Eerik's organizer of notes and stuff

Small handy linux tools

eerik — 2012-04-17T18:16:41Z

I'll be keeping a list here of the small Linux tools I often use so I don't have to go searching around when I forget the name.

Graphics

Color picker
apt-get install gcolor2

Security

Extract server signatures from profile
ssh-keygen -l -f .ssh/known_hosts

System

Check processor speed and info
cat /proc/cpuinfo

GKrellM
apt-get install gkrellm

... Hmm... by definition this page is for tools I forget the name of, so I can't think of anymore at the moment. I'll add more as I go searching for them.

Two lessons in computer security

eerik — 2012-04-15T18:20:16Z

If you've landed on this page, it's probably because you've come to realize there is no click-through solution to information security and so have been searching like a maniac for some real solution.

Indeed, it's the “click to solve my current problem” mentality which is largely the reason the internet is so insecure to begin with.

Here is the first part of a concoction of lessons I'm putting together of what I've learned in computer security.

So security lesson #1 is: Don't click!

Unfortunately, there's only 2 alternatives to Don't click, which is A. pay someome to take care of security or B. Understand things and take care of security yourself.

Though in a huge organization option A may be necessary on some level, it can never be real solution, since who do you trust? Even security firms are being publicly hacked these days (which just begs the question of how many more are covert-hacked, an issue we'll deal with shortly).

But probably you can't afford a high top level security consultancy to come in and vett your system and train you and your people to implement best practices. And any partial or non-expert requires understanding computer security to evaluate how secure it is and whether it's being done right.

So this security guide starts with the basics so you can at least know what people are talking about and the major issues to deal with. Bringing us to:

Security lesson #2: Research before implementing something.

Research the issues as you come to understand them, on this site and other sites. Run searches like “Thing is Bad”, “Thing is good”, “Thing vs” etc to get a survey of the different opinions concerning the thing in question.

Now, getting back to who can you trust, when it comes to computers the answer is surprisingly few. Even people that are trying to do their best may make a simple mistake which creates a huge security flaw, and not just people you're working with, but people writing the software you're using.

Software is so complicated that it's impossible for any one expert to go through any average program and say it's totally secure, much less a huge complicated system of programs working together. Indeed, software is so complicated that any claim to security should be viewed as highly dubious whoever it is, like free energy devices. The only solution is the “open the hood” approach. If anyone can look under the hood and verify for themselves the claim, only then do we start to have faith some outrageous claim is true.

But not only do many software companies not make it easy for others to verify, they make it impossibly hard. Unlike a chair or a hammer that any organization can buy and test for themselves the makers claims, software can be encrypted and obfuscated in many ways to make it essentially impossible to “see how it works”.

Switching to KDE

eerik — 2012-04-15T12:36:03Z

Gnome 2 is and was the greatest. Highly efficient and everything just works. Gnome 2 atop of Debian, which also just works all the time, was and is fantastic.

But alas, though Gnome 2 is still in Debian Stable squeeze, it's no longer in development so will die eventually.

I've thus decided to be pro-active and switch to KDE. There's of course other interesting desktops, such as LXCF that I use on older computers, but I also maintain computers for family members (and introducing people to Linux when I can) so I don't want to have to be configuring everything all the time or fixing bugs. So Gnome 2 was great for this purpose, and I hope KDE can fill those shoes.

Though I'm not specifically against Gnome 3 or unity, it's probably a lot better for new linux users or people who mostly do single-tasking, so hopefully Gnome 3 and Unity will bring more people into the Linux fold, which can only be a good thing.

But for intense multi-tasking it seems like KDE is taking this road so I feel I might as well jump on board now, while I can still fallback on Gnome 2 in a pinch.

It's also great to have a fairly modern KDE 4 in the Debian stable repositories, since on my work computer I only use a newer version of everything to correct serious bugs.

Thoughts on KDE so far

I've always liked a lot of things in KDE every time I've checked it out now and again ... but I always hit some strange bug or missing feature which brought me back to Gnome 2, which has always been bugless for me. Since it's not like Gnome 2 did any particular, rather it's great since it does nothing and does it well, leaving you to work.

But now with KDE 4.x things seem to be consolidating, and I seem to be able to get my work done without random things disappearing (or the network not working etc.).

KDE positives

Kate

Kate is really awesome for programming. Intelligent color/bolding of things based on the language your programming, and the options to change essentially anything.

