Pythoneer

Ich bin ja sonst etwas träger, aber als doch etwas umtriebigerer Internetnutzer muss ich hier mal um etwas Mithilfe werben. Es geht um eine Petition gegen das geplante Gesetz zur Sperrung von Internetseiten, diese Sperrung ist technisch fast komplett nutzlos und würde nur zur Abschreckung unerfahrener User dienen. Und die Kinderpornografie-Interessenten finden ihren Weg um die jetzt geplante Sperre so oder so ganz leicht herum. Aber das sind nur einige der Punkte die dagegen sprechen.
Weiterfürhrende Infos bekommt Ihr hier:

1) Das Video hier http://www.tagesschau.de/inland/petition106.html
gibt einen sehr guten Abriß.

2) Eine weiter Meinung eines geschätzten Freundes
http://schlueters.de/blog/archives/107-Guttenbergs-Eindruecke.html

3) Diese Petition wäre online zu unterzeichen
https://epetitionen.bundestag.de/index.php?action=petition;sa=details;petition=3860

4) Wenn Ihr Experten zu diesem Thema hören wollt, dann kann ich nur folgenden Podcast empfehlen
http://chaosradio.ccc.de/cr145.html

Solche oberflächlichen Äusserungen wie die von Herr Guttenberg “Das macht mich schon sehr betroffen, wenn pauschal der Eindruck entstehen sollte, dass es Menschen gibt, die sich gegen die Sperrung von kinderpornographischen Inhalten sträuben.” wären nicht entstanden wenn er auch nur diesen oben verlinkten Podcast gehört hätte (1h:56min Zeitinvestition gehören da schon dazu), dann wäre ein technisches Grundverständnis da, welches ihm selber die Nutzlosigkeit und den Tropfen auf den heißen Stein, der diese Aktion ist, verdeutlichen würde.

I have a Zyxel router here, that just didn’t want to work with WPA, after searching with the right terms I found this!
Update the firmware as given in the link above, be sure to have a matching version of the hardware!
Good luck.

When I have some time I might look into an alternative way to having a screen saver again, that shows me the uptime. The old version of showoff just crashed my system preferences, so I kicked it. It’s not that important that I spend a minute more. But some search brought up the above page and it’s a bit of work, but learning is always cool.

The plan for May 31st, 2008 is

15:00 meet at Weltzeituhr, as mentioned on upcoming, more info and a pic see here. When you are on Alexanderplatz you should find this clock pretty quickly.

19:00 - 22:00 we will be in Las Olas
and afterwards go to some bar, club, whatever we find, if you didn’t make it to join us until 22:00 your chances are only to search all of Berlin’s nightlife. Or catch one of us in the IRC channel #dojo before and get a mobile phone number (our nicks are klipstein, nonken, mccain).

And I will have a couple of T-Shirts with me, so be prepared to get undressed

see you

Actually the headline is not really what I want to say :-). But I don’t know any better yet.
I just want to write down how I solved sorting the following:

>>> var maxSpeed = {car:300, bike:60, motorbike:200, airplane:1000,
    helicopter:400, rocket:8*60*60}

This is a list of vehicles with their (approx.) max speeds. And now, I would like a list of the vehicles sorted by their max speed. This looks sooo easy, and actually it is :-). You just have to know how.
Look at this:

>>> var sortable = [];
>>> for (var vehicle in maxSpeed)
      sortable.push([vehicle, maxSpeed[vehicle]])
>>> sortable.sort(function(a, b) {return a[1] - b[1]})
[["bike", 60], ["motorbike", 200], ["car", 300],
["helicopter", 400], ["airplane", 1000], ["rocket", 28800]]

Since the sort function takes an extra argument, that allows to do your custom comparison we simply make sure that the right stuff gets compared, which is our speed. Unfortunately we can not sort the object directly, it has to be an array for doing so. But you can now convert it back, or just extract the vehicles, like so:

>>> dojo.require("dojox.lang.functional")
>>> vehiclesSortedBySpeed =
        dojox.lang.functional.map(sortable, function(i) {return i[0]})
["bike", "motorbike", "car", "helicopter", "airplane", "rocket"]

Didn’t I say it was easy?

[UPDATE] Thanks to Joe and others hinting me to the danger of purely using indexOf() without a check for -1. Added it in the last example.

I mean it’s not the first time I am doing it, but it happens always again that I have to look up how exactly it works. And every time I know it is very easy.
So here is how to remove an element from an array, mainly for my poor brain, so it doesn’t have to remember anymore. Though I know, now that I wrote it down I will never forget it again.

