The Germans came frighteningly close to establishing complete control of the skies with jet aircraft, rockets, hybrids of the two, and atomic weapons. Fortunately, in the latter area, they were not on the right track, but there is some evidence that they had built a small but effective breeder reactor by the end of the war.
Here are a few sites that point out their capabilities toward the end of the war - they were literally decades ahead in some areas, and one could argue we still haven't caught up in others....
Luft46.Com A site giving an unprecedented insight in to what the German Luftwaffe might have looked like had the war in Europe continued for another year or so. There were incredible things in the works - the world's first stealth fighter-bomber, the first guided missiles, dozens of extremely advanced jet aircraft designs, the atmoshpheric skipping SAnger Amerika bomber, and more. This one can be a significantly enjoyable time sink. You've been warned. (Don't miss the "Luft Art" section showing many very skillful renderings of these superplanes.)
John Walker's "Rocket-a-Day" Paper Read this to get a feel for the truly incredible capabilities of the German A4/V2 program, and an eye-opening comparison the the incredibly inefficient way we get things in to space today. This is particularly relevant in the wake of the Columbia disaster. NASA needs to die, and this paper points out one big reason why.
Stupid, useful lookup trick for Netscape/Mozilla
on
A Word a Day
·
· Score: 1
FWIW, I was asked the other day for these scripts by someone who hadn't snagged it back when this was Google's "toolbar" for Netscape. The original script allowes you to select text in any page, then click the Google toolbar link to initiate a new search. The thing that surprises me is how many people don't know you can do this stuff, so since it seems relatively on-topic, here it is...
The following is just a set of "Stupid JavaScript Tricks" that modify the original Google lookup script to allow similar easy lookups of other sites, including Dictionary.com and Thesaurus.com. (If you're really into wordplay, you can even build one of these to automatically pump the text into an anagram generator or the like...)
The GoTo one is a bit different, though: select the URL (making sure not to select the "http://" part due to the way JS processes things) and Voila! you're instantly at the site that some goon didn't bother to make a hyperlink.
Anyway, here they are, to use them just create personal toolbar items with these URLs:
Notes: 1. Columbine Bookmark Merge Users: These URLs are too long for that wonderful but dated program to handle and will be truncated in the mrege process. 2. One of these days, I suppose I should modify the scripts to open the lookup in a new tab or window... Any JS experts wanna fix that real quick?
<b>Do a Google lookup>/b> <A HREF="javascript:q=document.getSelection();for(i=0;i<frames.length;i++){q=frames[i].document.getSele ction();if(q)break;}if(!q)void(q=prompt('Enter text to search using Google. You can also highlight a word on this web page before clicking Google Search.',''));if(q)location.href='http://www.googl e.com/search?client=googlet&q='+escape(q)" >Google</A>
<b>GoTo a non-hyperlinked URL</b> <A HREF="javascript:q=document.getSelection();for(i=0;i<frames.length;i++){q=frames[i].document.getSele ction();if(q)break;}if(!q)void(q=prompt('Enter or Select URL.',''));if(q)location.href='http://'+(q)" >GoTo</A>
<b>Dictinary.com lookup</b> <A HREF="javascript:q=document.getSelection();for(i=0;i<frames.length;i++){q=frames[i].document.getSele ction();if(q)break;}if(!q)void(q=prompt('Enter text to lookup using Dictionary.com. You can also highlight a word on this web page before clicking Dictionary Search.',''));if(q)location.href='http://www.dicti onary.com/cgi-bin/dict.pl?term='+escape(q)" >Dict.</A>
<b>Thesaurus.com lookup</b> <A HREF="javascript:q=document.getSelection();for(i=0;i<frames.length;i++){q=frames[i].document.getSele ction();if(q)break;}if(!q)void(q=prompt('Enter text to lookup using Thesaurus.com. You can also highlight a word on this web page before clicking Thesaurus.',''));if(q)location.href='http://www.th esaurus.com/cgi-bin/search?config=roget&words='+es cape(q)" >Thes.</A>
So, pull out the battery, find the terminals that supply the power, and hook something up to them - a smallish DC motor with a piece of tape wrapped on the shaft (for air drag) will draw enough current, but if you really wanted to do it quickly, drop a 10-watt sand power resistor across the terminals, and point a fan at it (it will get DAMN hot, hot enough to burn if you don't cool it down with something - if you want to be extra careful, clamp a heatsink/fan combo onto the resistor - power the fan with the battery, too!)...
Sure, you could go to all that trouble, but why? I mean after all, you've got a Mobile Pentium Room Heater sitting right there - Running the right programs, it can suck the battery dry in no time.
A really good choice would be something that is both computationally intensive and does a lot of CD-ROM access, since that's another huge power-consumer in laptops. Something like a Exact Audio Copy with ridicuous settings for ripping and encoding should do the trick... (BTW, if your CD drive is removable, remove it (replacing it with an "air bay" makes the laptop lighter, too), or at least make sure there's no disk in it, for max battery life.)
A few weeks ago, I wrote up a text for an online petition with a long list of reasons why we need an industry standard for laptop batteries [hanno.de], similar to consumber electronics battery cells.
This has been tried before. Several years back, Duracell developed a set of standard laptop batteries that they offered for license to laptop manufacturers and even competitors.
The industry yawned, with only Compaq really giving them a try, but abandoning the idea shortly after.
There's just *way* too much money in proprietary batteries for them to leave it on the table. The whole situation is sort of a Yugoslavia of disintegrating standards - it's just as bad as memory cards, where we have Compact Flash (the most open of the group), MultiMedia Card, SD Memory Cards, and the hideously proprietary Sony Memory Stick all making our portable digital devices totally incompatible with one another...
Businesses still buy IBM hardware because of name recognition. They figure that the biggest must be the best. These decisions are made by suits whose VCRs flash 12:00, and who never ask any technical people for advice. As long as this is true, IBM will keep selling garbage. They know that suckers will buy their name.
No, people buy IBM because they make a very good laptop product. And as someone who was formerly a program manager responsible for laptops for one of their major competitors, I think I'm in a position to know. (I don't work for any of these companies any more, so I have no axe to grind here...)
IMO, Thinkpads are among the very best laptops out there, and distinct from all others because they're not simply OEM badge-engineered retreads of standard laptop designs from the likes of Compal, Quanta, and the other Taiwanese laptop mills.
As an experienced RoadWarrior I've had Dells (~6), Compaqs (3), Toshibas (4), T.I.s (2) and IBM ThinkPads (4) over the years, and of the group, the IBMs are my favorite by far, offering the best overall value, and the only tough thin-and-light machines. I'd take the IBMs any day over any of the others: Dells and Toshibas are a distant second. (Older Toshibas were nearly indestructable, but the newer ones have lost that vital trait - I once dropped my Toshiba Portege 610 down a flight of concrete stairs, resulting in nothing but a few scratches.)
Just a few weeks back I bought(brand-new in a factory-sealed box, thanks dot-com collapse)a ThinkPad 570E with the expansion base, the port replicator with built-in ethernet, and tons of accessories. It is ideal for me - thin and light when I want it, but also fully capable of the heavy lifting when required. I love it, and there is really no comparable product from any competitor. (Being older, it even still has serial ports to support my embedded development work, something that's vanishing from new laptops.)
I will say, though, that I've noticed that ThinkPad batteries only last about 1/2 to 2/3 as long as those of other brands, so there may be something to the allegations. They certainly make a nice income on replacement batteries.
Why the US still clings to imperial units is beyond me.
Actually, for the simple reason that they're what the technological world was built on, and also the not-inconsequential fact that English units often tend to relate to the real-world better than thier Metric/SI counterparts.
