Hacker News new | comments | ask | show | jobs | submit login
GitHub Is Microsoft’s $7.5B Undo Button (bloomberg.com)
351 points by danso 8 months ago | hide | past | web | favorite | 353 comments

Microsoft can not buy goodwill with me, personally. I worked at the company and was privy to "strategic conversations" whereby they were looking to regain trust with developers to get them back from Android and Google to Windows stack.

Microsoft practically forced us to use Windows stack for anything and everything, even when there were much better tools. Under Satya, it was better, but the emphasis is still very much on getting, as Steve Balmer said "Developers developers developers".

Microsoft knows its being challenged in the platform game and that developers have migrated. But having seen the inside of Microsoft corporate, legal, compliance and policy I don't have high confidence that this isn't another PR project destined for death of the startup.

It's also a terrifying lesson about the fragility of the open source movement.

I kept reading, waiting for the insidious part. So what you're saying is that Microsoft has a company wide strategic vision to improve its relationship with developers?

Did you leave some part out?

Nothing where they are going to gas all the developers or something. Nope.

It's more about priorities and intentions. Microsoft doesn't care about the developers. It cares about having the dominant platform. It recognizes that its image makes that difficult.

The GitHub acquisition: Microsoft didn't suddenly get more beautiful. It put on a mask. It will kill GitHub - or at least let it flounder - if that's what's best for its business. It will push Microsoft technology into the GitHub user base if necessary.

Developer relationships aren't an end in themselves - something Microsoft wants because its good. Developers are a means to an end, and its Machiavellian in its application of corporate strategy to achieve this end.

Microsoft would happily buy Ubuntu Canonical if they thought it would win them either a market advantage or a way to push Microsoft stack into the Linux userbase.

Some people probably don't have a problem with that. I do. But maybe I've got a chip on my shoulder from seeing this go badly so many times.

> Microsoft doesn't care about the developers. It cares about having the dominant platform.

"It is not from the benevolence of the butcher, the brewer, or the baker that we expect our dinner, but from their regard to their own interest." (Adam Smith.)

I don't trust benevolent people. I trust self-interest. I don't need to offer my undying allegiance to Microsoft - I can like them as long as they do things that benefit me, and dislike them when they don't.

I don't trust self interest.

Especially not in my butcher, brewer, or baker. Pure self interest gets you McDonalds, Bud Light, and Debi cakes. I consume exactly none of those things and have no desire to consume any of those things. They taste terrible, are bad for me, and the companies themselves are generally bad for their communities and the world.

A good butcher, brewer, or baker is driven by craft first - care for their product, care for their customers, and care for the environment. They devote themselves to their craft. Not out of self interest ("I'll make more money if I do it this way.") but out of sheer joy of craftsmanship and doing good.

I have known few producers who's products I would rate trust worthy who weren't driven first by craftsmanship, care for their customers, and care for doing good in the world. Self interest (of the "I need to make a living" form, not the "I want to be rich" form) in these situations is consistently a secondary interest. For these crafts people, if they can't meet their self interest, they simply stop crafting rather than pollute their product.

That's who I trust. Never the behemoth driven by fiduciary duty and self interest.

Adam Smith's philosophy is nearly 250 years old and predates the modern industrial revolution. It's time we stop putting it on a pedestal. We've learned so much about the way people, markets, society and people in markets and society function that he simply had no way to know.

I've met a large number of great craftsmen who I would rate trustworthy and who are driven first by craftsmanship and for whom self-interest is a secondary interest.

I'm privileged to work at a great shop with some 20 other people for whom profit is just a small part of the goal. We call fiscal health "oxygen": It's something we need, but if it's present it should be allowed to fade from the forefront of our minds, instead focusing on more rewarding targets.

This works great when you're working with a small number of people, and when the company is driven by and owned by a very small, close-knit group of people - ideally, you can count them on one hand. This allows non-fiscal values to be communicated. Unfortunately, communication is difficult among large groups, and the easiest and most common means of communicating values is in dollars.

I'm not sure that a $7.5B company can maintain this idealism. "Behemoths" are going to be present whenever you have economies of scale, and post-industrial-revolution we don't have craftsmen who are butchers, we have vertically integrated behemoth butchers with million-dollar tractors growing the grain that rolls in on conveyors to feed the cows that are butchered and packaged by automated machines, so each package of beef has less hands-on time than it takes the craftsman to wrap the finished product in paper and twine. The craftsman is going to be relegated to the edges of that industry.

Agreed. And we're suffering all kinds of negative consequences from this system. I think one of the great economic questions facing our generation is "How can we keep the efficiencies of scale we've gained while reintroducing the craftman's motivations and ethics that have been lost?"

I have some ideas towards that direction, but make no claims on a definitive answer.

Worker cooperatives seem promising, since they change the average worker from "grunt" to "owner / director". It changes the incentive structure around what gets communicated and how as a company scales. Worker cooperatives also change the incentives around growth itself -- disincentivising growth beyond a certain point because each new worker waters down the voice and control of the existing ones.

AI driven automation and robotics, in a context of worker cooperatives, presents the possibility of keeping our production efficiencies (that used to require huge scales) while maintaining small organization sizes. We're not there yet, but it seems like we're not too far off.

I think reconfiguring the economy around these two ideas, presents the possibility of a good answer to the question posed above. And has lots of extra benefits beside (like building democracy into our economic system and not just our political system, and creating a system that more equitably distributes wealth by nature).

>Especially not in my butcher, brewer, or baker. Pure self interest gets you McDonalds, Bud Light, and Debi cakes. I consume exactly none of those things and have no desire to consume any of those things. They taste terrible, are bad for me, and the companies themselves are generally bad for their communities and the world.

Whoever you do buy your food from has recognized there’s a portion of the market that values what you do and is happily selling it to you, out of their own self-interest.

I, for one, like McDonalds and sometimes prefer a Bud Light.

You've got it backwards. They are able to make a living happily doing their craft because there is a portion of the market that values their craft. They didn't choose their craft because they recognized their was a market for it. They chose their craft out of love of the craftsmanship.

Arguments like this fundamentally misunderstand human psychology. And this is why the economic model of the "rational man" is fundamentally flawed.

Lol this post shows an amazing savior complex. These poor altruistic artists just want to work for meagre amounts and no profit but thankfully heroes like you appreciate their craft and are unwilling to let them peddle their wares for anything lower than a healthy profit margin.

It has nothing to do with altruism and everything to do with the fact that most people are not driven by the hoarding of wealth. Most people recognize the need to make a profit because they need to survive and support their families, but it is not their primary motivator. If it were, there'd be millions more entrepreneurs than there are, and few people willing to work endless factory or desk jobs. For most people, they are primarily motivated to do things they enjoy doing. For the sorts of crafts people I'm talking about, those things are engaging in their craft.

Altruism has nothing to do with it. Although, psychological research has shown that economics has it wrong: most people are basically altruistic, not basically selfish. It's how we can be such a social species. But that's a different discussion for another time.

That’s totally backwards. Most people do their jobs because it pays the bills. Most people don’t have careers, with some glorious internal narrative in which they’re the protagonist. They have jobs. If they’re lucky or good spirited, they’ll find satisfaction or even enjoyment in their work. But they work because it gives them the means to get what they need and do at least some of what they really want to do.

Tho sarcastic, I upvoted you because you’re on target.

People who care about their craft are still driven by self interest: it's just that their internal value/utility puts a higher weight on the craft itself: since they get more out of the process of creating things that have craftsmanship, they are not distracted by the opportunity to make 5 cents more per loaf (or whatever).

> That's who I trust. Never the behemoth driven by fiduciary duty and self interest.

Trust is orthogonal to craftsmanship.

> Adam Smith's philosophy is nearly 250 years old and predates the modern industrial revolution. It's time we stop putting it on a pedestal.

It's less a philosophy than an understanding/insight regarding human behavior. You might as well say "it's time we stop putting newtonian physics on a pedestal", to the extent that both are reasonable models of human behavior, and are helpful in many, many real-world situations.

Arguing that people "shouldn't" behave according to Smith's expectations is going to be about as successful as arguing they shouldn't fall when they jump off a ladder.

> I consume exactly none of those things and have no desire to consume any of those things.

Nor do I, but many people do, and that decision makes them different than me, not less than me.

You are conflating self-interest with greed. That is a great mistake in trying to understand Adam Smith. They are not the same thing.

Values, ethics, morals... are part of self-inrerst for lots of people.

Sure, you could argue that, and maybe Adam Smith was at the time. It's been a long time since I've read him, I'd have to go back and re-read it to decide to what degree I think he's making that argument. But that is not how most people understand self interest in an economic context and that is not how Adam Smith is mostly wielded today (see the post I'm responding to for an example). Most people understand "self interest" to be "financial self interest" of the sort that supposedly is the primary driver of the modern economic idea of the "rational self interested person" who operates in markets.

Further, the vast majority of stock corporations have no values, ethics, or morals. They have only the maximization of stock holder value -- which is to say: greed. We know this from piles and piles of data on corporate behavior.

People and organizations of people are very different creatures and behave in very different ways.