DON’T
Unfortunately the most obvioous solution leaves a “hole” in the array :-(.

>>> var list = [4,5,6];
>>> delete list[1];
true
>>> list
[4, undefined, 6]

Sad, but understandable.

DO
So do it right and use splice().

>>> var list = [4,5,6];
>>> list.splice(1, 1); // Remove one element, returns the removed ones.
[5]
>>> list
[4, 6]

Useful DO
Actually, what I mostly need is: remove a certain value from an array.
I.e. I have some list of visible IDs and I want to remove one because it’s not visible anymore. So just extend the above example a bit.

>>> var visibleIds = [4,5,6];
>>> var idx = visibleIds.indexOf(5); // Find the index
>>> if(idx!=-1) visibleIds.splice(idx, 1); // Remove it if really found!
[5]
>>> visibleIds
[4, 6]

I hope I didn’t bore anyone with that

… we will be able to deliver much quicker, much cooler applications to the user. No installation needed, just an online connection. Not that I was not already talking about that for a while, but there was no hook between the position that your device knows and the web app you are running. Google Geear LocationAPI will provide it. It’s great to see the technologies go the right way to really provide the power to enable easier writing of really good apps. Let’s see what people come up with …
Btw, watch this video it is very nice in showing what gears is good for and what is yet to come, and you can hear dojo over and over again :-)!

I just changed the config of my application and tested it if it still works, since I only switched DB names. Everything is cool. It works, all the pages show the same stuff it seems to work. Great thing!
A couple minutes later, I go to the shell window where I did that change and am wondering what server I am on. Geeeeeeee, the online version. I changed the DB to use on the live version. God damn it.

It was no big deal, since I hadn’t restarted the server yet, since my local dev server (django) automatically restarts when I change any python file. So I just undid the change online.
Something has to change that this won’t happen again, searching the web a bit and with the help of Daniel I found this article “Colorful Terminal”, where he had solved the same problem. Great solution.
Just one little thing, when I accidently forget to start the script which logs me into the server and changes the background color of my terminal … what then? Can the script not trigger on “ssh me@server.com”? Any ideas?

I open this screencast (and I love to watch screencasts before trying the demo). And I see it is 10 minutes long, it has no chapter markers and I don’tknow which parts I can skip to really only see what I am interested in.
So let me just adjust the playback speed! That would be so cool! Whenever I think it’s interesting I just slow down the video speed (if I have not yet gotten used to the speed :-)).
Just give it to me I am an information junkie, but I need to consume faster, so I can consum more!

I always wanted this! A stand-alone web application, kinda. I just mean put gmail into it’s own application, so I can separatly launch it from the rest, especially when the browser crashes I don’t want to reopen all the sites, or reconnect my meebo, etc.
Prism, a mozilla labs project, is here to do it. Great I will follow that! When is that coming based on WebKit?

The article How to mount a remote ssh filesystem using sshfs very well describes what I have to do to have my disk mount on my Mac disk on my Ubuntu system.

After I got that working I only need to add the following two commands to “System > Properties > Sessions > Start programs”:

# Run synergy-client on ubuntu, so I can use only mouse
# and keyboard of the Powerbook after startup.
synergyc -n ubuntubox 192.168.0.100 

# Mount my user's Powerbook directory to the directory
# /media/macbox on the ubuntu maschine.
sshfs macbox:/Users/cain /media/macbox

(”ubuntubox” is the name for the ubuntu maschine and “macbox” is the name for my Powerbook, and 192.168.0.100 is the fixed address of my Powerbook).
Now I can just turn on my Ubuntu maschine and it comes up, starts everything and is just ready to go, of course I also turned on the auto-login, since there is nothing intersting on there anyway and it’s just the most convinient … cool

I just finished listening to the podcast on TalkCrunch (mp3 download) where Michael Arrington interviewed the founder and CEO of PayPerPost, Ted Murphy, Josh Stein a Director at Draper Fisher Jurvetson and Rob Hof Silicon Valley Bureau Chief at Business Week. It was one of the best podcasts I had heard so far.

PayPerPost connects advertisers with bloggers that are willing to blog about certain products, that the advertisers choses and pays them for blogging about them. The advertiser has to approve the blog article before paying.

