I personally think that people will install one version of this per server farm and call RH when any of the boxes has a problem
I suggest reading the license - it states that the licensing is one copy of software per system (search for "Installed Systems"). Underreporting of systems can lead to a 20% fine.
Your licensing prices exclude support costs. That's all well and good for home users, but businesses generally want support. Bundle in support costs on that Win2k Server and you're well over $800.
If you're building a file server, then the client access libraries are going to kill you fast... even at $180 per WS license you'll end up ahead with RH.
As far as going with another distro - you're simply missing the point. What's costing money here isn't the software - it's the support. Most of the other distros don't offer support to the level that RH does, and that's why businesses gravitate toward RH if they're going to do Linux at all.
So why not go with the Standard service for $1500?
I'm guessing you're wanting 24x7 support, but not with 1 hour turn around or unlimited incidents. And if web support isn't sufficient (and, in general, I can't imagine a 4 business day turnaround being sufficient) then, yeah, I guess you're out of options.
But just how many calls would you need to put into MS or Novell before Redhat becomes cheaper? What about turnaround time? How long do your servers need to be down before that 1 hour turnaround starts paying back?
The second question is really the key one -- I suspect most shops could get along just fine with the Standard option, which is pretty dang cheap. And if you're just replacing a file server or the like then go with ES - which is $350 or $800 depending on your support needs.
As for who they're targeting - I'm not a sysadmin, but it would seem to me that ES is targeted more toward the Novell/MS and AS more toward Solaris/AIX/HP-UX. It's certainly not a hard line though. But, in general, it's a lot easier to port an application from another Unix to Linux than it is from Windows/Novell to Linux.
formaldehyde released by everything from your insulation, plywood, and flooring to your couch and carpets
Yup. You're right. And at what levels? Measurable? Didn't think so.
Tyvek or some other non-permeable barrier means you are breathing deadly gasses the entire time you are home
Hell, I breathe carbon dioxide all the time and it's a deadly gas too. So is oxygen. So is nitrogen. I'm freaking amazed I'm not dead yet!
Of course it might all have something to do with concentrations, but lets not let science enter into this. It'd just confuse things.
Unless you keep your house hermetically sealed with the HVAC off then it's a non-issue. Even with Tyvek or similar housewraps (which are used to dramatically increase the energy efficiency) houses are designed to leak - primarily through the attic (another energy efficiency thing).
CCM and CCA pressure treated wood is regularly used by people to build raised bed gardens
Yes, and the arsenic does leach out. And it's up to 10x the concentration in the surrounding soil. And still far, far, far below toxic levels. Plants can be concentrators, but they're unlikely to do so at any appreciable level even then.
if most people knew exactly how they make PT lumber, they would _never_ use it for their kids to play on or in their homes
I know how they make it, as well as other engineered wood products. And I still use them. Because, frankly, they are safe. The leeching levels are very, very low over a very, very long time.
I wouldn't chew the stuff, but I wouldn't chew untreated lumber either.
The thing that REALLY annoys me is that things like Cedar and Redwood are naturally just as resistant to decay, and a fully renewable resource.
Sure. And you're willing to wait how many decades for that cedar grove to become big enough to supply the wood for even one house? Or maybe you'd rather wait a couple centuries for the redwood.
Of course, I have this lovely southern pine which will be ready for use in under a decade.
Oh and what was that you were saying about toxic fumes? Because, of course, cedar doesn't give off any fumes at all, right? Uh huh.
She is, honestly, just this side of whacko
Fortunately, the side you are both on is the opposite side from the majority of humanity.
Modern wood isn't very chemically "safe" -- various treatments involve chemicals you wouldn't otherwise like to be exposed to. One of the primary agents used in pressure treated wood is aresenic -- and there's some evidence that it does leak out into the surroundings (although still at a non-harmful level). Engineered wood (plywood, pressed wood, OSB, etc) often contains formaldehyde, along with other chemicals used in the bonding process (again - low levels - less than 1% of the mass is the bonding agent).
Modern wood floors are coated with polyurethane and aluminum oxide. I've done more furniture finishing than I care to think about and poly isn't the nicest thing on earth.
There are various chemicals used in the tinting of bricks and mortar, which I can imagine would be problematic. Modern concrete is also nasty - there's a reason they warn you not to handle with bare hands, since you can get chemical burns in short order.
Glass would seem to be relatively inert, but who knows.
Paints and wallpaper all have fun and interesting chemical compounds.
There probably are some psychosomatic symptoms in this illness, as there are in many, but I doubt that covers all of it. There's a high likelyhood that he actually does have severe negative reactions to a vast amount of chemicals -- although why this is is an interesting question. Genetic defect? Too many antibiotic sprays and cleaners as a kid (yes -- overuse of these is bad and reduces the overall effectiveness of your immune system)? Exposure to some high doses of chemicals that caused a trigger effect?
And while he claims the need for EI-friendly housing is "extreme", it's extreme only in a very, very, very small community. I won't question his need for it, but I do question the number of people in need. And the fact is, it's going to be expensive. Hideously so. Removal of modern building methods and resources means a lot of human intensive labor along with some very specialized resource requirements. An adobe home for $125/sq ft may be one of the cheaper alternatives.
The question isn't so much what regulation would I think to impose as it is what would occur to the rest of the spectrum without regulation.
There is regulation on 2.4GHz -- namely the spectrum allowed and power allowed.
The reason that 802.11 networks degrade when a microwave or phone are in range is that they don't play "nice" with 802.11 devices - they don't transmit intermittently but blast full power almost continuously (there are gaps, but far fewer than 802.11 devices like).
Can you imagine applying that to the rest of the spectrum? Tragedy of the Commons indeed.
There's also the need to seperate and regulate spectrum to ensure services... and I'm not talking about broadcast AM/FM/TV here - more like emergency services radio (police/fire/ambulance). An unregulated marketplace rarely makes concessions to the public need unless forced (by government, which means regulation).
I agree with your final statement on regulation - we could certainly head toward more open airwaves and less regulation, but the people who think no regulation at all is viable - ever - don't understand the problems. And while you didn't say this in your original post, some people may very well have implied it.
