Microsoft CTO: Thoughts on OpenAI (2019) | Hacker News Reader

Microsoft CTO: Thoughts on OpenAI (2019)

twitter.com

323 pointsmfiguiere158 comments

To add more context, about a month later after Kevin Scott's email...

July 12, 2019 OpenAI announces Microsoft's $1 billion investment

- https://openai.com/index/microsoft-invests-in-and-partners-w...

- https://news.microsoft.com/2019/07/22/openai-forms-exclusive...

It was interesting how K Scott was objective and candid about Microsoft's internal ai efforts falling behind but also phrasing the email diplomatically to not criticize anyone. It's interesting to see some of the behind-the-scenes thinking.

I don’t think they want a repeat of missing out on web and then mobile again.

Was gonna say, the email reads similarly to this famous thread about Java (with the old CTO of MS!): https://www.techemails.com/p/bill-gates-im-literally-losing-...

Really makes you think about the structure of mega corps and how powerful the “defender’s advantage” is. These giants knowingly sleep on disruption and wait to time their entry, and are generally rewarded. I dont know if its good or bad, it probably depends, but I think the capitalism game devs need some balance tweaks.

Are they generally rewarded? Wondering is there are statistics on that because the narrative is usually that defenders end up unable to compete even with technologies they themselves invented and die out, with maybe apple and Microsoft recently breaking that trend.

I think in the long term Google could still win the AI wars. They're behind right now but they have the talent, money, and infrastructure to win. That said, I would absolutely not bet on it. I sold almost all of my GOOG.

The defender's advantage only counts if the corporation is run by people smart enough to do something about it.

History is littered with out-innovated companies that are gone or clinging to relevance. Xerox, IBM, cable companies, telephone companies, every train company ever. All had a huge opportunity to disrupt their own business model and make even more money, almost none of them managed it.

It's funny that people only seem to remember the successes. Seemingly dominant giants fall all the time. Kodak, Nortel, Blockbuster, Blackberry, ...

A current model in danger: flight travel in the US.

It's becoming too expensive and too much of a hassle. On top of plane tickets, the airport rental car system is also absurdly expensive. Flying, while fast, is full of inconveniences that make it less worthwhile.

The first airline that rebrand itself as (1) from your door to (2) your real destination, and starts including effective ways to get to/from airport while having faster security lines will become dominant.

E.g. if Delta started connecting ATL to the local region by rail or fast buses

Microsoft didn’t miss out on the web. They won it so thoroughly that they assumed it was theirs to keep and stopped competing.

Seems we collectively forgot about IE era.

Nightmares are best forgotten.

You still live on them but now it is called Safari

The IE era, a shoddy browser by a sleepy dormant pseudo-monopoly, is what allowed for Chrome to be so good.

Lots of low hanging fruit in optimizing browser rendering and JavaScript. Google engineers plucked these fruit. Pichai took all the credit.

In an alternate universe, some IE PM focused on making a great, fast browser, Chrome would’ve been killed by Google after not taking sufficient market share to deem worthwhile, and Pichai would be working at Oracle as an L5 TPM.

It was Firefox, and to a lesser extent Safari, that initially calved into IE. While Chrome did indeed strike the fatal blow, it came a little later into fight.

You also missed the GPs point, the reason Microsoft could be “sleepy dormant” on the web was because they had already won the browser wars and because the dominant platform for a decade. And that wasn’t by accident, it was very much a strategic move to conquer the browser market.

You're spot pn on both accounts. Firefox was the one that chipped at IE initially.. remember Firefox 3 campaign and how everyone put banners for/of it on their sites? good times

They did fall behind initially, before "The Internet Tidal Wave" memo was written and Microsoft fully mobilized to fight Netscape. Bill Gates was convinced that proprietary networks like the original MSN were the future, but he changed his mind fairly quickly.

That sort of was the future, with Facebook reviving the proprietary network.

Huh. Exactly right.

We ended up with Xanadu's vision of walled gardens. Without all that pesky Xanadu technological innovation. Two-way links, OIDs & tumblers (zippers? Now I forget.), transclusion, micropayements, etc.

Another example of worst is better.

Parent is referring to the decade before that.

Berners-Lee created the WWW in 1991. Netscape 1.0 launched in December 1994. Bill Gates’s big Internet turnaround came in 1995. By 1999, IE was the dominant browser and Netscape was going broke.

Things moved so fast at the time, it left the impression that Microsoft was slow to react. But actually everything happened within just a few years.

There are multiple eras here.

MS missed the tpc/ip train completely and ended up just copy pasting the stack from one of the BSDs in Windows 95 I think, or it wasn't a default install, I forget which.

