You're partly describing the Fermi Parodox and partly describing one of the commonly proposed explanations. There are several other well known explanations, none terribly encouraging. I do think you are correct that the SETI effort will fail, but I very much think it's worth the experiment.
http://www.faughnan.com/setifail.html has more, and links.
I am an extreme PDA user. Slashdotters who get by without a PDA (or a thick, heavy, insecure and unbacked-up paper planner) may either work all day at a desk (and thus have their machine calendar at hand), or have good (young) memories, or have relatively straightforward lives. The grayer I get, the more I need my PDA. If I forget it, I'm basically 30% demented.
So what killed PDA use for middle-aged folks like myself who are often away from a desk? (And yes, the PDA is dead.)
A collapse in reliability.
The first and second generation Palms were manufactured in the USA. They were ultra-reliable and very rugged. Data loss was rare. Then came the PocketPCs. Microsoft and Compaq pushed the envelope too far, producing very ambitious devices that could not be manufactured at a competitive price. So they cut corners to trim their losses. Anyone try the early iPaqs for long? If you drained the battery you lost EVERYTHING. And, with that screen and memory/CPU combo, it was easy to drain the battery.
The PocketPCs looked great though, and they sold buckets (though most of those early buyers will never buy again). Palm reacted by cutting quality. Power switches that stopped working after 6 months on the V/Vx and m5xx series. The IIIxe with its notorious total data loss problems. Clie dropped its warrantee to 3 months... HandSpring devices with bad memory...
Again and again, I heard from people who'd put a lot of data into their PDA, and then lost it all. Even those who had synched shortly before a crash find the experience intolerable. People will tolerate a desktop that crashes (though Win2K doesn't crash on me). They won't tolerate a PDA that wipes data.
In the end, the collapse in quality control meant that PDAs started to get a bad name.
The final blow was synching a corporate Palm with an Exchange server. Microsoft, strangely enough, doesn't make it easy. The Palm/Exchange data models are different enough that only a lightweight or an expert Outlook user will be able to work around data model distinctions in the long run.
I just hope that someone keeps manufacturing something like my Vx (now running with software hacks to get around the defunct on/off switch). If not, I might as well check myelf directly into the nursing home.
In 1998 I was one of thousands of victims in an international hundred million dollar credit card fraud. Some of the suspected principals of that case are said to be back in operation.
I had a few minutes of limited fame back then, including an appearance on Japanese tv. The story of that fraud, and a dicussion of cc fraud in general, is here. (Alas, the site is hosted by myhosting.com, and as on many Sunday mornings it is now down!)
Only the banks can fix the problem, but with the very notable exception of American Express they've done very little. I now use AMEX for all recurring internet transactions, and if they ever got their Quicken support working reliably (they've failed for 3 years) I'd use them for all online transactions. AMEX has the best attention to security, and the best response to fraud, and the most sustained interest in combating fraud.
Barring litigation, the VISA/MC franchise will only fix this problem if customers stop using their cards. So use AMEX instead.
john faughnan jfaughnan@spamcop.net www.faughnan.com
There were only two changes my wife and I would have made. We were ok with the relatively angst ridden Aragorn and the augmented Arwen. All in all it was a superb result for the most difficult movie of the trilogy. (BTW, note the sword was never reforged, expect Arwen to bring it out to Aragorn in #2.)
The two changes we'd have voted for:
1. Less music. It was too intrusive in places where silence or birdsong would have worked better. The visuals were so powerful the music was often annoying.
2. They set everything up for Gimli to fall in knightly love with Galadriel, then they didn't show it! (In other words, we agree about more Lothlorien.) They could have shaved 3-4 minutes from 'Frodo pinioned by the cave monster' and Sam drowning, and used the time to restore the Gimli/Galadriel scene. I suspect they filmed it but made a misguided cut. Hope it's on the DVD.
Overall, a great result.
john
Gas at $10.00 a gallon, rental biz - it's good
on
This is IT?
