Thursday, 30 July 2009

Lit-bots and Comp-scripts: how the world is getting faster...

I saw this while on holiday published in the International Herald Tribune I was going to blog about it then decided not to becuase it was an isolated case, well I thought it was then I read this about half way down Ben describes this:

"Professional complaints followed in May, mostly about individual chiropractors’ claims. Then, in June, blogger Simon Perry found the BCA database of 1,029 members online, containing 400 website URLs. He wrote a quick computer program to automatically identify all the chiropractors in the UK claiming to treat colic, locate their local Trading Standards office, and report them (more than 500 in total) automatically, followed up with printed letters."

This made me look up. If you haven't read the links here is my main interest in these reasonable disparate stories: they both use simple bots within a legal framework to spam requests. Respectivly one spams buy orders (buy for 1cent more and get a lot of money) the other spams legal letters requests for investigation. This has been online for a while in the form of trawling take down letters, these are simple bots that look for things that may be copyrighted material posted illegally and then spam the ISP of the concerned website with cease and disist letters. My main interest in this is that this sort of thing will produce an arms race, already this is the case on the stock exchange where groups are trying to out do each other with faster systems and smarter algorithms, legally similar systems are likely to evolve: clouds of company bots that exist to absorb the flak of other's take down bots and simlar...

for more ideas read accelerando by charles stross

anyway just thought I'd flag that up as it interests me

Saturday, 18 July 2009

DNA databases, Charlie and Ben

This is going to be a pretty short post as there isn't much to add to this discussion other than what is said in these two reports: first Ben Goldacre on the bad evidence used to justify long term retention (ie 24 years retetion) of DNA data taken from those who are arrested but not convicted or cautioned. Second is this article by Charles Stross on the odds of being a false positive in a government database check (in this case CRB check and why he doesn't give school readings of his books).

The only thing really that I want to add to the debate on a database of our DNA etc run by the government is that the risk in terms of security of a database that will ultimately link ALL your information in one place if breached is pretty horrible. I'll briefly expand this: the DNA database would be likely allowed to expand (through ID cards etc) to cover everyone and be multi-use, ie your medical history would be there you NI number what ever. Even if this doesn't happen a single source of information would make identity theft a utterly crippling crime, especially as it would be most likely a random person who would be used to act as a patsy for someone else. Although this is less likely given the current technological state it is likely only a matter of time before the current DNA system starts leaking.

Wednesday, 15 July 2009

Adventures with VHDL part 1.

Well I've finally got my arse in gear and started playing around with VHDL. I am subjecting myself to this for two reasons: 1) being able to program firmware is cool and useful as well being a reasonably rare skill 2) it's going to be useful for my PhD.

So what am I doing I hear you cry? Firstly I have been lent a very fun piece of kit: a Digilent NEXYS 2 board this is a programmable board with all sorts of ports and switches LEDs and similar to play with. Attaching it to my pc is a basic USB cable and then I have the Xilinx WebPack (a free download containing IDE and other useful programming gubbins). The final thing needed was the digilent program 'Adept' which allows you to upload your program via USB.

Armed with these I have ventured forwards into the world of chip programming.

So what have I learnt? thus far not much, I've got a basic program running that depending on the configuration of 4 switches the 7-segment display will give the number converted from binary into hexadecimal this was done by follow thing this fairly basic tutorial.

I learnt two major things: firstly check the EXACT setting of your card. I had a lot of issues in that I was getting a "device IDCODE conflict" error every time I tried to load the program on to the hardware this was because I had it set for the XC3S500E not the XC3S1200E in the 'Design properties' (right click the any of the files in the 'design' pane). The second thing I learnt was that to set the start up clock you need to right click 'Generate Programming File' in the process pane.

This sort of programming is interesting it that everything has two entries into your code: the first is called the entity which describes the number of inputs and outputs to a given structure; the second is the architecture that actually describes what is to be done with those inputs and outputs.

In my case the inputs are the switches and the outputs are the 7Seg display as well as anodes which switch of the bits of the display that I don't want (there are 4 separate 7seg displays I need only one).

The architecture is where the interesting programming happens (as far as I can tell) with '<=' directing inputs to outputs. In this case 'select' was used select a binary string for controlling the 7seg when a certain value was input via the switches (x0 through to xE). The anodes were all set manually via the line 'anodes <= "1110"; ' (remember that for this '1' is off '0' is on) this was done so that only the right most display was on.

On the entity side ports (inputs/outputs) were declared as a block of '3 downto 0' giving ports 3,2,1,0. each being declared "STD_LOGIC_VECTOR" which I assume defines the arithmetic to which it subscribes.

Finally once your .hdl file is finished you need to declare what physical ports everything attaches to. This is via a .ucf file that simply translates the port from the VHDL file to an actual port (in this case by looking up the number on the board itself).

