#Opinion: Fight your own roots

Thank you, Red Hat. If I needed one more proof for your state of mind, you have delivered it to me.

The Advantage of Enterprise Linux over Community Projects

I totally get, why Red Hat is sending out such an email: They want to convice their potential customers of buying their Enterprise Linux. I actually second that – they need to pay their bills.

But this tone?

First of all: Linux IS a community effort. Even Red Hat Enterprise Linux (RHEL) is in great parts developed by a community inside and outside Red Hat. Linux on its own is community-driven, has always been and will always be.

Second: They are completely right in regard to TCO. The TCO is not defined by the list price, but by reaction times, quality of reactions, quality of support, etc. Absolutely right. But: That does not imply the usage of an Enterprise Linux. What is this, anyway? A vendor-locked version of Linux?

Third: What do I actually need an Enterprise Linux for, anyway, if I consider a cloud stack? Back in the days, when we were running our workloads directly on Linux, one needed something special, I get that. A linux with the ability to run for years, with awesome patching mechanisms avoiding downtimes.

But those times are gone, specificially for Container Workloads. They are running on top – or, to be more precisly, operated by – Kubernetes. And Kubernetes is not tied to one only machine, it works on a pool of resources. One can easily remove a worker node (or even a master node) from a running cluster, without affecting the workloads too much. This basically eliminates the need for an Enterprise Linux with awesome uptimes, a Linux (without Enterprise) would be good enough. Because Enterprise simply doesn’t matter anymore – your resource pool is breathing anyhow, all the time. Any modern community Linux would work just fine here!

There is no need for an Enterprise Linux anymore!

So, there you have it. This email is an evidence of fear. But is not your fear, although they want to scare you. It is Red Hats fear, it is Microsoft’s fear, it is the fear of those, who understand their Linux or their OS as the basis of everything. Truth is: Linux and any other OS are just commodity. They don’t weight as much anymore as they did back then.

And, if you think further about it, the same message is true for your (Vendor Locked) Kubernetes-stack. And for your (Vendor Locked) Cloud Foundry-stack. And for your (Vendor Locked) OpenStack-stack. What do you need them for? They basically offer the same functionality as the community version (give or take some special features who appear to just a very small minority), but they only run on Red Hats Linux, with Red Hats Storage, etc. – which typically is called Vendor Lock!

Featurewise, in regards to functionality, with respect to patching and fixing, the community versions are on the same level as their Enterprise cousins. But they don’t lock you in, and with the right software stack (read: VanillaStack), they even come with the last things that set those Enterprise Stacks apart from basic community projects: Commercial Support (with up to 24/7 availability) and easy-to-use integration. Which directly impacts the TCO – towards being more reasonable and affordable than compared to those of Red Hat products, for example.

Vendor lock always makes you pay more.

Then again: What exactly do you need your Enterprise Linux for? What would you attend such a webinar for? What exactly is your benefit, besides the nice swag you get?

Think of that.

Open as in Vendor-Lock

If you’re in the market for cloud software, specifically something based on Linux and Kubernetes, you’ll be penetrated by companies who claim their products would be open-source, and therefore would not lock you in.

Well, that’s a lie…

Let me explain that.

If you look at companies who claim to be the “biggest open-source company in the world” or “the biggest independent open-source company in the world“, their products are open-source. This means, one can look into their sources.

For most people, this implies it would impose no vendor lock on them. Because open-source. At least that is, what they interpret, when reading of “true open-source” or “open open-source”. Truth is, that this is a lie.

Obviously, Kubernetes and related technologies are open and actually impose no vendor locking on you. But if you were forced to run a specific Linux (which costs a lot and provides challenging subscription terms), then one would call this: A vendor lock.

Purposefully implemented vendor locks

This kind of vendor locking is considered to be very negative, since it is willingly imposed on a platform (Kubernetes or OpenStack or Cloud Foundry, just to name a few), which is actually open and on its own undemanding in regard of its underlying Linux flavour.

The same is true for something some companies call “Enterprise storage”, which typically is Ceph. Which is only running on these companies Linux distributions, opposite to what Vanilla Ceph is able to do. Or for company-specific versions of OpenStack or Cloud Foundry – all which are limited to the specific platforms.