This is NOT just an artifact caused by familiarity: For ordinary use, the English units are often just more convenient because thier sizes are more applicable to the problem at hand. For instance, in machining and design of precision parts, thousandths of an inch turn out to be considerably more useful than metric units, just simply because of the mechanics of material removal using common machining processes. This is one reason almost all machining in high-precision industries like Aerospace and Oil/Petrochemical/Energy is still done in English units. (Note that the recent NASA Mars probe debacle only happened when one group deviated from accepted industry practice of using English measurements and switched to Metric. (And without even telling anyone, at that!) The simple reason the error was not caught is that no idiot (except maybe a French idiot, they still haven't got over thier Napoleonic pride in the moronic Metric system) would use metric measurements in an aerospace context - it's just not done.)
Another good example of the oh-so-awkward size of metric units is the liters/100km unit that has to be used to measure fuel econonomy in reasonably sized numbers. Ugh. There are dozens of other examples.
Units are somewhat arbitrary, but to be honest, in my engineering career, I've seen many more errors with Metric units (decimal point errors, imagine that!) than I have with the English system.
HELP STAMP OUT THE METRIC SYSTEM!
P.S.: Of course, what we really need to adopt is a correct measurement system based on Dublins, that perfect unit of length between a yard and a meter, where the acceleration of gravity here on earth would be 10 Dublins/s^2. Physics and engineering students worldwide would celebrate my birthday with fireworks and parties.:-)
Someone please mod the parent postup as insightful (and correct!) Far too much of what I see on/. these days is knee-jerk political bashing, usually of Republicans...
Sometimes you just need to bite the bullet and do it. Screw the bean counters.
Remember, this is exactly what Enron did. I agree that significant businesses have resulted from big gambles (sometimes literally, as in Fred Smith's taking the the embyonic FedEx's meager cash to Vegas and winning enough to make payroll.) But public companies, by virtue of the fact that they *are* publicly owned and traded, do not have the same freedom to gamble as private firms, and if they choose to do so anyway, then the executives may properly do jail time.
FWIW: I worked with (not for) Enron and some of its predecessor energy trading companies like NorAm, and I'm convinced they just thought they were smarter than everyone else. For a while, they were. They did indeed bite the bullet and say, in effect, "screw the beancounters..." They just didn't know when to quit, and by that time they had so obliterated "the line" in thier own minds that they were unable to even acknowledge they had crossed it.
There is large support from some for the full development of a Blended Wing Body (BWB) airliner, and there are significant arguements for that development. The concept is over 50 years old (Northrop), the current design is at least 10 years old (acquired when Boeing bought McDonnell), and an implementation exists as the B-2 stealh bomber.
Good comment, but a couple of corrections: First, the BWB concept is far older than 50 years, and was pioneered by Vincent Burnelli in the 1930s, not McDonnell-Douglas in the 90s. Northrup was undoubtedly among the real pioneers (along with the Hortens and Lippisch in Germany, both of whom predated much of Northrup's work) in true flying wings, but there's a distiction between flying wings or nurflugels and BWBs or airfoil-body designs like Burnelli's (which have a notable lack of blending...)
For more good information on these vehicles, check out the following links from my own bookmarks collection on this topic:
Aircrash.org: A good site about the Burnelli wing/body design and it's history. Unfortunately it's marred by repeated and rather shrill accusations of a giant conspiracy to keep aircraft unsafe. A good resource, but there are more than a few sour grapes here. Too bad, becuase the idea has obvious merit.
Luft46.com: A site that catalogs many of the incredibly advanced and innovative aircraft that the Germans were working on (in in many cases beginning prodction of) at the end of WWII. The Horten 229 (known variously as Ho IX, Ho229, and Gotha 229) is an interesting comparison to the current B2. Not only was it a fairly high-performance jet-powered flying wing fighter/bomber, but the German engineers had already begun researching stealth: its body was made of carbon granules sandwiched between two thin plywood skins, forming the first know radar-absorbing aerostructure. (One warning though: don't blame me if you spend all your Christmas holidays reading this site. There's some amazing stuff here, and it tends to be a pretty engrossing once you get into it a bit...)
Now that the Sonic Cruiser has been canned, it will be interesting to see if the other (r)evolutionary design, the BWB, gains traction and sees a greater chance of production.
Indeed. I worked on the GE Unducted Fan in the mid 80's, and there is a LOT of room for improvement in current designs. Even the UDF itself might have made it had they paid a high-schooler's attention to basic physics: The UDF had two rows of either 8 or 10 counterrotating blades. As is entirely predictable, the thing made a very effective siren when running, and the acoustic and air-impulse energy literally beat the adjacent body skins nearly to death. Why this problem didn't occur to GE when they know enough to always use a prime number of blades in turbine sections is beyond me...
RTF has been in office for years and it is an open, portable standard readable on many platforms and with many programs.
Obviously you haven't tried it. RTF has gotten more complaints from users than raw word Docs does!
Replace "RTF" with "HTML" and you've got a winner, though.
OK, Let's see you put a page break in that HTML document... Seriously, an extended HTML could make a very nice document format, some of the better ones, like the one used by HTMLDOC actually *do* let you put in line breaks and such. I've started using HTMLDOC to generate lots of my documentation now, because it does a pretty good job of retaining the gist of the formatting and produces very nice PDFs from the same web pages I have to generate anyway. This product has really improved lately. In fact, the only thing wrong with HTMLDOC, IMO, is that it uses the GPL rather than a truly free license.
Now if only the Netscape/Mozilla team would add support for the HTMLDOC extended tags in Composer, and make HTMLDOC a standard output filter option (which would dramtically improve their ability to print web pages, anyhow...) we'd really have something.
make a PVR. All the cool functionality, but VERY limited fastforward/rewind. Make it so you can't skip commercials if you're watching the show. Then, put 2-3 tuners in it, and advertise the hell out of it. Charge people $5 a month for it. More TV watched, and more ads watched. I'm surprised the networks haven't done this
That's essentially what Time-Warner's new VOD "Digital Cable" solution does, but they charge $10/month for the PVR feature + another $15/month for the mandatory Digital Cable box and service.
Gee, I wonder why people aren't falling all over themselves to sign up for this rip-off? (Of course, they're also forcing much popular programming onto digital-only to arm-twist the masses that find analog cable just fine. (To be honest, I get a far better picture with analog cable than those that pay extra for the crappy pixellated "digital" feed, but heck, I only have cable to get RoadRunner anyway...)
Although the economics *should* work for PVR/VOD functions run from the headend, I can't see the MSOs (cable companies) to let this make economic sense *for the users* anytime soon. As a result, PVRs will continue to grow in popularity - after all, *something* has to give us a reason to need Terabyte microdrives...
So we could poison this system by actually responding to every spam and providing erroneous payment details, mailing details, etc to the companies who want to hawk their products by spam. Obviously they would waste plenty of money processing and shipping these orders, only to find out that they are getting no profit for it. This way, Ralksy's customers go under. Essentially, Ralksy's air supply would be cut off.
This idea has a lot of merit, and realistically, most of us here on/. have had addresses for so long that they're on darn near every spam list anyway.
Perhaps the thing to do is the following: Respond to *everything*. Click every link in every spam you get (well, almost every spam - I refuse to find out what Wanda really does with animals), including the almost-always bogus "unsubscribe me" links. Yes, this will make things worse short-term, but if enough people do it, then the spammers will have an incentive to filter those people out, thus for the first time giving the spammers an incentive to *remove* people from thier mailing lists.
It's ugly, but poisoning the waterhole has been known to be effective in getting people to move on, and I don't know about yours, but mine is already pretty thoroughly poisoned. Perhaps we need to poison the waterholes of the customers of Mr. Ralsky and his slimy ilk.
