The TTYtter 2.0 beta with Streaming API support is now available for the incautious.

Blow yourself away here: ttytter 2.0.0beta

get_iplayer

Firstly, the videos I needed to shorten were from the BBC.. There’s a command line tool called get_iplayer that enables me to grab live or recorded video from the BBC site. It’s way too complex to fully describe here but the command I used to grab my video clip was:

get_iplayer http://news.bbc.co.uk/democracylive/hi/house_of_commons/default.stm --stop 1800 --get

That command retrieves the embedded live video on that particular BBC web page and saves the file as an *.avi. The --stop 1800 argument recorded 1800 seconds of video (60secs x 30mins = half an hour – I was having my tea!). Once I’d retrieved the video I had to find the start-time of my desired clip and then calculate its duration. I used ffmpeg to create a *.flv file from the original *.avi.

I found that the original downloaded file’s audio and video were out of sync, by approximately 1 second. More Googling and I found a way to encode the *.flv file and offset the audio/video. Here’s the command I used:

 ffmpeg -i downloaded.avi -vcodec copy -itsoffset 1 -i downloaded.avi -map 1:0 -map 0:1 -ss 848 -t 399 result.flv

That took the video from the downloaded file, offset by 1 second from the audio in the same file, starting at 848 seconds and lasting 399 seconds. Cool.

Copy of a Copy

Much in the way a photo-copied document starts to degrade after multiple recopying, the same goes for video. Downloading a video, shortening it and then uploading the result to YouTube produced acceptable but not very clear looking videos. I had to reduce the number of copies so the answer was to locally host the file. Well, hosting the file on Amazon S3 and making use of their content delivery network (CDN).

Google led me to the marvellous site by Mark Pilgrim, Dive into HTML 5. The video section of the site goes into great detail about how best to display video in modern browsers and a link through to another awesome page, Video for Everybody. That site outlines how to display video in older or incapable browsers; yes Internet Explorer, pay attention. The information described in those two sites is invaluable, however since I primarily use WordPress for blogs and sites, I decided to use a plugin to display either *.flv or *.m4v files.

Not as pure as Video for Everybody but nevertheless, everyone should be able to see the video.

Flowplayer

Flowplayer is a tool that allows you to display video on your site in a YouTube style enclosure, complete with controls. There are a number of WordPress Plugins that utilise Flowplayer but some of them collide with jQuery. One that doesn’t is FV WordPress Flowplayer, so that’s what I’m using here on my site.

The plugin uses short codes to embed your video files, like this:

It seems to work pretty well. You can see an embedded video here: Embedded video using flowplayer. If I wasn’t using WordPress or something similar I’d definitely use the Video for Everyone technique for video embedding.

I’ve put this entry on my blog so that I don’t forget how to do it next time. It’s fairly straightforward to do so if you ended up here from a search engine, I hope you find it useful.

Well. I’ve decided to stop using it and here’s why.

Too Many Blogs!

I keep realising (and then forgetting and then re-realising) that if I set-up too many personal sites or blogs, I spread myself out way too thinly. I’m not the most prolific of bloggers at the best of times and if I’ve got to race round updating multiple sites, I’m just not going to and indeed didn’t bother. My Sweetcron site somehow became my lazy-blog. I knew I could fire off a few tweets here and there and the blog gods would be happy because I was doing my bit. So then this site site right here becomes like a ghost town. I even started giving out my Sweetcron address as the URL of my web-site. I’m such a cheat.

What’s sort of swung it for me are a couple of reasons. The first is that the Sweetcron Author himself has stopped using his own software and cites the impersonality of lifestreaming as one of the reasons. He’s actually moved onto Posterous as his blogging platform and makes regular updates. The second reason is that there’s a now a Lifestream Plugin for WordPress. I’ve installed it and without doing any customisation it already looks pretty neat. You can see it on this page. I’ll probably try and get my own style going there when I can get round to it but like I say, it looks pretty much OK as it is.

So What Now?

So, now that the Sweetcron site has bit the dust I’ll have to write in here more. I’m still going to have more than one site running but this place here will be for personal stuff. A proper blog with the attraction of an added Lifestream. I’m going to build a second site for geeky/linuxy/webby stuff and that will be more the start of business enterprise. We’ll see what happens. I’m going to need to find another income stream anyway since The Form Analyst has now closed its doors and is becoming a free site. Which kinda sucks!

