First, this isn't three vulnerabilities, it is TWENTY, addressed with three patches to make it look less severe. (And I don't really think this once-per-month patch cycle is to make adminsitrators' lives easier; I think it's to make Microsoft look better.)
Second, Microsoft has also increased the load on their servers by, oh, thirty times. While they have enough money to provision themselves with thirty times the incoming bandwidth to handle the huge burst of patch traffic once per month, at this point they don't appear to have actually DONE THIS. I am just barely able to get the Windows Update page to display at all, much less actually do anything useful like, say, download patches.
So, here I sit with a machine with twenty vulnerabilities, which they didn't tell me about all month to save face, and now that they HAVE told me, I can't patch because I can't reach their site.
I think the biggest thing about gigabit is that PCI isn't really fast enough to support it. You can shovel 133MB/second over a PCI bus, or 1064Mb.... very slightly more than a gigabit, but that's with NOTHING else happening on the bus. Generally, since the hard drive controller is also on the Southbridge, I think about the best you're going to get off most PCs, even very, very fast ones, is about 300 megabits sustained.
To really take advantage, you're going to need machines that run the network card off the Northbridge. Presumably, PCI-Express network cards will also keep up pretty easily. From what I can see, you're probably best to wait another year to eighteen months before upgrading; by then, PCI-X should be pretty common, and gigabit networking shouldn't be very expensive.
Note that I don't have any direct experience with gigabit: these are just back-of-the-envelope calculations. I could be completely off, so pay attention to replies.
I sold Amigas from 1987-1989, and our store sold both kinds... I would often send people over to the ST side of the store, depending on their needs. That said, I didn't like the ST very much; I KNEW the Amiga was a lot better, and it really annoyed me when ST people would insist their obviously inferior machine was the best.:-)
The ST was actually a very simple machine, in comparison to the Amiga. It was designed and assembled very quickly, and IMO, it showed. That simplicity gave the ST a big advantage early in its life, as NOBODY understood what the heck to do with a multitasking operating system at the time. (and the fact that the early AmigaOSes were pretty unstable didn't help much either:)).
The ST's two main areas of advantage were MIDI and desktop publishing. DTP was a really big new idea at the time... being able to lay out a page, graphically, and then print out what you could see and have it look the same but be in high resolution (not just screen resolution) was a BIG DEAL. They actually coined the WYSIWYG acronym at the time... "what you see is what you get". This is kind of funny to me now, why WOULDN'T you get what you see? But at the time, it was a big step forward. Anytime someone asked me about DTP, I pointed them at an ST. The ST had a very nice monochrome, high-resolution screen... it was nicer than a Mac for a LOT less money, and there were emulators that let you run most Mac software.
It also had a built-in MIDI controller, so for a long time I pushed musicians toward the ST also. I think there may still be some STs in production use for MIDI. And of course, the first true multiplayer games in the home were on the ST... Midi Maze was HUGELY popular at our company parties. You could hook up to 16 STs together by daisy-chaining them with MIDI cables. Midi Maze was a very simple game, but a very addicting taste of what multiplayer Quake would someday be like.
The sound was weak, though. I don't remember the details, but I think it was just a slightly-enhanced version of C64 sound. And the graphics were very simple; you had 320x200 in either 16 or 32 colors, 640x200 in 4 colors, or 640x400 in monochrome, and that was IT. Nothing else. The main CPU had to do all the work, there wasn't much of anything hidden away to take advantage of. This simplicity made it easy to program initially, but it meant the system didn't have much headroom.
The Amiga, on the other hand, was probably the single largest advance ever taken by 'home' computers. The Mac's big deal was a GUI, which was important... but the Amiga offered 4096 colors, a sprite engine, video processing (with overscan capabilities), the ability to have several separate screens at different resolution and color depth showing on the same monitor at the same time, incredible graphic flexibility (anything from 320x200x1 color up to about 680x450x4096 with smoke coming out of the video chip:) ) four-channel stereo sound, multitasking, and VAST expandability, all at once. It was actually the logical offshoot of the Atari 8-bit processors. It was kind of amusing -- the Commodore Amiga was the grown-up Atari 8-bit, and the Atari ST was the grown-up Commodore 64.
The Amiga was SO advanced, in fact, that nobody really knew what the heck to do with it for probably the first whole year... everyone was lost in the complexity, and of course, the 1.0 version of the operating system was really weak and crash-prone. But after that first year, things just kept improving and improving. Without a doubt, it was absolutely the most capable computer you could buy in overall terms for a number of years.
If Commodore had had a bloody clue, and had treated their genius tech staff with the respect and awe they deserved, there would probably still be Amigas being made today (I mean, for real, not just a fringe offshoot), and if Apple had owned this technology, they would probably be where Microsoft is today. It was that s
The advent of bitmap graphics at all was a big deal. In text mode, the smallest addressable unit is 'one character'. This means that an 80x24 screen takes 1920 bytes to represent. Processors are so fast these days that you could update that screen, sheesh, a hundred thousand times a second, probably. But in the days of kilohertz machines, that was quite a bit of data to push.
The early 8-bit home computers could do bitmap graphics, and in fact it was a big selling point.... "Game XYZ, fight monsters in actual bitmap graphics!' Check out Castle Wolfenstein on the Apple 2 emulators for an idea of what 'good graphics' once meant. I don't remember the resolution of those early screens anymore, but it was very low... certainly not higher than 320x200.
When the Mac shipped, computers really changed. Instead of a text OS with occasional, fully-focused graphical programs, the machine was so incredibly powerful (8mhz, 16 bit) that it could do graphics all the time...they could actually draw a user interface on a 512x384 screen and have time left over. That's 196,608 pixels. I don't know how many bits per pixel the first Mac used... I keep wanting to say "one", but I think I remember grays on those first Macs, so that might be wrong. If it WAS one bit per pixel, they could represent that screen in about 24k. That's still a lot of data to push around, compared with the 2k for a text screen, and could be as high as 196K if it was 8 bits/pixel. I'm pretty sure it wasn't that high... the first Mac had only 128k of RAM. Maybe it was just black/white.
