Gee , a couple of 2-axis Analog Devices AXDL accelerometers to detect tilt (and hence general orientation), a GPS for position (also altitude, and atomic time), a camera, a background process doing anonymous publish-to-web (or publish-to-P2P-mapping network) turned on by default (and publishing updates when it detects cheap bandwidth is available)... cartographic and mapping agencies could be replaced in a few decades (or augmented by 3-D world synthesis, automated geographic recognition based on image matching (a la the movie Red Planet)...
GPS-based position detection needs the open sky though (which is why I think this guy's idea depends on cellular networks for position detection) -- unless the GPS and acclerometers were augmented by tiny gyroscopes to become a rudimentary intertial navigation system.
"How could the EDR data make it into court at all? I would have hoped the guy's lawyer would have jumped on the 5th Amendmend problems with this from the very beginning."
This makes no sense at all. By this token... Imagine you shot someone. The police suspected you of the shooting; then took in your gun in for ballistics analysis of the firing pin and barrel markings, etc. Should you be able to ban the ballistics evidence from being presented up in court, since it could be incriminating?
I think the 5th amendment is protection against your own confession being used against you - you can't use it to supress evidence. Just because you own the gun (or you own the EDR that was part of the car that you killed someone with) doesn't give you the right to subvert justice. The right of the law to consider evidence to make a fair judgement, especially in a criminal case, ****overrides**** other rights, including property right, that you may have.
Paper #1) Danish and Middle East population could have diverged 4,500 years ago ----> Fits with the Biblical description of human dispersion occuring after the flood (around 4,500 years ago as well).
Paper #2) 20 times faster observed mtDNA Mutation Rate ----> Genetic bottlenecking can be approximately just 150,000/20 = 7,500 years old. Fits Biblical description of "bottlenecking" down to Noah's family 5,000 years ago.
Paper #3) 1 male root lineage / 3 sub-lineages / only 1 of these 3 has 7 sub-sub-lineages that populate the world outside of Middle East and Africa. ----> Remarkable fit with Biblical story of Noah, his 3 sons, and the 7 descendants of only one of the 3 sons ("Japeth") populate the rest of world. The other 2 sons and their descendants populate the Middle East and Africa.
The PocketPC *does* support cursive handwriting recognition and has for some time. My older Ipaq 3870 with Windows CE 3.0 (not PocketPC 2002) recognizes my cursive handwriting with about 80-90% accuracy.
I am a Linux enthusiast too, but the Zaurus you mentioned looks interesting only as far as the hardware specs are concerned. WinCE/PocketPC would run well on this hardware too. It's the software application support that, collectively, are the "killer app" for the PocketPC.
I don't know if I would classify excluding X a mistake. But I would love to know of a Zaurus developers program, with nice integrated developer tools like MS offers.
"Those numbers don't mean much: the best-selling PDA was the Zire, at sub-$100 prices. That skews dollar-based market share numbers in favor of the more expensive PocketPC handhelds.
Hi 73939133 - it appears these numbers aren't dollar-figures -- they are based on units sold. For instance, the report states PocketPC sales went up from 2,800,000 to 3,100,000 units in 2002. "3,100,000 units" tallies with the 25.7% share of 12.1 million units that the Gartner report estimates for PocketPC.
If Palm lost numeric market share despite cheap Zires, it must be doing something wrong by customers. MS is doing a pretty good job offering heaps of functionality with PocketPCs and I think customers are simply appreciating that. I've not used the Zaurus or the Yopy, but from what I understand, these two still lag a little behind the PocketPC in the technology stakes -- for example, no cursive handwriting recognition (?).
> "I want to put a GPS unit in my car... using Bluetooth? "
Funny... I had the same idea as yours, except I was also thinking of using a Solar cell to power the Bluetooth/GPS unit instead of using the car's battery. This would make the unit completely cordless. It would mount under in the front of the dash, under the windshield.
I was surfing this discussion in the hope someone would point to cheap Bluetooth modules. Didn't fine any, but I've posted this comment previously with a source for $60 Bluetooth modules.
"PocketPC doesn't really have an advantage over Linux on PDAs... That's, ultimately, why Palm won out over Newton."
Well, PocketPC may win the PDA war.
July's issue of "PocketPC" magazine, as well as Issue 11 of "PDA Essentials", a British PDA mag published every 6 weeks, both report a Gartner study of PDA shipments in 2002. In general, PDA sales declined by 9% in 2002. However, while Palm lost 12% of it's market share (down to 55%), PocketPC *gained* 5% market share (up to 25.7%). Total sales were 12.1 million units.
I personally think Pocket PC offers a lot of sophistication that users appreciate and Palm is now scrambling to offer.
"The Sharp Zaurus- all models- have character recognition, not HWR. " "Is it not considered proper handwriting recognition what the Zaurus does?"
(I haven't used the Zaurus) I have an IPaq Pocket PC - the MS-provided software includes a cursive writing recognizer which, for my writing, is about 90% accurate.
