Disqus - Latest Comments for aksyn

Re: Shopping at the OVI store…

aksyn — Fri, 23 Oct 2009 05:47:22 -0000

Genius.

Re: Gents, let’s not SpinVox this up, ok?

aksyn — Thu, 27 Aug 2009 06:06:32 -0000

I'm guessing they're the exception to the rule "no such thing as bad PR" :-)

Re: MIR London Mixer: This Wednesday at Charlotte Street Blues

aksyn — Mon, 10 Aug 2009 13:40:02 -0000

Well will ya look a that.. it's only 30 seconds from the Howler HQ. Anyone'd think I planned it that way ;)

Re: Nokia’s got game: Android handset coming in September

Matt Bland — Tue, 07 Jul 2009 15:24:59 -0000

You should have your own blog, column or something mate. Concise, well presented and thought out.

Re: Nokia’s got game: Android handset coming in September

aksyn — Tue, 07 Jul 2009 08:17:42 -0000

Not at all, go ahead!

Re: Nokia’s got game: Android handset coming in September

smstextnews — Tue, 07 Jul 2009 08:03:11 -0000

Do you mind if I publish that as a stand-alone post, Jay?

2009/7/7 Disqus <>

Re: Nokia’s Ovi let’s the market decide. The market says, ‘no’ (and how to fix it)

aksyn — Tue, 07 Jul 2009 07:48:24 -0000

(I wrote this for another post elsewhere on the site, but this one seems more apt and current so I'll copy it here)

I've wanted to write this for a while, because if they're not going Android, I don't know what their next move is going to be.

I really don't get Nokia these days. The hardware is good, as it always has been for the most part, but they've always been let down by OS. I think their biggest mistake was in not getting fully behind their very own PyS60 project (Python for S60). For those of you not familiar, a huge amount of work went into the PyS60 project at Nokia internally, almost as a grassroots project, to port the Python programming language to the Symbian platform.

The project was/is a great success, in that they succeeded in getting it working, and went the extra mile developing all sorts of simple hooks so that you could write Symbian apps, in Python, on any Symbian device.

The only thing it needed (or needs, as they could still pull it off) was for the company as a whole to get behind it, sort out the run-time distribution (install it on the device, include it in firmware upgrades, make it available OTA), signing process and (to allow deeper integration into the device) allow native modules to exist in \private directories (or some other solution).

Instead it's now fallen to the open source community to sort out the mess that is PyS60 app distribution (distributing the run-time with your app, different versions, different forks, overlapping modules, everything in one directory, the signing process etc.), and it needed to remain in the hands of Nokia so that it could be a common de-facto run-time across all devices. At the end of the day, only Nokia can sort out the mess that is security signing on Symbian - it's never going to be solved by the community.

So, why should they have gotten behind this? Because Apple came out from nowhere with an incredibly productive development platform and people (developers) took notice. Apple had the IDE (Xcode and Interface Builder), the debugger, the profiler, the language (Obj-C... somewhere between C and C++, but a little better thought out) and an API and it all made sense and was easy to use from day #1.

Symbian C++ is still to this day (8 years since release) considered a black art, crafted in basements by men with pointy hats and long beards. The tools provided are archaic.. as if they came from another era (which is pretty much the case given its roots in EPOC and Psion) and make you feel that we've learned nothing in the near 20 years of software development since EPOC was born.

Python on S60 by comparison gives you a nice, safe, productive language that almost anyone can pick up - its clean and concise. The guys at Nokia (and the open source community) have covered most bases of Symbian development in terms of user interface design, networking, sound and integration into the device, and wrapped it all into nice, easy to understand, Python API functions.

On top of that, you can fairly easily (caveat to the above run-time distribution issues being sorted out) include native modules with your Python application - so you can access the phones innards and do all the wonderful low-level pointy beard stuff that you need to on the device for games and whatnot, but add a great consistent Symbian UI and your higher level logic with very little development effort on top.

