As long-term dedicated readers of this site may remember, one of the reasons I find technology so interesting is convergence. Whenever I save up for a new gadget, one of the first things I will do is look at what can be done with the input or output stages (or indeed the middle – often processing of data) to hook it up into a system, sometimes beyond the original design concept. I started with simple things back in the late 90s/early 2000s, like creating a Heath Robinson style security system. It involved my first digital camera acting as a webcam, some software that analysed a flow of JPGs (a raw video stream wasn’t accessible), looking for changes in the image. Luckily, the tolerance was adjustable, as the digital noise the webcam created, combined with shadows on the floor meant initially that a lot of false alerts were created. Next I had to consider resources – this system was hosted on our one and only PC at the time (to consider such a scarcity of devices now is scary 🙂 ), which spent most of the day downloading (Public Domain or CC-licensed material obviously!) from FTP servers.
The problem was the image processing hogged most of the K6-2 (or it may have been an early Duron?) CPU, meaning downloads sometime stalled, or if I attempted to use Messenger to have a video chat with someone, the system would slow to a crawl. The solution? Masking – basically telling the software that 95% of the JPG could be ignored. There was not much point in monitoring the ceiling for example, and there were only two doors to the room. So, by telling the application to only monitor areas around the gap between the double door, and the other door’s handle, our PC was suddenly a lot more responsive. If a change was observed, a few actions were triggered, a 30 second video was recorded to the hard drive (which was immediately uploaded to an FTP server, in case the PC was stolen), an e-mail was sent to an address that I set up just for this purpose, and finally a 30 second stream of JPGs were copied to the folder my HTTP server used for media. My mobile phone was set to check the e-mail account every 15 minutes, and alert me if a message had arrived. If someone entered the room when I was at work, I could at the very least check the video from the (off-site) FTP server, and if the PC hadn’t yet been stolen, I could check the still images (higher resolution and quality) via the HTTP server to check who had come into the room. There was a very rudimentary one way communication system, where I could remotely record a low-bitrate WAV file that would then playback on the PC (assuming I had remembered to leave the speakers on).
Anyway, catching up a little closer to present time, I looked at a business’ premises last year with DanS. He was intending to put a bid in to supply IT support, and had me in mind to do the actual work on the ground. We went through the different locations, making a note of the equipment in place, and the somewhat ad hoc network infrastructure. I did wince when I saw an old (non-server edition) Mac mini being envisaged as a network file server. Probably 25% plain old prejudice against anything Apple based, and 75% lack of familiarity with the network technologies involved. As we finished the tour, I asked Dan (who had visited previously) what their telephony system was based upon. I had expected to find a few normal landlines had been installed, with a single ADSL line for internet browsing and VoIP shared amongst all the offices. Instead, there were no wires outside of the premises “plumbed” in at all! It all went through a WiMax connection, with the aerial on top of the middle building. Although I had heard of a WiMax trial being put in place five or six years ago, our application was eventually in vain, as local businesses (on our street at least) had complained about potential interference. It turns out they also had some sort of system in place that could automatically route calls to various phone points in the building, or out to staff members in the city, via nominated mobile phone numbers. Virtual PBX is a technology whose label I had come across, but didn’t really understand the specifics until I spent some time reading up on the topic. I ended up editing the Wikipedia page on it a little, as the original ending to a sentence had some how been lost during an unrelated edit. Unfortunately, no one ended up with the company’s contract (they were very local, I could have even walked there during summer months, and the money would have meant I could pay off my share of Irina’s credit card bill a lot quicker), but it did at least get me to open my eyes to a few areas of technology and convergence that I hadn’t considered important/relevant before.
Although with Anna and Tim now taking up a lot of our time, I do still occasionally think of new approaches to old problems, and to start to wonder what answers lie at our finger tips, if only we would consider new uses for existing hardware and software. These are mostly thought experiments, as new gadgets cost money, and bodging old hardware with a soldering iron requires time when no little hands are likely to pull a red hot stick of metal on to themselves, and a steadier hand than I have now, especially with the increasing use of surface mounted devices (tiny little components which take up a tenth of a finger nail, instead of capacitors and resistors that hurt when you threw them at people!).
What ever happens, I’m sure that Heath Robinson style (in terms of make do and mend, rather than the Rube Goldberg pretty, but unnecessary eccentric aspect) solutions will always indicate that a proper company will soon step into the breach and supply professional solution to a problem. Until that point, raise a glass to the home based bodogers!