For example the green at 510nm and the green at 520nm are both 'green' but with sufficient technological enhancement can be distinguished from one another.
Sure. Except that in order to communicate they need to change in some way - usually either in amplitude or frequency. Frequency modulation is used a whole lot more than amplitude modulation because it's more versitile and less subject to interference.
So let's say you give 5 nm of frequency modulation to each source -- how are you going to differentiate them when they're both at 515 nm? And do you honestly think that there's never going to be any leakage in the modulation? We've gotten a lot better about this over the past century, but still quite a ways to go - which is why you have guardbands in broadcast systems.
Oh, and yes, there is a way to differentiate them, even when they're at the same frequency - if your receivers are directed/sensitive enough then you can determine what direction a signal is from and latch onto it. Works great. Until one signal is behind the other, which is eventually going to happen if any one of the three subjects (receiver, two transmitters) is moving. And you need an incredibly sophisticated receiver and electronics to do this accurately... and hope that you don't get multipath transmissions (hah - a city may as well be a freaking hall of mirrors when it comes to radio frequencies - good luck decoding that one).
Simply put you still need regulation. Certainly we have more usable bandwidth now than we did even 20 years ago -- modern transmission and reception capabilities have improved vastly, plus signal processing capabilities have progressed in march with Moore's Law -- but that just means that there are even more ways to stomp all over your transmission. People who think regulation is a thing of the past are living in a dream world.
I believe what the poster was saying is that Toronto has the largest land area of any city in North America. Their site (which you referenced) lists it as 630 sq. km.
I've given up on finding "official" land areas of US cities - the best I can find is data from Sprawl City, which claims that Atlanta, GA has a land area over 700 sq. mi, Houston, TX over 630 sq. mi., and NYC over 541 sq. mi. -- they don't appear to include Mexico or Canada in their charts though, and so it's not a very good comparison, and I question their data - I've lived in both Atlanta and Dallas and don't believe that "metro" Atlanta is over twice the size of DFW -- Atlanta is freaking huge (and threatening to expand into both Alabama and Tennessee), but so is DFW.
None of which has anything whatsoever to do with this topic.
Georgia, and it does provide exemptions for charities and politicians (including pollsters). No exemption for surveys. There is an exemption for businesses for which you have had a previous or current relationship.
In the case of all but the politicians/pollsters you can still request to be placed on their Do Not Call list, which they must abide by thereafter.
Since signing up for my state's DNC list I've received precisely three telemarketing calls. All from DirecTV who I (kinda) had a pre-existing business relationship with. After the third I told them to place me on their DNC list.
Ever since then, zilch. Which is down from 3-4 per week.
Of course, you can decide that it will do absolutely nothing and not sign up for it. Then your desires will be fulfilled - it will do absolutely nothing for you.
Have you read the other Known Space books? There's much more to it than just Ringworld and sequels (and both sequels, IMO, aren't worth reading). Neutron Star, Flatlander (Gil Hamilton stories), and a good bit more.
Lucifer's Hammer, Mote in God's Eye, Dream Park, and Integral Trees are also worth reading, although not in the Known Space series. Most of his more recent works aren't very good - and I say this as someone who owns pretty much every book he's published - but Ringworld is without question worth reading, and Known Space is on the scale of the other worlds the OP mentioned.
Why would they poke the T-Rex that is IBM with a stick, unless they think they can bring it down?
Because, as Mr. Perens points out, they don't want to bring it down. They want to be bought out. Again.
You'd have an amazingly hard time proving infringement in court by IBM -- the bits that are most worrisome (such as SysV IPC) were in place long before IBM touched Linux or viewed SCO source. They were implemented because they were widely documented in Unix manuals, books, and taught in schools.
SCO's legal brief has quite a few sections that are laughable:
82. Linux started as a hobby project of a 19-year old student. Linux has evolved through bits and pieces of various contributions by numerous software developers using single processor computers. Virtually none of these software developers and hobbyists had access to enterprise-scale equipment and testing facilities for Linux development. Without access to such equipment, facilities, sophisticated methods, concepts and coordinated know-how, it would be difficult or impossible for the Linux development community to create a grade of Linux adequate for enterprise use.
84. Prior to IBM's involvement, Linux was the software equivalent of a bicycle. UNIX was the software equivalent of a luxury car. To make Linux of necessary quality for use by enterprise customers, it must be re-designed so that Linux also becomes the software equivalent of a luxury car. This re-design is not technologically feasible or even possible at the enterprise level without (1) a high degree of design coordination, (2) access to expensive and sophisticated design and testing equipment; (3) access to UNIX code, methods and concepts; (4) UNIX architectural experience; and (5) a very significant financial investment.
Section 82 is humorous. Section 84 is downright absurd. Point by point:
1) It's called a mailing list and revision control. The very same methods that are used in a vast amount of corporate development.
2) What expensive design and test equipment? Earlier in the brief SCO admitted that x86 hardware was vastly less expensive. The design and test equipment is these very same inexpensive boxes.
SMP wasn't that absurdly uncommon in the early 90s, and lots of people had access to large scale equipment, especially at a university. I know people who had unfettered access to early 90s supercomputers (Crays, etc), as well as SP-2s. Or built a cluster of SMP boxes running on Linux for PhD projects -- all of this in the early to mid 90s.
3) Code? No need. Methods and concepts? Sure. They're documented in man pages, thousands of books, and taught as part of most university CS curriculums. They're not difficult concepts really, and re-implementing them may not be trivial, but it's not impossible either.
4) Yes, because nobody knows the UNIX architecture except SCO. Uh huh. It's not in the very same books and courses mentioned previously.
5) There is a large financial investment - look at Redhat, Slackware, FSF, or just start counting man-hours donated to the kernel. If volunteer efforts were incapable of accomplishing anything then Habitat for Humanity would've gone belly up over a decade ago.
To top it all off there's a good bit of questioning with regards to Caldera Linux, the GPL, and SCO. If SCO knew that there were IP violations in the Linux kernel then it willfully violated the GPL in distributing them in Caldera Linux. That doesn't mean that those IP rights suddenly get lost, but it does mean that their legal case becomes a whole lot more hairy.