They actually managed to get a fairly good GUI up on the 1Mhz C64 with GEOS, but it was the Mac that first showed the mainstream that it was even possible.
Everything after that has been about accelerating that basic idea. For a long time, neither the Mac nor the PC was really fast enough to animate the whole screen at once at a reasonable framerate. Games had to be very clever to work around this; even though they'd done a GUI on the 64, it was still very, very hard to animate a full screen on a PC. As I recall, that was mostly due to bus speed; the system simply couldn't shovel enough bits out to the graphics card over an ISA bus. The processor was more than capable, but the bus just wasn't up to it.
For the last 15 years, the whole evolution of computers has been about making graphics go faster. First there were Windows (2D) accelerators, then full motion video, which flopped as a concept, because it didn't make good games and didn't work very well. A number of years ago, we finally got to the point that pretty much every computer in the world can do very smooth full motion video, and nobody even noticed, the idea was that dead. Then 3D accelerators, then GPUs, then hardware T&L.... the driving force in PC development has been graphics.
Sometime in the last couple of years, PCs really hit a plateau; they've gotten fast enough to do practically anything we can think of, at least for now. We can generate, manipulate, and output graphics of unbelievable quality... and we're mostly pretty blase' about the whole thing.
I'll tell you, though, if I showed my desktop machine (Athlon 2800+, GeForce FX5950, dual 36gb Raptors in RAID-0, Audigy 2 Platinum, Klipsch 5.1 speakers) to my 15-year-old self, I'd fear for my life. In 1985, I'd have killed someone with a big smile on my face to own a machine like that.
Phew, I kinda went off on a tangent there. Getting back on track..... GUI means a very specific thing. If the OS can turn individual dots on and off, and draws the user interface that way, it's a GUI.
You know, the thought occurs to me..... if, as you say, 90% of the people are doing 'wrong' and only 10% are doing 'right'... well, I'd say that the definitions of 'right' and 'wrong' are pretty messed up.
In this case, I'd say that 'right' is being defined by about 0.0001% of the population, and seriously needs revisiting.
I am really thinking that maybe no money at all changed hands.
SCO stated that the deal was 'worth' millions, but they didn't say that they actually RECEIVED millions. And they announced it just a day or two AFTER they announced results for their prior quarter; I think they are hoping that they will sign some more licenses this quarter, and that three months from now, people won't realize that the EV1 deal was, like, a dollar.
I would paint this as a desperate gamble. If this is true, I think SCO has no more than a quarter or so left before their BS is truly exposed.... unless, of course, Microsoft gives some other company money to 'freely sign a license for the valuable SCO IP'.
Grim Fandango was one of the best adventures ever done. It had great graphics, great voices, tough but solvable (for the most part, anyway:)) puzzles, incredible creativity, and a truly superb storyline.
It also sold dismally. LucasArts lost their shirt on GF. And that was BEFORE the huge slump in per-title PC sales.
It's possible that gamers really are 'hungry for' this kind of title, but given how most titles are selling these days, that's an awfully big risk to take. They've been burned badly a number of times in providing exactly what gamers are asking for. We may want these things, but apparently there aren't enough of us that want them to make the projects profitable. LucasArts is literally cutting their losses.
It's a shame... I imagine for this kind of game to really see a renaissance, it'll have to be developed in a low-cost country, which would allow them to sell, say, 25K copies and still be profitable.
I'm still trying to figure out why my original comment was marked 'flamebait'.... I've been here since, geeze, forever, and I'm pretty sure this is the first Flamebait I've ever gotten. It was probably a little Offtopic, but Flamebait??
It was *intended* as a wry, witty observation, meant to spark a little thought, and I fear that the moderators kneejerked before they did so.
Think about this a minute: what is a dollar, but t politician's promise? There's nothing backing them. They're not a promise to pay anything. And the Fed has been on a crazy printing spree for the last few years, desperately trying to paper over the damage from the tech bubble (which WE are mostly feeling, regardless.) There's a reason the dollar has been dropping so fast over the last two years. The Dow may be at 10,000 now, but the measuring stick has changed.
You say 'it's the best we can do', but that just isn't the case. I am of the firm belief that floating currencies is a lot of what is behind the steady siphoning of wealth from the general public and into the hands of the very, very rich. Floating money allows for the rise of a speculative class, and currency speculations are zero-sum; if I win, you lose. In essence, currency traders extract a lot of value out of the economy while providing nothing in exchange; they are vultures. Commodity money prevents this, except in cases where governments are abusing their money supplies. (ie, increasing the currency base without increasing the reserves.)
Remember, what you have in your wallet is a politician's promise to pay nothing. I was trying to point this out, because the anti-counterfeiting stance struck me as amusing....it's okay if the government does it, but if private citizens do it, we'd have too much of it. I believe that ANY counterfeiting is too much, and that's pretty much what's going on every day.
Well, okay, I guess it's not quite counterfeiting. It's a 100% genuine promise to pay nothing.:-)
So basically, HP and the government are both trying to stop you from taking pieces of paper, putting ink on it, and declaring that it is 'money' and 'has value'.
This is essentially what the government does every day. They print, at next to zero cost, boatloads of worthless paper, foist it off on a captive population, and extract valuable goods and services in exchange.
It's a great racket. No wonder they're worried about competition.
The weird thing about this argument is that it can't be contradicted; if you insist that all people who don't watch television are very loud about that fact, then anyone who disputes this is sort of proving the point.
I think it's going to become a very strong meme. It makes sense on the surface, and ALL arguments, both pro and con, can be construed as data supporting it. I bet practically everyone will believe it within a year or two.