I used Kate in Gnome 2 as well, but it takes a long time to load up (as it has to pull all the K libraries), so for quick edits I generally used Gedit, which ok ... I guess. So, it's good to have Kate open instantly.

Widgets

I'm not going to lie, widgets are good.

The Desktop widget prevents crazy proliferation of files on the Desktop eventually overlapping each other and rendering the Desktop useless to keep track of small things you should take care of ... someday, and on that day it's because the file happens to be on the Desktop that you remember anything about the issue.

The unit conversion widget's fairly handy as well as the RSS feed. Monitoring system resources you may as well pop on there as well.

There a paste widget for pasting things, so far only using the default generating random passwords but I may find other uses.

You can also download plenty of widgets, but I haven't gone through them yet.

Differences to watch for

So far the biggest difference has been finding the keyboard and show-desk top icon.

I need to configure the keyboard to be able to switch to French layout, in Gnome 2 it's under "Keyboard", in KDE it's under "Input devices",.

The show-desktop icon you have to right click on the panel where you want it and add it as a widget. Though on the newer version of KDE it's there by default.

Things I don't like

Multi-Desktops tabs make no sense

Biggest problem is that when you switch desktops the tabs at the bottom don't switch, which defeats the purpose of multiple desktops.

I do a lot of crazy multi tasking, and like to organize each task on a different desktop so I can easily switch back and forth depending on what I feel like accomplishing at any given moment. Generally I don't close windows at all until I close the computer.

So having all the tabs shown on each Desktop is a fairly big drawback.

But this maybe configurable, or accomplished through this "activities" concept which at the moment I haven't checked out what it does or how it works yet.

Update: minimized windows resolved!

Took surprisingly long to find how to change this beharviour though it was right infront of me.

Problem is that there's "computer > system settings > desktop" which I assumed would have the option. It has the option of different activity on each desktop, which allows a different set of widgets on each desktop, which is interesting ... but it crashed continually ... and it still maintained the minimized windows on all desktops.

The key is to close all your windows, right click on the now empty bar and choose task-manager settings. Then under filters select "only show tasks from current desktop". You can also turn off grouping by program which just complicates things in my opinion.

File browser weird

By default the file browser shows temporary files, and doesn't show "last modified" column in details mode. I assume it can be configured somehow or is "fixed" in the next versions.

Remote-server bugs

Connecting to remote servers with fish seems a bit buggy, sometimes it says I can't save ... and other times it opens folders but doesn't show the content. Don't know if it's really a bug or whether I've configured something wrong, or whether it's fixed in newer version, will report when I understand more.

Only real complaint so, but not annoying enough to abandon the KDE attempt so far, works most of the time.

The Wants Theory Refuted

eerik — 2011-06-02T15:12:28Z

I knew a guy once that was of the disposition that one could not exist without wants and took as given that fulfilling these wants was happiness.

My counter argument was that if ever one got what one wanted, but was not made happy, then one must at all times doubt that what one wants will bring happiness, either by doubting that what one thinks one wants is actually what one wants, or by doubting the entire framework.

In the first case, the framework is seriously undermined as the repair presupposes discarding the premise that wants are empirical facts that one simply “knows”. For if one get's what one wants but is not made happy, the only way to salvage the framework is to conclude that one might not know what one “truly wants”. One is now on an endless search to uncovered what one truly wants, without ever knowing for sure, and thus the system breaks down. At some point in time one must simply decide what one wants, which seems equivalent to the alternative of the entire framework: that intentions are decisions not empirically verifiable facts.

In general, the premise that fulfilling “wants”, of any given time not the tautology of wanting to be happy, equals happiness can only be proved if one actually fulfills all of one's wants: one cannot base one's decisions on what one doesn't know. In practice then, the theory is simply unworkable and unverifiable until one has carried the system through, which cannot be a justification for adopting it now. However, there are even further problems with the theory as we see with friend number two.

Friend number two argued more tangibly by appealing to what he thought at the time was the irrefutable virtue and general superiority of his wants: “This is what I want: to develop and spread solar technology for the good of the world and to get fifty million dollars in the mean time.”

I defeated his reasoning by pointing out that these two goals were competing against each other. Every dollar that comes under his control he can either put into goal A or goal B. He of course needs to put enough money into himself to carry out goal A efficiently. But sustaining oneself and improving one's abilities to be more effective in developing solar technology is not to be confused with having fifty million dollars and doing solar unrelated things.

