At 6:27 PM -0600 5/29/13, ben wrote: I currently use Plastic record crates from Storables. Though I haven't bought any since about 2009. Take a look at the Expedite shelves from Ikea. That's the standard solution for Record Collectors. Zane -- | Zane H. Healy | UNIX Systems Administrator | | healyzh at aracnet.com | OpenVMS Enthusiast | | | Photographer | +----------------------------------+----------------------------+ | My flickr Photostream | | http://www.flickr.com/photos/33848088 at N03/ | | My Photography Website | | http://www.zanesphotography.com |

On Thu, 30 May 2013, ben wrote: The old phonograph here is a bit particleboard! ...The back panel only. ;) All the rest seems to be Cherry. -- Cory Smelosky http://gewt.net/ Personal stuff http://gimme-sympathy.org Experiments

On 05/30/2013 08:10 AM, Liam Proven wrote: ...with whale oil... -- Dave McGuire, AK4HZ New Kensington, PA

On 5/30/2013 7:47 AM, David Riley wrote: No, I want real wood products ... Sure I use the $29.99 stuff, but I want real wood! * cry, pout , stamp feet *. 30 years ago they had good products, now it is the same any where. So all you IKEA people send me your old Heavy rusty big iron products, and I'll return you a phone, that has the latest mindcraft 16 bit computer on it*. Ben. * Or the 1st google hit on computer.

I currently have 26 Ikea Flarke shelf units holding books, databooks and manuals which I think are no longer available and they are chipboard so have "some" wood in them. They do bend a bit, but why worry about what they are made of it is the contents that matter. I see the archive purists screaming about use of plastics (out gassing) etc, but in the case of records and sleeves you have a large percentage of plastics in the make up of the item. I see plastic coatings in use in a lot of items, we just have to get used to the degradation over time until someone finds a real cure or method of stabilizing random unknown plastics. Dave Caroline

On May 30, 2013, at 1:05 PM, Dave Caroline <dave.thearchivist at gmail.com> wrote: I found the Fl?rke to be a pretty decent bookshelf in college, not least because they cost as little as $15 each for a decent amount of capacity when they were on sale. Alas, they seem to have discontinued them. It should be pointed out that while they stood up to just about any amount of books I wanted to put on them, they did not withstand moves very well. They tended to prefer a non-right parallelogram shape, and of course once chipboard does that there's really no going back. I have rather little problem with plastics in my bookshelves. Plastics in a lining for my raised vegetable bed is another matter, since a) we will be eating what comes out of it and b) the plastic in a raised bed gets warm enough that I would be genuinely concerned about the leaching of various estrogen analogs into my food. - Dave

On Thu, 30 May 2013, David Riley wrote: ...*Third* time?! How much did you pay for that garbage? -- Cory Smelosky http://gewt.net/ Personal stuff http://gimme-sympathy.org Experiments

On 29/05/13 10:11 PM, Dave McGuire wrote: An SS1 would be enough for DNS. Heck, you probably wouldn't even notice if it were a Sun 3 or a MicroVAX. But the power bill.... :( Yep they are great. And can run quite recent (current?) NetBSD, and even Linux, for example. --Toby

Even that, provided you stick to the parts of the Web that could have existed back then (ie, avoid all the stupid "we're damn well going to look slick no matter how much it impairs usability" crap). I use lynx, often on a '20 - but to me, the Web is generally unpleasant and difficult to use, and when I do use it it is never for anything that I can't just shrug off with "oh well, too bad" if it doesn't work. (Exception: when I'm paid to, ie, for work - but then it's on a work-provided, -administered, and -maintained system, usually a peecee of one stripe or another, not relevant to this list, and still relatively difficult and unpleasant.) Compared to a SPARCstation of the SS2, IPX, etc era? Yes. They're throwaway devices in that sense - but they're priced like it, too. /~\ The ASCII Mouse \ / Ribbon Campaign X Against HTML mouse at rodents-montreal.org / \ Email! 7D C8 61 52 5D E7 2D 39 4E F1 31 3E E8 B3 27 4B

On Thu, 30 May 2013, Dave McGuire wrote: Agreed. GNU/Linux likes to be a bit...big. -- Cory Smelosky http://gewt.net/ Personal stuff http://gimme-sympathy.org Experiments

On 30 May 2013, at 15:32, "Dave McGuire" <mcguire at neurotica.com> wrote: The GNU/Linux usage was because I use some lightweight-linux distros that contain nothing GNU project. True. Server stuff is fairly lean. The kernel is still getting pretty damn big, though.

On 30/05/13 4:38 PM, Cory Smelosky wrote: Kernel 3.7.10 on this system (x86/32 bit) is "2495k kernel code". On the package bloat issue, well, I run Gentoo. --Toby

On 31 May 2013, at 01:11, "Zane H. Healy" <healyzh at aracnet.com> wrote: Alpine Linux. uClibc and busy box OOTB.

On Thu, 30 May 2013, Tothwolf wrote: I'll still prefer 4.3BSD...which can happily fit on smaller than even an RP06. ;) -- Cory Smelosky http://gewt.net/ Personal stuff http://gimme-sympathy.org Experiments

On 05/30/2013 07:00 PM, Tom Uban wrote: Progress! -- Dave McGuire, AK4HZ New Kensington, PA

On Sat, 1 Jun 2013, Liam Proven wrote: It isn't quite what I'm looking for. I've yet to find a currently maintained Linux distribution that does what I need and isn't bloated in userspace. I'm looking more for what a Linux distribution once was in the late 1990s, with full (advanced) networking support, plus normal network services, and still have a small userspace footprint (and low memory usage). A GUI is non-essential and for both space and security reasons (note all the recent X and X-libraries security bulletins) and I'd prefer to be able to leave it out completely in many applications. Are there /any/ currently maintained Linux distributions out there these days that are geared towards embedded use where they don't blindly assume multiple GB of memory and drive space? Surely I'm not the only one looking for something like this? The lack of i386 support with TC isn't very encouraging, either (they state i486 as a minimum, which is the same as modern Debian, et al.) I have boxes upon boxes of i386 stuff and I'd hate to at some point have to create a separately maintained fork of the Linux kernel just to keep them supported (everything from Intel 80386SX-16 to AMD 386DX-40, with and without FPUs). The idea of using either an obsolete (and likely unmaintained) Linux kernel or MS-DOS on them isn't terribly appealing.

Maybe. In particular, if you want 386 support, you'll be stuck with older versions. (Mind you, I personally don't see anything wrong with that; I run 1.4T on those of my machines which don't need something newer for hardware-support reasons.) Troll indeed. :-) I'd say it _once was_ the reference etc. It's been headed downhill for a long time, though; I now see it as trying to take on Linux at its own game (desktop market share), and, of course, losing - and, in the process, alienating the user base its former strengths attracted. /~\ The ASCII Mouse \ / Ribbon Campaign X Against HTML mouse at rodents-montreal.org / \ Email! 7D C8 61 52 5D E7 2D 39 4E F1 31 3E E8 B3 27 4B

On 2 June 2013 08:59, Mouse <mouse at rodents-montreal.org> wrote: That's not NetBSD. If anything - and I think it's still a very unfair description - then what you are describing is the significantly different FreeBSD. Especially PC-BSD or GhostBSD. But "bare" FreeBSD, without X, is still pretty lean & minimal - enough so that I've never managed to bootstrap from a core install to a working, Internet-connected GUI on my oldest working Thinkpad, which runs BSD better than it does Linux. I don't have the Unix skills. I think it's much closer to what "Tothwolf" wants, rather than Linux, which is evolving in the direction of a full-function OS for modern hardware, with all that that implies. -- Liam Proven ? Profile: http://lproven.livejournal.com/profile Email: lproven at cix.co.uk ? GMail/G+/Twitter/Flickr/Facebook: lproven MSN: lproven at hotmail.com ? Skype/AIM/Yahoo/LinkedIn: liamproven Tel: +44 20-8685-0498 ? Cell: +44 7939-087884

