Hacker News Reader

Comment by capableweb | Hacker News Reader

> Microsoft needs to think hard how hostile they want to be to its competitors and users, because two people can play this game. I don't get Microsoft, have they no pride or desire to become a great company? Have they just become content to be an old corporate software house who only manages to keep users through dark patterns and anti-competitive behaviour because they have given up on making products which people enjoy to use?

You're talking about this like Microsoft ever did anything differently than what you wrote? Since when have they focused 100% on just building great products and competing fairly?

Microsoft has a loooong history of the behavior they still have, nothing is new here. Forcing people to use Microsoft EdgeXplorer? They been doing this since the very creation of their own browser.

Don't act all surprised when a company who have been acting one way, continues to act that very way still.

View on HN

> You're talking about this like Microsoft ever did anything differently than what you wrote? Since when have they focused 100% on just building great products and competing fairly?

They did a wonderful job with C# and .NET Core.

But you need to remember your org charts; the Edge and Windows 11 decision-makers are not the same people with the same incentives.

> They did a wonderful job with C# and .NET Core.

Ah yes, the ecosystems where features such as hot reloading would only be in paid offers unless the ecosystem didn't start to shout at them when their plans changed from "of course it's free" to "sorry, paid customers only".

> But you need to remember your org charts; the Edge and Windows 11 decision-makers are not the same people with the same incentives.

I couldn't care less about their org charts. That's just a way for employees to feel better about what their colleagues are doing, "but that's not what I do at the company, I'm doing great stuff".

Just like Googlers trying to convince themselves they're doing great stuff for the world because they happen to hack on FOSS stuff while their colleagues spend most of their days optimizing ads and siphoning as much data as they can from their users.

We often look at companies as a single monolith but that isn't conductive to productive discussions.

OP is giving a different perspective, a more detailed and granular one.

One interest counterpoint would be that the problematic behavior is encouraged by the CEO, then the different units wouldn't matter much (well, unless they have a tendency to go rogue). Anyway, all idle speculation on my part but I welcome the perspective.

Maybe if they stopped trying to become monoliths, people would stop considering them as monoliths?

What purpose could buying GitHub serve if they don't want GitHub to be a part of Microsoft? They are obviously trying to do everything, via a centralized organization which is Microsoft, so by all measures, they are or want to become, a monolith.

Regardless of what other people in the organization is doing, the organization as a whole is represented by everyone working their, no matter if they work on ads, cloud, FOSS or office products. You cannot work in one area and somehow ignore the work of your colleagues,

That's sort of like condemning all of the U.S. because one state does something you don't agree with. In a large organization different departments may do different things and have different incentives, and until they overstep too much may have quite a bit of autonomy.

Considering all actions taken as part of some overarching plan is probably assuming too much. Some actions probably are towards some larger strategy, but myriad other ones are likely sub-level decisions. Always looking at large corporations such as Microsoft as one entity with a cohesive goal is easy, but it's also simplistic, which doesn't always yield a useful and productive discussion.

> That's sort of like condemning all of the U.S. because one state does something you don't agree with.

It's more like condemning all of the US because it has implemented a national policy that only one state really wanted.

It doesn't matter that only one state wanted it. It's still national policy and thus fair game to criticize the entire nation for.

Is it a company wide policy that Microsoft apps open links in edge, or a Teams and Edge thing?

Do office apps open links in edge? Does VS code?

If they do, then I agree with you. If they don't, then I don't, for what should be obvious reasons.

If a company has implemented it in a product that is publicly released, it's a company policy even if only the dev team was responsible for it.

Eh, no. That's a reach. A company policy would be something that the whole company is expected to follow. If only the teams that coordinated on it follow it, that's not a policy in any way, it's just a choice that a portion of the company made, it could be something that the company doesn't care about one way or another or it could be something that wasn't addressed prior to release (if you think every decision a team makes gets complete review and is totally understood, I think you're wrong).

Even a monolith is made up of many atoms all sticking together.

Just because your particular outcropping isn't coated in digital blood doesn't mean you're not part of the problem.

If a company sent out two crews of people and one crew gave out flowers to strangers and the other crew punched little old lady's faces in, would you still crow about how you're only handing out flowers?

