• ip:72.233.127.217
• ip:158.136.1.105
• ip:72.51.52.15

• The Loop: 38 requests; 38.66KB; 1.89 secs
• Daring Fireball: 23 requests; 49.82KB; 566 milliseconds
• Macworld: 130 requests; 338.32KB; 8.54 secs
• Ars Technica: 120 requests; 185.99KB; 2.08 secs
• Apple: 46 requests; 419KB; 1.39 secs
• CNN: 196 requests; 269.41KB; 4 secs
• BGR: 368 requests; 2.74MB; 35.33 secs
• AppleInsider: 141 requests; 649.39KB; 5.64 secs
• Facebook: 137 requests; 993.54KB; 11.19 secs
• MacStories: 119 requests; 2.16MB; 2.13 secs

John Gruber started this by calling out The Next Web for it’s slow performance:

• TheNextWeb: 342 requests; 6MB; no time info

More benchmarks can be seen at Browsermob.

.speechbubble:after {
    content:"";
    position:absolute;
    bottom:-15px; /* value = - border-top-width - border-bottom-width */
    left:50px; /* controls horizontal position */
    border-width:15px 15px 0; /* vary these values to change the angle of the vertex */
    border-style:solid;
    border-color:#f3961c transparent;
    /* reduce the damage in FF3.0 */
    display:block;
    width:0;
}

More examples on Nicolas’ site.

What does AirParrot do?

AirParrot lets you AirPlay your Mac’s screen to a second or third generation AppleTV. What you see on your Mac’s screen will appear on the AppleTV, wirelessly!

How do I use AirParrot?

Once you’ve opened AirParrot, click on the icon in your menu bar. Select the AirPlay device (such as your AppleTV) and then select which screen you want to mirror. To stop mirroring, simply click on the same AirPlay device in the menu!

Ugh, isn’t there a better way?

• Placeholder Text
• Autofocus Fields
• Email Addresses
• Web Addresses
• Numbers as Spinboxes
• Numbers as Sliders
• Date Pickers
• Search Boxes
• Color Pickers
• Form Validation
• Required Fields
• Further Reading

Basic Installation

First, get root and update the OS:

sudo -s
yum update

With that done, let’s get the basic packages and services installed:

yum install mysql mysql-server mysql-devel httpd httpd-devel mod_ssl php php-devel php-mysql php-gd php-dom php-pear php-json memcached svn gcc pcre-devel make

That gets us Apache HTTPD with SSL, PHP with a number of modules, Memcached, and a few system tools.

Let’s set the packaged services to start on reboot:

chkconfig --level 345 httpd on
chkconfig --level 345 mysqld on
chkconfig --level 345 memcached on
chkconfig --list

APC and the Memcahe module are both essential for performance, but they need to be installed using PECL. Simply accept the default options for everything, then execute the echo command as shown to create the ini file.

pecl install apc-beta
echo extension=apc.so > /etc/php.d/apc.ini

pecl install memcache
echo extension=memcache.so > /etc/php.d/memcache.ini

And with all that you should have a working, if not fully configured, system. Let’s start the services to take a look:

/sbin/service mysqld start
/sbin/service memcached start
/sbin/service httpd start

MySQL Configuration

MySQL should output a number of messages about configuring itself on the first startup, but all the services should start.

Now we need to create the database and user:

mysql -u root -e "CREATE DATABASE mydb CHARACTER SET utf8 COLLATE utf8_general_ci;"
mysql -u root -e "CREATE USER 'myuser'@'localhost';"
mysql -u root -e "GRANT ALL ON mydb.* TO 'myuser'@'localhost';"

And then import the database from an export:

mysql pro -u root < mydb.sql

Apache HTTPD+SSL Configuration

Installing your app is up to you, but I usually fetch it via SVN into /var/www/appname.

And now to configure HTTPD services:

In /etc/httpd/conf/httpd.conf set AllowOverride All as shown below:

<Directory />
    Options FollowSymLinks
    AllowOverride All
</Directory>

Delete the vhost definition in /etc/httpd/conf.d/ssl.conf and create a new vhost file in /etc/httpd/conf.d/vhost.conf with something like the following:

<VirtualHost *:80>
	ServerAdmin support@example.com
	ServerName falcon.pro.gostage.it
	DocumentRoot /var/www/appname
	ErrorLog logs/appname-error_log
	CustomLog logs/appname-access_log common
</VirtualHost>
<VirtualHost *:443>
	ServerAdmin support@example.com
	ServerName falcon.pro.gostage.it
	DocumentRoot /var/www/appname
	ErrorLog logs/appname-error_log
	CustomLog logs/appname-access_log common
	SSLEngine on
	SSLProtocol all -SSLv2
	SSLCipherSuite ALL:!ADH:!EXPORT:!SSLv2:RC4+RSA:+HIGH:+MEDIUM:+LOW
	SSLCertificateFile /etc/pki/tls/certs/localhost.crt
	SSLCertificateKeyFile /etc/pki/tls/private/localhost.key
</VirtualHost>

Now restart the HTTP daemon:

/sbin/service httpd restart

Scanwiches: scans of sandwiches for education and delight. Above is Parisi Bakery’s ham, swiss, tomato, lettuce, mustard, mayo, on a hero. Prints were said to have been available — I’d like the dagwood, thank you — but the store seems in a sad state.

19% of adults own a tablet computer – iPad

To clarify, that 19% does not include ebook readers (they’re tracked separately). Here’s the whole slide:

The tablet market, which is essentially an iPad market, began with the introduction of the original iPad on April 3, 2010, and has grown now to count nearly one out of every five adults in the US as a consumer (the Census Bureau tells me there are over 240 million people aged 25 and older in the US).

Rob Reid’s Copyright Math at TED2012: the claimed effect of entertainment piracy to US economy is larger than value of most of our agricultural output.

The issue is that a large portion of the research done in the US is performed by faculty paid by academic institutions and supported by public money, often grants from the NIH. A significant condition of promotion in academic careers is publication of original research in trusted journals, which is entirely reasonable to most everybody involved, except for the librarians who have to pay for the journals. The journals, you see, are really expensive, but the problem isn’t the price of the journals, it’s the fact that they pay nothing to support the research they publish.

In more direct terms, universities and governments pay to do the research, then they have to pay again to buy the journals in which that research is published. Meanwhile, the publishers collect the research for free and then sell it back to the people who did the work for very large sums of money.

This has been a problem for decades, but it’s become more significant in the past decade because technology has changed how libraries buy those journals. It used to be that a library that couldn’t afford to continue a journal subscription still kept the shelves of back issues they’d purchased previously, but electronic access is licensed, not sold. A library facing a budget crunch this year can’t cancel a subscription without giving up access to all the back issues as well. One bad budget year can empty a modern library of the resources we expect from them.

It is both a surprise and not a surprise that several years into the longest and hardest recession most living Americans had ever suffered, this issue is now getting broader public attention. Michael Wilkes, the voice of KCRW’s Second Opinion made it the point of a recent episode (mp3).

I’d been out of library circles for too long to realize this outburst was triggered by a blog post by Timothy Gowers that triggered a boycott of the largest journal publisher, but an episode (mp3) of On The Media took up this issue in some detail. Meanwhile, some of the federal agencies responsible for the largest share of research grants are making open access to raw data a condition of the grants (NSF’s policy) and the new data.gov is has been launched as a clearinghouse for currently available data.

There’s a Microsoft Store right across from the Apple Store in the Valley Fair Mall. Cliff and I realized this after exiting the Apple Store there with a new keyboard and headphones. We’d never been in an MS Store before, so we ambled over with our clean white Apple-branded accessories in hand. The Windows Phone 7 display was in the back corner, attended by a nice woman who offered to fetch a Nokia Lumia 800 phone from the back for us to inspect. I have one of the earlier WP7 phones I use for browser testing. It’s really a pretty good looking mobile OS, and the browser is competent, and in the Nokia Lumia body I have to admit it’s quite a nice package.

I said as much to the woman who’d handed me the phone and the three of us exchanged a few words on the matter. Then she pointed at our Apple-branded boxes and told us it was a problem we’d brought them in. She was happy to help us dispose of them.

I smiled and offered the phone back. Other guests wanted to play with it as well, and I didn’t want to delay them or her. I felt sad for her, though I couldn’t say it. I can’t imagine we were the first people to walk in with Apple products clutched under our arms, and I’m surprised to see them perceived as a threat worthy of such a tone-deaf comment.

I can, however, remember a couple decades ago when I saw the Windows logo that way and probably said similar things myself.

The story would just be a startling realization of how significantly the two companies’ fortunes have shifted since then, except for my time with the Windows tablets. They were on display with a keyboard and mouse attached, but I was poking at the screen trying to bring up the Windows menu when another salesperson asked if I needed help. I asked if I should expect the touch screen to work and the salesperson said I should, unless it’s been turned off. He did some fiddling, then restarted it with the explanation that that usually fixes most problems. It didn’t, so he continued fiddling and two other salespeople came over to help. Together they pointed and argued about how what might be wrong and how to fix it.