Of course: that is why Adam Smith speaks of people and not corporations. And I was speaking of people as well.

Corporations are a different subject matter.

Fair enough. However, the context of this discussion -- and the modern economic context -- is made up primarily of corporate actors.

Which is one more reason why the application of Adam's philosophy doesn't make sense.

I like your answer. But you say : "are part of self interest". In which way ? What puts self interest above those traits to make it encompass them ? 'cos I'd argue that if values, morals, ethics are part of self interest, then self interest is not much more than "personality". If so, self interest is not a concept that is very useful in itself... (no I didn't read A.Smith)

Evidently not to the extent that we can rely on self-interest as a viable regulatory mechanism.

then it’s practically a tautology

Most of western philosophy is just a bunch of tautologies. That is what makes it interesting.

how do you figure?

Careful, Friedman and Hayek basically took Smith's theories warped it beyond recognition, and it is these two men that most of western economics has been tirelessly focused on this last few decades.

Believing that humans are merely information processors acting at all times only in their 'rational self-interest' is an incredibly narrow and simplistic view of humanity.

Well, exactly, that is in another way, what I am stating: rational self interest is inherently unmeasurable because "rationality" is not something that can be objectively counted. So at any time one should assume that people act in their "self-interest": what at the time they deem "better" (in whatever their will decides is "better" at that point). I am not arguing that people never change or never act wrongly or never do things harmful to themselves. I am arguing that if one wants to "objectivize" their actions then one should assume that their decisions show their interests and their values at each time.

I'll never understand the middlebrow hate directed at America's most iconic brands. McDonalds made real business ownership a possibility for an entire generation of people that would have never had the resources to do something on that kind of scale. I actually like McDonalds' food, it's simple but endearing.

Bud Light might not be the most flavorful beer, but have you ever tried the original Budweiser? You know it's actually pretty darn good. And when you compare Bud Light to truly bottom-market beers, you begin to see the value of Budweiser's offering. Also, I don't know how many foreign countries you've visited, but a lot of them really desperately need a Bud Light just to give the local brewers some real competition. Some of the beers I had were just bad.

I can understand if you don't want to consume mass-market products, but really, you shouldn't just dismiss something because it looks big to you. These things solve real problems that people have, at real scale. Your local brewer might have great beer, and your local butcher might have great meat, but you know what would happen if everyone was at the mercy of local business? Everything would be 4x more expensive and wages would be half as much. Just like it was before those brands emerged.

I’d argue your craft beer, artisanal cheese, and hand-made distressed-wood furniture are all made out of self interest. You’ve simply chosen to believe the marketing positioning statements delivered by their million-dollar companies: that they operate out of some passion for a craft rather than increasing shareholder value.

I think GP's point is that the nature of MicroSoft, as a platform, isn't lined up with your interests. They aren't looking to be developer friendly--that is, present you, as a developer more and better options.

Since MicroSoft is a _platform_, their interest is to _limit_ your options to MicroSoft so that you aren't developing and enriching competing platforms.

GP, if I am understanding correctly, is implying that their purchase of Github will either go to waste (Github can continue, but with no new resources from MS, and will slowly dry out) or MS will leverage it to push their own tools while limiting how developers can interact with it for non-MS code bases.

Trusting self interest is fine, but trust doesn't always mean _trust in the good_. Trust means we can visibly see what their intentions are. The butcher depends on our money, therefor we can trust that they will provide us with a good service so that we return. MS depends on us using their platform, therefor we can trust that they will limit our options of which platforms we develop for.

This is getting tinfoil hattery.

Microsoft sells a platform. The success of that platform is based on convincing users and their agents to buy the platform. The platform is appealing because developers make money by creating apps or it. Has Microsoft done awful and possibly illegal things to competitors? Of course. But they've been the best for developers. Best dev tools, most powerful and widely deployed OS, at least in the PC era that made them famous and infamot.

Historically, Microsoft made a good platform. People wanted to use it. Then Microsoft consistently, over a period of years, in several different directions, tried to use the fact that people used Microsoft's platform to make it harder for people to use other platforms or products.

That's not tinfoil hat stuff. It's historical fact, and I was around when it happened. They tried to, in their term, "leverage" their OS dominance. But on the receiving end of that leverage, it feels like being forced.

> But they've been the best for developers.

Microsoft has screwed developers on their own platforms too, when they didn't fall 100% in lockstep with Microsoft's whimsical decisions and about-turns.

> Best dev tools, most powerful and widely deployed OS

I strongly disagree they built the best dev tools or the "most powerful and widely deployed OS", but I'm going to talk about a more important point here.

There's a difference between making a product that is better that everything else, and preventing others from making a product better than yours. When you let one maker — irrespective of whether they make the best product or not — bully and cheat their way into preventing other makers from competing, you are and shall be left with a sub-par product.

One example: Microsoft single-handedly held back the web for close to a decade, all so it could continue to sell Windows licenses. As a result, the world suffered.

There's a lot of progress that was stifled and/or killed off — some of which we may never get again — by Microsoft, just to protect its OS, by abusing its position on the market.

It wasn't their dev tools or their "most powerful and widely deployed OS" that made them infamous, it was their own actions.

> I don't trust benevolent people. I trust self-interest. I don't need to offer my undying allegiance to Microsoft - I can like them as long as they do things that benefit me, and dislike them when they don't.

Sound logic, were it not for the fact that you may be trapped by the time you dislike Microsoft. Given Microsoft's history, it's a risk I will never take. I don't trust them, never will.

We nearly avoided a monolithic web in the browser wars. Microsoft pushed, successfully, Internet Explorer to be the only game in town. You didn't write websites against the standard. You wrote them to run on MSIE.

Mozilla, Google, the iphone and MS's hubris made IE 6 so noncompetitive that the web evolved past that blockage. It was luck, and we must recognize the near miss. The open Internet could have ended up closed and a lot worse. We almost ended up writing webapps in Activex.

Microsoft successfully pushed internet explorer to the front by making it cheaper and more accessible than Netscape. If they didn’t, the internet may have still lacked competition today and you’d still be paying money for Netscape. The foresight and competition that Microsoft brang to the browser market was probably the most significant factor for the adoption of the internet by the whole world back in those days.

I never, ever paid for a version of Netscape. Hardly anyone did.

I - like everyone - paid a huge cost in wasted time and non-existent security when I tried to use IE or develop a site compatible with it.

If you never paid, well that’s because you entered the scene after Microsoft’s commoditising effect of the browser market had taken effect. You now had a choice.

You've got your history wrong. Free browsers were always available. Mosaic first, Netscape and then Firefox, with plenty of other lower market share offerings. Microsoft didn't commoditize anything.

Netscape was basically shareware - it cost money, but hardly anyone paid for it.

Since shareware is kind of an alien concept these days, many people remember Netscape as "always free" but it was not really.

Many users would have obtained Navigator from ISP and magazine cover disks etc. So those copies would have been paid for as part of the price of the magazine or connection.

I trust those who exhibit exclusive self-interest to throw me under the bus the moment it is good for them. The term for it is selfishness. And you can usually count on a large corporation to be selfish sociopath. It is the job of society and the public to constrain the selfish they do not harm others through their greed.

the problem here is that the self-interests of github and microsoft might differ fundamentally.

github wanted to make money from their paid services to create a profit. microsoft couldn't care less about the money paid github accounts make (as long as it's not hemorrhaging too much), it's unlikely they'll get a ROI from paid accounts in anyones lifetime.

they need github to increase their earnings elsewhere, e.g. by getting developers to develop for the microsoft store or for their cloud services. so my guess is they'll at least start advertising their tech heavily on github and add features integrating their own tech while omitting the same for rivaling corporations.

the old github didn't have an interest in limiting integration with other platform providers, as this would decrease their usefulness for potentially paying developers.

FOSS will interpret Microsoft as damage and route around it.

GitHub is going to turn into the FOSS version of the MySpace wasteland. I'm sure GitLab and other alternatives are seeing a stampede of new sign-ups right now.

The question isn't really what MS will do, because - as is often the case - it won't matter.

MS bought Nokia Phone, trashed the company, flailed around dramatically for a while, and in the end nothing much changed in the phone space - except that WinPhone hung on for a year or two longer than it would have otherwise.

MS bought Skype, trashed the product, bolted it awkwardly onto Office, and in the end nothing much changed, because anyone with any sense moved to an alternative.

It's going to be the same with GitHub. There will be drama in the short term, but in the longer term MS will receive little or no benefit, except for maybe some closer integration with VS.

The money will simply be wasted.

The real winners will be the investors and owners of GH, who have walked off with a big pile of MS stock they can convert to money and use to buy nice things.

Looks like a pretty strong exit from GitHub to GitLab.

Just hope most go to a common place which appears it will be GitLab.

Just hope most go to a common place which appears it will be GitLab

You know that Gitlab is also VC backed, right? They have to exit in the same way as Github did. It’s baked into their DNA. The only question now is who buys them. What will you do if that’s Oracle?

