with Microsoft ownership of GitHub, they have effectively sized "open sources" means of production. They now own the defacto superhighway that much of "open source" reproduces itself.
Is GitHub becoming less a free host for git projects (not that they're suddenly charging) and becoming more a capital accumulator of free labor, enclosing open source software behind an invisible fence?
my concern is that github is becoming a marketplace where consumers of open source software (corporations, etc) can accumulate labor by reducing their own costs, they do this by externalizing the cost of developing software by appropriating it from open source projects.
it both makes it easy to for for-profit companies to externalize the cost of producing software and cheap (free) to consume OSS, and it drives the perception of the value of software to to nothing (the actual labor-time of the software creation is quite high). Much of the OSS labor being done is subsidized by the contributors paid day jobs.
I think you're describing an inevitable endgame of information technology: Value of software tends towards 0, as it can be copied with very low cost, while value of data tends towards the value given by scarcity and actionable information of that data. In there you've encoded the value of business processes within the data economy. Which is why long term, only the services model can endure while pure software vendors will have to either adapt or die. Among the traditional software companies, Microsoft has been the strongest in adapting to this new paradigm.
I only wonder though, how long this economy is going to last, i.e. how long until AI is so commoditized that the value of data goes towards 0 as well, along with the value of most education in IT.
> I only wonder though, how long this economy is going to last, i.e. how long until AI is so commoditized that the value of data goes towards 0 as well
Well, first - AI in general is not very commoditized at all. It's just machine learning that is becoming commoditized. And the distinction is meaningful.
Going from that - data is the stuff you use to build ML models, and it's absolutely essential to doing better ML. ML becoming commoditized increases the value of (good) data, because it's an essential ingredient for any ML company trying to gain an edge in an increasingly competitive market. It's not entirely unlike how oil being a commodity does not mean that oil drilling equipment is cheap.
We see this play out in the companies that are big users of machine learning. Facebook, Microsoft, Google, etc. are all quite happy to publish the software they use to build their models. But the data is kept under lock and key and buried under the bottom of the sea. And a huge, huge, huge amount of their competition is centered around attracting the people who produce that data, because it's the ability to generate more data that gives them a competitive edge.
I absolutely see the point about the value of data going up, and again, that's the paradigm we're living in now. It's just that I'm not sure this is going to last for long. What's new in ML is, at its core, increasingly powerful pattern matching. But with even some moderately strong cognitive functions on top of that, the data processing monster could start feeding on itself, i.e. go out to get new data on its own. We don't have enough images of donkeys feeding on grass for our object recognizer to match against? Let me coordinate with our drone network to go get a couple of terrabytes...
Again, I'm talking about possible future technology. What if you can infer labels some other way? The fundamental element you need is not labels but a cost function - that could e.g. be curiosity and fear (of disappointing the end user). The most fundamental part that ML is missing to be AI, is to have a model of the world, with which such a cost function could evolve organically. Once you have that, AI can go and explore on its own.
That makes no sense. If you can already infer the labels beforehand, then you already have a working model, right?
The job of the model is to infer those labels. The model needs to be trained to accurately predict those labels by using examples that were labeled beforehand.
By model of the world I mean something much more than what's called an ML model today. I mean the cognitive ability to predict the past and future of some part of the real world based on observation. Most simple example: Seeing two video frames of a massive object moving in front of a reference background and predicting the curve it's going to do, and where it came from. Only having such an understanding of the world (physical laws, human interactions with it etc.) allows to program in much more general cost functions that can get the system to explore on its own, not just within a given rule set, but the rule set itself.
Again, what you're proposing has no basis in today's reality of how ML is implemented. Especially not for classifying image data.
What you're proposing is some kind of machine intelligence that can only exist right now in a William Gibson novel.
Reinforcement learning is something similar, along with unsupervised learning. But you still need to define some kind of objective function and rules to play by. The AI cannot come up with this on its own. And the real world is far too complex to model all of these physical rules you've mentioned.
Reinforcement learning can work for Go because the rules of the game are relatively simple. It is the number of possible moves/outcomes that make it enormously complex.
Until humans stop providing free input to their machine (AI) adversary, e.g. because of new business models that pay some humans for high quality feedback. That will leave parasitic AIs competing for low quality signals.
If you can accurately, inexpensively, and automatically figure out what the labels on the classifier's training data should be, then you have no need for the classifier.
That doesn't quite match the reality that I can see.
A static piece of code may tend to 0 (business) value, but software that is actively being used and maintained tends to have significant value. That is probably one of the key reasons why the service model is so successful. Static software's value rots away quickly, but software that is actively being used appears to be getting more valuable over time.
If and when the point of advanced AI that you describe is reached, does it still matter? Everything will have lost its value by then unless society changes its mind about what a machine with that level of intelligence is.
I get the impression that we're actually in agreement, just with different wording. For me the term 'software' is describing exactly that, a static piece of code. The usage part is encoded in data that goes back to the developers - but if all they do with it is provide another static piece of code (e.g. an upgrade), we're at the same starting point again. Now, if you merge together development and operations, such that usage data is always there to grab, analyse and push down a development pipeline - that's where the services model is shining, but not thanks to a piece of code, but due to tons of business processes running that others can't easily replicate.
I think the distribution channels are exactly what changed - 25 years ago they were only available to large organizations, while now they're free and commoditized.
> github is becoming a marketplace where consumers of open source software (corporations, etc) can accumulate labor by reducing their own costs
This is kinda the perpetual quid-pro-quo of Open Source, though, right? It's mutually beneficial so it continues like this. Folks who fear misuse/abuse from licensees usually gravitate towards GPL -- it's much more user focused and IMO makes some sense.
Currently they mention that they collect "aggregated, non-personally identifying information ... about how our users, collectively, use GitHub", which is perfectly reasonable for a corporation providing services which many users consume at no cost. But if GH was under foundation control it would be nice if that same data were kept private.
The privacy policies of Mozilla, Apache, and Linux Foundation say more or less the same thing. Do you have an example of a relevant foundation that doesn't collect information on their users?
I don't believe those organizations typically share that information with other corporations. Sorry if I've let this aspect overshadow the more general point of foundation control -- I was just trying to avoid a potential loophole in a very hypothetical model.
It very well might be. But why is it bad? If there are more incentives to open source more things, I say let them go nuts. Our incentives are aligned.
What OSS does is to democratize access to tech building blocks. There is still a ton of wealth generation possible by using these blocks to build useful products.
How so? As far as I can tell there’s no evidence of that? I mean, I get MSFT’s past, but for quite some time they’ve shown they support OSS and developers in general. I haven’t seen any change in GitHub and MSFT, in general, has become much more supportive of OSS efforts.