Use tabs and the anotherClass will never, ever line up in any useful manner. Use spaces and it's happy. And lets not forget the problems with mixing tabs and spaces, because you're too damn lazy to change your tabstop. Then anyone who edits the code with a different tabstop gets horribly misindented code. Fun fun fun.
Tabs are the devil.
Also, with spaces, many editors will require you to hit the right-key for every space to get to the next indentation level. With tabs, most editors will let you jump in increments of the size of the tabs
Use a crappy editor and you get what you deserve. Any decent editor will do the right thing. Vim allows you to set the tabstop, the indent level, and translate tabs to spaces properly.
Which is pretty irrelevant if you keep it maximized in the browser. Opera 5.02 even had the option to open all windows as maximized.
Although, as others have pointed out, it could be a drawback of tabs vs true MDI, since you can't view two pages at once without going to separate program instances.
I found Opera5's MDI/tabs slightly better than IE's multiple windows, but not vastly so. Which is why I keep wondering what I'm missing, or if the people who never used tabs before also didn't use multiple windows previously. Shrug.
Actually what I missed most from Opera were the mouse gestures... those rocked. I'm a keyboard guy, but browsing pretty much dictates using the mouse. Being able to do most common operations from the mouse alone solved most of the keyboard/mouse switch problems. I believe there are plugins/COM objects to do the same for IE (and I know there are for Moz), but I haven't bothered hunting.
Serious question on tabbed browsing
on
Hyatt Discusses Tabs
·
· Score: 3, Interesting
How does tabbed browsing differ from MDI (which I've used in Opera5) or from simply opening multiple browser windows? As best I can tell it's just the same thing as MDI...
As far as MDI vs multiple windows, it's a tradeoff. With MDI you only need to minimize one app to get it out of the way, and don't have to sequence through a ton of browsers to get to something else - neither of which may be an issue for many people. With multiple windows you can see the titles for everything in the task bar, instead of on a tab bar, so it's a more consistent interface - again, may not be an issue depending on how you do things.
Switching between them is a wash - ctrl-tab vs alt-tab. Opening stuff up in another window/tab is also a wash, although being able to open stuff up in the background is a nice addition for tabs (it's just an additional keypress/mouse action with multiple windows).
I guess I just don't see the wonderfulness of tabs, even having used Opera5 previously. What features am I missing here? And no, I'm not trolling.
No, it's not a conflict of interest. Holding corporations often hold companies that have diverse businesses, even ones that butt heads. Furthermore, the individual companies are usually given pretty free reign to do as they wish, as long as they're putting up enough profits to satisfy corporate.
Probably the best example is Sony Corp - which owns both Sony Electronics and Sony Music. At the same time Sony Music are asking for stronger protections in consumer electronics due to alleged piracy, Sony Electronics is fighting against any additional protection measures due to increased cost and reduced consumer satisfaction.
As Mr. Perens states, if IBM takes this really personally and decides to cut business with the holding corp as a whole then they just drop the contract with TT. And encourage their partners to do the same. Which is a whole lot of partners.
Video cards are obsolescing faster than any other part of the system. An Athlon 1200 is still adequate for most uses, but had I bought a video card the day I bought the CPU, it would be worthless today.
A GF3 (about the same era) is not worthless today. When UT2k3 came out I played it on an Athlon 750 and a GF2 (original, w/ 32MB of RAM). It was playable at low enough features and resolution. I used the same system to play DungeonSiege and NWN with no issues.
That Athlon 1200 is about as useful as a GF3 as well... it does most things just fine, but throw something really big at it and it'll choke. Try doing realtime encoding of video to MPEG2 or MPEG4 for instance. It can't handle it.
As it happens, the GF3 has held its value far better than the CPU too - a GF3 (of any flavor) sell for about $80 new, down from a maximum (for a GF3 Ti500) of $400. An Athlon 1200 might get you $25 if you're lucky, and they retailed for nearly $1000 when new.
The price of a "reasonable" gaming card is rising faster than inflation
No it's not. Your comparison is false. The original Voodoo cards sold for $300 when brand new. The Voodoo2's sold for about $300 as well, and a lot of people bought two of them for the SLI feature. The TNT sold for about $200 new, but it was priced as an underdog from a new company that had no marketshare and didn't run most of the games available. The GF and GF2 premiered at about $350 while the GF3, GF4 and GFFx premiered at $400 for the flagship card. The ATI 8500, 9700 Pro, and now 9800 Pro all came out at roughly $400, and the ATI AIW cards premier for closer to $500.
The lower end cards have always premiered at lower prices as well, and both ATI and Nvidia know that they need to have a well performing card at the $200 price point at debut. Even with the GF4 Ti4600 at $400 the Ti4200 came out at $200 - and most sites recommended the 4200 because the price/performance wasn't (and still isn't) there for the 4600. Much the same is true for the Radeon 9500 vs 9700 Pro.
Hell, the GF3 still gives great performance in modern games for $80. It probably won't do that great in Doom3, but you asked for modern games.
When you do this upgrade, it provides only a narrowly focused benefit. Non-gaming video card benefit has diminished to insignificance
Take that Athlon 1200 you mentioned. It probably has 64-128MB of memory and a 10-20GB HD. What's the advantage to upgrading the CPU? Or the memory? Or the HD? If all you're doing is surfing the web, email, and maybe writing a doc or two in your favorite word processor then there's little to no benefit in each. Heck, the computer is massive overkill even for that. The major reason to upgrade any of those nowadays is gaming - sure, there are others, but there are other reasons to upgrade your video card too. Newer cards do a much, much better job of offloading the CPU for DVD playing (or anything else involving MPEG2). They're better for CAD/CAM too - although if you're deeply serious then you'll spend the extra $400-800 for the "pro" version of the video card which has 10% more functionality for 300% of the price... but that 10% makes a huge difference.
I mean, really, how much more are you going to get from productivity apps with a P4 3.06GHz CPU vs a P3 1.2? Not much. The app will spend most of its time waiting for input. Unless, of course, your productivity app is one of the increasingly few that's CPU bound.
