Kickstarter

I’ve been watching Kickstarter from afar for a while, not particularly looking for projects to back but watching as various people/companies/sectors I’m interested in have been experimenting with it. This week I finally backed a project.

The reason I haven’t backed any projects before even if I was interested are varied, but high on the list are combinations of: they’re not something I feel that strongly about; they’re already well backed (ie, well over the goal); or I just can’t justify the cost/benefit.

Some notable Kickstarter projects that I can remember looking at before are:

  • The Ouya, which I’m still skeptical will ever survive it’s infancy
  • The first Penny Arcade “no ads on the site” project, which I didn’t feel strongly about since I almost never visit the site, I just read the RSS feed and pop in for the comics when the come up. However I think this one is when I created my Kickstarter login, so I was tempted.
  • Amanda Palmer’s project, which at the time I thought was awesome, but over-funded by the time I saw it, and I wasn’t as attached to her as I am now. If/when she does another I’ll almost certainly back it.
  • The Lizzie Bennett Diaries DVD project — I loved the series, but don’t have a huge need to watch it again (though I am following the new series), and let’s face it this was destined to have huge backing as soon as it was announced

The project that finally got me to pony up some cash this week was the Penny Arcade ‘DLC’ podcast project. (note: runs until May 31st, 2013!)

Part of the reason was definitely the “impulse buy”, and Kickstarter has lowered the bar for this as much as possible. I think when I first looked at Kickstarter you couldn’t even back projects if you didn’t live in the US, which is obviously no longer an issue, and the whole site is just super easy to use.

Another reason is that I generally trust the PA folks to deliver on what they promise, so I’m not worried about handing over money and them cutting and running, or just not delivering. They’ve proven with PAX (which I really want to go to someday) and Child’s Play charity and various other projects that they’re good people in my books, and I’m willing to front them a few $$ for something I value.

Which brings me to the main reason — like many people featured in the video on the Kickstarter, I’ve really missed the PA podcasts. They were one of my favourite podcasts ever, I really looked forward to them coming out on their previous irregular basis, I was sad when they stopped, and I’d really like to get them back. And I’m willing to express how much I value them by sending along some money.

Which I think is pretty much the whole point of Kickstarter.

Upgrading to Ubuntu 13.04 – uxlaunch is gone

Leave a reply

Late last evening I made the questionable decision to start upgrading my xbmc Ubuntu home theatre box — note: don’t start upgrading anything late on a weeknight if you actually need to sleep.

Almost everything worked exactly as expected, I’ve upgraded several other machines to Ubuntu 13.04, but there was one notable issue that I thought I’d mention for posterity.
Continue reading →

Writing more

Leave a reply

I’m once again thinking I need to start writing more here. It’s sometimes hard to find the time, and also I’m overly attached to phrasing things in just the right way, so writing (actually, mostly editing and revising) tends to take more time than it really should.

So I’m going to try and just get things to a reasonably good state and post them up, with reasonable fact checking and basic grammatical review.

Good luck to me!

Moving from AWS to Rackspace

I’ve moved my little cloud instance from Amazon AWS to Rackspace, which includes hosting of this blog. It was a simple process, but I thought I’d write a few notes about it. There are a few reasons I wanted to move, but the straw that broke the proverbial camel’s back was lack of proper IPv6 support on AWS.

It is possible to get very basic IPv6 support on AWS by putting an Elastic Load Balancer in front of one or more EC2 instances that are IPv4-only, but I wanted to have native IPv6 on the server instance for testing outbound connections as well. And really, setting up a load balancer in front of a single instance just to proxy IPv6 requests just seems dumb.

We use both Rackspace and AWS at work, and I’ve been liking the steady improvements in services and support I’ve been seeing from Rackspace in the past couple of years. I also really like that they’re openly talking about how they do things (I follow their devops blog) and participating in the building of open source tools that they use in their business, like OpenStack.

