A break from the norm
Thursday was an unusually busy day for me.
In the afternoon there was the UK Enum Consortium AGM, a chance to catch up on the progress of UK Enum. My connection to the project is fairly tangental and mostly relates to my time at Telnic. Enum uses the same NAPTR technology as dotTel to support interconnection of PSTN and Internet routing for SIP services. Enum can actually be used for much more and the link with Telnic's .tel top-level domain is that the former offers an easy way to redirect a telephone number to a range of services whilst the latter does the same with names.
The link between UK Enum and .tel runs deeper than just a similar application of DNS technology though, including many of the same architects in their construction: Julian Rose, spikyblackcat, Jum Reid and Lawrence Conroy. The last I heard Jim and Lawrence were still busy down at Telnic HQ, spikyblackcat is off doing security for a major international financial network, and unfortunately Julian is no longer with us. So that leaves just me with both an understanding of the UK Enum technology and the freedom to follow the project.
Anyway the AGM was a small and relatively intimate gathering: some of the team from Nominet who control the registry along with the UKEC directors, the obligatory Man From The Ministry, a smattering of techies and interested parties, and yours truly. Once the formalities were out the way with the reelection of board members and acceptance of accounts by the members the meeting got down to the meat of the matter: namely how to raise awareness of Enum, and what to do about reaching a critical mass of registrations.
For a long time the preferred approach has been CRUE (Carrier Registrations in UK Enum), and indeed checking back through my notes I see that it was the main topic of conversation two years ago at the UK Enum Group meeting I attended with spikyblackcat. Unfortunately the debate doesn't seem to have progressed far beyond where it was at that time and I can see CRUE as being a confusing issue for sometime to come. On the plus side we also had the chance to discuss other aspects of the system and my suggestions for greater engagement of application developers were well received and I've offered to join the Policy Advisory Group when its formed in the near future. I'm not sure I'm suited to developer outreach as such, but if nothing else I can whip up some demo web apps for provisioning and manipulating NAPTRs.
One other change of note: UKEC has introduced a membership category for private individuals to complement the previous corporate memberships. Currently it's £100 + VAT per year and depending on how active I become in the Policy Advisory Group I'm strongly considering taking it out.
In the evening I attended a Club of Amsterdam conference on connectivity at the RSA. This is one of those unexpected events that turn up from time-to-time amongst my facebook invites without any obvious rhyme or reason, so I wasn't entirely sure what to expect but was determined not to miss an opportunity to hear Professor Cochrane talking on his pet topic. I'm glad I made time to attend.
The evening kicked off with a welcome from Club Chairman Felix Bopp followed by an interesting address by the Dutch Ambassador covering the thorny issues of privacy and data ubiquity by reference to Alfred Bester's "Tiger Tiger" (also published as "The Stars My Destination") an then the usual formalities from James Cridland of BBC A&Mi.
Dr Ir Egbert-Jan Sol of TNO gave an engaging presentation, something of a surprise given that it was mostly about the competitive pressure towards ever smaller digital components and the implications for pervasive processing power and bandwidth which follow. If the growth curves stay on track we're not that far from the point where computation and bandwidth will be pervasive which could make for an interesting democratisation of computability and data. Let's face it, one trillion connected devices are going to break all of our current routing mechanisms and require a far flatter peer-to-peer internet than the one we're currently used to - probably one in which the bandwidth available is genuinely unlimited but in which we will have to get used to higher latencies on round-trip communication.
Dr Sol also covered an area particularly close to my heart, that of autonomous vehicles. I would have liked him to mention some of the stuff people are doing with Sensor Networks and geolocation but twenty minutes is a tight runtime and he also had to fit in coverage of cutting-edge chip fabrication techniques, an area where Holland is currently the world leader.
Peter Cochrane in a sense picked up from this by giving us a quick tour of intelligence, both the limited human form and his hopes for a more pervasive and clueful digital form. There's a certain whiff of Vernor Vinge and his Technological Singularity about this, a line of reasoning that whilst valid mathematically strikes me as just one more example of Minsky's obsession with intelligence and consciousness as arising from algorithmic process. I find such a classical viewpoint itself strangely linear in light of the quantum revolution and also suspect the hard AI lobby underestimate the capabilities of the brain (with its 3.7 billion years of natural selection pressure), but that's a rant for another occasion.
Anyway, despite my reservations about the hard AI undertone, it was a thought-provoking presentation that for the most part focused on the non-linear nature of technological and social change, alongside mankind's supposed inability to think in this manner. I've certainly been frustrated by this tendency in others over the years and I agree with his assessment that simplistic solutions to complex social problems generally do more harm than good. Hopefully future generations will be blessed with a more open-minded educational system which encourages diversity of knowledge because as Cochrane himself recognised in his presentation, the polymath is no longer a significant element in our intellectual life and we're all the poorer for that. Of course Cochrane argues that's because we've passed a certain critical mass of knowledge at which it's no longer possible to be a polymath: it'll be interesting to see if that's really the case...
