This discussion has been locked.
You can no longer post new replies to this discussion. If you have a question you can start a new discussion

Local Network Webinars

Tuesday 15th November saw the first 'Webinar' sponsored by the IET Somerset and West Wiltshire Network, “Where Are We ? - A History of Time Measurement & Navigation” presented by Emily Akkermans. While introducing the event Peter Shiret canvassed the opinions of the viewers on webinars in general, here are my thoughts, with the hope that it will encourage others to add their opinions.



 



Local network-sponsored webinars are clearly a sensible substitute for the conventional local event, given the restrictions imposed as a consequence of the spread of Covid-19 but might they become the 'normal' way to conduct events?



 



I believe there are four groups of players involved in holding an IET event, the IET centre, IET local network, the presenter and the audience, each having aspirations that may not completely align. Broadly the centre wishes to promote the profession, the network to build and sustain the network, the presenter to inform and the audience to be informed and entertained. In theory a webinar can meet all of these desires but these aren't the only ones sought by the players.



 



For the centre there might be a temptation to sponsor all webinars. That would certainly take a load off network sponsors but at the expense of diminishing their role and possibly support for topics of a local interest that might be considered too specialist for a national or international audience. For the centre there is also the possibility that the audience becomes yet more acclimatised to getting engineering information online and discovers that there is some excellent material 'out there' and the institution loses its relevance.



 



The network runs the same risk, it only has relevance if its members share some common connection, i.e. the locality, but we are all equally 'local' (or distant) on the internet. The distinction between the centre and the network can easily become blurred.



 



For the presenter they possibly gain by not having to travel, (some might enjoy it). There might be some relief at not having to face an audience, (which might encourage the shy to take on the task), but I am sure most get some pleasure in the sense of connecting personally with the audience. It must be very odd not to receive even a clap at the end. I would think there might be a temptation to move from a webinar format to a (polished) internet lecture that can be viewed on demand. It has been said that radio and television killed off the music hall as a stage performer can present the same act all week at one town and then repeat it at the next. 'Perform once' might be commercial suicide, but then most presenters are not being paid so that is perhaps not so important for a technical talk.



 



For the audience the big plus is perhaps not having to travel, traditionally our events seem to run when it is dark and cold! The webinar has the possibility of providing better on-screen information, no problems of screens too far away, too dark or obscured by someone's head. The downsides include losing that personal connection, the presenter that picks up on an audience reaction, the feel of being part of a larger body. The chance to meet others in the network is gone too. (Personally I have always felt it is a pity that people don't meet up after an event, at least we would all have something in common to talk about. I often feel that some of the ideas brought up in a talk could do with being teased out, perhaps after a bit of self-reflection – I envy those smart enough to compose and ask a relevant question at the right time!).



 



My quick comment, better than nothing, but nothing beats a good live speaker.



 



Any thoughts?



 


Parents
  • Thank you for joining in Dominic!


    On Tuesday, 12 January the second local webinar, "Light Years Ahead: The 1969 Apollo Guidance Computer" was presented by Robert Wills, replacing the cancelled live event of the same name from early last year.


    Continuing the discussion of webinars in general and the local ones in particular I noted that more people took part in this event than in the previous one, 400 or so against 300, I believe. These are very large numbers compared to the traditional local event, (excluding 'flagship' events when the IET president and the government scientific advisor presented). It does make me wonder how 'local' these webinars were as, in theory, the audience could be world-wide.


    I noticed that this time a lot more use was made of the 'chat' facility, perhaps Peter hadn't issued as many cautions this time? There must be a danger that the presenter/moderators may get overwhelmed by too much 'feedback'. Some people used it to thank the presenter, nice in theory, not so good if there are 400 'chats' to pick up on 'real-time'.


    As in the previous event the preferred 'query' facility was a bit hit-and-miss. Clearly the presenter/moderators were seeing more questions than I ever saw. There was some talk of these questions being captured for later use. Did that mean that individuals might receive private responses later? Questions in a live event usually get answered live and the response no doubt satisfies more than just the original questioner.


    This particular event was, we were told, was being recorded. I wonder when and how it will be made available and will it include any extras like answers and links? One of the reasons that in the past I had set up a discussion topic about live events was to provide a forum for further questions and answers.