The point is: These limitations are artifical. There might be reasons for doing that, but for whatever reason it is done – it still remains a vendor lock.

Open-Source != Vendor Agnostic

So, next time, when you look into one of their offerings, you should read what they actually say: “We offer you our heavily customized versions of those awesome vanilla projects, so you have to stick within our ecosystem”.

And you should remember, that “open-source” does not mean “vendor neutral” or “vendor agnostic”. These terms have nothing to do with each other – proprietary solutions can be vendor-agnostic and open-source software can be vendor-locking.

And you should consider deciding for a true open-source and vendor-agnostic solution. Which would be VanillaStack.

Running Ubuntu

Written on Linux

As I pointed out several weeks back, Cloudical is transforming (and actually storming) to become a true open source company. And so am I (again)…

Over the past days, I (re-) installed Linux on several machines. It started as a VM on my iMac Pro, later I (again) transformed my Surface Book 2 into a Linux machine (only the camera is not working at the moment, but I couldn’t care less), my old 2009 iMac became the Linux treatment and today, I also deleted MacOS from a 2012 MacBook Pro.

Running Ubuntu
Running Ubuntu

Why?

Because I love the power I have over the system. I can adjust it to my needs, I can rest assured about privacy issues and concerns, and I have the freedom to choose my own desktop environment. My first experiences with Linux happened 20+ years ago, when I installed Mandrake and later run Ubuntu for several years. So, I know what I buying into, and I know why that feels right.

But, there are things that don’t feel right, though: My 2009 iMac has an old ATI Mobility Radeon HD 4850 graphics card, which is simply not supported anymore (at least not without workarounds, resulting into disabled HW-acceleration). Yes, my Surface Book can not utilize the camera. And Yes, for my 2012 MacBook Pro, I had to install WiFi-drivers by hand.

But the reason for all these things is not Linux, it is – and this is something I am working against for a long time – vendor lock and proprietary drivers. AMD, Apple, Microsoft, Broadcom are to blame. And that is why I won’t even try to install Linux directly on my iMac Pro – T2-chip, many Apple-specific glitches and traps. No, thank you.

In consequence, I am currently working around proprietary and vendor locked solutions. My next computer will be chosen with that in mind – as much as I love and admire Apple, as consequently I am willing to move away from it, because it is not open and therefore not sustainable and customizable. I get why the closed approach of Apple and Microsoft is a very reasonable one, but it is not my approach (anymore, again).

So, here I am: Back at Linux, happy with this, frustrated with the limitations imposed by proprietary approaches – and willing to live with it!

Written and published on Linux.

The a**h*liness of companies

When I look at some ongoing events in tech, I see companies having the same characters as in real-life. To me, this is very revealing – and you should understand this as well…

Disclaimer: I know a lot of people within at least one of the organizations I am referring to in this posting. I expressively don’t accuse them or say anything about their integrity, I totally accept their choice of employer, I actually respect the companies I am referring to – and I am not too sure, if I would behave in a different way when being in the same situation.

Example 1: Trump vs. TikTok

Trump hates TikTok. Trump wants TikTok to either disappear or become a controllable entity. So he issues an order and suggests its parent company ByteDance should sell it off. To Microsoft. And they happily see a chance there, they don’t think of any consequences or implications – they just see the chance to take over a successful platform for a bargain. The same is true for Twitter.

If it would be humans, I would know how to call them: Opportunistic. And if it would be humans, I would be ashamed to be associated with them.

Example 2: Epic vs. Apple and Google

Epic hates paying AppStore-fees to Apple and Google. So they circumvent them, get thrown out of the stores due to breaking of rules. Then they sue Apple and Google with pre-prepared paperwork and present themselves as the ones fighting for freedom and independence. Truth is: It’s only about earning more money, not contributing back and not following the rules.

During congressional hearings, other companies behave the same way (i.e. Microsoft…). They deliberately ignore the fact that Apple and Google created their ecosystems and own them. They perhaps even try to make us forget about their own failures.

If it would be humans, I would know how to call them: Egoistic a**h*les. And if it would be humans, I would be ashamed to be associated with them.