Previously he had simply imagined that these goals where mutually inclusive and reasoned things would just magically work out, but he quickly realized that this wasn't so.

How then could he decide to draw the line between goal A and B.

Furthermore, what if he accomplish these goals? Solar technology has grown beyond his ability to contribute further (something that cannot actually occur since solar technology can always be improved), and he has fifty million dollars, what does he do now? He'd do other things he wants, he answers. But what, he didn't really know. Would having fifty million dollars really solve all his problems: would a loving wife and exuberant children be the necessary corollary? And would this state of affairs simply continue indefinitely? He began to doubt himself, thus falling victim to criticism number one, regardless.

So, I reasoned that his actual goal was simply to have goals. His first principle was his last principle. And so, the harder the goal the longer his ambition to have goals is satisfied.

In general, if one's wants are finite, and the purpose of life is to fulfill wants, if one actually fulfills one's wants, one no longer has purpose, and so one would collapse into an motionless or apathetic state, which hardly seems happy.

Thus, the only way to avoid the question “what should one do?” (as in ethics), is to have not only wants that never prove dissatisfactory but an infinite number of them. One must be able to not only imagine the fulfillment of one's current wants and be totally convinced that such would be true happiness, but one must be able to know exactly what one's next want would be, and the next, and be able to repeat this process indefinitely. Otherwise, one cannot argue that wants are empirical in nature: a simple fact and the actual job of thinking is to fulfill these wants efficiently. The alternative is that one must decide what one's should do through some method of reasoning, a method of reasoning that would have to be consistent and complete (as in an ethical system).

Furthermore, since the only way to circumvent ethics, the abstract question of what one should do, devoid of any pretense, is to mystically know exactly what one wants all the time and never be failed by this “sense” (which is another avenue the wants theory falls into disarray, as empirical data is sensory, and so one must have a “wants sense” to establish these empirical facts), one would hardly waste one's time arguing about ethics with a bunch of strange creatures who seem to lack empirically verifiable wants through this sense.

Not only would such a person be too busy fulfilling wants, but the generalization of the theory is extraordinarily difficult. For, the only way for the theory to work is to assert that everyone not only has these wants but that these wants never contradict each other in any individual person; for if wants did contradict each other then that human is bound up in irrationality, by definition impossible to escape, and has scant use of a theory. Any philosophy that asserts: humans are irrationality, is pointless to accept. If one accepts it, one is irrational (whether it is true or false), if one rejects it one is still irrational if it is true.

Other obvious questions are, where do these wants come from? How exactly do we “observe them empirically”? How exactly is the activity of the mind separated from one's wants? If there is no distinction, one cannot argue that “one should do one's best to fulfill these wants” as, if one doesn't that is clearly what one wants to do (but how can one want to do what one doesn't want to do, which seems to be the state of anyone who rejects the theory?). How does the wants theory resolve the usual problem of self reference?

The general counter argument is, what alternative is there? The alternative is the total reconstruction of one's entire decision making framework. Starting from nothing, no pretenses or pre assumed “goodness” or “desirableness”, and contending what one should do. I call such a state the void, as one is devoid of all ambitions: one does not even possess the ambition to form ambitions, which gives this state of mind its meaning.

After rejecting my previous mode of behaviour which I concluded was meaningless and everything I had thought was important was clearly unimportant as I wouldn't even remember in a few years, and so coming to this void, I of course still moved and talked and slept and so on, but the difference is I could not answer why I was doing so. And so this is the beginning of actual philosophic contemplation.

Some might argue that removing all identity from oneself and starting from zero is some psychological process to some meta account of humans. I disagree, I view it as the intrinsic basis of ethics. In general, if one finds a flaw in one's decision making process one must remove everything that has been based on that flaw, as the first part of wisdom is the absence of foolishness, and so if the flaw is recognized as the basis of one's thinking, one must remove everything. And so, if this is reasonable, if one did find a flaw in one's fundamental reasoning, as in a contradiction attributed to one's most basic assumptions, clearly one should as quickly as possible remove the flaw from one's reasoning and any action that results from it. The most reasonable course of action is then to become devoid of any reasoning system, but without a reasoning system one does not have the intention to form a reasoning system. Thus I refer to any such first principle decision as a pure decision, not simply the product of more fundamental decisions (I decide to eat potato because I have decided to continue living, and eating this particular potato I understand to be an efficient fulfillment of my decision to live), or in renaissance terminology, an act of pure will.

