Microsoft loved to act like it had the best programmers programming applications for their platform, while in reality Microsoft had made huge strides in dumbing down programming into Visual Basic [&&.NET]
Which is another way of saying they created some libraries and languages that just did a lot of work, but I don’t see anything convincing me that it was “dumbing down” programming. The whole article seems to take a negative stab at the company.
(also holy shit does this author shill the Innovator’s Dillemma pretty hard)
Right off the bat, the author’s writing style is annoying:
The article is also broken out with the historical context, then I’ll dig into the actual dichotomy of the juxtapositions that are in evidence.
Not the best use of the language, and using expensive words where cheap ones will do. That’s the only example I’ll give of the writing style, and will now just focus on the substance. I’m just going to barrel-ahead with some examples and arguments.
~
The author deeply, deeply wants to deny MSFT any credit for innovation.
The history is pretty clear however, that Microsoft has resoundingly been fairly poor at innovations that come forward and stick.
Other than, say, NTFS? Or AJAX#History)? Or Direct3D? SMB? COM? C#/LINQ? Visual Fucking Basic?
We can all argue about how much those innovations are “Microsoft’s”, but the fact is that they threw serious money, time, and engineering and testing talent to get those widely deployed, and they’ve managed to stay relevant for, in some of the above cases, almost 30 years.
Most companies these days innovate and expand their product lines through aquihires, so if we want to complain about MSFT for that we need to similarly complain about Google, Facebook, IBM, et al.
If the author wants to chide MSFT for not “making things stick”, by pointing out the number of their technologies that didn’t work out, then they need to acknowledge the massive number of false starts in the 90s by Apple and other companies. Apple couldn’t save itself no matter how much engineering flights of fancy they went on, see also the Newton and various failed operating system attempts.
Windows was largely inspired by, with some slight customizations, from the work done at Xerox PARC), a research and development company.
As was Apple. Additionally, the author completely fails to mention–probably because they are a hack–the OS2 debacle between IBM and MSFT that resulted in some parts of the look-and-feel of Windows as it evolved. Or other interesting things that had been happening in the GUI space, like anything involving X-Windows or the Blit).
The author fails to capture just how well MSFT supported developers who wanted to build things for Windows, especially when compared to other vendors (cough Apple cough) at the time. Or even touching on things like Windows CE that were hugely impactful for embedded deployments.
Azure - Nothing really original in this space at all, it’s almost entirely derived from Amazon Web Services (AWS) and related cloud computing technologies. The first release of Azure was a complete and utter failure and purposely glossed over by anybody involved in cloud computing at the time.
As opposed to the runaway success of GCE, presumably. Even AWS took a while to take off. The same arguments about Azure not being an innovation could probably apply to early versions of AWS–i.e., nothing really innovative, mostly just productizing shell scripts and storage and DB management and deployment stuff that big companies had already been doing on their own for years quietly.
MSFT’s Azure was perhaps not an amazing “leap”, but it’s a pretty impressive accomplishment seeing as how only a couple of companies have even tried to do that kind of thing at scale and on-demand–and on an effectively non-supported stack! Everybody and their dog can spin up and spool down Linux VMs, but MSFT had to find ways of supporting (however roughly) the massive enterprise stack they’d developed and that others had built on. That’s not a small thing.
~
And then we get to the bullshit about “innovation” and other dumb misunderstandings of the industry by this author.
Microsoft was years late to this game (don’t even get me started on the laggard luddite beast that Oracle is!).
If the author can’t appreciate that a) Oracle makes dumb amounts of money and b) Oracle singlehandedly killed Sun via absorption, there’s not much going on between their ears.
The basic premise is, big companies simply do not innovate or invent without either creating small breakaway companies (i.e. like startups), research entities, or some other organization that can operate almost entirely outside of the large parent corporation.