On Sun, 2 Jun 2013 07:37:18 -0400 David Riley <fraveydank at gmail.com> wrote: Well. See. Thats how it goes. Time is not standing still. Things evolve. Things change. Things grow. Thats live. If software, be it an OS like NetBSD or applications / server daemons like apache, would not keep up with the general growth of things in the entire eco-system, it will die. If you run a current release of a software system, you have to expect that it needs at least a halfway current hardware to run on. If you have really old hardware you have to accept that current software will run slow. If you want fast software on ancient hardware you have to use contemporary, i.e. ancient software. If you want "modern" features, you have to run modern software and this in turn requires "modern" hardware. You have to pay a price for the improved functionality. Either it will run slow on your 11/730 or you have to upgrade your hardware. Elsewise stay with older software that runs fast but lacks newish featurism. Nevertheless: NetBSD is still compact and efficient in comparision to most Linux distributions. It is also not tainted with some Linuxisms like systemd... -- \end{Jochen} \ref{http://www.unixag-kl.fh-kl.de/~jkunz/}

On 06/03/2013 02:01 PM, Liam Proven wrote: Do you mean "stuff designed now being sold five years from now"? If so, you'll likely see that. The example I cited (the only one of which I can speak from direct personal experience) is in medical equipment. That stuff is designed, FDA-approved, and sold for 10-20 years without modification. Your local Freecycle groups are astounding. I live in a VERY high-tech area and even I don't see stuff like that given away. (well, actually, thinking about that a bit more...There are "clueful" people around here; stuff like C2D machines are more likely to be found running Linux or NetBSD around here.) -Dave -- Dave McGuire, AK4HZ New Kensington, PA

On 06/03/2013 02:41 PM, Liam Proven wrote: Hey, whatever it takes! Very well-written as always, but this time I couldn't get past the first paragraph. I have great difficulty imagining a day when I could do my job on a tablet. The press (as a whole) really needs to understand that "what sells best in department stores" does not define the entirety of, or even a significant part of, society's computing activities. Everything is WAY too tainted by journalists' personal (and sometimes myopic) points of view, something that journalism is, well, sorta supposed to be about not doing. But either way...fantastic writing, as always. -Dave -- Dave McGuire, AK4HZ New Kensington, PA

On 3 June 2013 19:55, Dave McGuire <mcguire at neurotica.com> wrote: Then don't. Imagine it's a laptop. Imagine it's a dual-head desktop with 24" displays. Whatever it takes. For now, the old form-factors will stick around, but in a decade, if a "PC" is a flexible A4 tablet, the thickness and weight of thick card, which connects wirelessly to its peripherals and is driven by touch, speech and the view from its multiple cameras, I submit that few will prop it in a stand and drive it from a keyboard. But some will, and I am sure that for them it will be perfectly possible - possibly driving a tiled array of large screens which are the thickness (and approximate power-draw) of paper and similarly can be rolled up for transport or storage. Computers are shrinking and using less power. This is more or less a fact of life. They are not going to remain humming beige desk-side boxes; those are already in decline, have been for a few years now, and it's steepening. You're right. In businesses, increasingly, the beige boxes are being replaced by graphical terminals to OS instances running in VMs on remote hosts. Thank you. -- Liam Proven ? Profile: http://lproven.livejournal.com/profile Email: lproven at cix.co.uk ? GMail/G+/Twitter/Flickr/Facebook: lproven MSN: lproven at hotmail.com ? Skype/AIM/Yahoo/LinkedIn: liamproven Tel: +44 20-8685-0498 ? Cell: +44 7939-087884

On Mon, 3 Jun 2013, Dave McGuire wrote: https://www.youtube.com/watch?v=KyLqUf4cdwc 100% relevant. Tablets don't work for what I do...I traded 2 for a reason. THE CLOUD IS COMING!!!!!! -- Cory Smelosky http://gewt.net/ Personal stuff http://gimme-sympathy.org Experiments

On 4 June 2013 00:12, Dave McGuire <mcguire at neurotica.com> wrote: Yes! Yes, I really do. Huge big ones, two foot across. And I think the designers will /love/ it. No more separate graphics tablets and screens - they'll have several and they'll all be /both/ tablets /and/ screens, clustered or coupled, backed up live and continually to the cloud, able to draw on huge online libraries of pre-drawn components and parts for the drawings. And if they want a keyboard, stylus, puck, whatever, of /course/ they'll still be able to buy them. As I said, none of this stuff means keyboards are going to be banned or anything. Open your mind a bit. Anything and everything can have a keyboard today: Nintendo Wii, iPhone, iPad, Android device, Windows 8 tablet - all of them. They all accept keyboards. I know. I submit that actually I have more contact with this stuff than you do. Yup. Which means I see a lot of new technology, and the trends are all around us. Here, you might find this discussion on my blog relevant: http://liam-on-linux.livejournal.com/34896.html I have other mates who have views just like yours. I explain in considerable detail at that link why I think you're both being far too short-sighted. I do not own a tablet and I don't want a tablet. I am typing on a 1992 IBM Model M connected by an honest plain-jane PS/2 port, with an optical mouse and two LCDs. (Sadly my right CRT died last week & I had to junk it.) I don't use the things, but I watch everyone around me do it. You know what? My lodger (house-mate) is a professional developer. Mainly Perl, but C/C++, Scala, PHP, anything. He works on a MacBook Pro. But at home, he only watches movies on it. The computer he uses in the house? iPad. I do, more than you think. Please go read the blog comments. I'm not typing it all twice. It's 3AM here already. -- Liam Proven ? Profile: http://lproven.livejournal.com/profile Email: lproven at cix.co.uk ? GMail/G+/Twitter/Flickr/Facebook: lproven MSN: lproven at hotmail.com ? Skype/AIM/Yahoo/LinkedIn: liamproven Tel: +44 20-8685-0498 ? Cell: +44 7939-087884

On Tue, 4 Jun 2013, Dave McGuire wrote: "Everything will be pop-rivets and glue. Any day, now."

On Tue, 4 Jun 2013, Toby Thain wrote: and what "tools" do you use to disassemble a PS/2 model 50?

On Tue, 4 Jun 2013, ben wrote: Use renewable resources. Stick with Well Whale Oil.

On 4 June 2013 15:04, geneb <geneb at deltasoft.com> wrote: What about a 36" tablet? There is absolutely no reason why not. What I am talking about is not pocketable devices, nor even portable devices. What I am talking about is ubiquitously touch-sensitive screens, with the "smarts" of the PC built into the back of them. CPUs, RAM and Flash storage are all getting cheaper, smaller, cooler-running and less power-hungry. Soon, they will be built into monitors /even if you don't want to use them/, just like for well over a decade, some LCD monitors have speakers, integral USB hubs and things, /just in case,/ even though most people don't use them. In some of these, one of the functions of the USB circuitry is a full display adaptor - so if you have no free monitor ports, you just plug the screen into a spare USB2 port and the PC sees it as another monitor. And you can daisychain them. No, most people don't use it - but once it is essentially free to include, why not? Some monitors have a full ARM computer built in, so you can detach the screen and keep working on the move, running WinCE or Android. Those free built-in computers are probably going to grow up to be full-power PCs, once it's cheap enough to just add a full-power PC just to run the internal control circuitry. To talk about doing full-scale big-screen layout work on something the size of an iPad is just egregiously stupid. If you're doing it in the hopes of point-scoring in an argument, you just lost. -- Liam Proven ? Profile: http://lproven.livejournal.com/profile Email: lproven at cix.co.uk ? GMail/G+/Twitter/Flickr/Facebook: lproven MSN: lproven at hotmail.com ? Skype/AIM/Yahoo/LinkedIn: liamproven Tel: +44 20-8685-0498 ? Cell: +44 7939-087884