This would require multi-link PPP support from the ISP end though, right? I mean, the ISP would need to "stitch together" the TCP/IP packets so they appear to originate from a single real-world IP address.
I guess what I was trying was to use a proxy so that MPPP support wasn't required anymore (so for example, it would still work if one PPP connection was made to AOL, and the second to a different ISP)
Regarding: " "the companies that are sending these items are directly bearing the cost of your DoS." Costs passed on to the consumer." etc..
Dude (Guppy06(410832)) -- You are wrong.
"If you're not giving them Ralsky's address, rest assured that they're probably interested in buying his address... "
Firstly - Two wrongs don't make a right - If Ralsky is does something wrong, it isn't right to *lie* to hundreds of companies to get them to send him junk mail.
Secondly - as others have pointed out here, your "it's not hurting anyone else" argument is false. For example, this is one of your justifications: " depending on how much they're shipping and where, it may actually be cheaper for them to add in a few extra addresses to bump the mailing into the next rate " This is a classic example of wishful thinking -- the lucky company that hits this "price break" only gets a fractional reduction per catalog. On the average, most companies will *not* hit the "lucky break". End result -- real money, real trees, real petrol, real effort -- are being expended in mailing Ralsky catalogs by companies who have been lied to, to get them to do this.
> Well, actually you'd need multiple telephone lines:-) Hey, thanks for the reply. Yes, I meant telephone lines:-)
> Diamond (or whatever they were most recently named before > finally going out of business) used to advertise something > they called "shotgun" modems; you bought two of them for one > computer and used one on one phone line and the other on a > second phone line. I remember "Shotgun":) So they're out of business, are they? I wanted to avoid that approach -- they were what I was thinking of them when I mentioned avoiding "channel bonding".
> Apparently in addition to whatever software came with the modems > (probably Windows only) you had to get your dial-up ISP to go > to some effort to support it as well. Yes, their approach required the ISP to have similar modems + software too. I was hoping to avoid this by using a proxy server to split the HTTP requests among multiple interfaces (each interface corresponding to a net connection). While this approach is more coarse-grained than TCP/IP-level channel bonding, it does have the advantage of not requiring ISP or hardware support at all - you could use accounts from different ISPs assigned different IP addresses; carried over any combination of phone lines, network connections, cellular connections, WLAN hotspots, etc. Besides the speedup, this could be useful in network quality of service issues as well, since different channels have differnt latency, bandwidth and jitter properties.
If I have multiple telephones, can a proxy combine two or more 56K connections to get a larger pipe?
i.e. If I view a page that references p1.png and p2.png; can a proxy server take the two HTTP requests for p1.png and p2.png, and send each request out through a seperate 56K connection?
I know some ISPs support channel bonded modems - but these are pretty expensive. Has anyone heard of a pure proxy server that can implement this sort of thing?
Hi ramen - I guess you're referring to the same point made here.
I don't disagree -- a USB-only mac would have helped spur USB adoption. But Apple was clearly a follower in this case, leveraging the widespread adoption of a royalty-free standard. According to the article itself, USB was already supported by a "vast army of cloners" before the imac came out (the article is dated Aug 98). In other words, Intel and party had been successful at getting motherboard makers to provide USB ports! The device support was just a matter of time.
This Firewire v/s USB article gives a good perspective on the whole issue. While Firewire is the better engineered standard, there were questions about Firewire's power usage and the high license fees being charged by Apple.
All -- Thank you for your polite responses to my original post. Some moderator took a rather dim view of it though - modding me down to zero.:)
Can someone enlighten me how the usage (described below) of two Ecommerce dialects -- CommerceOne's xCBL and Ariba's cXML -- is classified?
I don't think the usage below can be classified as either Web Service, XML-RPC, SOAP, or REST. However, this usage comprises the majority of XML-based Ecommerce transactions I see in mainstream business these days (I work for a large office supplies company).
"The Usage" ----------------- A business procedure (say, "place a purchase order") is executed by a simple transfer of an XML document. This XML document conforms to the standard it follows -- i.e. it is valid XML according to the standard's DTD/schema for that particular transaction.
For example, to send an XML purchase order via HTTP POST, the buyer would do the following sort of POST (HTTP protocol approximated below):
HTTP POST.... Content-Type: text/xml
<PurchaseOrder>
<Blah....
Note: the content type is "text-xml". There are no CGI params -- the XML document itself comprises the body of the HTTP POST. The seller sends back a synchronous response document (same TCP connection as the POST, but reverse direction). This 'sync response' must carry an HTTP 200 response header, but can also carry an XML document (this depends on the standard).
HTTP 200/OK Content-Type: text/xml
<ConfirmationOfPurchaseOrder>
<Blah....
The submitter party processes this response.
Note, there are no standardised 'sections' of XML in the document (like there are with SOAP or Web-Services (unsure?) ). Also, unlike REST, there is no usage of other HTTP methods - DELETE, GET, etc -- only POST. The XML document itself specifies the procedure being carried out (a richer, and hence better, approach).
For example, to do an order cancellantion, another XML document is POST-ed.