The business model seems to be discussed heavily in the blogosphere, as you can already see at the number of comments and links to the article about it on techcrunch. The podcast is a very interesting discussion about the business model, if it is ethical or not, the possible future of such service and other companies and facts around the topic. It is really worth listening to!
My opinion here is that PayPerPost does not need to be regulated by any higher power or is unethical. I found Michael Arringtons arguments a bit weak here, mainly he says it is not ethical when a blog post does not say that the author has been paid for. And he thinks that it is even worse when journalists write “anonymously” besides their employment (or contract). I just simply agree with the guy who comments that PayPerPost is just the same as “like when an actor in a movie just happens to pull out a Motorola cell phone and use it in a way that the brand is noticeable”. That is simply true, the movie doesn’t tell upfront “Attention: advertising being spilled on you, if you want or not”. And honestly there were so many concerns about Weblogs, Inc. and paying bloggers. And all this is forgotten and Jason Calacanis is moving on towards bigger goals. There is no big difference about how Weblogs, Inc. does it and how PayPerPost does it, PayPerPost just offers the platform without being the blog provider itself (which Weblogs, Inc. is) and they just connect those people. If people surf on Weblogs, Inc. they don’t know that the articles on there are from payed authors, until they digg deeper and want to find out.
Just let it go, I will be watching, it looks interesting.

What I like a lot about the way Michael Arrington is interviewing people is that he is walking the thin line of getting as many details as possible to driving them nuts by drilling for information as deep as possible. I had already a couple of grins on my face about his techniques and courage to do it. Great! Also the podcast with the Pageflakes founder Christoph Janz one episode later on TalkCrunch was driven very strongly by Michael to get the number of users of Pageflakes out of Christoph. Fortunately in the right moment Michael gave up and turned on to other important facts. That is what I mean by walking the thin line.
I remember very well how I was crossing a street and turning to go into the park when I started to have a big grin on my face because I was listening to the podcast and I felt that Christoph had to find good arguments to stay strong in this discussion or simply scream at Michael “we don’t disclose those numbers” :-). Fortunately when I had walked another two minutes (or so) when I was aside the nice river Isar and Michael said “I think we have beaten this topic to death” and I was releaved that I didn’t have to listen to another round of well argumented questions from him. One can hear that he is a lawyer. Great stuff Michael, keep it up! Beside Venture Voice and FLOSS weekly this is my favourite podcasts.

16th-17th December there is a BarCamp in Nuernberg, which is not too far from Munich. I will plan to go there, let’s see how many people I can gather that will join. I am not really sure if I can really contribute something for a session … I will have to think about it, may be Dojo, Django, MySQL or Python … But I would also be interested to get some connections, may be find someone interesting to kick off some project(s) with … we will see.
Anyone from Munich around who would accompany me?

Thanks to this blog article I found Katapult, which is basically the same as QuickSilver for Mac OS X, without which I couldn’t live anymore. And since I have my Ubuntu screen placed right beside my Mac’s screen and am using only one mouse and keyboard for both thanks to synergy, I was moving over to the Ubuntu screen and wanted to start an app by hitting APPLE+Space … doh … no way. So finding Katapult just really makes working with Ubuntu easier. Though it is a KDE app Ubunbtu let’s you run it and you don’t even realize that.
Ubuntu is just great.

Finally I got around to setup a proper system at home, after a while of thinking that a laptop is the overall best solution I am happy to have two 19" BenQ FP93G X, one connected to my main machine, the still in use G4 15" PowerBook and the other one hooked to some older AMD about 2GHz running Ubuntu.

And today I connected the awesome sound system SoundSticks II from harman/kardon. After a little of searching I also found out how to turn up the volume. And they even had thought of allowing to adjust the direction the satellite speakers point, this sound is just awesome. And now listen to Pet Shop Boys Sodom (Trentemoller Remix) (link to iTunes) and this system was worth every cent!

Actually I do only want to know about objects in my application, I don’t really need a relational database in the common sense below my application. It’s just one more level where data have to be mapped to something unnatural. So why should I not start looking a bit closer into real object storage, like db4objects does it. I will have to take the time, may be I can cut another unnecessary layer of complexity from my application.