On 4 June 2013 16:48, geneb <geneb at deltasoft.com> wrote: [Sigh] You have no imagination left, old man, none. :?) Nobody is asking you to, as I keep saying, but you seem to be so fixated on hating the new and telling me I don't know shit that you're not actually reading what I'm writing. I *SAID*, repeatedly, that PCs will continue to support mice and keyboards and pens and pucks and so on for many years to come. Hint. I know many professional writers working on iPads these days, for portability reasons. They don't do it by poking the screen. They use a keyboard. Shock horror, keyboards still work with iPads and iPhones and Android and Windows Mobile and Blackberry 10. With *every touch interface there is*. Most of 'em can support pointing devices, too, and that support is set to improve, not get worse. Touch does not mean vertical. Touch does not mean phone. Touch does not mean tablet. Let go of your obsession, FFS, before you burst a blood vessel. Here, look at this. The video is under 2min, you can make it. http://hci.rwth-aachen.de/benddesk -- Liam Proven ? Profile: http://lproven.livejournal.com/profile Email: lproven at cix.co.uk ? GMail/G+/Twitter/Flickr/Facebook: lproven MSN: lproven at hotmail.com ? Skype/AIM/Yahoo/LinkedIn: liamproven Tel: +44 20-8685-0498 ? Cell: +44 7939-087884

On 06/04/2013 03:00 PM, Cory Smelosky wrote: Trying to ride on the coattails of the iPad, or telling people how to work? Both are popular these days. I tolerate neither. I dumped OS X (in part) because of that crap. And the eleven people running Windows 8 aren't happy about it either. (so Microsoft is bringing the "start" button back, so they'll once again be able to shut down their systems! ;)) -Dave -- Dave McGuire, AK4HZ New Kensington, PA

On 5 June 2013 00:30, Cory Smelosky <b4 at gewt.net> wrote: I've fallen victim to the "trying to be too clever" PEBCAK error myself, a good few times. (E.g. I spent *ages* trying to work out the command to tell my first Apple Newton to shut down. Eventually I consulted the manual. Press the on/off button, it said. I think I actually blushed.) I tried to learn from it. I don't always win. For Simplicity Computers, I argued hard against there being a shutdown option in the GUI. I said, just tell 'em to press the power switch. I lost. We added one. It's completely redundant. It's like a "sleep" option on a notebook. You don't need one. Close the lid. Sure, provide an option to turn this off, for instance if you want to be able to use the machine with its lid shut - but most people don't need it. I think this is something really quite profound right here, actually. There is a new model of human-computer interaction evolving, right now, today, in the world of phones and tablets. It's more direct, less mediated. Instead of right-clicking and double-clicking and rooting through menus, it's direct: press here, then press here, then press here. Occasionally, press and hold or press and drag, but it's an important element that people need not find that - that they can do everything without it. (This latter aspect is directly parallel to using a multibutton mouse on a classic Mac. My G/F about 2005 had a Blue & White G3 with Mac OS 9 on it, and a replacement mouse - a Logitech USB wheelmouse. She explained to me that it was a PC mouse from the local PC superstore, as hers died, and because it was a PC mouse, the right button and wheel didn't work, because they didn't on Macs. Yes they do, I said. No they don't, she said. So I checked. They didn't. No Logitech driver installed. So I downloaded it and installed it and showed her. Yes, they *do*, I said. Look, empty the bin with a right-click. Scroll through Finder windows or web pages with the wheel. She loved the wheel, and constantly forgot the right mouse button. That's what years of Mac use do to you. ;?) But the thing is, the OS did support right-clicks. Not super-widely, but they were there. If you didn't /have/ a right mouse button, if you had a single-button Apple mouse, then you could get at right-click functionality by just holding down Ctrl and left-clicking. Worked fine. Being used to having a RMB, I used it all the time, to occasional deep befuddlement of Mac users I was supporting. Sometimes, it's the quickest, easiest way to do stuff.) But, on late-era classic MacOS and on iOS, *you don't need to know this stuff.* You can do everything without it. This is something Windows tended to miss and which today Android sometimes misses. You have to cater for the users who only want the simplest, most basic interaction modes. On Windows, some things are quite hard to accomplish unless you right-click. Using Acorn RISC OS without middle and right clicks was impossible. You /needed/ them. Then I started working with Macs and thoroughly learned a GUI which removed that small complexity. To this day, I still occasionally have arguments with RISC OS users who maintain that the Acorn was is unequivocally /better./ Well, it isn't. I'm sorry, but it's not. Easier is better. Simpler is better. /So long as it doesn't limit you./ That's the key thing there in the rider. PC products have all too often had a "simple mode" and a "full mode" as a concession to ease of use or unskilled users. That's better than nothing, but it is a crutch and it is inferior. Make 'em able to do everything they want with the simplest interactions. I know iPhone users - skilled ones, very fast, with hundreds of apps - who didn't know about holding down icons to remove stuff, or double-pressing Home to access the recently-used-apps list. But they did not /need/ to know. /That/ is how good UIs are done. Offer extras, but don't demand them. Converse example: I had a chap on Google+ criticize Ubuntu's Unity desktop because, he said, you can't open a second copy of an app, all you can do is switch to it and press Ctrl-N for a New window - if that app supports that. Did you read the help screen, I asked. What help screen, he said. Ah. Well, you middle-click, I said. Oh! He goes. Well I gave up on Unity because you couldn't do stuff like that. So he could not do all he needed to, and got frustrated, and blamed the product. Actually, the product has the feature he needed, but he didn't go looking for it. Now, granted, that is a failure of design, but it's not the failure of design he thought it was. He thought the feature was missing (which is silly). It wasn't, but it was hidden under something he didn't try, and he didn't go looking. Either way, though, still a failure. Back to Windows 8. :?) The thing is, coming to a touch style interface from a desktop WIMP, traditional PC users seem to get very frustrated, and even smart people don't think of Googling an answer. I find that very odd. But more na?ve users, without the burden of expectation, find their way around simpler interfaces like the iPhone and iPad with considerable aplomb, and soon, it seems second nature to them. Well, the thing is, and I speak as someone who's worked in support for a quarter-century, even most PC users don't really understand how to work a desktop WIMP. Even people who use one 8+ hours a day for years on end. And there are several *billion* more potential users out there who don't know how to use a desktop WIMP than all those who do know put together. And if you use commercial OSes, then their vendors are going to follow the money. And that might well mean we lose our desktops and get glorified iPads. Those of us who use FOSS OSes are all right, of course. At least for now. At least as long as someone still wants to maintain the code. But I think it will prove to be like the DOS-to-Windows transition all over again. Those who love the old way will moan and grouch and complain, then gradually, they'll discover that actually, they get used to the new way, and after a while, hell, that they quite like it. And gradually, most of 'em will move across. And then there won't be anyone maintaining the FOSS desktops and they will fade away. It's like Max Planck said: ?A new scientific truth does not triumph by convincing opponents and making them see the light, but rather because its opponents eventually die, and a new generation grows up that is familiar with it.? -- Liam Proven ? Profile: http://lproven.livejournal.com/profile Email: lproven at cix.co.uk ? GMail/G+/Twitter/Flickr/Facebook: lproven MSN: lproven at hotmail.com ? Skype/AIM/Yahoo/LinkedIn: liamproven Tel: +44 20-8685-0498 ? Cell: +44 7939-087884

On Tue, 4 Jun 2013, Cory Smelosky wrote: Did you try a hammer?

On 6/4/2013 5:38 PM, Liam Proven wrote: Me... Fuck that and pull the plug! IT NOW OFF! If one has to gripe, the real problem is rebooting every update for consumer products. Ben.

On Tue, 4 Jun 2013, Dave McGuire wrote: If you're a Windows user, you might want to check out Classic Shell. http://www.classicshell.net - it's open source and works on every edition of Windows from Vista onward. g. -- Proud owner of F-15C 80-0007 http://www.f15sim.com - The only one of its kind. http://www.diy-cockpits.org/coll - Go Collimated or Go Home. Some people collect things for a hobby. Geeks collect hobbies. ScarletDME - The red hot Data Management Environment A Multi-Value database for the masses, not the classes. http://www.scarletdme.org - Get it _today_!

For those so OCD as to have clear space, howzbout we reinvent "the desktop"! For the rest of us, howzbout a 4x8 foot holographic projected screen. Just like the movies and TV have been showing for decades. It should be feasible in a few hundred more years. For the "zoom" issue of pointing resolution, I wonder how long it would take to get used to a 3D interface where preciseness of motion is controlled by the distace you finger is from it.