·
· Score: 1
I agree with your comments, and I'd like to add that gas won't always be $1.00 a gallon. It's certain that SOMEDAY we'll use up a lot of our oil reserves, and it's somewhat likely that in the near term Saudi Arabia will be unstable.
It would be easy to combine the scooter with public transportation to replace a lot of automobile use in warmer cities. It even sounds like it does well on ice, so perhaps it might work in November in Minnesota (don't know about January though). Scoot to your light rail train, hang your palmprint locked scooter on the side, get off and go to the office. Check your scooter at planeside, use it at the airport...
I could see renting the scooters around cities, as taxi alternatives. Car rental companies might want to look into this business. Or rent a scooter with your car rental for a small fee. That might be one way that a lot of the $3000 devices will see popular use -- as rentals.
The price will fall like a rock (what exactly is the cost here? silicon? ummmm look for $400 devices within a few years). There will be a place to attach a standard bag for carrying things. Eventually there will be trailers. Go to Europe, and tour your favorite town via scooter.
Best of all, I can scoot around when I'm 85 yo, even more demented than now, and can barely shuffle from spot to spot. Sounds good to me.
I won't buy one at $3000, but I'll definitely have on at $600! And before that I'll have rented one to get around on my business trips.
If these things really work as advertised, they will indeed succeed.
The CERT alert is explicit that the worm only infects NT/2000 machines running IIS. Why is the Windows 98 Personal Web Server (a limited version of IIS 4) not vulnerable?
Since in scored a big fat [1] for this piece, I suppose I might as well reply to it. It turns out, not surprisingly, that someone else has thought of the implications of the Singularity for SETI.
See Kurzweil's article: and search on SETI.
I may have thought of it earlier (a year ago or so), but I didn't think I was the only one who thought of it.
see www.faughnan.com/setifail.html --
John Faughnan
1. Fermi's calculations assume a civilization with light speed technology that expands from planet to planet every few hundred to thousand years. No highly advanced technology is required for such a civilization to colonize the galaxy within tens of thousands of years -- just exponential growth. See for a Rumanian example.
2. The point of my argument is that it's not likely that post-Singular civilizations are driven by the same things that drive biological organisms (growth, expansion, etc). For one thing their time scales are different from biological organisms; it's not hard to imagine that they exist in a time-space that's thousands of times faster than ours.
Here's my argument in summary:
Fermi's paradox implies that no technologic civilization ever survives.
Singularity is a good example of a disaster or transformation that may strike all technologic civlizations of all forms, probably before then are able to travel between stars.
Since there not likely to be any star-faring civilizations, and since Singularity is a reasonable transformation event common to all technologic civilizations, it is probably that it is Singularity that ends a civilization's interest in expansion and exploration.
One of the best pieces of indirect evidence for the inevitability of the Singularity is the Fermi paradox, and, to a lesser extent, SETI's lack of success (to date).
Fermi showed that, given reasonable assumptions, we ought to expect "ET" to be ubiquitous. Since extraterrestrials are not all about us, this suggests either technologic civilizations are exquisitely rare or that they rapidly lose behaviors like migration and radio communication. By rapidly I mean within two to three hundred years.
The Singularity is the kind of event that would do that. If technologic civilizations always progress to a Singularity they may well lose interest in minor details like reproduction and out migration. Among other things they would operate on very different time scales from pre-Singular civilizations.
Wardk wrote "If IBM wants to make ammends for the OS/2 debacle, they would port the Workplace Shell to Linux/BSD/Unix. The Workplace Shell was and still is the finest GUI design on the planet..."
Agreed -- at least about the pleasure of using the Workplace Shell -- when it worked! Of all the shells I've used, it was my favorite. The implementation was flawed (desynchronization between file structure and shell representation, no ability to uninstall applications, vulnerability to misguided applications, etc) but the ideas were sound.
