Out of the 4 regular readers of this blog, this post may connect with two of them (you know who you are). I was going through Don’t start from scratch, which details the story of a guy who wanted to build his own toaster, literally from scratch. Starting from the iron ore up to the finished product:

“I realized that if you started absolutely from scratch you could easily spend your life making a toaster.”

Upon reading this, I was reminded of another post I had read back when Google Reader was still a thing and I was still reading stuff from academics (the pain of my PhD failure is still there, but my job is more demanding and I do not read theory any more):

Certainly you should never read anything from the 1960’s or 70’s or you will realize that it all has been done before — by Real Programmers who had to code in assembly on a trinary architecture with sixteen levels of virtual address space segmentation and only two registers

I do not know why, but these two connected themselves; memory is a funny thing.


xkcd 378

I suppose that I am one of a few people who had an interest in Loudcloud even though they had no vested interest in it. Maybe because it was Andreessen’s second company and I thought it would it would go south, I am not sure though.

However, because of its CEO, who also has three children and one of them autistic, the book was even more interesting to me. Horowitz is only a few years older than me.  The first part of the book describes the origins of Loudcloud, the efforts to keep it afloat after its clients went under, its transformation to Opsware and finally the sale. You can read the first part really fast; in fact you can feel the adrenaline.

Then comes the second part, equally interesting but a slightly slower read, where Horowitz explains his strategy and tactics and how he came to learn the job on the job, how mentoring helped him and how he built and transformed his management team. Without doubt the hardest things discussed are how to «properly» fire people and how to develop a HR process that can handle growth and personnel management.

However, due to two incidents briefly discussed in the book, I strongly believe that the author does not even come close to the hardest decisions that were made during this 8 year journey:

My father turned to me and said, «Son, do you know what’s cheap?»

Since I had absolutely no idea what he was talking about, I replied, «No, what?»

«Flowers. Flowers are really cheap. But do you know what’s expensive?» he asked.

Again, I replied, «No, what?»

He said, «Divorce».


Because you see, the hardest choices were made by his wife and she has not written a single page.

There is also another thing I take issue with the book. Horowitz describes himself as a wartime CEO and he complains that there is no wartime management literature and that most likely there cannot be a CEO who can run things both at peace and during war. First, this is not a war. Let’s get rid of the analogy, no matter how much Sun Tzu has helped you manage things. You did not kill any people. And second of all, if you ever need advice on how to run an organisation at peace and during turbulent times just ask a ship’s captain or figure out their curriculum if you’re so much in need of a citation.

Happy SysAdminDay


Since monitoring is at the heart of managing infrastructures, time-series and measurements also, I believe it is time to remember what the first derivative is. Because all those graphs we make with Graphite, Graphana and friends are not only eye-candy. They have some meaning too. «The Art of Monitoring» speaks of the rate of change in its first chapter, but this has a formal name.

Now go hug your SysAdmin.

dotplotting Melania


Terry Tao spoke using Shannon. Bob Rutledge figured out a probability of 1 in 87 billion. So how does this look visually?


x axis – Melania | y axis – Michelle


Melania’s 2016 speech is 1377 words. Michelle’s 2008 speech is 2223. In the diagram you see three distinct diagonals which are the verbatim phrases copied. I am sure that someone with more interest in these things can apply some NLP knowledge and graph something better.

PS: I just noticed that both speeches close with the same phrase: «Thank you, God bless you, and God bless America» (the phrase is seen at the top right corner of the plot).

While several German religious leaders, writers and artists, industrialists and politicians mounted strong opposition to Nazi rule at great personal cost, sometimes of their lives, there was nothing comparable to be found in German science.

I stumbled upon the book thanks to this article that mostly dealt with the effort of constructing an Aryan version of Physics. What an interesting book. By using the stories of giants like Max Planck, Peter Debye and Werner Heisenberg during the Nazi rule the author explores the apolitical stance many scientists used to have (and depending the field continue to adopt) on the results of their work and the repercussions of them being used. Planck remained glued to the Fatherland regardless of who was ruling it (and whether being true to the Fatherland), Debye tried to walk though the rain without getting wet and Heisenberg with the arrogance that came with his ability tried to rebuild German Physics by assuming the role of its curator.