They were late to the http train and had to give away explorer for free. You can tell how desperate they were since they didn't do the same thing with office.

Then between 1999 and 2012 they completely dominated.

Then they missed the mobile train completely and lost the browser wars. And here we are.

I blame IE being free for the state of the internet we have today. If people were trained to pay for it then we wouldn't have spyvertisment as the only model for it.

The death knell of MS's mobile ambitions came when the iPhone was released mid 2007. Windows Phone 7 launched end of 2010 as an unholy abomination based on Windows CE and a Zune-based interface, which created a whole generation of phones that would be unsupported by the NT-based Windows Phone 8 (released 2 years later), which is what they should've released in the first place.

I remember installing the TCP/IP stack from a treasured special floppy onto win 3.11 machines (giving them RIPE addresses and using NT4 as a router to expose the entire network onto the public internet - not a great idea in retrospect, but it was 1994 and everyone seemed so polite).

They didn't miss out on mobile. They had a strong mobile system with more than 10% market share in Europe but they only focused on the US market where it was not successful. Then they killed the platform with the god awful Windows 10 Mobile. The biggest enemy of Windows Phone was Microsoft themselves.

I actually really liked Windows Phone, but it's obvious why it didn't take off - iPhone captured all of the "sheep" consumers (the ones that just want what everyone else has) successfully (and still owns them somehow), and Android captured all the rest (which tend to be smaller target groups). The largest target was probably price point, but cheap androids sucked so much, it just created more Apple users.

By the time WP came around, this was already been done. Phones were costing near a $1000 and a 2 year contract, and opinions have been made. No real amount of people was going to risk a windows phone when you'd have to fork over that much money and that much of a contract term.

>The largest target was probably price point, but cheap androids sucked so much, it just created more Apple users

Doesn't this statement contradicts itself? Nobody buys an iPhone because cheap Android phones suck(ed) - you would have to shell over ~10 times the money. Android still dominates the European market (~65 % market share).

Good Windows Phone were 120 to 250 € and WP8 was optimized for low-end hardware so that was no problem. You got a cheap phone without an annoying expensive locked-in contract. I saw a lot of people using a Windows Phone exactly ten years ago. I had a brand new Lumia 535 myself and only paid 130 €.

No, it doesn't. Android on cheaper devices sucked so hard, that people decided forking the money for an iPhone that just works was now worth it if they could afford it. Paying $100 for a POS isn't helpful to budget conscious customers. If they couldn't afford it, they maybe still found a way to via various forms of financing.

Windows Phone came too late to dispel this problem. I still run into people who remember their first and only Android and switched to Apple and never looked back. WP never had a chance to wins those customers.

> iPhone captured all of the "sheep" consumers (the ones that just want what everyone else has) successfully (and still owns them somehow),

This is not a good take. Android was a highly fractured ecosystem that didn't really have good unified product experience until Nexus and Galaxy S3 rolled around. You don't go to a $1000 phone because a $100 phone sucks.

It’s fair to say that the CTO’s job is to make sure Microsoft never misses a trend. Their number one existential threat is a new tech trend that makes their existing model fail. Whether that thing is the rise of smart phones or generative AI, Kevin Scott’s job is to give early warning.

“Do you care to know why I’m in this chair… why I earn the big bucks? I'm here for one reason and one reason alone. I'm here to guess what the music might do a week, a month, a year from now, that’s it, nothing more.” John Tuld, Margin Call

I think about this movie all the time now that I’ve been in the corporate world for a while and I have realized one consistent outcome: the leaders who “guess” right a lot survive a hell of a lot longer than those who get it wrong.

I could perfectly hear in my head Jeremy Irons delivering those lines. What a fantastic scene.

"And please, speak as you might, to a young child; or, a Golden Retriever" -- I, when I'm reading academic papers. Also I, when I'm trying to review a pull request.

That is one of the best movies I've ever seen, and I've seen a lot of movies. But I am a little biased since I like that kind of stuff.

I agree, it is as near perfect as a film can be. It does exactly what it sets out to do, the acting and casting are entirely on point, and there is no violence.

Spoiler: A phone did get destroyed, though. Arguably a (large) part of the economy as well.

You don't need to guess correctly, a large company with enough capital can just buy a huge amount of new developments.

Wait how was the story again with Ballmer and the Iphone?

Isn't that exactly what an LLM does? Predicting the next token?

Exactly.

That's what struck me as well. It's extremely important for such positions to be clear and open about their position and not masking their reality. You'll often find marketing speak internally between levels which masks truth and then hinders actual strategic moves. In this context CTO would've said they're not that far off, if only this and that.. but no, he fessed up and CEO also understood; raw awareness.