My memory is unreliable on this, but I believe the shell sprang in part from IBM research labs and that it incorporated some forward thinking about user interfaces and information management. The introduction of the shell in OS/2 Warp was somewhat miraculous, but in retrospect it was probably a bridge too far -- it languished too long with its problems unresolved.
While I agree that a single productivity suite would not suffice to push the Linux desktop, my recollection was that SmartSuite on OS/2 was "too little, too late".
IBM had committed to put SmartSuite on OS/2, but after WordPerfect abandoned the platform they seemed to lose interest. They eventually put out a partial implementation, but it was quite incomplete and fairly awful. (Fonts didn't work at all, a chronic OS/2 application problem). They did make SmartSuite/Windows work well in the Windows emulation layer and IBM came to recommend that approach. (I complained about the crummy OS/2 version and they sent me the Windows version to use -- which I rather liked.)
One of the many traps OS/2 ran into was their Windows emulation layer. It was very strong, so that vendors had little motivation to write native software. Alas, it was all too easy for Microsoft to change Windows, and too hard for IBM to keep the emulation working -- especially after Win95 came out. Not the only wound OS/2 suffered of course, but a contributor.
A lesson perhaps for Linux and Windows emulation -- it's a treacherous path to take. If the clientside platform ever became a threat, Microsoft could again change Windows so the emulation layer would break.
john
www.faughnan.com --
John Faughnan
Batteries, Outlook/Exchange sync, Reliability.
on
Palm In Trouble?
·
· Score: 1
Posting this late in a slashdot discussion is an exercise in futility (no ratings, no readership). Still, some odd instinct urges me to toss out some ideas I think both relevant and not yet presented.
1. Batteries: In our corporate testing not only did the iPaq batteries die relatively quickly, but when they did the device lost all data. Restoration was somewhat painful. Users who do well with iPaq's seem to work at a desk and recharge very frequently. I expect Microsoft/Compaq will eventually fix this.
2. The main impact on battery drainage appears to be the display, not the CPU or OS. (The Palm OS, however, is very good at saving power and managing low power states).
3. The Palm conduits and sync mechanisms are robust. My experiences with iPaq syncing have been mixed. HUGE EXCEPTION: Outlook's data model is SO complex and powerful I've a long history of frustrating Palm syncs with Outlook. I assume the iPaq HAS to be better. If you want to look for why Palm is in trouble, the dominance of Outlook/Exchange in the corporate setting and the difficulty of syncing with Outlook/Exchange may be issue #1.
4. I am having more software and hardware reliability issues over time. The most reliable devices I had was a Palm III with OS 2.x. As the OS gets more complex (communications layers) and as price pressure grows reliability is being sacrificed. This is HUGE problem #2. My wife made a very strong effort to switch from a Franklin Planner to a Palm IIIx; until it started crashing with total data loss (hardware problem). She now uses the replacement IIIx for games and phone numbers. Declining reliability is the kiss of death for the original purpose of the Palm (calendaring, PIM).
5. The Palm OS is said to handle network traffic poorly. If true this is another major limitation, though in practice I have found little value and much cost to the current wireless web.
6. Security is very weak on the Palm, I think the iPaq does better. This is a very big issue for me, but not for most users. (I have had major usability and reliability issues with both TealLock and OnlyMe.
CONCLUSION: Things do not look great for any product in the marketplace. The future was brighter about 3 years ago, but fundamentally I don't think there's a big enough user base for a truly reliable product of the nature of the original Palm III. I think we'll see a profusion of devices with multimedia functionality, lower prices, more features, and less reliability and shorter battery life. We made it further down the road this time than we did with the Newton, but we may have to wait for the third wave.
PS. Yes, I do wish the Psion had done better. Unfortunately I am addicted to the size and weight of the Vx.
--
John Faughnan
The interesting part of this product announcement is the scheduled downloads. The examples given (retrieve email, etc) are of interest with a very slow dial-up or wireless connection, but there are other broadband applications.