So, does this sort of usage have a name (besides the name of the business standard)? What advantages do other approaches like Web Services bring to the table?
While Apple is responsible for many innovations, USB isn't one of them. USB was initially (co-)developed, and pushed heavily by Intel.
From this site: Q2: Who created USB? A2: USB was developed by a group of seven companies that saw a need for an interconnect to enable the growth of the blossoming Computer Telephony Integration Industry. The seven promoters of the USB definition are; Compaq, Digital Equipment Corp, IBM PC Co., Intel, Microsoft, NEC and Northern Telecom.
I'm pretty sure Apple adopted it only later; Firewire is considered faster, but a more expensive standard to implement.
This may seem a strange idea, but what about "Email address portability"?
This could be offered by a group of ISPs who competed to offer you dialup or broadband service, but kept your email address constant by sharing administration of the mail-servers for a set of generic domains (For eg: John.Smith@email.com. Or, using the.name TLD,: William@Smith.name, billy63@William.Smith.name)
Email forwarding services like iname.com do exist, but that option suffers the same lack of competition as the "ISP-provided-email" option. i.e. If you switch your email forwarder (say due to a hike in annual fees), you lose your email address.
The existence of companies like Propel.com implies that performance can still be squeezed from plain old 56K connections. So how can we get implement similar functionality for free?
From Propel's website (the link in your post above):
* Compression. Propel delivers text and graphics more efficiently, using a patent-pending compression technology that allows for intelligent encoding of Web pages and page elements sent to your browser. It looks like Propel has implemented something similar to Akamaitech - a network of geographically dispersed network caches available to subscribers. This is impossible to replicate for free. (Right ?)
* Caching. Propel intelligently retains and re-uses Web pages and page elements that have previously been sent to your PC. That's why the longer Propel Accelerator is in use on your PC, the faster your Web pages will load. Now how would this work? Maybe I could set my browser to cache *everything* (even disobeying a page's cache directive). And, in the background, the browser could figure out if the remote page had been updated (say, with HTTP HEAD) and if so, display a transition effect to the updated page.
* Persistent Connections. Propel uses proprietary techniques to carefully manage and optimize the communication between your modem and our servers using a persistent connection. Hmm. Sounds like HTTP pipelining could help here. Also, how about implementing a DNS cache and a proxy server on the local machine (the one running the browser) -- would that help?
Re:CDAC setup to build supercomputers
on
TiVo++ from India
·
· Score: 1
> > "I think every country needs a *balance* of free trade and protection of weaker industries" > Sure, if you want to be poor. In the short term, allowing weak > industries to fail may be harmful to the people who work in them, > but in the long term, "protecting" them makes everyone poor.
You have a point, but so many rich country impose a (small) amount of protectionism and still prosper. For instance, even the US imposes anti-dumping duties when it thinks excessively cheap goods are being imported. IIRC, it also has farm subsidies, as well as massive amounts of investment for industries deemed nationally important, like defense. The same with the EC (farm subsidies, aerospace industry subsidies), Japan (protectionism for rice imports, electronics, and cars ), etc. I agree though, that the richest countries have the highest levels of free trade - free-trade does make an economy more efficient. But it works only when the country has a functioning welfare-net -- so that people booted out of a job are guarenteed basic food, housing and shelter. If people are kicked out of a job and starve to death, the government will be forced to change it policies pretty soon, or will be changed itself.
If I understand correct, Western countries and Japan set up functioning welfare networks in the early 50s (??), and maintain it to this day. Many developing countries are just setting up working welfare networks now.
> unless they're selfishly trying to keep their own job at other people's expense. Democracy is the rule of the people and most people are selfish. They task their elected representatives to balance the benefits of free trade, with the pain associated with it.
CDAC setup to build supercomputers
on
TiVo++ from India
·
· Score: 5, Insightful
... if this is one of the dreaded consequences of outsourcing jobs to India.
Hmm - No. Not due to job outsourcing... but its certainly a result of technology that was born in the US.
I said "No" because the people behind this are (from the article) "founder members of CDAC, the brains behind India's PARAM-supercomputer". If I recall correctly, CDAC was setup by the Government of India in the late 80s as a direct consequence of the US *withholding* export of supercomputers to India for fear it would be used for defense research (more specifically, nuclear research). As a result, the CDAC people built massively parallel supercomputers from off-the-shelf CPUs (IIRC, they still used American CPUs - off-the-shelf 8086s (?) to begin with). They have some very cluey guys with a lot of experience born from research efforts like creating the complex electronics for interfacing supercomputers. Now it seems some of those people are moving to the private sector - kind of like with Govt. spending jumpstarting the computer revolution in the US.
A sign of what's to come? Is this the result of the US losing their position as main providers of R&D? What will be left afterwards? An economy of service?
I think every country needs a *balance* of free trade and protection of weaker industries. A "we can sell to you, but you can't sell to us" mentality is ultimately is bad for everyone concerned; from what I understand, 2/3rds of US income derives from exports.
At the end of the day, I'm sure your leaders have an eye on industry and employment figures. If not, you elect new ones.