Onwards and upwards…

I found myself with the option of working out what had gone wrong (the server wouldn’t even boot properly) which I may have managed, or to bite the bullet and start again with a fresh install. I went for the latter because I really couldn’t be bothered with the former and it would enable me to build a few applications in a way I wished I’d done on the first install. Fortunately due to WP-DB-Backup and rdiff-backup I was able to rebuild and restore quite quickly. DefCon 1 at 8.30am, multi-user subscription website and server back up and running at 2.00pm. I even managed to get a temporary forum up and running so that the users had somewhere to hang out. That was on-line at 9.00am.

The new set-up has Nginx proxying to Apache. I’d wanted to do it for ages and so now Nginx serves up the static content, Apache looks after all the PHP and other swanky malarky. I also did a review of the plugins on the site and found that one was really slowing things down to a crawl. It was Social Access Control that seemed to be strangling the server. I swapped it out for Justin Tadlock’s Members plugin. Things are a lot swifter. Maybe I didn’t have Social Access set-up correctly, who knows, it’s toast now. Members takes a little bit more messing about with templates to get it running sweet, but I like the concept and it’s there to stay over at TFA.

I was thinking of writing a few posts on the way I’ve got Nginx and Apache working in tandem but all I did was Google their names with the word proxy thrown in for good measure and all the help you need is already out there.

Google is your friend.

• Somewhere to test stuff
• An Apache Web Server (I use Nginx on my own server)
• As many domains/sub-domains as I like
• 5OGB of backup space
• Lots of funky stuff, like subversion!

All for less than that ten dollars. Cool or what!

If you don’t know what subversion is:

Subversion (SVN) is a version control system initiated in 1999 by CollabNet Inc. It is used to maintain current and historical versions of files such as source code, web pages, and documentation. Its goal is to be a mostly-compatible successor to the widely used Concurrent Versions System (CVS).

So my plan is use the following workflow:

• Create/Edit files on my Desktop machine
• Save/Open them on my local server via SSH
• Sync my local server files with my repository at Dreamhost
• Sync the live site on my remote server with my repository

I already backup any sites I have or am working on via rsync or rdiff-backup to my Dreamhost backup account. This new subversion set-up will be used for current development synchronisation with the bonus of historical svn backups.

It all sounds lovely and geeky to me which is great. Trust me it’s good.

I was thinking about carrying on with the theme but the experience has served its purpose and I’m ready to move onto something else. I’ll just leave the theme as it is and perhaps I’ll come back to it someday. So, the next plan is to develop a theme that doesn’t instantly look like Thematic (or thesis for that matter).

I’ll start tomorrow.

What is SteadyOffload?

SteadyOffload is a content delivery service. Here’s how the service describes itself:

SteadyOffload is an innovative service designed to reduce the load of your web server seamlessly. It enables you to offload all kinds of static web content – images, documents, movies, presentations, software downloads, you name it. The content you offload remains stored on your server while synchronized copies are delivered from us. It all happens transparently – all you have to do is point the files to offload, the rest is done by our cachebot.

So, what’s all that about then. Well, in all honesty if you visit the SteadyOffload site they do a good job of explaining what they are all about. In a nutshell they save your server from excessive work by serving up static files held on the SteadyOffload servers. It works much in the way that Amazon S3 does but with hardly any hassle at all. You simply upload your static image, document or whatever file and it gets mirrored and then subsequently served by the SteadyOffload server. All your own server need do is then serve up the dynamic content at a fraction of the server load.

So How Does it Work?

In order for the SteadyOffload integration to start working you obviously need to sign up for an account. New signups get 5GB of bandwidth free. Once that’s expired it works on a pay as you go basis but the fees are very competitive compared to other more well known delivery services. Once you have your account up and running you simply have to put a small piece of HTML code in pages that show static content. Here’s the code:


script type="text/javascript" src="http://steadyoffload.com/serve/name of script.js">/script


Then, by using custom tags, your static content is offloaded to SteadyOffload and your server can take a breather. Here’s an example custom tag:


img srcx="/home/www/site/example.jpg" /


Note the custom scrx tag

There’s also a funky plugin for WordPress Users that takes care of all the business for you. Grab it from here: WP-Offload.