I'm seriously considering clicking spam links and responding to spam messages whenever I find myself on hold, for instance. (If we spent half of our current/. time on responding to spam, we could make a big dent.) Widespread adoption of this habit could fundamentally change the ecomical attractiveness of spamming until legislation with real teeth can be implemented nationwide, and enable spam filters to be placed at all international network PoPs. It's about time to blacklist entire countries for support of spam...
My grand-dad went to buy a computer at a big chain store. He just wanted to look around on the web and email some friends/family. The sales droid tried to sell him a P4 2Ghz with all the bells and whistles. I ended up putting to gether a Duron 1.2G for $250 that does all he wants to do. Unless you are a big game freak or a geek (like most of us), people just don't need that much computing power.
Hear, hear. For all practical purposes, once we passed 300 MHz, it's been impossible to tell a difference in performance from a user perspective.
This is especially true with a relatively lightweight OSes like Win98/SE/ME and contemporary versions of applications like Office 97. I can't believe I'm saying that, but sadly, Linux is now far more of a resource pig than 9x, especially with the incredibly obese Gnome environment and other bloat that's been thrown in over the past couple of years in the mad rush to ape Microsoft. (The days of Linux running on machines like my 75 MHz 16MB Libretto are over, which is why it's stuck on Mandrake 7.1 with the older, much slimmer KDE, which it runs just fine, thank you.)
Sadly, both Windows and most current Linux distros expect 500+ MHz CPUs and 128 MB+ or RAM, which is just ridicuous - that's far more power than most full 3D Sun or SGI workstations had just a few years ago.
Win9x has its warts, that's for sure, but it runs great on slow hardware, as do older versions of Linux. (Sure, it's possible to build a slim Linux distro on your own, but almost all of us have better things to do with our time...)
For 99.5% of computer users, Win9x will do all they need - browsing, e-mail, and light document production. It also has the substantial advantages of having copies available all over the place (legitimate, cheap, used copies, even), and the bigger advantage of interoperating seamlessly with the rest of the world, something that's still not true for Linux and its associated office apps, which is why I prefer a Windows desktop with Unix extensions rather than the other way around - one way works flawlessly, the other is a constant battle.
Re:IDE Raid, inexpensive but major hassle
on
IDE RAID Examined
·
· Score: 2
so by the time you get the good drives and the ones with big caches.... you're spending the same money as us SCSI guys do.
No way. And big drive caches don't really buy you much in a RAIDed environment. It's far better to have that cache in the RAID controller itself. Drive speed doesn't matter much, it's spindle spreading that wins big, and IDE drives are now nearly as fast as SCSI at a quarter or less of the price!
Funny... I've enjoyed Ultra 160 raid for over 2 years now... and what have you IDE guys had?
Speaking for myself, I've had high-performance IDE RAID arrays that will easily saturate an Ultra 160 SCSI connection at a cost so low I can use RAID 0+1 instead of 5 and still save money. You just can't do that with SCSI.
IDE is for consumer and home use. SCSI is for serious work. even the manufacturers admit it... my scsi drives STILL come with 5 year warranties... IDE drives come with 1 year now... funny that eh?
Red herring if ever there was one. The mechanisms are the same and have been for years from most manufacturers. The only difference is the logic board. Aren't you glad you're paying a 4x premium for that one-drive-at-a-time SCSI interface? (All high-performance IDE RAID units have one channel per drive, totally eliminating bus contention. That approach is way too expensive to execute in SCSI.)
Finally, drive manufacturers are building "enterprise class" IDE drives now that have the same warranties as the very expensive SCSI drives. Sure, they're more than the consumer grade IDE drives, but still several times cheaper than SCSI drives that are only slightly faster. And you might want to check the reliability specs. I know of at least two drive vendors that quote *higher* MTBFs for their IDE drives than they do for their SCSI equivalents. Tell me again why I should waste my company's money on an anachronistic love-affair with an obsolete technology when that decision could result in additional layoffs?
If you don't believe me, get a clue about what's happening in the storage industry. I'd suggest InfoStor or SearchStorage as starting places. And if you think IDE is strong now, just wait until high-perfromance Serial ATA RAID controllers start to become common in a few months.
SCSI is dead. Not because it's a bad technology (although it isn't that great, either), but because it cannot compete economically.
Re:IDE Raid, inexpensive but major hassle
on
IDE RAID Examined
·
· Score: 3, Informative
Thats because one of the major limitations of current generation IDE is that only one device on a channel can "talk" at a time. So even if you're using a RAID card with two devices on a channel, it will be no faster than a standard IDE connection, since only one drive read/write can be done at a time.
Not at all true. There are a good many IDE (ATA, actually) RAID controllers out there that use one drive per IDE channel, and connect to the host via SCSI or Fibre Channel. (Of course, in this case there is *never* channel contention, and the weak spot is in the SCSI or of FC connection, both of which are using the SCSI protocol. This approach is FAR faster than almost all SCSI-based RAID systems out there, and much cheaper to boot. One of the advantages of using serious IDE Raid subsystems (not the cheezy desktop variety) is that the cost savings can allow you to replace RAID 5 with RAID 0+1 (sometime called 10) and still save money. I know because I've engineered and built multi-terabyte storage servers on this technology that are 2-3x faster and an order of magnitude less expensive than high-end storage servers like the IBM Shark or EMC Symmetrix. IDE *will* squash SCSI, it's not a matter of if, but when, mostly because SCSI will never be able to compete with the volume economics that produce IDE's 5-6x cost advantage. The performance advanstage of individual SCIS drives is already becoming marginal, and the speed of individual drives is nearly irrelevant anyway in a RAID environment where most of the poerformance comes from spanning mutiple splindles, not the speed of the individual disks. (This is why a properly configured RAID array of disks with average access time N can deliver average access times of significantly less than N.)
With SCSI, all of the drives on a channel can talk at the same time until the 160 MB/s that SCSI can handle is saturated.
Not even close. SCSI is a one-talker at a time bus architecture. This is one reason a good IDE RAID controller can so easily kick SCSI butt. The largest clusters and multiprocessor computers are all going to high performance IDE RAID arrays because of their superior cost, performance, and yes, reliability, since electrical problems in physical SCSI are one of the most common causes of data corruption in high performance environments, which is one of the chief reasons Fibre Channel has been so widely adopted. It too uses the SCSI protocol, and so has real weaknesses, but at least it avoids the hideous flakiness of SCSI's connector and termination scheme.
E-mail and the web are about to undergo the collapse phase of their CB-radio-like growth.
I mean this seriously. For a great many non-technical people, pop-ups, viruses, spam and the death of Napster are reducing the net value (groan) of the Interent to near zero. The inevitable result is that those people will simply decide that they can do without a marginal benefit that makes thier lives more difficult and complicated. Like CB radio, the shine is wearing off the Internet, and it will soon begin to *lose* people fo rhte first time ever. (Note that unlike CB, I certainly don't expect the Net to slip into technological obscurity and irrelevance, but people *will* leave - and in fact, they already are, and that's newsworthy.)
Case in point: I know one local company that is seriously considering pulling the plug on Internet access to thier network and instead designating just a few "browsing and e-mail computers" connected to the Internet. They claim they can easily justify the cost of the additional machines just by avoiding viruses, and I don't doubt them. (They've been hit by Linux viruses/worms, too, so don't get too cocky about blaming Windows.)
Spam filters are irrelevant until they can be made easy enough for the masses. I *hate* spam, but don't use much in the way of spam filtering because I don't want to spend half a week learning the obtuse intricacies of a dozen different mail processing packages, after evaluating several dozen alternatives. There needs to be an embedded box made for serious firewall, spam filter, and content filter use that is as easy to set up and use as e-smith and clarkconnect. It *has* to be embedded, because most people are NOT going to get another PC just to protect the one they use. ISPs could help, but most aren't interested, especially given their declining fortunes as the result of stupid management and decerasing revenues from the CB radio bust mentioned above.
