About this blog…

I am employed by Netnod as head of research and development and am among other things chair of the Security and Stability Advisory Committee at ICANN and very active in the UN multistakeholder process IGF (Internet Governance Forum). You can find CV and photos of me at this page.

As I wear so many hats, I find it being necessary to somewhere express my personal view on things. This is the location where that happens. Postings on this blog, or at Facebook, Twitter etc, falls under this policy.

The views expressed on this post are mine and do not necessarily reflect the views of Netnod or any other of the organisations I have connections to.

patrikhson on Twitter...

Powered by Twitter Tools

Roasting coffee with i-Roast 2

Nicaragua Pacamara RFA

Nicaragua Pacamara RFA

So the other week on recommendation I bought an i-Roast 2. I bought it in Sweden, the 240V model, from Alnö Kaffehus. They have a package with roaster and some coffee. After some experiments, I found the following actually work pretty well. All together, smell of roasted coffee in the office(!), and the taste of coffee made out of freshly roasted beans… The improvements between each roasting is so large, and quality of the result so high, that I really look forward to where this road leads.

I first tried with one of the programs in the i-Roast, but that did not lead to the 2nd crack. Some thinking and reading on roasting of coffee have lead me to the following program:

  1. 5 minutes on 188°C
  2. 4 minutes on 198°C
  3. 3 minutes on 208°C
  4. 2 minutes on 216°C
  5. 1 minute on 228°C
  6. 4 minutes cooling

Some programs on roasting I have found (including the one I first tested out of the box) started with a short burst on high temperature. Maybe to get the chaff off the beans? The more chaff on the beans, the higher the first kick? Maybe I should try that next time?

Anyway, the beans I use now are Nicarague Pacamara RFA, arabica, which works quite well also for the hard roast I am after. I have used the roasted beans now for espresso and grounded it on 6 on my grinder that have a scale 0-12 (not 11!) and tried a press. The press had tendencies of not really be as complex taste I wanted. It was more pointy in many different directions. Maybe a finer grind will make the coffee marry better with the extremely hard water I have.

I will let the beans now sit for a day or two before I try to grind some more and see what it is like.

Portuguese wine – Douro

Friday evening I was lucky enough to participate at a tasting of four red wines from the Douro region in Portugal. The tasting was run by Fredrik Åkerman that has his web site http://www.portugisiskaviner.se/. The wines where the following.

  • Mogadio da Calçada Tinto 2008
    Producer: Niepoort
    Grapes: Aragonez, Touriga Franca, Touriga Nacional
    Price in Sweden: 159SEK
    Number on Systembolaget: 93998
    I found this be very strict wine, closed, with very clear tones of the stems of cherries. Acids. Nose give clear indication of oak and smoke from the barrel. The grapes have been 100% pressed manually by feet, stored on steel containers and then 16 months on small barrique. 100% of them new french barrels.
  • Quinta do Vallado Tinto Reserva 2007
    Producer:
    Quinta do Vallado
    Grapes:
    Aragonez, Trincadeira, Touriga Franca, Touriga Nacional, Sousão + others
    Price in Sweden:  388SEK
    Number on Systembolaget: 92142
    Full bodies, plums, much more tannins than the first one. Grown on a southern side of a hill, with wines more than 70 years old. 50% of the grapes pressed by feet, 50% with press. 17 months on new barrels.
  • Quinta do Crasto Reserva Vinhas Velhas 2009
    Producer:
    Quinta do Crasto
    Grapes: More than 30 varieties
    Price in Sweden: 225SEK
    Number on Systembolaget: 71734
    Very old vineyard and because of that unclear what grapes are used. The grapes are mixed already in the vineyard. All grapes picked at the same time regardless of variety. A very well balanced wine. Smell of the leaves of roses. Some stem in the taste. Very nice on the nose. Also clear is the oak. 15% of the barrels used is new american oak.
  • Pintas Character 2007
    Producer:
     Wine & Soul
    Grapes: More than 30 varieties
    Price in Sweden: Unknown
    Number on Systembolaget: was 99809, now out of stock
    Raspberries, blackberries. God tannins. 100% of the grapes pressed by feet. 1.5 years on new french barrels. The producer also have a higher class wine also named Pintas that is very well priced. And the vintage port they make under the same name is said to be one of the best there is, although not produced every year. The producers are very new on the market, finished their education in 2001. This, Pintas, is their hobby project as they are employed as wine makers on other houses as well.