The top of the line cards are insanely priced... and most of the people buying them are doing it to brag about the few extra fps they get compared to a card half the price. But the midrange cards ($100-200) are extremely good values and last a couple years unless you're wanting the absolute top of the line all the time. In which case you're digging your own grave.
Which is one of many reason's I don't read Tom's anymore. Repeated use of badly done graphs (guys... having the graph range from 150 to 160 so that the guy that scored 157 vs everyone else's 154-55 is crap), sensationalistic reviews, obvious bias for or against particular vendors, and other such yellow journalism tactics made me (and many others it seems) stop reading it years ago.
The question isn't to ask "where's the money coming from" -- it's to read the review independantly and see if they were fair and unbiased at all or just "rah! rah!"-ed the subject (or vastly overplayed weaknesses).
Honestly, while I like AnandTech, I disliked the bit of spin at the end of the ATI article (and I dislike ATI - but the R300/350 beat the crap out of NVidia). Talking about a graphics chip that's 6 months out still is, well, nothing but FUD.
From PCWorld: "The 9800 Pro card clearly has the edge is in anti-aliasing tests"
Even Anandtech shows that the 9800 Pro slides slightly if you don't turn on AA or AF as compared to the 9700. There's a bit more overhead it appears. But it's a matter of a few FPS at best.
Turn on AA/AF and you get as much as a 30% advantage on the 9800 Pro, which is significant. Further tests show you essentially get the next higher ansiotropic filtering level for "free" compared to the 9700 Pro.
Frankly, the PCW article is a piece of crap, giving little actual data. Go read AnandTech, Tech Report, Ars Technica, Sharky's, or even Tom's Hardware.
Oh, and I'm not an ATI fanboy either - I still won't buy one of their cards because of previous (lack of) driver support.
We're talking about graphics cards that cost as much as entire systems... and not fancy workstation cards, cards designed for games!
We're talking about graphics cards that have more transistors and processing power than most CPUs. Have you looked at the R300 or NV30 GPUs? The shaders are fully programmable... just like CPUs are. Except they're a whole lot faster for the operations they're designed for.
You're also talking about video cards with 128 MB of memory that's 2-3x the speed of the stuff you put on your motherboard. Of course, a few years ago, 128 MB was more than you'd put in anything short of a workstation.
In otherwords, that $150 video card has more horsepower than the an entire workstation did just a few years prior. Oh, and the workstation cards are based off the same chips but only cost about 4x as much now - which is a considerable improvement over how it used to be.
Hell, I still remember seeing one of the first VR systems in the early 90s from GVU at Georgia Tech. It was designed to help reduce acrophobia and consisted of a SGI Onyx with a RealityEngine2. It could usually do 30 fps at 640x480 in 8 bit color with non-textured simple solids. Put more than a dozen or so objects in the FOV though and you started stuttering badly. The system cost roughly $600,000 - without the VR goggles.
About a year or so later you could go out to CompUSA and buy a 3DFx Voodoo card for $200 that could handle 100x the polygons, with texturing, at the same resolution with a higher frame rate.
Heck, companies are now looking at the GeForce FX and ATI Radeon 9700 cards and considering doing movie-quality rendering on them. Because they're getting that good. And you can do it in a tenth the time it would take otherwise. Trading a $10,000 workstation for a $400 video card sounds like one helluva deal to me.
Let's chain down the game developers and make them use $40 SiS305 cards, or better yet, $20 second-hand Matrox G400s and Voodoo3s
Why? Those cards are all cheap for a reason - they're crap. They don't support any of the graphics capabilities desired nowadays (the G400 and SiS305 don't even support the graphics capabilities of their time). You may get UT2k3 running on a G400 or V3, but not at a reasonable frame rate, and in order to get that reasonable frame rate you have to ditch visual quality features. There's simply no way around it.
Doom3 on such a card? Yah, right.
If you're happy with graphics from 5 years ago, then keep playing those games. But whining about cost and "it's not a workstation" just shows how amazingly ignorant you are.
You are misinformed. According to the stated philosophy of _their_own_ organization (the FSF) and the overall project (GNU) of which they are a part, they *are* less free. The L in LGPL stands for lesser.
Yes, and that's because of a wonderful bit of spin doctoring on the part of RMS. The L used to stand for Library.
The LGPL is a reasonable license for libraries. If you apply the GPL to a library then it's just as viral as its detractors say. Sorry, but if I utilize a published set of interfaces then there's no reason your code should force me to reveal all of my code. But that's what applying the GPL to a library does, and it's the reason pure GPL licenses tend to founder and fail. Funny that. The only reason that Qt has made it is because there are other licensing options.
So don't use the library? Damn straight I won't. And it doesn't mean I'm developing something proprietary either - I may want to release under the GPL, BSD license, or even to the public domain.
It's funny, really... RMS claims that he doesn't object to core logic being proprietary, but then he wants every library licensed under the GPL and not the LGPL -- but linking a distributed work to a GPL library will force you to reveal your core business logic.
I suppose everyone should just become an ASP... that way you don't have to distribute.
Nifty test... although slide 19 sucks. Kept trying to find the line between the two X's, until I re-read the explanation and it said that "normal" should not be able to see a line.
As far as traffic lights go - a good bit of that is rote memorization of "top/leftmost == red == stop", so even if a severely colorblind person saw a light they shouldn't be confused as to its state. It may, however, take longer to process the status than it does for a color-aware person (particularly for more complex traffic lights).
My sister is one of the rare women who is red/green colorblind. As I recall she sees green as shades of brown - and can distinguish (somewhat) between different shades of green (and thus likes Pothos type plants). I've never asked her what she sees red as.
Slashdot Mirror
I personally think that people will install one version of this per server farm and call RH when any of the boxes has a problem
I suggest reading the license - it states that the licensing is one copy of software per system (search for "Installed Systems"). Underreporting of systems can lead to a 20% fine.
Your licensing prices exclude support costs. That's all well and good for home users, but businesses generally want support. Bundle in support costs on that Win2k Server and you're well over $800.
If you're building a file server, then the client access libraries are going to kill you fast... even at $180 per WS license you'll end up ahead with RH.