I think the long-term impact of spam, pop-ups, and viruses could be far, far greater than most people expect. At the risk of raising the perennnial usenet cry of wolf, these things, left unchecked, could indeed herald the "imminent death of the Net" - at least as we know it today. The scary thing is that the cure could be worse than the disease...
The CD I have (admittedly, given to me by a friend - I thought it was a standard CD, but perhaps he altered it) is designed to run Smoothwall entirely from CD, and that way, even if the machine is compromised, there's no real damage done...
Tablet PCs strike me as an opportunity for Linux to crack the desktop market. From what I've read of initial impressions of Tablet PCs, the thing that sucks the most is the crappy Microsoft inking software, that is neither easy to learn to write for (sucks worse than Graffiti) nor will it learn your writing style (sucks worse than Newton).
I've spent a LOT of time as an evaluator and beta tester for non-Tablet PC tablet devices, including several based on Linux. I can tell you with some assurance that Linux is AT LEAST 5 years behind in this area. the Ms recognition is very good, in some (not all) respects better than the Newton, and as good as Graffiti if you print anything like legibly.
Right now, Linux and the BSDs have a very long way to go just to catch up, and a longer way to go before they could have any real advantage. Several fundamental problems really bite you in the butt in this environment: from the general unsuitability of X to the fact that writing a generic handwriting recognizer is even harder given the architectural (message-passing, etc.) differences in things like KDE and Gnome.
Given the fact that we're just now getting ACPI to sort of function correctly in Linux four years after MS added ACPI support to NT and Win98, I have no reason to expect that Linux developers will get good at supporting hardware built to MS specs anytime soon...
I really think the tablet PCs would have to be thin clients to conserve battery life. I also don't think that a M$ OS is the way to go. The bloat in OS would just use up so much memory and processor cycles, that the battery life would suffer.
As I point out in another post, we can always hope that having XP in Tablet PCs will finally motivate MS to put XP on a badly-needed diet. That would be a win-win all the way around, except that Intel would find it harder to convince people they need 4 GHz CPUs to check their mail...
Re:if patents hold up, pen computing is in trouble
on
Bricklin on Tablet PCs
·
· Score: 2
Furthermore, with the release of TabletPC, Microsoft has shown again that they simply can't innovate. Microsoft's TabletPC software is the same old stuff we had 10 years ago, only in a more bloated software incarnation. The only thing that has really gotten better is the hardware and processor speed, as well as the quality of real-time graphics those machines support.
At the risk of defending Microsoft, you're wrong. I've been a beta tester for several types of tablet computers (although not a Tablet PC), and nearly all of them are atrocious. Microsoft has done an excellent job at taking another swipe at building what should be the highest and best form factor for the "Dynabook" concept. The biggest thing they've done is to finally build a tablet that has all of the following:
1) a real OS (Whether or not you like it, XP is a real OS with pretty nearly all the things that were the domain of Unix alone just a few years back. You can do pretty much anything with XP that you can do with any other real OS. (With the possible exception of heirarchical mounts: Is the totally broken MS-DFS still the only way to fake that, or has something been added to XP - I haven't used that since W2K was brand new...))
It's hard to overstate the importance of having a Tablet PC device that can run and leverage the ordinary software run on ordinary PCs. That's just never been done before. Even Microsoft's own previous attempts were marred by ignoring this important aspect and trying to pretend CE was a real OS, when it is anything but, or grafting seriously substandard pen facilities onto Windows.
2) Real handwriting recognition that works as well as is possible given the state of the art, and digital ink to allow either raw ink capture and/or deferred recognition. In my estimation, the only other machine to have truly usable recognition is the Newton (although Calligrapher on CE came close), and as I said, I've used a lot of these things.
3) A nice, color, high-resolution display. This is vital for non-trivial applications. Ultimately, we still need native resolutions of paper-sized screens that are an 4-6x what we have today, but this is still a big first step. Interestingly, total screen pixel count may grow here before the desktop catches up.
4) Real networking capability. It's surprising how many devices like this assume they're islands and really don't communicate well with the rest of the world. The fact that 802.11 is taking over the world just now and can be fully levereaged by this device is just good timing for MS, and one more reason these things will probably be wildly successful.
5) To allow the good handwriting recognition, a high-quality, high resolution digitizer suitable for all but the pickiest of graphic artists. Try the awful digitizers in some other tablet devices and you'll see what I mean. My only gripe is that MS specifies active digitizers and pen buttons, maening the stupid things need batteries. Grrr.
6) Not vital, but nice: enough storage to carry around all the data most users will care about. This has staggering implications, if you think about it...
The only real downside now is another one related to XP: the devices require a LOT of compute power, and consequently suck down a lot of battery power. (With luck, the Tablet PC will give MS a good incentive to slim down the bloat in XP - there HAS to be quite a bit of room for slimmer code there... ) Fairly quickly, they need to get to the point that a Tablet PC can run an entire shift on a single charge - that's particularly important for vertical markets like healthcare, which is expected to be an early adopter of this technology. That won't happen on a widespread scale until a full shift is possible. Ultimately, they need to produce a Tablet PC that can be thrown in the briefcase for a one week trip without bothering to pack a charger. I can do this with my Kyocera/Palm SmartPhone, and should be able to do it with my computer, too.
In one sense, it;s fair to say that MS has not brought much new to the table. But at the same time, it;s also quite fair to say that no one has ever before brought together so many of the the things that a Tablet PC must have to be successful, and Microsoft deserves some credit for that. It's a good integration job, even if most of the pieces and ideas came from elsewhere. Ultimately, this is the sort of thing Microsoft occasinally shows themselves to be good at. Let's wish them well, since these capabilities will not be available in other forms unless they can succeed in establishing a market for these things - that's the lesson of volume economics in the computer and electronics space...
They said the Tablet-PC would open "new markets", since people who usually avoid using computers because of the mouse (Have you ever tried to teach it to you mother?:-) now have a more "natural" way of interfacing with it - the pen.
I believe it was Ted Nelson that said (possibly quoting (or jabbing at) Doug Engelbart) something to the effect that, "People discovered thousands of years ago that it was much, much easier to draw with sticks than rocks, but that lessson won't be absorbed by the computer industry any time soon..."
Lastly, I humbly suggest that you haven't had your fingers in anything other than the most exceptional of open source efforts. I have seen plenty of open source, and it is as bad and worse as the worst closed source software that I've had the misfortune of seeing. Open source is no cure for software quality; software quality happens when people care about it, open source or no.
Somebody mod this up as Insightful!
There is an incredibly misguided tendency (esp. here on/.) to regard all open source software as inherently of higher quality than commercial software. In reality, nothing could be further from the truth. I fully support open source software, but the reality is that as bad as much commercial software is, most open source software is worse. There are the notable exceptions (Apache, Samba, etc.), but they are popular only because they are exceptions to the usual sub-mediocre quality of most open source software. (How many open source projects have any sort of formal testing process other than, "toss it out there and see what bug reports we get back?" Commercial software has its weaknesses, but the concept of regression testing, etc. is far more common in commercial development than it is in open source development - that's a fact. In addition, there are very few open source projects that are not primarily attractive to propellerheads - there is very little suitable for the average user, and what's there is generally of far lower quality than the usual commercial equivalents. (BTW: Star/OpenOffice doesn't count, since it was originally developed as a commercial product and opened up only through Sun's benevolence after they bought Star Division.)