> Nor is science ruled by appeals to authority, which is exactly > what you did when you pointed out that your creationist priest has a PhD.
The only reason I mentioned Dr. Batten's credentials is because you said concerning people like him: "Do you really think those crackpots have any idea what they're talking about?"
> isn't a surprise because creationists often seem to have no sense of humor. You bore me.
OK:) I take it you won't reply. Bye then. I hope you will seriously consider the issues raised in those 3 papers again someday.
Paper #1) Danish and Middle East population could have diverged 4,500 years ago ----> Fits with the Biblical description of human dispersion occuring after the flood (around 4,500 years ago as well).
Paper #2) 20 times faster observed mtDNA Mutation Rate ----> Genetic bottlenecking can be approximately just 150,000/20 = 7,500 years old. Fits Biblical description of "bottlenecking" down to Noah's family 5,000 years ago
Paper #3) 1 male root lineage / 3 sub-lineages / only 1 of these 3 has 7 sub-sub-lineages that populate the world outside of Middle East and Africa. ----> Remarkable fit with Biblical story of Noah, his 3 sons, and the 7 descendants of only one of the 3 sons ("Japeth") populate the rest of world. The other 2 sons and their descendants populate the Middle East and Africa.
No no, not theoretical as in, they didn't exist: theoretical as in conceptually defined. Look, we're splitting hairs here. OK, you think they're 'conceptually defined'. At the end of the day, can you just give me real reasons why a *starting population of two* does *not* fit this data? And please don't give me something like "but Y-Adam and m-Eve are millenia apart". You know well that, given experimental error in these techniques, the few millenia difference in the estimate of around 150,000 years (from the NY Times article) could well converge down to 0.0.
> The bottleneck itself... it reflects a basic geometric necessity. > there would still be a relatively _recent_ mEve,
Can you explain how a genetic bottleneck is a "geometric necessity"? After all, the NY Times article does state the m-Eve bottleneck occured when *all* the other women had only sons, while the Y-Adam bottleneck occured when *all* the other men only had all daughters, at some point in time. As the general population grows, I would think this would be a rarity, rather than a geometric necessity.
--If the amount of radioactivity increased dramatically due to the flood (as described in my earlier post), fossils created before the flood would look artificially "old" since post-flood fossils would have higher amounts of radio-carbon to begin with.---
An increase of such a level would have also irradiated life on Earth, including any Noah.
Yes that's right. It did. But perhaps you didn't get my point, which is: The post-flood increase in radio carbon caused it to spike and eventually stablize at the levels we see today. Its likely there was an initial spike in radiation levels right after the flood. This caused Noah's direct descendants to have higher mutation rates -- the distinct "1/3/rest-of-world-7" mutation pattern seen in the NYTimes article (that corresponds to the Biblical description of human ancestry.) Eventually though, the Radio-carbon levels would have subsided down to the (still *much* higher than pre-flood) levels we see today. IIRC, a paper I'd read (don't have a link handy) observed a similar effect among children born after the Chernobyl nuclear accident. The Bible does note a vastly decreased human lifespan after the flood.
Note, I'm not saying radio-carbon dating is hocus-pocus. I'm saying the assumption evolutionists often make, that we can date simply everything since we know all there is to know about ancient concentrations of radio-carbon levels and they've been really smooth and constant -- now that's hocus pocus.
---Critics thrust back saying techniques like Radio-carbon dating give older ages....---
> So, I see. If we pick and choose whatever evidence... > radio-carbon dating is harly the only technique used for these various issues, > and indeed it is not even the most important for determining key ages of the earth.
A well-known example of a problem with isochron dating (taken from this article) occured when a prestigious lab dated recent samples as very old: The samples were sent progressively in batches to Geochron Laboratories in Cambridge, Boston (USA), for whole-rock potassium-argon (K-Ar) dating... Geochron is a respected commercial laboratory, the K-Ar lab manager having a Ph.D. in K-Ar dating. No specific location or expected age information was supplied to the laboratory. However, the samples were described as probably young with very little argon... The 'ages' range from which were observed to have cooled from lavas 25-50 years ago.
---Ultimately I can't *prove* God to you without a doubt.---
> Why would you need to prove God to me?... > The existence of God is an indepedant question.
OK...Sorry, I'll narrow down my phrasing... It should have read:
"I can't prove the existence of the God whose works are described in the Bible..."
The Bible tells us of a God (named simply "I AM"), creating the earth and all life on it. For example, in the old testament, God says "I have made the earth and created man upon it" (Isa 45:12). And in the new testament, Jesus says "But from the beginning of the creation, God made them male and female" (Mark 10:6, talking in the context of divorce.) The Bible also lists the lifespans and lineages of personalities from Adam to Christ, tracing a chain used to calculate the Biblical age of the earth.