I guess the thing that concerns me (not speaking for anyone else) has largely been the (admittedly late) realization that Github can be bought.
I had, for some indiscernible reason, thought of Github as this near-benevolent entity that existed exclusively to facilitate open-source and distribute Octocat stickers. When they were purchased by one of the largest megacorporations on the planet, it pulled-back the curtain, and showed that, no, they are a for-profit business, not a charity.
I shouldn't have thought that way, but I did, and as a result the MS-buying-Github thing has not sat well with me.
a16z invested $100MM in Github (when they did not need the money) to help Github become the dominant "OSS Social Network". a16z was always going to want a return on that investment. Prior to their acquisition, Github was pursuing the enterprise market.
Yeah, I'll admit that my mentality was flawed and short-sighted. Developer-time isn't free, of course they need a way of making money. I don't even blame the investors or Github for the acquisition; if I were in their shoes of course I would do the same thing.
Fair enough; I understand that point of view, and even had it myself, to some degree. But I also think that has little to do with MSFT. The same would be said if SF, or Amazon, or (insert any other large corp) would have bought them.
I also think GitHub, essentially, positioned themselves that way. However, that, if viewed a bit cynically, could easily be seen as just marketing or PR.
Oh, I don't know that I made this entirely clear, but I'm not saying that this is MS's FAULT necessarily. I agree, if Amazon or Google had purchased Github, it would have been the same thing.
I think the slight difference is that MS does have a history of shitting all over open-source software (Ballmer famously said Linux was cancer years ago). While I'm aware that companies change people all the time, it's unsurprising that some people get a little bit uneasy about them owning the largest distributor of open-source.
Yup, I definitely agree. And rereading your original comment I think it’s fairly clear and I probably just misconstrued your intent: /any/ large corp buying GitHub would’ve resulted in a similar reaction and it’s just a bit stronger given some of MSFT’s history.
I mean, it’s not that difficult to make the case that their support of OSS over the last decade was specifically to buy developers’ goodwill, though I don’t think that’s the case and think GitHub was headed this way anyhow.
On one hand, I think the history is enough to be highly skeptical of MS: how many times does someone have to show you who they are before you believe it?
But that aside, I don't see how any corporation should be trusted to host such a large percentage of OSS. Whether it's a year from now or 10 years from now, what do you expect to happen when Microsoft's material interests cease to align with supporting OSS? Do you think they will subvert their own interests in the support of some "greater good"? Given the history of how corporations behave, I think that would be very unlikely.
> how many times does someone have to show you who they are before you believe it?
It's dangerous to anthropomorphise corporations (or nations).
Yes, corporate culture is sticky but you can more easily change a corporate culture by a radical demographics change both at the base and at the top, and redesigning the incentives.
That goes both ways and it's why you cannot have inherently "good" corporations either (no matter what their positive track record is).
You're absolutely right, and I was probably being a bit too poetic in my wording. What I mean to say more precisely is: when a corporation has a documented history of using benevolence to acquire market share, and then leveraging that market share to pivot in an anti-competitive, anti-consumer direction, then it is more rational to approach their new-found benevolence with skepticism rather than optimism.
I think your second point is really the crux of the matter. Even an independent GitHub controls far too much of the ecosystem. There’s really no way around that in the current situation. Had they followed a similar path as Gitlab, we might not be having this discussion (regardless of an acquisition). Sourceforge is an obvious point here.
How much time would be considered enough? I’m not disagreeing, just an honest question. Would it need to be 50% of the company’s life, rather than ~11%? 25%?
My semi-facetious answer to this case is "The time span of 3 Microsoft post-Ballmer CEOs." The first CEO to implement open source (Nadella), the second to continue the strategy, and the third to prove the second CEO wasn't a fluke!
The honest answer is at best I can forgive their bad behaviour. But I'll never forget it.
Consuming software cost producers zero, practically. You do have overhead from 'how does this work with docker', of course. However, don't get offended at the 'useless software eaters'. The greatest consumer beneficiaries are driving their populations out of poverty, like China and India. China is the #1 software 'cost externalizer'. Developing nations are the #1 relative gainers (proportionally) from consuming open-source. All this means lower infant mortality, less unemployed women, and longer life expectancies. Let them eat. The US, the largest and wealthiest, is the #1 open-source contributor. However, in absolute terms, we (USA) are the greatest beneficiaries of open-source software. Just look at how high of a valuation our tech companies still fetch and how we have most of the top corporations. Yet still, the millions of Indian persons getting online every year and other gains has a greater relative impact than a 3% growth in our real GDP.
Also the addition of new programmers from developing nations is going to drive development costs down. And with the intro of ML code recommenders and distributed platforms, dev costs will fall in half every few years in that not so distant future. Software is incredibly expensive to develop at the moment. See tech hubs real estate and housing costs. The global trends that affected middle America is going to enter Silicon Valley like a tornado.
Yet you make a could point that from an accounting standpoint, costs are not clearly accrued by the companies that contribute their workers time to their open-source dependencies. There is some inefficiencies in determining which contributor pull requests get accepted; corporate sponsorshing projects and hiring contributors is a payment, in some cases, to have a higher probability of getting code merged. The largest contributors of the most valuable projects, like the Linux kernal, are large stakeholders that are dependent on the project. And most successful projects I've seen have competition from developers seeking to add stuff that helps their company projects better interoperate with the dependency. So the accounting is not very accurate.
However, all open-source developer contributions will soon be weighed by others so that those not so high up on the project food chain can be directly compensated. This is getting worked on, now. Every one who contributes value will be rewarded.
>where consumers of open source software (corporations, etc) can accumulate labor by reducing their own costs, they do this by externalizing the cost of developing software by appropriating it from open source projects
Isn't that effectively the end result of OSS ... for everyone? I use it too.
Not that I don't use the products. I use VSCode and Github.
But I just know that MS current behavior is simply money oriented: it started exactly when they lost their top dog position in several important markets.
So now they have to attract people, clients, devs... They can't get away with insulting communities anymore. Hence they hired more friendly people, and told the ones they already had that now it was ok to love open source and make the world better.
It's great for us really, and we should enjoy it as long as it last. Currently Google is going toward the reverse trend, and is probably going to be the new Microsoft in 10 years.
Then who knows, maybe they'll do vice-versa all over again.
But it's not about the company core identity, it's about what's profitable. It will change when the wind turns.
I will believe the heart of MS changed when their attitude will have been mostly positive for a long as they haven't been in the past. So see you in 30 years.
Companies aren't people, they don't have an identity or a personality. That's anthropomorphism that benefits them by causing users to trust them as if they were actual people. They aren't good or evil; they are better seen as profit seeking automatons. I will never "trust" a company as if it were a person. I just take what they give me and use what I need.