Add this to the fact that most open source software is simply "me-too" functionality (best whiny voice: "See, Linux can too play DVDs, and it can sort-of print, well, to a pitiful handful of expensive printers, anyway...") rather than anything really new and innovative, and it's easy to see why we may always be chasing Microsoft's heels.
As a final for-instance, take Longhorn: Whether or not one likes Microsoft, this is a set of technologies that will have real and significant value to the users. Given the fairly fundamental nature of some of the changes (DB-based filesystem, transparent integration of information across applications, etc.), Longhorn will set open source software back another two years until its functionality has been duly duplicated in a different, fatter, CORBA-compliant way. (If you don't believe me, check out how dot-Net has thousands of open source developers tied in knots chasing their tails and adding bloat that may eventually eclipse that of Microsoft's own implementations.)
Utimately, Courageous is dead on about one thing: "Open source is no cure for software quality; software quality happens when people care about it, open source or no."
Slashdot Mirror
The Germans came frighteningly close to establishing complete control of the skies with jet aircraft, rockets, hybrids of the two, and atomic weapons. Fortunately, in the latter area, they were not on the right track, but there is some evidence that they had built a small but effective breeder reactor by the end of the war.
Here are a few sites that point out their capabilities toward the end of the war - they were literally decades ahead in some areas, and one could argue we still haven't caught up in others....
Luft46.Com A site giving an unprecedented insight in to what the German Luftwaffe might have looked like had the war in Europe continued for another year or so. There were incredible things in the works - the world's first stealth fighter-bomber, the first guided missiles, dozens of extremely advanced jet aircraft designs, the atmoshpheric skipping SAnger Amerika bomber, and more. This one can be a significantly enjoyable time sink. You've been warned. (Don't miss the "Luft Art" section showing many very skillful renderings of these superplanes.)
John Walker's "Rocket-a-Day" Paper Read this to get a feel for the truly incredible capabilities of the German A4/V2 program, and an eye-opening comparison the the incredibly inefficient way we get things in to space today. This is particularly relevant in the wake of the Columbia disaster. NASA needs to die, and this paper points out one big reason why.
The following is just a set of "Stupid JavaScript Tricks" that modify the original Google lookup script to allow similar easy lookups of other sites, including Dictionary.com and Thesaurus.com. (If you're really into wordplay, you can even build one of these to automatically pump the text into an anagram generator or the like...)
The GoTo one is a bit different, though: select the URL (making sure not to select the "http://" part due to the way JS processes things) and Voila! you're instantly at the site that some goon didn't bother to make a hyperlink.
Anyway, here they are, to use them just create personal toolbar items with these URLs:
Notes:
1. Columbine Bookmark Merge Users: These URLs are too long for that wonderful but dated program to handle and will be truncated in the mrege process.
2. One of these days, I suppose I should modify the scripts to open the lookup in a new tab or window... Any JS experts wanna fix that real quick?
So, pull out the battery, find the terminals that supply the power, and hook something up to them - a smallish DC motor with a piece of tape wrapped on the shaft (for air drag) will draw enough current, but if you really wanted to do it quickly, drop a 10-watt sand power resistor across the terminals, and point a fan at it (it will get DAMN hot, hot enough to burn if you don't cool it down with something - if you want to be extra careful, clamp a heatsink/fan combo onto the resistor - power the fan with the battery, too!)...
Sure, you could go to all that trouble, but why? I mean after all, you've got a Mobile Pentium Room Heater sitting right there - Running the right programs, it can suck the battery dry in no time.
A really good choice would be something that is both computationally intensive and does a lot of CD-ROM access, since that's another huge power-consumer in laptops. Something like a Exact Audio Copy with ridicuous settings for ripping and encoding should do the trick... (BTW, if your CD drive is removable, remove it (replacing it with an "air bay" makes the laptop lighter, too), or at least make sure there's no disk in it, for max battery life.)
A few weeks ago, I wrote up a text for an online petition with a long list of reasons why we need an industry standard for laptop batteries [hanno.de], similar to consumber electronics battery cells.
This has been tried before. Several years back, Duracell developed a set of standard laptop batteries that they offered for license to laptop manufacturers and even competitors.
The industry yawned, with only Compaq really giving them a try, but abandoning the idea shortly after.
There's just *way* too much money in proprietary batteries for them to leave it on the table. The whole situation is sort of a Yugoslavia of disintegrating standards - it's just as bad as memory cards, where we have Compact Flash (the most open of the group), MultiMedia Card, SD Memory Cards, and the hideously proprietary Sony Memory Stick all making our portable digital devices totally incompatible with one another...
... and others that are just a few years old that don't work worth a Beowolf cluster....
That's great - almost as good as "ugly as a new Cadillac", an unexpected laugh line at a talk I gave recently.
They say in order for something to be really funny, it has to be true...
Businesses still buy IBM hardware because of name recognition. They figure that the biggest must be the best. These decisions are made by suits whose VCRs flash 12:00, and who never ask any technical people for advice. As long as this is true, IBM will keep selling garbage. They know that suckers will buy their name.
No, people buy IBM because they make a very good laptop product. And as someone who was formerly a program manager responsible for laptops for one of their major competitors, I think I'm in a position to know. (I don't work for any of these companies any more, so I have no axe to grind here...)
IMO, Thinkpads are among the very best laptops out there, and distinct from all others because they're not simply OEM badge-engineered retreads of standard laptop designs from the likes of Compal, Quanta, and the other Taiwanese laptop mills.
As an experienced RoadWarrior I've had Dells (~6), Compaqs (3), Toshibas (4), T.I.s (2) and IBM ThinkPads (4) over the years, and of the group, the IBMs are my favorite by far, offering the best overall value, and the only tough thin-and-light machines. I'd take the IBMs any day over any of the others: Dells and Toshibas are a distant second. (Older Toshibas were nearly indestructable, but the newer ones have lost that vital trait - I once dropped my Toshiba Portege 610 down a flight of concrete stairs, resulting in nothing but a few scratches.)
Just a few weeks back I bought(brand-new in a factory-sealed box, thanks dot-com collapse)a ThinkPad 570E with the expansion base, the port replicator with built-in ethernet, and tons of accessories. It is ideal for me - thin and light when I want it, but also fully capable of the heavy lifting when required. I love it, and there is really no comparable product from any competitor. (Being older, it even still has serial ports to support my embedded development work, something that's vanishing from new laptops.)
I will say, though, that I've noticed that ThinkPad batteries only last about 1/2 to 2/3 as long as those of other brands, so there may be something to the allegations. They certainly make a nice income on replacement batteries.
Why the US still clings to imperial units is beyond me.
:-)
Actually, for the simple reason that they're what the technological world was built on, and also the not-inconsequential fact that English units often tend to relate to the real-world better than thier Metric/SI counterparts.
This is NOT just an artifact caused by familiarity: For ordinary use, the English units are often just more convenient because thier sizes are more applicable to the problem at hand. For instance, in machining and design of precision parts, thousandths of an inch turn out to be considerably more useful than metric units, just simply because of the mechanics of material removal using common machining processes. This is one reason almost all machining in high-precision industries like Aerospace and Oil/Petrochemical/Energy is still done in English units. (Note that the recent NASA Mars probe debacle only happened when one group deviated from accepted industry practice of using English measurements and switched to Metric. (And without even telling anyone, at that!) The simple reason the error was not caught is that no idiot (except maybe a French idiot, they still haven't got over thier Napoleonic pride in the moronic Metric system) would use metric measurements in an aerospace context - it's just not done.)
Another good example of the oh-so-awkward size of metric units is the liters/100km unit that has to be used to measure fuel econonomy in reasonably sized numbers. Ugh. There are dozens of other examples.