But GitLab has an open core and GitHub is closed source.

that doesn't matter much, as the platform is only one side, while the hosting/service angle is the other. ultimately, hosting your code yourself is the only safe option, but of course it comes with its own downsides.

The butcher, the brewer and the baker have no ability to alter my biochemistry so I can't consumer other people's meat, beer and bread. So I am unafraid that they will harm me with vendor lock-in.

Microsoft, on the other hand...

> The butcher, the brewer and the baker have no ability to alter my biochemistry so I can't consumer other people's meat, beer and bread.

Yet. :-)

You don't trust the creators of Python or Linux then, eh?

Feels like a page out of The Dictators Handbook:

> First, politics is about getting and keeping political power. It is not about the general welfare of "We, the people."

Or I guess here, 'we the developers.'

I don't think Microsoft is at all alone in this spot. Github got a huge payout from this. It benefited them. We don't think of them as a Microsoft style company, but they made the same decision we might have seen in reverse. They didn't sell out before, but now the payout was significantly high enough. It's the old "You'd be stupid to turn down that much money" line from Thank You for Smoking.

But keep in mind, Adobe, Oracle, Intel, Amazon .. they'd probably all do something like this if some essential backers on the board thought it was in their best interest, with the goals always being around maximizing profit (so you can pay all those CFOs, CTOs, board members and controlling stock members). The people who make these decisions care about staying in power .. and are most likely very far removed from actual development .. or their userbase (in this case the same thing).

Just look at what happened with a userbase of gamers and the Facebook purchase of Oculus.

Oculus is vaporware, right? I don't see anything that encourages me that whatever thing they were aiming at will be a thing?

Vaporware is software/hardware that is never actually manufactured or released to the public. Oculus has released the Rift, the Go and (with Samsung) the Gear VR, which are all available in stores now. All are successful, if relatively niche, products performing as intended, which is the opposite of vaporware.

It's not vaporware. It's available for purchase, and it fulfills its advertised purpose as a video game peripheral. The ecosystem of available games could be better, but it's far from stillborn.

If I'm wrong, I'd like an example of something that justifies the hype

What have you tried so far? Doom VFR Edition? Skyrim VR? VR porn?

VR’s hype is easily justified.

> "its Machiavellian in its application of corporate strategy"

I'd say, rather, simply, "it's corporate in its application of strategy". Which is to say, Microsoft is a business.

As someone with no real stake in this deal, why should anyone care whether Microsoft kills Github or not? There are multiple alternative Git hosting competitors and switching costs are low.

I don't care if they kill Github. It's the reverse. I'm worried that they will be successful.

My main concern (having spent about a decade working in MS shops) is that they will try to create a divide between "professional" developers and "hobbyist" developers. Microsoft's fear has always been that their next major competitor will be someone who comes out of their garage. They have always tried to control the programmers so that these programmers will be dependent upon Microsoft. I was literally forced to use Visual Source Safe on more than one occasion because the company I was working for had a contract with MS which prevented us from using any other source code repository (in exchange for discounts on other MS development tools). At the same time, Microsoft did not use VSS internally, because it was horrible. Similarly, we were always playing catch up with APIs. MS would be using them for 18 months or so before any of the rest of use would get any documentation for them (Ha ha! Source code??? Ha ha!)

So my feeling is that this is the embrace part of embrace, extend, extinguish. If I see an effort to rewrite libgit2, or something similar with extensions, then I will be very worried. The reason I can imagine "Don't worry, it will support the git protocol and any hobbyists will be able to make contributions, but if you want a professional interface to Github it is available on reasonable terms (prick your thumb and sign here)" is because I've seen it many times before.

It's not that MS is the only company that pulls this nonsense. Apple is consummately skilled at it: join our fold and have a privileged place in the walled garden.

Free software is intended to level the playing field. As much as MS has improved, I do not believe for an instant that they wish the playing field to be levelled. GitHub, though proprietary originally had no position where it would make sense for them to try to control developer. Now they do.

I'm not entirely understanding your point. You state that Microsoft's fear has been that their next major competitor will be an independent/hobbyist, yet then go on to describe how they've made life great for these guys while simultaneously putting the squeeze on more established studios. And I generally agree with this, though I don't think it's particularly nefarious - it's the established studios that can afford to drop 4 figures a seat a year on licenses - by trying to charge that to independents, they mostly just push people potential customers away from their toolchain.

A really great example of this is the fact that they now give away Visual Studio Professional (rebranded to 'community') for free and even allow unrestricted commercial development/usage for independent studios. I don't recall their definition of independent precisely, but it's something like <= 4 developers, with gross revenue < $1 million/year - the exact definition is similarly fair.

The cost to copy a repo over is low, but you lose all the social aspects. You don't copy over forks or stars. People who want to fork your repo and make pull requests need to sign up for a new service. Old links to your repo on GitHub won't work anymore.

I copied my personal projects to GitLab today and that was very easy to do but I'm not sure what to do about the GitHub version for the one that had a little popularity.

I did the same. I'm replacing the README on GitHub with "Moved to https://gitlab.com/... and a description of what the software does. If people finds it interesting they will follow the link.

A fork between services becomes more complicate but not so much: git pull from GitLab and git push to GitHub or whatever. Everybody knows how to do it.

Right, one loses the "fork" link between the two repositories but really, who cares? We're writing software for work or for a cause, we shouldn't be playing a game of collecting stars and forks.

Even when I disregard the inate desire of humans for popularity, forks and esp. stars are a sometimes-useful signal for the size of your user base. If I have 10 repos with 1-3 stars and 1 repo with 100 stars, that is useful information to consider when planning the next project.

I'm not worried about myself with the stars and forks, I'm worried about the people who starred and forked my repo and the costs that moving my repo has on them.

For the stars, they presumably did it to get update notifications. I guess they'll get a notification if I update the README, and they can decide what to do about that. I'm making things just a little bit harder for each of these people though.

Forks are even easier to maintain than you're describing. You just need to do a change the url for your upstream. The issue for me is that I have several forks of other peoples projects and in almost every case it's because that's required to make a pull request. For each of these projects that moves, they're multiplying the small effort by the number of people who need to move their forks and/or learn how to make pull requests on the new service.

You sound like gitlab is the "right" choice.

Maybe, or maybe there is an emerging space for a service that would count stars, forks, etc on all versions of a repository and aggregate the values. But I would hate another centralized service.

Isn't GitLab a centralized service?

This could be a kind of git search hub.

This is in part some naming annoyance. Bitbucket let you once use your own domains for repos. Which was nice as you could migrate if you wanted.

The forks thing could be tackled. With a notification, something like I forked you. Or a code search engine which is clever enough to detect and link repos.

The user management gubbins, is the normal irritation of user management and access control.

The stars part is already a bit of a horror across all software - not knowing how to gauge the popularity and quality of a project. Nevermind searching for it.

I'm not the biggest user of these silo repos, but I do appreciate that thin barriers to entry, and improved usability or a good usability wrapper, is charming.

I don't mind not getting notifications or having my stars tracked. I just want people to be able to easily star/fork my repo if they want to keep track of or personalize it. That's harder when it's a moving target.

You could pull in all forks also.

I don't have control over forks people make of my projects. Luckily, this concern is more of a hypothetical for me, as only one person has forked one of my personal projects on GitHub. I'm much more concerned about stars and hyperlinks no longer leading to the real repo when people follow them.

Do you think the founders of GitHub, who after the sell to Microsoft will be the 2nd largest owner's of Microsoft stock, started GitHub because they "cared about developers"?[1]

It's always about the money. No one "cares" about developers not even your own employer. They only care about making money and we are all just a cog in the machine.

[1] https://timesofindia.indiatimes.com/business/international-b...

> Microsoft doesn't care about the developers. It cares about having the dominant platform.

It'll need to start caring to a degree, and catering to them, in order to achieve this

I think the original comment is trying to convey is that perhaps developers should be wary of goodwill.

Microsoft have done this in the past. They actively eat the browser market share with IE and left it alone after being the monopoly in web browser. This forced Netscape to become Mozilla. They also tried to be the gateway and believe OS is the gateway for all program and was left in the dust with web app and web search engine. Their past modus operandi was always to stay a monopoly not to create a good environment for developers.

And I think it's a good thing to keep in mind because microsoft's past actions we should all be wary and keep in mind when we decide to do anything with Microsoft.

Google is pretty much doing the same thing with Chrome. Opera has switched to its rendering engine and Firefox's share is dropping (after Google pumped tons of money in them and used their work to enhance their own offering). Gecko is barely on the map while Google's render can be found in newer versions of Opera, Vivaldi, and others.

I mean if you go back far enough, Microsoft did this with IE. IIRC, they licensed a bunch of stuff from NSCA Mosaic, who thought they were going to make some good money off a percentage of sales. IE was originally part of the Plus! pack which you had to pay for with Windows95, but then MS released their browser for free.

> Firefox's share is dropping

Firefox's share was never very high to beginning with, because Mozilla doesn't have the resources to play the vendor bundled game that Microsoft and Google have been playing.