I remember when Microsoft bought Github, they flooded open source-related sub-reddits with astroturfers who kept posting slight variants of the the same talkpoints, the funniest and most persistent of which was, "You have to admit that Microsoft has changed in the last four years", as if that was common knowledge that people were not willing to accept, as opposed to a slogan that they came up with in some meeting in Redmond.
Since we're on the topic of OS/2 wrapped up in a discussion about Microsoft's messaging versus actions in open source:
OS/2 could be open source. There was a push for this a little over 10 years ago. IBM actually took notice and issued a statement. They were willing to make OS/2 open source, to the extent that their interests in the IP allowed them to, but that alone wouldn't be sufficient. So the reason that it's not open source, right now, today, is because of Microsoft.
If MS really is this new and changed philanthropic messiah of free and open source software, then let's see it. Open source OS/2. I'm not saying "open source Win10". I'm not even saying "open source XP". We're talking about friggin' OS/2—a product that barely fared higher than DOA status at its height on the market.
I agree with you that assuming that they are "paid off" is illogical and damaging to the common discourse - but it does happen at times.
I was interviewed on this topic specifically and quoted in an article. It made me somewhat nervous, because the author wanted to source the quote with my full name, title, and employer - understandable, but my feelings on the issue were somewhat nuanced. As a result I had to place more than a little trust in the journalist writing the piece. We spoke for quite a while and I'm 100% certain that she could have quoted me out of context to make it sound like I held whatever viewpoint she thought sounded most controversial. She represented my statements accurately and clearly, and I'm glad I worked with her.
I referenced MS's history of "embrace, extend, extinguish" several times and basically said that while it appears that they've turned a new leaf and are not the same leviathan they once were, their reputation for hostility toward F/OSS was not built overnight and they will not lose it overnight.
I'm more than happy to give credit where credit is due - Microsoft is doing great things right now for F/OSS. They've built much of the Azure platform around it, and continue to promote the use of F/OSS alongside or even in preference to their own offerings. If this continues, they could be the next Red Hat... but if it doesn't, and a few years from now we look back and find that they were still the "devil we knew", then they have the potential to cause a lot of damage. I don't think it's at all unfair to be cognizant of that fact and to be hesitant to trust them fully.
MS has always been a really developer-friendly company (which is not the same as an open source friendly company).
Small tweaks to that model, coupled by really solid products that everyone can use since they're free and open source (like TypeScript and VS Code), make it easy to bring that developer-friendliness to the masses.
And the substantive move of making .NET open source with the apache license (and .NET core)x really opened up yet another solid technology to the masses that didn't want to run Windows.
Kudos to them.
Update: before people pounce on me, I mean they're developer friendly to developers in their ecosystem. .NET has had the best docs and tutorials and improvements for many years.
> MS has always been a really developer-friendly company
I beg to differ. Microsoft historically saw developers as customers they wanted to profit from - I remember how much Visual Studio and MSDN access used to cost back in the day, if you weren't a student. If it hadn't been for the competition from open source tools, I'm pretty sure they would have continued that way.
This was true even to the point of hurting themselves. The most egregious example I remember was not including any installer for Windows together with the paid Visual Studio - that was a separate package. I wonder how many millions of users got their DLLs and settings messed up because of this choice, and how much brand value Microsoft lost to Apple because of it...
> Microsoft historically saw developers as customers they wanted to profit from
I think that's exactly why they've tended to be so developer-friendly.
Historically, there have been three categories of dev tool producers. There were the IBMs and friends, who packaged the dev tools as freebies baked into a product that was mostly marketed toward business folks, and only made life good for developers to the extent that the business folks demanded it. There were companies like Microsoft and Borland, who sold more directly to developers, and therefore put out some really nice, productive dev tools for those who could afford them. And there are the open source projects, which are by and for developers, and are therefore also generally quite nice as long as you don't mind the things that aren't fun to work on in your spare time being kind of janky.
Microsoft falling into category #2 does not make them inherently evil. Nor does it morally obligate them to market toward hobbyists. The prices they charged were high, but not unreasonable for corporations, especially given the level of support that tended to come with those packages.
Apple falling into something more akin to category #1 doesn't make them inherently good, either. It's just a part of their product strategy as a for-profit company. Besides, does anyone actually like working in XCode? Like all cost center-produced software, it's kind of famously flaky.
To be fair, I think Microsoft considered Installers a Windows "bug" for many years.
Installers have always been a problem on Windows and Microsoft has always treated it as the unfortunate problem child it was. A lot of guidance from Microsoft has been "avoid Installers [because you shouldn't need Registry settings, shouldn't be putting things in C:\Windows\System32, etc] and don't use things that need Installs, but if you do need an installer, here's MSI which is super complex, not least of which because the preferred tool to build MSI has only recently changed to an XML-based build tool from a crufty collection of worse command line tools and direct binary manipulation".
Then Microsoft tried too fast to move everyone to AppX as the One True Installer. But they forgot backwards compat and the slow pace of Enterprise. So now they are working to turn the ship on AppX with the rebrand/rebuild to MSIX (now with Windows 7+ support in preview), and this time trying to get more of the Traditional GUI Installer Builders in on the game (so that Enterprises with InstallShield licenses dating back to their VB3-based Windows 3.1 apps don't have to learn anything new).
MSIX seems to be a great solution to move forward with, but yes the road to get there has been really painful, and the damage of how many people are still using Installer Builder systems built only with Windows 95 in mind rather than anything more recent is a painful legacy that Windows will likely be hurting from for a long time.
> To be fair, I think Microsoft considered Installers a Windows "bug" for many years.
And that's the cusp of the problem. A really developer friendly company would have given a simple solution to a developer who had finished his product and wanted to distribute it!
They certainly tried to make things better at times, though, they just often didn't have all the answers or solve the problem for enough use cases. ClickOnce had some good ideas, but the implementation lacked some details and the .NET-only focus left a lot of developers out.
It's also why I find it funny that a lot of developers are so angry at Microsoft for the effort put into the Microsoft Store. They absolutely wanted to make the installer/deployment world a lot better for the average user. Sure, the Microsoft Store in its focus on trying to be the best experience for users hasn't always been the best experience for developers, but they've been refocused on developer experience the past few years, and the MSIX efforts really are interesting. Particularly because they've learned from the mistakes of earlier projects like ClickOnce. (I don't think a lot of people realize yet that MSIX already supports better-than-ClickOnce style application updating from file shares and web servers, bringing ClickOnce-style deployment essentially on par with Store deployment for MSIX.)
MSDN is still pretty damned expensive... That is a significant profit center. They've also had a lot of free and/or open tools over the years.