A friendly nitpick: the word 'conductive' is a physics term. The term which means "tending to promote or assist" is 'conducive' with no 't', pronounced 'cundoosiv'.

> Something conducive "leads to" a desirable result. A cozy living room may be conducive to relaxed conversation, just as a boardroom may be conducive to more intense discussions. Particular tax policies are often conducive to savings and investment, whereas others are conducive to consumer spending. Notice that conducive is almost always followed by to.

* https://www.merriam-webster.com/dictionary/conducive

What the hell? They’re separate words?

How did I go at least half my life without knowing this.

Wait till you learn that "deprecated" and "depreciated" are different words, too.

But "flammable" and "inflammable" are the same. English is fun!!

Unfortunately I knew this one already. It’s be too much to hope I could have my mind broken twice in a day ;)

They are close enough.

Thank you. English is not my first language so I appreciate the correction. TIL.

Pedantry around the meaning of words with common stems is against the code of conduct. In other words, do Conduits not conduct water?

You really picked poor thing to try to bash on since .NET ecosystem is one of the most sanest out of all ecosystems

Give me something as sane, coherent and friendly as .NET ecosystem for C and C++ and I would be willing to pay for that

I didn't pick .NET, parent did. And regardless if the ecosystem is nice to work in or not, Microsoft tries their true and tested strategy of EEE with that ecosystem as well.

The example is not .NET in general, but that specific event when Microsoft reneged on open development tooling[1]. For some people, that was the moment they stopped trusting "new Microsoft" to keep their word (though for me, it was when the Python language server was replaced with a DRM-locked, LSP-noncompliant one[2] a bit before that; unlike with .NET hot reload, they didn't backtrack there). I can think the company makes great open .NET tools and at the same time not trust them to close those down on a whim.

Does anyone know where the open xlang implementation of MIDL[3] went, by the way? (Unlike the original 1990s MIDL, you can't reimplement this one from the language grammar in the docs, because there is no language grammar in the docs[4].)

[1] https://dusted.codes/can-we-trust-microsoft-with-open-source and links there

[2] https://github.com/microsoft/pylance-release/issues/4

[3] https://github.com/microsoft/xlang/pull/529

[4] I vaguely remember a GitHub issue where a Microsoft employee explicitly refused to provide one ("not a priority" or some such synonym for "we don't care, fuck off"), but I can't find it

> since .NET ecosystem is one of the most sanest out of all ecosystems

Is it? This is the same ecosystem that has .Net, .Net Core, .Net Standard, and .Net Framework as all different (and overlapping? who knows) products?

Its only .Net now.

The Core and Standard things was just there the aid the transition from Framework, which is now complete, any remains will soon be phased out entirely.

> They did a wonderful job with C# and .NET Core.

You mean the products Microsoft created when it became apparent they were losing the lawsuit about their embrace, extend extinguish tactics with J++ and their custom Java implementation. And also had to cough up billions of dollars in a settlement with Sun [1].

[1] https://www.infoworld.com/article/2667124/update--sun--micro...

I think he means the newer versions of the stack, from the release of .NET Core 1 in 2016 and onward.

Ok, but even if it's about .NET Core; As far as I can tell the creation of .NET Core was a business decision and not because MS suddenly wanted to create a better product.

More then 50% of the VM's in Azure run Linux and not Windows. On AWS and GCP the percentage VM's running non-Windows OS'es is probably even higher. That's a realistic threat to the continued existence of .NET if .NET only runs on Windows.

Aside from the cost of licensing Windows which makes small-sized non-Windows VMs cheaper, there's been a strong pro-Linux bias in academia for a long time that nobody likes to talk about.

It’s a preference, not a bias

The choice is open, free vs closed, stifling, costly, dark-patterns.

I don't think that's bias.

I just moved to openSUSE.

It's really nice not having advertisements in my start menu.

> there's been a strong pro-Linux bias in academia for a long time that nobody likes to talk about.

Who doesn't like to talk about it? It's so much easier to integrate components on Linux over windows. Why would anyone willingly use windows for data analysis?

Everything every public company does is a business decision

Isn't wanting to make a better product a business decision?

M$ only improved their product because they want to make more money! Unlike me, who works for free.