From what I have gathered, the SPEWS philosophy isn't just indifference to collateral damage (ie, 'civilian casualties'); they actively do this damage in order to try to force ISPs into changing their habits. And they are extremely difficult to both reach and reason with; you can post on a newsgroup and hope someone pays attention to your pleas.
I don't know if the actual newsgroup replies come from people who make decisions with SPEWS, but those replies are amazingly hostile. "Oh, you're blocked? That's because you're on a crummy ISP that allows spammers. You're on a contract and can't switch? Well, you'd better start calling your ISP, because the block on your addresses isn't going away until the spammer adjacent to you does, and maybe not then, because you're a whiner."
(ok, ok, that last part was a bit of hyperbole, but it's not that far off... check dejanews!)
Admittedly, they're not killing anyone, but the tactic of deliberately attacking people who are only tangentially related to your real target is often called 'terrorism'. The consequences here are far less serious, but the fundamental tactic remains the same.... someone is doing something you don't like, and so you hurt a whole lot of people to try to force them to stop. So I don't use SPEWS.
There are a number of other, much saner, blocklists available, and the advent of Bayesian filtering is a VERY big deal. I am personally using a combination of postfix, maildrop, SpamAssassin and bogofilter, and I get amazing results; I only started training about two weeks ago, and the spam I have to deal with has dropped by over 99%. I get 1 or 2 false negatives per day, and I have had only one false positive since I started using this system. It does take a little maintenance, but it's much less annoying and intrusive than the constant attention digging through spam takes.
It is possible, in other words, to do an exceptional job of stopping spam without contributing to a form of terrorism.
I've happily paid for a number of games where I didn't get anything physical. In all cases, of course, I got a file that will let me play the game forever (no need for a central server), but I don't have anything physical. And most shareware is that way.... or don't you register your shareware? And of course almost all Free Software comes without anything tangible... you can pay extra for a CD, but hardly anyone does in the era of broadband.
PC gaming is probably going to become mostly distributed over the Web. As other, smarter people have pointed out, it's a great way for a PC publisher to make money: with no middleman, they keep a much higher percentage. Since the market for PC games is shrinking so fast ANYWAY, the old tradeoff of accepting a lower percentage in order to make many more sales doesn't really work anymore.... going for the boutique market, instead of the mass market, seems the only likely way for them to survive.
In the electronic distribution field, I've seen three major models: Everquest, Valve, and Stardock. EQ and the other MMORPGs are a little different than anything else; they require a huge investment of servers and bandwidth to allow people to play the game, and a monthly subscription fee is the only way they could possibly pay for that. This model doesn't bother me at all....I'm a happy Second Life user, for instance.
Valve's method, on the other hand, involves spending a whole bunch of money on servers and bandwidth, but it's not for MY benefit, it's for THEIRS. They do this to make sure that I'm not stealing their software... there is no benefit to me WHATSOEVER. And there's no WAY they're going to get me to pay them for servers to make sure that I'm paying them!
Their games would work perfectly well on the old model of "sell it to me once and provide patches". They claim they'll be 'streaming content', but their content doesn't particularly need to be streamed. There are two main reasons for Steam; to prevent piracy, and to guarantee Valve a monthly revenue stream. They want to charge me monthly for features that benefit only them. Steam will not only cost me monthly, it will also provide me a service that is inferior to the one I've been getting for free. Because of that, I don't think it will fly.
If HL2 comes out in the standard "all you need is the CD to play" model, I'll buy it. If I'm required to use Steam, I will be much less likely to purchase, and there is NO WAY I will cough up any extra money to subscribe after purchasing it. Valve claims they "provide lots of extra content", but I just don't see that.... almost all their content comes from the mod community, FOR FREE. If I can't get that stuff for free anymore, I'll go play something else... it's not like I'm short on options.
Finally, there's Stardock's model, which I like a lot. I can buy an individual game if I want, or I can buy a subscription to everything they do. They have two subscriptions, one for their "serious" (Object Desktop) stuff, and one for their "fun" (Drengin Network) stuff. Anything I download during my subscription will continue to work even if I stop subscribing, which is critically important to me. If I still want to play the game I downloaded today ten years from now, it'll work fine (assuming the OS will run it, at least); there's no artificial barrier. They provide enough servers and bandwidth to provide me what I paid for; they're not building this complex copy-protection system and expecting ME to pay for it. I appreciate that they have no copy protection on their games... and I pay for it.
Ultimately, I think the EQ and the Stardock models will fly. I very strongly suspect that Steam is going to fail miserably: if HL2 is good enough, it may carry them for awhile, but I think ultimtaely the idea of charging customers for inconvenience is not workable.
I don't know what planet you're from, but on EARTH, we Linux admins have been scrambling just as desperately as Microsoft admins for the last year or so.
I've had a hypothesis for some time that the security flaw rate in Linux would decline over time and eventually approach zero, where Microsoft's would stay essentially constant. I believed this would happen because the Linux source was open and all the security holes would gradually be found and squashed, where the Microsoft source, being closed, wouldn't be as closely examined and would remain a fertile field for new exploits forever.
Well, in 2003, my pretty little hypothesis sure wasn't looking too good. I haven't actually compared numbers, but I felt like there were just as many bad critical bugs on Linux as there were on Microsoft. From my perception, the Linux rate rose, while the Microsoft rate dropped, which is exactly opposite what I was expecting.
I still believe that closed source is "fake" security, and that the only way to get REAL security is for everything to be open, but in terms of actual number of published exploits, both systems appear to be about equal at the moment.
And the standards to which Microsoft needs to be held are pretty much immaterial; only Microsoft can fix that code, where anyone can, in theory, fix bugs in OSS. Personally, I think we can use them as a yardstick, but we shouldn't be flinging mud.... very many more years like 2003, and they'll be flinging lots more of it back at us.
In 2003, OSS security sucked. I hope 2004 is better.
Their network quality is better than anything I've used except PacBell's.... when I was on PacBell DSL, it was very very low latency and really great. But I was one of the very first customers of PB when they lowered the price, and I heard horror stories about the later connections. Mine was always super, although the slow upload bugged me.
