Re:A real pleasure to work (remotely) with LNZ
on
RIP: Leonard Zubkoff
·
· Score: 1
Wow, I'd almost forgotten about the Concept-LNZs. Very cool in their day. I didn't know Leonard that well back then, but appreciated his efforts and enjoyed the results of his work. Am sorry that I didn't follow his work more closely. Sad to see such a bright flame extinguished so soon...
Consider the market for "souped up" car chips. At a recent local car show I attended, one of the salesy folks mentioned swapping chips between the Audi TT and the VW Beetle (I'm pretty sure that was the pair, but its been a while), then swapping them back in order to pass inspection. Lots easier for the common bloke to increase the performance of the car instead of having to bore out the cylinders, etc. etc. etc.
The real question will be how stable can such small devices be in the conditions under the typical car hood (vibration is either an issue or a non-issue, probably not a middle ground there). With that little size involved, attacking the problem with hyper redundancy would be feasible. When enough redundant copies fail, pop, on comes the chip-is-brokey idiot light (or just overload meaning for the current "check engine" idiot light).
If you get some kind of assembly noids (nano androids/bots) on the chip, it could actually change itself to tune for your style of use/driving (who said these had to be digital? They could just as easily be analog, with all kinds of funky non-linear response characteristics). He! Everyone could have their own custom chip. Probably enough room on the chip to replicate the control circuity to have a copy for each of the drivers who uses the car on a regly basis.
Moving one atom/electron at a time is quite precise, but also very very slow to assemble any quantity of anything interesting. My personal guess is that these devices will first achieve practical application using some kind of massively parallel assembly mechanism that is chemical rather than mechanical. Either vacuum chamber deposition or perhaps some kind of repeated bathings of various soupy mixtures. Having nanobots roam the surface of a chip and create the circuitry is interesting, but potentially slow. Better to have 'em swarm over a finished board/chip and tweak/disable the brokey parts.
Interesting idea, but bandwidth is too ephemeral. How do I know if there will be enough updates in any particular month to make it worthwhile? How do I know that if they are swamped with subscribers that they'll upgrade to the requisite capacity to satisfy the contract?
Now, if they were offering to send me a CD every month. Heck, send me a CD-RW that I could erase and reuse, say 1 CD-RW per month, plus 1 CD-RW per quarter, containing everything from the beginning of the year until that CD-RW is burnt.... The monthly CD-RW would be for stuff changed since the previous month, the quarterly CD-RWs would be "rolling patches", now that I'd buy (if YDL or some PPC Linux did it).
A great idea! Only you don't necessarily have to false floor the entire room, just enough to put the "farm" on. Oh, wait, that's just a rack system.
I've always wanted to do it the other way round, and hang everything from the ceiling. Stabilization would be an issue, but just think how cool it'd be to have industrial looking room (paint the walls black with silver accents) and have the hardware hanging from mondo big chains. (mondo -> more load capacity than you need by a factor of at least 10.) A few industrial pulleys and you could crank the servers up and down as needed. Everything up on the ceiling when in use, lower only those boxes you need to when you need to actually put your hands on the hardware. You'll need a bit more cable to handle the distances involved when one end is raised and the other lowered. 'course doin' this in an attic only takes the hardware up to the hotest air in the summer. So I'd probably do it in the basement, which also means the ceiling won't be that high, so I might just punt on the pulley part of it, except maybe for effect.
I still like the raised floor idea, all I need is an excuse to raise the entire floor.;-)
Bickering proprietary 'nix vendors lost the chance to shut MicroSoft out. Now bickerin' open source 'nix lovers are doing the same. *BSD or Linux is irrelevant. "Its the interface, stupid."
Apple at least has a chance to push past that and get to the meat'n'taters of selling apps built on a real multitasking protected memory O/S. Building on 'nix clone was a biz decision, not a political one. MicroSofts unity of vision (at least as presented outside the company) gives it enormous advantage over what should be an insurmountable enemy of open source fanatics working their asses off for nary a penny. Except for Divide and Conquer. MicroSoft didn't have to divide the 'nix community, its quite capable of doin' that itself... of shootin' itself in the feet, kneecaps and elbows.
Here's hopin' that a strong market presence can bring some unity to the open source community, even if it is starting off with a few baby steps.
Re: the previous rants/trolls about Art and Programming soiling each other... check out Playing both against neither (pdf format) which gets into the fallacy of the "exclusive or" mentality which seems to be apparent in many of the messages post here (so far).
The world is much too interesting, even the parts, like programming languages, which we've created, to be so easily pigeoned into such small compartments.
Ok, so lots of messages about how it is the maintainers of the packages, not the particular packaging tools, which are the issue.
