AWS has been historically ripping off open source projects, forking them, and building new products without any attribution or support. The premise that they are the best cloud provider is kind of tainted to me, since they aren't building anything from scratch but rather, taking the work of thousands of volunteers and incorporating it into a paid for service. Examples would be in addition to DocumentDB (an ironic name since Azure called their NoSQL offering that first), Redshift, Maria, and even their hypervisor.

Then you have Mongo who builds a product and markets it like it's the cure to cancer, and early on had a lot of success in convincing people they needed it (they mostly don't -- read up about Diaspora when you find time). Now that they are in the position of being outed for their proprietary connection interface (which is what makes AWS offerings compatible) they are changing the licensing. I don't think Red Hat's move hurts them as much as people don't want to maintain a fucking Mongo cluster any more.

So who do I feel sorry for? Fuck both of them. Azure and GCP are better cloud providers anyway -- Azure for the enterprise, GCP for 'startups' that don't need the enterprise tooling. If you start your business on AWS it's because you ride the bandwagon either have enough money to burn, or are stupid.

I thought that people were over the "M$" as many folks have grown up in a different era but clearly, the old folks still haven't grown up at all yet. Tools are less important than business outcome and generally tends to favor Windows vs Linux. But hey, I guess the "M$" jokes or talk about how it costs money (so does losing time, by the way) is just a bit old considering I've been on this site for more than 20 years.

In an enterprise there are two big costs... licensing, and support.

Linux the cost of support is pretty high -- for models like Red Hat the cost often is higher than Windows because you don't get as high a per-seat discount. Then there are the other ancillary costs like productivity, accessibility, data governance, etc... which are harder to materialize but also make an impact.

Ultimately most organizations use Windows because it meets the needs for those things that are ancillary while also staying competitive with the two larger costs of support and licensing.

This isn't a religious conversation much as Slashdot would like it to be; I am a big fan of the best tool for the job and on the desktop sad to say, Linux still doesn't do the job. Server side the uptick is huge, which is also why Microsoft products like SQL Server or .NET Core can literally run on Linux, and are supported in Docker. Microsoft saw the light because being agnostic is financially rewarding, and lock-in doesn't work. But the desktop? Still the realm of Windows.

Nothing of value was lost.

Just seems a weird choice to put in the article.

Yes, and if you bothered to dig through the whole history rather than picking up a few comments about me "liking" Microsoft (which I generally do, as far as tools go), you'd realize I worked in finance (and still do today).

And despite that -- does that make my comment less wrong? Or my previous comments? I won't deny that I'm in a minority on Slashdot who actually likes Windows AND Linux because they are good tools for specific things. My entire career I've treated things like tools and as a result, been pretty successful. It's the people who get 'religion' about platforms that don't inroads to success, and are relegated to doing sysadmin work for the entirety of their careers.

I'm a realist, and cursory glancing at my comment history aside from my thoughts on Windows phones which turned out dead wrong, though I might yet be proven right in the incarnation of Windows 10, I don't think there is anything there that says I hate Linux or open source. It's always been the best tool for the job.

So again, your point was?

Why would a startup go with an MS solution? Because they want to turn around the software quickly, have easily available programming talent, and high quality products?

I mean not for nothing, but .NET is an amazing language suite -- this is noted by almost anybody who is respected by the Slashdot community. .NET skills are readily available for hire, the IDE is top notch, and the ability for you to execute an idea is pretty darned fast compared to "open" languages.

But that's just the languages. SQL server on Azure is braindead simple to use, and cheap to boot. So are a lot of the PaaS apps for things like websites, caching, etc. And realistically the "cost" isn't really in the software, almost EVER. It's in the development time. And if you can shave development time by going with an MS stack then you'll do it. Gone are the days you have to pay big $ for a license for SQL or Windows... it's cheap as dirt now. And cloud makes it accessible. While AWS has the hearts and minds of startups, as engineers understand the landscape better Azure will catch on more and more. They offer you the ability to speed up your entire development process, a claim that AWS cannot commit to because the management of all the overhead (storage, network, backup, monitoring, etc) while automated, is still an overhead you don't have to worry about at ALL on Azure if you implement properly.

So yeah... if I was doing a startup today, I'd do it in .NET and on Azure.

Which I expected to see on Slashdot in a laughably fast manner did not disappoint.