As far as going with another distro - you're simply missing the point. What's costing money here isn't the software - it's the support. Most of the other distros don't offer support to the level that RH does, and that's why businesses gravitate toward RH if they're going to do Linux at all.
So why not go with the Standard service for $1500?
I'm guessing you're wanting 24x7 support, but not with 1 hour turn around or unlimited incidents. And if web support isn't sufficient (and, in general, I can't imagine a 4 business day turnaround being sufficient) then, yeah, I guess you're out of options.
But just how many calls would you need to put into MS or Novell before Redhat becomes cheaper? What about turnaround time? How long do your servers need to be down before that 1 hour turnaround starts paying back?
The second question is really the key one -- I suspect most shops could get along just fine with the Standard option, which is pretty dang cheap. And if you're just replacing a file server or the like then go with ES - which is $350 or $800 depending on your support needs.
As for who they're targeting - I'm not a sysadmin, but it would seem to me that ES is targeted more toward the Novell/MS and AS more toward Solaris/AIX/HP-UX. It's certainly not a hard line though. But, in general, it's a lot easier to port an application from another Unix to Linux than it is from Windows/Novell to Linux.
Yes, and that house is going to be wonderfully energy efficient and cheap, right?
Uh huh.
There's a reason we've moved on from such basic building methods. They're labor and resource intensive - both upfront and in continuing costs.
The srticle states that 40% of Internet traffic is Spam
No, the article states that 40% of email is spam.
Which, frankly, seems low. But perhaps they're including corporate email, which often sees a much lower spam level.
I'm still trying to find estimates on how much of all Internet traffic is from SMTP -- I've seen estimates of anything from 5% to 30%.
formaldehyde released by everything from your insulation, plywood, and flooring to your couch and carpets
Yup. You're right. And at what levels? Measurable? Didn't think so.
Tyvek or some other non-permeable barrier means you are breathing deadly gasses the entire time you are home
Hell, I breathe carbon dioxide all the time and it's a deadly gas too. So is oxygen. So is nitrogen. I'm freaking amazed I'm not dead yet!
Of course it might all have something to do with concentrations, but lets not let science enter into this. It'd just confuse things.
Unless you keep your house hermetically sealed with the HVAC off then it's a non-issue. Even with Tyvek or similar housewraps (which are used to dramatically increase the energy efficiency) houses are designed to leak - primarily through the attic (another energy efficiency thing).
CCM and CCA pressure treated wood is regularly used by people to build raised bed gardens
Yes, and the arsenic does leach out. And it's up to 10x the concentration in the surrounding soil. And still far, far, far below toxic levels. Plants can be concentrators, but they're unlikely to do so at any appreciable level even then.
if most people knew exactly how they make PT lumber, they would _never_ use it for their kids to play on or in their homes
I know how they make it, as well as other engineered wood products. And I still use them. Because, frankly, they are safe. The leeching levels are very, very low over a very, very long time.
I wouldn't chew the stuff, but I wouldn't chew untreated lumber either.
The thing that REALLY annoys me is that things like Cedar and Redwood are naturally just as resistant to decay, and a fully renewable resource.
Sure. And you're willing to wait how many decades for that cedar grove to become big enough to supply the wood for even one house? Or maybe you'd rather wait a couple centuries for the redwood.
Of course, I have this lovely southern pine which will be ready for use in under a decade.
Oh and what was that you were saying about toxic fumes? Because, of course, cedar doesn't give off any fumes at all, right? Uh huh.
She is, honestly, just this side of whacko
Fortunately, the side you are both on is the opposite side from the majority of humanity.
Modern wood isn't very chemically "safe" -- various treatments involve chemicals you wouldn't otherwise like to be exposed to. One of the primary agents used in pressure treated wood is aresenic -- and there's some evidence that it does leak out into the surroundings (although still at a non-harmful level). Engineered wood (plywood, pressed wood, OSB, etc) often contains formaldehyde, along with other chemicals used in the bonding process (again - low levels - less than 1% of the mass is the bonding agent).
Modern wood floors are coated with polyurethane and aluminum oxide. I've done more furniture finishing than I care to think about and poly isn't the nicest thing on earth.
There are various chemicals used in the tinting of bricks and mortar, which I can imagine would be problematic. Modern concrete is also nasty - there's a reason they warn you not to handle with bare hands, since you can get chemical burns in short order.
Glass would seem to be relatively inert, but who knows.
Paints and wallpaper all have fun and interesting chemical compounds.
There probably are some psychosomatic symptoms in this illness, as there are in many, but I doubt that covers all of it. There's a high likelyhood that he actually does have severe negative reactions to a vast amount of chemicals -- although why this is is an interesting question. Genetic defect? Too many antibiotic sprays and cleaners as a kid (yes -- overuse of these is bad and reduces the overall effectiveness of your immune system)? Exposure to some high doses of chemicals that caused a trigger effect?
And while he claims the need for EI-friendly housing is "extreme", it's extreme only in a very, very, very small community. I won't question his need for it, but I do question the number of people in need. And the fact is, it's going to be expensive. Hideously so. Removal of modern building methods and resources means a lot of human intensive labor along with some very specialized resource requirements. An adobe home for $125/sq ft may be one of the cheaper alternatives.
The question isn't so much what regulation would I think to impose as it is what would occur to the rest of the spectrum without regulation.
There is regulation on 2.4GHz -- namely the spectrum allowed and power allowed.
The reason that 802.11 networks degrade when a microwave or phone are in range is that they don't play "nice" with 802.11 devices - they don't transmit intermittently but blast full power almost continuously (there are gaps, but far fewer than 802.11 devices like).
Can you imagine applying that to the rest of the spectrum? Tragedy of the Commons indeed.
There's also the need to seperate and regulate spectrum to ensure services... and I'm not talking about broadcast AM/FM/TV here - more like emergency services radio (police/fire/ambulance). An unregulated marketplace rarely makes concessions to the public need unless forced (by government, which means regulation).
I agree with your final statement on regulation - we could certainly head toward more open airwaves and less regulation, but the people who think no regulation at all is viable - ever - don't understand the problems. And while you didn't say this in your original post, some people may very well have implied it.