My biggest complaint with Speakeasy was that they had relatively high latency -- never packet loss, but much higher latency than I used to get with PacBell. They've been doing some major upgrades to their backbones, however, and over the last few months my average pings have dropped by half.... right this second, I have several hundred Call of Duty servers that are 50ms or under from me. (I don't have a filter set up for an exact count, sorry.)
I've troubleshot dozens of home connections over the last few years, and I can tell you with great assurance that if you want the best possible network connection for a reasonable price, Speakeasy is the way to go. Liberal AUP, awesome network, great support.... just an excellent, excellent company to deal with. And they like Linux!
The companies that are fundamentally built around selling you something that they can't actually provide are the ones complaining. A good provider, like Speakeasy, doesn't care what you do with your connection, because they are paying for enough upstream bandwidth to handle your traffic.
Speakeasy doesn't say 'unlimited', they sell you bandwidth, and you can do whatever you want with it. Run servers, do VPN, run Bittorrent 24/7 -- it's all good. It's your bandwidth, you paid for it. As long as it's legal and isn't disruptive to other users, Speakeasy is happy to have you as a customer. (ie, you can't DOS people, spam, or scan/attack networks you don't own/manage, but pretty much anything else goes.)
They're linux-friendly, can do either DHCP or static IPs, have good latency, essentially zero packet loss, and they're happy to HELP YOU share your network connection with your neighbors.
As far as I'm concerned, Speakeasy should be considered the Gold Standard in ISPs. Obviously, they can't reach everyplace cable does, but if you can get Speakeasy and aren't, you may be doing yourself a disservice. Yes, they're probably a little more expensive than your current provider, and you probably won't be able to download as fast as you sometimes can on cable, but you will always get the bandwidth you were promised, you'll get low latency, good support (although the web-based support is pretty slow about responding.... call them if you're in a hurry), and best of all, you'll never get The Letter.
Some local providers can be great, too. Sonic.net in Northern California was excellent when I was there five years ago, and my brother says they're still great now. But national providers, by and large, suck rocks.
BTW, my relationship with Speakeasy is strictly 'I send you money, you give me bandwidth.' Other than that, I'm not affiliated with them. I'm just a very happy customer.
Ok, so here's a couple of questions..... the piezoelectric effect is based on physical vibration. In essence, we are introducing moving parts into a power supply. This prompts the questions:
1. Do they wear out?
2. Will other motion or vibration cause voltage spikes or sags?
Well, it appears you and Bruce differ. This is what he says:
"UserLinux is intended to be a system for business people. Central to its design is a network of competing for-profit service providers, who perform engineering and support services for the system."
"Business people", not "end users in business". You have a pretty theory, but it doesn't match what the architect is actually saying.
Given his stated goals for the project, the name is completely wrong. How the heck is a marketer going to work with this? (and we're all marketers when we're trying to sell someone on a project.) It is a fundamentally confused message, and it's going to hurt this project. Badly.
Any given brand can own only one or two words in someone's head. Linux mostly owns "quality" and "free". (which is a very, very rare combination!) Adding User to Linux, but then changing the definition of User to mean Business means that the marketing/branding effort is largely wasted. They can't spend much time on "quality" and "free" because "user" is exactly wrong.
The closest analogy I can come up with is Nabisco deciding to launch a new line of premium cookies, something like the Pepperidge Farm line, and calling them "Saltine Cookies".
Likely ensuing debacle left as an exercise for the reader.
Sure, business users are users.... but not all users are in business.
UserLinux implies the set of ALL users, when Bruce really wanted the subset of BUSINESS users.
It is just a terribly bad name, and I believe it will rob the project of much of its potential success.
Think about it a minute.... if it WERE called BusinessLinux, and Bruce said "GNOME only".... how upset would you be? I'd be a little bummed, because I like KDE, but I'd respect the decision. From a business perspective, it makes good sense. But excluding KDE from a distro called UserLinux seems impossibly stupid.
Until this brouhaha started, I didn't fully understand that UserLinux is meant for business. I was paying peripheral attention to UL because I thought I might someday move to it when the kinks were worked out. I do Linux every day. I've been doing Linux for a long time. Yet, even near the center of my knowledge base, I was confused about the project's focus. What are management types going to think?
Thousands and thousands of times, consultants are going to have to explain, "No, no, UserLinux is meant for business."
(this is an almost verbatim copy of a post I made at Linux Weekly News, so if you've seen this before, my apologies.)
Bruce says: "UserLinux is intended to be a system for business people."
OK, that's great, but why on earth call it UserLinux then? Shouldn't it be BusinessLinux?
Names are important. UserLinux sounds like a Linux distro intended for end users. Someone like my Mom, not someone like HP. Bruce may be right about GNOME being a better solution for business. I will, however, bet nickels to dollars that much of the controversy is because people assume that a distro called UserLinux should be about, well, users, and that's KDE's main focus.
I have assumed ever since the initial announcement that UserLinux might end up being my distro of choice, and I was upset when I heard about KDE's exclusion. Now that I read further, I see I have no reason to be upset, because UserLinux isn't intended for me.
It wouldn't surprise me to see the whole project fail because of this fundamental naming problem. Is a distro called UserLinux even going to register on a CIO's radar?
BusinessLinux might have. I don't think UserLinux will.
It is SO lame that they refuse to let you try the download version for free. I'm interested in Xandros, but there is no way I'm going to pay for a Linux distro I haven't thoroughly run through its paces.
And I pay for my software if I like it. I have given RedHat over 3 grand (including 2500 for an RHCE course a couple years ago [and MAN am I pissed at them for invalidating that investment by abandoning the market that spawned them]), and I've given Mandrake about $300. If I switched to Xandros and liked it, I'd absolutely support them.
They are, in other words, losing any chance at my money because they're not giving away their product.
