Source Reference: https://www.youtube.com/watch?v=Kusuu4NDEfg
Comment Reference: https://www.youtube.com/watch?v=Kusuu4NDEfg&lc=Ugy7PhPWQSLSf-OttPt4AaABAg
1:59:21 One problem why anarchism couldn't work was that all the people would have to talk all the time about everything. Wonder if the computer could help with that, and indeed, aspects of anarchism are quite prominent on the net now (as well as other aspects from other governance models, no need to be ideologically about it, just pick the best aspects of everything).
Comment Reference: https://www.youtube.com/watch?v=Kusuu4NDEfg&lc=UgwhAhSuKccmTk8ZYwR4AaABAg
1:46:35 So please, Sam and Harry, tell me please if you're typing on a Dvorak keyboard, and if not, why not. Furthermore, what do you think, is it a technology/tools problem or has the keyboard setting/arrangement more to do with the human system, or is it some kind of training/infrastructure cost?
Comment Reference: https://www.youtube.com/watch?v=Kusuu4NDEfg&lc=Ugze7LQflfWOrZzrvhB4AaABAg
1:46:18 This is where I really get upset: how can you call the iPhone "technology"? It's a proprietary piece of compliance device, with the sole distinction of artificial restrictions plus some branding and packaging. How can you talk about "tools" and "technology" if you don't even know what they are?
Comment Reference: https://www.youtube.com/watch?v=Kusuu4NDEfg&lc=UgwiRe7ipASPQNy7LIF4AaABAg
1:45:51 Erm, our problem is that there aren't any really good violins around, we don't know where to get them, we don't know how to make them. Therefore, there aren't a lot of good violin players around (if at all), or at least we don't get to hear a lot of good music at the moment, we don't know how it sounds like, we can only imagine how it would, and that it must exist somewhere and should be possible, but it's also possible that we'll remain in our very poor state just as it is now. No matter how good the players are, they won't be able to find a good instrument to exercise their skills (by the way, how did they even learn them?) to the full extend.
Comment Reference: https://www.youtube.com/watch?v=Kusuu4NDEfg&lc=Ugyh_4ljb5_wprSLpC54AaABAg
1:41:39 Oh my, is this what the group is about, why you guys are meeting? Do you even have a glimpse of understanding the problem in that space, sociologically?
Comment Reference: https://www.youtube.com/watch?v=Kusuu4NDEfg&lc=UgwF3m1JDwGrUBgPaKd4AaABAg
1:36:26 This is a huge, huge field, I don't want to even enter to discuss it in detail if there isn't explicit demand to do so.
Comment Reference: https://www.youtube.com/watch?v=Kusuu4NDEfg&lc=Ugzrwe_CUh_FSXiVz7t4AaABAg
1:33:35 The time on my disposal or the time in the day doesn't double every 2 years. Anyway, I think the effects at play here are different ones: not Moore's law, but increased availability and networking. A lot more people can now create their own tools and paradigms, share and distribute them easier than ever before, plus, the old stuff goes never away and stays available permanently. It's with the explosion of knowledge at the time of the invention of cheap paper and the printing press. It takes one guy a few hours or weeks or months or years to create something, but when completed, he can distribute the thing to a lot of people, so each individual gets completely flooded with what others have created. Sure, creation takes more time than creation, but at a certain point the amount of people creating stuff will exceed the corresponding time it takes to consume their stuff for a single individual. Furthermore, learning new methodologies and tools might lead to their confusion, if they're not easily distinguishable. How do you know in advance what's worth the investment of learning? So the strategy is to learn one and then stay with it, it's pretty much like vendor lock-in, where changing is costly and might not offer substantial benefit.