Yes, because we all know that there are no interference problems between 802.11, 2.4 GHz phones, and microwaves.
Frankly, the 2.4GHz spectrum is a case in point of why regulation is needed.
For example the green at 510nm and the green at 520nm are both 'green' but with sufficient technological enhancement can be distinguished from one another.
Sure. Except that in order to communicate they need to change in some way - usually either in amplitude or frequency. Frequency modulation is used a whole lot more than amplitude modulation because it's more versitile and less subject to interference.
So let's say you give 5 nm of frequency modulation to each source -- how are you going to differentiate them when they're both at 515 nm? And do you honestly think that there's never going to be any leakage in the modulation? We've gotten a lot better about this over the past century, but still quite a ways to go - which is why you have guardbands in broadcast systems.
Oh, and yes, there is a way to differentiate them, even when they're at the same frequency - if your receivers are directed/sensitive enough then you can determine what direction a signal is from and latch onto it. Works great. Until one signal is behind the other, which is eventually going to happen if any one of the three subjects (receiver, two transmitters) is moving. And you need an incredibly sophisticated receiver and electronics to do this accurately... and hope that you don't get multipath transmissions (hah - a city may as well be a freaking hall of mirrors when it comes to radio frequencies - good luck decoding that one).
Simply put you still need regulation. Certainly we have more usable bandwidth now than we did even 20 years ago -- modern transmission and reception capabilities have improved vastly, plus signal processing capabilities have progressed in march with Moore's Law -- but that just means that there are even more ways to stomp all over your transmission. People who think regulation is a thing of the past are living in a dream world.
I believe what the poster was saying is that Toronto has the largest land area of any city in North America. Their site (which you referenced) lists it as 630 sq. km.
I've given up on finding "official" land areas of US cities - the best I can find is data from Sprawl City, which claims that Atlanta, GA has a land area over 700 sq. mi, Houston, TX over 630 sq. mi., and NYC over 541 sq. mi. -- they don't appear to include Mexico or Canada in their charts though, and so it's not a very good comparison, and I question their data - I've lived in both Atlanta and Dallas and don't believe that "metro" Atlanta is over twice the size of DFW -- Atlanta is freaking huge (and threatening to expand into both Alabama and Tennessee), but so is DFW.
None of which has anything whatsoever to do with this topic.
Georgia, and it does provide exemptions for charities and politicians (including pollsters). No exemption for surveys. There is an exemption for businesses for which you have had a previous or current relationship.
In the case of all but the politicians/pollsters you can still request to be placed on their Do Not Call list, which they must abide by thereafter.
Whatever you want to believe.
Since signing up for my state's DNC list I've received precisely three telemarketing calls. All from DirecTV who I (kinda) had a pre-existing business relationship with. After the third I told them to place me on their DNC list.
Ever since then, zilch. Which is down from 3-4 per week.
Of course, you can decide that it will do absolutely nothing and not sign up for it. Then your desires will be fulfilled - it will do absolutely nothing for you.
My coworker downloaded it this morning... something like 24000 x 14000. Well into the absurd category for any non-commercial use.
The ringworld series fall off to quickly
Huh?
Have you read the other Known Space books? There's much more to it than just Ringworld and sequels (and both sequels, IMO, aren't worth reading). Neutron Star, Flatlander (Gil Hamilton stories), and a good bit more.
Lucifer's Hammer, Mote in God's Eye, Dream Park, and Integral Trees are also worth reading, although not in the Known Space series. Most of his more recent works aren't very good - and I say this as someone who owns pretty much every book he's published - but Ringworld is without question worth reading, and Known Space is on the scale of the other worlds the OP mentioned.
Why would they poke the T-Rex that is IBM with a stick, unless they think they can bring it down?
Because, as Mr. Perens points out, they don't want to bring it down. They want to be bought out. Again.
You'd have an amazingly hard time proving infringement in court by IBM -- the bits that are most worrisome (such as SysV IPC) were in place long before IBM touched Linux or viewed SCO source. They were implemented because they were widely documented in Unix manuals, books, and taught in schools.
SCO's legal brief has quite a few sections that are laughable:
82. Linux started as a hobby project of a 19-year old student. Linux has evolved through bits and pieces of various contributions by numerous software developers using single processor computers. Virtually none of these software developers and hobbyists had access to enterprise-scale equipment and testing facilities for Linux development. Without access to such equipment, facilities, sophisticated methods, concepts and coordinated know-how, it would be difficult or impossible for the Linux development community to create a grade of Linux adequate for enterprise use.
84. Prior to IBM's involvement, Linux was the software equivalent of a bicycle. UNIX was the software equivalent of a luxury car. To make Linux of necessary quality for use by enterprise customers, it must be re-designed so that Linux also becomes the software equivalent of a luxury car. This re-design is not technologically feasible or even possible at the enterprise level without (1) a high degree of design coordination, (2) access to expensive and sophisticated design and testing equipment; (3) access to UNIX code, methods and concepts; (4) UNIX architectural experience; and (5) a very significant financial investment.
Section 82 is humorous. Section 84 is downright absurd. Point by point:
1) It's called a mailing list and revision control. The very same methods that are used in a vast amount of corporate development.
2) What expensive design and test equipment? Earlier in the brief SCO admitted that x86 hardware was vastly less expensive. The design and test equipment is these very same inexpensive boxes.
SMP wasn't that absurdly uncommon in the early 90s, and lots of people had access to large scale equipment, especially at a university. I know people who had unfettered access to early 90s supercomputers (Crays, etc), as well as SP-2s. Or built a cluster of SMP boxes running on Linux for PhD projects -- all of this in the early to mid 90s.
3) Code? No need. Methods and concepts? Sure. They're documented in man pages, thousands of books, and taught as part of most university CS curriculums. They're not difficult concepts really, and re-implementing them may not be trivial, but it's not impossible either.
4) Yes, because nobody knows the UNIX architecture except SCO. Uh huh. It's not in the very same books and courses mentioned previously.
