The Sect of Rama

By way of Levi, this TED talk, Blaise Aguera y Arcas: Jaw-dropping Photosynth demo:



Very slick stuff. I hate to be curmudgeonly, but my first thought was wondering how technology advancements like this will erode privacy. Combine:

1. the backlash to Google's map street view

2. the fact that you can find some of my usenet posts from the early 90s, long before people were thinking in terms of search engines and data permanence (or at least long before *I* was!)

And you get data mining of people's photos (or other multimedia) in ways they didn't think would be possible, and potentially wouldn't have shared would they have known. Moore's Law is a double-edged sword...

I didn't start blogging at Sun until July 2004, so my 'blog anniversary' isn't for another few months. But on this day three years ago, our group launched the blogs.sun.com site- the success of which is partly responsible for the formation of my team, Community Software (which is really an eclectic mass of services in its current state: ad server, mediacast, wikis*, blogs, forums, search, planets*, developers, bugs. (* = work in progress))

I don't have anything new to add to what others have said already (1, 2, 3). So here's to another 3 years of pioneering (as far as big companies go) community software at Sun!

After tagging several people during the last meme, I suppose its only fair I get tagged on this round [1, 2]. Here's the deal-- I'll do my best to answer, but then I'M OUT! No more tagging for me! We're even-steven, blasted blogosphere...

Name five songs (or artists) that you love but annoy your significant other.

Geez, thats too easy. A more challenging question would have been to name five that don't annoy her. Ah well- here are the first five artists/genres that come to mind:

1. Mos Def.

2. DJ* (DJ Danger Mouse, DJ Shadow, DJ anyone else).

3. Simon and Garfunkel.

4. A Tribe Called Quest.

5. Radiohead.

Not too long ago, Matthew turned me onto The BileBlog (probably as I was bitterly whining about some horrible technology we had to deal with, and he found it apropos to mention.) I added the feed to my reader and quickly forgot about it. Today, there was a new entry- and after my initial "wtf is the bileblog?" reaction and scanning the entry, I realized I largely agreed with what he had to say. And the entries are funny 'cause they're true! (Well 4 for 4 random entries of my choosing were, I haven't yet delved deeper than that.)

My only gripe- of which I realize I'm guilty of myself, and should soon fix- is that I couldn't obviously find a link to a "who am I?" page. Someone with such potent opinions, it seems, should be qualified with such a page.

I'm trying to get phpMyAdmin installed on one of my dev environments. I've never had trouble before, but something is amiss on this box. Its running:

Apache/2.0.59 (Unix) DAV/2 PHP/5.2.0 Server at large.sfbay Port 80
mysql Ver 12.22 Distrib 4.0.24, for pc-solaris2.10 (i386)
PHP 5.2.0 (cli) (built: Nov 4 2006 07:17:22)

and a test script printing phpinfo() reports all the right things. However, when I try to pull up the phpMyAdmin index page, I get errors suggesting its not groking my include_path:

[Wed Mar 07 09:29:31 2007] [error] [client 129.146.110.129] PHP Warning: require_once(./libraries/common.lib.php) [function.require-once]: failed to open stream: No such file or directory in /export/home/htdocs/phpMyAdmin-2.10.0.2-all-languages/index.php on line 36
[Wed Mar 07 09:29:31 2007] [error] [client 129.146.110.129] PHP Fatal error: require_once() [function.require]: Failed opening required './libraries/common.lib.php' (include_path='.:/usr/local/php/lib/php') in /export/home/htdocs/phpMyAdmin-2.10.0.2-all-languages/index.php on line 36

Even though the error itself says my include_path includes the current working directory.

Whats up with that? I hacked around it for the time being, by making all paths absolute:

find . -type f | xargs perl -pi -e 's=\./libraries/=/export/home/htdocs/phpMyAdmin/libraries/=g'
find . -type f | xargs perl -pi -e 's=\./lang/=/export/home/htdocs/phpMyAdmin/lang/=g'

Ick. Perhaps I'll set some time aside to migrate to Sun Webserver 7 and see if thats cures it. Its about time I ran some benchmarks against that container anyway...

I've been an avid Newshutch user for several months now, but the QoS has slowly degraded, so I've jumped ship to Google Reader for the foreseeable future. (hot on the heels of Allen, who jumped ship too.)

Newshutch, if this gets your attention, please fix the following (in order of importance to me):

1. Re-enable prefetch. For slower sites with more than one batch of entries, I'm waiting for the next batch of entries to load once I get to the psuedo-bottom. (if your servers can't handle this, you should limit the # of users you support until you can.) Its also annoying because I used the scroll bar as a "progress meter" (percent read in this feed), and now thats gone.

2. You've got some elusive bug that showed Engadget or Consumerist feeds under my Slashdot feed ("you got your peanut butter in my chocolate!") Its been that way for a while now.

3. Delays in feed updates. I've started subscribing to a few things that are time sensitive, like Dave does with Hudson reports. But delays between feed fetches render much of that functionality moot.

4. And lastly, as a DoS to prevent me from leaving, the OPML export was invalid XML. Ampersands weren't properly escaped, and I needed to fix a bunch before I could successfully import it.