And now it's rising again, because they've made great improvements in Quantum and are on a roll. I use Firefox and it really is the best browser, at least for me. The only problem that Firefox has at this point is that its developer tools are worse than Chrome's, however they've been making improvements there as well and I'm sure they'll manage to win back the developers they've lost.

> Google's render can be found in Opera, Vivaldi, and others

You mean WebKit, which originated in KDE and then forked by Apple? It's not really Google's engine ;-)

Also Opera, Vivaldi and others are at this point completely irrelevant.

Mozilla is playing this game correctly. If you don't have your own rendering engine, you can't compete and you don't count. Which is also why Google developed V8 and then forked WebKit into Blink.

Firefox share was about 30% in 2010 https://en.wikipedia.org/wiki/Usage_share_of_web_browsers

"You mean WebKit, which originated in KDE and then forked by Apple? It's not really Google's engine ;-)"

Google has since replaced WebKit with Blink in Chrom(e|ium), and the projects based on Chromium have been following suit. Even Qt is switching from QtWebKit to QtWebEngine, which is Blink-based.

Granted, Blink is a fork of at least some of WebKit, but still.

>Firefox's share was never very high to beginning with, because Mozilla doesn't have the resources to play the vendor bundled game that Microsoft and Google have been playing.

>And now it's rising again, because they've made great improvements in Quantum and are on a roll.

You say MS/Google are winning because they have money to bundle their browser, and at the same time Firefox is rising because they made improvements to their software. That's a contradiction.

Personally, I left Firefox for Chrome because how unbearably slow Firefox is/was, something Mozilla seemed to ignore for a long time.

Chrome is absolutely the new IE... it’s just in the heyday stage instead of the “ancient barely supported monster” stage that IE was in when developers began to hate it.

The need for greater variety in the spread of popular rendering engines is dire and real. Browser monoculture is dangerous and I refuse to use any browser that’s Chromium/Blink as my daily driver for this reason. I seek alternatives for sites that obviously haven’t been tested in anything else - if you can’t be bothered to properly service Gecko and/or WebKit you don’t need my traffic.

Chrome will never be IE6. Both Firefox and Chrome do constant rolling releases. There's no more "This works with Firefox 4" because in a few years we'll see versions over 100 and they'll be irrelevant. They move up, just like the web, together.

Even MS Edge is on a rolling release with IE11 just sitting back there in its corporate shit role. Safari is really the last big browser not to move to rolling releases.

So long as Chrome/ium remain open source, it is absolutely impossible for it to become the new IE. Chrome itself has no magic that makes it a better browser than base Chromium if you aren't invested in Google's ecosystem.

I disagree that Chrome it will become stagnated like IE.

Microsoft had no desire to improve IE and make the web a more desirable platform than Windows. (I'd suggest it still doesn't)

Google, however, directly benefit from the web being a better place, and they know first hand that a dominant browser can be toppled. And the competition is fierce, with MS back to bundling and it's so easy to switch browsers now that there's (currently) a good compliance of standards and no such things as ActiveX.

> Microsoft had no desire to improve IE

Microsoft was given an ultimatum by the US Department of Justice to stop improving IE, because at the time people thought browsers and the web shouldn't be a part of an OS base install.

Microsoft started what Electron/Cordova/PWAs are continuing today, where the web is a first class application platform in nearly every OS, and nearly every OS needs to bundle a web browser as core functionality.

> Google, however, directly benefit from the web being a better place

…to serve ads. As long as the web remains the most universal platform for ad serving, Google's interests follow.

> now that there's (currently) a good compliance of standards and no such things as ActiveX

Let's not forget how recently Google tried to make NaCL, its own play at an ActiveX-like thing.

It's only recently we've "killed" Flash, and Oracle and others are still holding a lot of Enterprises hostage to Java applets.

Mozilla may be the only group without sin here, and they live in a glass house, to mix metaphors.

I've noticed some unfair rot in Chrome. Like not being able to remove the start page without a plugin.

Even if Chrome never stagnates, it’s not a good thing for it to be overwhelmingly dominant with competitors stuck in firmly the margins. It gives google too much power, regardless of its intentions. It might be more convenient for web developers (if I had a dime for every time I’ve heard of desires to ignore everything not chrome...), but what’s convenient for devs is rarely what’s best for users.

So Google is bad because they poured millions of dollars into Mozilla for Firefox development, but even so their own browser is still better and more popular, and the code is open for other projects to use. Which is a bad, evil thing to do and we want none of that round here!

And this is the same as developing a crummy browser, not sharing the code and not funding any alternate implementations, and leaving it stagnant for years.


«So Google is bad because they poured millions of dollars into Mozilla for Firefox development, »

No. Google paid for having its search engine in firefox (as the default engine).

To be precise, Google/Yahoo deal(s) made by Mozilla only affect certain countries. Mozilla is free to include whichever default browser they want in the rest of the world.

In fact, they did that during the Yahoo deal. Yahoo was default only in countries affected by the deal, while Google was the default one... I would say in the rest of the world, but I'm not 100% sure if they made deals with Yandex in Russia or some other deal.

Google keeps enhancing Chrome and where do you think http2 came from?

Heck Google created SPDY and gave to standards group who changed and then Google went and moved their SPDY to the standard.

Heck Google has over 60% of the browser market and the 2 most popular web sites in the world and easily could have kept to themselves as a competitive advantage. Most would consider that just good business.

But instead Google did the right thing while consistently MS does the wrong thing.

Beware of Microsoft bearing gifts...

Because the implied part is a bit obvious to any programmer would had to go through the 90s under Wintel domination.

They want programmers to use their tools so they can exploit a dominant position and crush competition. They have been charged (and condemned) for abuse of monopolies as late as 2013.

I don't trust a company that used to be dominant and evil and suddenly as it becomes weaker tries to play nice.

The GAFA are problematic, but their ethos is nowhere as bad as what MS was when it was the dominant player of the software world.

Yeah except the implication is that it's at the expense of other players, not that they want to simply join the ecosystem and contribute things that everyone can use with low barriers to exit -- the antithesis of open source. Some companies just join, donate hard work and earn good will without the expectation to cash in with their platform play.

No one has a problem with any company trying to improve it's relationship with any population of consumers it serves, but Microsoft seems to still be trying to create walled gardens and get developers in then make sure they don't leave.

Almost every company works this way. They kind of are okay with the world growing but they want to help themselves in what they are really doing. Like hiring interns, it's mostly a way to get 'a really long interview'. It's not really about helping the interns to grow that much.

Like everything, it's a balance -- open source can also be used to make money (open core, closed source plugins/enhancement suites). At the very least, it's about tact and doing things that are just open enough to not betray your desire to create a walled garden.

Let's use Google's Kubernetes as an example -- it's a fantastic tool that doesn't require use of their ecosystem -- so open in fact that both Azure and AWS are spinning up their own hosted versions. Many of the architectural/organizational pieces are non-trivial as well, years of work at google went into it, and they open sourced it.

I think the bad part is the act of MS buying complete control of the worlds largest open source code (and other) repository. Maybe it goes without saying that this is not a good thing, given the MS track record of EEE and FUD.

An example of what could now happen is that MS could decide one day at will to drastically change how your data is accessed from github, maybe even requiring MS version of git, with added "security" and requiring you to login with your hotmail account or whatever they call it these days.

Microsoft is forcing employees to use Microsoft tools when there are better ones to solve the problem.

Read that again and think about the culture they have at Microsoft, then think about the open source culture if it's the same. Do Microsoft embrace open source?

As a Microsoft employee, I use Mac, Linux and Windows regularly. I write code in vim. I write quick-and-dirty scripts in Perl or Ruby. I’ve shipped code in Java, C and C#. I browse the web in Chrome. I push to GitHub and VSTS.

Microsoft doesn’t force employees to use Microsoft tools; Microsoft forces employees to think critically, empathize with our users and build great experiences for them.

empathize with our users

So why isn't this reflected in their actions?

Cmon, google does the same thing, Facebook does the same thing. They are all at the scale where they can’t adopt random JavaScript framework of the week.

Google cares about ads and Facebook cares about monopolizing social graph tech with ad targeting applications.

Microsoft historically wanted to monopolize everything. Your operating system, developer tools, office suites, Internet browsers... everything.

Yes you are right - Google definitely isn't a monopoly around search nor Facebook for social networks.

No citation handy but IIRC MS recently became the world's largest contributor to open source, by at least some measure. I'm no M$ apologist, having been in webdev for 20 years, but let's not let caricatures stand in for reality.

Microsoft practically forced us to use Windows stack for anything and everything

This is called “dogfooding” and all companies should do it. How can you sell to customers or understand customers if you’re not using the product yourself?

One day I’d love to visit SAP and see how they do their own vacations and expenses. It can’t be with their own software.

I've done some security contracting work for the folks in Walldorf over the years; I've seen first-hand some SAP FTE's getting insanely frustrated putting in their expenses in their own systems. Which was pretty hilarious to watch. And most folks there fully acknowledge that usability is not their strength traditionally.

Things like uptime, support-contracts, having someone on-site within three hours when your multi-millon-dollar-order-processing-SAP-cluster goes down etc are what matter to their customers more than usability (traditionally).