5) There is a large financial investment - look at Redhat, Slackware, FSF, or just start counting man-hours donated to the kernel. If volunteer efforts were incapable of accomplishing anything then Habitat for Humanity would've gone belly up over a decade ago.
To top it all off there's a good bit of questioning with regards to Caldera Linux, the GPL, and SCO. If SCO knew that there were IP violations in the Linux kernel then it willfully violated the GPL in distributing them in Caldera Linux. That doesn't mean that those IP rights suddenly get lost, but it does mean that their legal case becomes a whole lot more hairy.
And they royally fuck up when you continue a line, because the right place to put the line continuation is rarely a tabstop.
some_call(someClass->randomMethod(),
anotherClass->someValue());
Use tabs and the anotherClass will never, ever line up in any useful manner. Use spaces and it's happy. And lets not forget the problems with mixing tabs and spaces, because you're too damn lazy to change your tabstop. Then anyone who edits the code with a different tabstop gets horribly misindented code. Fun fun fun.
Tabs are the devil.
Also, with spaces, many editors will require you to hit the right-key for every space to get to the next indentation level. With tabs, most editors will let you jump in increments of the size of the tabs
Use a crappy editor and you get what you deserve. Any decent editor will do the right thing. Vim allows you to set the tabstop, the indent level, and translate tabs to spaces properly.
Which is pretty irrelevant if you keep it maximized in the browser. Opera 5.02 even had the option to open all windows as maximized.
Although, as others have pointed out, it could be a drawback of tabs vs true MDI, since you can't view two pages at once without going to separate program instances.
I found Opera5's MDI/tabs slightly better than IE's multiple windows, but not vastly so. Which is why I keep wondering what I'm missing, or if the people who never used tabs before also didn't use multiple windows previously. Shrug.
Actually what I missed most from Opera were the mouse gestures... those rocked. I'm a keyboard guy, but browsing pretty much dictates using the mouse. Being able to do most common operations from the mouse alone solved most of the keyboard/mouse switch problems. I believe there are plugins/COM objects to do the same for IE (and I know there are for Moz), but I haven't bothered hunting.
How does tabbed browsing differ from MDI (which I've used in Opera5) or from simply opening multiple browser windows? As best I can tell it's just the same thing as MDI...
As far as MDI vs multiple windows, it's a tradeoff. With MDI you only need to minimize one app to get it out of the way, and don't have to sequence through a ton of browsers to get to something else - neither of which may be an issue for many people. With multiple windows you can see the titles for everything in the task bar, instead of on a tab bar, so it's a more consistent interface - again, may not be an issue depending on how you do things.
Switching between them is a wash - ctrl-tab vs alt-tab. Opening stuff up in another window/tab is also a wash, although being able to open stuff up in the background is a nice addition for tabs (it's just an additional keypress/mouse action with multiple windows).
I guess I just don't see the wonderfulness of tabs, even having used Opera5 previously. What features am I missing here? And no, I'm not trolling.
No, it's not a conflict of interest. Holding corporations often hold companies that have diverse businesses, even ones that butt heads. Furthermore, the individual companies are usually given pretty free reign to do as they wish, as long as they're putting up enough profits to satisfy corporate.
Probably the best example is Sony Corp - which owns both Sony Electronics and Sony Music. At the same time Sony Music are asking for stronger protections in consumer electronics due to alleged piracy, Sony Electronics is fighting against any additional protection measures due to increased cost and reduced consumer satisfaction.
As Mr. Perens states, if IBM takes this really personally and decides to cut business with the holding corp as a whole then they just drop the contract with TT. And encourage their partners to do the same. Which is a whole lot of partners.
Video cards are obsolescing faster than any other part of the system. An Athlon 1200 is still adequate for most uses, but had I bought a video card the day I bought the CPU, it would be worthless today.
A GF3 (about the same era) is not worthless today. When UT2k3 came out I played it on an Athlon 750 and a GF2 (original, w/ 32MB of RAM). It was playable at low enough features and resolution. I used the same system to play DungeonSiege and NWN with no issues.
That Athlon 1200 is about as useful as a GF3 as well... it does most things just fine, but throw something really big at it and it'll choke. Try doing realtime encoding of video to MPEG2 or MPEG4 for instance. It can't handle it.
As it happens, the GF3 has held its value far better than the CPU too - a GF3 (of any flavor) sell for about $80 new, down from a maximum (for a GF3 Ti500) of $400. An Athlon 1200 might get you $25 if you're lucky, and they retailed for nearly $1000 when new.
The price of a "reasonable" gaming card is rising faster than inflation
No it's not. Your comparison is false. The original Voodoo cards sold for $300 when brand new. The Voodoo2's sold for about $300 as well, and a lot of people bought two of them for the SLI feature. The TNT sold for about $200 new, but it was priced as an underdog from a new company that had no marketshare and didn't run most of the games available. The GF and GF2 premiered at about $350 while the GF3, GF4 and GFFx premiered at $400 for the flagship card. The ATI 8500, 9700 Pro, and now 9800 Pro all came out at roughly $400, and the ATI AIW cards premier for closer to $500.
The lower end cards have always premiered at lower prices as well, and both ATI and Nvidia know that they need to have a well performing card at the $200 price point at debut. Even with the GF4 Ti4600 at $400 the Ti4200 came out at $200 - and most sites recommended the 4200 because the price/performance wasn't (and still isn't) there for the 4600. Much the same is true for the Radeon 9500 vs 9700 Pro.
Hell, the GF3 still gives great performance in modern games for $80. It probably won't do that great in Doom3, but you asked for modern games.
When you do this upgrade, it provides only a narrowly focused benefit. Non-gaming video card benefit has diminished to insignificance
Take that Athlon 1200 you mentioned. It probably has 64-128MB of memory and a 10-20GB HD. What's the advantage to upgrading the CPU? Or the memory? Or the HD? If all you're doing is surfing the web, email, and maybe writing a doc or two in your favorite word processor then there's little to no benefit in each. Heck, the computer is massive overkill even for that. The major reason to upgrade any of those nowadays is gaming - sure, there are others, but there are other reasons to upgrade your video card too. Newer cards do a much, much better job of offloading the CPU for DVD playing (or anything else involving MPEG2). They're better for CAD/CAM too - although if you're deeply serious then you'll spend the extra $400-800 for the "pro" version of the video card which has 10% more functionality for 300% of the price... but that 10% makes a huge difference.