When I logged onto Amazon tonight to see WTF is holding up my copy of Lightroom (that I preordered before it even released!), I stumbled across a new-to-me tab called Rama's Amazon.com. It piqued my curiosity, so I checked it out. Same typical cross-selling based on previous purchase history junk, but I see they've also implemented a tag cloud based on- best I can tell- my previous interests. Here's what mine currently looks like (links were relative on their end, so they're all broken-- don't bother clicking...):

Action All Categories Chamber Music Classics Comedy Contemporary Digital Photography Drama Electronica Equipment Fiction History Indie Rock International Military Nonfiction Object-Oriented Design PhotoDeluxe & Photoshop PL & SQL Pop Rap Software Design Software Development Spirits Trip-Hop Web Graphics Web Site Design

Best I can tell, Amazon thinks I'm an international man of electronica, equipment, and software design. Cool.

Anyone else feel like they need to at least scan the entries in a particular feed before marking them all read? I don't know why I feel obligated to do so, and on busier weeks, it feels like a curse.

On a side note, I frequently dump the OPML from my feed reader for backup purposes and noticed an unfortunate trend that certainly doesn't help matters. Back in Nov 2006, the file was 15k, now its 22k. A 50% increase in feed subscriptions in just four months.

About a month ago I signed up for the service from 103bees, a free search engine traffic reporting tool. This morning I received this week's keyword report. Its an interesting view of whats driving eyes to this blog. I'm surprised no one's put together a widget of the top search terms for a given site, a la a tagcloud widget. Seems about equally useful. Hmmm... maybe we should try something like that with Sun's search. (Matthew?)

THE SECT OF RAMA (http://blogs.sun.com/rama)
> bypass registration adobe lightroom
> ferrari 4000 fan continuously runs
> osx86 usb logitech
> how to inflate tires with an air compressor
> mac pro vs amd
> 6 steps to make wine
> mystery spot india
> zabbix solaris 10
> macbook pro right click windows 2007
> macbook pro windows right button


THE SECT OF RAMA QUESTIONS
> how to make one pot soup
> how to inflate tires with an air compressor
> how to build a smoker with cardboard
> who is majority sect in india
> what is rama's responsibility?

I received the email below from Flickr this morning, and as part of the engineering team that runs Sun's common identity and login services, I totally understand the need to migrate over to Yahoo ID's. But I've got a request for you Flickr, if you're listening. "Badge" the Old Skool members like you would other types of members ("Pro", etc). That way us early adopters still have some sort of recognition of such. (I know you guys were thinking along these lines by calling us the Old Skool to begin with- why not preserve it?)

Dear Old Skool Account-Holding Flickr Member,

On March 15th we'll be discontinuing the old email-based
Flickr sign in system.  From that point on, everyone will
have to use a Yahoo! ID to sign in to Flickr.

We're making this change now to simplify the sign in
process in advance of several large projects launching this
year, but some Flickr features and tools already require
Yahoo! IDs for sign in -- like the mobile site at
m.flickr.com or the new Yahoo! Go program for mobiles,
available at: http://go.yahoo.com.

95% of your fellow Flickrites already use this system and
their experience is just the same as yours is now, except
they sign in on a different page.  It's easy to switch: it
takes about a minute if you already have a Yahoo! ID and
about five minutes if you don't.

You can make the switch at any time in the next few months,
from today till the 15th. (After that day, you'll be
required to merge before you continue using your account.) 
To switch, start at this page:

http://flickr.com/account/associate/

Nothing else on your account or experience of Flickr
changes: you can continue to have your FlickrMail and
notifications sent to any email address at any domain and
your screenname will remain the same.

Complete details and answers to most common questions are
available here: 

http://flickr.com/help/signin/

Thanks for your patience and understanding - and even
bigger thanks for your continued support of Flickr: if
you're reading this, you've been around for a while and
that means a lot to us!

Warmest regards,

- The Flickreenos

Flickr is rolling out a new feature called "machine tags" that allows users to be more precise in how they tag, and how they search, their photos. Cool stuff, I'm glad Flickr is supporting a "deeper" way of associating data to images other than just keywords. But it too comes back to the folksonomy vs taxonomy problem:

# Are machine tag namespaces reserved?

No. Anyone can use a namespace for anything they want.

If you are concerned about colliding namespaces you should consider adding an additional machine tag to define your namespace. For example :

dc:subject=tags
xmlns:dc=http://purl.org/dc/elements/1.1/

Like tags, in general, we expect (hope?) that the community will develop its own standards by consensus over time.

Folksonomy "standards" is an oxymoron, so is this really just a fancy way of trying to put structure around chaos? Getting people to create metadata is difficult enough, let alone creating complex metadata. I'd much rather see Flickr adopt a "tag suggestions" feature, a type-ahead to offer tag choices based on their popularity. Its the low hanging fruit of the metadata clean-up problem- big pay off (cleaner tagging) *and* low effort.

Sigh. The more I play around with sites with tagging, the more curmudgeonly I get about enforcing cleaner data. :-/

If that subject line hasn't rung a bell with you, here's the gist: daylight saving dates have changed, requiring patching for everything from smartphones, routers, databases, servers, etc *before* March.

Edgeblog has a great summary of the problem along with lots of links to various vendor fixes, including Java, Solaris, Oracle, and MySQL.

With the year almost over, I thought it'd be interesting to see what kind of damage was caused via this blog:

First entry: July 30 2004
Total weblog entries: 339 (89 entries in 2006)
Total weblog comments: 417


I wish I had record of everything like this ("how many days did I play basketball in 2006 vs 2005?")-- alas, I need better record keeping. Sounds like a good new years resolution...

Via BoingBoing, Google Patent Search. A much easier way than the USPTO search. Makes me curious to revisit Google's privacy policy-- harvesting interesting patent search phrases seems like a good way for them to build out their portfolio...

Amazon's version of a Woot-Off. If you're willing to battle it out on Thanksgiving day, you've got a chance at an Xbox 360 for $100.