I learned C# with a big not MS book, and the command line compiler alone. In late 2001-2002, I was unemployed and it was nearly impossible to find work as a developer. I spent my free time learning the ins and outs of C#, and by mid 2002 was working FT again. By comparison, I also tried to get into Java, and it was painful. Not the Java language, but the tools. Getting a simple hello world in Java wasn't fun at all, getting anything more complex outright painful.
I didn't pay MS a single dime. Through the early 2000's I did a lot of ASP.Net development. Both in, and out of MS's sphere. I used Castle Monrail on Mono in Suse Linux. I also used .Net in Windows. I built a standalone kiosk system on Suse that used an embedded browser for the UI via Mono and a Mono backend.
I also worked in MS shops, where the Windows and MSDN licenses were paid by the company. I'm generally skeptical of MS. I was hugely skeptical of Silverlight. I don't like that MS doesn't have an official client for MS Team in Linux (despite MS basically having the best cross platform Electron tools) and generally falling short on Linux all around these days.
MS tends to make their money when developers using their tools makes money. Historically off of Windows licenses, today probably as much or more so from Azure. Not counting Windows and o365.
And while I would prefer to be on a Linux desktop/laptop at work, I do appreciate the work they've done with Docker and the WSL teams to make development for non-MS better overall.
As to installers, there's InstallShield, NSIS, Inno Setup, Wix and a lot of others that have come up over the years. I think there was a light version of Install Shield with VSTS. I don't think MS has been nearly as bad on developers as you seem to feel.
At least MS doesn't force you to turn over 30% of your income to them like Apple does.
The interests of platform vendors and developers are inherently at odds. Vendors benefit from lockin, while developers benefit from portability.
This applies to other OS vendors like Apple, Cloud vendors like AWS and GCP, and so on, just as much as it applies to Microsoft. It's a universal principle.
It's true, but Microsoft has also done a pretty good job of failing to lock people into their development platform, and seems to have largely given up on it. Does anybody use COM or WCF anymore, except maybe to maintain or talk to legacy systems?
It leaves us in a situation where it's hard to get any more locked into Microsoft than it is to get locked into Haskell: Sure, you can build up a whole lot of dependencies on libraries that don't have great equivalents on other platforms, and any wholesale migration is going to involve rewriting a mountain of code. But interoperability relies on the same open and de facto standards that everyone else is using these days, so your migration paths are exactly the same as they'd be anywhere else.
I agree. They really try and take care of their customers. I think it was the launch of 2017, and my team upgraded right after the release. (I mean like within hours.) We were having some problems with the upgrade and TypeScript (I think), and it asked me to rate it my experience with it so far. Admittedly, I was in a slightly bad mood and aggravated at the time, and I gave it a low score. They wrote me an email and scheduled an hour to talk to like 2 or 3 people from different product teams about what was wrong and how they could make it better. It's not often when a company does that.
Did anything come out of this? A lot of companies these days are into getting feedback but nothing never comes out of it. I feel like they have departments that collect the feedback and talk to customers with the best intentions but that data barely makes it to the people making the real product decisions.
Whether or not I did anything specifically, I can't say. It did get a lot more stable after a patch or two. I wanted to say that our main problem was it double reference counting the TypeScript files. I think because it was storing a shadow copy of it in another location and that problem went away. (Memory is hazy.)
I do remember complaining that F# was not an automatically included in the language selection like C# and VB.NET. I don't believe they ever fixed that, and I can't remember if it's the same way in 2019. You can't win em all I suppose though. :)
I wish MS open model all the best, but I think MS started seriously pursuing those efforts after it lost a lot of power during and after the dotcom boom. In the PC era, MS supported Windows developers (kind of) as they were seen contributing to Windows ecosystem and viewing others in a not to friendly way. Google for "Linux is a cancer", "GPL is a virus", etc.
Power corrupts. The transition of MS to the good side roughly coincided with Google's divergence from its earlier "Do not be evil" side as it got a chance to influence people on a worldwide scale. My 2c.
Microsoft acquired GitHub for $7.5 billion[1]. If Kubernetes were proprietary, could one value it at $7.5 billion? I think GitHub was acquired for the same reason Kubernetes was open sourced - to gain market share against AWS.
Microsoft has a similar service to GitHub called Azure DevOps[2]. It has been rebranded over the years. My guess is GitHub's brand will absorb Azure DevOps, even as far to rebrand everything in Azure to "GitHub Cloud".
The new war is in the cloud. Microsoft, Google, Oracle, and IBM have to pay a premium for a chance to catch AWS.
I don't think that Azure DevOps (AZDO) will be rebranded again anytime soon, but I agree that it will get a bit more muddled with GitHub. I still think "VSTS" is easier to say than "Azure DevOps". AZDO is a path towards CI/CD with Github, and can include private repos. I do think that GH's path for CI/CD is more interesting though.
I'm not sure where Circle-CI and Travis-CI will end up against the stacked competition though. I think they'll be the biggest losers from GH's acquisition in the end, which would be true regardless of who acquired them.
I feel that MS was probably the best option. Specifically because of the increased skepticism that MS draws despite it's recent moves in the developer community towards more open software. I spent over half my career in an MS-centric world, and the other half outside of it. I'm both appreciative and skeptical of everything that is out there that developers have to make use of.
There are two main reasons I remain skeptical about Microsoft's ambitions.
First and foremost is that they don't seem willing to make a commitment to use software patents only for defensive purposes.
Secondly is Microsoft's refusal to let regular users of Windows 10 fully and persistently disable telemetry.
If Microsoft were to change in both of those ways, without introducing some replacement form of sleaze, I would take their overtures more seriously. But as things currently stand, I see no reason to believe this is anything other than a charm offensive.
I dislike the forced telemetry stuff but, I hate the forced updates even more cos it wastes my limited data.
When developing Windows Phone, Microsoft aped Apple and tended to copy the stupid things from them - such as the locked store model, inability to share apps, developer registration fee...
Now for Windows, they are looking at Google. Ms prob copied the telemetry and forced update idea from Chrome.
Google is a cloudish company and has used open source extensively to get that big. Google tracks every thing, everyone and converts that to cash via ads.
Telemetry and forced updates are as much an Apple/iOS thing as a Google thing if you are playing "Microsoft followed whatever someone else was doing".
iOS has a hard deadline on updates before it forces an install, Windows was just less obvious about the window between when an update was available and when it would be forced. (It's actually very similar, most updates on Windows have exactly a week before they are forced from when they are first "advertised", but most people don't check the "advertisements" so never noticed. Each build of Windows 10 has worked harder to make advertisements more obvious and schedules more clear. Apple has more people respond directly and positively to the advertisements each update cycle.)
iOS and macOS also have as much, if not more, telemetry than Windows. (iOS since Day 1, macOS more slowly.) On a fresh iOS or macOS install you see as many Telemetry opt-in questions as on a Windows install.