SONY wants to deliver video and entertainment. Even with DSL/cable connections that will require background downloading (if only because we'll saturate the net infrastructure with heavy use). This device, with the BeOS and scheduled downloads is well suited (ok, ignore the display for now).
Go to a web page and request the video/music you want. Get billed. Servers queue up the material, perhaps based on how many customers vote for it. (Recent videos will get more votes). Deliver off hours, view in real time.
So, why not a letter-box display?! Cough, cough. Got me there. Look for another model within the the year with a letter box display...
The article is well written, but the discussion would be very familiar to readers of the Economist. Dr. Gordon's opinions are well researched, but I've also read convincing essays claiming total factor productivity is increasing quite broadly.
Jonathon Rauch's article in this week's Atlantic is a case study of the application of IT to the old economy. It's a very detailed exploration of arguments that partially refute Gordon's stance that productivity growth is very sector limited. (Even Gordon has shifted, I believe he used to claim few manufacturing improvements.)
The author also states that there has been no technological transformation to equal the mail to telegraph transformation. I think the web/internet, in its original (CERN) sense of a way to publish worldwide for pennies, has a potentially equal potential for transformation. A vast amount of human knowledge (yes, and noise, but humans are very good at sorting out noise) can now be published worldwide for an infinitesmal fraction of the pre-1980 cost. Twenty years from now that impact may seem much larger than it does today.
It all depends on where one measures from. I do think that the period from 1965 to 1990 may be considered a time of relative "calm", a period where much was happening below the surface. I suspect, however, that the period from 1990 to 2040 may rival the shocking transformations of the early 20th century. And that prediction, of course, assumes that we don't develop sentient machines (An event that would render all comparisons to past eras irrelevant.)
One last thought (sorry, long post), last year, in their famous millenium issue, the Economist published a graph showing economic output per person over the past 10,000 or so years. I wish I could find a way to link to it (the original may have come from the World Bank). It's a very slightly sloping line until the 18th century, where it starts to head up. Then it heads up very sharply around 1890 and goes exponential. There's another inflection to a higher exponent around 1980/90. Studying that curve is mind boggling.
I've used a variety of techniques, all of which are detailed here. I'll have to add the "real address that seems fake and is rejected by spambots" technique!
I've ended up relying on 3 different solutions, each with different strengths.
Trusted correspondents get my direct MindSpring (now, sigh, Earthlink owned) address. MindSpring has decent ISP spam filtering.
I pay a pittance every year for my spamcop redirector, and I use my spamcop address for non-trusted correspondents. SpamCop does very aggressive ISP level filtering; rejected correspondents get a SpamCop reply with an embedded URL. Correspondents can click the URL to send the message and bypass the filters. Unfortunately this step is too much for many users, and vendors have trouble with it to (security fears probably). If SpamCop had more subtle filtering options I would use only that address (and willingly pay more). SpamCop does use a form of positive filtering -- if you manually accept email from the SpamCop held box the sender is added to an "accepted" list.
On the client side I do a mixture of positive and negative filtering. Obvious spam goes directly to a spam box, addresses that do not match my accepted list go to a lesser garbage box.
Ultimately the solution is likely either signed email or adding finer filtering controls to SpamCop and integrate SpamCop type services with ISPs. At the moment though, netizens have not been willing to pay a premium for ISP-level spam filtering solutions and spam filtering has not figured in ISP rankings (to my knowledge).
Why do people choose to drink even when it costs them friends, family, employment, health, and life?
Our best answer in 2000 is (of course) both nature (genetics) and environment. Some persons brains are wired so that the sensation of intoxication is irresistibly pleasurable. They are cursed with an unfortunate gene that, in the modern world, is very maladaptive (in the ancient world perhaps this was not so). Not all these people will drink, but for many not drinking is a lifelong struggle.
Others share this prediliction, but most of the time they manage. Personal loss or social pressure can tip the balance.