Together with these we learn of the noble stance by Max von Laue and Fritz Strassman  («I value my personal freedom so highly, that to preserve it I would brake stones») and the grey stance of the rest who stayed and did science within the regime. Even Lise Meitner and Debye left at the very last moment they could. Correspondence between many prominent scientists from the beginning of the Nazi rule till after WWII is also included. Very interesting points of view and commentary of the situation these people were in.

While you cannot completely enjoy the book without remembering your Physics, you will highly enjoy the questions of ethics and moral compass that it poses. I know it troubled me and it makes me answer questions about the work I am doing and the sector it applies too daily. And DevOps is not exactly science yet.

While the stories we read about for 11 or so chapters serve largely as a preamble to the fantastic epilogue on Science and Ethics, they do describe the situation we are in today and questions we need to answer. Prior to reading the book I used the get a bit sad when a scientist or an engineer I admired failed me on other standards that I supposed they were concurrently holding. I do not get sad any more. Within the realm of politics they are equally fallible like the rest of us.

I make it no secret that one of the most impactful essays on me the last 20 years has been Bob Lucky‘s essay about the last electrical engineer. And I have seen other computer scientists taking it into account too.

Ever since cloud computing caught on, it comes more and more frequently to mind, because I see that the abundance of computing power makes people think that they can get away without really thinking about the problem. After all, computing has become cheap, right? The prototype you’ve just rigged up costs $10 / month to host. Well scale on the other hand does not come cheap and it will bite you.

Now that «cloud» has become a commodity word, «serverless» is coming to dominate. Because you really do not need to buy your infrastructure since, you know, you can have a data center as a tab in your browser for half the cost. And you really now do not need old-school sysadmins, since you make people all work together and deploy stuff n-times a day (DevOps).  But hey, even that is complex, because you just want to write the damn stuff and not think on where it runs, or how the hell it is supposed to run on scale. You only want profit on scale.

So let’s go serverless. And in the process, let’s make the prophecy about the world needing 5 computers true. Only they are cloud / serverless providers (just count the big players) where for some engineers, paraphrasing a bit Bob Lucky, life will be:

Projecting the current trends, future applications will be serverless.  No one will have the foggiest idea what is on that cloud.  Somewhere in the basement of the serverless provider or its successor will be a huge computer file with the listing of that infrastructure.  The last engineer will sit beside the file, handcuffed to the disk drive like a scene out of «Ben Hur.»  That engineer will be extremely well paid, and his or her every demand will be immediately satisfied.  That engineer will be the last keeper of the secret of the universe: E = IR.

Bootcamps and other vocational training institutions will turn out lots of serverless programmers, so prepare at least to be able to tick the Steinmetz chalk mark.

How to hire


Well this is the summary of my notes after listening to this podcast episode, about a subject I know nothing about. Five rules on hiring that can make you or brake you. They seem pretty normal to me, I guess I will try to make some use of them in the future:

  • Have a process and make it collaborative. It does not really matter what kind of process you have in place. It matters that you have one. Because once you have it and you have to shift through a lot of applications, patterns of interest to you will emerge. Make it a collaborative process in order to gain more perspectives and do not rush it, for you will end up hiring the first one that comes through the door without thinking it through.
  • Hire for attitude to work and accountability. When you’re hiring, you should try to figure out whether the person applying is taking ownership of their failures or missed goals, or whether it was always everybody else’s fault.
  • Be aware of your natural bias for sameness. Sameness can bring you closer to groupthink. You do not want that. Even Drucker advocated for dissent. To achieve it you need to be aware of your natural bias.
  • Know what is culture in your company and what is not. This is a pain point because when hired you can end up in the wrong environment. Not a happy arrangement for both parties. Just imagine being a 9to5 person at an all out, «sleep under the desk» company.
  • Hire for potential. This one I use to hear and real all over, including this podcast. Maybe this happens because everyone is looking for a specific type of guru? I do not know. But sure, if you cannot find (or pay) the specific worker that you’re looking for, why not build them? What if you build them and they leave, asks the old joke. What if you do not build them and you have none then?

Funny thing: This is the first podcast episode I ever listened to and kept notes (the bold lines, plus some keywords). And it is all transcribed.

PS: I believe the twins study mentioned in the podcast is this one.


Ενημερωθείτε για κάθε νέα δημοσίευση στο email σας.

Μαζί με 2.153 ακόμα followers