tom alison dot com

Helvetireader in Mozilla Prism

2010-05-19T00:00:00-07:00

I switch back and forth between Netvibes and Google Reader for reading RSS feeds. The Netvibes box layout is great for scanning lots of new items, but Google Reader lets you power through content more quickly.

Today I noticed a co-worker using Google Reader with the Helvetireader userscript. Helvetireader applies a clean theme to Google Reader with an emphasis on readability.

Because Google Reader is a browser-based application, using Helvetireader requires installing the appropriate extension for your browser.

I’m not a big fan of installing too many extensions, particularly just to accommodate a site-specific enhancement like Helvetireader provides. This is where creating a site-specific browser (SSB) becomes advantageous. An SSB turns a single web site into a self-contained application. This provides finer-grained control over the experience of a site like Google Reader without interfering with general purpose browsing functionality.

I’m currently on Linux, so Mozilla Prism is my best choice of SSB provider. On OS X, Fluid accomplishes the same goal and provides more features.

Before downloading Prism directly from the Mozilla side, I ran a quick check to see if there was already a pre-packaged version of it available for Ubuntu.

apt-cache search prism

Not only was there a prism package, but there was also a pre-packaged build of Google Reader for Prism. Nice!

sudo apt-get install prism-google-reader

I initially thought that Prism supported userscripts like Helvetireader natively, but it turns out you still need the Greasemonkey add-on. There’s a Greasemonkey add-on built specifically for Prism available on Github. Building it is simple:

git clone http://github.com/teramako/greasemonkey-for-prism.git
cd greasemonkey-for-prism
sh build.sh

From within Prism go to Tools > Add-ons > Install… and choose the XPI file created in the greasemonkey-for-prism directory above. Then restart the Google Reader Prism app.

In the settings menu, you can now install any userscript. Go to Settings > Greasemonkey > Download User Script… and enter the URL for the Helvetireader script (currently http://helvetireader.com/helvetireader2.user.js). Now you should see Google Reader with the Helvetireader theme.

References:

Command-line WebDAV

2010-04-03T00:00:00-07:00

These are some reference notes I put together for mounting WebDAV volumes via the command-line.

About six months ago I switched from Wordpress to Jekyll to maintain this blog. Consequently, I saved a couple of bucks by cancelling my full-featured web hosting service and switching to the simple file publishing offered by my Fastmail.fm account.

Fastmail doesn’t offer ssh access, so files must be transferred by FTP or WebDAV. I wanted fully scripted deploys for my Jekyll site, and started off using FTP with the NcFTP client.

The problem with FTP was that my entire site was uploaded every time, even if most of it was unchanged. This wasn’t that big of a deal because I don’t have too many files, but it bothered me knowing that a better solution was available if I could just mount my Fastmail directory via WebDAV and then rsync everything over.

One thing worth noting is that rsync normally looks at both the modification time and the size of each file in the set to determine whether or not it needs to be copied to the destination directory. Because Jekyll always regenerates every file, the modification times are always updated. This causes rsync to copy your entire site over to the destination and subverts one of the best features of rsync - the speedup you get from only syncing diffs.

The way to get around this is to use the --size-only flag. This causes rsync to ignore timestamps. Most of the time this shouldn’t be a problem, unless you make minor edits to a file that doesn’t cause a change in file size.

rsync -rvz --size-only --delete "~/mysite_src/" "/mnt/webdav/mysite_dest"

See man rsync.

I use Ubuntu and OS X. There’s a different solution for each. Here are my notes.

Ubuntu

Install davfs2:

sudo apt-get install davfs2

To mount as a non-root user:

Note: Must run dpkg-reconfigure davfs2 according to /usr/share/doc/davfs2/README.Debian to get around “/sbin/mount.davfs: program is not setuid root” issue when trying to mount as non-root user.

Note: Add ignore_home kernoops to /etc/davfs2/davfs2.conf to get around the “/sbin/mount.davfs: / is the home directory of user kernoops” error.

Add an entry in fstab:

http://example.org/dav   /media/dav   davfs   noauto,user   0   0

Add your user to the davfs group:

sudo adduser <user> davfs

Make sure WebDAV connection credentials are in ~/.davfs2/secrets with restrictive permissions (chmod 600):

/media/dav   webdav-username   password

Perform mount/unmount:

mount /media/dav
umount /media/dav

References: man mount.davfs2

Mac OS X

mount_webdav exists but is unreliable. Could not get workarounds to work for password-protected WebDAV resource.

The best thing to do is first mount the WebDAV volume manually. In the Finder, hit Command+K to connect to the server. When prompted for credentials, allow them to be saved in the keychain.

Then, launch Automator and create a workflow that simply consists of the action “Connect to Servers”. Save the workflow to disk.

The Automator workflow can be run from the command-line via:

automator -i "https://webdav.example.com" <workflow> > /dev/null 2>&1

The -i parameter specifies input to the workflow. See man automator.

My automator command-line utility printed a bunch of warnings about certain additions that couldn’t be loaded. That’s why the output is directed to /dev/null.