Google was a powerful place to work at 10 years ago. At Microsoft you had to convince people to give you quota for a few cores and wait for days or even weeks, if they even agree at all. At Google you could spin up a distributed batch job processing terabytes of data during your orientation codelab, literally on the first day. The contrast was very stark. Google was a very low friction environment where things just happened, and the environment was set up such that you couldn’t help but ship something. Note the past tense.

Internal research had low friction external product not so much. They even had a tool, ariane, to help you navigate the product launch friction.

IMO Ariane was pretty straightforward and reasonable, all things considered. In fact it’s preferable to navigating the usual wobbly maze of humans each of which takes days to respond.

That's in contrast to what OpenAI's David Luan "Why Google couldn’t make GPT-3" (https://www.latent.space/p/adept):

  And it turned out the whole time that they just couldn't get critical mass.
  So during my year where I led the Google LM effort and I was one of the
  brain leads, you know, it became really clear why. At the time, there was a
  thing called the Brain Credit Marketplace. Everyone's assigned a credit. So
  if you have a credit, you get to buy end chips according to supply and
  demand. So if you want to go do a giant job, you had to convince like 19 or
  20 of your colleagues not to do work. And if that's how it works, it's
  really hard to get that bottom up critical mass to go scale these things.
  And the team at Google were fighting valiantly, but we were able to beat
  them simply because we took big swings and we focused.”

The whole episode is very interesting.

Yes. Hence the past tense. There are more reasons why they fell so badly behind, from bureaucracy to eye wateringly decrepit, overengineered, and legacy ridden training code, to slow and hard to debug AI infra, to deliberate forking and siloing of critical projects, etc. Let’s just say Google is now very far from where it was a decade ago. I’m mildly surprised they released anything competitive at all. I’m not surprised that they failed to beat OpenAI (and therefore ironically Microsoft - how do you like them turntables?)

thanks for sharing it!

This video [0] was from 2010s, and seems like they had similar issues even back then. (never worked at Google)

[0] https://www.youtube.com/watch?v=3t6L-FlfeaI

These emails were released as part of the antitrust lawsuit against Google currently being pursued by the FTC. It seems to me that contrary to the FTC's claims about how Google's monopoly power leads it to stop innovating, exactly the opposite is true. If Google had stopped innovating it's clear that Bing eventually would have caught up in terms of quality. As these emails make clear though, Google kept its lead by continuing to invest in cutting-edge AI research.

Indeed, if anything it's Microsoft who should be scrutinized.

why were these emails released as part of the FTC's lawsuit against google.

How did the FTC get microsoft's email over an antitrust lawsuit against google

Both are acting with monopolistic power in numerous areas and both should be scrutinized. Just because one argument against monopoly power is that incumbents can sometimes rest on their laurels and fail to innovate doesn't make this the sole reason to pursue antitrust action.

Not on you though, this kind of reasoning failure is so common I think it deserves a fancy "cognitive bias" name. Off the cuff maybe something like "Single Rebuttal Fallacy"?

> These emails were released as part of the antitrust lawsuit

I never understood this: Why would hot shot, high powered people risk putting such things on email? Things that could backfire when released into public domain like from a lawsuit or a leak. They know this well, and still keep doing it. Why not setup an in person meeting, or just pickup the phone and talk ? Why email?

I almost feel they actually want it to happen but could never point my finger to how this could cover their asses, or an exit strategy?

Ultimately, if the people at the company have to do a job, they have to communicate somewhat, and putting it on an email or another recorded medium is the best way to prevent endless meetings and have the perspective documented in a single place.

Everything has a legal liability, it doesn't mean that it's worth it to move everything to an undocumented medium.

These emails might have been offered by Google to show the opposite right, they just are part of the trial.

How would Google have access to Microsoft's emails?

It is probably an observation and a forecast at the right time. I remember my days at one of the Top Home and Enterprise PC manufacturing companies over 15 years ago, when there was criticism around Smart phones.

People laughed assuming that a smart phone is of no use and people prefer a PC or a laptop. Everything else is history.

What is important at all times is the timing and Identifying something that can change the world at the right time.

This is where the Top Leadership roles come into play. Identify the gaps and introduce the immediate action plan to make the best of the best.

Sure and not to discredit your observation, but what other observations have you made in the last 5 years that didn’t pan out? Regarding politics, sports, stock market, covid, or other tech trends? The evaluation can’t be looking back, and if you’re right about 1/10 things, would that warrant a $1b investment in each?

Parent comment doesn’t claim to have predicted the rise of smartphones.

If you truly had a bankable 10% success rate, $1B on each spin would be a steal.