Though this may sound mystical, I think it is born from a lack of words: our language is designed to explain what are decisions are, which we almost always do by appealing to some more general generally accepted decision, or at least accepted by the person we are talking to (or more precisely we don't talk to people who do not share some general decision). The idea of not having made any decision at all, and removing one's adherence to any previous decision, and thus being in a decisionless state does not seem to exist with ease in our language, or any language I gather. What I am trying to describe is a logical necessity, assuming one can be logical, which we must assume for the assumption of the opposite is by definition useless.

Why Spip?

eerik — 2011-06-02T14:38:10Z

Spip is designed to be an ultra light, ultra fast labour saving device for professional webmasters wanting to give real time, fast and autonomous content access to users with fully customizable permissions (no risk), to solve common web problems quickly rather than reinvent the wheel, but still be able to design completely from scratch with total freedom.

What is Spip?

Spip is a open source Content Management System (CMS) running in PHP/MSQL. Spip is free software licensed under the GNU/GPL. Spip was originally made for a magazine, where an important criteria is that authors can start publishing and editing content without any training nor risk to the system.

Spip user experience?

Spip allows authors to be created to connect to the site and manage their own content, add and edit pages and upload images, files and video. Each author can be given publishing rights (admin) to one or a selection of folders. For folders outside an authors admin space, they can submit articles but only an admin can publish it on the public side.

The “back-office” interface of Spip is designed to be as simple as possible and require the minimum of actions of the authors to publish their content. The steps are navigating to their section (organized same as folders on their computer), click “write an article”, clicking “save” and then “publish online”. When logged in, authors can also stay on the public site and simply double click on any of their existing content to change it on-location.

Under this basic interaction, articles appear in the navigation automatically where they are supposed to and images are rescaled to what they are supposed to automatically. To augment their text, the author can select something and click on bold/italic/heading/link etc. Lists can be made by simply starting each new line of the list with a dash. Advanced users can create HTML code offline and simply insert as text, for more sophisticated presentations.

Comparison

Though all open source CMS's are fundamentally the same in the sense that, as they are open source, their code can be changed to do anything possible (there is nothing one CMS can do that another CMS can't be rewritten to do), they are not the same in terms of the effort required to solve any given problem. Clearly, the closer a CMS already is to the final objective, the easier the task will be.

In Spip there is absolutely no pre-defined design framework. This is important, since professional and unique web sites are designed from scratch, for the same reasons unique books are written from scratch and by modifying existing ones.

Most other CMS's are created to allow amateurs to build websites without any coding knowledge and configure with simple multiple choice. Though advanced users can tweak the “cooky-cutter”, the farther ones departs from the standard template framework the harder it is to code and the more confusing it becomes for the users to interact with.

Design Spip does not provide a design framework, but is better thought as providing six major services.

1. Content management 2. Templates 3. Markup Language 4. Language Management 5. Caching and compression 6. Plugins

1. Content management

The first service is allowing users to interact with the database in a secure and logical way. Content is organized into “sections” and “articles” exactly similar to folders and files on their computer. They can upload images, files, documents etc. decide if they want comments to be allowed for the article, choose a logo, associate key words, etc. (or not if they don't have the permissions to do so).

2. Templates

The second service is template management. The three main templates of a Spip site are for the “home page”, the “default section page” and “default article pages”. These templates are not required to have anything in common at all and can be completely dissimilar if need be.

However, when a common object is needed in multiple templates a “sub-template” can be made and inserted where required in the main templates; thus, repetition is reduced to zero.

For further differentiation of site design, other templates can be written for specific sections or/or all subsections of a section, or individual articles. A vast complex hierarchy of templates can be created if need be.

For all sections of the site, the design can be adapted to the content and not the content to the design.

3. Markup Language

The third service is providing the Spip LOOP (Boucle) language tailored to the common needs of the web developer. This language simplifies interaction with the database to solve common problems.

A basic example of a Loop is returning all the the articles in a section, ordering them under a given criteria, then extracting the Title and Text, and then padding these objects with code written by the webmaster (HTML/CSS, Javascrip, Flash, PHP, etc.). This task is very easy to accomplish in Spip because each section has a number, so only this number is required for the loop to return all the articles, there's quick code for all the usual ordering criteria, and the padding code is simply as it would normally be.

The LOOP language also easily allows taking only a part of the list (say the first or the last five entries), alternating the code padding (so the first entry is blue, the second grey, the third maroon, then repeating these for the next entries), entries can be excluded on an individual basis or by some criteria, as well as applying various filters to the objects. For instance, only the first 200 characters may be desired to make a list of introductions.