Slashdot Mirror
Gee , a couple of 2-axis Analog Devices AXDL accelerometers to detect tilt (and hence general orientation), a GPS for position (also altitude, and atomic time), a camera, a background process doing anonymous publish-to-web (or publish-to-P2P-mapping network) turned on by default (and publishing updates when it detects cheap bandwidth is available)... cartographic and mapping agencies could be replaced in a few decades (or augmented by 3-D world synthesis, automated geographic recognition based on image matching (a la the movie Red Planet)...
GPS-based position detection needs the open sky though (which is why I think this guy's idea depends on cellular networks for position detection) -- unless the GPS and acclerometers were augmented by tiny gyroscopes to become a rudimentary intertial navigation system.
"How could the EDR data make it into court at all? I would have hoped the guy's lawyer would have jumped on the 5th Amendmend problems with this from the very beginning."
This makes no sense at all. By this token... Imagine you shot someone. The police suspected you of the shooting; then took in your gun in for ballistics analysis of the firing pin and barrel markings, etc. Should you be able to ban the ballistics evidence from being presented up in court, since it could be incriminating?
I think the 5th amendment is protection against your own confession being used against you - you can't use it to supress evidence. Just because you own the gun (or you own the EDR that was part of the car that you killed someone with) doesn't give you the right to subvert justice. The right of the law to consider evidence to make a fair judgement, especially in a criminal case, ****overrides**** other rights, including property right, that you may have.
The Bible is correct.
Read this post discussing these papers/articles:
Paper #1) Danish and Middle East population could have diverged 4,500 years ago
----> Fits with the Biblical description of human dispersion occuring after the flood (around 4,500 years ago as well).
Paper #2) 20 times faster observed mtDNA Mutation Rate
----> Genetic bottlenecking can be approximately just 150,000/20 = 7,500 years old. Fits Biblical description of "bottlenecking" down to Noah's family 5,000 years ago.
Paper #3) 1 male root lineage / 3 sub-lineages / only 1 of these 3 has 7 sub-sub-lineages that populate the world outside of Middle East and Africa.
----> Remarkable fit with Biblical story of Noah, his 3 sons, and the 7 descendants of only one of the 3 sons ("Japeth") populate the rest of world. The other 2 sons and their descendants populate the Middle East and Africa.
The PocketPC *does* support cursive handwriting recognition and has for some time. My older Ipaq 3870 with Windows CE 3.0 (not PocketPC 2002) recognizes my cursive handwriting with about 80-90% accuracy.
I am a Linux enthusiast too, but the Zaurus you mentioned looks interesting only as far as the hardware specs are concerned. WinCE/PocketPC would run well on this hardware too. It's the software application support that, collectively, are the "killer app" for the PocketPC.
I don't know if I would classify excluding X a mistake. But I would love to know of a Zaurus developers program, with nice integrated developer tools like MS offers.
"Those numbers don't mean much: the best-selling PDA was the Zire, at sub-$100 prices. That skews dollar-based market share numbers in favor of the more expensive PocketPC handhelds.
Hi 73939133 - it appears these numbers aren't dollar-figures -- they are based on units sold. For instance, the report states PocketPC sales went up from 2,800,000 to 3,100,000 units in 2002. "3,100,000 units" tallies with the 25.7% share of 12.1 million units that the Gartner report estimates for PocketPC.
If Palm lost numeric market share despite cheap Zires, it must be doing something wrong by customers. MS is doing a pretty good job offering heaps of functionality with PocketPCs and I think customers are simply appreciating that. I've not used the Zaurus or the Yopy, but from what I understand, these two still lag a little behind the PocketPC in the technology stakes -- for example, no cursive handwriting recognition (?).
> "I want to put a GPS unit in my car... using Bluetooth? "
Funny... I had the same idea as yours, except I was also thinking of using a Solar cell to power the Bluetooth/GPS unit instead of using the car's battery. This would make the unit completely cordless. It would mount under in the front of the dash, under the windshield.
I was surfing this discussion in the hope someone would point to cheap Bluetooth modules. Didn't fine any, but I've posted this comment previously with a source for $60 Bluetooth modules.
"PocketPC doesn't really have an advantage over Linux on PDAs ...
That's, ultimately, why Palm won out over Newton."
Well, PocketPC may win the PDA war.
July's issue of "PocketPC" magazine, as well as Issue 11 of "PDA Essentials", a British PDA mag published every 6 weeks, both report a Gartner study of PDA shipments in 2002. In general, PDA sales declined by 9% in 2002. However, while Palm lost 12% of it's market share (down to 55%), PocketPC *gained* 5% market share (up to 25.7%). Total sales were 12.1 million units.
I personally think Pocket PC offers a lot of sophistication that users appreciate and Palm is now scrambling to offer.
"The Sharp Zaurus- all models- have character recognition, not HWR. "
"Is it not considered proper handwriting recognition what the Zaurus does?"
(I haven't used the Zaurus) I have an IPaq Pocket PC - the MS-provided software includes a cursive writing recognizer which, for my writing, is about 90% accurate.
Funny you say that... http://www.linradio.com/
This is a software-defined-radio PCI card.
> I could get around 25-30k/sec downloads
Wow - 25 KBytes/sec! over 2 modems - Wow!
