Windows is doomed
theweek.com
theweek.com
- there is a viable alternative to Microsoft Office on the desktop. This doesn't mean one that gets most of the way there; LibreOffice and company, Office on mobile, etc. don't quite cut it. Macro jockeys and VBA scripting are huge game changers for corporate clients, especially in the finance and accounting worlds.
- common CAD software has been entirely ported to a new platform and has complete feature parity. Considering how conservative engineering/manufacturing firms can be, this will likely never happen. No large engineering corporation will ever risk, or run the risk of ever having to run the risk, of even the slightest of hiccups in a large software migration. The only big swinger even making an attempt in this department is Siemens NX, and even we're still quite far out from feature parity for even that single package.
- Active Directory has a viable counterpart for a competing operating system. Some people are trying, none are truly succeeding.
Those three points alone are enough to prop up Windows for the next couple decades. Don't get me wrong, I'm not a fan of Windows, Microsoft, their business practices, or many of their ideas on how things should be done. However, it drives me up the fucking wall when people make dumbass claims that Windows will die within any reasonable timeframe (20-30 years).
EDIT: As a quick clarification, I bring up CAD not because it attracts a large userbase (it doesn't), but because switching away from Windows means that the manufacturing base of the entire world will undergo a massive change that could prove fatal in many ways. This means that any company with a significant CAD userbase will be using Windows, even if the HR, IT, etc. doesn't, in an attempt to consolidate overhead. This also means that many companies that work with engineering/manufacturing companies will stick with it for compatibility's sake. Many people are commenting that CAD/manufacturing doesn't account for many users, but that's beside the major implication. Engineering and manufacturing customers alone could ensure at least a workable level of Windows support, all while indoctrinating most of their employees in to Microsoft's ecosystem.
However, I do see it as a possibility that windows being doomed as a consumer OS might happen in our lifetime. As more things move to service oriented and more devices become connected it's very plausible that the majority drop their desktops and laptops and rely on mobile and chrombook level devices.
It's not doomed as a consumer OS. People want to use the same OS at home that they do at work, it feels familiar to them and they already know how to get stuff done.
Millions of professionals very good at their jobs, but with no particular technical proficiency, have been using Word and Excel for years, and convincing them to switch to something else would take much more effort that any IT manager with a budget can possibly muster.
OnShape is a viable online replacement for SolidWorks but it is hard to transition people from their favorite software package.
CAD is not the biggest deal and is not a major factor, even if some people think it is. Less than 1% of corporate office computer workers are CAD and it is probably way less than that even.
Office and Active Directory are a big deal though.
Adobe Suite is another major factor: Wacom Tablets are still superior to iPad Pencil and Surface Pen (although the Surface Pen provides a cheap entrypoint for these programs). And nothing on iPad has the featureset of Photoshop (or Manga Studio or whatever you prefer)
Non-CAD programs like Blender, or Maya and whatever, are still best on a PC.
----------
And as long as you're using PC software, you'll need an IT department that supports enterprise PC operations. Active Directory and Group Policy really does make life easier for any IT Administrator who is in charge of 100 to 1000 computers.
"Doomed" means a slow eventual death, not an immediate one. This reminds me of how people were so busy knocking down strawmen with the whole post-PC thing.
Reading the article, the problem becomes more obvious: without mobile, they have a huge gap in their offering and have to embrace competing OS's (Cortana goes nowhere with a non mobile windows).
Obviously this would require Microsoft themselves to undercut Windows, but Office is such a cash cow for them, they might see it as worthwhile.
Even Server Windows to an extent if their other on-prem cache cow (SQL Server) does well on Linux.
Windows is not their core priority these days. The Azure platform is, and Windows is just a part of that which could be swapped out. Maybe Windows will just become a free (though obviously not Free) product, with the money coming in from Azure + SQL + Office.
Back in the bad old days, Microsoft could barely be convinced to keep a Mac port of Office going, and that was basically to keep the government at bay. But they seem to be more open now.
Office...I'm in finance and massive numbers of people are either working on ancient versions of Office, or they've transitioned to alternatives. Office is absolutely enormous in scope, but it's one of those things that is detrimental to many workflows because of its powers. Many shops prefer webapps and on-demand generated consumables. Even Excel is very rapidly dying.
The criticality of Windows has absolutely imploded, and is nothing like it was. If someone wants to use Linux, or OSX, or Android, or whatever, in most cases it fits in many workflows with absolutely no issues.
On the home front the single thing keeping Windows running in most home situations is game compatibility. A massive number of home users never go outside of the browser, and those that do usually do just for games.
This is a sample:
dispatcher = createUnoService("com.sun.star.frame.DispatchHelper")
dim args1(0) as new com.sun.star.beans.PropertyValue
args1(0).Name = "ToPoint"
args1(0).Value = "$C$7"
dispatcher.executeDispatch(document, ".uno:GoToCell", "", 0, args1())
Now in MS Office I'm pretty sure this would be something like ActiveSheet.Cells(1, 5) = "foo". There is nearly no discovery for the LO Basic equivalent 'sheet.getCellByPosition(3,16) = "foo"'.
It really is worlds apart. I don't see how a moderately sophisticated user can learn to write LO macros without finding a reference manual and investing hours orienting themselves. I've seen people teach themselves to a useful standard using MS Office macro recordings.
EDIT More importantly, a novice wouldn't necessarily understand that the recorded macro isn't the accepted way to select a cell. Even realising that a cell can be selected with one line is hindered by the LO macro recording.
Which is why replacements seldom have more than the most trivial of automation -- the people who are making the move usually change their workflows. The offices that built around Office 95 are, however, still doing things the way they always do.
EDIT: What a changed world when this is a contentious comment.
