It's camping time again. This time it'll be Yosemite, not the valley but just outside. We have another Korean family joining us so we had to get two lots but valley camping is too competitive to get two side-by-side because the lots are snapped up too fast.
We haven't camped there before so we are going in blind. I hope the bears there are less noisy than the ones inside the valley (you can hear them breathing and sniffing heavily as they roam about the camp at night). If you just imagined me inside my sleeping bag clutching pots and listening to bears' phone-sex me into sleep, you got the picture. If you are planning to camp at Yosemite, avoid outer lots. It's a matter of danger but more a matter of sleeping soundly.
Speaking of sleep, we are leaving in a few hours but I couldn't sleep so I spent the early morning setting up a test server. I can't wait to have nothing to do but worry about bears. ;-p
I don't have an answer to the question but thought Annie Heckenberger's comment to Max Kalehof's post was funny:
I’ve noticed a surge…and consequently a mass exodus of their core users. My 20-yr intern summed it up best: when I asked her why she deleted her facebook profile/account this week, she said, “My 6th grade cousin tried to friend me on facebook and that was it. Facebook’s over.”
Free goosebumps for morning! A bit of drama and a lot of talent, catch 6-year old Connie Talbot singing Somewhere Over The Rainbow. Her voice was perfect and her smile is just precious. I just hope she doesn't turn out like Britney Spears did.
If your freshly installed Tomcat 6.0 on Windows is failing to start with a mysterious error about javajni.c, it's because a Tomcat native library depends on msvcr71.dll file which is often missing on Win32 servers. Get a copy from another (likely developer) machine and drop it into SYSTEM32 directory. Tomcat 6.0 should now start like this blog's server just did. ;-p
BTW, following server.xml entry is what you need to enable NIO HTTP 1.1 connector for Cometd and other uses.
<Connector port="8080" protocol="org.apache.coyote.http11.Http11NioProtocol" maxThreads="150" connectionTimeout="20000" redirectPort="8443" />
While I was busy humping ActionScript, Ruby installation on my Mac laptop broke. During reinstall, I noticed that MySQL gem failed to update. After collecting tips and misleads off the net, none of which worked as is, I found a combo that worked for me:
- Edit mysql.h to defined 'ulong' as unsigned long
- Use dash-dash space dash-dash for parameter
Apparently, mysql.c in MySQL gem uses ulong which is not defined anywhere so gem install fails. If you define ulong inside mysql.c then make/install, the gem won't show up in the gem list and gem install command will wipe out any change you make, so hacking mysql.h from local MySQL installation seems to be the path of least resistance.
MySQL is installed at /usr/local/mysql on my laptop so, for step 1, I edited:
then ran (watch the Morse-codish crap):
sudo gem install mysql — –with-mysql-dir=/usr/local/mysql
Total time wasted: 2 foul-mouth hours. Some folks reported success by switching to GCC 3.3 (sudo gcc_select 3.3) but not for me. Above fix works with GCC 4.0 (sudo gcc_select 4.0).
With increasing number of broadband ISPs choking BitTorrent traffics down to irrelevant level, I fear BitTorrent is no longer an acceptable platform for new applications. What do you think?
Having watched the Jazz.net demo video, I have mixed feelings. I don't want to trivialize the tremendous amount of sweat and heart that went into the project but, to put it bluntly, it feels too…stereotypical German. By that I mean too focused on functionalities and not enough attention paid to what gives the namesake, jazz, life.
As I watched Eclipse evolve, I noticed the same happening. Proliferating views and perspectives, each making perfect sense yet, as a whole, cluttered and bewildering. Jazz.net is an extension of Eclipse. Watching the video, I had to wonder what it would feel like being a cog in the machinery of Jazz.net. Jazz.net enables global projects but at what price? Where is the life in engineering? What will engineering be like if it's measured only in metrics and graphs? Precision can be fun but visibility and control are both blessings as well as curses.
I wish the best for the Jazz.net team but I wish there was more, more space, more room for us than details. I know that I am asking for something elusive because, even in midst of 40 close guild members at Molten Core, supposedly having fun, we can feel like less than a person and more like a switch waiting to fire in time. But then maybe there is no room for us in the machinery of global economy.
My copy of RESTful Web Services, by Leonard Richardson and Sam Ruby, arrived this morning so I spent some time reading and skipping through chapters. It's a well written easy to read book that attempts to shed more light on an increasingly popular yet still foggy and unfortunately politically charged area called REST (Representational State Transfer).
Overall, I think anyone interested in REST will find 400+ pages of examples and clear explanations very helpful. As for me, most of the content wasn't new to me but I ordered the book to read Chapter 8: REST and ROA Best Practices. As expected, the chapter insightfully enumerates a broad range of design issues REST developers are likely to encounter.
Good stuff but, like a tasty appetizer, the chapter left me wishing for more, a book that focused soley on the subject covered by the chapter, maybe a series of compendium is needed to cover the growing body of knowledge as the area evolves through wider use.
As I mentioned before, I am in midst of launching a webapp under Appily label. Underlying theme of Appily is a growing collection of simple yet practical webapps. While the first app is being preped for trial launch, I wanted to start thinking about the follow up which is intended to address common teleconferencing needs. One problem I have is that I am a frequent teleconferencing user but never hosted one involving more than a handful of people.
So I would like to have in-depth discussion with folks who frequently hosts medium to large teleconferences (i.e. financial teleconferences). If you are such a person and you are interested in helping me identify common problems, please contact me.
UPDATE: If you would rather just coredump teleconferencing problems and wants, please use the comments. Note that I am not looking to replace any existing teleconferencing systems (it'll be awhile before telephones go out of style), just augment them to improve their effectiveness.
GNUCitizen points out some potential XSS vulnerabilities in Google Gears. My only take away is that developers have to be more careful and scrub everywhere, not just at the server-side. More work is good, me think. ;-p