This should be the perfect forum to ask this question considering the whole mission of Y combinator.

The thing is that we had small notebooks/agenda/notepad, then we moved on to PDAs when things turn digital, it's not a big stretch to imagine how smart phones should work if the hardware is there (what I read is that manufacturers was cheaping out and locking things down). I still believe that what LLM does best is analyzing natural languages and producing coherent (not necessarily true) output. And there are maybe business needs for that, but still have not seen a truly individual tool, like personal computing is (you can go on a desert island with a laptop and compute). I agree that executives is to predict and plan strategical responses. But so far, it seems to be only useful for those that like quick answers, even if it maybe untrue.

I remember reading about Xerox people forecasting the smartphones and tablets. We have a picture of their brainstorming and they used small pages to model a handheld computer. Science fiction was and still is full of more and more integrated computer with human. Technology is driven by imagination more often than not.

Curious how Bill Gates is still main cced? Does he still play a significant role in the control of Microsoft?

He's still influential behind the scenes - there was a recent article published about it: https://www.businessinsider.com/bill-gates-still-pulling-str... (https://archive.ph/VmtON)

> Current and former executives say Gates remains intimately involved in the company's operations — advising on strategy, reviewing products, recruiting high-level executives, and nurturing Microsoft's crucial relationship with Sam Altman, the cofounder and CEO of OpenAI.

> "What you read is not what's happening in reality," another Microsoft executive said. "Satya and the entire senior leadership team lean on Gates very significantly. His opinion is sought every time we make a major change."

If true it would explain why Microsoft keeps getting better while Google management seems to ignore the founders and keeps getting worse.

I don't think it's so much that Google ignores the founders, so much as the founders are ignoring Google.

I was at G when L&S stepped away, and it really just felt that leading up to that, they'd become completely disinterested. S, especially, seemed completely out of touch. And L just tired.

BG is a shrewd, cutthroat businessman on top of having a tech background. I think he's far different from L&S in that respect. But yeah it helps he hasn't disengaged as well.

Google would have probably been better off if Eric Schmidt had stayed engaged.

I don't like the guy's opinions on many things, but he was pretty good at keeping the ship headed in generally the right direction.

Or any direction? I couldn't tell what the direction of Google was if I had to.

What's the direction of Microsoft, Amazon, or Apple? They're too big to have one direction.

Google does have one direction, the same direction it's had since IPO, really.

Maximize ad revenue, to maximize share value.

Just all the other window dressing is going away.

Never liked Eric, but Google had such a clear direction and vision under his leadership. Pichai, feels like the accountants are running the business. It is ONLY about the next quarters results.

I don't think Larry and Sergey is on the same level as Bill Gates. Solo founders who have raised to and led a big tech company is a different breed. I would pick Jeff Bezos, Zuck and Larry Ellison on the same level of Bill Gates. They don't manage day 2 day but big strategies don't go through without their blessing. Even though you can say Bill does not manage, the top level executives will always think "what would Bill do" or "what would Bill think about this".

Bill Gates co-founded Microsoft with Paul Allen. If anything, Paul Allen was the instigator.

Yeah and Jobs co-founded Apple with Wozniak. Wozniak was instrumental to Apple's initial success but clearly it was Jobs who took it to the mat. Same with Gates/Allen, by their IPO in 1986 it was billg at the helm and Paul Allen doing something else.

I don't dispute Bill Gates might have been the business mind that drove Microsoft to its status today as the biggest software vendor to businesses on the planet. However, it's not because he was some sort of solo founder sigma grindset.

Allen clued Gates in on the upcoming microcomputing revolution when the Altair made the cover of Popular Electronics in 1975. And there are stories about both writing code on planes while on their way to sales meetings. So Gates arguably combined Woz’s technical chops with Jobs’ business acumen (even though I always admired the latter’s sensibility and vision more).

I think my phrasing of "solo founder" was not very good. I should have used another word.

Zuck is not a solo founder at all but what happened to the rest of the guys. Some says Google is what it is today was largely due to Eric Schimdt's involvement who is not a founder. The idea that all co-founders hold large influence in the long run of the company is not true. However, there individuals who essentially established the culture, the vision and the framework of the company.

The idea is that within the original or pivotal executive board there are people who even though left the company still holds significant influence in terms of grand strategy or vision. And those people does not include all cofounders.

Which of their products is getting better?

They’re executing better.

Microsoft has never made great products with some exceptions.

Microsoft makes “good enough” products that integrate smoothly. That’s what enterprises want and smaller vendors can’t provide by their very nature.

Some exceptions like Windows, Office and Xbox?