Whilst the anti MSers here continue with that tirade, the rest of us are noticing a differentiation in how MS is conducting is business. It's not selling software any more -- it's selling services. Office 365 and Azure are pretty much the key to this differentiation. They want to build their partner ecosystem (this is what they've been pushing heavily for the last year+) and allow their partners to resell not only Azure, but the PaaS offerings Microsoft has built and is building.

If you've seen "Field of Dreams" this is the Microsoft version of "If you build it they will come." They are building the future of deploying applications to the cloud, and managing everything throughout. They are going to integrate with everybody, they will make their own software a commodity and use that benefit of wide integration to drive it home in terms of operational benefits. It means developers can *just develop*. They won't have to worry about infrastructure, networking, etc.

Compared to AWS, Azure is a far less configurable but far easier to manage platform. AWS builds all of the automation they offer into a base of virtual machines that still need to be managed on a storage, network, and VM level. Azure offers that with less configurations (ie, less machine types) but also offers you abstraction from all of it via their PaaS services. The only thing AWS has to offer in that space is Beanstalk and to be honest, unless you're running a lot of Java services it's not that useful.

This is the future of Microsoft, in my opinion. You can think it's "embrace extend extinguish" but since all of their offerings are open source and they are making a hell of a lot of OS contributions, I think the simplicity of the hate has to be expanded a bit to think what MS could be doing to make money given their moves recently.

Because they are losing game developers out to Azure. Look at Titanfall, for an example (whether you like it or not is irrelevant).

The problem AWS has is that it is entirely VM/IaaS based. There is little to no PaaS offerings so developers have to write integration layers for the VMs and at many times, use more resources than they need. With PaaS services they can tailor it to be much more dynamic and that's the allure Azure brings them. The scaling can be elastic but also not waste any resources.

AWS is providing an engine that is really powerful, adding its customization so that the integration for Twitch and AWS is inherent, and hoping you sign on. Because if you do, you're going to pay more money to run your game and as a result, they will make money as well. That's why they can give it away and to boot -- not open source it.

That said I think most people would take the power, flexibility and ease of use of the Unreal Engine for 5% revenue share rather than pay 20% more on utilization on AWS. But that's just my logic talking.

Glad to share, I always enjoy learning things myself so happy I could pay it forward :)

Cloudformation is piss poor compared to Resource Groups in Azure. The idea is similar but you can deploy 'stacks' of things with permissions for role based access as well, very easy administration.

As for the Lambda stuff -- that's WebJobs in Azure, and far more powerful (takes in .exe, .sh, .bat. cmd, etc) whereas Lambda does .zip and .js only. Beanstalk integration into .NET environments (which is what we are primarily) is very bad compared to WebApps, and MapReduce is pretty commonly usable on either cloud. If you use SQL the SQL Azure is a better offering than RDS, but if you're on PostGres/MySQL then definitely a point in the RDS favor. Dynamo is also available on Azure in another form, as well as API aggregation, mobile integration, etc.

Long story short is that I don't think that one is better than the other outright -- it's about choosing the tools that make sense for your environment and developers. For us that seems to be Azure given a heavy SQL/.NET platform and developer base. For others it may be AWS for other reasons. I just think that as Azure expands its capabilities, and some of the things they wind up doing a lot better than AWS, as well as the stuff they do that AWS can't do at all (ie micro services architecture support in Service Fabric, Machine Learning that is far beyond AWS' capabilities, etc), then AWS gets into a race to the bottom. Not to mention the segregated storage in AWS that they are working on consolidating (ie, the S3/Glacier/EBS/etc) whereas in Azure it's one storage platform that can be called by API for different purposes (like block store, drives, etc).

Right now Amazon's primary problem is the talent keeps leaving, and they don't pay enough, nor do they have the perks to keep people motivated to stay. They take existing open source projects and expand upon them and provide you a managed service primarily on the IaaS side, but nothing is really developed net new; it's all a rehash of things already made. And that's where MS is making strides... all of their software is a reimagining of what cloud should be, rather than taking their existing products and making them "cloud ready". In time I think MS will do very well, but the poor marketing as well as MS' poor image from years and years doesn't help them. That said... AWS can only stay in people's good graces so long until they just can't keep up the level of innovation that not only MS, but Google are doing in the Cloud space. And truth be told... if MS has to worry about anybody in Cloud, it's not Amazon -- it's Google.