Things are changing and a lot of effort has been put into getting better at UI/UX though. But I haven't been involved with them for several years now so I don't know where they are at or what their current offerings are (I keep tabs a bit on HANA developments but that's it really).

Customers and potential customers also use competing products, so you also need to use those to sell to, understand, and grow your customers.

Maybe SAP uses their own software internally and they've all become power users that aren't bothered by the issues that you're having. Perhaps if they used competing products they'd have a better understanding of where theirs is lacking.

You can always spot software that is or isn’t used by the people writing it

I'm not suggesting that people shouldn't use their own software. I'm saying that they shouldn't use it exclusively.

SAP Is actually pretty good in eating their own dog food, usually the problem is that the customers are super slow in adopting the new stuff that SAP releases...I worked at SAP for 9 years.

I work at SAP. I book absences via a web UI for a SAP system. I record my time using a very nice HTML5 UI that hooks into an SAP system via RPC. I record travel expenses in Concur (the company of the same name was acquired by SAP, so that counts as dogfooding too).

SAP still needs to get better at their dogfooding game in some areas (e.g. using our own ticket system products instead of internally-grown idiosyncrasies), but we're getting there.

I wish they dogfooded on spinning rust and limited RAM.

It’s always been the case that developers have better hardware than end users, I don’t see that changing anytime soon!

It's pretty obvious that the only reason they pretend to give a shit is because they have no other alternative.

I'm not seeing the problem.

I mean, a big problem to me is that Github wasn't open source. It was a closed source, SaaS product that a lot of people depended on.

An open source, open identity, self-hostable platform would be a lot better. And, we're running into issues ourselves in trying to migrate, because running your own-self hosted GitLab isn't really good enough, we don't have the team to ensure the uptime we need, but I would feel comfortable if there were more hosting providers for GitLab than just GitLab itself. Just to get that extra redundancy.

On the other hand, I imagine that's something that could exist quite readily if GitLab shut down or got acquired. Someone could step forward and do the SaaS for all the companies that don't really have the capability to self-host at the same reliability at a pro.

>It's also a terrifying lesson about the fragility of the open source movement

It's not fragile, it very much antifragile. The highs and lows make it stronger and more resiliant.

What I find interesting is the commentary about this deal is all about Microsoft, and very little about the people who sold Github. Presumably they were highly regarded by many, and they sold out open-source (not my view).

Microsoft isn't going for the developers as much as for the development managers and technology company management. They build up a grand picture of the universal MS platform that can be used to target any platform and try to sell it to corporate.

With Google recently investing $20 million in GitLab [1] and helping them migrate from Azure to Google Cloud Platform [2][3], what are developers going to do when Google inevitably acquires GitLab?

Move their repositories back to GitHub, switch to BitBucket or something else entirely?

[1] https://techcrunch.com/2017/10/09/gitlab-raises-20m-series-c...

[2] https://venturebeat.com/2018/04/06/why-and-how-gitlab-abando...

[3] https://about.gitlab.com/2018/04/05/gke-gitlab-integration/

>what are developers going to do when Google inevitably acquires GitLab?

Fork it and move on with their day.

You will not be able to easily move away from any dominant platform because it will be not just the code but the community you leave behind. Also it's quite possible that github will add CI stuff directly into Azure and if you start using that, there may be no other product to move to with the same functionality.

Vendor lock-in is a strategy that works even if you know its there. I'm building my code with the excellent CI stuff built into Gitlab today and I really don't want to move away from it because of its amazing ease of use.

> You will not be able to easily move away from any dominant platform because it will be not just the code but the community you leave behind.

I find it hard to believe that an audience of developers can't point their browser to a different url, or search for your project using Google.

Right but because they're otherwise decentralised it would take co-ordinated effort and ultimately it fragments the community. What we need, if we want to solve this issue, is a federated network of servers which provide the functionality of github today.

That is already the case. The underlying "protocol" is git.

And if the functionality of github became the basic minimum, each provider will still provide their value-added on top of that.

If that sort of thing is forbidden, then there will be no motivation for improvement.

Then I'm curious what kind of data would be shared on the federated protocol. Why can't projects be independent?

I think primarily identity and git locations.

I want stuff like the issue tracker/merge request dialog to be consistent across the federation (and of course for projects elsewhere in the network to be able to be the source for those merge requests)

I don't want an account on everybody's git{lab,hub,ea,xxx} instance I want a single identity that people can trust is me.

I want a way to discover and index other projects on the network. The same way that I often just go to github without knowing the name of a project just roughly what I'm looking for.

> I don't want an account on everybody's git{lab,hub,ea,xxx} instance I want a single identity that people can trust is me.

I know blockchain is excessively overhyped, but to my limited understanding this appears to be a good fit...

Look at how cryptocurrencies work, a wallet is singular (i.e your identity), exchanges and mining pools are numerous and independent (hosts and services for git repos).

The vital part to make this work is a specification for storing the social aspects of collaborating on these platforms in a git repo itself - it can definitely be done, git is so flexible, it doesn't have to interfere with the project code or even need to be present in the visible tree of any branch - but we need that for interchangeability. This first step could be made before figuring out how to make neatly interacting network to add portability to existing platforms.

Yes, I think identity management/authentication/trust is a problem space that should be separate from version control, because it can serve so many other interesting applications.

By the way, I'm not sure if something like a blockchain is necessary here. You could use a simple public key cryptosystem, where your identity equals the public key, and you prove that you have that identity by using your private key in a challenge/response scheme, and you can also sign messages/commits/whatever by using your private key.

> By the way, I'm not sure if something like a blockchain is necessary here. You could use a simple public key cryptosystem.

I think you're right public crypto is enough... i'm just applying blockchain unnecessarily.

I can see this working in a really nice way: right now people are dealing with "migrating" and having to choose to loose PR/issues starts etc that these platforms provide, all of that could be implicit to the repo with zero effort to use a different platform.

- Public crypto for repo authority (who can commit, who can merge PRs, who can manage issues etc - preferably more granularity than current platforms)

- A common Issues/PRs etc format is saved into the repo

- Hosts are networked, they sync repos with each other allong with any platform data... anyone can add (sync) a repo from github to gitlab without being the author (ownership is done by crypto not platform accounts, each platform/host would verify the repo content when syncing from some other host).

In other words, platforms like github and gitlab etc just become a viewport and a node in the network, like there are GUIs for git, github would be just another GUI for the collaboration protocols... at this point you could even do it offline with a headless server, the purpose of github/gitlab is to provide hosting and good UX etc. This would mean that a repo can simultaneously exist on any platform, there is no platform boundary, you could accept a PR from a user using gitlab on github... it wouldn't matter where they came from. Or if you hate all these platforms you can just use the cli and accept a PR from someone using these platforms (you would just need a public git host that the other host/platforms can sync with).

How the crypto integrates with the repo would have to be well thought out as it's essentially completely independent of platform accounts, anyone could locally clone the repo and attempt to push something to your public server (or a hosting platforms server on your behalf).

> I think primarily identity and git locations.

Accounts/authorization is already semi-federated through oauth etc. You can sign in to gitlab.com using your github.com account.

We still have only one kind of namespacing and for users and projects: domains, leveraged by Java (and go, others?) for packages, and email for identity.

Discoverability is really a matter of indexes and catalogs; npm, pypi, Google and duckduckgo.

Delegating login is actually useful for self-hosted vcs/issue trackers: not as low friction as just sending an email with a patch; but it's easier to be able to "sign in" with an existing account, add an issue/code - and get notifications via email - rather than having to manage an account for every program you use and find bugs in...

This is why I'm very wary of recommending any serverless solutions. All the consultants we work with are lusting for them. It's pretty hard to push back against it.

As long as GitLab remains also deployable on your own infrastructure and not become a SaaS-only platform, I'll worry less about this.

You know that GitLab offers commercial add-ons?

You know that in most cases, they are "neat-to-have" and not essentials?

Not all of GitLab is open source, you know that, right?

That's a valid argument against GitHub, since they open source a ton of code, but you can't run GitHub yourself.

That's not a valid argument against GitLab, which offers you the option to host it yourself, and then builds features on top of the free version.

While GitLab EE isn't released using an open source license, its source is publicly available and you're allowed to make modifications to it, satisfying 2/4 essential freedoms of free software, as defined by FSF[0]: https://gitlab.com/gitlab-org/gitlab-ee/

[0] Open Source Initiative's definition contains 10 rights, I'm too lazy to memorize them.

Dunno. The thing is, Microsoft knows they are ruthless. The thing that worries me about Google is, they are completely inaccessible, exist in a bubble and think they're the hero.

Okay, three things.

Yes, I do think this is the worrying part and worse, all the others try to become like them (Google). That future is scary.

Google in the past had Google Code, which was fine and had some level of success.

Microsoft had Codeplex and was rather similar. Didn't use it enough to form an opinion about it.

Both were largely left behind by Github.

You can also use another instance, like https://framagit.org, which is hosted by an independent org who fights for Libre Software since many years.