A lot of it just comes down to marketing. Apple has more marketing charisma to spare, and more people respect Apple's updates and telemetry. Arguably too, fewer overall users makes fewer obvious complaints, and historically there are fewer reasons to complain about Apple products than Microsoft products.
When did iOS start forcing upgrades? I have phones/iPads on iOS 5,6,9 and 11, and they don’t force anything (I don’t use them daily, only to check some things - but I do turn them on for a day every couple of months)
Most of the early MS open licenses were effectively MIT with a nuclear patent clause that would only be defensive.
Aside from that, the developer tools, azure etc have been under different management from Windows. Windows has specifically been chipped away slowly and methodically. Windows licensing has been reduced and the cost significantly lowered. And while I agree on telemetry, I'm not sure it's significantly worse than say iOS, Android or other OSes these days.
I'm likely to shift from Hackintosh to Linux (Manjaro or Pop) with my hardware refresh coming soon. I'm not a huge fan of Windows. But I don't really blame them too much. All of that said, MS is probably the most developer friendly corporation out there today. That doesn't mean they aren't trying to and don't profit from their positions.
Also the vampire keeps firebombing your house telling you that you should instead live in the house that the vampire is selling, filled with wiretaps and bathroom cameras so he can sell those to third parties for a profit.
remember that movie "mars attacks" how the humans kept falling for the ruse, martians want peace we just had some misunderstandings, lets try again, then the martians screw them?
This whole situation looks like that movie. MS has started convincing people to trust them once again, after capturing sovereign territory. The thing that allowed this was the hope that a centralized service [github] will stay as is forever. this should be the poster child for distributed peer to peer sharing.
I feed my cattle too. If they understood their situation, they would hate and fear humans even more than they hate and fear dogs. They've been bred for generations not to understand their situation...
It’s pretty clear here. Microsoft wins over skeptics, and who exactly are those won over skeptics? Don’t expect satisfying answers in the article. In general, I wish the passive voice was banned from headlines.
> In general, I wish the passive voice was banned from headlines.
I know most people can't detect the passive voice properly, but I'm mystified as to what in this headline even makes you think it's passive. Usually, misidentification of the passive comes from people identifying some variant of the verb "to be."
Microsoft hasn't gotten better. But other companies are now even worse. People used to be scared Microsoft would lock our devices down so we could only run approved software.
Guess what Apple and to a lesser extent Google did that.
Microsoft tried to move in that direction with the windows store, but they didn't go as far.
Microsoft has worsened on privacy but less so than others.
Microsoft has moved more things to the cloud where we can't see the binaries we run but less so than others.
The players in the competition you describe above are all vendors, and we should expect all of them to act according to their interests. Believing too fervently in any vendor will inevitably yield disappointment.
I'm sure that their whole "windows subsystem for linux" revival is actually a long game that started with secure boot.
Give it a couple of years and we'll be unable to boot an unsigned kernel on x86 (as well as arm) at which point the response is going to be "Why do you care? You can boot a linux kernel in windows!"
This may happen for workstations, but I think Linux will continue to be popular on servers. In that context, the cost/benefit of running it on top of Windows won't make any sense. But Windows could very well become the de-facto workstation, even for people who only deploy to Linux servers.
I'm pretty sure WSL is because of the number of server-side tools that don't work well in Windows, and that over half of Azure deployments are non-windows. And that in order to make Azure devs happy, supporting Linux without a fully abstracted VM is best.
Also the huge number of developers over the last few decades that switched full time to macOS and claim having a Unix shell was a big reason. You could very easily see that at any major dev conference all the Apple machines (often with at least one GitHub sticker), including Microsoft's own Xamarin conferences (and now that they own GitHub too, GitHub conferences and sticker counts ;).
That's an interesting audience of developers for Microsoft to try to make happy again on Windows.
True enough... and while I tend to use bash on windows for my work laptop, I'm giving Linux another go as my primary OS at home when I refresh my desktop in a few months.
I'll remain a skeptic until they stop treating their Windows customers like children, implementing deliberately deceptive dark patterns and exfiltrating data on a scale that rivals Google and Facebook.
Funny, I posted that phrase not 4 days ago, in response to a suggestion that Microsoft adopt Firefox. I was harshly downvoted, and every reply was along the lines of "that's a not a thing" and "not this tripe again" and "fake news".
Maybe they really have won over skeptics. In any case, there are other demographics on HN than on display in this thread.
I'm with you. I'm not won over, and how does the media presume to speak for skeptics? There's 0 evidence that any 'skeptics' have been won over, only that pro-MS people are more pro-MS at this time.
Doesn’t Microsoft still control almost 90% of desktops?
OS/2, Commodore, Atari, WordPerfect, Lotus 123, those DOS clones, Novell, BeOS.
They killed Netscape as a company.
Microsoft was a monopoly that certainly did some damage. The only reason they are less feared is because of Google, and because iOS and Android on mobile.
I understand that... I was only wanting to point out that Desktop platforms are a much smaller slice of the overall pie. Not to mention, chromebooks and similar devices will likely continue to eat into the Windows desktop space.
Yeah, I don't know where people get this idea that Microsoft isn't a monopoly anymore. You still can't buy a new computer without Windows on it (unless you trade Microsoft's horizontal monopoly for Apple's vertical one). And the captivity of the audience is evident from the way Windows treats its users these days. Last time I installed Windows it even had the temerity to ask me what kind of ads I wanted in my OS. Not whether I wanted ads - what kind.
I think the cultural shift around Microsoft is connected to a broader cultural shift where software freedom is losing importance. Maybe the old guard that weathered the intolerably closed conditions of the 80s and 90s is dying off, and the new guard, having grown up in an environment where free software is high-quality and plentiful, just doesn't understand what all the fuss is about (not realizing that it was that fuss that got us here today). Or maybe the idealistic countercultural element to hacker culture got swamped by capitalism. All I know is that even as Richard Stallman has never been more right, he's never been less popular.
> You still can't buy a new computer without Windows on it
Yes you can, although you have a much narrower range of machines to choose from. Dell sells machines with Linux preinstalled, and there's always System 76.
In almost all physical stores, I agree. But in my city (yours may have one too), there is an electronics recycler that rebuilds computers, installs Linux, and resells them. They won't be the latest machines, of course, and may not meet your needs, but it might be an option.
I would say that I buy around 90% of my computers from them.
That sounds wonderful. I can honestly say that I stick my head into damn near every store I walk past that has a computer in the Window, and I have never - never - seen one selling a computer with Linux preinstalled (with the minor exception of 2007-era netbooks, which just goes to show what a colossal threat to Microsoft those were). I wish this were not so - I would love to support such a place, with my money and with my time.