I hate Microsoft as much as your average Gen X tech nerd, but C# is good. I don't care what inspired them to do it.

You mean you don't care about Java?

I also hate Oracle. Now what?

OpenJDK, that's what.

Microsoft is terrible in almost every way, but I admit that I very strongly prefer C# over Java.

C# is great. Also VS Code.. how many Microsoft haters on HN daily drive it? I'd have to assume quite a few!

The Remote SSH extension has been a game changer for how I do development.

I like VSCode too. I use it for Clojure, Python, and C#, which I am working in this year and haven't for a very, very long time. I was pleasantly surprised at the progress it has made.

Ironically compared to Windows, MS has been willing to kill backward compatibility more to improve .NET to make it more cross-platform, something that must come from a very different part of the company, probably the one that got the "DEVELOPERS DEVELOPERS DEVELOPERS" mandate.

I make no apologies for their history, but I like these tools and I like them much better than Java, which also has its place. (This isn't a throwaway comment, notice I said I use Clojure, it's my favorite, which runs on the JVM.)

VS Code is pretty great, and I agree the remote dev extensions (ssh) are hands down the best feature added. I also like that it had a directory tree and the integrated terminal early on, that's what got me hooked.

C# is nice, but the experience in VS Code is sub-par and I think it's the VS (not code) managers that are responsible for that state of being.

Editors have had directory trees and integrated terminals for many years, why is that something you find special?

I'm not familiar with another code editor with an integrated terminal, at least not one as complete as VS Code's before VS Code. I could be mistaken though.

C# is great. If only they hadn't made it windows only from the get go there might be a huge ecosystem supporting it instead of java.

>They did a wonderful job with C# and .NET Core.

This is a joke, right? We never, ever, needed anything other than the Win32 GUI at the core of windows, and it still works to this day. You can write Windows programs without the new stuff, in Lazarus, for example.

C# and .NET were a huge step backwards. VB6 was something easy enough for domain experts to get working, without having to hire a programmer. Delphi for Windows make it very easy for even beginner programmers to ship a clean professional Windows GUI by just building the UI and hooking events to it, with almost zero boilerplate. C/C++/C# reverse that trend and lost us all a decade or more of productivity.

I spend all my time writing services and HALs and even when I do need a UI, it's in the browser, that I forget there are people who still code actual GUIs for Windows.

How so? Because they sort of open source the tech and made it possible to build for other platforms? C# is a great language taking over the advances of Java and C/++ making things a bit better, but intentions didn't make it any better: owning the devs.

VScode? A rebuild of Atom, "free" with un-disable-able telemetry with the same intention: owning the devs

Buy out of github, owning the devs

Rebuild of ms office for the web: owning users given gsuite was seriously taking over office use

OpenAI: since bing kept being a failure, they invested elsewhere and may have a chance, here again: own the general users and to a degree professionals.

Anything MS did and continues to do is lateral moves to own markets. The pretence of openness or innovation makes it even worse than the corp from its flamboyant 90s days, at least back then nobody was taken for a fool.

> Rebuild of ms office for the web: owning users given gsuite was seriously taking over office use

And Google let them win by coasting on GSuite. I use it for my business, my partner uses Office 365 for hers. In 2023 Office 365 is by far the better choice.

> And Google let them win by coasting on GSuite. I use it for my business, my partner uses Office 365 for hers. In 2023 Office 365 is by far the better choice.

If you don't mind me asking: what is better about o365? I know that Excel is the jewel in the Office crown, but if you aren't leaning on that heavily, what's the big difference to you?

Can't speak for GP, but the o365 Word is better than Google Docs as well. And while you can view Visio in o365, would be nice if you could edit the diagrams. Odd, but Visio is still imo much better than alternatives in that space.

Visio documents can be edited on ODfB and SPO within the browser.

https://support.microsoft.com/office/overview-of-visio-for-t...

Sweet... hadn't actually touched it for about a year and a half.

365 web apps are still shockingly terrible in 2023.