Hardy Schloer brought a different perspective to Peter Cochrane's thesis as he's involved with projects in self-programming systems and quantum computing, although yet again the main thesis was that computational advantage is driving us towards the singularity. I've yet to be convinced that we're only a couple of years from general self-programming and maybe half a decade from practical QC - for one thing the tools for mainstream programmers to access this stuff are probably more like a decade off at the earliest, even with the benefit of Schloer's Quantum Relations Theory. His contention that the decline of interest in chess was a direct consequence of IBM's Deep Blue beating Kasparov back in 1997 probably has a certain truth to it, even though the validity of the win itself as a proof of computer intelligence is still very much open to debate.
After the presentations were over there was a quick coffee break to allow people to mix and then it was back for the Q&A session. The questions covered the usual gamut from the utopianist visions of the obligatory Star Trek fan, to the somewhat more distopian issues of the third world. To be honest this was the weakest section of the evening, not because of any failing on the part of the speakers or the RSA but because the format was too carefully controlled by Mr Cridland with the lights down low and a reasonable immitation of that slightly supercilious manner David Dimbleby uses on Question Time. With a different style there could easily have been an additional quarter hour of discussion which would have allowed some of the thornier issues to be explored. In particular I would still like to know how the speakers see mankind's intrinsic tendency to form hierarchical structures being compatible with the democratising pressure of utterly pervasive data and computational power: is this a game changer in human behaviour, or will it ultimately be the stimulus for the replacement of our current species with one better adapted to individualism? I guess they're all only an email away if my curiosity overcomes that other great human trait - intrinsic laziness.
If the formal discussion wasn't all it could have been, I can't say the same for the informal discussion which followed over drinks (and in hindsight, possibly too many drinks). I discussed the missing aspect of the proposed technological utopia with two charming women who run businesses in the city, namely the lack of any role for human emotion. We came to the conclusion that without this the future will be a prison for mankind, the machines our gaolers to moderate our negative behaviours as opposed to a symbiotic extension of our individual nature. I hope that will not be the case.
The ghost of Deep Blue then lead me into an interesting discussion with one of the Fellows of the RSA and that spiralled as these things often do in a dozen different directions, the upshot of which is that I'm strongly considering becoming a Fellow myself. They fund work in several areas I'm interested in, ranging from effective education to neuroscience, and if the Fellows I met during my brief visit are even vaguely representative it seems I may have found some kindred spirits. The question of course then becomes one of whether the RSA will feel equally as keen to have me as a member...
In the afternoon there was the UK Enum Consortium AGM, a chance to catch up on the progress of UK Enum. My connection to the project is fairly tangental and mostly relates to my time at Telnic. Enum uses the same NAPTR technology as dotTel to support interconnection of PSTN and Internet routing for SIP services. Enum can actually be used for much more and the link with Telnic's .tel top-level domain is that the former offers an easy way to redirect a telephone number to a range of services whilst the latter does the same with names.
The link between UK Enum and .tel runs deeper than just a similar application of DNS technology though, including many of the same architects in their construction: Julian Rose, spikyblackcat, Jum Reid and Lawrence Conroy. The last I heard Jim and Lawrence were still busy down at Telnic HQ, spikyblackcat is off doing security for a major international financial network, and unfortunately Julian is no longer with us. So that leaves just me with both an understanding of the UK Enum technology and the freedom to follow the project.
Anyway the AGM was a small and relatively intimate gathering: some of the team from Nominet who control the registry along with the UKEC directors, the obligatory Man From The Ministry, a smattering of techies and interested parties, and yours truly. Once the formalities were out the way with the reelection of board members and acceptance of accounts by the members the meeting got down to the meat of the matter: namely how to raise awareness of Enum, and what to do about reaching a critical mass of registrations.
For a long time the preferred approach has been CRUE (Carrier Registrations in UK Enum), and indeed checking back through my notes I see that it was the main topic of conversation two years ago at the UK Enum Group meeting I attended with spikyblackcat. Unfortunately the debate doesn't seem to have progressed far beyond where it was at that time and I can see CRUE as being a confusing issue for sometime to come. On the plus side we also had the chance to discuss other aspects of the system and my suggestions for greater engagement of application developers were well received and I've offered to join the Policy Advisory Group when its formed in the near future. I'm not sure I'm suited to developer outreach as such, but if nothing else I can whip up some demo web apps for provisioning and manipulating NAPTRs.
One other change of note: UKEC has introduced a membership category for private individuals to complement the previous corporate memberships. Currently it's £100 + VAT per year and depending on how active I become in the Policy Advisory Group I'm strongly considering taking it out.
In the evening I attended a Club of Amsterdam conference on connectivity at the RSA. This is one of those unexpected events that turn up from time-to-time amongst my facebook invites without any obvious rhyme or reason, so I wasn't entirely sure what to expect but was determined not to miss an opportunity to hear Professor Cochrane talking on his pet topic. I'm glad I made time to attend.