(Even if you donated a Linux machine to a thrift shop, I've no doubt they'd wipe it and install Windows 10 in order to make it (in their eyes) saleable.)
Yeah, I guess that I'm spoiled on that front. Should I even mention that they sell almost all sorts of cables and mice for $1 each? They have large boxes overflowing with the things. There are exceptions -- I once had to pay $5 for a 20' long HDMI cable that was still new in the packaging. No, I better not, that would be mean.
The odd thing about when I buy machines from them is that I end up wiping the installed OS anyway -- they install Ubuntu, and Ubuntu and I don't get along very well.
> "The open-source world would’ve rightly looked at us at the time as the antichrist..."
Heavy PR campaign changed some minds, but the underlying issues remain.
Win32 persists but the thinking that brought it here is gone? .Net? All the lawsuits and bad blood aren't on the minds of Microsoft's sponsored CS programs. Not on the Mindcraft players radar at all.
Same code, name name, same culture. Counterclaims are pure PR.
> Same code, name name, same culture. Counterclaims are pure PR.
I don't think that's fair, there's major, objective moves in this direction:
- .NET Core is fully open source and they're moving towards it being the primary .NET implementation
- VS Code
- TypeScript
- React Native for Windows
- Windows Subsystem for Linux
- Broad support for major open source tooling on their Azure services
These are significant and measurable moves of the sort we wouldn't have seen from Microsoft a decade ago. While they may not have opened all their patents, they're also not pursuing implementers of open source software for them. It seems a shift of one sort or another occurred.
Don't forget little things like Calculator now being Open Source and developed on GitHub in the open. In box Windows Apps are now open source, and the Windows team is currently working to migrate the vast majority of the modern Windows Presentation stack (the "UWP" UI system) to open source in the WinUI 3.0 library. (Basically everything above the lowest level DirectX libraries will be open sourced.)
Even if they are playing nice now, MS is a profit-driven corporation and will behave as such. If and when they feel like they have sufficient market leverage to use their positioning to favor their own products, they will. They have a fiduciary responsibility to do so.
In many ways this open source tooling is making them money. Developers tend to use what they know and are comfortable with, if you have to move off of C# say in order to write servers with better Linux compatibility or mobile apps you're going to learn something else and may not stay in their ecosystem.
If instead you use .NET Core for your Linux based services and buy Xamarin for your mobile apps, they're in a lot better condition, especially given that they can offer to host those Linux based services for you on Azure as well as dozens of open source databases and tools which may support them.
They're making a killing off this, many companies have gone to pure Azure services given that they're able to host everything they need there, had they not been able to that money may have gone to Amazon or Google, investing in open source is generally a good move for them financially.
They're now less and less reliant upon everyone using their exact stuff - as long as your corporation still wants a Windows domain controller and MS Office, they're on top of the list for cloud considerations and will easily make money hosting open source software - they lock companies in in other ways than software, like Office 365 and Active Directory login integration with your Azure hosted applications. There's big money in hosting and managing open source software, Amazon and Google discovered this before Microsoft, but Microsoft has been quite competitive given these advantages.
There is nothing wrong with wanting to, and making money. As long as the decisions and direction of a corporation are generally reasonable there's nothing wrong with community interests and the actions of a corporation aligning.
Frankly, I trust MS a LOT more than I do Amazon, Google, Apple, Facebook or Twitter. MS tends to make money when you make money. The rest will bury you to make everything you used to.
As a user even... to my knowledge, MS hasn't actually sold my data to anyone, or manipulated search results in favor of a particular political campaign, or acted in mass-censoring, or established policy that favors a political party.
There's nothing wrong with people holding Microsoft Corporation to a high standard. So long as you also hold Facebook, Google, Oracle, Amazon and Apple to that same high standard.
IMHO, the world has changed since the 1990s and we have more to worry about from some of the others.
I've been a long-time .net developer and the changes in the last few years have felt really refreshing, even as someone deeply in the microsoft ecosystem.
.Net announcements seemed to feel like monolithic messages coming down from on high. Now, I see folks like Scott Hanselman and the asp.net team (Damian Edwards and David Fowler, mostly) being so open about their design decisions as core moves forward. VS Code seems to have been a smash hit from all indicators I can find.
Sure, Windows isn't OSS, and there's a lot of skeletons in the closet. But the amount of dismissal I read here is frustrating given the amount of earnest effort I see engineers putting into getting things right.
I was a teenager when MS was at its worst. In the late 90's and early 2000's. I still remember the Halloween papers. Embrace, extend, extinguish. I know rationally that MS is not that bad anymore. But emotionally its really hard to let it go. At maximum I manage to be "Meh" about whatever MS does, never excited.
I do try to avoid the influence it has on my decision making. But its hard. And its something that likely will never go out of my mind.
On the other hand for the companies that's the price one pays for that behaviour. Humans are not perfectly rational. If you behave like crap there will be people that will harbor grudge until they die, regardless of how you try to make up for past behaviour.
In the end, MS is a profit-seeking entity, as is Google, Apple, etc. It's best to take anything that any tech company does with healthy dose of skepticism, but at the same time it's OK to give them a nod when they do something beneficial. I agree that it can be hard to do this with MS when you are used to Ballmer's more predatory vision for the company.
For the record, I'm very happy with what MS is doing in the development space. I'm less happy with the "cloudification" of Windows and the infiltration of ads there, and I'm not at all happy with how MS continues to fund entities that aggressively pursue patent lawsuits in a sort of protection racket [0]. I'm on the fence (and somewhat nervous) about their embrace of Linux and their growing influence there.
Most people seem to treat technology companies as sports teams. This is foolish to say the least. Every company does bad things, and you shouldn't look past those things, even if you are generally happy with the company's behavior. Ignoring the negatives just ensures that there will be no pushback, and the companies will increase their bad behavior if there is an incentive to do so. At the same time, it's OK to like a good thing that a "bad" company does, and encouraging more of these things may help to shape incentives and behavior over the long term.
I know you know this, but their motto was "embrace, extend, extinguish".
How do we know they have given that up? Many still aren't convinced that non-free software is absolutely a good idea (though it hasn't bitten us yet that much).
Just because they are in the "embrace, extend" phase, doesn't mean they isn't an extinguish coming.
Understand though that "extinguish" may have changed over the years. Only time will tell.
Honestly, I have yet to see something come out of MS' open source work that I actually care about! I'm happy in my non-windows world. What have they even done that applies? TypeScript and VS Code? These are essentially irrelevant to me. So what reason do I have to extend my trust to them?