I would suspect Mr. Katz shared a full dose of a bad gene (or set of genes). It is a true curse.
Of course it is not that simple. 1950s middle-class euro-americans drank a staggering amount of alcohol by our (relatively) low-consumption standards, yet neither genetics nor 'happiness' appears to have changed that greatly in the past 50 years.
Last note, contrary to the quotes in the news piece, the original PKZip was not free. It was shareware. I believe Mr. Katz's unique wisdom was to make the file format public, in contrast to the proprietary compression format it replaced (who's name is now lost to my aging memory). I have often felt we spend too much attention to code and not enough to file formats. Requiring Microsoft to use a published XML format for Word and Excel might have a greater effect on competition than requiring them to open source either application.
I hope this guide becomes an official museum piece. It's very well written with a lovely layout. The HTML is also quite clean.
Let us not be too proud of our progress, however. Read the document, and think how LITTLE has changed in six years. There is more of everything, but that page looks as modern as any other. Note the discussion of neat MOSAIC features such as MPEG and streaming QuickTime video...
We have been standing still by comparison with the period from 1989 to 1993 (Gopher, HyperG, the web, etc.) Consider all the stillborn and unborn: RDF, VRML, VR conferencing, true hypertext, NNTP/HTML collaboration frameworks, Xanadu, alternative navigation frameworks (RDF, Apple's Project X). CSS-2 is barely implemented anywhere, and even CSS-1 is not implemented. Client-side Java has struggled.
Those were the days of the giants, and now we are still digesting those changes -- and fighting endless patent battles.
Ahh, when I was a younger man... now those were exciting days. Let me tell you....:-)
>I remember fondly the days wasted with my 286 >and a 1200 bps modem, racking up some impressive >phone bills. Not bad, but I'll trump that one. I used my 8086 and a pre-Internet packet switching network (tymnet or telenet) to visit BBSs around the nation. For a modest monthly fee one could access these corporate networks after hours, and avoid long distance charges. There were tens of thousands of BBSs nationwide. The setup process was rather baroque, but it worked. The net was around by then, but not for use by non-academics.
It is not possible to separate Microsoft's application division from their OS division. Each gained advantage from the other.
There is some interest in publishing the Windows APIs, but I think there would be more interest in requiring Microsoft to publish all current and past file formats, and to require all new file formats be published and documented prior to application release.
If this were paired with structural changes and API publication, Microsoft's strategic advantages would be diminished, and competition enhanced.
Is publication of file formats past and future a possible remedy?
Slashdot Mirror
You're partly describing the Fermi Parodox and partly describing one of the commonly proposed explanations. There are several other well known explanations, none terribly encouraging. I do think you are correct that the SETI effort will fail, but I very much think it's worth the experiment.
http://www.faughnan.com/setifail.html has more, and links.
I am an extreme PDA user. Slashdotters who get by without a PDA (or a thick, heavy, insecure and unbacked-up paper planner) may either work all day at a desk (and thus have their machine calendar at hand), or have good (young) memories, or have relatively straightforward lives. The grayer I get, the more I need my PDA. If I forget it, I'm basically 30% demented.
... HandSpring devices with bad memory ...
So what killed PDA use for middle-aged folks like myself who are often away from a desk? (And yes, the PDA is dead.)
A collapse in reliability.
The first and second generation Palms were manufactured in the USA. They were ultra-reliable and very rugged. Data loss was rare. Then came the PocketPCs. Microsoft and Compaq pushed the envelope too far, producing very ambitious devices that could not be manufactured at a competitive price. So they cut corners to trim their losses. Anyone try the early iPaqs for long? If you drained the battery you lost EVERYTHING. And, with that screen and memory/CPU combo, it was easy to drain the battery.