Note: The workflow is itself a folder. The folder can be given as the argument to automator. Within the main workflow folder, there is a subfolder called Contents with a file called document.wflow. This file can be passed directly to the automator command-line program, and the containing folders can be ditched.

Eclipse and Ubuntu Karmic

2009-11-20T00:00:00-08:00

Today, for the first time in a while, I wanted to work with some Java code. I’ve always used Eclipse for Java programming, so I headed to their web site to download the latest version.

I don’t keep up with the Eclipse project anymore. I’m sure they’re doing lots of great stuff, but every time I visit their site - about once every six months - I’m more and more confused by the menu of download choices. It used to be that you just downloaded Eclipse. Not anymore.

Having just done a clean install of Ubuntu 9.10, I decided on a whim to see if it a packaged version of Eclipse built just for Java programming. Sure enough they did:

sudo apt-get install eclipse-jdt

Perfect. Everything installed correctly and a nice little ‘Eclipse’ choice appeared under Applications > Programming.

Eclipse launched and things were copacetic…until I tried to install plugins. Oddly, when I chose “Install new software…”, there were no pre-configured repositories. In a normal installation you’d see the default Eclipse update site, which offers a dizzying array of add-ons. But here there were no platform capabilities to add.

I didn’t fret too much - after all, I just wanted a simple Eclipse installation optimized for Java programming. I did, however, also want to add the Subversion and Maven plugins.

I added their respective update sites to the my Eclipse preferences and tried installing both plugins. No luck. Each plugin reported unsatisfied dependencies related to ‘org.eclipse.gef’ or ‘org.eclipse.zest’.

I knew those dependencies would be fulfilled by the current Eclipse platform update site. I just couldn’t figure out what the URL was. I found the GEF update site, which seemed like a sure bet, but it failed to resolve anything.

Things did not become clear until I stumbled across this comment in an Ubuntu bug report:

org.eclipse.emf is not installed by default; try adding

http://download.eclipse.org/releases/galileo to your update sites and try again.

Ah…finally - the canonical Eclipse platform update URL. After following those simple instructions, my plugins were able to resolve their dependencies correctly and everything worked great.

The logical question, posed by a commenter in the bug report, is:

…shouldn’t this galileo repository configured by default after

installing the eclipse packages?

Seems reasonable. The answer:

Yes; that repository should be configured by default - most of our

plug-in install related problems are actually related to it. If we knew how to make the build do it, then we would.

While I respect any developer honest enough to say “I don’t know how to do it”, the answer is unsatisfying. Offering a packaged version of Eclipse without the core platform update site is pretty close to offering a broken software package.

I’m sure someone will figure out a solution and fix the problem. My wild-ass guess is it involves generating TouchPoint Instructions to add the respositories correctly.

Command-line Document Conversion

2009-09-21T00:00:00-07:00

If you’re simply looking for a way to convert a one-off document, upload it to Google Docs or Scribd and let them take care of the conversion for you. Be mindful of your privacy settings so you don’t accidentally share your document with the whole world.

For command-line conversion of documents, here’s how to get going on Ubuntu:

Install OpenOffice (headless)

OpenOffice provides the core conversion facilities. They do a pretty respectable job at conversion of most document types, including spreadsheets and presentations.

You need to actually run an instance of OpenOffice in order to send it the request to convert the document. Installing the headless version means you can run in on a server without a windowing system, which you need to be able to do if you want to run a massive document conversion farm on EC2.

apt-get install openoffice.org-headless
apt-get install openoffice.org-java-common
apt-get install openoffice.org-writer

Simple conversion with CUPS-PDF

One simple way to do this is to use CUPS-PDF.

apt-get install cups-pdf

Use OpenOffice to print the doc to a PDF file using CUPS-PDF. Note that the output path can be found in /etc/cups/cups-pdf.conf (I’m using cups-pdf v 2.5.0)

Here’s a script to do the conversion and open the output PDF in evince.

to_pdf1.sh

#!/bin/bash
#
# Prints a file to PDF using OpenOffice and
# CUPS-PDF. Opens output file with evince doc
# viewer.
##

CUPS_HOME="$HOME/PDF"

file=`basename $1`
prefix=${file%.[^.]*}
outfile="${CUPS_HOME}/${prefix}.pdf"

echo Printing ${prefix}.pdf

soffice -norestore \
	-nofirststartwizard \
	-nologo \
	-headless \
	-pt PDF $1

echo Sleeping 10 seconds

sleep 10

echo Opening $outfile

evince "$outfile"

Note that in the -pt PDF option, “PDF” is the printer device name of the CUPS-PDF printer device. It may differ on your platform. Check /etc/cups/printers.