On Wed, 5 Jun 2013, Liam Proven wrote: What box? I'm talking about a 4x8 foot flat surface to put on sawhorses or file cabinets and call it a "Desk". Optionally, it can be a touch-screen, with detachable keyboard. For "up-scale", a frame like an old-style drafting board. Screw classroom projectors. I want the chalkboard replaced with a touch-screen that size. Include a printer - I want students to listen, not transcribe. Can the quality be brought up enough to have a combination display and window? Without compromising either. Howzbout display panels instead of drywall? I'll have some difficulty getting used to the future, but I am looking forward to bigger and better displays. -- Grumpy Ol' Fred cisin at xenosoft.com

On 6/4/13 9:54 PM, Fred Cisin wrote: 3M developed something very nearly a combination of touchscreen and display projector in the late '90s. I was working for the applied archaeology labs at University of Texas and we got a demo/market-test model. It was one of those things that everybody went "Ooooh SHINEY", got all excited, then had zero use for. Accordingly, my memory of it is very hazy. Mostly what I remember was that the Windows-only software was a clusterf**k to use and nobody would sit still to get trained to it, so the thing sat in the corner. It looked like a big whiteboard on wheels, connected via serial(?) cable to the computer. The projector displayed the Windows desktop on the board and there were special markers to draw with. The board recorded marker strokes to something resembling a PowerPoint presentation. IF, that is, I remember correctly. Does anyone remember such a thing? Did 3M ever bring it out as a product? Doc

On 5 June 2013 03:54, Fred Cisin <cisin at xenosoft.com> wrote: OIC! Sorry, I thought by "desktop" you meant "a desktop computer", as in, a separate system unit, either vertical or horizontal, with separate I/O devices attached to it. Definitely agree with all that, sure, no problem. -- Liam Proven ? Profile: http://lproven.livejournal.com/profile Email: lproven at cix.co.uk ? GMail/G+/Twitter/Flickr/Facebook: lproven MSN: lproven at hotmail.com ? Skype/AIM/Yahoo/LinkedIn: liamproven Tel: +44 20-8685-0498 ? Cell: +44 7939-087884

On 05/06/2013 01:18, Liam Proven wrote: At a recent Computer Conservation Society talk in Manchester, England a researcher at Salford University said in essence we are already at that stage. What he actually said was that the conductors are now so small that quantum effects mean that only 80% (I think it was 80%) of the electrons that form a current travel within the conductor and if it were any smaller it would cease to be a conductor. I think he also said that the difference between a 1-bit and a 0-bit in modern RAM is about 19 electrons.... ... I am going to do some more digging but it looks as if recent changes in CPU architecture have really involved shuffling round things within the design and using bigger chips rather than reducing the size of the components -- Dave Wade G4UGM Illegitimi Non Carborundum

On 5 June 2013 00:56, Toby Thain <toby at telegraphics.com.au> wrote: Me too. (Well, except my twin 21" CRTs are no more, as the right one died. So I've downgraded to twin 19" LCDs. Not happy about the change, but the extra physical desk space is good. I might have room for 3 ? 19" LCDs now.) But, you know, most computers support external screens. Even iPads! Support for multiple monitors is commonplace now. On a laptop with a single display port, one can use something like this: http://www.samsung.com/my/consumer/pc-peripherals-printer/monitor/lcd-monit… At $JOB-1, I had a nettop as a thin client, with an LCD on its monitor port and (somewhat bizarrely, as it had an HDMI port as well) a USB display adaptor & a second, identical LCD on that. There was little to no noticeable display-performance difference between the two screens. If the nettop had been a tablet-format device, it could have laid on my desk as a handy mousemat-sized trackpad - all it needs is a bit of flexibility in the design of the OS. In other words, the easiest sort of thing to change, really. -- Liam Proven ? Profile: http://lproven.livejournal.com/profile Email: lproven at cix.co.uk ? GMail/G+/Twitter/Flickr/Facebook: lproven MSN: lproven at hotmail.com ? Skype/AIM/Yahoo/LinkedIn: liamproven Tel: +44 20-8685-0498 ? Cell: +44 7939-087884

On Mon, Jun 3, 2013 at 2:09 PM, Dave McGuire <mcguire at neurotica.com> wrote: Here in Columbus, P4 machines have been skipware for some time, and I'm starting to see complete, working Core 2 Duo boxes offered for free. No quad-core yet. As for running UNIX/Linux on old Intel hardware, I lament the loss of 386 support, but the oldest desktop I've run in quite some time is a 486 (I'm not counting running DOS on 8-bit and 16-bit boxes - those never ran Linux and only could run a sliver of UNIX spectrum). I do still have a few 486s knocking about - they tend to max out at 16MB and have been limited in scope as a result of lack of memory more than anything else. I think the "weakest" machine of this kind I still have set up is a PS/2-E (IBM 9533 w/486SLC, 16MB RAM shared between code and video - 12MB/4MB) running RedHat 5.something, and the only reason for keeping it around is the quad PCMCIA ISA board in it. I built it up in 2000 as a poor-man's PIX (loaded with four 10/100 cards) as a dev box for the company I worked for at the time. We had a dual-PIX for production (redundancy/failover) but the devs learned the hard way when they tried developing in a single network space using broadcast protocols that there was a lot to set up in a secure production environment (i.e., production deployments fell flat on their faces and the devs didn't know what ports had to be opened between what machines). The PS/2-E router was never meant to enforce security, but it did a good job of segmenting the machines on the network from each other, forcing the developers to a) pay attention to, and b) tell the Network Guy (me!) what their network requirements were. My trusty old 386-DX/40 that I purchased in Apr, 1992, for the express purpose of running Linux, is long gone. -ethan

On Mon, Jun 3, 2013 at 3:18 PM, Cory Smelosky <b4 at gewt.net> wrote: Four 10/100 PCMCIA cards on a single PCMCIA-ISA bridge board in the lone ISA slot on a PS/2-E. I never claimed the network segments were saturated. It was an in-house network for product development and carried very little traffic. -ethan

On 06/03/2013 03:44 PM, Ethan Dicks wrote: <obedient consumer> You need 10gigE! Anything else is OLD GARBAGE! </obedient consumer> -- Dave McGuire, AK4HZ New Kensington, PA

On Jun 3, 2013, at 1:51 PM, Jochen Kunz <jkunz at unixag-kl.fh-kl.de> wrote: I don't disagree. I'm saying it makes it unsuitable for my own needs, which admittedly shouldn't stop the NetBSD overlords from doing the things that need doing to make it efficient and useful on modern hardware. Essentially, to make it accomplish both goals at once would require a fork at this point because the userland is no longer efficient for older/smaller hardware. Or run modern software designed to run on older or smaller machines, which is usually targeted toward the embedded market. NetBSD has some of that, but the rest of the environment has gotten too big for the older PPCs. There are plenty of tiny OSes meant for embedded or vintage computing that I may turn my spare cycles towards. There's a really neat bare- bones RTOS-type one out there called AtomThreads that I started a ColdFire port to a while back. I should really pick it back up. Again, though, there is a middle ground. It's a much, MUCH smaller niche, though. That much is true. I still like NetBSD in principle, and it still runs a hell of a lot better on my 604 machines than Linux does. I think I just need to find a new horse to back for future dev, though I still want to finish my SWIM3 driver for NetBSD before I move onto something else. - Dave

See, that's one of the things I find broken about NetBSD: this wasn't true of NetBSD ten years ago. It mostly wasn't true of NetBSD five years ago. But, in recent years, NetBSD has gotten sloppy, no longer letting their old-hardware ports keep them honest with respect to doing things right instead of just hitting them with Moore's Law. And that's what's wrong - well, that's one of the things that's wrong - with "current software". "WordStar on a 4.077MHz 8086 could keep up with my typing. WinWord under Windows on a 300 MHz Pentium II can't." I saw that quote attributed to Seth Briedbart, of Briedbart Index fame (though I've never verified the authenticity of the attribution). It's also not really true; it amounts to painting all "current software" with the same brush, which is about as false (and as true) as any other such broad-strokes summary. That's a bit like saying that a wheelchair is fast compared to a tree sloth. ...instead, it's tainted with other Linuxisms like v3 of the GPL. /~\ The ASCII Mouse \ / Ribbon Campaign X Against HTML mouse at rodents-montreal.org / \ Email! 7D C8 61 52 5D E7 2D 39 4E F1 31 3E E8 B3 27 4B