You're listing widely used products, not great products. Fast food is not great just because it's popular.

> Fast food is not great just because it's popular.

It's not good for you, as a consumer. As a company, it's a trillion dollar market, nurturing some of the largest companies in the world.

Even Bill Gates has admitted that a big part of their success was just sheer luck. There were many competitors that were pushing far better technology than they had but simply could not get a foot hold in the market.

Just the other day on here I mentioned BeOS/Haiku OS, the tech vision on that thing is astounding but it simply did not pan out into any sort of market success. Same with parts of Commodore, IBM, NeXT. MS did capitalize and manipulate the market to assist themselves and keep competitors out, but just becaused they won doesn't mean they had the best technology.

Microsoft intentionally and specifically suffocated BeOS. That was not luck.

Not a single person who uses Word and Excel daily would agree with you. Apple the “great product” company has been at it forever and so have many open source competitors and none come close to how good word and excel are.

This seems to be the problem with engineers: “I use eMacs on Linux so that’s the greatest product for everyone.”

Microsoft is expert at vendor lock-in. Your company got Office in the 1990s because competing office suites didn't work with Microsoft's operating system "for some reason". Now all of your accountants' macros only work in Excel and it would take thousands of hours to convert them, so no one can switch until a competitor has bug-compatibility with the highly complex and under-specified document format. But because companies can't switch, it sucks all the oxygen out of the room for competitors who are deprived of the revenue, patches and donations they would otherwise get with a larger user base. So they don't have the resources to achieve bug-compatibility and the status quo sustains, not because Excel is so great, but because Microsoft ensures that competitors don't get better.

In theory Apple has the resources to do this, but "Microsoft compatibility mode" and "backwards compatible forever" are anathema to their brand, so they don't. Google, by contrast, has actually made an effort and taken a huge bite out of Office market share. You might even say the product is better. But it's still not 100% compatible, and there are companies not willing to put their most sensitive private data on somebody else's cloud, so there are still plenty of companies stuck with Microsoft -- not because it's great, but because there is no great alternative.

But it's not just about compatibility, open source (or any other alternatives) are just simply worse in most ways (UX/feature parity/etc.) than Excel and Word etc.

Google Docs is not worse. LibreOffice is sometimes worse for the reason already mentioned -- companies can't switch to it when they're locked in, but user base is the source of resources for improvement via support contracts, donations and code contributions.

> Google Docs is not worse

It's certainly worse" than than the desktop office apps for some use cases.

> improvement via support contracts

Which might end up being effectively more expensive than paying for MS Office licenses.

> It's certainly worse" than than the desktop office apps for some use cases.

You can add "for some use cases" and make the sentence true about any product in comparison to any other product. That's not what makes something great.

> Which might end up being effectively more expensive than paying for MS Office licenses

You're now comparing two entirely different things. With a support contract you get support. It's something different than the product itself with independent additional value.

> You're now comparing two entirely different things. With a support contract you get support

I was thinking about you needing to pay someone to add new features/fix bugs or just help you to reimplement the workflows which you already have on Excel.

> any product in comparison to any other product. That's not what makes something great

I was thinking about many common workflows and general "power user" stuff that's very important for many professional users.

It's like saying that Apple Numbers is not "inferior" to Excel/Sheets because it's a different product and that it's perfectly sufficient (maybe even superior) for certain workflows, however that doesn't change the fact that it's objectively inferior from the perspective of significant proportion of people who are working with spreadsheets.

I don't know. I use the iLife suite (Pages, Numbers, Keynote) and prefer them over Word/Excel. Obviously I'm not a power user, but for the standard tasks you need from such tools I find them much better designed. Keynote in particular stomps all over PowerPoint for anything except collaboration, where Apple's "only the Apple ecosystem exists" mentality holds them back.

Excel is a great example.

The idea is fantastic. Programmers are too harsh on excel.

Unfortunately Microsoft haven't really updated excel in decades. They've added tweaks here and there, and made a really clumsy attempt at integrating python, but it's potential is shockingly untapped considering it's at getting on for twice as old as me.

In my brain, Microsoft is two companies at once. Old Microsoft and New Microsoft.

The split is categorized by Satya's focus and specifically MS's acquisitions after he took over.

This new Microsoft does very well (Azure, Github, Linkedin, OpenAI investment, Unix focus (WSL, C#/.net on linux)

However the old Microsoft is still around and in some ways just as bad as ever. Namely Windows.

I agree Windows is getting worse, but there are still lots stable products in old Microsoft, namely the whole productivity suites like Office, OneDrive, SharePoint, Teams, etc.