Better conversion with unoconv

The printing method works okay, but it doesn’t detect the orientation of your document, so you may notice presentations are rendering in portrait mode instead of landscape mode in the output PDF.

unoconv is a Python utility that talks to an OpenOffice process via an Uno bridge.

apt-get install unoconv

Because unoconv requires an OpenOffice instance, it’s best to have a process running before doing a lot of document conversion. unoconv can start one for you:

unoconv --listener > /dev/null 2>&1 &

Verify the process started:

$ ps -ef | grep soffice

Look for something like this (line breaks inserted for legibility):

soffice.bin -nologo -nodefault
-accept=socket,host=localhost,port=2002;urp;StarOffice.ComponentContext

Now you can bust out a script like this:

to_pdf2.sh

#!/bin/bash
#
# Converts a document to a PDF file using
# unoconv and opens it with the evince viewer.
##

file=`basename $1`
dir=`dirname $1`
prefix=${file%.[^.]*}
outfile="${prefix}.pdf"

echo Generating $outfile

unoconv -f pdf $1

echo Opening $outfile

evince "${dir}/$outfile"

This option is much nicer.

Java folks should check out the JODConverter project, which provides similar functionality.

Taking it further

Document thumbnails

Want to grab a thumbnail of the first page of your converted PDF? First install Imagemagick:

apt-cache install imagemagick

Then run:

convert example.pdf[0] -thumbnail 120x120! -gravity center thumbnail.jpg

This will generate a fixed 120x120 thumbnail in JPG format of the first page of your PDF. Note the array-like syntax: example.pdf[0]. This constrains the thumbnail generation to the first page. If omitted, a thumbnail of every page is generated.

Indexing document content

Want to add document search? Run:

unoconv -f txt sample.doc

This generates a text version of the document that you can index with Lucene or the indexer of your choice.

Read documents in the browser

You can generate an HTML version of your document to display on the web:

unoconv -f html sample.doc

For presentations and some other document formats, separate HTML files are generated for each page. A little messy…

You could get fancy like Scribd and render the document using Flash.

apt-get install swftools

Check out pdf2swf. Put in a little work to build navigation controls for your document and you’ve got yourself a pretty nice viewer.

Moving the Location of Your Flip Videos

2009-06-13T00:00:00-07:00

Update, June 21 2009:

The latest update of the FlipShare software (version 4.5) supports moving your library natively. Get the latest software, then go to “Edit > Preferences > Library”.

Older method, pre-FlipShare 4.5:

My friend gave my wife and I a Flip Mino HD video camcorder as a gift when we had our first child. So far it’s been quite wonderful. It’s simple to use and we share lots of videos with our family that live far away.

Today I was going through a backlog of videos stored on the Flip camcorder and saving them to my computer. The problem was my hard drive was running low on space.

I figured I could just change a preference in the FlipShare software to point it to a different hard drive where I had lots of space, but no such luck. It turns out that FlipShare is hard-coded to place your videos in a particular directory. I suppose this is to reduce the likelihood of accidentally losing of your videos. But it also raises the question of what to do with your Flip videos when your hard drive starts to run out of space.

It turns out it is possible to move your Flip videos to a new drive. It’s not simple, but here’s how I did it.

Note that I’m using FlipShare version 4.1.2 on Windows Vista. You can upgrade Flipshare here.

Disclaimer:

This requires a bit of hacking. If it doesn’t work for you there’s a chance of data loss. Before doing any of this I recommend you backup your Videos folder, especially the “FlipShare Data” subdirectory.

Moving your default “Videos” folder

FlipShare is hard-coded to store all of your videos in the default “Videos” folder in Windows Vista. You can change the location of this folder.

Close FlipShare if it’s open.
Right-click on your “Videos” folder in Windows (my original path was “C:\Users\Tom\Videos”)
Click “Properties”
Click the “Location” tab
Enter the full path of the new location (my new location was “K:\Videos”)

Vista will create the folder in the new location if it doesn’t exist. It will also ask you if you want to copy all the subdirectories of the “Videos” folder to the new location. You do.

There’s a more detailed description of how to change default folder paths in Vista here.

I figured I was all done at this point. I’d launch the FlipShare software, it would see the new path to the “Videos” folder, and I’d be all set.

But it didn’t work out that way. The folders I had previously within the FlipShare application showed up, and the video counts were correct. But none of the thumbnails showed and I got strange error messages when I tried to click on any of the videos.

As it turns out, FlipShare uses an internal database to store the paths to your videos. Even though it could now save videos to the new location on my K: drive, it couldn’t find any of my previous videos because it’s internal database was still pointing to the old “C:\Users\Tom\Videos” location instead of the new “K:\Videos” location.

The good news is that FlipShare uses the open-source SQLite database to store all of it’s data. That means it can be read (and updated), using a SQLite database editor.

I’m not that familiar with SQLite tools, but a quick Google search yielded SQLite Admin. SQLite Admin is an Adobe Air app that makes it easy to view and modify a SQLite database. Download it here.

Updating Your FlipShare Database

The FlipShare database is located in your “Videos” folder, in the “FlipShare Data” subdirectory. In my case the new path to the database file was “K:\Videos\FlipShare Data\flipshare.db”.

