Sonar 2014.1alpha3 is out
First of all, for some months now, I’ve been working alongside Jonathan Nadeau and the ([Manjaro]https://manjaro.org/) team to redesign Sonar GNU/Linux around the Manjaro distribution, which is based on Arch Linux, adding ease of use, stability and a great out-of-the-box experience to the already great and venerable Arch Linux. I am pleased to announce that the redesigned Sonar 2014.1 has reached alpha3, and now includes the long-awaited talking login screen via gdm. It can be downloaded at the Sonar GNU/Linux SourceForge page. As always, it is available for both 32-bit (i686) and 64-bit (x86_64) systems.
TalkingArch gets a new team and a new website
In addition to my work with Sonar, I am very happy to be one of the new maintainers of TalkingArch, an unofficial respin of the Arch Linux live CD/USB image that adds speech and optional braille output so that blind and visually impaired users can perform eyes-free installations of Arch Linux. Chris Brannon, who faithfully maintained the TalkingArch iso since 2008, has named Kelly Prescott and myself as his successors, as he was no longer able to consistently maintain the project. This live image will still maintain consistency with the official install media as much as possible, but should now be synchronized more often, in time with the snapshots of the official install image. I am also providing hosting for the new TalkingArch website and the new support email, as well as seeding the torrent for the iso file. Currently, we still have the iso from August 5, but we are planning to release the next build to coincide with the February Arch snapshot. 2014 will indeed be a great year for Linux, especially Arch and Arch-based distributions.
I've been playing around with SimpleWebRTC.js and the very interesting-looking Talky website, which allows anyone to create a video chat room on-the-fly by entering it in a form or putting the name of the room at the end of the URL. I also looked at the SimpleWebRTC demo, because it seems to be a simpler version of Talky that I could use to get started. Apparently, something like this is extremely easy to implement, and can be set up in as little as 5 minutes. Sadly, it seems they were exaggerating just a bit when they said it could be set up in 5 minutes.
According to the website, SimpleWebRTC.js is a very simple modular library that allows developers to get a very basic WebRTC application up and running quickly and easily. There is some very basic documentation on the front page of the site that shows just how easy it can be, and a link to fork it on GitHub. All good so far. So I figured it would be fun to put a copy on my vps and play with it there. But the HTML and JavaScript on the page for some reason didn’t work for me. No problem, I decided that just for testing purposes, I would rip the source from the demo page. After all, it works there, so it should work on my VPS with minimal modifications. No such luck. Every time I enter anything into the form that is presented on the page, it just loops back to the same page with the same form, unlike the demo, which reloads the page, attaches the room name entered from the form to the end of the URL and appears to start the WebRTC chat room.
The SimpleWebRTC.js website does mention that if I’m planning to do anything other than development or testing, I need to set up my own signaling server. It recommends using a very small JavaScript library called SignalMaster. Fine, at this point, I’m thinking that my problems are caused by the fact that my copy of simplewebrtc.js is still configured to use the development signaling server that is provided for demo/testing purposes only, so I decided to try SignalMaster. And now I find an entirely new problem. I have read just about all there is to read about SignalMaster, which only has a GitHub page and a couple of very short blog posts around the internet, none of which tell me how I’m supposed to get this thing running as a server listening on a port. I did see something that mentioned that I may need node.js, but there were no instructions on how to get the server up and running. In the SignalMaster source tree, there are only 2 json files, a server.js file and the README.md file that shows up on GitHub, which contains very little other than a very short explanation of what it does.
Conclusion
I like the idea, and I like the fact that it's open source, but it took less time to write this rather extensive blog post than it took to find out that there was no way I was going to be able to make SimpleWebRTC.js work on my server. So I guess since it took me several hours of messing with this thing only to find that something that is said to be usable within 5 minutes doesn’t work at all no matter what I try, and since there is too little documentation on how to make the signaling server run, I will have to table this project for now, and hope that the situation will improve with time.
As some of my readers may be aware, I am a huge fan of Markdown, because it makes writing much easier, and because it is human-readable. After all, people have been writing email using something very similar to ([Markdown syntax]https://daringfireball.net/projects/markdown/syntax) since before Markdown was created, so the transition is quite natural.
I never would have moved my blog to WordPress if not for the Markdown On Save Improved plugin, because I was quite accustomed to writing my posts using Markdown. A big thankyou goes out to Matt Wiebe for putting all the pieces and parts together that make this version of the plugin truly rock!
Now for the good stuff for everyone else. For the first time in it’s 4-year history, The Kyle File now supports Markdown in comments, thanks to Evan Solomon's GitHub-Flavored Markdown Comments plugin, so everyone who reads my posts can easily write comments using Markdown syntax. If you have been writing your comments in plain text or HTML, don’t panic, nothing has changed for you. This is just a third option that can make adding links, lists, headers and other nice formatting to comments much easier. Go ahead and try it out by adding a comment on this post. Have fun!
I just received an e-mail from someone who believes that just because he refuses to use Internet Explorer and Outlook Express, that the viruses, spyware and other problems that affect these applications cannot impact his system in any way. He refuses to understand that Internet Explorer is still integrated so tightly into Windows that it always runs whether you use it or not, and is still capable of picking up viruses, spyware and other exploits that give attackers unparrallelled access to a computer at little to no cost of time or energy.
Strangely, the same person thinks there is no speech synthesizer in Linux that can speak the language he needs. Funny that I found one in all of 1 minute, and wasn’t even half looking for it. Even better, the voices I found are able to be used by two different freedomware speech synthesizers, Espeak and Festival.
I have heard it said that someone who has no knowledge of how a computer works shouldn’t even sit down in front of one. I never wanted to take it that far, but the person who e-mailed me and prompted me to write this blog post is starting to make me feel that this is more true than I ever imagined.
As part of the switch to a self-hosted blog, I decided to minimize the number of third-party services I depend on to make everything work. So after a good deal of fiddling, I have been able to cut the cord that was tying me to Feedburner. WordPress has built-in RSS feeds, both for posts and for comments, and I don’t have any problems with either of them. Even better, I was able to install a plugin that allows readers to subscribe to e-mail updates from a very easy-to-use form that I put right in the sidebar. No more clicking on a link that takes you to an entirely different website to subscribe to e-mail updates. It’s also better to have that subscribe/unsubscribe form in a visible location on every page instead of just having a little link that says "Get the email sitting up near the top of the page next to the archive. Maybe now I'll get some e-mail subscribers 😀 . Either way, I really like this whole layout better than anything I had with third-party services. I only hope my readers enjoy it as much as I do.