The technology is always making us take a detour, simply because it is not really a human thing to use. How human is it to communicate with each other by converting our interaction signals (voice, gestures, …) into digital streams of multiple zeros and ones. This is so awkward and simply feels unnatural. The same applies to writing web applications.
First we are limited by the capabilities of the browser. Then we have to support all the browsers, that in turn again implement the same technology in different ways. Beyond the browser we have to communicate over a dumb protocol, which does not allow for real interaction and actually makes us covert our data in a way so that we can send them over the protocol. Next and probably the worst of all, is we are trying to use those data in a way that has a completely different way of modeling the data then the actual usage case does it. Wouldn’t it be ideal to auto-determine what and how to persist data, states and combinations of them by “scraping” the user interface that is created for the user? Why do we have to map the user interface first to some low-end technology, some markup, later convert it into data streams, use it with objects and finally persist it in a completely different final format, in a database. Oh man, this seems soo much overhead and error-prone. But I am getting too philosophical. Back to work …

I just came across those three very interesting articles (they are linked one among another, so finding them was easy :-)). Since I am building a new web app I have to think about the following things. And I have to do it better than I thought before I read those articles.

The well-known article Cool URIs don’t change from Tim Berners-Lee.

URL as UI is also about URI design, it contains a couple more practical tips and some reasoning that I should really consider when creating my URIs like: easy to type, all one case, hackable, no longer than 78 characters, etc…

And I should read this one about Usability of Confirmation Email and Transactional Messages, it just costs a US$100.

As I was looking for which external displays I get to optimize my working environment, since virtual desktops are nice but still not the end of the road, I came across this discussion that was aiming towards the same. And by reading this it seems that it is not possible to run two external displays on my Powerbook.

Actually I didn’t just want to close the lid and run only the two external display, no I wanted to put one on the right and the other one left side, so that I get a total of three displays. That would really give me the ability to open the editor, browser and log-file viewer at once. That would be all I need in my working environment.
So I will finally only get the one extra display. What a pitty …

[Update] Well, obviously it’s not all impossible, you just need the money. They support even up to four displays. Once when the business is really going great I might get that :-).

Obviously the latest software update for my Mac had fixed something and my problems with crashing after an uptime of about 14 days are solved. That is very pleasent. And my fear of the “You have to restart your computer” screen is becoming smaller every day.

Of course I had seen the newest features of Leopard that Apple had shown yesterday at WWDC 2006. And for one night I was sure that they had built the same thing that I had wanted since I have a Mac. But Apple’s Spaces are only virtual desktops - bummer. My Spaces2.0 is not only a virtual desktop switcher. No, Spaces2.0 let’s you really switch between working environments. One Space in Space2.0 is like another Mac on your Mac. What? Let me explain as I would use it in real life.

My Work-Space - is my Mac with my opened Mail.app, Safari, IRC, Console etc. and all the applications that only relate to work. So

• the E-Mails I am reading are only those that I receive from my work accounts,
• the chat contains only the chat rooms I am using with my colleagues,
• Safari only shows the pages I opened while surfing for info related to my work
• and the Console does only show all my work in progress, that I am doing on the command line.

This does not allow any private E-Mail, Chat, News or what-not to interfer with my working life. I completely switch as if I had one computer only for my work, which has nothing to do with my private computer. buying two Macs just for that reason would be quite stupid, expensive and twice as much to carry :-).

My Private-Space - is my Mac with my opened Mail.app, Safari, IRC etc. and all the applications that only relate to my private activities. So

• the E-Mails contain only the private mails from my family and friends,
• the chat rooms I am in are only those I visit for my personal fun
• Safari shows only the pages I personally like to surf on.

There is no work related application opened, I don’t get distracted by any co-worker’s E-Mail or chat-query, while uploading my private photos and planning our next weekend trip. No, Spaces2.0 let’s you separate your lifes as it should be.

Why?
I am using my Powerbook as my exclusive computer, I am doing my work on it and my private stuff. Currently I have things separated by using either separate projects (in my editors, for programming) or by using Mail.app for my private mails and Thunderbird for my work mails. And I am also trying to open certain browser windows only for work related stuff and others only for private interests. But this kind of separation is very hard and work intensive to do. And I only got one chat client and one Skype installed, so when the icon starts jumping in the Dock I still don’t know if someone wants something from me related to work or privately.
That I want to separate. My Skype for work would only have my Skype contacts that are work related, all my family and whatever would not see me, since I am not online for them. I wouldn’t receive private E-Mails since I am not checking my private accounts but only my work accounts.
This separation is just what leads to less distraction and being able to better concentrate one the one current thing. There are many articles describing this syndrom of news-addiction. Spaces2.0 would be the cure.
And virtual desktops have to be a feature of each of my Spaces, of course.

May be I should tell Apple about my dream