At this point I highly recommend backing up flipshare.db

Close FlipShare if it’s open
Open SQLite Admin and open your “flipshare.db” database.
In SQLite Admin, press the “Open Query” button.
Enter the following query as a test to make sure things work the way you expect

	select replace (uri, "[old Videos path]", 
		"[new Videos path]") from MediaElementSource

Replace the items in brackets with the values appropriate to your system. After you’ve typed in the query, hit “Execute Query” to run it.

In my case the actual query was:

select replace (uri, "C:\Users\Tom\Videos",
    "K:\Videos") from MediaElementSource

You’ll see output similar to this:

Take a look at the paths in your results. Make sure that the paths that are shown are the actual paths to files in your new location. You can open up a Windows Explorer window to verify.

Nothing has been modified yet. The query above was just a sanity-check. To modify the paths, enter the following query (with your data substituted for the brackets) and hit “Execute Query”:

update MediaElementSource
    set uri=replace (uri, "[old Videos path]", "[new Videos path]")

In my case, the actual query was:

update MediaElementSource
    set uri=replace (uri, "C:\Users\Tom\Videos", "K:\Videos")

You’ll now see something like this in your SQLite Admin tool:

You’re almost there but not done yet. FlipShare also stores the paths of your video thumbnails it it’s database. If you don’t update the thumbnail paths to point to the new location, FlipShare will simply crash when you try to open it.

To update the video thumbnail paths we’re going to proceed similar to above. First enter this query and hit “Execute Query” to preview the results of the update:

select replace(PreviewImagePath,
    "[old Videos path]", "[new Videos path]") from MediaElement

In my case, the actual query was:

select replace(PreviewImagePath,
    "C:\Users\Tom\Videos", "K:\Videos") from MediaElement

Here’s what it looks like:

Again, check to make sure that the paths shown point to actual locations in Windows.

To go ahead and make the changes, enter this query and hit “Execute Query”:

update MediaElement set PreviewImagePath =
    replace(PreviewImagePath, "[old Videos path]", "[new Videos path]")

My query was:

update MediaElement set PreviewImagePath =
    replace(PreviewImagePath, "C:\Users\Tom\Videos", "K:\Videos")

And here’s what it looks like in SQLite Admin:

Now close SQLite Admin. Open up FlipShare. All of your videos and thumbnails should be intact, now referenced in their new locations. New videos that you copy over from your Flip camcorder will also be placed in the new location.

Enjoy and keep making videos!

Pastebins

2009-06-01T00:00:00-07:00

I came across my first pastebin while following the SpamAssassin mailing list. Users of the SpamAssassin mailing list face an interesting problem. They frequently need to share examples of spam with one another in order to help debug SpamAssassin. But the SpamAssassin mailing list is hosted on the same servers that run all of Apache’s mailing lists. Hence the SpamAssassin mailing list is filtered by SpamAssassin itself.

The consequences of this are logical, though ironic. If you want to share the contents of a spam message with the SpamAssassin mailing list, it will most likely get blocked by SpamAssassin. Doh!

So what recourse does someone investigating a spam message have? Answer: post the spam message to a web site, and send the link to the posted message content to the mailing list.

This is where a pastebin comes in. A pastebin is a web site that allows you to publish a snippet of code, a server log, a spam message, or any arbitrary piece of text. You can then share the link with people so they can review it.

The utility of a pastebin extends far beyond my SpamAssassin example. They are frequently used by developers on IRC channels. Pasting a 60 line code snippet into a chat window and telling someone “Hey look at line 32 for me” does not work. Lines wrap in weird ways, monospace fonts are not guaranteed, and line numbers are not preserved.

Similarly, if you want help with an HTML or Javascript issue, you’ll have a hard time pasting your code snippet directly into a web site. Most sites, for security reasons, will strip some or all of the HTML tags, and will almost certainly not let you paste Javascript that might execute when it’s rendered on the web page.

Here’s a list of a few of the pastebins out there and what they can do:

pastebin.com

This is one of the older pastebins and was the first one that I was introduced to via the SpamAssassin mailing lists. It allows you to do several cool things:

Choose a syntax highlighting mode for your paste. You can also use a shortcut to automatically select a syntax highlighting mode. If you know you’re going to paste some ruby code, go to ruby.pastebin.com and the ruby syntax highlighting option will be set for you by default.
Create a pastebin just for your community. For example, if you run a software company called “Flying Squirrel”, setup flyingsquirrel.pastebin.com for all your pastes. Note that these subdomains are not password protected so they can be viewed by anyone who guesses the URL.
Highlight certain lines in your paste by prefixing them with ‘@@’.
Choose the retention time for your paste: 1 day, 1 month, or forever.
Work collaboratively. For example, you can create a follow-up to someone’s paste. Your follow-up will be linked to the original, and you can generate a diff.
Pastebin.com provides it’s source code under the GPL

Here’s an example of a paste in pastebin. This is a snippet of Python code that generates awful poetry. It’s derived from an exercise in Programming in Python 3.

http://pastebin.com/f71708aec

I created a follow-up to this paste where I added a comment line and highlighted it:

http://pastebin.com/f4feb6a1c

Pastebin provides a diff between the two pastes here:

http://pastebin.com/pastebin.php?diff=f4feb6a1c

gist.github.com

Github now provides a pastebin. They are called gists. There are a few nice options:

Multi-language syntax highlighting. You can also specify a filename and the language will be auto-detected based on the naming convention.
Support for private gists.
Quick access to the raw version of the code snippet in a browser, so you can copy code without line numbers in it.
Git integration so you can clone or fork the gist.
Embed support (the embed times out so I’m not going to demonstrate it here).