    Finally some comments on the latest webinar that piqued my curiosity. Why should using 1's complement numbers be a problem (two ways of indicating zero)? Signed binary and normal decimal notations have the same 'problem' and the 'solution' is just not to use negative zero for input and the computer won't produce it during calculation. I rather suspect that the reaon for using 1's complement is a direct result of using NAND gates as the basic building block. Software engineers might think of 0 and 1 as being of equal 'value' but from a hardware point of view one state might be easier to assert, 'reset' signals invariably appear in the negated form, zero volts. Main Motor Burn and Main Motor Not Burn are logically equivalent but not in hardware terms. Similarly word size for a control computer would be determined more by input/output control word size, not by the need for mathematic precision. That can be achieved by using more than one word as and when required, a large word size increases the size of the processor and memory even when it is not needed. The popular 8-bit microprocessors, such as the 6502, 6800 and Z80, were able to handle floating point numbers if programmed to do so.


    Someone asked about using alternatives to core memory, such as diodes. Core memory was 'high-density' for its day. I built a 96 bit diode memory for a 'home job' and a few years later was working with a 4k x 12 bit core store of pretty much the same size. Talking of 'home jobs' I think the biggest project I made for home use had about 45 ICs, I can't think why anyone in this day and age would want to tackle a 6,000 gate project by themself. There is no guarantee that the extant AGC documentation is correct, (missing change notes?). Personally I would either simulate the circuit, (a lot easier than 24,000 plus wire-wraps) or emulate the software on something like a Raspberry Pi - or just use someone else's simulation! http://svtsim.com/moonjs/agc.html Each to their own - good luck!
Reply
  • Thank you for joining in Dominic!


    On Tuesday, 12 January the second local webinar, "Light Years Ahead: The 1969 Apollo Guidance Computer" was presented by Robert Wills, replacing the cancelled live event of the same name from early last year.


    Continuing the discussion of webinars in general and the local ones in particular I noted that more people took part in this event than in the previous one, 400 or so against 300, I believe. These are very large numbers compared to the traditional local event, (excluding 'flagship' events when the IET president and the government scientific advisor presented). It does make me wonder how 'local' these webinars were as, in theory, the audience could be world-wide.


    I noticed that this time a lot more use was made of the 'chat' facility, perhaps Peter hadn't issued as many cautions this time? There must be a danger that the presenter/moderators may get overwhelmed by too much 'feedback'. Some people used it to thank the presenter, nice in theory, not so good if there are 400 'chats' to pick up on 'real-time'.


    As in the previous event the preferred 'query' facility was a bit hit-and-miss. Clearly the presenter/moderators were seeing more questions than I ever saw. There was some talk of these questions being captured for later use. Did that mean that individuals might receive private responses later? Questions in a live event usually get answered live and the response no doubt satisfies more than just the original questioner.


    This particular event was, we were told, was being recorded. I wonder when and how it will be made available and will it include any extras like answers and links? One of the reasons that in the past I had set up a discussion topic about live events was to provide a forum for further questions and answers.


    Finally some comments on the latest webinar that piqued my curiosity. Why should using 1's complement numbers be a problem (two ways of indicating zero)? Signed binary and normal decimal notations have the same 'problem' and the 'solution' is just not to use negative zero for input and the computer won't produce it during calculation. I rather suspect that the reaon for using 1's complement is a direct result of using NAND gates as the basic building block. Software engineers might think of 0 and 1 as being of equal 'value' but from a hardware point of view one state might be easier to assert, 'reset' signals invariably appear in the negated form, zero volts. Main Motor Burn and Main Motor Not Burn are logically equivalent but not in hardware terms. Similarly word size for a control computer would be determined more by input/output control word size, not by the need for mathematic precision. That can be achieved by using more than one word as and when required, a large word size increases the size of the processor and memory even when it is not needed. The popular 8-bit microprocessors, such as the 6502, 6800 and Z80, were able to handle floating point numbers if programmed to do so.


    Someone asked about using alternatives to core memory, such as diodes. Core memory was 'high-density' for its day. I built a 96 bit diode memory for a 'home job' and a few years later was working with a 4k x 12 bit core store of pretty much the same size. Talking of 'home jobs' I think the biggest project I made for home use had about 45 ICs, I can't think why anyone in this day and age would want to tackle a 6,000 gate project by themself. There is no guarantee that the extant AGC documentation is correct, (missing change notes?). Personally I would either simulate the circuit, (a lot easier than 24,000 plus wire-wraps) or emulate the software on something like a Raspberry Pi - or just use someone else's simulation! http://svtsim.com/moonjs/agc.html Each to their own - good luck!
Children
No Data