Sheesh, no wonder regular companies have so much trouble understanding Linux.:-)
That would be exceptionally difficult, because the SD format is so thin. Hard drive heads need to float over their platters, and that takes a certain amount of thickness. They also Do Not Like being bent or squeezed, and the SD format just doesn't have a lot of structural strength.
A format the same height and width might work, but it would probably need to be thicker.
Slashdot Mirror
First, this isn't three vulnerabilities, it is TWENTY, addressed with three patches to make it look less severe. (And I don't really think this once-per-month patch cycle is to make adminsitrators' lives easier; I think it's to make Microsoft look better.)
Second, Microsoft has also increased the load on their servers by, oh, thirty times. While they have enough money to provision themselves with thirty times the incoming bandwidth to handle the huge burst of patch traffic once per month, at this point they don't appear to have actually DONE THIS. I am just barely able to get the Windows Update page to display at all, much less actually do anything useful like, say, download patches.
So, here I sit with a machine with twenty vulnerabilities, which they didn't tell me about all month to save face, and now that they HAVE told me, I can't patch because I can't reach their site.
I think the biggest thing about gigabit is that PCI isn't really fast enough to support it. You can shovel 133MB/second over a PCI bus, or 1064Mb.... very slightly more than a gigabit, but that's with NOTHING else happening on the bus. Generally, since the hard drive controller is also on the Southbridge, I think about the best you're going to get off most PCs, even very, very fast ones, is about 300 megabits sustained.
To really take advantage, you're going to need machines that run the network card off the Northbridge. Presumably, PCI-Express network cards will also keep up pretty easily. From what I can see, you're probably best to wait another year to eighteen months before upgrading; by then, PCI-X should be pretty common, and gigabit networking shouldn't be very expensive.
Note that I don't have any direct experience with gigabit: these are just back-of-the-envelope calculations. I could be completely off, so pay attention to replies.
I sold Amigas from 1987-1989, and our store sold both kinds... I would often send people over to the ST side of the store, depending on their needs. That said, I didn't like the ST very much; I KNEW the Amiga was a lot better, and it really annoyed me when ST people would insist their obviously inferior machine was the best. :-)
:)).
... Midi Maze was HUGELY popular at our company parties. You could hook up to 16 STs together by daisy-chaining them with MIDI cables. Midi Maze was a very simple game, but a very addicting taste of what multiplayer Quake would someday be like.
:) ) four-channel stereo sound, multitasking, and VAST expandability, all at once. It was actually the logical offshoot of the Atari 8-bit processors. It was kind of amusing -- the Commodore Amiga was the grown-up Atari 8-bit, and the Atari ST was the grown-up Commodore 64.
The ST was actually a very simple machine, in comparison to the Amiga. It was designed and assembled very quickly, and IMO, it showed. That simplicity gave the ST a big advantage early in its life, as NOBODY understood what the heck to do with a multitasking operating system at the time. (and the fact that the early AmigaOSes were pretty unstable didn't help much either
The ST's two main areas of advantage were MIDI and desktop publishing. DTP was a really big new idea at the time... being able to lay out a page, graphically, and then print out what you could see and have it look the same but be in high resolution (not just screen resolution) was a BIG DEAL. They actually coined the WYSIWYG acronym at the time... "what you see is what you get". This is kind of funny to me now, why WOULDN'T you get what you see? But at the time, it was a big step forward. Anytime someone asked me about DTP, I pointed them at an ST. The ST had a very nice monochrome, high-resolution screen... it was nicer than a Mac for a LOT less money, and there were emulators that let you run most Mac software.
It also had a built-in MIDI controller, so for a long time I pushed musicians toward the ST also. I think there may still be some STs in production use for MIDI. And of course, the first true multiplayer games in the home were on the ST
The sound was weak, though. I don't remember the details, but I think it was just a slightly-enhanced version of C64 sound. And the graphics were very simple; you had 320x200 in either 16 or 32 colors, 640x200 in 4 colors, or 640x400 in monochrome, and that was IT. Nothing else. The main CPU had to do all the work, there wasn't much of anything hidden away to take advantage of. This simplicity made it easy to program initially, but it meant the system didn't have much headroom.
The Amiga, on the other hand, was probably the single largest advance ever taken by 'home' computers. The Mac's big deal was a GUI, which was important... but the Amiga offered 4096 colors, a sprite engine, video processing (with overscan capabilities), the ability to have several separate screens at different resolution and color depth showing on the same monitor at the same time, incredible graphic flexibility (anything from 320x200x1 color up to about 680x450x4096 with smoke coming out of the video chip
The Amiga was SO advanced, in fact, that nobody really knew what the heck to do with it for probably the first whole year... everyone was lost in the complexity, and of course, the 1.0 version of the operating system was really weak and crash-prone. But after that first year, things just kept improving and improving. Without a doubt, it was absolutely the most capable computer you could buy in overall terms for a number of years.
If Commodore had had a bloody clue, and had treated their genius tech staff with the respect and awe they deserved, there would probably still be Amigas being made today (I mean, for real, not just a fringe offshoot), and if Apple had owned this technology, they would probably be where Microsoft is today. It was that s
It was actually 7.14Mhz. I assume that NTSC is therefore 3.57Mhz.
The early 8-bit home computers could do bitmap graphics, and in fact it was a big selling point.... "Game XYZ, fight monsters in actual bitmap graphics!' Check out Castle Wolfenstein on the Apple 2 emulators for an idea of what 'good graphics' once meant. I don't remember the resolution of those early screens anymore, but it was very low... certainly not higher than 320x200.