Just to add more examples of markets they are trying to take over by not being better, but by buying up the competition:

- Video games: Activision/Blizzard, ZeniMax, Double Fine Productions, Obsidian Entertainment, inXile Entertainment, Playground Games, Compulsion Games, Undead Labs, Ninja Theory, Mojang

- Developer ecosystems: npm, Dependabot, GitHub, Playfab, Deis, Xamarin, Havok, HockeyApp

This is standard business affairs for pretty much every big company. Apple, Google, Amazon, and many other fortune 500 companies, have all bought up dozens of businesses and their products and rebranded them as their own. Not sure why we're solely picking on Microsoft here...

Indeed, Microsoft didn't invent capitalism. They just have a strong position and understand the rules very well.

Because Microsoft is the subject of this thread. We could pick on other corporation, MS isn't the only one, but whataboutism isn't an argument.

And MS practices are not standard business, they certainly understand the rules well, but have have been making illegal moves, it is a fallacy to word it as if illegal moves were part of the capitalism game: Monopolies are to be prevented, that's why anti trust laws were put in place, which MS was already caught breakig.

> They did a wonderful job with C# and .NET Core.

And VSCode. One thing all three have in common is that they are all FREE to use--they don't make Microsoft any money directly.

And, for the segment of the developer tools market that wrangles C# code, if VSCode gets too good, it becomes a threat to a cash cow: Visual Studio.

Start here: https://github.com/OmniSharp/omnisharp-vscode/issues/5276#is...

The relationship between VSCode and commercial versions of Visual Studio is the worst part of .net, especially the omnisharp debacle.

Given much of the key feature set of visual studio is now in VSCode for free for .Net development, I wonder what future holds for Visual Studio. My assumption is that Microsoft will eventually stop making commercial IDEs and just give the tooling away, as it all helps drive Azure/cloud compute sales.

VSCode runs in a browser and can now be launched straight from a github repo page - I don't think you need a crystal ball to see the way things are going. Imagine onboarding someone with zero dependencies to install on their development machine - just pop a browser and get to work.

> https://visualstudiomagazine.com/articles/2021/08/31/github-...

The "portable" language that doesn't run on FreeBSD, Solaris, AIX, etc.?

Uhm, I'm currently running abour 4 binaries that have that famous `exe` file extension on FreeBSD. C# runs on FreeBSD just fine. There is no bootstrap available for FreeBSD - must either cross-compile or use linux emulation.

> They did a wonderful job with C# and .NET Core.

And typescript!

And Azure. But aside from C#, .Net, Typescript, and Azure, what have Romans done for us?

Yeah, I beg to differ about including Azure in that list.

No mentions of Excel? I still haven't found better spreadsheet software...

Its interesting to look at their 'product' history b/c they have a few shining moments of 'getting it' as far as OP means. .NET especially realizing Linux was a real thing and to at least outwardly 'embrace' it.. Then there are moments like these the OP points to .

I guess that's about .. literally.. average.

Nobody said that they have never made a good product. But in no way does this mean that the company focus has been on making good products vs strong-arming their way into things.

> But you need to remember your org charts; the Edge and Windows 11 decision-makers are not the same people with the same incentives.

People who I assume are MS employees keep saying this on HN, and to an end user it's really weird. I don't need to know squat about the internal org structure of some monolith in Redmond, they need to stop screwing around with this anti-consumer behavior and regulators need to slap their hand any time they try and reach for this cookie jar.

You think those licenses for Visual Studio and those Azure VMs sell themselves?

> They been doing this since the very creation of their own browser.

Before that https://en.wikipedia.org/wiki/AARD_code

To me, this is the proof that the dark patterns are a result of corporate culture that stretches all the way back to the DOS days.

They simply cannot resist the urge to cheat in this manner.

Arguably, Microsoft is why a browser's user-agent is so useless for determining supported features.

https://webaim.org/blog/user-agent-string-history/

The context is explained prior, but the part I'm pointing to is:

> And Microsoft grew impatient, and did not wish to wait for webmasters to learn of IE and begin to send it frames, and so Internet Explorer declared that it was “Mozilla compatible” and began to impersonate Netscape, and called itself Mozilla/1.22 (compatible; MSIE 2.0; Windows 95), and Internet Explorer received frames, and all of Microsoft was happy, but webmasters were confused.