An obvious benefit of SteadyOffload is that whilst the files are served remotely, if you cancel your account, your original files are still in place and ready to be served once again by your server, which probably won’t thank you too much for all the work you’re giving it back.

Control Panel

When you have your SteadyOffload setup running sweetly, you can monitor what’s being offloaded by visiting the Control Panel. Here you can set URL Filters to ensure no-one is stealing your bandwidth, monitor offloaded files and keep tabs on your subscription.

Added 28/5/09

I found a really useful Video at Google produced by SteadyOffload themselves that shows how simple it is to use the Control Panel. It also shows something that I neglected to mention and that’s dynamic image manipulation by using the “xmanip” tag. The video is here.

And so…

I think SteadyOffload is a great service that deserves support and can provide obvious performance benefits for your site. It’s simple to set up, free to trial and cheap once you use up your free bandwidth. Once I start paying (and yes, I haven’t used up my freebies yet) then I’ll make an effort to produce some stats and figures to show just how cheapo it really is.

Give it a go!

NB. Nearly forgot, I recently Tweeted on Twitter about SteadyOffload since I stupidly misplaced my password. As far as I can tell I was the first person to tweet about it. Ever. Unreal! My Original Tweet

What I’m going to do is write out some text on a sheet of A4. I’ll take a picture of it with my phone and then email the image from my phone to a Tarpipe workflow. Tarpipe will pass the image on to my Evernote account where the text should be picked out using Optical Character Recognition. Once done the text will be returned to Tarpipe and then tweeted to Twitter.

The images below describe the process:

Inspiration for this little stunt came from this blog post

I Use Ubuntu

Before anything else, I have to point out that I am an Ubuntu user. Funnily enough not a Desktop user but I run the latest Ubuntu (Jaunty) on one of my VPSs. I also ran Hardy for a while after my main SCSI hard drive curled up it’s toes a few years back. I needed a Desktop fast and I couldn’t wait for the lengthy Gentoo compile so I installed the brown monster. I think I ran it for about 3 months before I could hear Gentoo calling and I went back.

Actually, I recall being more than surprised at how more intuitive things felt once I was back on Gentoo but that’s not what I’m writing about here.

I Recommend Ubuntu

I have this dream of setting up some form of Linux Support business where I help people move away from proprietary OSs and embrace Linux. Lisa’s Dad had a fragged and sluggish Windows XP machine and he agreed to let me replace it with Linux. I really didn’t have to think about it at all, I just installed Ubuntu. Depending on hardware, I’d probably do the same with anyone else.

So What’s This Thing About Ubuntu?

The thing is this. It seems to me that the word Ubuntu is becoming completely synonymous with the word Linux. I keep seeing Tutorials and How-Tos that purport to be “How to Install Application X on Linux”, yet when I read the how-to, it’s really meaning how to install in Ubuntu. Then we have the release of the latest whiz bang application and I read “Now available on Linux!”. I go to the download site, only Ubuntu and perhaps RPM based distros supported. Adobe AIR is a typical example. I then have to jump through hoops to get things running on my Linux and should I cry out for help it’s often the case that I get an Ubuntu styled response.

Another little irksome thing, is that Ubuntu moves the new user a little bit further away from what’s going on underneath the surface. If I’m trying to get something working and off I go Googling, it’s more than likely that I’ll find some forum conversation between two or more Ubuntu users. They’ll most likely be newbies and they’ll be wanting me to double click on things and load up the preferences manager and click this that or the other. Double what? First sign of not being able to work something out and I’m straight in a shell, fishing around.

Cutting My Nose of to Spite My Face

It’s probably right and proper to commend Ubuntu for grabbing centre stage in the Linux world and making things easy for newbies to come over to the light side. If techy mumbo-jumbo is what scared Linux wannabes away and Ubuntu has made them lose their fears then that’s OK by me. I think about just caving in and jumping on the bandwagon myself because if something new comes out, it’ll work on and be available for Ubuntu first, guaranteed.

I just like things a little more adventurous though. I don’t want to feel all cosy in the Ubuntu comfort zone and I like getting my hands dirty when things go awry. I certainly don’t want to bloody double click anything that’s for sure!

So. I can’t slag off Ubuntu, and if things go all right for me I’ll be installing an Ubuntu CD into your machine some time soon. It just won’t be mine. For now.