Comment Reference: https://www.youtube.com/watch?v=Kusuu4NDEfg&lc=Ugxq_VYONzkk8TYuhmt4AaABAg
1:33:00 No, those two things are not related in any way. If a chip can do more instructions, that doesn't have any impact on the cleverness of the software it's running, except that we can run more of the same smart or stupid applications, having the calculations be performed in less time, no matter if they make sense or are totally stupid, plus they don't change their algorithm or programming just because the chip got faster. I think it's Bill Gates or maybe Marc Andreessen who addressed this point, that the cost in computing isn't the hardware any more and the IBM software for the machine being gratis, it's not even the costs of running the software at the client side in buying the computers and the power bill, but it's the tremendous shortage of software developer time, and that one isn't doubling every 18 months no matter what we do, because biological and mechanical/electrical systems follow quite different metrics. It's also not even necessary that the programming must improve just because the hardware got better, if the software is good, fine, we won't profit much from Moore's law, if it is bad, waiting 18 months won't make it better except we design/anticipate applications that need to calculate an immense amount of data, for which we can't affort the computing power at the moment, but still, it might be the case that we started cheating with Moore's law in quite some ways: "cloud" computing, multiple/vertical cores, as we hit the limits of physics (OK, there might be quantum computing one day, coming down in price in a similar fashion to have it in your pocket for a few bucks), then most older applications are written to just use a single core regardless of the other cores (because who can invest the many man-years to reprogram it for parallel execution?), and the prices on the consumer market seem to follow different rules, because a comparatively "slow" computer is still a very useful tool for writing e-mails/documents, watching videos, surfing the net, so prices don't fall much for this basic utility. Performance doesn't increase much because that isn't needed for those basic functions anyway. But then, more and more layers of software and background services are crammed into the standard operating system, so that performance needs to increase just to provide the same basic functions it did before with less. There are other limits as well, be it how much data the front-side bus can deliver to the CPU, or the frequency the machine is running with, or the instruction set, or cache sizes, etc., so more calculation power can't be used all the time in an ideal way. It also invites bad programming, because developers enjoy their powerful machines and trade expensive developer time for coming up with a good algorithm for the performance increase every 18 months that's expected for the machines of the client. In the end, the human and tool systems need to improve because with the tools we already have, don't really understand or don't care about their implications (cars, nukes, consumer electronics), we absolutely destroy the planet at a rapid rate, and we don't have an exponential way to fix or mitigate this trend, maybe not even a linear one. The problems are big, complex and systemic, so it's unlikely that we can work on them on the human level at all, on a global scale, but tools hopefully could at least contribute a little bit with learning and understanding the complex feedback loops of the systems, to model and simulate them. Feel invited to discuss what and how could be done, I continue to doubt that human or tool side or both together can make a significant impact.
Comment Reference: https://www.youtube.com/watch?v=Kusuu4NDEfg&lc=Ugw1Y6rCyFhINXxFlDN4AaABAg
1:24:48 It's the industrial center of the craft of toolmakers, especially driven by investment business. There are immense incentives to do other things than what Doug wanted to do, so what do you expect? What other centers are there for big world problems? Would it be impossible to find some software developers there?
Comment Reference: https://www.youtube.com/watch?v=Kusuu4NDEfg&lc=Ugw71xKQJH_BKCOowrh4AaABAg
1:21:53 Because implementing Doug's ideas are hard, people who did the work say that Doug didn't seem appreciative of the stunning work that was actually achieved, because it was only a few percents of his grand vision, so they went to a place where their effort was more welcomed for what it is. It was fine enough for a printer company, it wasn't for sovling big, complex world problems. How can the computer be a tool for the latter if you can't actually get the most simple tasks around text work on it properly to a decent statisfaction?
Comment Reference: https://www.youtube.com/watch?v=Kusuu4NDEfg&lc=UgwjhLQhPyTFQE07Rj94AaABAg
1:22:08 I don't know if that's the same. The GUI paradigm is about having a bitmapped interface as the primary way of interaction and less a text-based system with embedded drawing areas here and there. The Xerox PARC work is famous for the GUI as we know it because I'm not aware if NLS/Augment had things like the desktop, icon, control button, menu metaphors etc., which are not present in Doug's systems to this day, maybe for deliberate, good reasons, but they're not really the same I guess.