The question is: Are we doing any better?

Well, frankly: I doubt it. We’re talking about business and companies, not humans. If organiuations want to grow, they usually need to do that by fighting against other organization, by being opportunistic and by eliminating their competition.

The point I am trying to make is this: Companies tend to care only about themselves, regardless of which image they might give themselves. You don’t become huge without becoming an a**h*le. Period.

So keep that in mind when doing your next purchasing decision or when fanboying a company.

To them, the only thing that matters is business and profit, not being a remarkebly “nice” organization. Because “nice” turns into “niche” quite simply – and “niche” implies failure. At least to them.

Just wanted you to understand this.

PS: Thanks to Michael for pointing out that TikTok belongs to ByteDance and not Tencent. 🙂

#OpenSource: Trust vs. Usage

When I published my article about open-source and the way I see it as an ecosystem, I received some – very welcome and critical – feedback from a very valued Microsoft employee and architect, who rightfully pointed out, that Microsoft is among the biggest contributors to open-source frameworks and -foundations – and still I don’t trusted them.

He is absolutely right in his analysis: I don’t trust the company and most of its products.

Let me explain that.

First, I have some history with Microsoft. I have been Microsoft Most Valuable Professional for ASP and ASP.NET for several years in a row a decade ago. And I am very, very thankful for this – I got to know many very talented and experienced people, great products, loved the atmosphere and the cultural diversity at Microsoft. I still do, at least in regard to most of the aspects I just pointed out (subtract the products, which is what I am going to explain here). Later, I was working with a small group of awesome, smart and talented people in regard to Windows Phone and Windows Mobile – I was teaching on behalf of Microsoft, wrote several very successful applications for the platform and truly loved all of this. And I still do. I even considered several times joining them as Evangelist and Architect.

I lay that out to make clear that I am by no means a hater or a disappointed fan boy or something similar. I truly love many aspects of Microsoft, I have the fullest and most honest respect for many of its employees and from a technical perspective I can understand a lot of excellence within their products (and some of the excellence is way beyond my level, to be honest).

But: I don’t trust Microsoft.

And the reason for this is: They are not an open-source company. Their products are mostly closed source. Their platforms are closed source. Azure is closed source. And their business practices are embracing closed- and proprietary source ecosystems over open-source ecosystems.

It is absolutely true and needs to be acknowledged that Microsoft is amongst the biggest contributors to open-source ecosystems. And they even open-sourced many of their own products. Thank your for this, Microsoft!

But: I don’t trust Microsoft.

My point is this: Regardless of how much valuable and excellent work you put into open-source technologies – it does not increase the trustworthiness of a proprietary environment or a proprietary ecosystem. Because it is not completely transparent. It is not completely open. There are substantial blind spots in the software and the ecosystem.

Take a picture: If you add layers of glass (open-source glass for the matter :-)) to each other, it stays transparent. But just one layer of – say – concrete or wood in between makes the whole transparent stack in-transparent. You won’t be able to see through all the layers. And that is enough, because you don’t see what is going on with that stack of glass.

Transfer this picture onto a cloud-environment, and it immediately becomes apparent: How are you supposed to trust this environment, if it is not transparently laid out? What do you trust more: The words of the people who built it, revisors and auditors? Or a community of experts, who constantly review and audit?

To me, the answer would be obvious: I would trust the community of experts (plus additional auditors and reviewers). And that is regardless of the awesome quality, the overwhelming quantity and the sum of funding that is put into open-source projects, because – again – it is a matter of trust.

What does that imply for Cloudical’s offerings?

Good question! Thanks for asking!

The same measurements need to be applied here! If we ever create a software stack (and you could bet it would be a Vanilla software stack), we would need to open-source it. Without discussion and dispute. It would be a matter of trust!

Actually, the foundation behind our Managed PaaS– and Managed SaaS-service-offerings is an open-source software-stack. It consists of SUSE-products, such as CaaSP (Kubernetes) and CAP (Cloud Foundry), plus additional scripts and tools required to roll it out and to operate it.