Images filtering is also very powerful. Images inserted into the text can be scaled, cropped, rotated, colorized etc. all automatically.

Also, the syntax easily support “backup LOOPS” if no entries are returned by the primary loop. So it's easy to automate the “what if what we would normally expect isn't there” that often arise in web development. An example of this is that a primary loop can return all the articles in a section but if no such articles exist a secondary loop returns all the articles in the subsections of that section or something else entirely. All lists can be easily paginated as desired (if a list is too long for one page it is broken up into several pages).

For advanced uses, the LOOP language is designed to smoothly include standard PHP to solve uncommon problems not envisioned by the Spip developers. For instance, if a highly specific ordering or inclusion/exclusion criteria is needed that no one has ever needed before, PHP and or Regular expression can be coded directly into the LOOP. So there's never a need for completely independent PHP code to be superimposed on the system, which often creates “reinventing the wheel” type situations.

4. Language Management

Language management is one of the main reasons Spip became so popular in Europe, as multilingualism is the rule not the exception. Spip accomplishes this by associating as having a language to each page as defined by the site administrators. In the templates, the webmaster can insert things like Hello which will return “Hello” if the page using the template is English and “Bonjour” if the page needing the template is French and Hola for Spanish. The “multi” tag is the easy way but isn't a perfect solution for when a phrase needs to appear in many different locations.

To solve this latter problem phrase references can be inserted into the code appearing as <:home:> and language files created that will replace these references with the corresponding phrase for that reference. So, if a new language is added to the site, all that it is needed is to translate a centralized language reference page for all pages requested in that language to return the correct phrases. So, repetition can be reduced to zero and the webmaster doesn't even need to touch the templates for someone to translate the entire navigation.

Likewise, LOOPs entries can exclude/included by language, so a single loop can serve completely different content depending on what language requests use of the loop. So, it's easy to code multilingual templates.

5. Caching and compression

When a page is calculated via the content-template interaction, the result is a HTML/CSS, Javascript and/or Flach page served to the user. This page and all associated content (i.e. re-sized images, external RSS feeds) is then cached for a time decided by the webmaster or until it's recalculated manually when a user changes the main content associated with the page (a page with a long caching time can become outdated if content exterior to the page is changed, e.i. Blogrolls; for a super dynamic page the caching time can simply be set to zero, but this is generally not required).

When pages are recalculated, the load on the server is minimal since Spip is a very light engine providing only the bare essentials.

This caching system has been tested in Apache to not be measurably slower than serving individual files.

Spip can also be set to automatically compress HTML, CSS, Javascript pages to reduce transfer size (a service often paid for).

6. Plugins

There are less plugins for Spip than other CMS's, but this is because Spip is not intended to be built via a multiple choice of plugins. Rather plugins are designed to augment the engine to allow new classes of problems to be easily solved, for instance augment the Loop syntax for less common problems. This approach is meant to keep the core engine as simple as and as fast as possible, by allowing the webmaster to only include added functionality on a needs basis. The plugins are generally designed to be easily deactivated without causing any problems.

Havana to Paris

Eerik — 2011-06-02T14:28:46Z

I sat in the airport in Havana, I waited alone for the plane, and I was glad to be leaving after six months there. To where I was going, however, was an entirely different matter. continue on Natural Writers

Who am I?

Eerik — 2010-04-15T16:04:00Z

As I go about my day, I am mostly concerned with bridging the gap between ecological theory and implementation on the ground. Though I have a bit of training in mathematics from my days of mathematics, outdoor survival from my days as a conscript in the Finnish defense forces, and creativity from my days as a creature, I am mostly self taught in philosophy, music, sustainable technology, and programing.

I am involved in a few projects that actually exist, such as Solar Fire and Decentralized Democracy. ... as well as many that don't exist but I'm currently working on. Get back to me later about those.

I play chess as well as a form of scrabble where only invented words are allowed ... I regularly score 400 points or more. I enjoy the letter Q.

Contact

Eerik — 2010-04-15T12:10:00Z

Feel free to contact me at eerik(at)eerik.org .

- About

The Urgency of Local Solar Power

Eerik — 2008-07-13T12:06:25Z

http://www.solarfire.org/en/doc_en_...

Idea Map

Eerik — 2008-07-12T10:35:03Z

Here is a rough map of my basic ideas. Click here to open image outside the frame.