Perhaps not entirely the authors' fault, but this is such a bullshit definition. “Company’s can’t innovate, because the innovative parts of the company are, according to my definition, disjoint from the rest of the company”. You know, like how AT&T deserves not credit for Bell Labs. Or how IBM totally has no constant creation of patents. Or how HP and Silicon graphics totally didn’t make the STL for C++ or a lot of good systems work. Or how…oh wait no this is just a terribly stupid point by a terribly ignorant person.
The company has spent most of its history creating this “maintain the status quo“ type of ecosystem and employee base, in spite of the employee’s efforts sometimes to push forward the company into a leader instead of a massive monolithic follower.
Bullshit. Go read Renegades of the Empire and tell me about how there is no internal innovation at Microsoft. Sure, at the end of the day promising stuff might get killed off, but there was a time
Also, the author kinda tips their bias by saying “status quo” instead of “solid interfaces so we don’t have to keep rewriting working software”. Only a fool or a salesman sees no benefit in providing such a thing.
~
And then the author spins off into this “crossroads” business.
Microsoft’s Azure still has some significant features to overcome to be best of breed.
Being the “best of breed” has never been the Microsoft business model. Being the “largest market share” has, and it’s pretty obvious that the Azure goal isn’t “herp derp let’s compete with AWS and GOOG for startup funbucks” but instead “hey, let’s transition our mind-bogglingly fuckhueg customer base onto the next generation of Windows tech without disrupting their businesses”. Azure doesn’t have to be good at anything other than running the good Microsoft tech that companies are already using, and integrating with what they already have–something nobody but MSFT seems to give a shit about.
This type of misleading information is used to push specifically these types of decisions.
Author just discovered how marketing works, is confused and frightened.
honestly, the Internet itself was nothing more than a fad for Microsoft for so long it’s been a late comer of competence for the company
Author is missing basically the entire focus of MSFT in the mid to late 90s and perhaps even early 2000s. Even though MSFT wasn’t always great (cough active desktop cough), saying that they thought the Internet was a fad is hugely ignorant. MSFT’s own internal byzantine politics (see Project Chrome vs. the IE team) and external anti-trust stuff (well-deserved) caused them to have trouble being on the forefront of ‘Net technologies, but the sheer number of web pages ending in “.asp” gives a clue that they weren’t sitting around doing nothing.
~
And the bits about community:
The community could just ignore the “softies” and fanbois of yesteryear but Microsoft actually needs to get these individuals to update their viewpoints, get with the program, and step out of the ways they were previously taught in order for Microsoft to actually get ahead of itself.
What does this even mean? The vast majority of “boring” software is written on MSFT-dominated systems and platforms these days, regardless of what us folks in startupcanistan may think. These “myopic” devs are actually the bread and butter for MSFT, and will keep it in business and in the black for quite some time–especially since few other companies seem to give a shit about providing a stable technology base for the working dev (cough docker npm mongo cough).
This Microsoft Way™ way introduced much of the computer scientist community to a new form of software developers that was dubbed the mort.
Whoa whoa whoa the “computer scientist community”? You mean the same grandiose community that led that well-known MSFT fanboi Rob Pike to conclude that Systems Software is Irrelevant? The same community that is sucking on the teat of Linux and half-warm ‘nix derivatives and tools as hard as they can, trying to spelunk the same boring chunks of distributed systems research and type theory? That community?
Oh, and here I was thinking we were talking about industry, software engineering–you know, the place where code actually gets shipped to users.
while in reality Microsoft had made huge strides in dumbing down programming into Visual Basic [&&.NET] (and hey, I wrote some amazeballs stuff with Visual Basic too, shoutout to SCP Pool Corp!) and other tools
So, at a time when our industry is trying its hardest to open up the field of development to as many folks as possible, we are supposed to suddenly believe that MSFT’s leading the way at scale on this is somehow bad? That MSFT recognizing that their customers had more important things to do than wank about the “craft” of software development was somehow a bad strategy? Are you fucking kidding me dude?
The problem with this is Microsoft has also built up a huge number of developers outside of Microsoft inside corporations that latch onto the Microsoft Way™.