> multi-link PPP driver
This would require multi-link PPP support from the ISP end though, right? I mean, the ISP would need to "stitch together" the TCP/IP packets so they appear to originate from a single real-world IP address.
I guess what I was trying was to use a proxy so that MPPP support wasn't required anymore (so for example, it would still work if one PPP connection was made to AOL, and the second to a different ISP)
Gee, I didn't get what you meant by this comment. I mean, once you have cable, there's really no need for two telephone lines, right? :-)
Regarding:
" "the companies that are sending these items are directly bearing the cost of your DoS."
Costs passed on to the consumer."
etc..
Dude (Guppy06(410832)) -- You are wrong.
"If you're not giving them Ralsky's address, rest assured that they're probably interested in buying his address... "
Firstly - Two wrongs don't make a right - If Ralsky is does something wrong, it isn't right to *lie* to hundreds of companies to get them to send him junk mail.
Secondly - as others have pointed out here, your "it's not hurting anyone else" argument is false. For example, this is one of your justifications:
" depending on how much they're shipping and where, it may actually be cheaper for them to add in a few extra addresses to bump the mailing into the next rate "
This is a classic example of wishful thinking -- the lucky company that hits this "price break" only gets a fractional reduction per catalog. On the average, most companies will *not* hit the "lucky break". End result -- real money, real trees, real petrol, real effort -- are being expended in mailing Ralsky catalogs by companies who have been lied to, to get them to do this.
Is that good?
> Well, actually you'd need multiple telephone lines :-) :-)
:) So they're out of business, are they? I wanted to avoid that approach -- they were what I was thinking of them when I mentioned avoiding "channel bonding".
Hey, thanks for the reply. Yes, I meant telephone lines
> Diamond (or whatever they were most recently named before
> finally going out of business) used to advertise something
> they called "shotgun" modems; you bought two of them for one
> computer and used one on one phone line and the other on a
> second phone line.
I remember "Shotgun"
> Apparently in addition to whatever software came with the modems
> (probably Windows only) you had to get your dial-up ISP to go
> to some effort to support it as well.
Yes, their approach required the ISP to have similar modems + software too. I was hoping to avoid this by using a proxy server to split the HTTP requests among multiple interfaces (each interface corresponding to a net connection). While this approach is more coarse-grained than TCP/IP-level channel bonding, it does have the advantage of not requiring ISP or hardware support at all - you could use accounts from different ISPs assigned different IP addresses; carried over any combination of phone lines, network connections, cellular connections, WLAN hotspots, etc. Besides the speedup, this could be useful in network quality of service issues as well, since different channels have differnt latency, bandwidth and jitter properties.
If I have multiple telephones, can a proxy combine two or more 56K connections to get a larger pipe?
i.e. If I view a page that references p1.png and p2.png; can a proxy server take the two HTTP requests for p1.png and p2.png, and send each request out through a seperate 56K connection?
I know some ISPs support channel bonded modems - but these are pretty expensive. Has anyone heard of a pure proxy server that can implement this sort of thing?
Hi ramen - I guess you're referring to the same point made here.
:)
I don't disagree -- a USB-only mac would have helped spur USB adoption. But Apple was clearly a follower in this case, leveraging the widespread adoption of a royalty-free standard. According to the article itself, USB was already supported by a "vast army of cloners" before the imac came out (the article is dated Aug 98). In other words, Intel and party had been successful at getting motherboard makers to provide USB ports! The device support was just a matter of time.
This Firewire v/s USB article gives a good perspective on the whole issue. While Firewire is the better engineered standard, there were questions about Firewire's power usage and the high license fees being charged by Apple.
All -- Thank you for your polite responses to my original post. Some moderator took a rather dim view of it though - modding me down to zero.
I don't think the usage below can be classified as either Web Service, XML-RPC, SOAP, or REST. However, this usage comprises the majority of XML-based Ecommerce transactions I see in mainstream business these days (I work for a large office supplies company).
"The Usage"
-----------------
A business procedure (say, "place a purchase order") is executed by a simple transfer of an XML document. This XML document conforms to the standard it follows -- i.e. it is valid XML according to the standard's DTD/schema for that particular transaction.
For example, to send an XML purchase order via HTTP POST, the buyer would do the following sort of POST (HTTP protocol approximated below):Note: the content type is "text-xml". There are no CGI params -- the XML document itself comprises the body of the HTTP POST. The seller sends back a synchronous response document (same TCP connection as the POST, but reverse direction). This 'sync response' must carry an HTTP 200 response header, but can also carry an XML document (this depends on the standard).The submitter party processes this response.
Note, there are no standardised 'sections' of XML in the document (like there are with SOAP or Web-Services (unsure?) ). Also, unlike REST, there is no usage of other HTTP methods - DELETE, GET, etc -- only POST. The XML document itself specifies the procedure being carried out (a richer, and hence better, approach).
For example, to do an order cancellantion, another XML document is POST-ed.So, does this sort of usage have a name (besides the name of the business standard)? What advantages do other approaches like Web Services bring to the table?
