competent foreign workers is well taken as well. To re-use my analogy, it's not as if there are not skilled foreign contractors also... but those people don't hang out at Home Depot, waiting to do day labor for under-market wages, they have higher paying jobs closer to home. The people who are being rented out as "cheap" foreign labor are, in most cases, "cheap" foreign labor, and you get what you pay for. It's just that in tech, more than other industrial areas, you generally get less productive value out of rote labor (in my experience).

This is very true. In the software industry, especially, there is a vast difference between people who are good developers, and people who are "just able to write code". For the organizations who employ a lot of the latter (either though legitimate need, or simply inability to attract and/or hire the former), outsourcing can be economically viable... as long as you are able to still stay in business, that is.

I know, anecdotally, that several "smarter" organizations who experimented with outsourcing software development for cost reduction have since "in-sourced" it back for quality purposes. I know others who would not have made that error in the first place. For those organizations, ability can still have value.