OK Enough Database Related Teasing
I know Database technologies are hardly the most exciting subject to be teasing readers with cliff hangers about, but here is our Cunning Plan.
The Answer in one word, HBase. HBase is an Apache open source project which builds on top of Hadoop (another Apache project) to provide a clustered database engine. Its actually based on the ideas set out in a paper Google published about their proprietary DB technology called Big Table.
[Read More]
I Apologies Yet Again....
First as usual I am feel I have to apologise for the lack of posts on this blog. I am still here working hard. The IPTV project has taken a bit of a back seat whilst the company does some consulting projects for clients. Don't fear the IPTV project is still in progress its just going to be delayed by a couple of months more.
[Read More]
The Technical Approach to Management
If you can remember far enough back to when I last made a post I was talking about Google Sites and how we were using it as a sort of internal wiki. I have discovered another piece of software which has got me even more excited (if thats the right word).
Before I dive head long into explaining what I have found and what I like about it I think I ought to explain why we need it. When developing software apart form the actual tools for writing code there are a couple of types of management tools which are quite essential.
The first of these are the Version Management Systems. These basically keep track of the changes made to all the files which make up a project. So at any point you can see whats changed between two versions over a period of time and revert back if needs be. They also let multiple people work on multiple versions of the project at the same time.
The second set of tools are Issue Tracking tools. These are usually in the form of web applications, which keep track of bugs and feature requests. These often integrate with the Version Management Systems to enable you to see recent changes which likely caused the bugs and issues which are listed in the Issue Tracking tool.
All big software projects need to use these tools as its far too complicated to keep track of things without them. Of course open source projects are no different, except they make their Version Management Systems and Issue Tracking tools accessible to the public. It could be argued that Open Source projects need them more than most commercial projects, because Open Source developers tend to be so geographically separated from others in the project. Rarely can they just wonder down the hall and actually ask the person that wrote that piece of trouble some code.
If you look at most open source projects they use Subversion for version management and Trac for issue tracking, both are open source projects. Both are quite effective and serve a purpose. However Subversion in particular and Trac to a lesser extent both have flaws in their design, which can make day to day use a pain.
We have started using a version management system called Git, (silly name I know). It was created by the some of the Linux Kernel developers specifically to look after the source code for Linux Kernel. which as you can imagine consists of a very large collection of code.
I have also found a really good alternative to Trac which is also Open Source. This has built in support for lots of version management systems including Subversion and Git. It also has its own wiki system like Trac so you can keep project documentation there as well. Whats more it has lots of planning and time management tools. And the name of this wondrous tool? Redmine.
If anyone is in the market for a Trac alternative Redmine is defiantly worth a try. Git on the other hand is very good, but is lacking an easy to use GUI especially on Windows, where TortoiseSVN is so popular. Still I am happy to use the command line if it means getting Git's amazing power and features.
Google Sites has arrived
Google has launched a new service called Google Sites. It has a very similar effect as a Wiki but it has a fill wysiwyg interface which cuts out the need for learning Wiki syntax which most Wikis use. We use Google's Enterprise Applications at KISS Intelligent Systems quite a lot, so this morning I got an email informing me it was available to add to our enterprise account.
[Read More]
Are Mobiles About to Finally Get Smart ?
The last few months seem to have seen a new resurgence in the mobile phone industry with first Apple and now Google throwing their sizeable hats into the mobile arena. Its not surprising their competitors (although they will never admit it) and the press are taking note.
Apple's iPhone had been the stuff of wild rumour at the start of the year. It was announced a few months before release during which the hype had grown almost completely out of hand. Which is about the time the cracks started to show.
Its clear the iPhone is a really amazing device from a design point of view. The multi touch interface really is something new and innovative. The problem is the end user experience doesn't really live up to that. For a start the hardware is somewhat lacking, the lack of a 3g internet connection is a real deal breaker for me.
[Read More]
A New Look
Welcome to our new look website. Since we are getting ever closer to our public BETA we decided we should produce a more detailed site explaining a little more about the IPTV Platform are currently developing.
You will also notice that the Labs section has been renamed Below Deck. Apart from allowing for an unhealthy amount of nautical metaphors, I think the new name also better sums up what we want it to be. Its supposed to be a look behind the scenes where as the old name made it sound like it only contained technical content. That's not to say there won't be any technical content. AMPE our Open Source project is still part of the new Below Deck site.
Its also given us the opportunity to change the themes of the various below deck sections as well as update the Apache Roller blogging engine we are using. The old labs URLs should still work for the time being, but its best to use the new below deck ones.
The BBC's Online DRM Crusade
Some of you may have noticed the press generated by the BBC's first public release of the BBC iPlayer. On the face of it the iPlayer is a brilliant idea (if unoriginally named). Once a TV show has aired on the BBC you can view the up to 7 days later on the internet. To achieve this in sensible almost TV like quality the software downloads the shows form a private Peer to Peer network on to your computer.[Read More]
The Migration
You may have noticed some outages on our website for the last couple of days. This was caused by us moving to new servers. This puts us in a better position as we get closer to launching our public beta.