All your links will point to gitlab.com, you'll have top change everything again. The problem is centralization.

Which links ? Because if I host my projects on framagit, there’s no link to gitlab except maybe for documentation

There'll be a link to framagit.org. Links from other parts of the web. When someone blogs about your software, it'll be a link to that URL etc

Ok, you can also setup your own domain name I guess if you want to avoid that (like with github pages)

Google acquiring Gitlab is very inevitable. GV (which is a Alphabet and not Google subsidiary) always stated that their priority is to be a VC that is as neutral as possible, and should a Google competitor be interested in acquiring a portfolio company they are open to that.

Of course if you have them as an investor they will obviously try to help you with their network which can definately steer a company to a more Google friendly path.

> Google acquiring Gitlab is very inevitable.

From the rest of your post, I think you mean “...very much not inevitable.”


If the portfolio company wants to be acquired by Google, then taking the network they get via GV certainly makes that route easier for them.

If the portfolio company decides if acquisition by another tech giant is the better route, they are free to do that and GV won't sabotage that deal. AFAIK there have been GV exits to all of the big Google competitiors.

> Google acquiring Gitlab is very inevitable

This means "Google is going to acquire GitLab".

> If the portfolio company decides if acquisition by another tech giant is the better route, they are free to do that and GV won't sabotage that deal.

To me, this reads as "Google might not acquire GitLab".

Yeah, wow had a brainfart their. I thought about "unavoidable" every time I wrote inevitable.

"Google acquiring Gitlab is very unavoidable" would also mean that Google IS going to acquire GitLab!

Yeah... I think I should get off HN for the day :D

Except for the “No”, that whole description is consist with “not very inevitable”.

“Inevitable” means “unavoidable", but what you've described is “possible, and maybe slightly more probable than if it wasn't a GV-backed firm, but by no means guaranteed.”

Buy a $5/mo VPS and self host your repositories? It’s really not that hard.

And spending how many hours a month maintaining it and applying updates and security patches and setting up monitoring and dealing with downtime?

The promise of these hosted services isn't the free storage/servers, its the maintenance of them.

We're running our own private GitLab server but we thought long and hard before doing it.

> And spending how many hours a month maintaining it and applying updates and security patches

Looks like GitLab puts out one release a month. Not a big deal. One day a month you can roll out the latest and greatest if you really want to. I probably wouldn't even do it that often, depending on what's included.

As far as all the underlying stuff, depending on what Linux distro you pick it should be pretty self maintaining nowadays for everything that's required to self host GitLab. nginx, PHP, etc. will all self update with your package manager of choice.

> setting up monitoring and dealing with downtime?

If you're buying a VPS you shouldn't need to worry about these. Any decent host will make sure your VM is up and have a high enough uptime to be solid for any use case short of enterprise. If you're an enterprise you might want to spend more than $5/mo.

> We're running our own private GitLab server but we thought long and hard before doing it.

Well that's up to you but it's really not hard.

To be fair, maintaining an (internal) gitlab server really only costs you the upgrade time. You make a snapshot, upgrade (and optionally rollback if something breaks). It takes exceptionally little time - unlike say, running your own email server.

I think Gitlab will be bought by Google soon

> what are developers going to do when Google inevitably acquires GitLab?

There are downsides to MS owning Github, but Google is in a whole separate universe of Evil.

So the answer is: Move out, don't hesitate, and if there's no other viable choice, yes, move back to Gitgub.

I think the author nailed it. This is Microsoft's chance to reach out to developers and prove that they've had a change of heart regarding open source.

We've seen this over the years with previous code hosting platforms. If the platforms betrays it's users, developers are not scared to mass migrate to something else. The onus is now on Microsoft to keep developers happy or risk losing their $7.5B investment.

> prove that they've had a change of heart regarding open source.

Change of heart? Why? I think they are doing a pretty smart move to be friendly towards open source community, but it's not because I believe they changed their mind and want to contribute to the open source world, instead, I simply believes they embrace open source community is because it's good for business.

You see, open source software these years became pretty good and infrastructural. If Microsoft can't be a part of it, then they will soon be obsolete. And they don't want to be obsolete, so they just can't afford to not be a friend with open source.

The real thing that I curious about is what would happen when Microsoft managed to re-dominates the software market (in both user and enterprise end). Will they still be friendly with open source community after that?

All companies that embrace open source do it because it's good for business. It's a seemingly unlimited resource that accounts for billions (made up number, feels reasonable) of dollars of unpaid labor.

It was always stupid to be against OSS from a business perspective and so long as developers hand over their IP on a platter it always will be - Microsoft would be really stupid to turn away free money, and they seem to realize this now.

The idea that open source involves “handing over ip on a platter” is what’s stupid here. The vast majority of open source code is developed on the job. The fact that it’s open source means that the developer can take it with them when they leave and possibly even get some name recognition from it. If it was closed source, the developer would be handing it over on a platter to their employer, never to be seen again.

That's a very contemporary development. Open source was originally hobbyist, research or academic.

Often, it was "programmer solve thyself".

Businesses have only been employing significant numbers to work on open source projects for the last decade.

2 decades

So are you saying that we don't encourage unpaid labor with open source? Are you saying that companies choose OSS for non-business reasons?

That to me is what DotNet Core is, along with Asp.Net, Entity Framework many years ago, and everything developer related becoming open source.

GitHub is just a signal to people who have been out of the loop, bloomberg and many linux developers.

When I start dropping my apps on their systems, they might be surprised how well they run.



DotNet Core was a real surprise to me. As was Windows Subsystem for Linux.

With DotNet Core we have an extremely well crafted language available on all Windows and Linux. It's a great move on part of Microsoft.

I've heard it told that WSL was actually a repurposing of the Android emulation layer for winphone. If that's true it's a hell of an awesome pivot.

People may or may not remember, but most of this started with ALT.NET movement, with a number of influential people and developer supported or created a number of projects.

Before ASP.NET MVC, there was Castle Monorail. Before Entity framework or linq2sql, there was Nhibernate.

It was amazing times. It was clear Scott Guthrie started moving microsoft in that direction back in the day, after ALT.NET movement started proving success and showing an alternative to people.

I remember ALT.NET, and I would not retroactively declare a success. What it did was push into the open the fact that the Microsoft of the day could not effectively handle having rival technologies on their platform, open or closed source. Their strategy at the time was to try to provide a solution for every single product category, to the point of maintaining various no-cost but proprietary tools and libraries just to fill gaps (most of which are now abandoned).

> That to me is what DotNet Core

DotNet Core debugger is proprietary . You can only debug apps within Microsoft environment according to their license.

Same thing for most of the azure tools , you are only allowed to use them within Visual Studio.

I don't really called that "all in open source" .

It's gimmick to give C# devs the illusion of a choice.

Try to build , run & deploy a C# app on AWS with Atom. You're gonna have a bad time.

You've heard of Open Core, I imagine? That's what everyone does. Even the GitLab everyone's talking about in this context is Open Core.

And regarding your claim, you can do all of that, even with Vim if you want. You just have to build some scaffolding.

After all, they can't be expected to implementing everything for everyone. More than that, they were actually nice and created LSP (https://langserver.org/) and now they're thinking about a Debug Adapter Protocol (https://github.com/Microsoft/vscode-debugadapter-node/issues...) that would do the same thing for debugging.

>You've heard of Open Core, I imagine?

It's not "Open Core" if the core is not open. In this case a debugger is core part of a language. It's a very clear movement in order to keep C# devs within their ecosystem and guarantee revenue with Azure. They loose money here , they must get it somewhere else.

This guarantee a total monopoly over the .NET ecosystem , Rider is probably one of the only tools that tried to offer an alternative they are stuck with the exact issue of not being able to bring the debugger because of that hostile license policy.

>Even the GitLab everyone's talking about in this context is Open Core.

I never talked about Gitlab and I don't care about them.

Don't be mistaken about MS , almost every move they do as a double intention behind it.

Just to name a few :

One MS engineer is core member to Webpack & Angular , why ? Windows Server dashboard is written in Angular , it's a critical product, they need some controller there.

One MS engineer is core contributor to Electron, why ? VS Code is powered by electron , it's a critical product, they need some control here as well.

One MS engineer is core contributor to Vue.js , why ? Azure is investing hundred of millions in China and Asia due to insane growth over the past 5 years, they need a good image in asia.

If you think MS has changed , you are delusional about the situation.

Influencing the things you care about in a way you like, seems like a sensible thing to do...

>DotNet Core debugger is proprietary . You can only debug apps within Microsoft environment according to their license.

People keep bringing this up as if it's some evil plan, not a legacy problem.

The debugger is based on Visual Studio's closed source debugger. Visual Studio is a major paid product. It's not simple to just extract an integral feature from a paid product.

> It's not simple to just extract an integral feature from a paid product.

Why? This isn't a technical challenge. As far as I know, the actual debugging engines are pretty isolated modules with some alternative interfaces already available (e.g. cdb). So where do you see the real challenge?

No, it's not a technical challenge. It's a business issue. You know, the whole Visual Studio makes us a lot of money can we really afford to just open license its main feature?