Although we are speaking of a service on a platform stack everyone can easily roll out on its own (well, actually, no, there is a LOT of effort and a TON of knowledge in it), we will open-source it, within the next two to three months. Because we are committed to this kind of thinking, to transparency, trustworthiness and open-source.

Thanks again for asking. And thanks for commenting – and being critical on this and with me.

And yes, I still love Microsoft. 😉

#OpenSource is the way to go!

There are several reasons for this, but seeing my LinkedIn-timeline flowing over with proprietary software- and infrastructure-stack-related news, I feel I should point out some crucial points for open-source:

Open. Source.

The source is open – you can actually see, what is happening! This gives you something which you won’t get with proprietary software and ecosystems: Trust!

Trust

Trust is not only what you need in times where more and more workloads are sourced out into public cloud- and hyperscaler-environments, it is a strategic asset! Trust needs to be at the center of your actions, but there is another aspect being strongly related to that: Security!

Security

With open-source software, literally everyone can check the source code and find security issues. Yes, it might be painful and it feels better with proprietary software, since their security issues are handled way more silent – but this is security-by-obscurity! Just because security issues are not that well-known, it does not mean they don’t exist. Instead, there is a higher chance for them to be exploited, since there is no sense of danger. Open-source-projects fix their security issues regularly and often – proprietary software vendors might not do this. This leads to another important aspect of open-source: Support.

Support

With open-source projects and solutions, you get an awesome level of support if you dare to ask. Yes, the tone might be sometimes a bit … nerdy, and yes, there are no SLAs, but this then again is where you can rely on professional vendors such as Cloudical, providing you with SLAs, professional support and managed services. The advantage of those vendors give you over proprietary software vendors is quite important: They are independent of a specific (proprietary) software stack! So they give you support and consultancy, but in an unbiased and open way.

If you compare this to vendors of proprietary software, they usually want to sell you their products – which might not be what you want. The open-source ecosystem provides you with many vendor-neutral frameworks and solutions – and you find a lot of skilled experts for your operational needs. And even if not (or if it would not be enough to hire those experts) – there are several very affordable and well-executed managed service offerings in place!

Managed Services

Since experts, specifically cloud infrastructure- and software-experts, are a rare things nowadays, there is a huge struggle for finding and hiring the brightest minds. Nonetheless, you need the know-how and knowledge to run your infrastructures and workloads – regardless of the cloud environment and regardless of the workloads you are throwing at them. Managed Services for open-source ecosystems are platform- and vendor-agnostic, so they can work on every cloud and on every platform.

Since the required knowledge is reusable and (mostly) independent of the platform, it will give you peace of mind and sustainability in regard to operations. This would be more complicated with vendor-specific software and infrastructure-stacks, since they are way more limited in regard to their usage and usability scenarios. Another aspect of open-source projects is that they are usually widely adopted and utilized – in their original, vanilla form.

Vanilla

The foundation of products such as Red Hat OpenShift, SUSE CaaSP or Rancher K3S is Vanilla Kubernetes. The foundation of Red Hat Open Stack or Mirantis Open Stack is Vanilla Open Stack. The foundation of Pivotal Cloud Foundry or SUSE Cloud Application Platform is Vanilla Cloud Foundry.

And the good thing is: All the things you or your partners know from vanilla projects apply to the commercial distributions. With open-source projects, knowledge is transferrable between distributions, knowledge is shareable – and knowledge can be achieved without expensive certification trails (although some very useful certificate trails exist). And: Knowledge is shared within a community. It is understood as an asset, not something someone owns (because it was so expensive to gain). Community stands at the center of open-source.

Community

With community, open-source starts and with it, the circle closes: Without a community of enthusiasts, interested industry partners and supporting foundations, open-source would be not even half as appealing, as it is. Open-source is for a very long time matured, processes and governance exist for dozens of years, and even companies such as Microsoft, Oracle, SAP or Amazon contribute to open-source. They form a community, big foundations support and steer the process and create the sustainability in the projects they chose to sponsor.

Open-source is all about openness, trust, security, support, managed services, vanilla and community. And this combination of factors make open-source a strategic aspect and a strategic factor in organization’s sustainability. With open-source, you decide against vendor-locks and golden cages, surpressed issues and horrendous prices. Open-source is trustworthy, cost-effective and driven by intelligence, not by greed.