Here’s a sample gist: http://gist.github.com/120863

snipt.org

snipt.org has a lot of the features of the previously mentioned pastebins, including syntax highlighting, embedding, access to the raw text, and private snipts. It also offers a few unique options:

The ability to choose a color theme (with great names like “Glitter Bomb” and “Black Pearl”)
Short urls, with one-click options to post to Twitter or save to del.icio.us or Google Bookmarks.
A rating system for public snipts.
Customization options for the snipt viewer such as increasing and decreasing the font size as well as changing the theme.
An option to turn off line-numbering.

Here’s my awful poetry python script with the “Glitter Bomb” theme:

http://snipt.org/kqj

SyntaxHighlighter

This is not a pastebin, but offers syntax highlighting and line numbering of code snippets that you include on sites that you host. It’s written in Javascript and offers support for multiple syntaxes. As of version 2.0 it also provides theming support via CSS.

There are SyntaxHighlighter plugins for popular content management systems like Wordpress, Drupal and MediaWiki. I came across an example SyntaxHighlighter in a great post on human-readable sorted du output.

There’s a lot more pastebins out there. See the Wikipedia entry on pastebins for more detail. If you’re thinking of starting your own, pay attention to the ”Abuse and spam” section on the Wikipedia page.

Weekly Links : May 29

2009-05-29T00:00:00-07:00

Found this week: “fail” is the new schadenfreude, malwebowent trolls, avoiding burnout, enjoying the silence, the paradox of cancer in whales.

Why is everyone saying "fail" all of a sudden?Christopher Beam, slate.com

“Why has fail become so popular? It may simply be that people are thrilled to finally have a way to express their schadenfreude out loud. Schadenfreude, after all, is what you feel when someone else executes a fail. But the fail meme also changes our experience of schadenfreude. What was once a quiet pleasure-taking is now a publicand competitive sport.”

The Trolls Among UsMattathias Schwartz, nytimes.com

“Two female Yale Law School students have filed a suit against pseudonymous users who posted violent fantasies about them on AutoAdmit, a college-admissions message board. In China, anonymous nationalists are posting death threats against pro-Tibet activists, along with their names and home addresses. Technology, apparently, does more than harness the wisdom of the crowd. It can intensify its hatred as well.”

BurnoutScott Boms, alistapart.com

“Ultimately, burnout results from a lack of equilibrium. When you lose your balance, physically, you fall over. Burnout is very similar, except that once you’re down, it can be a real challenge to get back up.”

In Praise of SilenceLane Wallace, theatlantic.com

“But what concerns me as much or more about incessant connection through Twitter, texting, Facebook, Crackberrys, and yes, even 24/7 instant news … is that all those technologies enhance an already bad inclination humans (and especially Americans) have. And that is: an overweening desire to be distracted from being alone in silence … or having to come to terms with whatever we might find there, if we slowed down enough to let it catch us.”

Why don't all whales have cancer?John Nagy et al., oxfordjournals.com

“Larger organisms have more potentially carcinogenic cells, tend to live longer and require more ontogenic cell divisions. Therefore, intuitively one might expect cancer incidence to scale with body size. Evidence from mammals, however, suggests that the cancer risk does not correlate with body size. This observation defines “Peto’s paradox.” Here, we propose a novel hypothesis to resolve Peto’s paradox. We suggest that malignant tumors are disadvantaged in larger hosts.”

Infant Growth Charts in WolframAlpha

2009-05-24T00:00:00-07:00

A bit over a month ago I was searching for infant growth charts because I was concerned my daughter wasn’t gaining weight sufficiently. At our previous check-up our pediatrician told us she had dropped 1 standard deviation on the standard growth curves since the last visit.

It was nothing to be overly concerned about. Weight gain rates fluctuate in the early months of an infant’s life. Nonetheless, as a new father I was anxious to see how well she was doing a few weeks later when my wife had her weighed at a local group for new mothers.

I used Google to search for growth charts. It returned a bunch of interactive tools, many of which did not work well or were difficult to interpret. The PDF files on the CDC site were useful, but excruciating to find.

Flash-forward several weeks later. My daughter’s weight is doing fine; our pediatrician had already verified that at our subsequent visit. But I was fooling around with WolframAlpha, trying to find some interesting data beyond what percentage of the population is named "Tom". On a whim I typed in ”infant growth” and found exactly what I had been looking for a month back: a concise and easy-to-interpret answer to the question of what percentile my daughter was in based on her age and weight according to the standard growth curve.