I've always wondered - Do serious apps/products deploy directly from an IDE into AWS/Azure?

I wouldn't say that, as siblings echo. I genuinely think that, for instance, Microsoft is genuinely doing a better job of doing open source than Google right now (With Facebook at the top, MSFT #2, Goog #3 from FAANG).

That might be because I'm deeper into web development though, and both TypeScript, and VS Code are great contributions.

Maybe I'd feel different if I worked in ML, and used Tensorflow daily.

I wouldn't say that Google is high on my list of trust, either. But TypeScript and VS Code do not jump you into a #2 ranking. Go and Chrome easily trump them for starters. Then you get things like node that ride off V8, and Atom that rides off Chrome. The Google ecosystem of community building blocks is still much, much more significant than MS. After all, Google actually dogfoods OSS.

FWIW, I don't and wouldn't use a source code repository of Google's or Facebook's, either. In fact, rather than open a GitLab account, I started self-hosting a year ago. GitLab made that easy. I also looked at Pagure.

wait, how does go and chrome trump open source c#, f#, vb.net, .net core, typescript, asp.net, visual studio code, monaco, powershell core, mono, and the many, many other things they have open sourced?

they’ve gone insane.



and all of these are active and well managed with microsoft employees engaged in communication.

Angular, Polymer,

Dart, Go,



Protocol Buffers, Java Guice/Collections, C++ Abseil

Plus a bunch of minor things


and stuff Google contributes to but doesn't lead.

Kubernetes, CGroups (for containers) VP9, VP10 codecs

Microsoft open source is inherently less valuable because it is all built around being closed source until recently.

So there is much less of a network effect which makes it less valuable. One example: most open source modern OO languages, e.g. Kotlin, Scala, target the JVM, and so are interoperable. Whereas C# is in a closed Microsoft bubble of open source. If it had an LLVM or Java backend, it'd be much more useful.

none of that makes any sense. c#, f#, and vb.net all target .NET (and thus the CLR) and are interoperable. they are also cross-platform with .NET core.

and saying “most open source modern OO languages ... target the JVM” is not accurate. if you mean modern OO JVM languages, then yea, of course, but that is satisfied analogously by the .NET languages.

the c# and f# compilers, .NET core (libraries and framework), and core CLR (the runtime) are all open source and cross-platform. there isn’t anything closed about it. i don’t know what you mean by “closed microsoft bubble of open source”.

why do the .NET languages need LLVM or JVM implementations to be useful? again, that doesn’t make sense.

Nothing which a normal 'non microsoft' developer would profit from. Even their contributions to Linux were mostly about running Windows on Linux and the other way around. Nothing a non Microsoft dev would even care about.

Google however maintains a few products that can easily be considered standard, or at least will find their way in any other developer stack.

what is a non-microsoft developer? these are open platforms that happen to be made by microsoft.

if you use google’s open source projects are you now a google developer?

and visual studio code is probably used more for web development than it is .net work. it certainly is the major selling point of it on the website.

Agreed that is weird wording. When I look at my environment there are basically 2 kind of developers (from this point of view) those who do Net and use the MS stack, and those who don't do Net and at max worked with VS code.

I really don't know a lot of people who do anything with the MS stack. And those who do code on and for Microsoft (which already is a minority)

I personally don't know anyone who switched from Atom or Sublime to VS. Only VS users I know have been using real VS before.

Microsoft might be doing a better job wrt contributing to and creating open source software, but the trust of Microsoft is not there.

My SO migrated everything off Github back on Sunday when rumors broke, deleting the account & repos that had commits and contribs going back years.

Consistently, one of the first questions I ask friends and acquaintences who go to work across the lake to work for MS in Bellevue or Redmond is whether they are on an Embrace or Extend team. Most are on Embrace/Rewrite our own version but "better" teams.

Microsoft has changed, but they're no Red Hat or Suse.

Considering Google has over 2000 plus open source projects I find it hard to believe how Microsoft is doing a better job of doing open source than Google.

Additionally, I find it comical that you would even suggest that a company that has sued other companies for using open source software is even in the conversation with companies that have championed open source software since their inception.

There is no undo button. All you can do is be better than what you were. If Microsoft really wants to start doing a better job in the open source community they can start by agreeing to never sue another company for using Linux.

I assume you're not using Google's GitHub profile as a measure since Google's has fewer repositories than Microsoft's. Where does the 2000 number come from?

To get the developers to help write obnoxious garbage like win10 which forces users into all sorts of things they don't want?

They want an undo button? It's right under their nose: stop with the bullshit auto-update that re-enables itself, document all telemetry and make a big button to disable it.

But they don't do that, do they? Any idea why? Spoiler: it's all fake bs again, take one flashy but insignificant step forward and take 3 covert snake-like steps back.

Updates are lose-lose. Either you get complaints from people about how they're being forced on them, or you get complaints that Microsoft isn't doing more to protect people from exploits because they've all shut off updates.

I'm not a fan of the telemetry either, but they do have it all documented now, which is more than, for example, most phone manufacturers: https://docs.microsoft.com/en-us/windows/privacy/basic-level....

And I'm not sure what Windows is supposed to force users into things they don't want. I haven't had an issue with turning off functionality and having it be switched back on during updates - I guess it must happen based on the complaints, but it's not something I've experienced.

There's a middle ground somewhere around 'not re-downloading updates every time they fail' and 'not disguising Minecraft as a security update'. A lot of the annoyance at the updates is with how frustrating the implementation is.

Microsoft aren't the only company that need constant security updates: I've never seen any complaints about Chrome's.

Chrome's disabled using backspace as "back" button.

But really that's my only complaint!

> Updates are lose-lose. Either you get complaints from people about how they're being forced on them, or you get complaints that Microsoft isn't doing more to protect people from exploits because they've all shut off updates.

When updates that only add telemetry (Windows 7), or that upgrade an installed Windows 7 / 8 to Windows 10 without even asking, among other garbage, are classified as "critical security updates" by Microsoft, it really should come as no surprise that people disable all automatic updates in response.

> Updates are lose-lose. Either you get complaints from people about how they're being forced on them, or you get complaints that Microsoft isn't doing more to protect people from exploits because they've all shut off updates.

They didn't exactly help themselves, firing their QA team, and using their users as the test platform.

Nothing was done about Telemetry (maintain state after updates, better options to disabling PARTS of it, documenting it) till AFTER people discovered it and there was a public outcry. And they still didn't document it till it was becoming clear that GDPR would go through and they'd be legally liable for not doing so. And there's currently still no way to completely disable Telemetry's data collection. It's a simple on/off toggle and potentially a registry entry that determines if the startup scripts of the OS should start the Telemetry application at all. I'm sorry but at best that's pretty myopic for a company who claims (paraphrased) "trust is our most important resource", or (literal) "judge us by the actions of our recent past, our current actions and our future actions". And it certainly doesn't help to shake the notion that they have larger and more maligned plans for what they intend to do with their greater OSS influence, certainly not when you consider their history. But even if we're to take those statements at face value they still don't completely align with their actions.

even chunks of Azure are open source.


Azure used to bill themselves as the "open cloud". That's kinda gone from their marketing, but they are WAY more open than AWS.

Only OpenStack is fully open source and fairly feature complete, with a few dozen providers using the OpenStack platform.

OVH, who owns the largest datacenter in North America uses OpenStack, they're actually one of the largest contributors. Red Hat & IBM both seem to be really pushing private deployments as of late for their larger clients that already own large DCs.

Knowing that OVH uses a piece of software makes me want to never use that piece of software.

They are the worst quality cloud provider, IMO.

Definitly not the worst, but very far from the actually good ones for sure.

Did they kill your puppy? I've got multiple servers and VMs with them, uptime and connectivity is great.

Realistically, if Microsoft wants to be profitable on that investment, they will have to make changes to earn more money. Which will alienate developers looking for free good repo.

Even if they plan to keep good will support for strategic relational reasons, that good will is easy to go away with strategy changes.

> The WWDC, sacred nerd summit of Appledom, is where they announce things like a new “night mode” for the operating system and try to convince programmers that Apple Watch matters.

Bloomberg got an honest laugh out of me.

Apple Watch is apparently the best selling watch (not just best selling smartwatch) in the world [0], with millions of users, and many copycats. Why shouldn’t it matter?

Of course, there’s only so much you can do and make on such a small screen, and it is an accessory rather than a “primary” device, but that is another issue altogether, if one.

I think it may even function as a motion control input for the rumored VR headset they’re planning.

And is anyone seriously trying to pass off Dark Mode as a MINOR feature? It can be a significant productivity and wellness boost for many people who have to look at a screen all day. Microsoft has struggled for decades to achieve universal consistency across their OS even for a single system theme, and they still haven’t gotten it right, whereas Apple is already doing seamless switching of UI appearances across many apps (this is more than just changing the colors; I’m currently watching their WWDC session about it, and the amount of thought they’ve put into it and how easy they’ve made it to implement is impressive imo.)

[0] http://www.businessinsider.com/apple-outsold-the-entire-swis...