When the Mac shipped, computers really changed. Instead of a text OS with occasional, fully-focused graphical programs, the machine was so incredibly powerful (8mhz, 16 bit) that it could do graphics all the time...they could actually draw a user interface on a 512x384 screen and have time left over. That's 196,608 pixels. I don't know how many bits per pixel the first Mac used... I keep wanting to say "one", but I think I remember grays on those first Macs, so that might be wrong. If it WAS one bit per pixel, they could represent that screen in about 24k. That's still a lot of data to push around, compared with the 2k for a text screen, and could be as high as 196K if it was 8 bits/pixel. I'm pretty sure it wasn't that high... the first Mac had only 128k of RAM. Maybe it was just black/white.
They actually managed to get a fairly good GUI up on the 1Mhz C64 with GEOS, but it was the Mac that first showed the mainstream that it was even possible.
Everything after that has been about accelerating that basic idea. For a long time, neither the Mac nor the PC was really fast enough to animate the whole screen at once at a reasonable framerate. Games had to be very clever to work around this; even though they'd done a GUI on the 64, it was still very, very hard to animate a full screen on a PC. As I recall, that was mostly due to bus speed; the system simply couldn't shovel enough bits out to the graphics card over an ISA bus. The processor was more than capable, but the bus just wasn't up to it.
For the last 15 years, the whole evolution of computers has been about making graphics go faster. First there were Windows (2D) accelerators, then full motion video, which flopped as a concept, because it didn't make good games and didn't work very well. A number of years ago, we finally got to the point that pretty much every computer in the world can do very smooth full motion video, and nobody even noticed, the idea was that dead. Then 3D accelerators, then GPUs, then hardware T&L.... the driving force in PC development has been graphics.
Sometime in the last couple of years, PCs really hit a plateau; they've gotten fast enough to do practically anything we can think of, at least for now. We can generate, manipulate, and output graphics of unbelievable quality... and we're mostly pretty blase' about the whole thing.
I'll tell you, though, if I showed my desktop machine (Athlon 2800+, GeForce FX5950, dual 36gb Raptors in RAID-0, Audigy 2 Platinum, Klipsch 5.1 speakers) to my 15-year-old self, I'd fear for my life. In 1985, I'd have killed someone with a big smile on my face to own a machine like that.
Phew, I kinda went off on a tangent there. Getting back on track..... GUI means a very specific thing. If the OS can turn individual dots on and off, and draws the user interface that way, it's a GUI.
You know, the thought occurs to me..... if, as you say, 90% of the people are doing 'wrong' and only 10% are doing 'right'... well, I'd say that the definitions of 'right' and 'wrong' are pretty messed up.
In this case, I'd say that 'right' is being defined by about 0.0001% of the population, and seriously needs revisiting.
SCO stated that the deal was 'worth' millions, but they didn't say that they actually RECEIVED millions. And they announced it just a day or two AFTER they announced results for their prior quarter; I think they are hoping that they will sign some more licenses this quarter, and that three months from now, people won't realize that the EV1 deal was, like, a dollar.
I would paint this as a desperate gamble. If this is true, I think SCO has no more than a quarter or so left before their BS is truly exposed.... unless, of course, Microsoft gives some other company money to 'freely sign a license for the valuable SCO IP'.
"Run away, intruder, or I shall gum you to death!"
:-)
Suggestion for rev 2.0: fangs and claws are much more visually impressive than smoothly rounded corners.
Grim Fandango was one of the best adventures ever done. It had great graphics, great voices, tough but solvable (for the most part, anyway :)) puzzles, incredible creativity, and a truly superb storyline.
It also sold dismally. LucasArts lost their shirt on GF. And that was BEFORE the huge slump in per-title PC sales.
It's possible that gamers really are 'hungry for' this kind of title, but given how most titles are selling these days, that's an awfully big risk to take. They've been burned badly a number of times in providing exactly what gamers are asking for. We may want these things, but apparently there aren't enough of us that want them to make the projects profitable. LucasArts is literally cutting their losses.
It's a shame... I imagine for this kind of game to really see a renaissance, it'll have to be developed in a low-cost country, which would allow them to sell, say, 25K copies and still be profitable.
It was *intended* as a wry, witty observation, meant to spark a little thought, and I fear that the moderators kneejerked before they did so.
Think about this a minute: what is a dollar, but t politician's promise? There's nothing backing them. They're not a promise to pay anything. And the Fed has been on a crazy printing spree for the last few years, desperately trying to paper over the damage from the tech bubble (which WE are mostly feeling, regardless.) There's a reason the dollar has been dropping so fast over the last two years. The Dow may be at 10,000 now, but the measuring stick has changed.
You say 'it's the best we can do', but that just isn't the case. I am of the firm belief that floating currencies is a lot of what is behind the steady siphoning of wealth from the general public and into the hands of the very, very rich. Floating money allows for the rise of a speculative class, and currency speculations are zero-sum; if I win, you lose. In essence, currency traders extract a lot of value out of the economy while providing nothing in exchange; they are vultures. Commodity money prevents this, except in cases where governments are abusing their money supplies. (ie, increasing the currency base without increasing the reserves.)
Remember, what you have in your wallet is a politician's promise to pay nothing. I was trying to point this out, because the anti-counterfeiting stance struck me as amusing....it's okay if the government does it, but if private citizens do it, we'd have too much of it. I believe that ANY counterfeiting is too much, and that's pretty much what's going on every day.
Well, okay, I guess it's not quite counterfeiting. It's a 100% genuine promise to pay nothing. :-)
So basically, HP and the government are both trying to stop you from taking pieces of paper, putting ink on it, and declaring that it is 'money' and 'has value'.
This is essentially what the government does every day. They print, at next to zero cost, boatloads of worthless paper, foist it off on a captive population, and extract valuable goods and services in exchange.
It's a great racket. No wonder they're worried about competition.
The weird thing about this argument is that it can't be contradicted; if you insist that all people who don't watch television are very loud about that fact, then anyone who disputes this is sort of proving the point.
I think it's going to become a very strong meme. It makes sense on the surface, and ALL arguments, both pro and con, can be construed as data supporting it. I bet practically everyone will believe it within a year or two.
That does not, however, make it true.