USB... Apple
While Apple is responsible for many innovations, USB isn't one of them. USB was initially (co-)developed, and pushed heavily by Intel.
From this site:
Q2: Who created USB?
A2: USB was developed by a group of seven companies that saw a need for an interconnect to enable the growth of the blossoming Computer Telephony Integration Industry. The seven promoters of the USB definition are; Compaq, Digital Equipment Corp, IBM PC Co., Intel, Microsoft, NEC and Northern Telecom.
I'm pretty sure Apple adopted it only later; Firewire is considered faster, but a more expensive standard to implement.
I've been looking for LCD controllers as well. Can you post some links to companies to procure them from?
This may seem a strange idea, but what about "Email address portability"?
.name TLD,: William@Smith.name, billy63@William.Smith.name)
This could be offered by a group of ISPs who competed to offer you dialup or broadband service, but kept your email address constant by sharing administration of the mail-servers for a set of generic domains (For eg: John.Smith@email.com. Or, using the
Email forwarding services like iname.com do exist, but that option suffers the same lack of competition as the "ISP-provided-email" option. i.e. If you switch your email forwarder (say due to a hike in annual fees), you lose your email address.
The existence of companies like Propel.com implies that performance can still be squeezed from plain old 56K connections. So how can we get implement similar functionality for free?
From Propel's website (the link in your post above):
* Compression. Propel delivers text and graphics more efficiently, using a patent-pending compression technology that allows for intelligent encoding of Web pages and page elements sent to your browser.
It looks like Propel has implemented something similar to Akamaitech - a network of geographically dispersed network caches available to subscribers. This is impossible to replicate for free. (Right ?)
* Caching. Propel intelligently retains and re-uses Web pages and page elements that have previously been sent to your PC. That's why the longer Propel Accelerator is in use on your PC, the faster your Web pages will load.
Now how would this work? Maybe I could set my browser to cache *everything* (even disobeying a page's cache directive). And, in the background, the browser could figure out if the remote page had been updated (say, with HTTP HEAD) and if so, display a transition effect to the updated page.
* Persistent Connections. Propel uses proprietary techniques to carefully manage and optimize the communication between your modem and our servers using a persistent connection.
Hmm. Sounds like HTTP pipelining could help here. Also, how about implementing a DNS cache and a proxy server on the local machine (the one running the browser) -- would that help?
> > "I think every country needs a *balance* of free trade and protection of weaker industries"
> Sure, if you want to be poor. In the short term, allowing weak
> industries to fail may be harmful to the people who work in them,
> but in the long term, "protecting" them makes everyone poor.
You have a point, but so many rich country impose a (small) amount of protectionism and still prosper. For instance, even the US imposes anti-dumping duties when it thinks excessively cheap goods are being imported. IIRC, it also has farm subsidies, as well as massive amounts of investment for industries deemed nationally important, like defense. The same with the EC (farm subsidies, aerospace industry subsidies), Japan (protectionism for rice imports, electronics, and cars ), etc. I agree though, that the richest countries have the highest levels of free trade - free-trade does make an economy more efficient. But it works only when the country has a functioning welfare-net -- so that people booted out of a job are guarenteed basic food, housing and shelter. If people are kicked out of a job and starve to death, the government will be forced to change it policies pretty soon, or will be changed itself.
If I understand correct, Western countries and Japan set up functioning welfare networks in the early 50s (??), and maintain it to this day. Many developing countries are just setting up working welfare networks now.
> unless they're selfishly trying to keep their own job at other people's expense.
Democracy is the rule of the people and most people are selfish. They task their elected representatives to balance the benefits of free trade, with the pain associated with it.
Hmm - No. Not due to job outsourcing... but its certainly a result of technology that was born in the US.
I said "No" because the people behind this are (from the article) "founder members of CDAC, the brains behind India's PARAM-supercomputer". If I recall correctly, CDAC was setup by the Government of India in the late 80s as a direct consequence of the US *withholding* export of supercomputers to India for fear it would be used for defense research (more specifically, nuclear research). As a result, the CDAC people built massively parallel supercomputers from off-the-shelf CPUs (IIRC, they still used American CPUs - off-the-shelf 8086s (?) to begin with). They have some very cluey guys with a lot of experience born from research efforts like creating the complex electronics for interfacing supercomputers. Now it seems some of those people are moving to the private sector - kind of like with Govt. spending jumpstarting the computer revolution in the US.
A sign of what's to come? Is this the result of the US losing their position as main providers of R&D? What will be left afterwards? An economy of service?
I think every country needs a *balance* of free trade and protection of weaker industries. A "we can sell to you, but you can't sell to us" mentality is ultimately is bad for everyone concerned; from what I understand, 2/3rds of US income derives from exports.
At the end of the day, I'm sure your leaders have an eye on industry and employment figures. If not, you elect new ones.
Or, if range isn't an issue, how about a DIY bluetooth module for the wireless connectivity?
BTDesigner.com seems to be selling 20m-range modules for about $60.