AWS has a lot more features than Azure? No, not really... if you want to talk about IaaS components, Azure matches them almost 1 for 1. There are things that AWS has like Kinesis, or OpsWorks that don't exist in Azure so I'll agree to a point, but there are things in Azure like Service Bus, Service Fabric, WebApps, etc that don't exist in AWS. It's a different set of features and if you are IaaS heavy, the nod is definitely in AWS' favor.

However when you think about operational overhead, the amount of time you have to spend 'managing' the platform, the cost is far higher on AWS. It's primarily IaaS and therefore while you have elasticity and scaling, you still have to manage a hell of a lot of stuff. And Microsoft hasn't given you the flexibility or customization on this stuff, but as an organization you have to ask yourself -- do you need it? I'm sure on Slashdot everybody wants 100% control of everything and for that reason I can't tell you Azure is better than AWS, when AWS offers more finely grained control. However where Azure's benefit is, is in the PaaS offerings that allow you to focus on code development and product development, and less time on operational overhead. I have had conversations with some Netflix engineers and they LOVE AWS, but the amount of operational stuff they do to maintain the platform is well... huge. But it's part of their business now so it's not a big deal, and they've automated a lot of that though their own code. For organizations who have no foot in either cloud, the easier one to maintain is Azure over the long term.

And finding articles that say "AWS is better than Azure" are a dime a dozen, but it takes some real work to find out whether it's right for your organization. If you can tell me what you can do with AWS (outside of Kinesis and a few other services) that you can't do in Azure, then it will make sense for you to say that. But I can give you a LOT more I can do with Azure that I cannot possibly do in AWS. Not to mention, all of what I do in Azure can be taken back in house and run on my own servers; with AWS it's a one way street and if they screw you on the service or price... you are basically fucked. Microsoft plays that strength very well.

I am a happy AWS user as well, and have been for years. But they exist in the IaaS space (which is why you saying Softlayer is kind of apt), and their developer tooling pretty much sucks. I've found this out only over time, but they have a very narrow way of doing things for developers where MS seems to be giving a lot more options. For Lambda for example, you really only have a few ways of inputting a script, and it only goes against AWS resources. Azure WebJobs (much less sexy a name) is far more powerful with a lot more options for code than Lambda has.

AWS has a lot of options in terms of IaaS deployment, and you can customize a lot. But Azure meets most needs on the IaaS side (and even some moreso with things like resource manager which are far more elegant than CloudFormation) and blows them out of the water in the PaaS space. In terms of future development for my team, we are shifting from spending almost 100% AWS to migrating into Azure. I think operationally the costs will be lower and the barrier of entry is a lot faster and easier.

I was also at ReInvent but that's like saying that there's a lot of people excited about baking at this cupcake convention... it's not a surprise.

As for the account reps well... MS has always been great in the enterprise space in general, so not sure if it's a $ issue or your existing relationship with them, but for any EA holding customers, they are basically on call 24/7 for anything you need. Granted I don't work for them, so fuck if I know... but guess you know better than me :)

I have friends who work and who have worked there.... the job climate is basically ridiculous. Lots of hours, the pay not as great as their neighbor in Seattle (Microsoft), the advancement not very good either. Not to mention, Amazon is basically falling down in the enterprise space. They have made a lot of gains with CIOs/CTOs who are infrastructure focused and have a mission to "cut costs," so they have companies like GE and the like moving over to use their IaaS, but their platform services are a joke.

Everything at AWS is rehashed open source that is made to fit into a 'cloud' world... nothing wrong with this of course, but most of the basis of their products never really was meant for humungously distributed systems. Microsoft on the other hand (love them or hate them), made a totally new stack for cloud and the development community is embracing it on the enterprise side. This is Amazon's game to lose, but given the way their storage is segregated, their platform is one-way (come to us, no migration path anywhere else!), and their costs are nothing to write home about (because everybody price matches the IaaS pricing now).... I dunno, it's not going to be great for them going into the future. Of course I could be wrong, but right now I think given their human resource problems, their platform issues, and their inability to focus on developers (since they cater more to the ITPro crowd with IaaS solutions), it doesn't look good long term for them.