The evening kicked off with a welcome from Club Chairman Felix Bopp followed by an interesting address by the Dutch Ambassador covering the thorny issues of privacy and data ubiquity by reference to Alfred Bester's "Tiger Tiger" (also published as "The Stars My Destination") an then the usual formalities from James Cridland of BBC A&Mi.
Dr Ir Egbert-Jan Sol of TNO gave an engaging presentation, something of a surprise given that it was mostly about the competitive pressure towards ever smaller digital components and the implications for pervasive processing power and bandwidth which follow. If the growth curves stay on track we're not that far from the point where computation and bandwidth will be pervasive which could make for an interesting democratisation of computability and data. Let's face it, one trillion connected devices are going to break all of our current routing mechanisms and require a far flatter peer-to-peer internet than the one we're currently used to - probably one in which the bandwidth available is genuinely unlimited but in which we will have to get used to higher latencies on round-trip communication.
Dr Sol also covered an area particularly close to my heart, that of autonomous vehicles. I would have liked him to mention some of the stuff people are doing with Sensor Networks and geolocation but twenty minutes is a tight runtime and he also had to fit in coverage of cutting-edge chip fabrication techniques, an area where Holland is currently the world leader.
Peter Cochrane in a sense picked up from this by giving us a quick tour of intelligence, both the limited human form and his hopes for a more pervasive and clueful digital form. There's a certain whiff of Vernor Vinge and his Technological Singularity about this, a line of reasoning that whilst valid mathematically strikes me as just one more example of Minsky's obsession with intelligence and consciousness as arising from algorithmic process. I find such a classical viewpoint itself strangely linear in light of the quantum revolution and also suspect the hard AI lobby underestimate the capabilities of the brain (with its 3.7 billion years of natural selection pressure), but that's a rant for another occasion.
Anyway, despite my reservations about the hard AI undertone, it was a thought-provoking presentation that for the most part focused on the non-linear nature of technological and social change, alongside mankind's supposed inability to think in this manner. I've certainly been frustrated by this tendency in others over the years and I agree with his assessment that simplistic solutions to complex social problems generally do more harm than good. Hopefully future generations will be blessed with a more open-minded educational system which encourages diversity of knowledge because as Cochrane himself recognised in his presentation, the polymath is no longer a significant element in our intellectual life and we're all the poorer for that. Of course Cochrane argues that's because we've passed a certain critical mass of knowledge at which it's no longer possible to be a polymath: it'll be interesting to see if that's really the case...
Hardy Schloer brought a different perspective to Peter Cochrane's thesis as he's involved with projects in self-programming systems and quantum computing, although yet again the main thesis was that computational advantage is driving us towards the singularity. I've yet to be convinced that we're only a couple of years from general self-programming and maybe half a decade from practical QC - for one thing the tools for mainstream programmers to access this stuff are probably more like a decade off at the earliest, even with the benefit of Schloer's Quantum Relations Theory. His contention that the decline of interest in chess was a direct consequence of IBM's Deep Blue beating Kasparov back in 1997 probably has a certain truth to it, even though the validity of the win itself as a proof of computer intelligence is still very much open to debate.
After the presentations were over there was a quick coffee break to allow people to mix and then it was back for the Q&A session. The questions covered the usual gamut from the utopianist visions of the obligatory Star Trek fan, to the somewhat more distopian issues of the third world. To be honest this was the weakest section of the evening, not because of any failing on the part of the speakers or the RSA but because the format was too carefully controlled by Mr Cridland with the lights down low and a reasonable immitation of that slightly supercilious manner David Dimbleby uses on Question Time. With a different style there could easily have been an additional quarter hour of discussion which would have allowed some of the thornier issues to be explored. In particular I would still like to know how the speakers see mankind's intrinsic tendency to form hierarchical structures being compatible with the democratising pressure of utterly pervasive data and computational power: is this a game changer in human behaviour, or will it ultimately be the stimulus for the replacement of our current species with one better adapted to individualism? I guess they're all only an email away if my curiosity overcomes that other great human trait - intrinsic laziness.
If the formal discussion wasn't all it could have been, I can't say the same for the informal discussion which followed over drinks (and in hindsight, possibly too many drinks). I discussed the missing aspect of the proposed technological utopia with two charming women who run businesses in the city, namely the lack of any role for human emotion. We came to the conclusion that without this the future will be a prison for mankind, the machines our gaolers to moderate our negative behaviours as opposed to a symbiotic extension of our individual nature. I hope that will not be the case.
The ghost of Deep Blue then lead me into an interesting discussion with one of the Fellows of the RSA and that spiralled as these things often do in a dozen different directions, the upshot of which is that I'm strongly considering becoming a Fellow myself. They fund work in several areas I'm interested in, ranging from effective education to neuroscience, and if the Fellows I met during my brief visit are even vaguely representative it seems I may have found some kindred spirits. The question of course then becomes one of whether the RSA will feel equally as keen to have me as a member...