Why, as a business seeking to maximize userbase, is this a problem? The author goes on to suggest that they are going to be a drag as MSFT realigns itself (somewhat of a fool’s errand, in my view, but still)–unfortunately, they then go on to list a bunch of HN-tier technologies that, outside of startupcanistan, aren’t really super important:
They’ve taken huge hits from any number of technologies: Ruby on Rails, Java, Node.js, JavaScript, Cloud, etc.
Really? How much marketshare has MSFT actually lost to RoR, Java, Node.js, JS, and Cloud (whatever the fuck that it)? A better author would link to figures here, but we can’t let statistics ruin a perfectly dumb point.
They still have a huge problem with their internal and external demographics. It’s almost entirely older males, and often from only one subset of society. Contrary to this the industry itself (also in spite of itself) is slowly but surely becoming more diverse: from men and women to cultural differences to age ranges.
Oh fuck you and your limp lip-service to diversity. Again, explain how that is somehow different than the old greybeards running ‘nix systems, or the young dudebros killing in on AWS brah with docker and macbooks. Explain how all of the nice programmers in SE Asia and India are all just white men. Explain how Microsoft is losing market share there. Fuck. You.
This is the reason why startups have their pick of top tier talent before Microsoft (and often before Amazon or Google) even though they offer slim chances at riches and very little stability.
Ahahahaha no. Some startups, flush with cash from investors, might have their pick of some good talent. Most startups are looking for the good enough programmer to get things out the door without being too large a cost-center (hint: the same as any other business).
~
And after this excellent review of historical events? After this thorough analysis of the market and trends? The author gives us these gems:
Microsoft should get serious about downtown Seattle office real estate and becoming a real prominent SEATTLE company. Ideally they ought to build a tower,[…snip…]
Why the fuck would MSFT want to do that? Why would they want to take on a multimillion (more likely billion) dollar construction project that would worsen the commuting burden of their existing employees, create a half-finished asset on their books for years to come, and take on a bunch of real-estate and property hassle when they already have many functioning campuses and offices throughout the world?
This is the most nonsensical thing in a long, long article of nonsense.
Do NOT dictate, which Microsoft hasn’t been doing much of this for years now, but don’t start and get rid of any notions that the company should dictate how a thing should be.
I mean, except for the part where this was their exact business model during their most successful period. And where, when they prioritised maintenance and testing, it made a hell of a lot of sense, and why the MSDN is leagues ahead of whatever crap manpages we get on linux (OpenBSD is p. cool though).
Build software and tooling as stand alone entities. Stop trying to build software to lock one into other software.
The author clearly doesn’t understand the value of an integrated stack. If I’m stuck developing for Windows, I at least have the solace of one of the best IDEs known to man (Visual Studio), and a bunch of tooling that plays really well with it. The dirty secret of the Unix Way is that it results in software that is equally flexibly bad for every system you want to use it on–some walled gardens (just ask any iPhone dev!) have their advantages.
This is super abstract and kind of out there, but seriously Microsoft needs to figure out marketing.
Clearly the author never heard of the vagina monster used to promote DirectX, or any of the crazy shit Alex St. John and other evangelists pulled.
~~~
Seriously, this article is just garbage, making specious arguments based on a poor understanding of an incomplete history. This whole thing was bad, and the author should revise or delete it.
Which is another way of saying they created some libraries and languages that just did a lot of work, but I don’t see anything convincing me that it was “dumbing down” programming. The whole article seems to take a negative stab at the company.
Gees, what a rambling mess of an article. I couldn’t even read enough to figure out where he’s going with it.
This is one of the stupidest goddamn things I’ve ever read.
Once I have a proper tool with which to display my displeasure–e.g., a keyboard–I will at length back up that assertion.
