Vacuum - Edward Vielmetti in Ann Arbor, Michigan 48104

How to download all of your twitters into Excel

I'm scheming to make Twitter more productively personally useful, and as a part of that I'm finding a need to pull my archive of past twitters out into a format that's easy to parse with the tools that I have at my disposal.

A goal, then, is to write a command that generates a file which can be imported into a spreadsheet so that all of my spreadsheet-fu can be applied to it. (There isn't much spreadsheet-fu, but what there is is very handy; I'm particularly fond of interactively sorting through tables, and I long to be proficient in pivot tables.) The easiest way to write such a thing is to describe it enough detail that someone might have written it already, and then you can just use what they did.

The Google search terms would be
twitter (import OR export) (excel OR csv)
and I get 1.9 m documents, none of which are an exact hit, though there was a useful one about importing your Twitter contact list into Dopplr.

The del.icio.us tag looks like
http://del.icio.us/tag/twitter+export
but it doesn't get me what I want either, though I did discover that "social network" in German is Freundesnetzwerk.

Think a little bit, search for
Twitter API
and a few clicks later I'm at the Twitter Fan Wiki . That led me towards twitish, a Perl based shell for twitter. There's also python-twitter, a Python wrapper for the API.

that's fine but what I really just want is not to write code. Posting this as a draft, pls comment if you find it, otherwise I'll look again when next the urge hits.

using embedded Google Maps to automatically geocode blog postings

An idea (for someone else to implement).

Now that you can embed Google Maps into blog entries, it should be possible for whoever is doing blog search tools to parse those embedded links for geography and then auto-code those blog entries geographically. E.g. if you post a map with a piece of downtown Ann Arbor in it, it should be automatically geo-tagged (either by the blog authoring software or by the blog search engine or by a feed management tool as belonging to that geographical area.

This idea affects sites like outside.in , topix.net , and Feedburner , and would likely be implemented on the feed side as a GeoRSS based extension.

Technorati Tags: 48104, feedburner, for:ajturner, for:rklau, gmaps, google, kennecott, maps, outside.in, todo:add-category-maps, todo:add-category-neogeography, topix.net

It's hard to have great weird ideas when you're tracking every billable minute

It's hard to have great weird ideas when you're tracking every minute to determine who it gets billed to. The cognitive overhead of deciding which part of your brain gets dinged for each good thought gets in the way of flow.

Some recommendations on time tracking software via an evening on twitter and jabber from the cafe (I love mobile jabber as a way to unobtrusively get good ideas from someone else):

mitten recommends Slimtimer ("All your timesheet are belong to us") in her post internet finds.

This is my new favorite tool for freelance work. I don’t have to think, I don’t have to calculate, just press stop and start when I’m doing various tasks and spit out a report at the end of the month. And it’s free!

anotherjesse recommends Jabber::Simple by blaine as a toolkit for building Jabber applications, e.g. a Twitter based time keeper. I think all I'd have to have would be something that listened privately and kept a log, but then the mind starts to wander into nice things like ways to figure out where the bus is from an IM or Twitter and not from a browser.

credits: Peter Kaminski from Socialtext for the phrase "great weird ideas"; Caterina Fake for the phrase "cognitive overhead" as heard on a Tod Maffin interview on a CBC Radio One.

Previously: It's hard to have great weird ideas when you're busy closing trouble tickets.

Technorati Tags: greatweirdideas, cognitiveoverhead, twitter, jabber, blackberry, cariboucoffee, slimtimer, mitten, timesheet, ruby, aata, theride, annarbor, googletalk, socialtext, flickr, cbc, rt

Pimp my legos: teaching kids to program (Prentiss Riddle at SXSW)

Prentiss Riddle is working on a panel session on teaching kids to program for SXSW in Austin, TX:

It's four busy months later and we find ourselves having to put our PowerPoints where our mouths are. In addition to the experiments we are planning on our own captive guinea pigs, we're looking for two or three co-panelists in the Austin area with extensive experience teaching programming to (say) 16-and-under kids. Despite the reference to Legos in the title, we want the focus of the panel to be about software, not just killer robots (although we certainly hope to have a Mindstorms person on board). We have some sources of panelists in mind, but we hope word of mouth will provide us with more. Suggestions?

My kids aren't quite at the stage where programming is one of the things they do at the computer, but I'll be watching this one with interest. I do know that one of the best things I ever learned in school was how to type, and most of my early programming expertise had to do less with writing code from scratch and much more with playing games like Lunar Lander to try to figure out how they worked.