All of the new versions of those suck. They're absolutely awful performance and design wise. I don't know what the hell they're doing. It's this horrible mishmash of design decisions and it isn't even good looking. You can forgive a lot of usability issues if it looks fantastic, but it looks like a white screen with some random lines of text and feels like notepad via RDP over a DSL line.

In my mind everything they have acquired they are running into the ground.

Tell that to the massive increase in relevance and usage that Linkedin has seen, and the huge money maker that is now Github Copilot / Actions / Dev Environments.

This was sent in 2019, when Gates was still chairman of the board at Microsoft and involved day to day as technical advisor to the CEO. Post 2021 he has no official role at the company.

I found it curious he wasn't cc'd on the response from Satya.

I think Satya knows better than to waste Bill's time. His reply added no value to Bill and was done just to loop in Amy.

You can easily kill ambition or by requiring everything to go into slow processes, no freedom, and corporate politics.

If there's any ambition left, it won't be for that company.

How do such emails get leaked. Since the email is from top management with just few recipients, wouldn’t it be easy to find the leaker. Testing done of these private emails scares me a bit. If such top secret emails can be leaked, how about private citizens like me.

Knowing how something is done can be quite different from knowing all that it really takes to make it happen.

People actually use Mail in Windows 10?

Probably millions of people.

Us nerds need to habitually self-remind that we’re a sliver of the total user base and our experience and needs are not representative.”

I use Outlook, not super nerdy, idnk

If you’re reading HN, you’re so much nerdier than the average software user that the existence of super nerds is irrelevant

I think this was more a generalisation about HN as a whole than directed at you.

I did, it was a genuinely good little lightweight UWP email app. Now they've turned it into a bloated ad-ridden Outlook webview app.

Was gonna say, I wonder if he still uses it now that it is that Outlook thing :)

I’ve been helping people to Thunderbird. Most people are like: Mail>Thunderbird>New Outlook thingy. Good enough for me ;)

Don't get me wrong, Thunderbird is awesome for what it is (and the Mail thingy was kinda perfect as a lightweight mail client) but Thunderbird has some rough edges.

Like it took me long until I figured out how to show message size, filter by date and then sort it by message size. Of course, the Mail thingy wasn't able to do something like that, but I remember waaaaaaaaaaaaaay back, that Communicator (that Mail program which came with Netscape / Mozilla) was way simpler to use than Thunderbird. But maybe my memories are hazy as this was like ~15 - ~20 years ago.

I'm guessing that happened almost exactly a year ago, which is when I was astonished to find that they'd junked up their perfectly good Weather app with ads.

Microsoft does plenty of cool things these days, but it's bizarre how they're pushing these little Windows monetization efforts like it's some garbage F2P mobile game.

Microsoft knows most people aren't going to buy a Mac

I remember liking it when I found the app.

But then I had to use the web interface to do anything with filters, so when I had to reinstall Windows I didn't bother configuring it again. And now everything is on Teams anyway.

I don't know a single techie person who uses Windows (other than for gaming), let alone Mail in Windows.

But the other replies are right. We're not really representative of the entire population, especially if you consider the developing world.

This phenomenon of software engineers using MacBooks for work is a rare and primarily coastal phenomenon, at least in North America. Virtually all software engineering at utility companies, hospitals, state and local governments, and similarly boring but critical enterprise companies in industries like insurance happens on Windows on Azure using Windows or Microsoft oriented stacks.

>Windows on Azure

Lol, citation needed if you actually believe this.

>or Microsoft oriented stacks.

is a very complicated way to say Windows desktop.

Re windows desktop, no, it's not a complicated way of saying that. I am talking about, for example:

- ASP.NET Core backends running on anything (e.g. Linux)

- Angular or Vue frontends running on anything

- applications using SQL Server databases running on anything

- applications targeting on-prem Windows servers or VMs or, sure, Windows desktops

- data / reporting systems that interact with AAS

Apparently I phrased my response poorly. I was responding to the narrow context of the idea that devs aren't using Azure VMs instead of Macs. I was not disputing the popularity of Azure.

Microsoft sales is surprisingly good at selling Azure to c-level at companies where the primary output isn't software, based on existing Windows/Office deployments. Unfortunately also true for Teams. Retailers are also often opposed to AWS because they consider Amazon a competitor in its entirety.

Just so you know this is a very ignorant comment. I don't mean that as an insult, but literally you are ignorant of the facts. If you care you should use this opportunity to educate yourself on the state of the Microsoft stack.

This is something "bothering" me as well. It's either Mac or Linux if you want to be "cool" in tech.

I am always like "wait do you not write any Windows software or is it all on the Web"?

