--
You received this message because you are subscribed to the Google Groups "FIGnition" group.
To unsubscribe from this group and stop receiving emails from it, send an email to fignition+...@googlegroups.com.
For more options, visit https://groups.google.com/groups/opt_out.
Hi guys,
I do want a reasonable set of programs to be written and available for FIGnition. It's just my priority has to be the firmware, because otherwise the firmware would become locked to programs that depended on features I wanted to change.
The 'problem' is that FIGnition turns out to have more potential than I imagined, and it's taking a while to get there. In the meantime, the best use is the kinds of thing Stuart & Carl have done with FIGnition, an exercise in understanding it; the fun of building it; and experimenting in coding it and expanding it.
-cheers julz
Hi Stuart,
" The challenge Micky is to write your own good software, and then share your learnings."
Like.
-cheers julz
Hi Micky,
In a sense both FIGnition & the R-pi were launched with the aim of getting more people (particularly kids) to code. But yes, the R-pi is much better for people who don't want to, because there's so many (tens of thousands of) programs already available. And for those who do, the R-pi will turn out to be no better than Mac, Linux or Windows, because it runs all the same programming tools & languages in the same way.
What the pi does excel at is as a hacking platform, though its I/O ports are fairly limited. It's great for bolting scripts & other people's code together. But the sheer availability of existing resources itself is a drag on learning to code.
FIGnition, I think, is not yet great for learning to code, but it could be good & is getting better. The good bit is that FIGnition is very direct and like early 80s machines is close to the hardware which means you're getting the real deal, not a fantasy language or environment.
Also like all 80s computers, the code base must be written by its users. This was true of the C64, Sinclair and BBC machines. That's part of the joy, Asteroids for FIGgy won't be a recompilation of MAME, it will involve real coding by someone who will learn how it's done. Blitz for FIGnition was (co)written by someone who couldn't code before he bought a FIGgy. That's part of the joy!
Check out my blog posts on the subject:
oneweekwonder.blogspot.com/2011/10/duplocode-fallacy.html?m=1
oneweekwonder.blogspot.com/2012/05/fignitionraspberry-pi-challenge.html?m=1
-Thanks for your comments Micky, cheers Julz
Hi James,
Perhaps strangely, I think so - lots of re implementations of retro programs that have been done before. Why? Because the value lies in how it teaches us to think. We don't judge that learning to multiply is pointless because it's been done before, we learned it because it gives us insights. Similarly, kids learn the same letters & words from generations back and then use them to write the same kinds of short stories their parents did, because it offers valuable insights into self-expression.
It doesn't follow that we're forever building on the past. Sometimes the new is grossly inferior to the old in some respects. Take for instance the android smartphone I'm writing this on. It is, simultaneously, the most powerful mobile hardware I've ever had: with a 1GHz Arm CPU, and the most dreadful mobile experience I've had to endure. The touch keypad is an utter pain, and with months of practice still dreadfully error prone. Vertically it's almost unusable, so right now it's horizontal. As a result I can only see 1.5 lines of text, worse than my Nokia N95 (2007) Sony Ericcson K900, W800 or K700 (going back to 2003) worse than my Siemens S45 (a tiny monochrome phone from 2001); worse than my Palm m505 (touch screen with graffiti 2001); worse than a Psion 5mx (2000? which is tough to beat) or a Series 3; worse than a Newton. And it's only as usable as it is, because I turned off predictive text (which kept mis-correcting correctly typed words because it was so dumb it couldn't even recognize "it's" without substituting some other word for the "'s" part) - dreadful, dreadful implementation that made me want to smash the phone on the pavement a dozen times a day, even after the OS upgrade that was advanced enough to keep the cursor in sync with the thumb tag!!!!!
But how can this be when Android is built on decades of progress? My view is it's a mis-nomer to reduce technology to the idea of "progress", because really it's human insights and thoughts we can learn from and to subsume them in this way inevitably obscures the insights themselves.
So, sure some aspects of these systems are a real improvement over older tech - their connectivity is fantastic and their input and sensing systems (gps, gyro, mems, multitouch) offer real advantages. Yet, it's wasted without real understanding; in this case to the point of near-unusability in basic operations (though the wonderful whizzy stuff works well).
I wrote a version of oxo, based on a 1K ZX81 Basic version (the one in the examples). It's only hundreds of bytes big, but offers more insights into the game than a brute-force algorithm. But then I came across a Practical electronics mag (or Everyday electronic) from 1978, where they implemented an oxo game (using Leds) in just a few dozen gates. Which just shows how much of an insight I don't have. Somewhere, someone is implementing an LED-based oxo game using LEDs through its GPIO port, with code running into mere 100s of Kb ;-)
-cheers julz
Hi Romilly,
Of course, you're right about the actual benefits of the pi. I do
think it's a good machine for what it's actually used for. And what
you're doing is impressive, 5000 orders is almost an order of
magnitude more than I've had in total.
I was just going on the original premise of the Raspberry PI. If you
go back to the BBC video from April 2011 on FIGnition.
http://www.bbc.co.uk/news/technology-13201254
Then look at the video from the BBC on the Raspberry PI (actually a
MicroDB [Micro Development Board?]) a week later:
http://www.bbc.co.uk/blogs/thereporters/rorycellanjones/2011/05/a_15_computer_to_inspire_young.html
It's very centred on them learning to program. You'll notice Braben
talks about it being a 'simple' computer for programming (in later
features, the BBC says it's a 'simple to use' computer).
The way the pi's being used is more hacking-oriented, but that makes
sense and at that price doubly sensible (a computer to learn to
program on can be more expensive, since Spectrum-era computers are
equivalent to £600 computers today, 3x the price of a netbook).
But, Wow - 5K Units, perhaps I'll be there in 5 years :-) !
I would think that a community that is energised and shares discoveries would bring a platform would it be a Pi or a FIG on leaps and bounds. The trick is to give enough to *so* others give more.
Thanks Romilly - Encouraging stuff & you're right, my dev budget was much <£30K! cheers julz