Unfortunately the solution is not open-source/gnu/whatever your fav. term is, friendly. With RH, Soose, SnackWare, YD, or whomever, in control, how can a regly joe developer fix a packaging bug? A bug in the code, yes, post your patches, etc. But packaging is a meta-data thing that is mostly immune to the usual bug fixing process. It is the delivery of the code, not the code itself, so its 'above the law' in a sense that the code itself is not. Sure, I could take a distro and just fix the prereqs or other packaging stupidity, but:
would I take the time if I weren't really sure the distro maker was going to take them?
would I feel this might be infringment on the distro itself in some way? (the feeling I'm going for here is elusive and hard to describe, but I hope this gives you the idea).
is the open source/gnu/whatever community going to give me props for doing it?
just what is my ROI for doing this?
I cannot install a fix to buggy/stupid package delivered to me on a CD. Once it ships, its a done deal. Which means getting packaging right the first time is even more important (though often undervalued and underappreciated). Electronic distros can be fixed more easily, but releasing a new copy of a package with the same version is bad juju, and how many folk wanna upgrade versions just to get a fixed package which they've already managed to get installed?
Re:As a certified electrician...
on
Wiring A New House?
·
· Score: 4, Insightful
My SO, a data wrangler from a local university says:
Great list. Esp. marking the strings, you'll really want to know what is where.
25 pair sounds cool, however it is a termination nightmare unless you're just going to use it specifically from point to point as your major house backbone and have it premade and preterminated for your situation. If you terminate it yourself you might be able to maintain 100Mb, but likely not and certainly not Gig. Unless there is a whole new 25 pair solution out there. Big conduit to run a plethora of different cables (except power of course) is a more general solution.
I would also add that you want more than one outlet per room. Depending on your builder, it could be something that you'd have to octopus up (or down) from a central point, or you might be able to run in between the floor boards to get opposite wall coverage (or three or four wall coverage depending on your layout.
One thing is for sure, there is never enough bandwidth, and you don't know what you'll want to run in the future.
The house I bought was already built, and has pathetic insulation which is not easily fixed. I'm planning on moving the "server farm" and "main hub" between the attic (in winter) and somewhere lower in the house in summer.
Consider also that where-ever you've picked for your central hub might become ideal for something else (kid's attic room, enclosed hot tub in basement, whatever), and since its so easy now, I would be inclined to run the conduit so you could have two (or maybe three, depending on the structure of your house) alternative spots.
It might also be cool to run a room or three with a few extra outlets for gaming/multimedia/"record"-making/"movie"-making parties, efforts, etc. (I'm thinking of "Duality" ((Lost the URL, darn!)).
What you want, really, is two jukeboxes (jukeboxii?). One that has real music CDs in it to be ripped, the other which can feed a CD writer. Depending on your writer, you'd need a gig or two of free space. Rip rip rip until you had an MP3 CD's worth of music, then write, write, write... Condensing music CDs down to MP3 cds, with a nice little back up of the music in the process. The software would let you either file the on-line MP3s somewhere or remove 'em to make room for the next batch.
I disagree, it was never a good concept. The web is distributed, WebRings only "value add" was to have a centralized place to "maintain" the ring. If page 'a' links to page 'b' which links to page 'c'... which links to page 'a' what do you need a central site for?
Slashdot Mirror
Wow, I'd almost forgotten about the Concept-LNZs. Very cool in their day. I didn't know Leonard that well back then, but appreciated his efforts and enjoyed the results of his work. Am sorry that I didn't follow his work more closely. Sad to see such a bright flame extinguished so soon...
The real question will be how stable can such small devices be in the conditions under the typical car hood (vibration is either an issue or a non-issue, probably not a middle ground there). With that little size involved, attacking the problem with hyper redundancy would be feasible. When enough redundant copies fail, pop, on comes the chip-is-brokey idiot light (or just overload meaning for the current "check engine" idiot light).
If you get some kind of assembly noids (nano androids/bots) on the chip, it could actually change itself to tune for your style of use/driving (who said these had to be digital? They could just as easily be analog, with all kinds of funky non-linear response characteristics). He! Everyone could have their own custom chip. Probably enough room on the chip to replicate the control circuity to have a copy for each of the drivers who uses the car on a regly basis.
"Control, control... you must learn control."
"Control, control... You must learn control!"
Now, if they were offering to send me a CD every month. Heck, send me a CD-RW that I could erase and reuse, say 1 CD-RW per month, plus 1 CD-RW per quarter, containing everything from the beginning of the year until that CD-RW is burnt.... The monthly CD-RW would be for stuff changed since the previous month, the quarterly CD-RWs would be "rolling patches", now that I'd buy (if YDL or some PPC Linux did it).