From what I have gathered, the SPEWS philosophy isn't just indifference to collateral damage (ie, 'civilian casualties'); they actively do this damage in order to try to force ISPs into changing their habits. And they are extremely difficult to both reach and reason with; you can post on a newsgroup and hope someone pays attention to your pleas.
I don't know if the actual newsgroup replies come from people who make decisions with SPEWS, but those replies are amazingly hostile. "Oh, you're blocked? That's because you're on a crummy ISP that allows spammers. You're on a contract and can't switch? Well, you'd better start calling your ISP, because the block on your addresses isn't going away until the spammer adjacent to you does, and maybe not then, because you're a whiner."
(ok, ok, that last part was a bit of hyperbole, but it's not that far off... check dejanews!)
Admittedly, they're not killing anyone, but the tactic of deliberately attacking people who are only tangentially related to your real target is often called 'terrorism'. The consequences here are far less serious, but the fundamental tactic remains the same.... someone is doing something you don't like, and so you hurt a whole lot of people to try to force them to stop. So I don't use SPEWS.
There are a number of other, much saner, blocklists available, and the advent of Bayesian filtering is a VERY big deal. I am personally using a combination of postfix, maildrop, SpamAssassin and bogofilter, and I get amazing results; I only started training about two weeks ago, and the spam I have to deal with has dropped by over 99%. I get 1 or 2 false negatives per day, and I have had only one false positive since I started using this system. It does take a little maintenance, but it's much less annoying and intrusive than the constant attention digging through spam takes.
It is possible, in other words, to do an exceptional job of stopping spam without contributing to a form of terrorism.
I've happily paid for a number of games where I didn't get anything physical. In all cases, of course, I got a file that will let me play the game forever (no need for a central server), but I don't have anything physical. And most shareware is that way.... or don't you register your shareware? And of course almost all Free Software comes without anything tangible... you can pay extra for a CD, but hardly anyone does in the era of broadband.
PC gaming is probably going to become mostly distributed over the Web. As other, smarter people have pointed out, it's a great way for a PC publisher to make money: with no middleman, they keep a much higher percentage. Since the market for PC games is shrinking so fast ANYWAY, the old tradeoff of accepting a lower percentage in order to make many more sales doesn't really work anymore.... going for the boutique market, instead of the mass market, seems the only likely way for them to survive.
In the electronic distribution field, I've seen three major models: Everquest, Valve, and Stardock. EQ and the other MMORPGs are a little different than anything else; they require a huge investment of servers and bandwidth to allow people to play the game, and a monthly subscription fee is the only way they could possibly pay for that. This model doesn't bother me at all....I'm a happy Second Life user, for instance.
Valve's method, on the other hand, involves spending a whole bunch of money on servers and bandwidth, but it's not for MY benefit, it's for THEIRS. They do this to make sure that I'm not stealing their software... there is no benefit to me WHATSOEVER. And there's no WAY they're going to get me to pay them for servers to make sure that I'm paying them!
Their games would work perfectly well on the old model of "sell it to me once and provide patches". They claim they'll be 'streaming content', but their content doesn't particularly need to be streamed. There are two main reasons for Steam; to prevent piracy, and to guarantee Valve a monthly revenue stream. They want to charge me monthly for features that benefit only them. Steam will not only cost me monthly, it will also provide me a service that is inferior to the one I've been getting for free. Because of that, I don't think it will fly.
If HL2 comes out in the standard "all you need is the CD to play" model, I'll buy it. If I'm required to use Steam, I will be much less likely to purchase, and there is NO WAY I will cough up any extra money to subscribe after purchasing it. Valve claims they "provide lots of extra content", but I just don't see that.... almost all their content comes from the mod community, FOR FREE. If I can't get that stuff for free anymore, I'll go play something else... it's not like I'm short on options.
Finally, there's Stardock's model, which I like a lot. I can buy an individual game if I want, or I can buy a subscription to everything they do. They have two subscriptions, one for their "serious" (Object Desktop) stuff, and one for their "fun" (Drengin Network) stuff. Anything I download during my subscription will continue to work even if I stop subscribing, which is critically important to me. If I still want to play the game I downloaded today ten years from now, it'll work fine (assuming the OS will run it, at least); there's no artificial barrier. They provide enough servers and bandwidth to provide me what I paid for; they're not building this complex copy-protection system and expecting ME to pay for it. I appreciate that they have no copy protection on their games... and I pay for it.
Ultimately, I think the EQ and the Stardock models will fly. I very strongly suspect that Steam is going to fail miserably: if HL2 is good enough, it may carry them for awhile, but I think ultimtaely the idea of charging customers for inconvenience is not workable.
I've had a hypothesis for some time that the security flaw rate in Linux would decline over time and eventually approach zero, where Microsoft's would stay essentially constant. I believed this would happen because the Linux source was open and all the security holes would gradually be found and squashed, where the Microsoft source, being closed, wouldn't be as closely examined and would remain a fertile field for new exploits forever.
Well, in 2003, my pretty little hypothesis sure wasn't looking too good. I haven't actually compared numbers, but I felt like there were just as many bad critical bugs on Linux as there were on Microsoft. From my perception, the Linux rate rose, while the Microsoft rate dropped, which is exactly opposite what I was expecting.
I still believe that closed source is "fake" security, and that the only way to get REAL security is for everything to be open, but in terms of actual number of published exploits, both systems appear to be about equal at the moment.
And the standards to which Microsoft needs to be held are pretty much immaterial; only Microsoft can fix that code, where anyone can, in theory, fix bugs in OSS. Personally, I think we can use them as a yardstick, but we shouldn't be flinging mud.... very many more years like 2003, and they'll be flinging lots more of it back at us.
In 2003, OSS security sucked. I hope 2004 is better.
Their network quality is better than anything I've used except PacBell's.... when I was on PacBell DSL, it was very very low latency and really great. But I was one of the very first customers of PB when they lowered the price, and I heard horror stories about the later connections. Mine was always super, although the slow upload bugged me.