WolframAlpha defaulted the data to the set appropriate for a 12-month old, but it was easy to change via a simple input box at the top of the page. And while the small version of the growth plots were a bit tricky to read, the distribution charts where nice and clear. The best part was the accompanying data tables that showed the median weight, as well as the range one and two standard deviations out.

Infant Growth Distribution

WolframAlpha has received a lot of flak for not being useful outside of a small realm of users. A lot of that is justified - for now. When used with the expectation of a conventional search engine, you’re much more likely to find irrelevant or no results than anything useful. But if you want to ask a somewhat arcane, data-driven question - like how many seconds old you are - WolframAlpha does produce results.

Books Worth Reading

2009-05-01T00:00:00-07:00

Managing Humans: Biting and Humorous Tales of a Software Engineering Manager

The management section of software bookshelves typically has a bunch of books explaining various methodologies like Agile and XP. They’re typically written by consultants and tend to be a little heavy on process and dogma.

“Managing Humans” is the total opposite. It’s a humorous collection of anecdotes about the people and personalities that inhabit the software development world in both start-ups and larger corporations. And you don’t need to be a manager to enjoy it. For some samples, check out some articles from the author's blog; 1.0 and The Nerd Handbook are both fun reads.

Rainbows End

I’m not an erudite science fiction reader, but this 2007 Hugo Award winner is excellent. It’s author, Vernor Vinge, helped define the notion of the Singularity and Rainbows End provides an entertaining look at what that slippery concept might look like twenty years from now.

Weekly Links : Apr 15

2009-04-15T00:00:00-07:00

I’m tired of reading startup gossip, news about Twitter, and news about companies that are building stuff on top of Twitter. So I had to stop.

I recently set my home page to Hacker News, which unearthed some great content (new and old) and surprised me with refreshing community guidelines.

Here are a couple of links recently featured on HN. These days I think a lot about the challenges of startups, strategy shifts, and revenue generation. Most of these links touch on those topics.

Validated learning about customers

Nice post on measuring success via what you learn about your customers in each iteration. Also read Eric’s post about running out of iterations. Assume you’ll get a lot of things wrong before you discover what's right. Optimize your funding and spending to maximize your ability to adapt and make major strategy shifts.

The Development Abstraction Layer

An insightful article by Joel Spolsky about why great developers often start unsuccessful companies.

Solve the problem, not just a problem

It’s easy to take a technology and build a bunch of stuff that might be cool…someday. These guys did the opposite and focused only on what they needed to solve their problem. They found a neat use of git in the process.

Getting Attention Is a Stupid Business Plan

If you view the number of people following your business on Twitter as an important metric of success, then read this article. Multiple times.

Caring for your Introvert

This has nothing to do with startups, but as an introvert I enjoyed this a lot. Nicely distinguishes introversion from shyness and misanthropy:

“Leave an extrovert alone for two minutes and he will reach for his cell phone. In contrast, after an hour or two of being socially ‘on,’ we introverts need to turn off and recharge.”

Enticing User Engagement via Achievement Indicators

2008-12-10T00:00:00-08:00

One of the best ways to make your site “sticky” is to encourage users to use multiple features.

Encouraging users to engage with more features increases the amount of data they share with you. The more data they share with you, the more you succeed in creating a switching cost that serves as an impediment to abandoning your service for one that may offer a similar set of features.

The switching cost exists because in order to start using a site offered by one of your competitors, your user now has to take the time to set up their data on the new site, and often learn a new user interface in the process. This is not something users are very likely to do unless your competitor offers much more value.

Mint and LinkedIn both do a good job of enticing you to share more data with them. Mint asks “How Minty are you?” and answers that question with a personalized score - the highest being 100% - based on how much data about your financial life you share with them. They do this via a simple user interface shown below.

LinkedIn provides a similar user interface that gives you a score depicting how “complete” your profile is. They encourage you to add more data about yourself and solicit recommendations from others in order to achieve a higher measure of “completeness”.

Both of these user interface devices are “achievement indicators” that stimulate feature engagement by appealing to a user’s ego. As a person with a competitive, Type-A personality, I feel a sense of achievement in seeing that progress bar climb toward 100%. It’s not very rational, but it is very real. In striving for this feeling of achievement, I increase my commitment to the service by providing it with more personal data.

So do you want to get users to engage with more features on your site and create a switching cost? Give them a user interface that appeals to their ego by rewarding them with an incremental sense of achievement for each piece of data they give you.

Five Holiday Gifts to Avoid In 2008

2008-12-01T00:00:00-08:00

There are a lot of gift guides out there to help you determine what to buy this holiday season. The Ars Technica staff released their holiday gift guide last week, the New York Times offer their Critics Picks for the season, and Amazon.com will be hawking gifts of all types for the next three weeks.

