Explain the world with maps. - UUorld
via uuorld.com
Support outil d'analyse de données.
Explore Posterous
Search posterous
Search all posts and users. Type a name, type a favorite song title, whatever! See what comes up.
More posterous blogs
Here are posterous posts filed under data...
yepp, genau. die c't ist eine alte dame…
Ich lese die c’t ja schon länger nicht mehr, vor allem weil es mich genervt hat, daß die Computer-Welt sich massiv schnell bewegt, aber in Hannover das Web immer noch keine Rolle spielt, sondern nachwievor Hardware-Specs, Testlabore und irgendwelche abstrusen Anwendungsszenarien, die irgendjemand unbedingt niederschreiben mußte, noch dazu sind nahezu alle Artikel längst “durch”, man hat das meiste schon vorher im Web lesen können.
Nun bin ich eben gerade bei heise.de über folgenden Artikel gestolpert: Chrome OS: Der “Windows-Killer” fällt aus – dort schreibt Oliver Diedrich:
Seit dem Sommer geistert Chrome OS durch die Schlagzeilen, jetzt hat Google eine erste Entwicklerversion als Chromium OS veröffentlicht. heise open hat sich das Google-Betriebssystem angesehen und kommt zu dem Schluss: Ein Windows-Killer ist es nicht – und will es auch gar nicht sein. Chrome OS, an dem auch Ubuntu-Entwickler mitarbeiten, zielt auf Web-Appliances, die immer online sind und im Wesentlichen für Internet-Aktivitäten genutzt werden. Das System spielt in einer ganz anderen Klasse als Windows und Mac OS X, aber ist auch kaum mit den etablierten Linux-Distributionen vergleichbar.
Natürlich wird Google Chrome OS der Windows-Killer sein, und zwar präzise, weil es nicht mit Windows vergleichbar ist, sondern sich auf das konzentriert, was die Masse der Anwender interessiert: Anschalten, Nutzen.
Diedrich schreibt dann weiter, um seinen Punkt zu untermauern:So verzichtet das Google-Betriebssystem komplett auf einen Desktop, die einzige Anwendung in Chrome OS ist der Google-Browser. An die Stelle lokal installierter Programme treten Web-Applikationen; in seiner Standard-Konfiguration ist es ohne Google-Account gar nicht benutzbar. Für den Einsatz auf Web-Appliances bietet es allerdings interessante Ansätze.
Ja, eben. Das ist ja gerade das Spannende an Google Chrome OS. Ok, aus Sicht der c’t wäre das natürlich total fatal, wenn künftig nicht mehr irgendwelche Viren-Programme getestet, keine 1500 Superduper Office-Tipps niedergeschrieben werden müssen und so weiter und so fort, sondern wenn die Nutzer einfach einen kleinen Rechner hätten, den sie einschalten und dann gleich loslegen können, mit lauter Nutzungsszenarien, die alle sofort realisierbar sind, da die Applikationen oder Inhalte irgendwo in der Cloud zu finden sind. Diedrich versteht nicht, oder will es nicht verstehen, dass die Welt sich schneller ändert als man es in Hannover wahrhaben will.
Google Chrome OS ist der Windows Killer – und zwar eben genau aus einem Grund: es ist nicht mit Windows vergleichbar und macht damit den dramatischen Unterschied. Es “spielt in einer ganz anderen Klasse” – sagt Diedrich selber, aber die richtigen Schlüsse kann er nicht ziehen. Schade, er war dicht dran. Das Web wird den Desktop ersetzen und Google Chrome OS ist ein massiver Schritt in diese Richtung.
Share:
via lumma.de
Opendata campaigns can wait.. There's analogue data to save first!
At the #flookup last night, @cpchannel sparked a great conversation about data and how we're about to lose a lot of the important stuff...
At the moment it seems there are many campaigners for what I see now as a luxury of opendata. This is a worthy cause. I think though there is one to deal with before it. Paper data. @cpchannel (aka Chris in the real world) pointed out that in Henley is the office of an organisation which manages all of the data on small green spaces in the uk; not big parks but those little areas of green which exist all over. Their location data, rights, sizing etc is all stored by this organisation. It is all stored on paper though.Another example is the organisation who keep all of the data on public water fountains (of which there are now only 15 in London @cpchannel informs me, down from arund 2,00 at one point). This organisation too has plenty of useful data but it is all analogue on paper. Which prompts the question, what happens when one of these organisations gives or burns down. All that data is lost. Culturally they're (without exaggeation) run by OAP's who are not bought in to buying the IT mans services for document management and spreadsheets etc... Will a uni graduate want to go and work for a dinosaur of an organisation that is 100% paper based; the concept would be very alien (note there's another topic here as I found out kids at school from 5 learn powerpoint!) So there's all this paper data... There really should be a movement to get it digital... Then we can worry about it being open... At the moment we risk losing information.
"2012" - prediction come true?
After watching the movie "2012", some friends treat it as a fiction and give it a smile, some become worried what they can do to save the world, and some (may include you and me) are googling around to see if the Mayan predictions have come true in the history. Here I would like to crunch the numbers to see if it is just a joke or a real end, after all, "If you torture data long enough, it will tell you anything you want!" The 2012 phenomenon originates from the Mayan calendar which ends at December 21st, 2012. I tried to find all the Mayan predictions online and failed to do so, the only prediction I found other than 2012 if the 911 in 2001, well, the Mayan long court calendar ends at 2012 does not mean the world ends at that year. Maybe the guy who made the calendar just don't think the humans can live that long. All in all, I don't have enough data for the torturing.