Maybe I am ignorant and stuff and maybe it's just bothering me. :)

Windows computers tend to also get the most overreaching invasive MDM / Endpoint Slowdown Software.

Can't say for Windows, but at work I manage a bunch of iPads with JAMF and it's really comfy to use.

I write applications for Linux, MacOS, and Windows but I use cross-compiling frameworks like Wails or Tauri to do so. If I write in Windows Forms (old Xamarin forms) then my application only works on Windows. MacOS is similar. All of my development occurs in Linux as a result.

There's a lot of work that is neither Windows software nor web. And much (most?) of that runs on Linux.

I find this whole conversation amusing. Growing up and going to school in the 2000s, Windows was the default choice, and I felt like I had to fight to use anything else. Now the conversation has reversed to some extent.

I, personally, would not chalk that up to a phenomenon. Operating System selection largely depends on the work you do paired with the industry you're in. Some extrapolations:

I work in systems engineering for web technology companies. Most of the applications I build run on Linux so it makes sense to write on a Unix like OS. I could use Linux or MacOS, but Apple has a strong preconfiguration and leasing pipeline so usually the companies I work for offer MacBooks Pros.

When I was building software in the US South I would have to look out for companies that were Windows shops because I don't do Windows systems engineering and there were an abundance of shops that did before the .NET Core rewrite that enabled you to run on Linux. Those shops would've definitely shipped me a Windows laptop. Anecdotally, I buy servers out of a DC in Houston and nearly all of them come with Windows Datacenter edition. Most of those companies fell into certain industries that didn't include what I typically worked in.

That's all to say, region can be roughly correlated, but it's not the actual reason. It largely has to do with who sold who the software stack they have trouble moving off of today which influences everything else.

Note that my comment was observational, I'm not saying companies should use Windows, and I'm not even talking about reasons for using one tech stack or another. I'm just reminding the parent that this association of Windows with "non-technical" is nonsense -- lots of extremely talented folks work on difficult and important problems and critical systems on Windows, targeting Microsoft stacks. Further, it seems likely that most software is written on Windows.

Yeap, that's all fair. I was mainly poking at your word choice of "phenomenon" because it can be reasoned. I'm not quite sure about your last sentence but that's an entirely different discussion with much more noise than signal to parse through.

"Phenomenon" just means "thing that can be observed". It doesn't need to mean something that is unusual or surprising.

Huh; according to Merriam-Webster the word has basically lost meaning: https://www.merriam-webster.com/dictionary/phenomenon

The first definition is how you and OP used it, the second is how I used it (exclusive of things that can be "sensed" as opposed to reasoned).

I think your usage is actually more like definition 3.

I both game and program on Windows. I don't know what people are crying about. I've got Docker. I've got WSL. I've got a high quality IDE (IntelliJ). Everything works and runs great.

I also run Debian at work and that's also perfectly good for writing code, just not gaming.

But no, I don't use Mail in Windows. I've been using Gmail since it came out in 2004....and was perfectly happy with it until just this second. I just found an email from 2004 wherein I had emailed myself a project, and Gmail has blocked it because it thinks it's a virus. The solution? Export the EML and open it in Mail on Windows. Funny.

> I don't know a single techie person who uses Windows (other than for gaming)

I'd say that Windows actually has some nice software, like MobaXTerm: https://mobaxterm.mobatek.net/ which in my eyes is better than Remmina or pretty much anything I've found on nix, short of just running the same thing on Wine.

WinSCP is also pretty cool, albeit nothing particularly special: https://winscp.net/eng/index.php

PowerToys (and other customization software) also make the OS feel more pleasant to use, especially with something like FancyZones which feels nicer to use than the window snapping in XFCE or Cinnamon: https://learn.microsoft.com/en-us/windows/powertoys/

WSL2 actually seems nice to use and even Hyper-V is pleasant.

The vertical taskbar in Windows 10 worked better out of the box than my current customized Cinnamon desktop.

Oh also the task manager is really nice and the Linux remake does tend to eat resources: https://github.com/KrispyCamel4u/SysMonTask

There's probably more nice things that someone can say about Windows and personally I don't mind doing development on it because most of my software works on it anyways (VSC, JetBrains IDEs, GitKraken, Docker, browsers etc.), but Linux distros do feel better for that particular type of work otherwise.

On the other hand, even with Proton, I still enjoy gaming on Windows more, far less of a hassle and curiously the graphics control panels seem to mostly only be available on Windows and something like CoreCtrl on Linux isn't always good enough (e.g. if I ever want to set a power limit for the GPU easily).