I can believe that it is the best selling watch in the world.

People don't buy many watches, because they last a long time.

Last time I bought a watch was 7 years ago, and it still works fine. It's probably got another 3 years before I need to change the battery.

I think the last time my old man bought a watch was some time in the 80's or 90's.

> People don't buy many watches, because they last a long time.

Smartwatches can fix that problem. Obsolete in 1 year, battery failure in 2 years, incompatible with app ecosystem in 5 years.

My Series 2 Apple Watch still runs perfectly after 2 years of daily usage, and is compatible with all current apps as well as the upcoming watchOS 5.

How do we fix the problem of people seeing a problem in selling long lasting products ?

By changing the system to internalize extra costs of products with low lifetime. If companies can earn more by building products that last decades instead of years, they will do it. To make this work, you'd probably need to come up with a pretty convoluted taxation scheme.

Yes, I’m still using the Series 2 I bought two years ago, and have no desire or need for the newer series.

Best selling watch is believable, but who the heck buys and wears a watch for anything besides fashion these days? It's a totally niche market IMO.

I like having time easily accessible.

Though I have some philosophical and design issues with Apple, the biggest thing, the #1 requirement of any smartphone for me, is a notification LED for messages. I like seeing at a glance who or what is notifying me. Why is an LED so important? Because I want to identify a message quickly, without my phone always in my pocket (e.g. it's on my desk or next to the bed). With a watch accessory, for many, that need would go away.

Don’t underestimate where the watch trumps LEDs: discretness and gentleness. After owning one for a few years I can’t stand the abrasive sound of a vibrating cell phone on a table.

> the abrasive sound of a vibrating cell phone on a table.

Indeed. Waking up to that cacophony can mar one's entire day.

It's not even that: Timex probably outsells Apple watches 100:1, but they have a crazy variety of models instead of just 1 or 2.

My Apple Watch is surprisingly useful: I can figure out which conference room my next meeting is in with a single tap of the home screen, control music, etc. It honestly surprised me how much value I got out of it.

Same experience here - I got it essentially so I could have something being like "Hey buddy, you should be somewhere in 15 minutes..." but I've gotten a lot of use out of it.

I actually had exactly the same experience with my Pebble. I got it out of casual nerdy interest and was sure I'd use it for a couple of months and then pack it away.

Having notifications and basic functionality on your wrist is surprisingly useful though, and I used that watch until it broke (after Pebble went under unfortunately).

I've got a Ticwatch E on it's way as a replacement.

Pebble was nice - had one but its uncanny how similar the Palm Treo analogy goes - I loved my Treo till I got my iPhone. I loved my Pebble till I got my Apple watch.

Stuff the old ones do better (even today) but the Apple replacements are all that + constant years of OS upgrades and a stronger app ecosystem.

I hated watches - had a wrist ornament years ago and stopped wearing it because it interfered with my active life.

Apple's strategy is a blue-ocean one - they brought Apple users and active-lifestyle folks into a relative staid luxury goods market and (aside from an initial stumble with series-0 edition) co-opted the market and transformed it.

It wasn't exactly as powerful as the move to smartphones, and tech was moving there already but it's really great that my Apple watch is part of my work (2nd factor), lifestyle (Strava), and family (timer+music controls).

There's room for improvement but it gets the useful parts right.

I'm not a watch person, but bought a series 2 on sale last year. After a week I was cursing Apple because here is a device that I already feel lost without. Workout tracking is great, alerts are nice, quick views to upcoming meetings, and the list goes on. It ends up as a super useful device.

Also, niche? I literally see Apple Watches everywhere. At first I was surprised because all I read online was how the AW was a flop. I hope I can be part of a niche product as successful as the AW one day. /s

I would guess that watches are number one fashion item for men? A watch and a wedding ring are the two wearables that are acceptable for men even in the most conservative communities.

Health and fitness, and urgent notifications without bringing the phone out. It’s actua remarkably useful.

I use my Apple Watch (LTE) as a minimal smartphone.

I take the dog for a walk at lunchtime, or take an afternoon off. I can listen to music (Apple Music is crap) and receive texts. Clients can call me.

But I don’t get the normal smartphone distractions. And going places with just your keys, headphones and watch is really nice on my pockets too.

> Apple Watch is apparently the best selling watch (not just best selling smartwatch) in the world

That's rather "big fish in a small pond" in 2018, though, isn't it? It's like making the best-selling model of recumbent bicycle in the world.

If we consider a small pond to be worth billions of dollars, sure.

Even worse than that, because it glosses over how the other manufacturers sell a hundred times as much stuff, just categorized out into thousands of different models and trim levels.

Maybe another analogy would be some kind of specialized shoe.

It matters for Apple, for developers? not so much. Millions of users and not a single third party "killer" app.

The health and fitness stuff is the killer app there, imo. Once you get hooked on that you are hooked.

Which developer wants to miss out on serving potentially millions of users? What kind of “killer” app do you envision for a watch?

Being able to check messages, calls, Uber, flights, etc. right on my wrist are all killer applications for me, along with the fitness features of course.

I suspect the upcoming walkie-talkie will be fun too (something that should’ve been there from the start.)

Any potential the Watch has is clearly not enough for a strong app market to develop.

> Which developer wants to miss out on serving potentially millions of users?

You are forgetting that most developers already have millions of potential users on the iPhone with a market at least 30x bigger than the Watch's, and iOS has none of the limitations the Watch has.

Right now a Watch app requires an iPhone app, is hard to monetize, it cannot be sold on its own, it's slow and frustrate your users, etc. The only upsides are for Apple. It might sell really well because the core functionality provides enough value, but developers need to see a clear return for their efforts and right now is not there.

As others have said, think of your Apple Watch app as an extension to your iOS app's UI: it enhances your iOS app vs. the competition that doesn't offer the wrist UI.

I've seen users requesting a Watch version in reviews for iOS apps. Even I have chosen some apps in the past over others specifically because they offered a Watch app, like navigation apps and flight checkers etc.

As a user though, I must say that if "monetizing" and "selling on its own" are more important concerns for you than make your product/service a better experience for ME, then I probably wouldn't want to use any of your apps.

That is all well and good, but that is not the point. The point is the App Store is bleeding watch apps every day, and Apple needs to convince developers that the watch is worth it. Even you admit that the only reason for a developer to have a watch app is as a defensive move, and that is what needs to change.

The upside for you is that if you have a watch app and your competitor doesn't, customers are going to choose your product over theirs.

That is what in theory you think happens. What actually happens is that in very few app categories that is the case. For example, not a single game is losing market share to a competitor because they do not have a watch app.

You picked the worst example here: you really can't play (serious) games on your watch, so the comparison is invalid. But for a fitness app, or a messaging one, or a music service, I would significantly favor the service that has a watch counterpart, if not shun the other one outright.

I picked the biggest category of apps just to show you having a watch app is irrelevant when picking an iPhone app for everything in the app store but a very narrow set of categories, and you just demonstrated that. Yes, fitness is great, messaging somewhat ok, but what about the other millions of apps in the app store? what about a writing app? or a video editor? or a measuring app? or Dropbox?

Right now there is no reason for Youtube to make a watch app, and no reason for you to pick Vimeo over Youtube because it has a watch app.

I think you make your own point. There is no killer app, so why would developers care about millions of people with a watch they can’t do anything novel by a third party?

I've switched my app usage several times to ones that have better watch integration.

It's popular enough that, if your competitors make a watch app, a large userbase would begin to prefer using the competitor app.

The watch is a UI for iOS apps. There are zillions of iOS apps.

It is funny, but (for the benefit of those who haven't been to one) also a complete mischaracterization of WWDC. The keynote has always been "dumbed down" for press/analysts desperate for between-product-launch news, with most of the juicy stuff saved for in-depth developer technical sessions.

And that's the problem. It's a World Wide Developer Conference. And then, in the keynote and later in news and announcements all they come up with is emoji and dark mode?

The Platform State of the Union is the real Keynote and the sessions are the meat of the conference.

I don't see why this is relevant for this article, though…

Because Apple and Microsoft are still at war.

Or something.

The whole article is tongue-in-cheek. It's relevant mostly because it's funny and partially because it provides context. Also, later in the same paragraph:

> For Microsoft to trot this out during WWDC is a real thunder-stealer. It’s nice to see global-platform capitalism played with a little verve.

they're not wrong.

They aren't right, either. I wasn't exactly wowed by the public-facing announcements (other than the Siri-related stuff) but there's some nice stuff in the side sessions...and the Apple Watch is somehow the hottest product in both wearable devices and watches at the moment.

And are they not expected to make some product announcements at this conference?

I guess it depends how you define "products," but if we're talking hardware then WWDC as a product announcement platform is a relatively recent phenomenon. It is a developers conference, after all.

Right but there is always some announcement that is not developer related.

In 2017 it was iMac Pro, HomePod. In 2016 it was Apple Watch Updates. In 2015 it was El Capitan.


Applications are open for YC Summer 2019

Guidelines | FAQ | Support | API | Security | Lists | Bookmarklet | Legal | Apply to YC | Contact