That Nokia haven't entertained it at all, and I know people have tried to evanglise it from within and FAILED, strikes me as a company that really has no clue as to what they're doing in this market anymore.

Re: Nokia’s got game: Android handset coming in September

aksyn — Tue, 07 Jul 2009 07:36:08 -0000

I've wanted to write this for a while, because if they're not going Android, I don't know what their next move is going to be.

I really don't get Nokia these days. The hardware is good, as it always has been for the most part, but they've always been let down by OS. I think their biggest mistake was in not getting fully behind their very own PyS60 project (Python for S60). For those of you not familiar, a huge amount of work went into the PyS60 project at Nokia internally, almost as a grassroots project, to port the Python programming language to the Symbian platform.

The project was/is a great success, in that they succeeded in getting it working, and went the extra mile developing all sorts of simple hooks so that you could write Symbian apps, in Python, on any Symbian device.

The only thing it needed (or needs, as they could still pull it off) was for the company as a whole to get behind it, sort out the run-time distribution (install it on the device, include it in firmware upgrades, make it available OTA), signing process and (to allow deeper integration into the device) allow native modules to exist in \private directories (or some other solution).

Instead it's now fallen to the open source community to sort out the mess that is PyS60 app distribution (distributing the run-time with your app, different versions, different forks, overlapping modules, everything in one directory, the signing process etc.), and it needed to remain in the hands of Nokia so that it could be a common de-facto run-time across all devices. At the end of the day, only Nokia can sort out the mess that is security signing on Symbian - it's never going to be solved by the community.

So, why should they have gotten behind this? Because Apple came out from nowhere with an incredibly productive development platform and people (developers) took notice. Apple had the IDE (Xcode and Interface Builder), the debugger, the profiler, the language (Obj-C... somewhere between C and C++, but a little better thought out) and an API and it all made sense and was easy to use from day #1.

Symbian C++ is still to this day (8 years since release) considered a black art, crafted in basements by men with pointy hats and long beards. The tools provided are archaic.. as if they came from another era (which is pretty much the case given its roots in EPOC and Psion) and make you feel that we've learned nothing in the near 20 years of software development since EPOC was born.

Python on S60 by comparison gives you a nice, safe, productive language that almost anyone can pick up - its clean and concise. The guys at Nokia (and the open source community) have covered most bases of Symbian development in terms of user interface design, networking, sound and integration into the device, and wrapped it all into nice, easy to understand, Python API functions.

On top of that, you can fairly easily (caveat to the above run-time distribution issues being sorted out) include native modules with your Python application - so you can access the phones innards and do all the wonderful low-level pointy beard stuff that you need to on the device for games and whatnot, but add a great consistent Symbian UI and your higher level logic with very little development effort on top.

That Nokia haven't entertained it at all, and I know people have tried to evanglise it from within and FAILED, strikes me as a company that really has no clue as to what they're doing in this market anymore.

Feels good to get that out. Rant over!

Re: AMR-NB G.729 CDMA iLBC DSP + Howler Tech. Nice.

PatrickatJoshuaPR — Thu, 12 Feb 2009 08:07:43 -0000

Jay, it's probably good timing - there's not that much actual sunlight around at the moment anyway!

Re: AMR-NB G.729 CDMA iLBC DSP + Howler Tech. Nice.

aksyn — Thu, 12 Feb 2009 07:30:21 -0000

Thanks Patrick - been in the basement so long I'm going to need keep some shades on!

Re: Private: Review: Teleware PMN Rapid Deployment Unit (RDU)

aksyn — Fri, 28 Nov 2008 12:32:35 -0000

I have no doubt that it /is/ possible, but I couldn't find an option to do it in amongst the hundreds of configurables (we also didn't have a manual, and I did try to call support several times but... well, lets just say the support wasn't a great experience).

Re: Your mobile will cook a hardboiled egg. In 80 minutes.

aksyn — Fri, 10 Oct 2008 08:52:09 -0000

OMG we've soft-boiled James Body!!