In 2017 there are shockingly few cases where macro programming an Office document is an ideal solution, and it is usually a really clear sign that you're approaching the problem wrong. It was seldom a good solution even in 2010.
It's funny that in these conversations people always bring up finance. I work in finance, for a wide variety of companies making a wide variety of solutions. Excel automation is a curse. Every Excel automation is an abomination from a decade plus earlier -- islands of data, extraordinarily full of errors, security nightmares, data exposure risks, and on and on.
It may not be the best workflow, but it is the most accessible one for an individual to start with.
Edit: for what it's worth, I agree that it can be a hindrance to a company when it becomes a core operating process. However, that doesn't mean that it is inherently wrong for many needs
But in 2017, it is almost always a very bad sign. Which is why the competition have mostly abandoned automation of the 1990s Office sort.
Windows is on the same path in part from the windows 10 fiasco, but also from the stagnation in computing hardware.
Consider Direct X with ever improving hardware it's was a hard target to keep up with. But, now they would have to make changes for the sake of change while providing backward compatibility...
I agree that the scripting is a moat around Window's dominance but most small businesses run Windows because the hardware is more affordable and a lot of people are comfortable with Windows. Office is also super cheap now and free alternatives kind of suck and they address common pain points incrementally better than they did 10 years ago. In other words, it doesn't feel like free Office alternatives are keeping up. For example, LibreOffice stopped trying to improve compatibility with MS Office documents. I get the philosophical reasons but if you want to migrate end-users, you have to care about things like that.
At my son's public school, he was being taught to use Google Docs because (of course) it's more affordable than buying a bunch of copies of Microsoft Office for the kids. He was doing homework one day and he was trying to do some common formatting thing in Google Docs. Can't remember specifics. After a few minutes of trying to help him figure it out, I was like "let me introduce you to this little program called Microsoft Word". It blew his mind after using Google Docs.
For years high end Unix workstations dominated in the 3D graphics and design field, in engineering, industrial design and fabrication, basically anywhere a computer+engineer generated the kind of income which justified a $50,000 computer (epitomized by SGIs and Dec Alphas) plus equally or more expensive software licenses.
What happened is the bottom dropped out of the hardware market. The Pentium Pro came along, Windows NT 4.0 was released with OpenGL support, Nvidia brought powerful 3D to the consumer desktop, computers with 256MB or more of RAM were commonplace, and all of this was available for a few thousand dollars. Over time, almost everyone making software for these platforms either jumped ship or went broken.
So while there's a historical precedent for this same migration in the computer field, something's different this time. There's no room at the bottom.
The only one that has a remote chance of killing either one at this point is Microsoft themselves through continuum and X64 code running on ARM. Even then, phone horsepower is still not enough to fully replace my laptop and desktop usage. I don't see that changing anytime soon with the issues fabs are having getting to smaller process sizes.
In any case, the use of the word "doomed" is hyperbolic or at the very least ill defined scaremongering.
I really don't think many here actually read the article and instead just focused on the byline without any context.
Phones are not computers and are deviating ever more from being mistaken for devices that can be used for actual productive work.
So Microsoft lost the mobile handset/OS race.. so far.
I'm not encouraged that any of the reasoning in this article represents nails in MS's coffin.
As long as finance managers want to download their financial data from SharePoint to Excel, their reports generated on Crystal Reports and edited on Word and as long as finance managers are the ones paying for new software and computers then both Windows and Microsoft have a bright future ahead.
It is called synergy. The same "walled garden" approach that Apple uses for consumer market is working very well for Microsoft on the enterprise market.
Without Mac OS there'd be no iOS apps (or at least, many fewer). And without Windows there'd be many fewer Xbox and likely Android apps too.
Non-mobile computing is needed until (at the very least) the bulk of development for mobile is done _on_ mobile (hint: I doubt that will happen any time soon).
If you ignore the lag from the CPU or memory constraints, the interface is just not conducive for working. Who wants to edit text documents or spreadsheets, used a shell, or edit code on a 5 inch screen? Who wants to use a soft keyboard for any non-trivial amount of work? It's not that it can't be done; there are much more attractive options readily available.
There there's the argument about connecting mobile devices to monitors, using adapters/bluetooth for peripheral devices, and thin client setups. If that's the case, why don't you grab a cheap laptop? Mobile is gimping you at that point.
One can develop Android apps just fine on Linux. It's a superior experience to Windows or macOS IMHO.
Honestly, I find the modern Linux desktops friendlier to non-power-users than modern macOS or Windows.
I think the rule is if you're marked as doomed incorrectly more than three times over the course of a decade, then you are too big to fail and are, by definition, not doomed.
That'll keep it in the home, and AD & Office will keep it in the workplace. Seems not so doomed to me.
I hate that the platform vendors cannot ally themselves with those interests and must instead always attempt to lock developers in, damaging us economically and wasting our time and human potential.
When a vendor suffers because their lock-in attempts falter, I can't help but feel schadenfreude. Now, if only their competition were consistently better...
But...even if Windows goes away, there's AWS and Google Cloud, but the ability to migrate existing legacy enterprise applications into Azure, as well as leverage new features like machine learning and cognitive vision, will let Microsoft stay alive.
How successful will it be in the long term is the question.
Ultimately MS (like IBM) will not to be able to sell anything outside the C-suite, because anyone closer to the technology knows there are better cheaper solutions.
(But Xboxes...hmmmmm...maybe my view is a little to simplistic.)
https://docs.microsoft.com/en-us/azure/azure-resource-manage...
Thanks for the time capsule experience though, sometimes I miss the legendary zealotry of Slashdot.
It totally misses the scale and intertia around Windows in the enterprise. It may not show up as the hot platform for new applications on HN, but it's not remotely "doomed".