Cool stuff

Frstly: the Chap Olympiad, a very daft but fun event recently held in London here's a link to an ITN report on it.

Secondly: big news in engineering and opto-electronics (if it gets off the ground) scientists have set up 'circuits' such that light can be used to open and close gates. This is important as it means that fully controlable gates can be made either way here is the report.

Sorry about the lazy reporting feeling tired and working on a programming project. Will probably update some information on it soon

Sunday, 12 July 2009

Update and first games review: 'War on Terror'

I haven't posted for a while, mainly because I've been skimming the news (busy week because of new job) and not seen anything too interesting.

Anyway I'm going to try and instigate some interesting blog material which will be to do with the various games I've been playing. I have plans to play some D&D at some point soon so that will be going up here but today I'm going to write about 'War on terror' (WoT).

I can't review it fully as I've only played a 1v1 game and it requires a few more people for the full affect but what I have seen so far is very good, very very good in fact.

At its most basic level WoT is a souped up version of risk with elements of settlers of catan; from the risk side comes the global domination/war aspect and from the catan side comes the resource production/settlement building side. The basics are fairly simple: each go you can build or upgrade a certain number of settlements as determined by the action dice, these can only be built in countries adjoining your existing settlements. Once you've built all you want to you roll two dice and gather oil money from the countries with the corresponding number (these are hidden until you build upon that country). As well the building there are two decks of cards (empire and terrorist) and the terrorist counters. The decks allow you to attack other countries or terrorist groups as well as other actions (free builds etc). The terrorist groups can be bought by anyone and placed anywhere on the board but once they're on there they are usable by anyone, this means that your opponent can move them into your countries or anywhere else. There is one final rule which is that a random player will be made the 'evil empire' this allows you to draw cards from both empire and terrorist stacks giving you a greater range of options but also giving others a bonus for attacking you.

These three mechanics work very well together, the building is simple and quick, attacks can be resolved by rolling two dice (unlike risk's hundreds) and once someone goes out they can control the terrorists.

In terms of actual gameplay WoT is highly intuitive, fast paced and funny; a game can take quiet a while but no one gets left out because they lost early which is very nice. The balance of it makes the terrorists very powerful and a dangerous group to deploy. It will take at least a few goes to get the balance of cash right as you need to spend money to make money as well as defend or weaken oponents with terrorists keeping money in reserver is also vital as some of the terrorist cards demand that you pay a hefty ransom.

The other advantage that I have found with this game is because of how you start (only 4 teritories each for a 2 player game) you don't have to start in direct competition with someone (like risk) and that global position is not quiet as crippling as in risk (ie starting in Australia doesn't mean instant win for you).

Outside of gameplay the game is very nicely designed, colourful with lots of nice touches it is wonderful just to look through the cards and around the board for the tons of little comic touches.

Generally this is a game for a pretty much anyone it is a long game and it does require a bit of thought to play BUT that being said it is excellent fun and doesn't penalise people who go out early.

Friday, 3 July 2009

Nextgen screens.

I've been recently reading 'Halting state' which is an amazing book by Charles Stross on the what will happen as the Internet becomes more and more ubiquitous. Among the technologies he uses in the plot is continuous Internet as a HUD (heads up display) this sort of HUD would be projected onto what you see from a pair of glasses giving you anything from the best route to your destination (think an iPhone whose screen you see all the time).

Most of this technology already exists (see the sixth sense prototype), iPhones supply constant GPS capabilities, 3G phone systems allow for some degree of Internet access almost anywhere (the newer 4G will improve this massively). Things like the cloud and the grid give constant wifi connections. MMORPGs (Massively-Multiplayer-Online-RolePlaying-Games like World of Warcraft) are becoming mainstream and as they become less esoteric and video games become a common media more MMORPGs will crop up. Urban gaming is already appearing with examples such as geocaching or the more paranoia inducing StreetWars.

In terms of non-existent technology only the ability to create overlays onto the real world are missing and this has moved a lot closer to reality with the creation of a .97" screen by the American company Kopin the big thing about this is that at less than an inch they have created a fully functioning 1280x1024 screen (that's the same resolution as a standard 19" monitor). While an inch screen sounds like the ultimate in eye-strain-o-vision at the distance of a glasses lens it is reasonable to view.

At most much of this technology will happen within the next 10 years or so, the full overlay technology may take a little longer but if nothing else we are now (pretty much) fully capable of hooking a good webcam to a pair of glasses and running the lens as screens. Of course what happens if someone hacks this would be pretty horrible.

UPDATE 1340 03/07/09: this news just broke. The theft was via standard abuse of privileges (something that is not punishable in eve although the subsequent sale of the kredits is) while the value of the theft was ~£3,000 and technically hard to punish legally it will be more and more common and is only a matter of time before someone starts testing this legally.