Anyone know of cheaper sources? By now, Bluetooth modules were meant to cost around $5/module (in large quantities).
> Nor is science ruled by appeals to authority, which is exactly
:) I take it you won't reply.
> what you did when you pointed out that your creationist priest has a PhD.
The only reason I mentioned Dr. Batten's credentials is because you said concerning people like him:
"Do you really think those crackpots have any idea what they're talking about?"
> isn't a surprise because creationists often seem to have no sense of humor. You bore me.
OK
Bye then. I hope you will seriously consider the issues raised in those 3 papers again someday.
Paper #1) Danish and Middle East population could have diverged 4,500 years ago
----> Fits with the Biblical description of human dispersion occuring after the flood (around 4,500 years ago as well).
Paper #2) 20 times faster observed mtDNA Mutation Rate
----> Genetic bottlenecking can be approximately just 150,000/20 = 7,500 years old. Fits Biblical description of "bottlenecking" down to Noah's family 5,000 years ago
Paper #3) 1 male root lineage / 3 sub-lineages / only 1 of these 3 has 7 sub-sub-lineages that populate the world outside of Middle East and Africa.
----> Remarkable fit with Biblical story of Noah, his 3 sons, and the 7 descendants of only one of the 3 sons ("Japeth") populate the rest of world. The other 2 sons and their descendants populate the Middle East and Africa.
No no, not theoretical as in, they didn't exist: theoretical as in conceptually defined.
... it reflects a basic geometric necessity.
...
... ...
...
Look, we're splitting hairs here. OK, you think they're 'conceptually defined'. At the end of the day, can you just give me real reasons why a *starting population of two* does *not* fit this data? And please don't give me something like "but Y-Adam and m-Eve are millenia apart". You know well that, given experimental error in these techniques, the few millenia difference in the estimate of around 150,000 years (from the NY Times article) could well converge down to 0.0.
> The bottleneck itself
> there would still be a relatively _recent_ mEve,
Can you explain how a genetic bottleneck is a "geometric necessity"? After all, the NY Times article does state the m-Eve bottleneck occured when *all* the other women had only sons, while the Y-Adam bottleneck occured when *all* the other men only had all daughters, at some point in time. As the general population grows, I would think this would be a rarity, rather than a geometric necessity.
--If the amount of radioactivity increased dramatically due to the flood (as described in my earlier post), fossils created before the flood would look artificially "old" since post-flood fossils would have higher amounts of radio-carbon to begin with.---
An increase of such a level would have also irradiated life on Earth, including any Noah.
Yes that's right. It did. But perhaps you didn't get my point, which is:
The post-flood increase in radio carbon caused it to spike and eventually stablize at the levels we see today.
Its likely there was an initial spike in radiation levels right after the flood. This caused Noah's direct descendants to have higher mutation rates -- the distinct "1/3/rest-of-world-7" mutation pattern seen in the NYTimes article (that corresponds to the Biblical description of human ancestry.) Eventually though, the Radio-carbon levels would have subsided down to the (still *much* higher than pre-flood) levels we see today. IIRC, a paper I'd read (don't have a link handy) observed a similar effect among children born after the Chernobyl nuclear accident. The Bible does note a vastly decreased human lifespan after the flood.
Note, I'm not saying radio-carbon dating is hocus-pocus. I'm saying the assumption evolutionists often make, that we can date simply everything since we know all there is to know about ancient concentrations of radio-carbon levels and they've been really smooth and constant -- now that's hocus pocus.
---Critics thrust back saying techniques like Radio-carbon dating give older ages....---
> So, I see. If we pick and choose whatever evidence
> radio-carbon dating is harly the only technique used for these various issues,
> and indeed it is not even the most important for determining key ages of the earth.
You go on to point me to isochron dating. However, you must be aware of widespread problems with radio dating in general?
A well-known example of a problem with isochron dating (taken from this article) occured when a prestigious lab dated recent samples as very old:
The samples were sent progressively in batches to Geochron Laboratories in Cambridge, Boston (USA), for whole-rock potassium-argon (K-Ar) dating
Geochron is a respected commercial laboratory, the K-Ar lab manager having a Ph.D. in K-Ar dating. No specific location or expected age information was supplied to the laboratory. However, the samples were described as probably young with very little argon
The 'ages' range from which were observed to have cooled from lavas 25-50 years ago.
---Ultimately I can't *prove* God to you without a doubt.---
> Why would you need to prove God to me?
> The existence of God is an indepedant question.
OK...Sorry, I'll narrow down my phrasing... It should have read:
"I can't prove the existence of the God whose works are described in the Bible..."
The Bible tells us of a God (named simply "I AM"), creating the earth and all life on it. For example, in the old testament, God says "I have made the earth and created man upon it" (Isa 45:12). And in the new testament, Jesus says "But from the beginning of the creation, God made them male and female" (Mark 10:6, talking in the context of divorce.) The Bible also lists the lifespans and lineages of personalities from Adam to Christ, tracing a chain used to calculate the Biblical age of the earth.