I’m also pleasantly surprised that Rackspace Cloud DNS is free. Not that AWS Route 53 is expensive, I think DNS for my few domains only costs around $4/month, but if I don’t have to pay that I’m happy not to. The Rackspace DNS is differently highly-available, using anycast name servers instead of Route 53 “brute force” distributed (separate AWS zones and separate TLDs for the name servers) — but either is fine for my purposes.

So, I’ll let you know how it goes!

IPv6 on my home network

I decided to finally setup IPv6 on my home network, for now running in parallel with IPv4. My DSL ISP (egate) has had IPv6 support for a long time, I received the info I needed from them to set it up over three years ago. I played with it a bit at the time, but had a few issues, and had no real need for it, so never got it working.

I use an old desktop running FreeBSD as my home router, currently running version 8.3-RELEASE-p7. There’s some fairly good instructions easily found with Google, and my ISP even has a sample FreeBSD configuration, but I thought I’d document what exactly I did and what remains to be done.
Continue reading →

Tiny Tiny RSS

Leave a reply

In my last post I was thinking about what to do now that Google Reader is headed for end-of-life. As a quick interim step, I installed Liferea on my Ubuntu desktop, which at least reminded me of how to export from Google Reader via OPML and prompted a bit of feed cleanup.

I found Liferea was not a good permanent solution, because it required that I be either at my desktop or at least have it running all the time, and usually I keep it suspended when I’m away from home. I needed something more like Google Reader that I could access from anywhere, but that I could run on my own server, so I started looking at Tiny Tiny RSS.

Tiny Tiny RSS is a RSS reader web app not entirely dissimilar to Google Reader. It has a PHP/MySQL backend with a nice looking, heavily JavaScript-driven frontend. It’s pretty easy to install, has a ton of features, a choice of methods for updating feeds (daemon, cron, background PHP while using the site), and even a plugin architecture. And it’s open source and you can install it on your own server — excellent. :)

The install is simple — setup a DB in either MySQL or PostgreSQL and import the appropriate schema, make sure you have some web server with PHP and the PHP dependencies installed, then drop the source files in a folder, edit the config files, and go. I have it running on my EC2 instance using Apache with FCGI/PHP and MySQL and it seems reasonably fast, though I think I may need to do some tweaking to gzip and browser caching settings for some files.

The standard UI is clean, and mostly intuitive. I had to search for a couple of options, including the OPML import — the preferences section has both tabs and top and bottom bars within each tab for separate sections of settings — but once you get used to the interface most things are in consistent locations.

If you access the app on a mobile device, it will automatically switch to a mobile UI (there’s an option to use the full UI). The mobile UI is heavily stripped down, so far that even simple functions like adding or removing a feed are removed. You can basically just read feed articles. Which is useful if you’re using a phone, I suppose, but on a tablet I found it restrictive, so most frequently opted for the standard UI. I may use the mobile UI more while commuting to reduce data and speed things up, and almost certainly if I’m using it on my phone.

So, all in all, two thumbs up for Tiny Tiny RSS, I think I’ve found my solution for replacing Google Reader. If anyone I know in person wants a login either to test or for long-term use just let me know. :)

RIP Google Reader

1 Reply

I’m sure I was as surprised as anyone else today when I got the popup in Google Reader, which I used constantly, informing of it’s impending demise on July 1st. I’ve been a user forever, and find it quite useful, but I can certainly understand why it’s not of huge value for Google.

I mean, sure, they get some data about what various feed items you’re reading and when you’re reading them for the ever-hungry Google data engine, and they can do some selling of ads alongside the feeds, but I imagine usage has been dwindling as social media sharing has spiked, and so those numbers are shrinking and becoming less relevant. The efforts of the people supporting Google Reader (there haven’t been many [any?] new features I’ve noticed recently) are better used elsewhere in the Google ecosystem.

Because most of the feeds I subscribe to are actually interesting, as opposed to just general news I can browse through, I prefer to have feed items presented more like messages in a mail inbox, where items remain new in a list until I read them. This means the more magazine-like readers like Google Currents or similar options frequently suggested are pretty useless to me. And I’m not in the mood for another webapp reader at the moment. I wanted an old-school RSS reader.