Units are somewhat arbitrary, but to be honest, in my engineering career, I've seen many more errors with Metric units (decimal point errors, imagine that!) than I have with the English system.
HELP STAMP OUT THE METRIC SYSTEM!
P.S.: Of course, what we really need to adopt is a correct measurement system based on Dublins, that perfect unit of length between a yard and a meter, where the acceleration of gravity here on earth would be 10 Dublins/s^2. Physics and engineering students worldwide would celebrate my birthday with fireworks and parties.
Please mod parent post up. It has a lot of good information in it, and presents a very valid (and truthful) view.
Someone please mod the parent postup as insightful (and correct!) Far too much of what I see on /. these days is knee-jerk political bashing, usually of Republicans...
Sometimes you just need to bite the bullet and do it. Screw the bean counters.
Remember, this is exactly what Enron did. I agree that significant businesses have resulted from big gambles (sometimes literally, as in Fred Smith's taking the the embyonic FedEx's meager cash to Vegas and winning enough to make payroll.) But public companies, by virtue of the fact that they *are* publicly owned and traded, do not have the same freedom to gamble as private firms, and if they choose to do so anyway, then the executives may properly do jail time.
FWIW: I worked with (not for) Enron and some of its predecessor energy trading companies like NorAm, and I'm convinced they just thought they were smarter than everyone else. For a while, they were. They did indeed bite the bullet and say, in effect, "screw the beancounters..." They just didn't know when to quit, and by that time they had so obliterated "the line" in thier own minds that they were unable to even acknowledge they had crossed it.
There is large support from some for the full development of a Blended Wing Body (BWB) airliner, and there are significant arguements for that development. The concept is over 50 years old (Northrop), the current design is at least 10 years old (acquired when Boeing bought McDonnell), and an implementation exists as the B-2 stealh bomber.
Good comment, but a couple of corrections: First, the BWB concept is far older than 50 years, and was pioneered by Vincent Burnelli in the 1930s, not McDonnell-Douglas in the 90s. Northrup was undoubtedly among the real pioneers (along with the Hortens and Lippisch in Germany, both of whom predated much of Northrup's work) in true flying wings, but there's a distiction between flying wings or nurflugels and BWBs or airfoil-body designs like Burnelli's (which have a notable lack of blending...)
For more good information on these vehicles, check out the following links from my own bookmarks collection on this topic:
The Nurflugel (German for "flying wing" more or less) Page: Great info on historical flying wing designs, and the difficulties that the problem posed for so long.
Aircrash.org: A good site about the Burnelli wing/body design and it's history. Unfortunately it's marred by repeated and rather shrill accusations of a giant conspiracy to keep aircraft unsafe. A good resource, but there are more than a few sour grapes here. Too bad, becuase the idea has obvious merit.
Luft46.com: A site that catalogs many of the incredibly advanced and innovative aircraft that the Germans were working on (in in many cases beginning prodction of) at the end of WWII. The Horten 229 (known variously as Ho IX, Ho229, and Gotha 229) is an interesting comparison to the current B2. Not only was it a fairly high-performance jet-powered flying wing fighter/bomber, but the German engineers had already begun researching stealth: its body was made of carbon granules sandwiched between two thin plywood skins, forming the first know radar-absorbing aerostructure. (One warning though: don't blame me if you spend all your Christmas holidays reading this site. There's some amazing stuff here, and it tends to be a pretty engrossing once you get into it a bit...)
Now that the Sonic Cruiser has been canned, it will be interesting to see if the other (r)evolutionary design, the BWB, gains traction and sees a greater chance of production.
Indeed. I worked on the GE Unducted Fan in the mid 80's, and there is a LOT of room for improvement in current designs. Even the UDF itself might have made it had they paid a high-schooler's attention to basic physics: The UDF had two rows of either 8 or 10 counterrotating blades. As is entirely predictable, the thing made a very effective siren when running, and the acoustic and air-impulse energy literally beat the adjacent body skins nearly to death. Why this problem didn't occur to GE when they know enough to always use a prime number of blades in turbine sections is beyond me...
RTF has been in office for years and it is an open, portable standard readable on many platforms and with many programs.
Obviously you haven't tried it. RTF has gotten more complaints from users than raw word Docs does!
Replace "RTF" with "HTML" and you've got a winner, though.
OK, Let's see you put a page break in that HTML document... Seriously, an extended HTML could make a very nice document format, some of the better ones, like the one used by HTMLDOC actually *do* let you put in line breaks and such. I've started using HTMLDOC to generate lots of my documentation now, because it does a pretty good job of retaining the gist of the formatting and produces very nice PDFs from the same web pages I have to generate anyway. This product has really improved lately. In fact, the only thing wrong with HTMLDOC, IMO, is that it uses the GPL rather than a truly free license.
Now if only the Netscape/Mozilla team would add support for the HTMLDOC extended tags in Composer, and make HTMLDOC a standard output filter option (which would dramtically improve their ability to print web pages, anyhow...) we'd really have something.
make a PVR. All the cool functionality, but VERY limited fastforward/rewind. Make it so you can't skip commercials if you're watching the show. Then, put 2-3 tuners in it, and advertise the hell out of it. Charge people $5 a month for it. More TV watched, and more ads watched. I'm surprised the networks haven't done this
That's essentially what Time-Warner's new VOD "Digital Cable" solution does, but they charge $10/month for the PVR feature + another $15/month for the mandatory Digital Cable box and service.
Gee, I wonder why people aren't falling all over themselves to sign up for this rip-off? (Of course, they're also forcing much popular programming onto digital-only to arm-twist the masses that find analog cable just fine. (To be honest, I get a far better picture with analog cable than those that pay extra for the crappy pixellated "digital" feed, but heck, I only have cable to get RoadRunner anyway...)
Although the economics *should* work for PVR/VOD functions run from the headend, I can't see the MSOs (cable companies) to let this make economic sense *for the users* anytime soon. As a result, PVRs will continue to grow in popularity - after all, *something* has to give us a reason to need Terabyte microdrives...
So we could poison this system by actually responding to every spam and providing erroneous payment details, mailing details, etc to the companies who want to hawk their products by spam. Obviously they would waste plenty of money processing and shipping these orders, only to find out that they are getting no profit for it. This way, Ralksy's customers go under. Essentially, Ralksy's air supply would be cut off.
/. have had addresses for so long that they're on darn near every spam list anyway.
/. time on responding to spam, we could make a big dent.) Widespread adoption of this habit could fundamentally change the ecomical attractiveness of spamming until legislation with real teeth can be implemented nationwide, and enable spam filters to be placed at all international network PoPs. It's about time to blacklist entire countries for support of spam...
This idea has a lot of merit, and realistically, most of us here on
Perhaps the thing to do is the following: Respond to *everything*. Click every link in every spam you get (well, almost every spam - I refuse to find out what Wanda really does with animals), including the almost-always bogus "unsubscribe me" links. Yes, this will make things worse short-term, but if enough people do it, then the spammers will have an incentive to filter those people out, thus for the first time giving the spammers an incentive to *remove* people from thier mailing lists.
It's ugly, but poisoning the waterhole has been known to be effective in getting people to move on, and I don't know about yours, but mine is already pretty thoroughly poisoned. Perhaps we need to poison the waterholes of the customers of Mr. Ralsky and his slimy ilk.
I'm seriously considering clicking spam links and responding to spam messages whenever I find myself on hold, for instance. (If we spent half of our current
My grand-dad went to buy a computer at a big chain store. He just wanted to look around on the web and email some friends/family. The sales droid tried to sell him a P4 2Ghz with all the bells and whistles. I ended up putting to gether a Duron 1.2G for $250 that does all he wants to do. Unless you are a big game freak or a geek (like most of us), people just don't need that much computing power.