Suggestions abound about what to give. But what if you need some tips on what gifts to avoid? Here are my top five holiday gifts not to give this year:

1) “I Love Wall Street” T-Shirt

Unless you’re shopping for someone who likes “ironic hipster” t-shirts, this one probably isn't a good idea. Your day-trading sister-in-law might not have done so well this year, and even if she did wearing this t-shirt in public might increase her chances of being physically assaulted, particularly in a US auto manufacturing city like Detroit. Put it on the shelf and hope the market does better next year.

2) The ManGroomer Electric Back Hair Shaver

Personal grooming kits, including ear and nose hair trimmers, have become de rigeur for the aging male who wants to keep things neat and tidy. And they increasingly serve as fun stocking-stuffers during the holidays.

A personal back hair trimmer, however, might be a little harsh as a holiday gift. While the utility of this item can not be questioned, its social repercussions could be devastating.

Surprising your gift recipient with his electric back hair trimmer in front of family and friends may expose more about the extent of his hirsuteness than anyone cared to know. And it will inevitably lead to calls for a live, on-the-spot demonstration that puts your trimmer recipient in a very awkward position.

If a loved-one you know needs this, slip it under their pillow discreetly and let them believe it was delivered by the back-hair fairy instead of giving it to them in front of people enjoying egg nog by the fire.

3) A Cat Stroller

Don’t get me wrong, I love cats. I have two of my own and can share a lot of weird stories about my adventures in feral cat rescue that stop cocktail party conversations dead in their tracks.

Let it be known that my affection for my cats does not imply I want to parade them around town in a cat stroller. Aside from the cat going berzerk due to overstimulation, “crazy cat guy” is a nickname that is incredibly difficult to shed.

4) A Snuggie Wearable Blanket

Nothing says “I belong to a cult” more than a wearable blanket. It’s especially creepy when worn by multiple people in the same room. Plus it comes with a free booklight, presumably so you can read and re-read your favorite mantra over and over and over and over and over…

If you want to give someone a blanket, go old-school and knit them a nice afghan like grandma used to do.

5) Motor oil themed shampoo

According to RoadKing magazine, this shampoo comes in a “one-quart motor oil jug”, is “maplewood and tobacco oil scented and doubles as a body wash.”

Emphasizing the “body wash” aspect, anythingsupermall.com notes that “it can be used from the top of your head to the bottom of your piston.”

I don’t think I have anything else to say about that.

iPhone Screen Grabs and Other Tricks

2008-11-25T00:00:00-08:00

Here’s 4 iPhone tricks that hardcore iPhone users probably know about but they took me a little while to figure out:

Taking a Screen Grab of Your Current iPhone Screen

You may know how to take a screen grab (or “screenshot”) on your PC or Mac, but did you know you can do it on your iPhone? I had no idea this was possible until my friend Dave Cascino showed me at lunch one day. Just simultaneously hit the round button at the bottom of the iPhone and the small silver button at the top of the iPhone simultaneously. You’ll see the screen flash white for a second and your screen grab will be available in the “Photos” section of your phone.

I wish I knew about this feature when creating instructions on how to configure the iPhone to connect to Zenbe for our email users.

Controlling the iPod When Your Phone is Locked

Your iPhone probably locks itself after several minutes of not using it and you need to use the “slide to unlock” feature to get to your home screen. This is kind of annoying if you’re listening to music and want to pause or advance the track (and don’t use headphones with iPod controls built in).

No problem. When the phone is locked, just double click the round button at the bottom of the phone. A little menu will appear at the top. It shows the current time, as well as controls for pausing, playing, adjusting the volume, and moving forward or backward a track. Here’s a screen grab of what it looks like with an audiobook I was just listening to (note I used the screen grab technique described above to get this image):

Jumping to the Top of the Screen in Safari

A lot of people know about this shortcut now, but I figured I’d throw it in anyway. If you’re browsing a long web page in Safari and are close to the bottom of the screen, just tap on the menu bar (where the time and battery indicator is displayed) and Safari will auto-scroll back up to the top.

Many other third-party applications that scroll now support this feature so try it out.

Getting Back to Your Home Screen

If you’re a hardcore iPhone user you may have several pages of applications you’ve downloaded or bought. As of the 2.2 firmware update, you can now click on the round button at the bottom of the iPhone and you’ll jump back to the very first page of applications.

The Jonestown Enigma

2008-11-22T00:00:00-08:00

I picked up a PBS DVD about Jonestown and The People’s Temple from the library last week. I had no idea that November 18th marked the 30th anniversary of the tragedy where over 900 people died. Though I knew Jonestown was a notoriously significant event in the cultural history of United States, I knew little of the details. It happened about 4 months after I was born.

After watching the documentary I was surprised how little mention there was of the 30th anniversary, although I later found some articles on CNN’s web site and by chance recorded a 2 hour television special with Jonestown survivors interviewed Soledad O’Brien (which I haven’t watched yet).

Most of the dead associated with Jonestown People’s Temple members lead by Jim Jones into what he termed “revolutionary suicide.” Other deaths included journalists and California Congressmen Leo Ryan, who were murdered after visiting the People’s Temple compound in Guyana to investigate and subsequently help a few defectors escape the compound. Ironically, Congressman Ryan considered Jonestown “a beautiful place” despite the fact that several members implored his help to escape. This video footage of his visit underscores the enigma of the entire situation.