What does this imply for Cloudical?

Well, with Cloudical we already decided to set more on open-source toolstacks. We decided to move away from Microsoft 365 and Google-GSuite-based approaches. We decided for an open-source CRM-system. We integrate them, we operate them, we trust them.

We are currently setting up, integrating, automating and starting to operate our own open-source ecosystem – and we will make this stack (i.e. Rocket Chat, Keycloak, Harbour, etc.) available to our customers as managed solution and service.

We already offer awesome consultancy and managed services, and we will step up the game later this year. There will be way more offerings from Cloudical, and we will broaden our support for the open-source ecosystem – so stay tuned on that!

What does this imply for me?

For me, personally, this implies way more usage of these technologies. I need to learn and adjust, even if it implies leaving well-known and comfortable ecosystems. But I feel it is worth it, because to me, trustworthiness, privacy and security are more worth than “just” comfort.

To me, it is worth every effort. I personally don’t trust Microsoft, Google or Amazon. I don’t want to be spied upon, don’t want to be bombarded with advertising and don’t want my profile to be sold to some random company. I don’t want my data and my workloads to run within environments we don’t trust (and we can not trust them, because it is their proprietary ecosystem!). I want to support alternatives, I don’t want this world to end with some dominating companies – I want freedom of choice, trust and security.

So to me personally, there is no alternative to this positioning and to this approach.

My new (old) travel policy

In the past years, I adjusted my personal travel habits to be more ecological and sustainable. Now I decided to step up the game.

Abandoned Airplane (Pexels)

I already decided for some fundamental rules to my personal, 90,000+km / year, travel- and commuting habits:

  • I don’t use any plane for distances short of 1,500km
  • I travel mainly by car and by train
  • When travelling by train, I use the 1st class and keep distance by booking single-seated options

My car has – for the past 12 years – always been a Diesel, with all modern cleaning technologies. I have used a BahnCard 50, which grants me 50% rebate of all inner-German travels via train, at least for what Deutsche Bahn covers. Since there is no such concept as peak- or offpeak-fares, this was very viable and it worked out once you had reached a spending of 500,– EUR per year.

After doing some research into feasibility and after wanting to improve upon my ecological footprint, I decided this as my plans for the next two years to be incorporated:

  • I don’t fly within Europe at all
  • My next car will be 100% electric
  • Commuting to the office will be done with zero carbon local footprint only
  • I will upgrade to a BahnCard 100 1st class
  • I will use my bike for short distance commuting (up to ~25km each direction)
  • I will just use a none-electrical car if it can not be circumvented
  • I will add solar panels to my roof to produce energy locally

The goal is to reduce my car-bound travels from roughly 70,000km in 2019 to appr. 35,000km in 2021. I shall use a car only for travelling, when there is no appropriate train connection available, or when I have to carry heavy or bulky luggage. All other trips can be done by train, given the health situation allows for it.

The most challenging aspect will be the switch to a fully electrical car, since I expect a minimum of 500km of range – not on paper, but in real-life situations, such as when driving on an Autobahn with 130km/h.

Tesla Model S (Source)

I understand currently only some Tesla Models matching these requirements – all of the German engineered cars may have such a range on paper, but not in real life. On the other hand, it will be very rewarding – I anyhow used 100% renewable energy for the past 10 years, I am planning to add some solar panels to my house’s roof, ideally resulting in being able to sustain my local travels from my own energy sources and circumventing charging points for most of the time.

So, my new travel habits will be more:

  • More sustainable
  • More ecological
  • More healthy

To me, this is the right way to go, even if it is more expensive then continuing with the traditional approaches.

What do you think?

They track me!

Following an internal discussion last week, I decided to do something I always wanted to do: I created a video. The topic is serious, although the preview picture is not: It is about why you are tracked by Google and Facebook (and many other companies) and what you can do about it.

I have to admit: It was a lot of fun making the video, and I see plenty of room for improvement. But it was a good start and I am planning to do more videos. So feel free to subscribe to my channel and leave a comment. 🙂