The most challenging part of the upgrade was not moving between the physical hardware, but moving to a new Operating System.
[Read More]
Its Open
We have decided to open source (LGPL) the video engine we built for our forth coming Content Distribution Platform. There's now a project page as part of the Labs site.
Just to clarify, the Content Distribution Platform its self will still be proprietary software, although at the moment the plans are to release the client software free of charge. The video engine that powers that part of the client is what we are Open Sourcing today. We feel we need more control over the Content Distribution Platform as its a service we are providing, rather than stand alone software.
Is The Writing On The Wall For DRM
Is it just me or is there a definite feeling that DRM is on its way out. Steve Job's recent essay seems to have been the final straw for many. People in the industry have been muttering about the evils of DRM and how its use is damaging to the industry for a while. There voices seem to have got lost under the constant noise the big four music labels make about the needs of content owners and how the use DRM is non-negotiable.
[Read More]
Happy New Year
Happy New Year all. At this time of the year everyone looks back at the year that's just gone. Its also the time that people make predications and resolutions for the year ahead. Well I have a predication, We will release the beta and possibly 1.0 version of our content distribution platform in 2007.
There have been a few technical hurdles we had to overcome in 2006, they weren't actually related to the content distribution technology but surrounding items, such as the user interface. These have all been resolved now and we are steaming on towards a releasable product.
In the wider technology industry this is going to be a big year. The move to multi core CPUs is now well under way and we are starting to see big jumps in processing power. The end of this month sees the launch of Windows Vista. We have been running pre release versions for testing our software for the last few months now. It is a big improvement in most areas, although you get the feeling we should have had it at least a year ago, still its hear (nearly) now. Virtualisation is getting interesting too in particular the Xen project has a lot of potential.
I am quite excited about this new year, I am not sure exactly why. I think it could be a good year both KISS Intelligent Systems and the Tech industry as a whole.
A Core Too Far ?
Will Harris from the excellent Bit–Tech.net posted an interesting article which was basically rebutting all the nay sayers about weather we need quad core CPUs. This got me thinking, most people approach this from the point of view of end users. As an end user, if I can afford more its got to be better, right? Well yeah, the counter argument is you wouldn't use the extra cores much so why waste your money.
[Read More]
1 Year Down, Many More To Go (hopefully).
Yep as of 14th of November we are one year old, on paper at least. The planning began at least a month before and much of the current code base was started soon after last Christmas. So depending on how you look at it we are about a year old now.
What's happened so far, well we have come up with lots of great ideas and written lots of code. Of course as few people have seen anything we have produced, it doesn't seem that impressive an achievement. By our second birthday that should have all changed. The beta should be well underway by then, so hopefully some of you reading this should have seen some of our work.
One thing everyone can see is our web site, and the eagle eyed amongst you will have noticed the labs page has changed a bit. This is due to an upgrade to Roller the blogging engine we use. Unfortunately the most recent update has completely changed the template system it uses. So we had to do a fair bit of work on the templates so I figured we may was well make a few cosmetic changes as well. Make it look like we have actually done some work.
What's With the Wait ?
This is a just a quick post to let everyone know what we are up to. I know I have neglected this blog for a month or two ,although hopefully I can keep it updated more now. I also intend to most more than just progress report posts, as I did in the early days of the blog.
[Read More]
Why So Quite?
The main reason is I have been on holiday in Dubai for the last 10 ten days. It’s an incredible place. Where else would you decide to build a large snow ski slope at the edge of a desert? The same place that realised there was a shortage of cost line to build five star hotels on, so they simply built more coast line. It’s certainly well worth a visit. Emirates the state airline for Dubai is also very good, even in cattle class. There should be some Dubai photos on my Fickr profile soon.
There is also another reason for the lack of activity on the web site, that’s a lack of time at present. Periodically KISS Intelligent Systems does consulting and other bespoke work for clients. This has usually comes about from contacts I had prior to founding KISS Intelligent Systems. I have spent a week out of the office with one of these clients before my holiday and the project may require me to spend more time out of the office.
This sort of work does delay our own projects a little but it helps to keep the company on a firm financial footing. I am quite wary of what happened in the dot com bust, when many new promising technology companies disappeared along with those which had the down right silly business plans we have herd so much about. Keeping the company reliant on its self during the “start up” phase seems to make sense. Who knows may be there won’t be a “Web 2.0 bust,” although I can’t see all of these Web 2.0 services surviving.
So what I am trying to say is we are still here working hard as ever (except for a well earned holiday of course). We are still working on our content distribution platform it just may take a little longer than we would prefer.