Also arguably, the ubiquity of user agent sniffing is why. You get what you measure. You want your UA to contain "Mozilla". By god, you'll get it one way or another.

Yeah, I was a little hesitant to suggest that this was caused by Microsoft given that. I actually agree that this problem is caused by user agent sniffing. I could see someone arguing the other way, though.

Ultimately, I think it's still on-topic because, at any point, any of these browser vendors (admittedly, not limited to Microsoft) could have started to work with webmasters for a standard of "feature claims" instead of "user agent claims". So a user agent can say, "I support frames," and a server can serve frames. Instead, Microsoft decided they'd just lie about it.

Switching to a system of "user agent claims" have been up to any one of the browsers, when they had a majority of marketshare, to push through. Mosaic/Netscape/IE all had a chance in the Web 1.0 world, unfortunately, none of them did.

I agree with that too, and admitted to such in my previous comment. Still, being the first practitioner of this comes with a certain responsibility for its proliferation. It's not a sole responsibility to your point, but it is primary in a way. It's a lot easier for others to decide to copy what someone else did than to decide to be the first to do it.

To the same point, Google is arguably the reason it persists today given the popularity of Chrome. Just as Microsoft is arguably the reason it started. (I suppose I could have been more careful in the phrasing of my initial comment since "is so useless" could imply both but I don't think it's particularly egregious.)

One thing I have always wondered about the story - were servers in the mid-1990s really sniffing user agents to decide what pages to send through? I remember most sites were static .html files back then.

Not sure if it's the server, or the javascript, but it was definitely a thing in the 90s.

I think user agent strings were a terrible idea from the get-go, so I'm OK with Microsoft (and everyone else) fudging them this way. I think it'd be even better if everyone used the exact same UA string.

> I think it'd be even better if everyone used the exact same UA string.

Then why send a UA string at all? Save the bytes on the wire.

I agree -- I don't think the UA string should exist. But it does, so the next best thing is having everyone send the same string.

“DOS ain't done till Lotus won't run"

That is not accurate, incidentally.

> "It's an interesting myth, and one I've heard about in general terms, although I've never heard the specific quote before. However, I have no recollection of any instance of its actually happening with 1-2-3 or with any other product I've worked on." And, "My memory of the early days (1984-85) is that we would get early betas of DOS to test with 1-2-3 and any errors that we found were 'bugs' in DOS and fixed by Microsoft."

> Try to imagine a customer with a working copy of 1-2-3 who installed a new version of DOS and 1-2-3 stopped working. Would they blame Lotus or Microsoft? Would their reaction be "1-2-3 sucks" or "DOS sucks"? Would their solution be to get rid of 1-2-3, or stop buying DOS upgrades?

[0] http://www.proudlyserving.com/archives/2005/08/dos_aint_done...

Speaking as a Lotus Notes survivor, it would have been really helpful if had prevented lotus notes from running on windows as well.

Lotus notes was like the ultimate in write-only software.

Encarta 96 was incredible.

100% agree, but it has been a while since that came out...

Spent countless hours on it as a child. I didn't even really know English.

"Flight mode" was a lovely curiosity at that time. Example from Encarta 98, but I thought 96 also had it.

https://youtu.be/7ytQQ4XdQmM?t=131

> Since when have they focused 100% on just building great products and competing fairly?

iirc They bought (or rather scammed) Internet Explorer from Spyglass. As they did with many of 'their' products back in the day (Windows NT/SQL server/...). Are MS building products themselves these days?

Like all big companies, when something comes along that is growing fast, where they don't have a product, they might choose to buy an existing one to get in the market. On some of them they might spend lots of time and money building huge amounts of new technology and improving on some of them so they were hardly comparable to the original. SQL Server is one example I worked on. The original sybase product had a pretty nice sql language with stored procedures, but the implementation didn't take us very far. It was basically awful. We spent years building out a new database with that same sybase surface layer, we added lots of new things over time - new execution system, new query optimizer, integrated it with .net, etc. Because we didn't have to develop a new surface or sql language, it helped a lot. I always wondered how much that cost Microsoft to buy it. I'd say even at 100 million it was worth it.

> have they no pride or desire to become a great company

Yes. But their definition of those terms are stuck in the 90s

Reply on news.ycombinator.com