Hear, hear. For all practical purposes, once we passed 300 MHz, it's been impossible to tell a difference in performance from a user perspective.
This is especially true with a relatively lightweight OSes like Win98/SE/ME and contemporary versions of applications like Office 97. I can't believe I'm saying that, but sadly, Linux is now far more of a resource pig than 9x, especially with the incredibly obese Gnome environment and other bloat that's been thrown in over the past couple of years in the mad rush to ape Microsoft. (The days of Linux running on machines like my 75 MHz 16MB Libretto are over, which is why it's stuck on Mandrake 7.1 with the older, much slimmer KDE, which it runs just fine, thank you.)
Sadly, both Windows and most current Linux distros expect 500+ MHz CPUs and 128 MB+ or RAM, which is just ridicuous - that's far more power than most full 3D Sun or SGI workstations had just a few years ago.
Win9x has its warts, that's for sure, but it runs great on slow hardware, as do older versions of Linux. (Sure, it's possible to build a slim Linux distro on your own, but almost all of us have better things to do with our time...)
For 99.5% of computer users, Win9x will do all they need - browsing, e-mail, and light document production. It also has the substantial advantages of having copies available all over the place (legitimate, cheap, used copies, even), and the bigger advantage of interoperating seamlessly with the rest of the world, something that's still not true for Linux and its associated office apps, which is why I prefer a Windows desktop with Unix extensions rather than the other way around - one way works flawlessly, the other is a constant battle.
so by the time you get the good drives and the ones with big caches.... you're spending the same money as us SCSI guys do.
No way. And big drive caches don't really buy you much in a RAIDed environment. It's far better to have that cache in the RAID controller itself. Drive speed doesn't matter much, it's spindle spreading that wins big, and IDE drives are now nearly as fast as SCSI at a quarter or less of the price!
Funny... I've enjoyed Ultra 160 raid for over 2 years now... and what have you IDE guys had?
Speaking for myself, I've had high-performance IDE RAID arrays that will easily saturate an Ultra 160 SCSI connection at a cost so low I can use RAID 0+1 instead of 5 and still save money. You just can't do that with SCSI.
IDE is for consumer and home use. SCSI is for serious work. even the manufacturers admit it... my scsi drives STILL come with 5 year warranties... IDE drives come with 1 year now... funny that eh?
Red herring if ever there was one. The mechanisms are the same and have been for years from most manufacturers. The only difference is the logic board. Aren't you glad you're paying a 4x premium for that one-drive-at-a-time SCSI interface? (All high-performance IDE RAID units have one channel per drive, totally eliminating bus contention. That approach is way too expensive to execute in SCSI.)
Finally, drive manufacturers are building "enterprise class" IDE drives now that have the same warranties as the very expensive SCSI drives. Sure, they're more than the consumer grade IDE drives, but still several times cheaper than SCSI drives that are only slightly faster. And you might want to check the reliability specs. I know of at least two drive vendors that quote *higher* MTBFs for their IDE drives than they do for their SCSI equivalents. Tell me again why I should waste my company's money on an anachronistic love-affair with an obsolete technology when that decision could result in additional layoffs?
If you don't believe me, get a clue about what's happening in the storage industry. I'd suggest InfoStor or SearchStorage as starting places. And if you think IDE is strong now, just wait until high-perfromance Serial ATA RAID controllers start to become common in a few months.
SCSI is dead. Not because it's a bad technology (although it isn't that great, either), but because it cannot compete economically.
Thats because one of the major limitations of current generation IDE is that only one device on a channel can "talk" at a time. So even if you're using a RAID card with two devices on a channel, it will be no faster than a standard IDE connection, since only one drive read/write can be done at a time.
Not at all true. There are a good many IDE (ATA, actually) RAID controllers out there that use one drive per IDE channel, and connect to the host via SCSI or Fibre Channel. (Of course, in this case there is *never* channel contention, and the weak spot is in the SCSI or of FC connection, both of which are using the SCSI protocol. This approach is FAR faster than almost all SCSI-based RAID systems out there, and much cheaper to boot. One of the advantages of using serious IDE Raid subsystems (not the cheezy desktop variety) is that the cost savings can allow you to replace RAID 5 with RAID 0+1 (sometime called 10) and still save money. I know because I've engineered and built multi-terabyte storage servers on this technology that are 2-3x faster and an order of magnitude less expensive than high-end storage servers like the IBM Shark or EMC Symmetrix. IDE *will* squash SCSI, it's not a matter of if, but when, mostly because SCSI will never be able to compete with the volume economics that produce IDE's 5-6x cost advantage. The performance advanstage of individual SCIS drives is already becoming marginal, and the speed of individual drives is nearly irrelevant anyway in a RAID environment where most of the poerformance comes from spanning mutiple splindles, not the speed of the individual disks. (This is why a properly configured RAID array of disks with average access time N can deliver average access times of significantly less than N.)
With SCSI, all of the drives on a channel can talk at the same time until the 160 MB/s that SCSI can handle is saturated.
Not even close. SCSI is a one-talker at a time bus architecture. This is one reason a good IDE RAID controller can so easily kick SCSI butt. The largest clusters and multiprocessor computers are all going to high performance IDE RAID arrays because of their superior cost, performance, and yes, reliability, since electrical problems in physical SCSI are one of the most common causes of data corruption in high performance environments, which is one of the chief reasons Fibre Channel has been so widely adopted. It too uses the SCSI protocol, and so has real weaknesses, but at least it avoids the hideous flakiness of SCSI's connector and termination scheme.
meaning the average american will get over 3600 of them a year
But at least my penis will grow by an inch or two.
I got one recently that promised to grow my penis 3 inches in 24 hours. Ouch!*! That's gotta hurt...
E-mail and the web are about to undergo the collapse phase of their CB-radio-like growth.
I mean this seriously. For a great many non-technical people, pop-ups, viruses, spam and the death of Napster are reducing the net value (groan) of the Interent to near zero. The inevitable result is that those people will simply decide that they can do without a marginal benefit that makes thier lives more difficult and complicated. Like CB radio, the shine is wearing off the Internet, and it will soon begin to *lose* people fo rhte first time ever. (Note that unlike CB, I certainly don't expect the Net to slip into technological obscurity and irrelevance, but people *will* leave - and in fact, they already are, and that's newsworthy.)
Case in point: I know one local company that is seriously considering pulling the plug on Internet access to thier network and instead designating just a few "browsing and e-mail computers" connected to the Internet. They claim they can easily justify the cost of the additional machines just by avoiding viruses, and I don't doubt them. (They've been hit by Linux viruses/worms, too, so don't get too cocky about blaming Windows.)
Spam filters are irrelevant until they can be made easy enough for the masses. I *hate* spam, but don't use much in the way of spam filtering because I don't want to spend half a week learning the obtuse intricacies of a dozen different mail processing packages, after evaluating several dozen alternatives. There needs to be an embedded box made for serious firewall, spam filter, and content filter use that is as easy to set up and use as e-smith and clarkconnect. It *has* to be embedded, because most people are NOT going to get another PC just to protect the one they use. ISPs could help, but most aren't interested, especially given their declining fortunes as the result of stupid management and decerasing revenues from the CB radio bust mentioned above.
I think the long-term impact of spam, pop-ups, and viruses could be far, far greater than most people expect. At the risk of raising the perennnial usenet cry of wolf, these things, left unchecked, could indeed herald the "imminent death of the Net" - at least as we know it today. The scary thing is that the cure could be worse than the disease...
The CD I have (admittedly, given to me by a friend - I thought it was a standard CD, but perhaps he altered it) is designed to run Smoothwall entirely from CD, and that way, even if the machine is compromised, there's no real damage done...