Interesting was to hear stories about the grapes. Aragonez is in Portugal what in Spain is Tempranillo (which explained the similarities between these wines and some spanish ones. Also that Touriga Nacional is said to be one of the greatest grapes from the region. It gives a lot of everything: acidity, color, taste, roughness, alcohol etc. It also has a clear taste of violet that can bend towards licorice and even bergamot.

 

TextMate will not launch

I have had some problems with TextMate, but I did not know when it started. Probably since my last reboot (which would in fact be a correct guess) but since I do not reboot my mac very often I have no idea when it really started. I thought it had something to do with some zombie process, some socket that stalled, or something like that.

I was almost correct.

After looking carefully at what happens at boot, looking around on the net etc, I found that TextMate has a file where it keeps track of the process id of itself. If the mac crashes in such a way that the file with this status is not removed, TextMate will not boot if the PID in this file happens to be used, but not for TextMate. It simply believes itself is already running. Etc.

The file is the following:

~/Library/Application Support/TextMate/TextMate.pid

If this file exists, and TextMate is not running, you can safely remove it1. After removing it you will be able to launch TextMate again.

1. Note that the path do have a space in its name, a space that must be quoted if you remove the file on the command line.

Adobe CS or something else?

I was notified in Twitter that my friend Rikard Nilsson was installing Adobe CS 5.5.

Rikard tweets

Rikard tweets

My response to that was of course “Why?” and he then stated he was doing it just because he has always done it.

I have myself as well just followed the stream, because I thought I needed CS as well (or whatever it was named before it was CS). But one time I asked myself what I was doing. Is it worth the money or should one do something else?

I have, to be fair, never used the advanced features in the Adobe series. I have never had any need for specific fancy Photoshop plugins for example. What I have used during the years are simple Photoshop editing and then more and more Illustrator work.

And for that I today use two other tools: Graphics Converter and OmniGraffle. They are not for free. Graphics Converter is €34.95 and OmniGraffle is $99.99. But much cheaper than the CS package.

Yes, I do have the older CS package installed, and yes, I do open some files in Illustrator still when I really need to edit some more advanced illustrator things. But I think that is fine. I do not need the newest shiny version, and the old license I have is still ok to use.

I think my message to Rikard and others are, yes, you might need the CS package, but do not just ignore for example these two great software packages. I can live without CS, but I can not live without these two.

Are we in a new bubble?

Update: Jocke just said in a tweet that I completely misunderstood the article, and he promised to clarify and educate me. No one would be happier than I if he have time to do that.

A blog post by my friend Jocke made me think, are we in a bubble? After tweeting a bit with Jocke, I find that I have to at least explain my thinking, even if I draw the wrong conclusions. Compared to some of my friends, I am not in the VC business, although I have been involved in a number of startups and do know what it implies selling, buying and investing in companies. Specifically regarding control over a company. And business models, yes, I know that.

Anyway, what is it that I see? If we look at what Jocke saw in the graph, the average successful startup raises $25.3M and sells for $196.8M. What I am looking for are things like what the revenue (or turnover) is, what the burn rate is, and have that compared with the value, the size of the future market and what success the products might have.

This because I see an investor wanting the money invested back and it can happen in basically two ways: by selling the shares and by getting dividend. And if companies in the startup phase is growing from $23.5M to $196.8M, that is of course growth in value that is extremely good for an early investor, but what is the situation for the later investor?

Should the value continue to grow another 7-8 times, or should the later investor be happy with just the normal expected growth of maybe 10-15% on a y2y basis? And where does this growth come from? Evaluation once again based on future value, or expectation that actual revenue will grow enough?

And if we look at a number of investors, and a number of investments, each investor of course does not expect revenue on every investment, but as an average an investor most certainly want growth that is above other investment instruments. And if every investor expects that, we as a total can say that investment as an instrument should grow with say 15-20% on a y2y basis. If it grows faster, I ask where that extra money is coming from? Just evaluations? Evaluations plus external capital that might come from borrowed money where the shares are security?

What would be interesting to see, as pointed out by CityNetworks, the graph do not show the non-successful startups. The ones the investors have invested in that failed.

The reason why I am nervous we are in a new bubble is if the investments are done in series where money invested comes from another sale based on another potentially grossly positive evaluation. And then repeated among investors and companies. And that no new money is injected in the spiral. Or if the newly injected money comes from for example various loans where the shares in the over evaluated organizations is security. Loans that must be paid back one day.

Anyway, where does the revenue from then? Well, call me traditional, but I am a guy that likes look at revenue, market share and such figures. Is the operation of the organization possible sustainable based on the revenue? Is further development possible to do without external capital? Is it in the future? If so, when? And, most important, is the business model so positive that it can start give money back in one way or another to the investors?