Again I felt sad and had to walk away.

Steve Jobs in 1996 explained to the Apple faithful, including me, that Apple’s success wasn’t dependent on Microsoft’s failure. Perhaps now the the MS faithful need to remember that the reverse is also true.

Legend has it that Marta Becket rolled in to Death Valley Junction in 1967 and has been performing at the Amargosa Opera House since, but tonight is her last performance. I visited in 2004 and took in the show then. It’s a certain kind of show and performer that can run 45 years non-stop (it was in its 37th year when I saw it).

Contour marketing video:

Click here to view the embedded video.

GoPro marketing video:

Click here to view the embedded video.

Drift marketing video:

Click here to view the embedded video.

Swann marketing video:

Click here to view the embedded video.

Three of the cameras compared:

Click here to view the embedded video.

This update allows users to search and explore WordPress sites using facets that represent the tags, categories and other aspects of the collection. It works with custom post types and custom taxonomies, and works for library records as well regular blog posts and pretty much anything else that can be represented in the posts table (and there are people putting business directories, classified ads, vacation rentals, job boards, real estate, and lots of other stuff in their posts tables).

Review the code now:

• Browse: http://plugins.trac.wordpress.org/browser/scriblio/trunk/
• Anonymous SVN: http://plugins.svn.wordpress.org/scriblio/trunk/

Because this is entirely rewritten it deserves all-new documentation. First, however, I need to explain that a number of features of the old Scriblio are not yet implemented in this version. In some cases this is because WordPress now offers similar features and it’s better to use those, in other cases it’s because they’re just not implemented yet.

This 2007 screenshot shows most of the public-facing search features (view it in Flickr for annotations):

Specifically:

• Display facets in widgets, clicking a term in the facet re-runs the search with that facet added
• Display the the terms of the current search and allow users edit them
• Suggest search terms based on the facets in the corpus as users type in the search box
• Render book records in HTML with cover images
• Display the availability and shelf location of the book
• Display links to Google Book Search and other resources related to the book
• Allow users to send an SMS with the book’s shelf location

And behind the scenes, Scriblio had some rather complex functionality to support those features:

• Crawl library database systems to extract book info for bulk imports, ongoing updates, and real-time status information
• Implement structured data on top of WordPress posts to represent books, archive records, and other media
• Replaced WordPress’ built-in query parsing that gets matching content from the database
• Support for using Sphinx as the keyword search engine instead of MySQL

For the purposes of this rewrite, I’ve been focusing only on the components related to searching and browsing a collection. In my 2009 diagram of the components, that’s the box with the Scriblio logo in it. Of that component, I see two features missing: live searching and user-specified sort ordering (this was implemented in 2.9, but never released publicly). I have not yet looked at the other components to determine the complexity of migrating them. Anybody, if there is anybody, who’s currently using Scriblio with those components SHOULD NOT UPGRADE.

Some things that need to be done in the short term:

• Reconnect with current Scriblio users; get an understanding of the needs and interests of the community
• Migrate the main Scriblio website to a new server and document this new version of Scriblio there
• Implement search suggestions/live searching
• Implement user-controlled result sorting (sort by relevance, a-z, set the sort field, etc)
• Review the Scriblio Schema components and make a plan for how to move those forward
• [Name your plans for Scriblio here]

The URL syntax is as follows:

• A comma (,) between terms will return posts containing either term (logical OR), like this http://maisonbisson.com/blog/post/tag/wordpress,mysql/ .
• A plus sign (+) between terms will return posts containing all terms (logical AND), like this http://maisonbisson.com/blog/post/tag/wordpress+mysql/ .

Sadly, there’s no native way to structure a URL that excludes certain terms, and I’m not really comfortable mixing term separators and logical operators.

Suggestion: Separate all terms with commas (or spaces, represented in the URL as %20), and optionally prepend each term with an operator.

Example: The following query would require both “wordpress” and “mysql” tags, but exclude posts that have the “nginx” tag, http://maisonbisson.com/blog/post/tag/+wordpress,+mysql,-nginx/ .

This syntax is similar to that many people are already familiar with when querying search engines (especially in Google, until recently). It could also probably be implemented in a way that’s backwards compatible with the current syntax.

Worth a core patch? I’m on it if anybody else thinks this is a worthwhile feature.

Update: Ticketed with patch in WordPress core trac, and the patch is live on this site now.

Update 2: The new version of Scriblio doesn’t yet support these new logical operators, but I’d love to integrate support if a feature like this is implemented in core WordPress.