Re: Private: Review: Teleware PMN Rapid Deployment Unit (RDU)

Mike42 — Tue, 30 Sep 2008 11:05:57 -0000

"Radio waves should be free, man."

No they bloody shouldn't! That's VC talk there. Standards, properly co-ordinated and managed, are it, man. This is the RF equilvalent of saying that IP should be free - We don't need that pesky IPv4/ICANN system. Free the Bits!

Proponents of free RF favour the 'Detect and Avoid' mechanism, relying on ultra/fast and sensitive detectors to sense when spectrum is in use by others. Nice idea, but there are fundamental flaws in this logic when operating at the noise floor, as CDMA systems do.

We've seen what 'Free' radio waves delivers - Any Blyk subscribers out there?

Funnily enough, I like call quality. I like 5 9's of reliability. I have no desire to bring the wonderful end-user experience that is WiFi RF "management" to the voice/sms world.

You'll be telling me mesh networks are viable next!

Re: Private: Review: Teleware PMN Rapid Deployment Unit (RDU)

Mike42 — Tue, 30 Sep 2008 10:59:12 -0000

It's the red velour chair / throne thing....

Re: Private: Review: Teleware PMN Rapid Deployment Unit (RDU)

kbateman — Tue, 30 Sep 2008 09:16:17 -0000

Yup complete agreement here ... if we can figure out the prices I'm sure the usual crowd (and the prom queen) could knock something together that works. Sounds like a good bit of fun too...

Re: Private: Review: Teleware PMN Rapid Deployment Unit (RDU)

aksyn — Tue, 30 Sep 2008 07:58:02 -0000

Prom queen? Have you been following me at the weekends again?

Re: Private: Review: Teleware PMN Rapid Deployment Unit (RDU)

aksyn — Tue, 30 Sep 2008 07:57:20 -0000

Any clue as to how much it costs to get a license? The PMN can only handle 4x calls at once (8-channel AMR BTS) and has limited range, so surely can't have cost Teleware too much? (famous last words, but otherwise Pico/Femto-cell deployments in offices would be prohibitively expensive).

The OpenBTS project looks great, but the lack of easy legal testing is obviously holding up its progress.

Radio waves should be free, man.

Re: Private: Review: Teleware PMN Rapid Deployment Unit (RDU)

aksyn — Tue, 30 Sep 2008 07:50:15 -0000

Thanks Terence. It wasn't roaming per-sae, as the PMN had no backhaul facility to T-Mobile and so they were totally out of the loop when I was within range of the PMN. You'd need to add an 'outside of coverage' divert on your T-Mobile number, to bounce the call to a real phone number that was routed to the PMN.

Ideally, you'd grab a number from someone like AQL.. then you can have Voice+SMS+Fax all on one number, and have it automagically route to wherever you are :) No roaming charges, and all your calls can be routed via VoIP out of the PMN when you're at your home/office - I imagine it'd pay for itself fairly quickly!

Maybe we can all club together and time-share a PMN?

Re: Private: Review: Teleware PMN Rapid Deployment Unit (RDU)

aksyn — Tue, 30 Sep 2008 06:04:54 -0000

If you enjoyed the review, please Digg it so other like-minded folk get to see: http://digg.com/hardware/Teleware_s_Private_Mob...

Thanks :)

Re: Iridium Networks in rude health

aksyn — Wed, 06 Aug 2008 11:05:06 -0000

I never quite understood why the data access was /so/ slow - isn't it like 9600bps or something?

But yes, you're right. I want one. What do they cost to run?

I bet it only works it slightly cheaper than Vodafone :-D

Re: Private Mobile Network: First look!

smstextnews — Thu, 03 Jul 2008 15:24:06 -0000

Hmmm... email me and let's see what we can do, I need to send it back soon!

Re: Private Mobile Network: First look!

aksyn — Thu, 03 Jul 2008 15:21:41 -0000

Ewan, I've just moved out t' country and the mobile signal here leaves something to be desired. If you can clear it with Teleware, I'd love to put it through its paces? (can they make a slightly cheaper version please?)