If it is the case that the evaluations are increasing in a spiral, simply because investors can invest more because the last exit was so successful, that series of investments and exits might one day collapse. Was that not what happened 2001? Not an IT-bubble, but an investment bubble. Of course we do not know that as we do not know the failures among the startups discussed in the article Jocke refers to, but just looking at difference between investment and exit values, well, that to me is a too simple picture (as Jocke says) as there are other parameters that one must look at before drawing conclusions. If we just look at the almost 8-fold difference between investment and exit, that growth to me is too large to be healthy.

How is it going with the Arduino?

The Arduino

The Arduino

I got that question today, and the answer is “just fine”. In fact, it is up and running, sort of doing what it is supposed to do. You see in the picture to the right what things looks like, and in this diagram, pump.pdf, what the circuit looks like.

All Arduino code is not written yet. I so far do only have one sensor and one relay managed by the code, but that works as it should.

If people have comments on the schematics, I do not mind getting ideas on how to improve it before I solder all together.

iPhone not detectable by iTunes

I managed to solve a problem that might be something that one of you encounter as well.

I installed a new MBP fetching with migration assistant stuff from a MBAir. On the air, XCode was installed. On the MBP after migration XCode was not available (one of those things Migration Assistant do not migrate properly, as network profile settings, VPN configuration, WLAN configurations and million other non-documented things).

I did installed again using the XCode Installer that was available in /Applications, but that did not solve the problem (i.e. XCode was not installed properly). I then removed XCode Installer, and installed from App Store again. And XCode was now fine. Part from crashing in an unexpected way because “it did not find some iPhone stuff”. I did not care, as I only wanted gcc.

That was yesterday.

Today I was to sync my iPhone, but the phone was not detected by iTunes. Weird…reboot everything, and then check system.log where I found this:

 Dec 31 10:54:14 ix com.apple.launchd[1] (com.apple.usbmuxd[764]): posix_spawn("/System/Library/PrivateFrameworks/MobileDevice.framework/Versions/A/Resources/usbmuxd", ...): No such file or directory

Very weird.

Some googling did though show me this blog entry.

I.e. an installation of XCode could in some cases ruin your iTunes.

Anyway, now when I knew I had a working XCode Installer, I did run it again, and voila, not only did XCode start without crashing this time, iTunes worked.

My first Arduino project

20111220-143606.jpg

20111220-145235.jpg
With the help of a ninja and lots of use of Google, my first Arduino project is up and running. Can not say I understand every detail but…

Arduino

Arduino Ethernet

Arduino Ethernet

I have a project that have been sleeping for a while. Basically it has to do with connecting a few sensors in my house to something and then keep track of what is happening. In the longer run also be able to report when something is wrong, and collect statistics.

The coolest thing (today) of doing this I think is to use an Arduino. The programming is simple, and the electronics (that I am most nervous about as my skills are about zero) is understandable. Almost. At least with help from friends.

Today I finally booted my Arduino Ethernet. I am running Mac OSX 10.7, i.e. Lion, and that created a few issues.

  • Ensure you have a USB-FTDI cable, or else it will be hard to connect the Arduino in the first place
  • The Arduino code that exists today is version 1.0, and it only downloads the Arduino.app file, no disk image as it is said in the instructions for Mac OSX. Do not be confused!
  • You do need a driver for the FTDI, and you can find it here. Download the 64 bit version.
  • You do not have to reboot when installing the driver. The install program is doing the right thing.
  • Install the driver before you connect the Arduino and before you start Arduino.app.
  • On the Ethernet board, a few ports are busy for the ethernet. For example, the LED (that sits beside the power port, that btw is 7-12V 2.1mm center positive) is on port 9. Because of that, the blink example must be changed to work.

Then you can boot, and run some examples.

I have not used the ethernet port yet, and will most certainly write about those findings as well.

Kent Berggren

Kent Berggren

Kent Berggren

One of my oldest friends on the Internet passed away quietly night between Saturday and Sunday. We have known each other since 1992 or even earlier when Swedish Radio contacted us at Royal Institute of Technology to discuss email deployment.

Kenta was one of the few persons in the world that although he was misunderstood now and then (partly due to dyslexia) was one of the most loyal persons there is. And not a single piece of him was evil or had evil thoughts.

Me and people on mailing lists in Sweden where Kenta has been active the last 20 years or so already miss him — deeply. Even people that say they have never met him in person.

I have met him. Thousands of times, if not more. He always helped people when he could.

I am missing you already Kenta.

So much.