(also holy shit does this author shill the Innovator’s Dillemma pretty hard)
Right off the bat, the author’s writing style is annoying:
Not the best use of the language, and using expensive words where cheap ones will do. That’s the only example I’ll give of the writing style, and will now just focus on the substance. I’m just going to barrel-ahead with some examples and arguments.
~
The author deeply, deeply wants to deny MSFT any credit for innovation.
Other than, say, NTFS? Or AJAX#History)? Or Direct3D? SMB? COM? C#/LINQ? Visual Fucking Basic?
We can all argue about how much those innovations are “Microsoft’s”, but the fact is that they threw serious money, time, and engineering and testing talent to get those widely deployed, and they’ve managed to stay relevant for, in some of the above cases, almost 30 years.
Most companies these days innovate and expand their product lines through aquihires, so if we want to complain about MSFT for that we need to similarly complain about Google, Facebook, IBM, et al.
If the author wants to chide MSFT for not “making things stick”, by pointing out the number of their technologies that didn’t work out, then they need to acknowledge the massive number of false starts in the 90s by Apple and other companies. Apple couldn’t save itself no matter how much engineering flights of fancy they went on, see also the Newton and various failed operating system attempts.
As was Apple. Additionally, the author completely fails to mention–probably because they are a hack–the OS2 debacle between IBM and MSFT that resulted in some parts of the look-and-feel of Windows as it evolved. Or other interesting things that had been happening in the GUI space, like anything involving X-Windows or the Blit).
The author fails to capture just how well MSFT supported developers who wanted to build things for Windows, especially when compared to other vendors (cough Apple cough) at the time. Or even touching on things like Windows CE that were hugely impactful for embedded deployments.
As opposed to the runaway success of GCE, presumably. Even AWS took a while to take off. The same arguments about Azure not being an innovation could probably apply to early versions of AWS–i.e., nothing really innovative, mostly just productizing shell scripts and storage and DB management and deployment stuff that big companies had already been doing on their own for years quietly.
MSFT’s Azure was perhaps not an amazing “leap”, but it’s a pretty impressive accomplishment seeing as how only a couple of companies have even tried to do that kind of thing at scale and on-demand–and on an effectively non-supported stack! Everybody and their dog can spin up and spool down Linux VMs, but MSFT had to find ways of supporting (however roughly) the massive enterprise stack they’d developed and that others had built on. That’s not a small thing.
~
And then we get to the bullshit about “innovation” and other dumb misunderstandings of the industry by this author.
If the author can’t appreciate that a) Oracle makes dumb amounts of money and b) Oracle singlehandedly killed Sun via absorption, there’s not much going on between their ears.
Perhaps not entirely the authors' fault, but this is such a bullshit definition. “Company’s can’t innovate, because the innovative parts of the company are, according to my definition, disjoint from the rest of the company”. You know, like how AT&T deserves not credit for Bell Labs. Or how IBM totally has no constant creation of patents. Or how HP and Silicon graphics totally didn’t make the STL for C++ or a lot of good systems work. Or how…oh wait no this is just a terribly stupid point by a terribly ignorant person.
Bullshit. Go read Renegades of the Empire and tell me about how there is no internal innovation at Microsoft. Sure, at the end of the day promising stuff might get killed off, but there was a time
Also, the author kinda tips their bias by saying “status quo” instead of “solid interfaces so we don’t have to keep rewriting working software”. Only a fool or a salesman sees no benefit in providing such a thing.
~
And then the author spins off into this “crossroads” business.
Being the “best of breed” has never been the Microsoft business model. Being the “largest market share” has, and it’s pretty obvious that the Azure goal isn’t “herp derp let’s compete with AWS and GOOG for startup funbucks” but instead “hey, let’s transition our mind-bogglingly fuckhueg customer base onto the next generation of Windows tech without disrupting their businesses”. Azure doesn’t have to be good at anything other than running the good Microsoft tech that companies are already using, and integrating with what they already have–something nobody but MSFT seems to give a shit about.
Author just discovered how marketing works, is confused and frightened.
