# Blog

2019-02-25

The goal of this blog post section of my web page is to combine my perception, ideas and solutions into wordy experiences. The experiences I can reflect on, the solutions which are maintainable and does make an improvement. Ideas and thoughts which exposes dogmatic experts. And some experiences of my own.

Since I have work to do, a book to read and a stride to swim I do not have time to perfecting my blog posts. Although I am learning to be better at it ;) Also comments are not supported because of maintainace costs. But if you have feedback and/or questions you can reach me over email akels14@gmail.com twitter account @graphitewriter or meet me in person.

## Making presentations with inkscape

About half a year ago I was in trouble for making presentation in a seminar talk. I had experienced strange bugs with libreoffice which I was trying to get rid of. I had also tried to use latex beamer templates, but I could not stand the empty spaces and misaligned pictures. For a time I thought I had solution with ipe which was designed for scientific drawing and could edit multipage pdf documents (created by ipe only) and is deeply integrated with latex. Unfortunatelly I did experience some click object bugs which made ussage of it irritating. Eventually I gave inkscape a second look.

## Using Digisparck for dispossable Android projects

I bought my Arduino about 6 years ago by being impressed how easally it is to make blinking lights and read value from potentiometer. But that was all for very long time. I lost my curiosity of making anythinkg more complex due to the ugly arduino ide, frustrations of dubugging and lack of easy options for deploying what I make. The Disgisparck is a new game. It only contains Ariny85 on it and still can be programmed as Arduino. Because of such minimalism, size and price I was quite interested what limits of applicaction one can reach and so I set up a task of automating a pump for my central heating.

## Email with emacs

Gmail web aplicaction is generally very good with many features. However I felt I was strugling with its usability - need to interact with a mouse in many places, browser spell checker, fast notificaction of a new email, multiple account managment and lack of emacs way of writing the text. In the process I tried Geary, which was good but was sometimes crashing, Nylas1, which had instant email notifications but became buggy and now is not activelly developed. I felt that my emails are still hard to manage and write. That was when I started to consider setting email with emacs.

## Moving from cloud to crowd

Data sharing between multiple computers is as old as desktop computers themselves, for example look at first computers Steeve Jobs presents 1. That was really outstanding achiewment that you can collaborate and share changes, but there is a catch - you need to maintain your own personal server. Although that is easy and still effiecient for memebers of the same companny that is hard for your own use to collaborate with others. That is where Dropbox had made a great benefit to our society with their cloud product. However I quickly reached the free membership limits of it and I was not ready to subscribe to their paid membership. A hope came to me with Resilio Sync allowing to host my own data infrastructure on raspberry pi.

## My thouhgts on electronic voting

Sometimes when I hear about new upcomming elections I dream that in future I could vote from my home or abroad for officials who would represent best the interests of our state. But this dream quickly becomes spoiled into nightmare by issues of trust such system could bring in if implemented into straighforward way with a central server or as a blockchain network (as it could suffer 51% attack which now becomes regularity). However in spite of this setting I believe that the issues of trust can be designed away and so I started this intelectual challenge as diletant assuming that all new cryptographic algorithms available are safe for production use.

## Electronic voting system based on an anonymous public key ledger

I am a theoretical physicist (a PhD candidate) and try to learn about cryptography as a hobby. One of the interesting aspects I had found is that many respectable scientists say that electronic voting is not safe, and so I started to follow the arguments which they use. For example, submitting the trust to authority to do their job honestly or sacrificing anonymity. Or impossibility to deal with anonymous bribers on the internet. However, a formal proof for that did not exist, so was interested if I could design a better e-voting system which would be anonymous, transparent and had a mechanism to fight bribery and coercion.

## A technical proposal on using Github to establish a democracy

As we know activism which is a necessary element of a working democracy is a subject to a free-rider problem and that diminishes its power. In the environment we live in, we have a limited capability to finance public goods such as media, open-source projects and political parties. Subscription model just does not cut it. Thus we rely on the state or from the advertisement model. Here I propose an alternative fundraising strategy based on transparent, anonymous, user verifiable electronic voting scheme.

## Could we save the world by democratizing democracy?

Currently, it is commonly accepted that one can not design an electronic voting system which would be transparent, user verifiable yet still anonymous and legit. And additionally which would be resistant against initiatives of bribery and coercion. During a discussion on twitter last week to find a real argument what makes electronic voting insecure, I started to sketch a better scheme. Now, I think I have nailed the design which solves all those problems and only uses asymmetric cryptography and hashing. Thus it is also cryptographically secure. (low rating I suspect is because of the narrow-mindedness of academia nowadays. The moderator did not find more significant flaw than family abuse.) The system seems to be safer than a paper bailout as one can check the set of citizens participating in the elections (so no fake identities) and can count the votes himself.