In the movie, the catastrophe starts from the volcano in Yellowstone national park. So the earthquake events in YellowStone should give us some information about what is happening in the heart of our planet. Fortunately, the US government does have regular data for the earthquake events in YellowStone from 2006. I spent some time digging some basic information from the data. In the following analysis, I only plotted the number of earthquake events observed, no consideration of the magnitude of individual or average earthquakes. Fig1 plots the number of earthquake events from 2006.11 to 2009.10. We can easily see a seasonal pattern. And notice in 2008.12-2009.1, the number is almost twice as large as those in 2007.12-2008.1. And it is the second largest in the recorded history based on quotes from The Yellowstone Volcano Observatory website. What will happen in 2010.1? Will it follow the red line I drew? We'll see. "The recent swarm (2008.12-2009.1) is well above typical activity at Yellowstone. Nevertheless it is not unprecedented during the last 40 years of monitoring. Swarms are the typical mode of occurrence of earthquakes within the Yellowstone caldera, with magnitudes ranging to > 4.0. The 1985 swarm on the northwest rim of the caldera lasted for three months, with earthquakes up to M4.9 and over 3000 total events recorded."Fig2 and Fig3 plotted the trend without seasonal effect and the seasonal effect respectively. We can see October is the safest, while Dec-Jan is the most dangerous one followed by April and July. The trend is absolutely go north, however, will it cause the volcano and then the end of world? No answer. The recorded maximum is 3000 per month and we still have a long way to reach that point. But who knows, we will check the data in 2010.1 and 2011.1 to get the latest information from the data. Don't forget the last point: It is our government who is providing the data.For those of you who want to analyze the data, please see attached.
How not to analyse Twitter's stats - pt.43
CAUTION: unfiltered rant...
Another day, another thrilling piece of navel-gazing by the tech industry. This time from the mighty TechCrunch who give us Blogging Vs. Microblogging: Twitter’s Global Growth Flattens, While WordPress’ Picks Up. Good to see they're picking up some tips from political journalism - stuff the analysis, it's a horserace!
The article ticks at least one of the boxes in the Heaths' classic Made To Stick about what makes stories memorable: it is 'unexpected'. All year it's been "Twitter Twitter Twitter" and "blogging is dead". But what's this? Mercy me! Twitter's "flattening off" and Wordpress is "picking up". Look, they've discreetly annotated the graph to help you out:
My problem (apart from possibly terminal sarcasm) is that, as almost anyone who uses Twitter knows, a lot of people use apps to tweet and may never visit the website at all. A *lot* of people. A random trawl through my stream shows up half a dozen different apps used in the sapce of a few minutes.
Actual research on this seems scarce. However, this effort from Sysomos has the web as by far and away the most common means of tweeting, used by 46%. However, that still means the majority of users *don't* use the website.
[Disclaimer: I have no reason to distrust these raw figures, however some dubious analysis elsewhere in the report dosen't fill me with confidence.]
If this is the case then what is the point in comparing Twitter website visits with Wordpress? Website drop-off could quite easily be offset by increased app usage, a not unreasonable hypothesis. *You are simply not comparing like with like.* I have no way of knowing what Twitter's trends are overall. They may well be flattening off after this year's insane growth. I'm no expert. What I do know is that there are a *lot* more interesting topics that could do with some serious analysis right now rather than a perpetual horserace propped up by dubious evidence.
The Boom of Social Sites
Simple Complexity brought this cool bubble graph (created by Alexa and TechCrunch) to our attention. It does a great job of charting the rise of social media on a macro scale.
Kada baigsis karantinas?
Dėmesio, Achtung, внимание!!!
Kaip galbūt jau skaitėte, karantinas turėjo baigtis lapkričio 26-ą dieną (ketvirtadienį), kada jau būtų reikėję eiti į gimnaziją. Tačiau mūsų žurnalistai visai nebijodami gripų išsiaiškino, kad gimnazijoje karantinas baigsis lapkričio 30-ą dieną, pirmadienį. Tai reiškia, kad jau pirmadienį visi laukiami pamokose.
Ligoms ir epidemijoms tarkime NE!
Ligoms ir epidemijoms tarkime NE!
ChartsBin, data visualization
ChartsBin meets a growing collection of experiments, performed with free software, playing with the creation of interactive charts and thematic maps based on statistical tables. For example, here are some samples.
Historical Population of World, 1 AD to Future
Worldwide Driving Orientation by Country
Average Age at first sex by Country
Today's Resource: Census On the Map
There are some great, public (read: free) resources for analyzing economic data, though many times you have to be willing to wade through and reformat information for presentation purposes. The US Census Bureau collects and analyzes a mountain of useful data.
"LED on the Map" (for the curious, that stands for Longitudinal Employer-household Dynamics)
- Provides a map-based view of where people live and work, with several additional layers of analysis on wages, industry, and commuting patterns.
- Data lags a bit - the current most-recent analysis only covers through 2006 so the current economic situation is not reflected.
Here's an example:
This map shows where workers earning $1200 per month or less live who are employed in Tulsa County. Remember - this is up through the end of 2006.