On 3 June 2013 20:43, Fred Cisin <cisin at xenosoft.com> wrote: More to the point, run, in a cross-platform manner that works well on Linux, Mac and Windows plus phones, a web-based app that everyone can use to schedule meetings and perform online collaboration, using FOSS on the client and freeware online services. That is a rather big deal, actually. :?) Interesting usage of it. The thing is that Moore's Law stopped delivering 100% more CPU power every 18mth in about 2007 or so. Since then, we just got more cores for a few years, then incremental tweaks: the return of Hyperthreading, better cacheing, better branch prediction, improved MMX, etc. - small, incremental improvements. It's gaining maybe 10-15% every 18mth now. In its place, we have Koomey's Law, which seems to be delivering something like twice the power efficiency every 18th. Which is nice, and why I predict lots of thin, lightweight, wireless, battery-powered gizmos, but which are not massively more powerful than we have now - they're just talking constantly to vast back-end servers in cryogenic vaults, using massive parallelism and brute force to deliver something that looks and feels a bit like an intelligent computer that understands you. It won't, but it will feel sort of like it. True. -- Liam Proven ? Profile: http://lproven.livejournal.com/profile Email: lproven at cix.co.uk ? GMail/G+/Twitter/Flickr/Facebook: lproven MSN: lproven at hotmail.com ? Skype/AIM/Yahoo/LinkedIn: liamproven Tel: +44 20-8685-0498 ? Cell: +44 7939-087884

On 6/3/2013 7:17 PM, Fred Cisin wrote: Fred, the folks at my university can do one better. They send email with trivial messages in MS word format (bad), and actually attach two copies of the doc... one in a serif font and one in a sans serif font. I have no idea why. Funny, because when I was interviewing for some positions in some "big data" companies (Google, Bloomberg, etc), they specifically ask about these kinds of skills because for all the horsepower they have access to, there's always more data. -Matt

Does anyone read/study Don Knuth's "Art ..." (TAOCP) any more? Sorting using tape drives ... what fun! Memory, what memory ... don't need no stinkin' memory! On Mon, Jun 3, 2013 at 11:41 PM, Fred Cisin <cisin at xenosoft.com> wrote:

I wish that more were at least willing to look through it. 7 years ago, when they discontinued all of the "advanced" programming classes, I was re-assigned to be a librarian. I insisted on the library getting Knuth and K&R. The head librarian tried to veto them, "They are WAY TOO OLD! We should only get current stuff." I pushed it. Hard. We got them over her objections. I have always had a copy at home, a copy at my business, a copy at my college office. Now that I am RETIRED, maybe I'll eventually get rid of a few copies. NOT YET. On Tue, 4 Jun 2013, Chuck Guzis wrote: That's a great illustration of an extremely important concept, of adapting the sort to the data and hardware configuration, and not being pushed to look at a sort as some sort of sealed black box, based on assumptions of "normal" hardware and "random" data. I saw that particular fail most often with people who could not accept that the "best" sort was NOT best when dealing with other situations, particularly the situation where there already exists significant amount of partial existing order. -- Grumpy Ol' Fred cisin at xenosoft.com

On 6/4/2013 5:31 PM, Dave McGuire wrote: I got thinking about that, since they got rid of the "advanced" programing why would you be able to get any *real* text books? "????? for dummy's" seems to be today's library mind set. Ben.

In article <51AE9638.20607 at jetnet.ab.ca>, ben <bfranchuk at jetnet.ab.ca> writes: Chunks of the forthcoming volumes have been publishes as "fascicles"; were you aware of those? -- "The Direct3D Graphics Pipeline" free book <http://tinyurl.com/d3d-pipeline> The Computer Graphics Museum <http://computergraphicsmuseum.org> The Terminals Wiki <http://terminals.classiccmp.org> Legalize Adulthood! (my blog) <http://legalizeadulthood.wordpress.com>

On Wed, 5 Jun 2013, ben wrote: Unfortunately, I can benefit from e-readers due to having terrible eyesight. I do still like real hard-cover books though! -- Cory Smelosky http://gewt.net/ Personal stuff http://gimme-sympathy.org Experiments

Fred Cisin wrote: I'd be willing to pay even more than that for a K&R hardcover. Peace... Sridhar

Am 05.06.2013 22:25, schrieb Fred Cisin: Public libraries once in a while re-bind paperbacks into hardcovers in order to have the book survive a longer time. And, as already mentioned, they often discard them after some time because "that book is old, and hasn't been loaned out for quite some time". Then you have good chances to acquire hardcover copies of CS books for very cheap, and it also happens that this is not the "Word 98 for beginners" crap. I got Knuth's trilogy this way. -- Holger

On 06/05/2013 02:50 PM, Sridhar Ayengar wrote: K&R in hard cover... Unlikely. The most interesting version was one given to me in the 8.5x11 looseleaf format with Bell Telephone Labs covers and notes inside by Brian. I'll never give that one up as it was a gift to me by Brian. I also have the early "white cover" version as reference. Both date before 1980 by a bit as they were my early aids to learning C using TinyC on Z80-s100-CP/M and later BDS-C. If one seriously wants hard cover it's still possible to find book binders to make it so. The trick is to find a copy thats clean and printed on decent low acid paper. IF one is to bind a book longevity is one good reason to make the effort, However my DEC Library are a lot of those pulp printed paper covers and they are very useful and very worn as a result. I have a number of 50$ plus books in paperback, the lack of hard covers does not dilute the value of them as its their content that makes them worth the price. I've always said in case of fire open my room window and toss all the books out, I'll be out after that. Allison

On Thu, Jun 06, 2013 at 09:27:45AM -0300, Alexandre Souza wrote: [...] I don't. The Kindle is optimised for displaying ephemeral content rented from the Kindle Store at prices higher than hardbacks. It's utterly hopeless at reading PDFs, which is the usual format for freely-distributed technical documents. (FWIW, I use an iPad 3 for reading PDFs. The retina screen is really handy for picking out the fine details in circuit diagrams, etc.)

-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 On 06/06/13 13:27, Alexandre Souza wrote: I don't think your belief in data preservation using the Kindle is well justified at all if the company you get your e-books from is freely able to wipe your data. At the very least, I hope you didn't buy your paper-free books from Amazon. See http://stallman.org/amazon.html - -- Mateusz K.

On Jun 6, 2013, at 9:37 AM, "Alexandre Souza" <alexandre.tabajara at gmail.com> wrote: I like the Kindle app on my phone for the convenience factor, but our corporate overlords at Amazon have done some pretty objectionable things. The most amusing/chilling episode I can recall (which RMS doesn't mention in that article) was when they remotely deleted copies of several Orwell books including 1984 from customers who had paid for them. Legally, they were actually doing the correct thing because the party that had submitted it to the Kindle store didn't have distribution rights (we can get into the ethics of copyright law elsewhere), but it was still rather ominous. I mostly stick to the public domain versions of classic literature that Amazon has up for free, and I didn't have to pay a dime for the app on the phone. The only books I've paid for are electronic editions of the Song of Ice and Fire series, because I don't plan on buying the dead-tree version until it's done (or Martin dies) and I can get a snazzy box set. - Dave

On Thu, 6 Jun 2013, David Riley wrote: I bought one of the "ad supported" Kindles - and didn't bother to let it have access to my wireless network. :) Any books I buy that have DRM on them have it stripped off before it hits the Kindle. (So far, one title. :) ) A Kindle is great for reading novels, but I don't see it adding any value over a paper reference book. It can't meet the speed of navigating a paper book. :) BTW, if you want a wonderful source of DRM-free science fiction and fantasy books, I can't recommend Baen Books highly enough. g. -- Proud owner of F-15C 80-0007 http://www.f15sim.com - The only one of its kind. http://www.diy-cockpits.org/coll - Go Collimated or Go Home. Some people collect things for a hobby. Geeks collect hobbies. ScarletDME - The red hot Data Management Environment A Multi-Value database for the masses, not the classes. http://www.scarletdme.org - Get it _today_!