Author is missing basically the entire focus of MSFT in the mid to late 90s and perhaps even early 2000s. Even though MSFT wasn’t always great (cough active desktop cough), saying that they thought the Internet was a fad is hugely ignorant. MSFT’s own internal byzantine politics (see Project Chrome vs. the IE team) and external anti-trust stuff (well-deserved) caused them to have trouble being on the forefront of ‘Net technologies, but the sheer number of web pages ending in “.asp” gives a clue that they weren’t sitting around doing nothing.
~
And the bits about community:
What does this even mean? The vast majority of “boring” software is written on MSFT-dominated systems and platforms these days, regardless of what us folks in startupcanistan may think. These “myopic” devs are actually the bread and butter for MSFT, and will keep it in business and in the black for quite some time–especially since few other companies seem to give a shit about providing a stable technology base for the working dev (cough docker npm mongo cough).
Whoa whoa whoa the “computer scientist community”? You mean the same grandiose community that led that well-known MSFT fanboi Rob Pike to conclude that Systems Software is Irrelevant? The same community that is sucking on the teat of Linux and half-warm ‘nix derivatives and tools as hard as they can, trying to spelunk the same boring chunks of distributed systems research and type theory? That community?
Oh, and here I was thinking we were talking about industry, software engineering–you know, the place where code actually gets shipped to users.
So, at a time when our industry is trying its hardest to open up the field of development to as many folks as possible, we are supposed to suddenly believe that MSFT’s leading the way at scale on this is somehow bad? That MSFT recognizing that their customers had more important things to do than wank about the “craft” of software development was somehow a bad strategy? Are you fucking kidding me dude?
Why, as a business seeking to maximize userbase, is this a problem? The author goes on to suggest that they are going to be a drag as MSFT realigns itself (somewhat of a fool’s errand, in my view, but still)–unfortunately, they then go on to list a bunch of HN-tier technologies that, outside of startupcanistan, aren’t really super important:
Really? How much marketshare has MSFT actually lost to RoR, Java, Node.js, JS, and Cloud (whatever the fuck that it)? A better author would link to figures here, but we can’t let statistics ruin a perfectly dumb point.
Oh fuck you and your limp lip-service to diversity. Again, explain how that is somehow different than the old greybeards running ‘nix systems, or the young dudebros killing in on AWS brah with docker and macbooks. Explain how all of the nice programmers in SE Asia and India are all just white men. Explain how Microsoft is losing market share there. Fuck. You.
Ahahahaha no. Some startups, flush with cash from investors, might have their pick of some good talent. Most startups are looking for the good enough programmer to get things out the door without being too large a cost-center (hint: the same as any other business).
~
And after this excellent review of historical events? After this thorough analysis of the market and trends? The author gives us these gems:
Why the fuck would MSFT want to do that? Why would they want to take on a multimillion (more likely billion) dollar construction project that would worsen the commuting burden of their existing employees, create a half-finished asset on their books for years to come, and take on a bunch of real-estate and property hassle when they already have many functioning campuses and offices throughout the world?
This is the most nonsensical thing in a long, long article of nonsense.
I mean, except for the part where this was their exact business model during their most successful period. And where, when they prioritised maintenance and testing, it made a hell of a lot of sense, and why the MSDN is leagues ahead of whatever crap manpages we get on linux (OpenBSD is p. cool though).
The author clearly doesn’t understand the value of an integrated stack. If I’m stuck developing for Windows, I at least have the solace of one of the best IDEs known to man (Visual Studio), and a bunch of tooling that plays really well with it. The dirty secret of the Unix Way is that it results in software that is equally flexibly bad for every system you want to use it on–some walled gardens (just ask any iPhone dev!) have their advantages.
Clearly the author never heard of the vagina monster used to promote DirectX, or any of the crazy shit Alex St. John and other evangelists pulled.
~~~
Seriously, this article is just garbage, making specious arguments based on a poor understanding of an incomplete history. This whole thing was bad, and the author should revise or delete it.