I've always wanted to do it the other way round, and hang everything from the ceiling. Stabilization would be an issue, but just think how cool it'd be to have industrial looking room (paint the walls black with silver accents) and have the hardware hanging from mondo big chains. (mondo -> more load capacity than you need by a factor of at least 10.) A few industrial pulleys and you could crank the servers up and down as needed. Everything up on the ceiling when in use, lower only those boxes you need to when you need to actually put your hands on the hardware. You'll need a bit more cable to handle the distances involved when one end is raised and the other lowered. 'course doin' this in an attic only takes the hardware up to the hotest air in the summer. So I'd probably do it in the basement, which also means the ceiling won't be that high, so I might just punt on the pulley part of it, except maybe for effect.
I still like the raised floor idea, all I need is an excuse to raise the entire floor. ;-)
Apple at least has a chance to push past that and get to the meat'n'taters of selling apps built on a real multitasking protected memory O/S. Building on 'nix clone was a biz decision, not a political one. MicroSofts unity of vision (at least as presented outside the company) gives it enormous advantage over what should be an insurmountable enemy of open source fanatics working their asses off for nary a penny. Except for Divide and Conquer. MicroSoft didn't have to divide the 'nix community, its quite capable of doin' that itself... of shootin' itself in the feet, kneecaps and elbows.
Here's hopin' that a strong market presence can bring some unity to the open source community, even if it is starting off with a few baby steps.
check out Playing both against neither (pdf format) which gets into the fallacy of the "exclusive or" mentality which seems to be apparent in many of the messages post here (so far).
The world is much too interesting, even the parts, like programming languages, which we've created, to be so easily pigeoned into such small compartments.
Unfortunately the solution is not open-source/gnu/whatever your fav. term is, friendly. With RH, Soose, SnackWare, YD, or whomever, in control, how can a regly joe developer fix a packaging bug? A bug in the code, yes, post your patches, etc. But packaging is a meta-data thing that is mostly immune to the usual bug fixing process. It is the delivery of the code, not the code itself, so its 'above the law' in a sense that the code itself is not. Sure, I could take a distro and just fix the prereqs or other packaging stupidity, but:
I cannot install a fix to buggy/stupid package delivered to me on a CD. Once it ships, its a done deal. Which means getting packaging right the first time is even more important (though often undervalued and underappreciated). Electronic distros can be fixed more easily, but releasing a new copy of a package with the same version is bad juju, and how many folk wanna upgrade versions just to get a fixed package which they've already managed to get installed?
Great list. Esp. marking the strings, you'll really want to know what is where.
25 pair sounds cool, however it is a termination nightmare unless you're just going to use it specifically from point to point as your major house backbone and have it premade and preterminated for your situation. If you terminate it yourself you might be able to maintain 100Mb, but likely not and certainly not Gig. Unless there is a whole new 25 pair solution out there. Big conduit to run a plethora of different cables (except power of course) is a more general solution.
I would also add that you want more than one outlet per room. Depending on your builder, it could be something that you'd have to octopus up (or down) from a central point, or you might be able to run in between the floor boards to get opposite wall coverage (or three or four wall coverage depending on your layout.
One thing is for sure, there is never enough bandwidth, and you don't know what you'll want to run in the future.
The house I bought was already built, and has pathetic insulation which is not easily fixed. I'm planning on moving the "server farm" and "main hub" between the attic (in winter) and somewhere lower in the house in summer.
Consider also that where-ever you've picked for your central hub might become ideal for something else (kid's attic room, enclosed hot tub in basement, whatever), and since its so easy now, I would be inclined to run the conduit so you could have two (or maybe three, depending on the structure of your house) alternative spots.
It might also be cool to run a room or three with a few extra outlets for gaming/multimedia/"record"-making/"movie"-making parties, efforts, etc. (I'm thinking of "Duality" ((Lost the URL, darn!)).
128 bit encryption and MAC security not withstanding.
What you want, really, is two jukeboxes (jukeboxii?). One that has real music CDs in it to be ripped, the other which can feed a CD writer. Depending on your writer, you'd need a gig or two of free space. Rip rip rip until you had an MP3 CD's worth of music, then write, write, write... Condensing music CDs down to MP3 cds, with a nice little back up of the music in the process. The software would let you either file the on-line MP3s somewhere or remove 'em to make room for the next batch.
I disagree, it was never a good concept. The web is distributed, WebRings only "value add" was to have a centralized place to "maintain" the ring. If page 'a' links to page 'b' which links to page 'c' ... which links to page 'a' what do you need a central site for?