On Thu, 6 Jun 2013, David Riley wrote: Remotely deleting 1984...;) -- Cory Smelosky http://gewt.net/ Personal stuff http://gimme-sympathy.org Experiments

On 6/6/13 11:27 AM, "Cory Smelosky" <b4 at gewt.net> wrote: Yeah, when I read that I thought: if you made something like that up, no one would believe it....

On Thu, Jun 6, 2013 at 1:27 PM, Alexandre Souza <alexandre.tabajara at gmail.com> wrote: How many kindles do you have to buy to replicate getting a few fold out diagrams text pages etc out of a manual when servicing a device Paper therefore rules and I am a kindle free zone Dave Caroline

On Thu, 6 Jun 2013, Dave Caroline wrote: Several hundred I'd guess. -- Cory Smelosky http://gewt.net/ Personal stuff http://gimme-sympathy.org Experiments

I _much_ prefer books on paper. You can flip between pages quickly (somthing that you can't do on any bookreader type thing I've tried), you ccan have foldouts opened and read the reat of the book. You can have several books open at once and readable at once. And paper books do not need cheging, they don;t crash, and the publisher can't make them disapper from your bookshelves. So NO WAY am I going to replace a paper book with an ebook, and given the choice I wil lbuy the paper book even if it costs a lot more BUT... Information that is inconveniet to use is better than no informatio nat all. Recenely a household hardware shop chain in the UK (Robert Dyas in case it matters) were selling off a cheap E-book reader for \pounds 19.00. I bought one with the intention of experiemntign with the colour LCD display module. Of course I openeed it up, it contaisn a RockChip SoC, an SDRAM, a Flash memory device and not a lot more. The CPU in the SoC is not suprisingly an ARM of some flavour. But befoe I raided the dispaly from it, I decided to try it out. It will display most .pdf files I've thrown at at. And while readign a DEC printset on that size display involves a lot of scrolling around, it's preferenable, as I said, ot no scheamtics at all. So I went abck and bought a second one to raid the dispaly from, use as apares, etc.... Things I like about it is that it has real buttons, not a touchscreen. I loathe touchscreens. I also has the feature that it appears as a couple of USB mass storage devices (one for the itnernal memory, one for the micoSD card you cna slot in). Which means it cab be trivially mounted by the RPi. It does not depend o na particualr OS to be useful. -tony

On 06/06/2013 10:40 AM, ben wrote: Ranch style house, ground floor, window sill around 4ft, quantity, procedure is just save them. If they get beat up OK, wet is bad. There is about 80 linear ft of shelving filled with data manuals and system specific books and manuals. Also much of it is general electronics books and manuals from my work including 40+years of engineering notebooks. Add to that the ScFi collection, cook books, Technical Encyclopedias and other its a lot of paper. About 45 years accumulating them and every move they were packed and traveled with me. They represent knowledge and I know of nothing more valuable. While I can and have verified much of the DEC, S100 and CP/M stuff is on line its often easier to go to the dog eared book. The actual procedure established is close door to buy time, cat goes out first, then ... Allison

--- On Wed, 6/5/13, Fred Cisin <cisin at xenosoft.com> wrote: Some years ago, I happened to be talking with a rep some brief discussion at one time about releasing an edition that was leather bound with gilded edges. I think the context was for the 25th anniversary and playing up on the nickname it sometimes has as the C bible. BLS

On 06/06/2013 09:25 PM, John Foust wrote: Seriously? Do tell! (after you've found it, of course) I've spent some idle time looking, but not for many years now. -Dave -- Dave McGuire, AK4HZ New Kensington, PA

On Jun 7, 2013, at 5:57, David Griffith <dgriffi at cs.csubak.edu> wrote: He actually doesn't issue real checks anymore because of problems he's had with people posting unredacted pictures of the checks online and other people committing check fraud with said info. http://www-cs-faculty.stanford.edu/~uno/news08.html - Dave

Yet, he was quite willing to sign a copy of his first published work: "Potrzebie System Of Weights And Measures" in Mad magazine (1957?)

On 6/7/2013 7:05 PM, David Riley wrote: Somewhere, somewhen, someone did an experiment to see how bad their signature would have to be before the banks would not honor the check. I think they ended up PRINTING a totally different name and still the checks were honored. This was several years ago. I might be mistaken -- maybe this was credit card receipts. But the same principle applies. - Joe

On 6/4/2013 4:31 PM, Dave McGuire wrote: Love that. I picked up 10 copies of K&R remainered almost new out of the box one time (sorry have given all of the extra copies away) for a couple of bucks each because of the date thing. Also our buddy who did the princeton ascii art pictures, Harbison's book was also treated the same way (C: A reference manual). I am still more frosted by the ones who discard antiquarian books for lack of circulation from libraries than I am this nonsense @ bookstores. Jim

On 06/04/2013 02:04 PM, Fred Cisin wrote: I think I could still write a polyphase merge sort without consulting a book. When you chart out how some external sorts operate, it's like watching a a Rubik's cube solution. Pure poetry. Knuth's book was excellent. I'd used a McGraw-Hill or Prentice-Hall text before the Knuth text came out. I still have Knuth, but not the other book, whose title I've long forgotten. But there's also the idea of not gilding the lily with a too elaborate sort. Case in point--I needed to print a sorted symbol table on a compiler I was working on. This was a single-user system with on-line (dot matrix) printer. I chose a selection sort because each pass through the symbol table took less time than it did to print the preceding line. The only storage required was a flag in the symbol table (at that point, I was through with the names, so I just clobbered the first character in each name as I printed it). The symbol table always printed at full printer speed. The code was minuscule. We were using floppies for storage, so even printing to disk was fast enough. Sometimes the "stupid" algorithms are good enough. -Chuck

On Wed, 5 Jun 2013, Tony Duell wrote: The English Lit courses do not use the original editions. They use whatever copy of Hamlet was published within the last 5 years. (He does get reprinted a lot) For Shakespeare or Bible, You can (and until my action HAD TO) get an edition published this year. Now, if a teacher wants to stick with an older edition or commentary, they can, by parroting: "It is a CLASSIC WORK IN THE FIELD" When I waged that battle, I used lots of examples in literature - What's wrong with an old edition of "Moby Dick"? What's wrong with an old edition of "Gulliver's Travels"? I used to use astronomy, but the dissing of Pluto argued in their favor. So, I switched to What new has happened in history of the Roman Empire? Are the dinosaurs no longer extinct? Has the year 1066 changed much lately? But, the argument that was most effective was: "If you won't permit use of older editions, then you must increase the budget for the library!" Good. On your advice, I bought a copy of "Art Of Electronics". Good books are worthwhile even if for a field on which you have only a casual interest. -- Grumpy Ol' Fred cisin at xenosoft.com

On Thu, 6 Jun 2013, Tony Duell wrote: In my mind, this is more the sort of book that you pick up once you've decided to "get serious" and really jump into the electronics field. It isn't a beginner or intro book by any means, and I think for the uninitiated, something more along the lines of "Getting Started in Electronics" by Forrest M. Mims III would be a better introductory text before diving into "The Art of Electronics". While even more basic introductory books from Irwin Math such as "Wires and Watts", "Morse, Marconi, and You", and "Bits and Pieces" would be ideally suited for grade school students, these would also be valuable for anyone completely unfamiliar with basic electronics. (Note that Math's books are ~30 years old now, so they are very outdated when it comes to modern electronics, however the basics never really change.)