Tablet PCs strike me as an opportunity for Linux to crack the desktop market. From what I've read of initial impressions of Tablet PCs, the thing that sucks the most is the crappy Microsoft inking software, that is neither easy to learn to write for (sucks worse than Graffiti) nor will it learn your writing style (sucks worse than Newton).
I've spent a LOT of time as an evaluator and beta tester for non-Tablet PC tablet devices, including several based on Linux. I can tell you with some assurance that Linux is AT LEAST 5 years behind in this area. the Ms recognition is very good, in some (not all) respects better than the Newton, and as good as Graffiti if you print anything like legibly.
Right now, Linux and the BSDs have a very long way to go just to catch up, and a longer way to go before they could have any real advantage. Several fundamental problems really bite you in the butt in this environment: from the general unsuitability of X to the fact that writing a generic handwriting recognizer is even harder given the architectural (message-passing, etc.) differences in things like KDE and Gnome.
Given the fact that we're just now getting ACPI to sort of function correctly in Linux four years after MS added ACPI support to NT and Win98, I have no reason to expect that Linux developers will get good at supporting hardware built to MS specs anytime soon...
I really think the tablet PCs would have to be thin clients to conserve battery life. I also don't think that a M$ OS is the way to go. The bloat in OS would just use up so much memory and processor cycles, that the battery life would suffer.
As I point out in another post, we can always hope that having XP in Tablet PCs will finally motivate MS to put XP on a badly-needed diet. That would be a win-win all the way around, except that Intel would find it harder to convince people they need 4 GHz CPUs to check their mail...
Furthermore, with the release of TabletPC, Microsoft has shown again that they simply can't innovate. Microsoft's TabletPC software is the same old stuff we had 10 years ago, only in a more bloated software incarnation. The only thing that has really gotten better is the hardware and processor speed, as well as the quality of real-time graphics those machines support.
At the risk of defending Microsoft, you're wrong. I've been a beta tester for several types of tablet computers (although not a Tablet PC), and nearly all of them are atrocious. Microsoft has done an excellent job at taking another swipe at building what should be the highest and best form factor for the "Dynabook" concept. The biggest thing they've done is to finally build a tablet that has all of the following:
1) a real OS (Whether or not you like it, XP is a real OS with pretty nearly all the things that were the domain of Unix alone just a few years back. You can do pretty much anything with XP that you can do with any other real OS. (With the possible exception of heirarchical mounts: Is the totally broken MS-DFS still the only way to fake that, or has something been added to XP - I haven't used that since W2K was brand new...))
It's hard to overstate the importance of having a Tablet PC device that can run and leverage the ordinary software run on ordinary PCs. That's just never been done before. Even Microsoft's own previous attempts were marred by ignoring this important aspect and trying to pretend CE was a real OS, when it is anything but, or grafting seriously substandard pen facilities onto Windows.
2) Real handwriting recognition that works as well as is possible given the state of the art, and digital ink to allow either raw ink capture and/or deferred recognition. In my estimation, the only other machine to have truly usable recognition is the Newton (although Calligrapher on CE came close), and as I said, I've used a lot of these things.
3) A nice, color, high-resolution display. This is vital for non-trivial applications. Ultimately, we still need native resolutions of paper-sized screens that are an 4-6x what we have today, but this is still a big first step. Interestingly, total screen pixel count may grow here before the desktop catches up.
4) Real networking capability. It's surprising how many devices like this assume they're islands and really don't communicate well with the rest of the world. The fact that 802.11 is taking over the world just now and can be fully levereaged by this device is just good timing for MS, and one more reason these things will probably be wildly successful.
5) To allow the good handwriting recognition, a high-quality, high resolution digitizer suitable for all but the pickiest of graphic artists. Try the awful digitizers in some other tablet devices and you'll see what I mean. My only gripe is that MS specifies active digitizers and pen buttons, maening the stupid things need batteries. Grrr.
6) Not vital, but nice: enough storage to carry around all the data most users will care about. This has staggering implications, if you think about it...
The only real downside now is another one related to XP: the devices require a LOT of compute power, and consequently suck down a lot of battery power. (With luck, the Tablet PC will give MS a good incentive to slim down the bloat in XP - there HAS to be quite a bit of room for slimmer code there... ) Fairly quickly, they need to get to the point that a Tablet PC can run an entire shift on a single charge - that's particularly important for vertical markets like healthcare, which is expected to be an early adopter of this technology. That won't happen on a widespread scale until a full shift is possible. Ultimately, they need to produce a Tablet PC that can be thrown in the briefcase for a one week trip without bothering to pack a charger. I can do this with my Kyocera/Palm SmartPhone, and should be able to do it with my computer, too.
In one sense, it;s fair to say that MS has not brought much new to the table. But at the same time, it;s also quite fair to say that no one has ever before brought together so many of the the things that a Tablet PC must have to be successful, and Microsoft deserves some credit for that. It's a good integration job, even if most of the pieces and ideas came from elsewhere. Ultimately, this is the sort of thing Microsoft occasinally shows themselves to be good at. Let's wish them well, since these capabilities will not be available in other forms unless they can succeed in establishing a market for these things - that's the lesson of volume economics in the computer and electronics space...
They said the Tablet-PC would open "new markets", since people who usually avoid using computers because of the mouse (Have you ever tried to teach it to you mother? :-) now have a more "natural" way of interfacing with it - the pen.
I believe it was Ted Nelson that said (possibly quoting (or jabbing at) Doug Engelbart) something to the effect that, "People discovered thousands of years ago that it was much, much easier to draw with sticks than rocks, but that lessson won't be absorbed by the computer industry any time soon..."
Lastly, I humbly suggest that you haven't had your fingers in anything other than the most exceptional of open source efforts. I have seen plenty of open source, and it is as bad and worse as the worst closed source software that I've had the misfortune of seeing. Open source is no cure for software quality; software quality happens when people care about it, open source or no.
/.) to regard all open source software as inherently of higher quality than commercial software. In reality, nothing could be further from the truth. I fully support open source software, but the reality is that as bad as much commercial software is, most open source software is worse. There are the notable exceptions (Apache, Samba, etc.), but they are popular only because they are exceptions to the usual sub-mediocre quality of most open source software. (How many open source projects have any sort of formal testing process other than, "toss it out there and see what bug reports we get back?" Commercial software has its weaknesses, but the concept of regression testing, etc. is far more common in commercial development than it is in open source development - that's a fact. In addition, there are very few open source projects that are not primarily attractive to propellerheads - there is very little suitable for the average user, and what's there is generally of far lower quality than the usual commercial equivalents. (BTW: Star/OpenOffice doesn't count, since it was originally developed as a commercial product and opened up only through Sun's benevolence after they bought Star Division.)
Somebody mod this up as Insightful!
There is an incredibly misguided tendency (esp. here on
Add this to the fact that most open source software is simply "me-too" functionality (best whiny voice: "See, Linux can too play DVDs, and it can sort-of print, well, to a pitiful handful of expensive printers, anyway...") rather than anything really new and innovative, and it's easy to see why we may always be chasing Microsoft's heels.
As a final for-instance, take Longhorn: Whether or not one likes Microsoft, this is a set of technologies that will have real and significant value to the users. Given the fairly fundamental nature of some of the changes (DB-based filesystem, transparent integration of information across applications, etc.), Longhorn will set open source software back another two years until its functionality has been duly duplicated in a different, fatter, CORBA-compliant way. (If you don't believe me, check out how dot-Net has thousands of open source developers tied in knots chasing their tails and adding bloat that may eventually eclipse that of Microsoft's own implementations.)
Utimately, Courageous is dead on about one thing: "Open source is no cure for software quality; software quality happens when people care about it, open source or no."