"How do we know they have given that up? Many still aren't convinced that non-free software is absolutely a good idea (though it hasn't bitten us yet that much)"
Companies are not people. Companies are composed of people and the people working there now on important positions are not the same as were back in the days of old. Most importantly CEO has changed. That's the crucial difference.
On an emotional level I cannot get myself to understand this. Thus I still feel like they are the same as they have always been, basically as if they were human being.
Do you know anyone who works at MS? If so, you would probably know that they have plenty of employees who have been there 20+ years. There are many arms of their business that are steeped in the culture of the 90s: windows, office, etc.
Now, corporate culture can and does change. But, we generally want evidence of change to be proportionate to the original problem/wrong committed. What has MS done that is more remarkable than any other major tech corporation, at this point?
Additionally, what reason do I have to trust them? If I were looking to host my website on a cloud provider, I may look at azure. But people act like MS making a change means we should all start using Windows, writing in office, etc. Why? These are still, for me, substandard tools.
At the end of the day, I am looking for a real, substantial change. Release Office as GPL, then I will really be surprised. So far, they have just done a MS version of what the other players have done:
Typescript: Dart
Edge: Literally a clone of Chrome, which was literally a clone of WebKit, which was literally a clone of KHTML
> Citizens United decided that companies are in fact people
No, it really didn't.
The "people" status of corporations is the same as it has been for longer than you and I have been alive: legally speaking, corporations are not people but do have a small set of rights that are normally afforded people. This isn't a bad thing -- for instance, the right to enter into a contract seems perfectly reasonable.
I just wanted to say I really appreciate the honesty in this comment. I find it truly irritating when people talk about EEE like it was yesterday, and obviously still totally company policy just in secret or something.
There are also companies I admit I will probably never want to deal with or get excited about, even if it's not for a particularly rational or currently relevant reason, but it sticks with me too.
But how to you deal with the fundamental disconnect of worrying about a twenty year gone policy when Google has successfully already completed the extinguish phase? How many truly open source mobile operating systems have died under the foot of Android, an open source Trojan horse?
I mean, heck, look at the hijinks Google has engaged in with Edge after Microsoft agreed to switch to their open source browser engine! Suddenly it would stop working with modern YouTube, and then Google's like "Oops", and undoes it. But ex-Mozilla devs have pointed out that this is a recurring strategy for Google: https://www.zdnet.com/article/former-mozilla-exec-google-has...
Or the impending death of ad blocking in Chrome with Manifest V3 shortly after Google introduced their own ad blocker that doesn't impede their own ads?
Why obsess about the possibility that a company which hasn't extinguished in a while, when their competitors are actively doing it right now? I'd say if your focus is on Microsoft, you're worrying about the wrong target, at least for the moment. The return of evil Microsoft is certainly possible and given enough time, even likely, but evil Google and evil Facebook and evil Amazon are here today.
Oh I am with you 100% about Google, trust me there. But, they aren't the topic of this article. I stopped caring for Google when they dropped the moniker "do no evil" and went into censored search in China at about the same time. I think this was like 2006/7?
What I more dislike is how people seem to think that corporations are inherently moral, just, good, or "on our side". I mean, come on with a title like '"Great Satan" no more', this is clearly a moral conversation.
Corporations will generally act in the most profitable way, sometimes it seems altruistic because there's value in winning customer loyalty or developer respect. But they are fundamentally amoral and should never be trusted to act in your best interests.
For this reason, holding a twenty year grudge against a company is just as illogical as standing by one for twenty years because you think they're the good guys. In twenty years, any given business' strategy has probably shifted five times.
If not more! My main concern is it shifting back. MS made a TON of money, and was at the height of its glory in the old ways. That kind of thing leaves a mark that fades very slowly.
I've always been worried about how centralized the FLOSS community has become w/ Github. Now its even more of a problem!
I'll be the first to admit that I have a emotional dislike for MS. However, I still have reasons for why I think its a bad idea to for the industry to trust them, aside from my personal feelings. For a long time I had generally positive feelings toward Google. I still generally like Github (though they have their share of problems) while acknowledging it is problematic for the larger community.
Magnitude of correction needs to be proportionate to magnitude of wrong. I don't think anyone would disagree that MS is more friendly towards FLOSS. But, does that mean they are suddenly "the good guys", as many seem to say?
MS is the only company I know of that actively tried to kill FLOSS and had a motto similar to "embrace, extend, extinguish". This makes me seriously worried about trusting the company in any serious way while they seem to be "embracing" by purchasing one of the most important companies in FLOSS and "extending" by making VS Code/Edge.
The world doesn't just end up good. We make choices that influence others and influence outcomes.
What seems more bizarre to me is why are there so many people eager to ignore MS stated historic policies and give them a free pass for the damage they done just because they've done the minimum viable amount to appear friendly to open source?
VScode is not open-source. Even when Microsoft do open source, it's not open source... It's bundled with closed source proprietary telemetry/spyware. If anyone care, there is VSCodium which is open source
Not really, or rather it very much depends on MS subdivision. Their browser and Alliance for Open Media efforts are commendable, but their gaming division with its DirectX lock-in straight from the '90s is still as nasty as it gets. At least they dropped their obnoxious UWP push. But let them start supporting Vulkan on Xbox, then they can come back claiming they aren't that bad anymore.
Also, don't forget their other cash cows that thrive on lock-in, such as Outlook, Office, exFAT and etc. Opening up stuff there could indeed take MS out of their dark empire stigma into something positive. But that didn't happen yet.
I switched to Visual Studio Core on daily basis for Go and Ruby.
I switched to Typescript for everything JS including React Native.
I was already using Github and was ok with it, now under Microsoft it feels like it's moving faster and actually going somewhere.
I'm in the market for a Windows-based PC since I can get Linux with it courtesy of WSL2. Probably keep using my current crappy MBP until the ARM ones come out.
With every release of .NET Core I feel tempted to use it instead of Go.
How does all this translates to me wanting to try Azure Cloud the next time I need to deploy something? I feel stupid because I don't know, you may need to ask Microsoft.
No, they didn't. Yes, I use VSCode, but that is it. And still, I have zero trust on Microsoft with it. I also still use github, but the first sign of misstep will give me the little push I need to migrate of off it.
> first sign of misstep will give me the little push I need to migrate of off it.
To the company that will own GitLab? The one that blocks ad-blocks in their browser? The one that primary revenue comes from advertisement and user tracking?
That's correct. I don't care about these issues. The repos are open source, so track away. That's a large part of what my code doest, ironically. I work in Adtech and I have no problem with adblock-blocking. YMMV
"Embrace, extend, extinguish" can be thought of as pattern of incentives, not just a policy that might be renounced.
Embrace. We care about our developers and customers. If someone somewhere does something that would be better for them than what we're doing now, of course we're going to pick it up and adapt it for them. Failing to do so would be to fail them.