On Fri, 7 Jun 2013, Ian King wrote: The frequency of editions in educational textbooks is high enough that usually the only noticeable differences will be page numbering, a few illustrations (particularly the cover picture), and maybe the exercises at the end of the chapter. I think that they need "MINOR" edition numbering (2.1, 2.2, 2.3 V 3,4,5) Since the students are buying new, and therefore getting the "current" one, the teacher is stuck with using the newest. The publishers will send free "desk" copies to the teachers, so they aren't financially affected. The schools still discourage letting the teachers create their own materials. Some of that is due to a few abuses of some teachers overcharging for a "reader" published and available through local copy centers. One teacher was accused of such profiteering inspite of hiw reader being cheaper than any of the available textbooks. The school bookstore, which is often a franchise from a national conglomerate such as "Follet", wants a monopoly. But, they often refuse to carry readers by instructors, which means that those are cold by the local copy center. Some more of the resistance is that the textbook is listed in the "course outline" (which is now up to 20 pages, with sections on "content", "instructional methods", "Student Learning Outcomes", etc.) and THAT document is used for articulation - getting universities and grad schools to accept the course for transfer. "You claim to have taken the same course at your previous school before transferring here; give us a copy of the "Course Outline", and if it is virtually identical to ours, including the same textbook, then we will let you transfer the course." So, there are some strange, demented, external pressures preventing the teacher from selecting book based on what's best, Another battle that I fought was to permit bot "required" and "recommended" textbooks. Then, if I was forced to select an inappropriate textbook, I could list it as "recommended", and go ahead and use an appropriate one. For example, Rosen "Discrete Math" is a great book, but it's content was not a good fit for what I felt needed to be covered in "Computer Math" (I insisted on good coverage of 2's complement, binary/hex/octal, floating point storage format, etc. ) I was under a lot of pressure to "teach Microsoft C". I insisted that the course was "C Programming". THEREFORE, I used and showed a bunch of different com[pilers in class, and made the students (over screaming objections) do one program using a command line compiler (GCC, DeSmet "Personal C", or any other), and then do one program using an integrated development environment (such as Micrtosoft C), although I suggested that they use the Borland free download of Turbo C. "The course is about C! It is NOT about some specific comiler! Those who teach a compiler, instead of teaching programming and the language are ignorant suits." MY students could sit down with whatever compiler was available. I told my students that if they intended to ever use C for anything, that they should get a copy of K&R, but that they could probably get through the class with a used copy of Deitel or Waite/Prata/Martin -- Grumpy Ol' Fred cisin at xenosoft.com (retired 2 weeks)

On Jun 7, 2013, at 17:28, Ian King <IanK at vulcan.com> wrote: I have always HATED highlighting. It's never been a useful method of learning for me, and I had enough teachers throughout middle and high school attempt to force it on me that it's left quite a bitter taste in my mouth. Taking notes on index cards has a similar reaction for the same reason. And there's nothing I despise more than having to wade through someone ELSE'S highlighting/notes, because even when they're not the product of a diseased or deficient mind, it's still someone else's thought process, which is necessarily alien. On the other hand, I've found taking notes in a notebook to be very useful. I never, ever refer back to the notes unless it's months later, but the act of writing it out helps me remember because I remember when I DO things better than when I HEAR things. It also doesn't mar an otherwise perfectly fine book. - Dave

Reminds me of a course I nearly ended up on, but avoided just in time... The lecturer showed me the previous year's exam papers. There was a question on the interface between the 80386 CPU and 80387 floating point coprocessor (These were current devices at the tiem). I said something like 'That's trivial, just let me see the data sheets', only to be told that uyou didn't get the data sheets in the esam,you were expected to learn them. My first though was that no engineer _ever_ learns data sheets, and I've yet to meet a designer who doens't have half a dozen databooks ope around him while he's working. I pointed out how silly that was and siad that if I was teachign the course I would lecture on say the 80386/80387 but set an exam question on the 68020/68881 interface. Of coruse all data books would be availalbe. That way, the successsful students would be the ones who actually understood the subject and could read a databook and extrace the information they needed. A much more valuable skill than knowign a particular device. I tried to learn C from a variety of books and got nowhere. I bought K&R and it all made sense within an hour. Yes, it's expensive, and it's not a thick book, but it's well worth the money... -tony

On Fri, 7 Jun 2013, Dave McGuire wrote: Last year I bought a copy of the ANSI version. Turns out it was the green cover, "not for sale outside of asia" copy. Same content though. Funny thing was, when I put it on my bookshelf, I discovered I already had a copy. :) g. -- Proud owner of F-15C 80-0007 http://www.f15sim.com - The only one of its kind. http://www.diy-cockpits.org/coll - Go Collimated or Go Home. Some people collect things for a hobby. Geeks collect hobbies. ScarletDME - The red hot Data Management Environment A Multi-Value database for the masses, not the classes. http://www.scarletdme.org - Get it _today_!

On Fri, 7 Jun 2013, Dave McGuire wrote: I just checked and the paper isn't that bad. It actually isn't THAT bad. I just checked again (couldn't remember the paper quality) and it turns out the one I already have is a K&R syntax, first edition (1978). I'd like to get a white cover ANSI version, but I'm not spending $44 on it. g. -- Proud owner of F-15C 80-0007 http://www.f15sim.com - The only one of its kind. http://www.diy-cockpits.org/coll - Go Collimated or Go Home. Some people collect things for a hobby. Geeks collect hobbies. ScarletDME - The red hot Data Management Environment A Multi-Value database for the masses, not the classes. http://www.scarletdme.org - Get it _today_!

On Fri, 7 Jun 2013, Dave McGuire wrote: Amazon is listing it at $45 for the paper copy, $29 for the Kindle edition. g. -- Proud owner of F-15C 80-0007 http://www.f15sim.com - The only one of its kind. http://www.diy-cockpits.org/coll - Go Collimated or Go Home. Some people collect things for a hobby. Geeks collect hobbies. ScarletDME - The red hot Data Management Environment A Multi-Value database for the masses, not the classes. http://www.scarletdme.org - Get it _today_!

On 07/06/13 11:26 PM, Dave McGuire wrote: That's not even the most egregious example. Norvig's Paradigms of Artificial Intelligence Programming - a book I am looking for, in paper form, but I'd take a good quality ebook: http://www.amazon.com/Paradigms-Artificial-Intelligence-Programming-Studies… The KINDLE edition is $79 for me! --T

On 06/07/2013 06:36 PM, Cory Smelosky wrote: Yes, the first edition. My apologies for my imprecise language; it has been a bit of a day. -Dave -- Dave McGuire, AK4HZ New Kensington, PA

On Fri, Jun 07, 2013 at 06:16:03PM -0400, Dave McGuire wrote: I think a huge part of C's success comes from how well-written that book is. The language itself (especially pre-ANSI C) is pretty undeserving -- it's exactly the wrong kind of "low-level" language, i.e. it's designed to be easy for the compiler to compile, but it's clumsy about giving you control of bits and bytes. So you get the non-portability of assembly code without the speed. (Code that compiles OK but doesn't *work* is still non-portable.) If they'd made it possible to declare variables/pointers/bitfields with a specific word size and bit/byte order, and leave it to the compiler to take care of ntohs() etc. w/o making it your problem on every single ref, it would have saved a lot of people a lot of gray/missing hair... John Wilson D Bit

On 4 June 2013 00:17, Fred Cisin <cisin at xenosoft.com> wrote: It certainly was, and I miss it already. But Koomey's Law promises a different kind of changed computing experience, and I am really quite interested to see where that leads... ... At the same time that I want to get the hell out of IT as a trade, because it's no fun anymore. But writing about it is. -- Liam Proven ? Profile: http://lproven.livejournal.com/profile Email: lproven at cix.co.uk ? GMail/G+/Twitter/Flickr/Facebook: lproven MSN: lproven at hotmail.com ? Skype/AIM/Yahoo/LinkedIn: liamproven Tel: +44 20-8685-0498 ? Cell: +44 7939-087884

On Tue, 4 Jun 2013, Dave McGuire wrote: You'd think they'd like to actually SOLVE the problem... Unfortunately...I doubt it stayed in their heads -- Cory Smelosky http://gewt.net/ Personal stuff http://gimme-sympathy.org Experiments

On 06/04/2013 05:22 PM, Fred Cisin wrote: These people are SUITS. And as such, they suck. -Dave -- Dave McGuire, AK4HZ New Kensington, PA

On Tue, 5 Jun 2013, Fred Cisin wrote: I know some CS students with actual talent! I also know some that don't know anything whatsoever. (I expected someone with an interest in CS to know what a cascade failure was...did I expect too much?) -- Cory Smelosky http://gewt.net/ Personal stuff http://gimme-sympathy.org Experiments

On 08/06/13 9:32 PM, Alexander Schreiber wrote: Indeed, and this is why "C is fast" is effectively a fallacy. It is very easy to write a slow C program, and lots of people do it. It's not clear how an average developer gets exposed to the concepts of time/space complexity outside a CS degree. --T

So what can you drive *Up Stairs*.