Comment Reference: https://www.youtube.com/watch?v=Kusuu4NDEfg&lc=UgzWLVSI8NKaq0Fs3kh4AaABAg
1:20:50 No. You have to realize how hard it is to implement all the things Doug wanted, it's really, really hard, back in the day and today too (for other reasons). Bill English and Jeff Rulifson point out that Doug's demands simply weren't possible to realize due to time, money, etc. constraints. If you've talked to Doug and didn't find difficulty to work with him, then probably because you didn't have to built his systems, and even your conversations with Frode, both of you don't implement, let's say, the Time Browser, and it's still very hard to even come up with a reasonable design, let alone bring it to live. Sure, computers can be made to do anything for us, whatever we can imagine, but it's also very, very costly, one could easily spend his life's work and never see a return on it. The reason why we only have one type of GUI etc., is, that you don't easily write window libraries that are too efficient enough and decent in design, just to find out that nobody wants to learn them as users are already used to the one we somehow arrived from ARC to Xerox PARC to Apple/Windows/X11.
Comment Reference: https://www.youtube.com/watch?v=Kusuu4NDEfg&lc=UgxB2CcZJ2aSZxh-laN4AaABAg
1:15:47 There's https://www.youtube.com/watch?v=8za_4g5zCOM now, but without credit card, I can't order something from the US.
Comment Reference: https://www.youtube.com/watch?v=Kusuu4NDEfg&lc=Ugy5dyeIyInlwtwtPvN4AaABAg
1:15:42 The Doug Engelbart Institute dismisses it. What now?
Comment Reference: https://www.youtube.com/watch?v=Kusuu4NDEfg&lc=UgxYWemCyMfD2aMXY014AaABAg
1:12:20 This is an excellent way to look at it. I'm a lot in favor of Jon Bosak's "principle of doing things based on principle, not expediency", also promoted by Bret Victor's "Inventing on Principle" and "The Future of Programming".
Comment Reference: https://www.youtube.com/watch?v=Kusuu4NDEfg&lc=UgwKhlfrCKCO2pb5vc94AaABAg
1:09:21 I think, in part, it has to do with the problem of references/links to all those versions. If a single version would change and the links to it not, they would end up broken. Now, there might be ways around this, but I'm not aware of a very, very good one, especially if it's not only limited to text, but also expanded onto semantic markup. It's much easier to never change revisions and keep links to a specific version intact. This is where work needs to be done.
Comment Reference: https://www.youtube.com/watch?v=Kusuu4NDEfg&lc=UgxJ4AR25jQv9fvzlgt4AaABAg
1:06:47 I just want to talk about hypertext and word processing. As a Christian believer, I'm seriously suspicious of self-salvation attempts by atheists for deep, fundamental reasons. I don't find these addressed with Doug nor in Doug@50 nor in this group, so text-/media-based tools are at least helpful for learning from the past as well as supporting philosophical discourse. Plus, Doug got the hypertext and word processing part working 50 years ago, and we don't have those capabilities any more nowadays, so that leads to a grim indication in regard of our abilities to replicate that or even do the bigger things, with text being the old, primitive, easy one.
Comment Reference: https://www.youtube.com/watch?v=Kusuu4NDEfg&lc=UgxfN7bh3GzsTcdnrCN4AaABAg
1:07:10 Sure, Silicon Valley is a scheme to make money, just themed around technology, like it was or is themed around steel, oil, banking. It's unfortunate because the technology in itself, not so much seen from an investment opportunity, could do great things for humanity, as most relevant fundamental tech-related research is publicly funded anyway (privileges for AT&T and Xerox, funding for ARPANET, CERN, non-profit SRI) as those are risky long-term investments and not clever short-term investments for a good ROI bottom-line, and as this money isn't available any more without the cold war and space race, good luck with expecting funding for humanitarian endeavors from the Silicon Valley money machine.
Comment Reference: https://www.youtube.com/watch?v=Kusuu4NDEfg&lc=UgwItui6v-7JdSZFCmt4AaABAg
2:15:26 Heiner, you probably know about this stuff, the structure we would like to use is pretty much the xFiles, EDL, ZigZag, NOSQL or whatever else form/name it got over the time, so there can be many dimensions. But as you're talking about 17 and just words or just images, I guess you're referring to input sensory, and sure, the solution for text should be generalized to all kinds of media/data.