According to Wikipedia, the “events in Jonestown constituted the greatest single losses of American civilian life in a non-natural disaster until the events of September 11th.”

The deaths at Jonestown are commonly referred to as the “Jonestown Massacre”, which I think is more appropriate than labeling it a mass suicide, although the semantics are admittedly complex given the opaqueness of the details. While most members “voluntarily” ingested the cyanide-infused concoction that lead to their demise, they were surrounded by armed security guards prior to doing so. Most disturbingly, the PBS documentary replays audio of the initiation of the mass deaths that features Jones hysterically encouraging the adults to begin by poisoning their own children.

The phrase "Drinking the Kool-Aid" is derived from incidents at Jonestown in reference to one of the ingredients of the poison concoction, though the Kool-Aid brand itself was likely not involved. I used to find that phrase somewhat amusing when it was used flippantly in business meetings or other contexts, but now it’s just disturbing.

For anyone not familiar with the rise and fall of Jim Jones and the People’s Temple, I highly recommend checking out the DVD below. It’s highly fascinating and profoundly sad, but it’s also a uniquely compelling narrative in that it relies on former members of the People’s Temple and Jonestown survivors to tell the story of what happened.

Buyer's Remorse? Maybe not...

2008-11-21T00:00:00-08:00

“The gallows humor of this time was brought home to me by a reader who told me he no longer felt badly about having stopped renting and instead buying a home in 2006. Sure the house has lost value, he said, but if he had not bought it, he would have left his money in the stock market and be even worse off now.”

Floyd Norris in the New York Times

Last night I was reviewing a few of the decisions I’ve made over the past two years in light of the latest bad economic news. Within that time I bought a condo in the NYC metro area and left a job at a successful hedge fund to co-found a start-up. My wife and I are also about to have a baby, which puts me in a much more contemplative mood than usual.

Over two years ago I remember our mortgage broker saying to my wife and I, “Wow, you guys are being really conservative. You could qualify for a much larger mortgage.” I don’t think he was trying to up-sell us - we were already working with him on closing on the condo we had chosen. He was surprised we didn’t want to buy a bigger place, but commended us on what I suppose he considered financial prudence. I was shocked because I was looking at what I thought were pretty big numbers on the papers he was preparing for us.

At the time I was still working at the hedge fund and my wife had a job as a Quality Assurance manager at a web site in New York. I didn’t have a good feeling about the housing market - in fact just a few months earlier I had made some slightly profitable trades in the stock market on the expectation that home builders were going see lower earnings.

But, a lot of our friends were buying homes and my wife and I were fortunate to have profited a little from employee stock options we converted and sold after leaving the company we used to work at together. That money was sitting in the stock market. I remember my accountant encouraged me to look at buying a home from the perspective of re-allocating of some of our investments. Take some money out of the stock market and put it into your own home. Your home is an asset, it’s value will likely increase, you’ll build equity and diversify your investments.

My wife and I knew we would be in our home for at least 5 years, hopefully enough time to ride out a likely dip in the housing market. We calculated what our monthly payment would be given our likely mortgage payment, taxes, and maintenance fees. We found a place we liked, obtained a 30 year fixed-rate mortgage and wrote the biggest checks of our lives at our closing.

The monthly mortgage payment and related expenses ended up being about twice what we were paying in monthly rent at our previous place. That felt like a steep increase, but we still had a good cushion given our monthly income. It’s a good thing too. A few months later my wife quit her job and took a steep pay cut to start her own local business. A few months after that I quit my job and took a steep pay cut to co-found a start-up.

So last night I found myself thinking “I wish we hadn’t bought that condo. We’d have so much more cash on hand to face what might be coming down the road with the economy.” But when I woke up this morning and read the quote at the beginning of this blog post I thought a little differently.

The money that we invested in our home in 2006 had not been cash stuffed between the springs of our mattress. It was in the stock market. And since we moved that money out of the stock market and into our home, the S&P has fallen 42%, the Nasdaq 44%, and the Dow 33%. My guess, based on recent home sales, is that the value of our home has fallen about 8%.

Given those numbers, taking money out of the stock market and putting it into a home over the last two years was not a terrible move (unless you need immediate liquidity in which case selling your home is a little harder than selling your stocks). Though I anticipated the housing bubble would eventually burst (as everyone had been predicting for at least a year in 2006) and our home value might go down a little, I certainly did not anticipate the enormity of the credit crisis and the intense impact it would have on the stock market.

As for changing careers, that didn’t turn out so bad either. A few weeks after my wife quit her job the company laid off two-thirds of their employees. As for me, the hedge fund I worked at is still in business and presumably doing well (assuming they are benefiting from the current volatility).

So financially I still can’t say changing careers was a good choice, but it wasn’t a financially motivated decision. When you leave an established company to do the start-up thing, it’s about the experience more than the compensation. And the experience has been extremely good, regardless of what happens to the start-up.