A cow? (Sorry, could not resist. Just try driving her down again, though...) -tony

On Mon, 3 Jun 2013 14:42:27 -0400 (EDT) Mouse <mouse at Rodents-Montreal.ORG> wrote: I call this pragmatic. You can spend endless brain cycles in microoptimizations or just give it the CPU cycles. Yes, it is a bit sad that this happens. But beeing a volunteer project, the amount of brain cycles that the NetBSD project can spend on platforms like VAX or HP300 is quite limited. I know that you know that the GPL v3 is _very_ much disliked by the NetBSD developers. The problem is that there are some components of NetBSD like GCC are taken from external sources. These sources decided to go GPL v3. So NetBSD has no real choice. In case of GCC: Stay at the last GPL v2 version is no solution and clang / llvm does not support all the CPU architectures that GCC and NetBSD support... So we have to swallow the bitter pill and accept third party components to be GPL v3 infected. -- \end{Jochen} \ref{http://www.unixag-kl.fh-kl.de/~jkunz/}

On Sun, 2 Jun 2013, Jochen Kunz wrote: Meh, I think not. I'm trying to cut down on the number of different platforms I have to maintain, not add another. ippool also wouldn't really be a suitable replacement for the way I use ipset, and the idea of trying to use pf in place of ebtables and iptables for bridge-mode firewalling is very unappealing. In fact, if I had to switch to something else, I'd probably go with OpenBSD (which I already use for other stuff) over NetBSD given OpenBSD's hardware support. I plan to stick with Linux for most embedded applications, but the widespread assumption by many userspace "developers" that everyone has 4GB or 8GB of ram and a 2TB hard drive isn't helping anyone except help hardware companies sell newer gear year after year. Right now there seems to be a gap in the market for a Linux distribution tailored for small/embedded systems which is also easy to maintain/update with a halfway decent package management system. I have nothing against pkgsrc, Ports, etc, but when you are dealing with lots of different systems, and especially those with limited CPU/memory/storage, compiling everything from scratch (as with Gentoo) just isn't a viable solution.

On 02/06/13 6:40 AM, Mouse wrote: You don't need to "compile everything from scratch" with Gentoo. --T

On Sun, 2 Jun 2013, Liam Proven wrote: The Linux kernel is still only as big as you compile it. It can still be built pretty small. The issue of dropping i386 is a different matter -- some developers were "inconvenienced" by some of the old code, and rather than redesign it so that they could handle some newer stuff while still supporting the i386, they simply removed that code entirely. Where is that goal defined, and for which "project"? There is no reason why a modular Linux distribution such as Debian can't support both cases, and handle both very well. The issue I've witnessed first hand is that many (not all) such "developers" don't want to support what they consider "old obsolete hardware" (look at Firefox, for example). The general attitude seems to be more "that's old hardware, just go buy a new PC". That doesn't work when you have an embedded application where you aren't dealing with a "PC". Some examples would be a single board computer used in an information kiosk (where you would want a small GUI) or a CompactPCI based application with lots of network interfaces (where you don't want a GUI at all). Neither of those can be replaced inexpensively (and certainly not with a cheap "PC") and since the issue isn't CPU-performance related, there is (or at least should be) no reason to replace them. Not even decades old, some as little as 5 years old (or less) can't handle more than 256MB or 512MB of ram. Embedded boards generally aren't marketed for applications that require multi-GB of memory. Many of these (fanless) boards are designed to remain in service for decades.

On 2 June 2013 18:34, Cory Smelosky <b4 at gewt.net> wrote: No, but Linus doesn't like to be tied down or held back. That's a deliberate policy of his. No, I don't really understand it, either, nor share it. The thing that hath been, it is that which shall be; and that which is done is that which shall be done: and there is no new thing under the sun. -- Liam Proven ? Profile: http://lproven.livejournal.com/profile Email: lproven at cix.co.uk ? GMail/G+/Twitter/Flickr/Facebook: lproven MSN: lproven at hotmail.com ? Skype/AIM/Yahoo/LinkedIn: liamproven Tel: +44 20-8685-0498 ? Cell: +44 7939-087884

On Mon, 3 Jun 2013, Sean Conner wrote: Good point. What'd be the middle-ground? ;) -- Cory Smelosky http://gewt.net/ Personal stuff http://gimme-sympathy.org Experiments

On 3 June 2013 17:05, Andrew Hoerter <amh at pobox.com> wrote: Heh. Yes indeed. I knew Geoff Chappell slightly online, the main uncoverer of the obfuscated code that lost MS the lawsuit and won DR hundreds of millions in damages. -- Liam Proven ? Profile: http://lproven.livejournal.com/profile Email: lproven at cix.co.uk ? GMail/G+/Twitter/Flickr/Facebook: lproven MSN: lproven at hotmail.com ? Skype/AIM/Yahoo/LinkedIn: liamproven Tel: +44 20-8685-0498 ? Cell: +44 7939-087884

On 2 June 2013 17:21, Tothwolf <tothwolf at concentric.net> wrote: I'm sure you're right, but nonetheless, I think my reasoning still stands. There are far better choices for embedded CPUs than an 80386 these days. For an old device, use an old Linux. Most of 'em won't be Internet-connected & so security fixes should not be an issue, surely? It's where it is heading. It's basically a desktop/server OS and always has been. Sure, but is it worth the additional developer effort for an edge case? And why shouldn't their software remain in service for decades, as well? It's not my field but I would have thought that something like QNX was a better fit for that kind of thing, anyway. -- Liam Proven ? Profile: http://lproven.livejournal.com/profile Email: lproven at cix.co.uk ? GMail/G+/Twitter/Flickr/Facebook: lproven MSN: lproven at hotmail.com ? Skype/AIM/Yahoo/LinkedIn: liamproven Tel: +44 20-8685-0498 ? Cell: +44 7939-087884

On 06/03/2013 08:53 AM, Liam Proven wrote: This is one of those rare occasions in which you and I agree. There's nothing good about the 80386 (as an architecture) for embedded use. We have very, very good processors now; there's no excuse for using that crap other than incompetence and kickbacks. (the same is true in other areas of computing as well) You'd think this, but many of them are connected, and that's on the rise. That's why I suggested that I expect someone will likely re-add i386 support to Linux. We'll see. QNX has a big part of the market, but it *sucks* compared to more modern OSs. -Dave -- Dave McGuire, AK4HZ New Kensington, PA

On Sun, 2 Jun 2013 11:21:58 -0500 (CDT) Tothwolf <tothwolf at concentric.net> wrote: NetBSD had (or still has, I didn't check) the same problem. The problem is that the i80386 lacks some functionality that all following i80x86 have. So the code is full of if (i80386) { ... } else { ... } or #ifdef i80386 ... #else ... #endif This creatas a mess, makes the code less readable, is a constant source of bugs and even hurts performance. The i80386 is ancient PeeCee shit. I don't see any point in keeping this legacy. At least: One idea was to create a special i80386 port of NetBSD that supported the i80386 and only the i80386. Quite simple, given the superbe multi platform support of NetBSD. But nobody found it woth the work to keep that old i80386 shit alive. -- \end{Jochen} \ref{http://www.unixag-kl.fh-kl.de/~jkunz/}