Extend. We care about our developers and customers. We want to improve the experience we provide them. Sometimes that involves extending standards with additional functionality. Just as everyone else does. We're not going to deny them those improvements while standards processes grind slowly, and especially not when those are retarded by competitors with interests other than the best well being of our developers and customers. And we value maintaining backwards compatibility. [...] Similarly for community infrastructure.
Extinguish. We care about our developers and customers. If we can help that community grow and strengthen, attracting new members from elsewhere, of course we're going to do that for them. Helping competitors, their developers, and their customers, isn't where our focus is. Resources are limited. And if competitors think to steal our hard-earned intellectual property, of course we're going to stop them. (Ok, that last bit is policy.)
"Windows Everywhere"... and scorched earth everywhere else. That systematically torching the surrounding valley, now has less prominence as policy, does seem an improvement. It remains to be seen how different the effects are, of say allocating available water based on the best interests of our developers, customers, and competitive advantage.
One question is what Microsoft is willing to sacrifice for the common good. Taking a hit on DOC, for example, to avoid undermining ISO. Sacrificing the competitive advantage of offensive patent use. There doesn't seem much track record yet, in prioritizing the broader there's-more-here-than-Microsoft ecosystem and community and profession.
Community-scale bulk analysis of code and development artifacts is now finally arriving. One issue with the open-source community's choice of GitHub over a more open provider, was the leverage to help or hinder access to that data. The issue continues, and manifests as what portion of GitHub data can be downloaded, under what licensing terms, and using what tooling. It would be unsurprising to find those "limited, and less than Microsoft is using for it's own value-added products", "restrictive of competition, and thus of innovation", and "Azure-only".
I never in a million years thought a day would arrive where I would be supportive of Microsoft and skeptical/suspicious of Google, but yet that’s exactly where I am.
Looks like everyone's forgotten about EEE again[1]. MS is simply going through the same motions as the last dozen times they wanted to take down a competing technology.
Microsoft has certainly not "won over" skeptics; if anything, the opposite.
Maybe Microsoft hasn't changed that much, but others in the industry do way worse?
I remember late 90s Microsoft phones home via one web request and all hell broke loose that they try to track everyone and take total control of everything.
Then Google came along and did just that, and noone complaint because it was free (as in beer).
Let’s stop idolizing or demonizing any big companies. Basically they are all the same. They may look friendly or hostile from time to time. But in the end they are not your friend and should be viewed with skepticism because of the power they gain through their size and wealth. They can cut the little guy off whenever they feel like it and can trample on people when they want to.
I think we would be better off if we had more smaller companies but I guess capitalism lends itself to power concentration.
I couldn't agree more. Choice and competition are good for consumers.
These days because of what I happen to be working on, it's more Apple that's getting in my way than Microsoft (because of Vulkan vs. Metal).
This article raises my hackles, because it is selling a vendor as benign. It's wrongheaded -- skepticism is always appropriate when fundamental interests are at odds.
Agreed. Embrace, extend, extinguish isn't a warning against Microsoft, it's a warning against a pattern of behavior that any corporation may attempt at any time it seems beneficial and likely to succeed. And it's a warning not about the first step (embracing a technology or standard is purely beneficial), but the second (extending it in proprietary ways for lock in).
So keep an eye on every embraced technology of Microsoft and everyone else, and when you see someone extending it in incompatible ways, make sure you have an escape plan. The obvious example these days? Google Chrome.
A quote from Chris Hitchens about religion springs to mind:
"Many religions now come before us with ingratiating smirks and outspread hands, like an unctuous merchant in a bazaar. They offer consolation and solidarity and uplift, competing as they do in a marketplace. But we have a right to remember how barbarically they behaved when they were strong and were making an offer that people could not refuse."
Not sure how relevant it is but Microsoft’s acquisition of github has kiled atom’s written-in-rust successor xray. This along with what feels like slower updates and less focus on atom feels very much like embrace extend extinguish. So in my books Microsoft is still quite evil.
I haven’t switched yet but you are correct the general consensus is that vscode is superior to atom. That said Microsoft slowly suffocating vscodes largest competitor still feels like extinguish.
Here’s to hoping Xi does well and delivers on its promises
I don't think this phrase applies for an acquisition. It's more like "(1) buy. (2) do whatever suits my agenda. (3) no step 3."
De-emphasizing atom doesn't sound evil to me, really. Re-licensing it with some crazy terms like "you can't say bad things about MS or sue MS" would be evil. But you can still use Atom and modify and distribute it with relatively few restrictions from MIT license. Just the feature pace sounds like it's slower. Maybe that's natural with an acquisition?
I think Embrace Extend Extinguish applies perfectly here, but with respect to git and not github. Github has already become the way that a vast majority of people interact with git, and it already has lock-in potential with the value-added features it implements: I've worked at companies which use github almost exclusively for project management and issue tracking, and it would be a huge expensive burden for them to change to something else.
Owning GitHub puts Microsoft in an excellent position to advantage their own products, and disadvantage competitors. Given Microsoft's history, I can not see it as anything other than foolish to assume this time they are going to act altruistically.
"To repeat the same experiment and expect a different result is the definition of insanity."
The endless Microsoft bashing in this thread makes me so sad, I think they've done really well and put a lot of work and effort behind it.
VS Code, Ts, .NET Core, WSL (v2 soon), Github (yes I know they merely bought it, but changes like allowing free users private repos wouldn't have happened without MS) among others are all quite amazing.
It's foolish to anthropomorphize a corporation and have pity for it. A publicly traded company is an entity which is legally required to maximize value for its shareholders. Microsoft is "putting a lot of work and effort" into open source software because, by their calculations, this is good for the bottom line. When that calculation changes, they will certainly be less developer friendly.
Don't forget: nothing is for free. If a corporation is pouring resources into something and you're not paying up front, someone's going to pay for it sooner or later.
I appreciate skohan's perspective and their comments in this thread very much. It won't take much to adapt the points to be technically correct while preserving the gist.
with Microsoft ownership of GitHub, they have effectively sized "open sources" means of production. They now own the defacto superhighway that much of "open source" reproduces itself.
Is GitHub becoming less a free host for git projects (not that they're suddenly charging) and becoming more a capital accumulator of free labor, enclosing open source software behind an invisible fence?
my concern is that github is becoming a marketplace where consumers of open source software (corporations, etc) can accumulate labor by reducing their own costs, they do this by externalizing the cost of developing software by appropriating it from open source projects.
it both makes it easy to for for-profit companies to externalize the cost of producing software and cheap (free) to consume OSS, and it drives the perception of the value of software to to nothing (the actual labor-time of the software creation is quite high). Much of the OSS labor being done is subsidized by the contributors paid day jobs.