I really don't want to deal with Windows 11 though (which is inevitable because of updates and also work computer) and I have very few positive things to say about Windows Server, however. But hey, even .NET now works on RHEL/Ubuntu and other popular distros, so Microsoft tech stacks also feel decidedly more sane, in addition to something like C# just being a decent language in general.

That said, Thunderbird runs everywhere and does so well, so for me, it's the obvious choice.

Regarding techie people not using Windows, you’re living in a bubble.

John Carmack, ryg, people making demoscene, are the hackerest of the hackers, and they use Windows mostly. These people derive from the Wirthian pascal heritage. Unix people would not know.

Like the other replies are saying: you're in a bubble, like most Apple users.

MacOS needs third party apps to not be incredibly frustrating, and often the apps are subscription based. (Window management, clipboard history, and power management are the big ones - even if you go with a free option like I do, it's not good that the base OS is missing essential features in these areas.) There are many little annoying things that can't be easily disabled, like Apple Music launching every time you connect your bluetooth earbuds. There are incredibly bizarre UX choices, like not having "pin" in Finder be a right click option.

As for development, Windows + Powertoys can do everything MacOS can better, plus game (and game dev). For anything Linux I can use WSL2. I can natively develop any kind of software for literally any non-walled-garden platform on Windows (and use a VM if I'm forced to make an iOS app, god forbid). Why would I intentionally choose to limit what my computer can run and what development I can do?

I'm forced to use a Macbook for work, but would never use it if given the choice. MacOS is for people that value style over substance and non-technical people, just like everything else Apple.

> I'm forced to use a Macbook for work, but would never use it if given the choice. MacOS is for people that value style over substance and non-technical people, just like everything else Apple.

You’re obviously free to prefer whatever platform you please for whatever personal reasons you please, but the conclusion that macOS is for non-technical people is insanely ignorant.

I said "people that value style over substance and non-technical people," you appear to have read "non-technical people."

>ignorant

In fact the only techy people I know that enjoy MacOS seem to be ignorant about what they're missing. I've had multiple people be very surprised when I showed them the clipboard manager built into Windows and explained how insanely slow switching between windows over and over to copy and paste multiple things on MacOS feels without it. They genuinely had no idea because they, again, never leave their Apple bubble.

Like the iPhone the one thing Macbooks objectively have going for them is (overpriced) good hardware.

I did, till they took it from my claws and tried to convince to use that Outlook thing. :(

It was kinda perfect for peeking into mails, deleting a few, answering some other and stuff. Not a great client to do your business communication, but to drop someone a quick mail, it was neat.

Funny to see Satya using it. You'd think he'd be on the full Outlook. Maybe it's a case of dogfooding.

To me the biggest news is not that he is using Mail on Windows but that he is using Windows. You wouldn't tell that the CEO of the company is using the product given how badly they let it decay.

I’m more surprised he doesn’t seem to know basic comma rules.

It's an interesting read. Especially if you consider that all it needs today is llama.cpp and a model of your choosing from huggingface.

(Okay Okay, granted it takes a bit more and someone had to train those models, but point it, the fear from way back seems to be....dunno.)

If you can't train the model, you don't really have control over a product based on it.

Previous submission from 4 days ago:

https://news.ycombinator.com/item?id=40226114

I don't understand why large companies use email for such important topics knowing that they can be required by law to disclosure them in the future.

I know that large companies have email retention laws, but why not just keep sensitive topics on Signal or similar?

becasue it can be interpreted as acting against the company in first place

They are definitely doing more of that, but it's not always a get our of jail free card, and may actually put them at legal risk for destroying documents:

https://www.theverge.com/2024/4/26/24141801/ftc-amazon-antit...

https://arstechnica.com/tech-policy/2024/05/judge-mulls-sanc...

Not being allowed secrets by law is such a weird concept

It's not weird for a publicly traded company.

I find it weird that these get released as public records, but otherwise, laws against destroying evidence seem fine. And laws requiring disclosure of documents in a court to determine wrongdoing also seem fine. And laws requiring proper record keeping of decision making at large companies also seem fine. Particularly when the burden of record keeping (ensuring emails are not deleted) is not very large.

They write these emails knowing that they'll be public record at some point. The audience is as much internal as it is for the history books. The details get hashed out offline, while the record gets preserved as an email

Why are you so sure of this?

Retention laws are neutral towards technical protocols. They apply to instant messaging and electronic chats just the same.

The contents of these particular emails is actually not very sensitive, and the courts do allow for redactions of more sensitive things (e.g. the actual numbers involved).

Things that should not be discussed in writing happen in person. Meeting rooms, offices, hallways, golf courses.

The emails are written knowing it might be public. And at the C suite, emails are to be very intentional.

Reply on news.ycombinator.com