My biggest complaint with Speakeasy was that they had relatively high latency -- never packet loss, but much higher latency than I used to get with PacBell. They've been doing some major upgrades to their backbones, however, and over the last few months my average pings have dropped by half.... right this second, I have several hundred Call of Duty servers that are 50ms or under from me. (I don't have a filter set up for an exact count, sorry.)
I've troubleshot dozens of home connections over the last few years, and I can tell you with great assurance that if you want the best possible network connection for a reasonable price, Speakeasy is the way to go. Liberal AUP, awesome network, great support.... just an excellent, excellent company to deal with. And they like Linux!
Speakeasy doesn't say 'unlimited', they sell you bandwidth, and you can do whatever you want with it. Run servers, do VPN, run Bittorrent 24/7 -- it's all good. It's your bandwidth, you paid for it. As long as it's legal and isn't disruptive to other users, Speakeasy is happy to have you as a customer. (ie, you can't DOS people, spam, or scan/attack networks you don't own/manage, but pretty much anything else goes.)
They're linux-friendly, can do either DHCP or static IPs, have good latency, essentially zero packet loss, and they're happy to HELP YOU share your network connection with your neighbors.
As far as I'm concerned, Speakeasy should be considered the Gold Standard in ISPs. Obviously, they can't reach everyplace cable does, but if you can get Speakeasy and aren't, you may be doing yourself a disservice. Yes, they're probably a little more expensive than your current provider, and you probably won't be able to download as fast as you sometimes can on cable, but you will always get the bandwidth you were promised, you'll get low latency, good support (although the web-based support is pretty slow about responding.... call them if you're in a hurry), and best of all, you'll never get The Letter.
Some local providers can be great, too. Sonic.net in Northern California was excellent when I was there five years ago, and my brother says they're still great now. But national providers, by and large, suck rocks.
BTW, my relationship with Speakeasy is strictly 'I send you money, you give me bandwidth.' Other than that, I'm not affiliated with them. I'm just a very happy customer.
Ok, so here's a couple of questions..... the piezoelectric effect is based on physical vibration. In essence, we are introducing moving parts into a power supply. This prompts the questions:
1. Do they wear out?
2. Will other motion or vibration cause voltage spikes or sags?
Either name would be better. I like Enterprise Debian myself, it has a really nice sound.
With a name like that, it could be extremely successful, IMO. Even with icky GNOME. *grin*
I'll bet you that if it were ALREADY called Enterprise Debian, it would never have made the front page of Slashdot.
"Business people", not "end users in business". You have a pretty theory, but it doesn't match what the architect is actually saying.
Given his stated goals for the project, the name is completely wrong. How the heck is a marketer going to work with this? (and we're all marketers when we're trying to sell someone on a project.) It is a fundamentally confused message, and it's going to hurt this project. Badly.
Any given brand can own only one or two words in someone's head. Linux mostly owns "quality" and "free". (which is a very, very rare combination!) Adding User to Linux, but then changing the definition of User to mean Business means that the marketing/branding effort is largely wasted. They can't spend much time on "quality" and "free" because "user" is exactly wrong.
The closest analogy I can come up with is Nabisco deciding to launch a new line of premium cookies, something like the Pepperidge Farm line, and calling them "Saltine Cookies".
Likely ensuing debacle left as an exercise for the reader.
UserLinux implies the set of ALL users, when Bruce really wanted the subset of BUSINESS users.
It is just a terribly bad name, and I believe it will rob the project of much of its potential success.
Think about it a minute.... if it WERE called BusinessLinux, and Bruce said "GNOME only".... how upset would you be? I'd be a little bummed, because I like KDE, but I'd respect the decision. From a business perspective, it makes good sense. But excluding KDE from a distro called UserLinux seems impossibly stupid.
Until this brouhaha started, I didn't fully understand that UserLinux is meant for business. I was paying peripheral attention to UL because I thought I might someday move to it when the kinks were worked out. I do Linux every day. I've been doing Linux for a long time. Yet, even near the center of my knowledge base, I was confused about the project's focus. What are management types going to think?
Thousands and thousands of times, consultants are going to have to explain, "No, no, UserLinux is meant for business."
Names matter. This project has the wrong one.
Bruce says: "UserLinux is intended to be a system for business people."
OK, that's great, but why on earth call it UserLinux then? Shouldn't it be BusinessLinux?
Names are important. UserLinux sounds like a Linux distro intended for end users. Someone like my Mom, not someone like HP. Bruce may be right about GNOME being a better solution for business. I will, however, bet nickels to dollars that much of the controversy is because people assume that a distro called UserLinux should be about, well, users, and that's KDE's main focus.
I have assumed ever since the initial announcement that UserLinux might end up being my distro of choice, and I was upset when I heard about KDE's exclusion. Now that I read further, I see I have no reason to be upset, because UserLinux isn't intended for me.
It wouldn't surprise me to see the whole project fail because of this fundamental naming problem. Is a distro called UserLinux even going to register on a CIO's radar?
BusinessLinux might have. I don't think UserLinux will.
And I pay for my software if I like it. I have given RedHat over 3 grand (including 2500 for an RHCE course a couple years ago [and MAN am I pissed at them for invalidating that investment by abandoning the market that spawned them]), and I've given Mandrake about $300. If I switched to Xandros and liked it, I'd absolutely support them.
They are, in other words, losing any chance at my money because they're not giving away their product.
Sheesh, no wonder regular companies have so much trouble understanding Linux. :-)
That would be exceptionally difficult, because the SD format is so thin. Hard drive heads need to float over their platters, and that takes a certain amount of thickness. They also Do Not Like being bent or squeezed, and the SD format just doesn't have a lot of structural strength.
A format the same height and width might work, but it would probably need to be thicker.
Stay alert!
Trust no one!
Keep your laser handy!
Trust The Computer.
The Computer is your friend.