Page Saver and Pearl Comments: new releases from Pearl Crescent

Ready for Firefox 2.0, Pearl Crescent has new releases out of their Pearl Comments and Page Saver Firefox plugins. Pearl Comments release 1.6 allows a team to work on edits to a web site by sharing in-browser comments on page elements. Page Saver 1.3 is an image capture tool for Firefox that lets you save the entire browser window, not just what you can see on your screen.

Mark Smith notes:

In addition, our partner companies are increasingly asking us to design and implement custom software for them. The "hot" area right now is Firefox extension development, although we are talking with various clients about Internet Explorer and Thunderbird-related projects too.

If we can do anything to help you, please let us know.

Technorati Tags: firefox, plugin

Results of a Google code search

According to Google Code Search, my code or ideas (or documentation or organization or answers to FAQs) have been incorporated into the following tools and software:

traceroute, an Internet debugging tool (1989)

gnus.el, a Usenet news reader, part of GNU Emacs (1990)

prospero, a web file system (1992)
alex, a web file system (1992)
ange-ftp.el, an FTP client, part of GNU Emacs (1992)

gn, a Gopher server (1993)
pine, a Unix mail user agent (1993)
mh and nmh, a Unix mail user agent (1996)
netics, an extensible network statistics collector (2002)

Notably, Google Code Search is not finding some more recent work, so I need to push a few more packages into better code repositories.

Jose Nazario was talking about using Google Code Search today to support the effort to find bugs in software - there are idioms that are regular and predictable that result in failures (he noted one where bitwise tests are coded wrong as an easy one to pick out with a regexp).

Interarchy 8.2 for OS X has Amazon S3 support

Interarchy (the program formerly known as anarchie) now has S3 support:

With Interarchy you can efficiently and reliably fetch, edit or SFTP/FTP transmit files to any kind of Internet server using FTP, SFTP, WebDAV, Amazon S3 or HTTP. Common uses include setting up web sites, long-distance transfer of data or remote server administration. Interarchy also supports other common Internet protocols including ping, traceroute, DNS lookup and packet sniffing. Because Interarchy is standards-compliant it will work with servers running on any operating system including Windows and Unix.

It's a $59.95 purchase via Kagi.

I wrote about Jungle Disk earlier, an Amazon S3 client for the Mac - Interarchy is miles ahead in the way of overall system maturity, since started out as an FTP client back in the olden days and expanded out from there.

Peter Lewis, the developer, wrote a tutorial on How Do I Make An Amazon S3 Hosted Website Using Interarchy - it's a start to finish description of using S3 as your backing store for a web server.

Technorati Tags: amazon, s3, amazonwebservices, aws

day.sh - date arithmetic in the shell

The little shell script day.sh gives you a convenient way to get a date in the present, past, or future into a shell script. Use it as follows:

% day.sh -o +100
2007-01-09

If you're using the Hiveminder shell script, this lets you do things like say

% todo.pl hide 4J $(day.sh -o +100)

which I have aliased to

% someday-do 4J

JOB: US/MI/Ann Arbor, Python/MySQL, publishing - contract (Rosenfeld Media)

Lou Rosenfeld shares this job listing - please contact him directly for more information.

Hi all; I'm looking for a local (Ann Arbor) Python developer to help finish up a project (and possibly work on its future development). The work would support the launch of an innovative vertical mashup for a new publishing house. Work would start immediately.

The technology: a web application (a simple Python MVC CGI wrapper) using CherryTemplate, SQLObject, some FormEncode, and MySQL.

Currently, it's nearly engineering complete and has a few minor bugs left to address. The site is nearly ready to launch, but there are a few tweaks to make immediately. Going forward, the site has some new features to add as well as some bugs and tweaks to make after a public launch. New features would get added over the next few months. It could also use some refactoring to improve maintenance. A seasoned Python developer responsible for the work to date would be available for limited assistance.

Please feel forward to share with anyone who might be interested; thanks!

cheers

--
Louis Rosenfeld :: http://louisrosenfeld.com
Rosenfeld Media :: http://rosenfeldmedia.com

Again, direct comments and questions back to Lou. Thanks!

Technorati Tags: python, cherrytemplate, sqlobject, mysql

Amazon EC2

I've signed up for Amazon EC2, the elastic computing cloud they are running, which promises you a server farm on demand at $0.10/hr/CPU. I'll be taking notes here and in the comments as I get it going.

The project I have in mind mostly needs a flexible infrastructure that is emphatically not running except when I'm actively using it, and my current hosting setup is somehow not quite agile enough to match. I think if I do this right I'll be able to turn off one of the accounts I keep for just-in-case offsite computing. We'll see.

Technorati Tags: amazon, ec2