I mean, really, how much more are you going to get from productivity apps with a P4 3.06GHz CPU vs a P3 1.2? Not much. The app will spend most of its time waiting for input. Unless, of course, your productivity app is one of the increasingly few that's CPU bound.
The top of the line cards are insanely priced... and most of the people buying them are doing it to brag about the few extra fps they get compared to a card half the price. But the midrange cards ($100-200) are extremely good values and last a couple years unless you're wanting the absolute top of the line all the time. In which case you're digging your own grave.
Which is one of many reason's I don't read Tom's anymore. Repeated use of badly done graphs (guys... having the graph range from 150 to 160 so that the guy that scored 157 vs everyone else's 154-55 is crap), sensationalistic reviews, obvious bias for or against particular vendors, and other such yellow journalism tactics made me (and many others it seems) stop reading it years ago.
The question isn't to ask "where's the money coming from" -- it's to read the review independantly and see if they were fair and unbiased at all or just "rah! rah!"-ed the subject (or vastly overplayed weaknesses).
Honestly, while I like AnandTech, I disliked the bit of spin at the end of the ATI article (and I dislike ATI - but the R300/350 beat the crap out of NVidia). Talking about a graphics chip that's 6 months out still is, well, nothing but FUD.
Did you?
From PCWorld: "The 9800 Pro card clearly has the edge is in anti-aliasing tests"
Even Anandtech shows that the 9800 Pro slides slightly if you don't turn on AA or AF as compared to the 9700. There's a bit more overhead it appears. But it's a matter of a few FPS at best.
Turn on AA/AF and you get as much as a 30% advantage on the 9800 Pro, which is significant. Further tests show you essentially get the next higher ansiotropic filtering level for "free" compared to the 9700 Pro.
Frankly, the PCW article is a piece of crap, giving little actual data. Go read AnandTech, Tech Report, Ars Technica, Sharky's, or even Tom's Hardware.
Oh, and I'm not an ATI fanboy either - I still won't buy one of their cards because of previous (lack of) driver support.
We're talking about graphics cards that cost as much as entire systems... and not fancy workstation cards, cards designed for games!
We're talking about graphics cards that have more transistors and processing power than most CPUs. Have you looked at the R300 or NV30 GPUs? The shaders are fully programmable... just like CPUs are. Except they're a whole lot faster for the operations they're designed for.
You're also talking about video cards with 128 MB of memory that's 2-3x the speed of the stuff you put on your motherboard. Of course, a few years ago, 128 MB was more than you'd put in anything short of a workstation.
In otherwords, that $150 video card has more horsepower than the an entire workstation did just a few years prior. Oh, and the workstation cards are based off the same chips but only cost about 4x as much now - which is a considerable improvement over how it used to be.
Hell, I still remember seeing one of the first VR systems in the early 90s from GVU at Georgia Tech. It was designed to help reduce acrophobia and consisted of a SGI Onyx with a RealityEngine2. It could usually do 30 fps at 640x480 in 8 bit color with non-textured simple solids. Put more than a dozen or so objects in the FOV though and you started stuttering badly. The system cost roughly $600,000 - without the VR goggles.
About a year or so later you could go out to CompUSA and buy a 3DFx Voodoo card for $200 that could handle 100x the polygons, with texturing, at the same resolution with a higher frame rate.
Heck, companies are now looking at the GeForce FX and ATI Radeon 9700 cards and considering doing movie-quality rendering on them. Because they're getting that good. And you can do it in a tenth the time it would take otherwise. Trading a $10,000 workstation for a $400 video card sounds like one helluva deal to me.
Let's chain down the game developers and make them use $40 SiS305 cards, or better yet, $20 second-hand Matrox G400s and Voodoo3s
Why? Those cards are all cheap for a reason - they're crap. They don't support any of the graphics capabilities desired nowadays (the G400 and SiS305 don't even support the graphics capabilities of their time). You may get UT2k3 running on a G400 or V3, but not at a reasonable frame rate, and in order to get that reasonable frame rate you have to ditch visual quality features. There's simply no way around it.
Doom3 on such a card? Yah, right.
If you're happy with graphics from 5 years ago, then keep playing those games. But whining about cost and "it's not a workstation" just shows how amazingly ignorant you are.
You are misinformed. According to the stated philosophy of _their_own_ organization (the FSF) and the overall project (GNU) of which they are a part, they *are* less free. The L in LGPL stands for lesser.
Yes, and that's because of a wonderful bit of spin doctoring on the part of RMS. The L used to stand for Library.
The LGPL is a reasonable license for libraries. If you apply the GPL to a library then it's just as viral as its detractors say. Sorry, but if I utilize a published set of interfaces then there's no reason your code should force me to reveal all of my code. But that's what applying the GPL to a library does, and it's the reason pure GPL licenses tend to founder and fail. Funny that. The only reason that Qt has made it is because there are other licensing options.
So don't use the library? Damn straight I won't. And it doesn't mean I'm developing something proprietary either - I may want to release under the GPL, BSD license, or even to the public domain.
It's funny, really... RMS claims that he doesn't object to core logic being proprietary, but then he wants every library licensed under the GPL and not the LGPL -- but linking a distributed work to a GPL library will force you to reveal your core business logic.
I suppose everyone should just become an ASP... that way you don't have to distribute.
Nifty test... although slide 19 sucks. Kept trying to find the line between the two X's, until I re-read the explanation and it said that "normal" should not be able to see a line.
As far as traffic lights go - a good bit of that is rote memorization of "top/leftmost == red == stop", so even if a severely colorblind person saw a light they shouldn't be confused as to its state. It may, however, take longer to process the status than it does for a color-aware person (particularly for more complex traffic lights).
My sister is one of the rare women who is red/green colorblind. As I recall she sees green as shades of brown - and can distinguish (somewhat) between different shades of green (and thus likes Pothos type plants). I've never asked her what she sees red as.