## Julia experience - multiple dispatch

Every useful computer program applies some operations to input and generates output which can be represented in a list of machine instructions. Fortunately we can make these programs with abstracting data structures in types and operations which acts on them. Depending on way we organize types and operations we can write code in object oriented paradigm (OOP) or functional/imperative paradigm (FP) as myfoo.hello() and hello(myfoo) is the same thing. But nevertheless of paradigm chosen we turn out to spend more time on organizing rather than in benefits of creative process.

This web page is hosted from my github repository with Github Pages automatically for free with no advertesments. The tradeoff is that you can serve only static pages and content. This is where I and many others come up with creative solutions since noone likes to write a plain html. Since I use .org files for my journaling and other note taking I chose to use them as basis for the source of my web page which I then compile to .html with a pretty elisp code. The results is quite marvelous with low maintainability (except for the content ;).

## Mans CERN piedzīvojums

Divus mēnešus šajā vasarā es biju CERN vasaras studentu programmas ietvaros (CERN summer students programme). Mani tur sagaidīja darba vadītājs ar uzdevumu izpētīt iespējas noķert vai pavirzīt zināmās robežas vienai no supersimetrijas daļiņām ATLAS eksperimentā pēc lielajiem CERN jauninājumiem. Arī šogad ir iespējas pieteikties šai apmaiņas programmai, tādēļ nolēmu šo iespēju pareklamēt, izstāstot savu stāstu. Tātad, kas mani aizveda uz CERN…

## Mans pirmais mēnesis Delftā

Ceļš uz Delftas tehnoloģisko universitāti bija pārdzīvojumu (un intrigu) pilns un vēl joprojām nespēju aptvert tās veiksmes, kas mani šeit noveda. Delfta visnotaļ ir forša vieta - salīdzinošs klusums, droša un patīkama pārvietošanās ar kājām un velo un tā patīkamā mazmājiņu arhitektūra. Ne gluži Valmiera vai Mazsalacas lauki, kur var izskriet pa meža takām vai ar velo braukt uz Cēsim peldēties, tomēr patīkami. Iespēju ir daudz.

2017 bija nervozs, bet ar panākumiem bagāts gads. Publicēju savu pirmo rakstu ar ko noslēdzu savu zinātnisko darbību Latvijā un ar mērķtiecību un līdzcilvēku atbalstu nonācu doktorantūrā Delftā. Pilnīgi svešā vidē, kurā esmu spiests sevi pilnveidot, kas sakrīt ar manu apņēmšanos arī nākamajam gadam. Krietni piezemētāk par to ko esmu paveicis pagājušajā gadā :)

## My first year in TU Delft

I remember that after the first month I promissed to make a blog post every month on my studies here. I admit I had different expectations and so being under the constant change makes hard to see the reflection of how I feel, perform and what future I see. But now when my first year had passed in the TU Delft and as I am required to submit my self reflection for the Go/NoGo meeting, I made concious effort on where I actually stand. Over this time I have had some unusual thoughts which might form an opinion and recepies posted in my webpage. But for now about my part of TU delft.

## Julia for Cluster

A time once comes that computation is too expensive to run. If you are lucky, you can adjust the parameters of the numerical method or find a better one. If that is not the case, you are now looking at what the phrase parallel computing promises. Often the numerical tasks can be reduced to actor-model meaning that there is a single master giving orders to the computing slaves. Most programming languages give many tools for these capabilities however the way and complexity can make us alienated to use them. I believe this is another strength of Julia which shows us a better and more elegant way and about that is this blog post.

## A ReMarkable tool on my desk

I used to spend countless hours to make reading and note-taking experience better with the tools I had a laptop and kindle eink reader, but nevertheless the efforts I was investing made me unsatisfied with the result. Perhaps I am easily disturbed by nuisances - eye irritation, unreadability, synchronization, and convenience - which stood in front of my goals and started to form a goal itself. If only there would have been A4 sized e-reader on which one could write at that time. Meet ReMarkable!

## A fairy tale: supercondoctors, Andrew bound states and quantized conductivity

Superconductors. Where are they actually used real life? Examples are actually pretty cool. A floating wheelless trian, a means to steer speedy protons in large hardron colider and if you have been lucky one you might have seen a levitating superconductor on top of a magnet.