I’ve recently installed Ubuntu on my gaming desktop (which I need to do a blog post about), so I did some quick Googling and decided to go with Liferea. It looks nice, seems to work well, and it’s open source which is always good. I will need to actually backup my configuration, but it stores it’s subscription config in .opml by default which is awesome. And encrypted backups of my entire home folder is near the top of my geeky ToDo list, so I’m not too worried.

Two things that surprised me in this process — how many “RSS readers” are really just front-ends for Google Reader, which I guess means they have a fairly short shelf-life now; and how many of the blogs/sites in my RSS feed haven’t had any updates in the past months/years. Some feeds hadn’t been updated since 2008, and I’d completely forgotten I’d subscribed to them. Some (a lot actually) I’ve left in the new reader in hopes of someday being surprised if posts start up again, and others I’ve removed as my interests have changed. A good spring cleaning exercise.

Git and libcurl with older OpenSSL connecting to newer OpenSSL – error 1112

There’s an interesting edge condition we ran into at work last week around git connecting to a repo over HTTPS.

Specifically, the error “error:[number]:SSL routines:SSL23_GET_SERVER_HELLO:reason(1112)” was being returned to all clients after an upgrade of the git server.

After much Googling, including reading through a lot of comments on git and openssl and libcurl bug trackers, I came across this comment on libcurl bug 1037 that shed some light on the root cause.

Summary of problem:

When libcurl linked against OpenSSL 0.9.8* connects to Apache running mod_ssl linked against OpenSSL 1.0.*
AND
There is a mismatch between the host name in the client request and/or the Common Name (CN) in the SSL certificate and/or the ServerName in the Apache config
THEN
During the initial SSL handshake which determines SSL protocol version, OpenSSL returns a TLS Alert Warning about this mismatch
AND
libcurl interprets this as a fatal error and aborts the connection

Once figured out, the resolution was simple — make sure the CN in the SSL cert and the Apache ServerName (or a ServerAlias) match the host name being sent by the client (in this case, ServerName in Apache was set to 127.0.0.1:443).

This issue does not arise when OpenSSL 0.9.8* clients connect to an OpenSSL 0.9.8* server, or in any other combination than the affected one, which is why we didn’t see it before the upgrade.

OpenSSL 0.9.8* is still the version shipped with the current version of MacOS X. I didn’t find the bug in testing because I use a macports install of git and OpenSSL 1.0.1c, so everything worked fine for me. :(

Movies I’ve watched recently – Feb 2013, part 2

1 Reply

Continuing on from part 1, thoughts and ratings on a bunch of movies I’ve watched recently. Just realized I didn’t split these very well, and this part has way more movies… oh well.

Movies in this post: Green Lantern: Emerald Knights (2011), Salt (2010), Inception (2010), 2012 (2009), The Hunger Games (2012), Brokeback Mountain (2005), Elf (2003), Tower Heist (2011), The Recruit (2003), DOA: Dead or Alive (2006), Iron Man 2 (2010), Captain America: The First Avenger (2011), Dredd (2012)

Note: may contain spoilers.

Continue reading →

Movies I’ve watched recently – Feb 2013, part 1

1 Reply

Wow, it’s been almost a year since I posted anything here about movies. Sooo… I’ll go through what I rated things on IMDB since my last post and see what I can remember. Unfortunately, this isn’t going to include anything I’ve re-watched and didn’t re-rate, but there’s still a ton to go through, so I’ve split it into two parts.

Movies covered in this post: The Expendables (2010), The Expendables 2 (2012), Fantastic 4: Rise of the Silver Surfer (2007), Ghost Rider: Spirit of Vengeance (2011), Battleship (2012), Conan the Barbarian (2011), I Am Legend (2007), The Dark Knight (2008)

Note: may contain spoilers, and I may be fairly brief with some if they didn’t make a big impression!

Continue reading →