The PocketPCs looked great though, and they sold buckets (though most of those early buyers will never buy again). Palm reacted by cutting quality. Power switches that stopped working after 6 months on the V/Vx and m5xx series. The IIIxe with its notorious total data loss problems. Clie dropped its warrantee to 3 months
Again and again, I heard from people who'd put a lot of data into their PDA, and then lost it all. Even those who had synched shortly before a crash find the experience intolerable. People will tolerate a desktop that crashes (though Win2K doesn't crash on me). They won't tolerate a PDA that wipes data.
In the end, the collapse in quality control meant that PDAs started to get a bad name.
The final blow was synching a corporate Palm with an Exchange server. Microsoft, strangely enough, doesn't make it easy. The Palm/Exchange data models are different enough that only a lightweight or an expert Outlook user will be able to work around data model distinctions in the long run.
I just hope that someone keeps manufacturing something like my Vx (now running with software hacks to get around the defunct on/off switch). If not, I might as well check myelf directly into the nursing home.
john
[meta: 021124, jfaughnan, jgfaughnan, Palm, Palm OS, PDA, PocketPC]
In 1998 I was one of thousands of victims in an international hundred million dollar credit card fraud. Some of the suspected principals of that case are said to be back in operation.
I had a few minutes of limited fame back then, including an appearance on Japanese tv. The story of that fraud, and a dicussion of cc fraud in general, is here. (Alas, the site is hosted by myhosting.com, and as on many Sunday mornings it is now down!)
Only the banks can fix the problem, but with the very notable exception of American Express they've done very little. I now use AMEX for all recurring internet transactions, and if they ever got their Quicken support working reliably (they've failed for 3 years) I'd use them for all online transactions. AMEX has the best attention to security, and the best response to fraud, and the most sustained interest in combating fraud.
Barring litigation, the VISA/MC franchise will only fix this problem if customers stop using their cards. So use AMEX instead.
john faughnan
jfaughnan@spamcop.net
www.faughnan.com
There were only two changes my wife and I would have made. We were ok with the relatively angst ridden Aragorn and the augmented Arwen. All in all it was a superb result for the most difficult movie of the trilogy. (BTW, note the sword was never reforged, expect Arwen to bring it out to Aragorn in #2.)
The two changes we'd have voted for:
1. Less music. It was too intrusive in places where silence or birdsong would have worked better. The visuals were so powerful the music was often annoying.
2. They set everything up for Gimli to fall in knightly love with Galadriel, then they didn't show it! (In other words, we agree about more Lothlorien.) They could have shaved 3-4 minutes from 'Frodo pinioned by the cave monster' and Sam drowning, and used the time to restore the Gimli/Galadriel scene. I suspect they filmed it but made a misguided cut. Hope it's on the DVD.
Overall, a great result.
john
I agree with your comments, and I'd like to add that gas won't always be $1.00 a gallon. It's certain that SOMEDAY we'll use up a lot of our oil reserves, and it's somewhat likely that in the near term Saudi Arabia will be unstable.
...
It would be easy to combine the scooter with public transportation to replace a lot of automobile use in warmer cities. It even sounds like it does well on ice, so perhaps it might work in November in Minnesota (don't know about January though). Scoot to your light rail train, hang your palmprint locked scooter on the side, get off and go to the office. Check your scooter at planeside, use it at the airport
I could see renting the scooters around cities, as taxi alternatives. Car rental companies might want to look into this business. Or rent a scooter with your car rental for a small fee. That might be one way that a lot of the $3000 devices will see popular use -- as rentals.
The price will fall like a rock (what exactly is the cost here? silicon? ummmm look for $400 devices within a few years). There will be a place to attach a standard bag for carrying things. Eventually there will be trailers. Go to Europe, and tour your favorite town via scooter.
Best of all, I can scoot around when I'm 85 yo, even more demented than now, and can barely shuffle from spot to spot. Sounds good to me.
I won't buy one at $3000, but I'll definitely have on at $600! And before that I'll have rented one to get around on my business trips.
If these things really work as advertised, they will indeed succeed.
john
The CERT alert is explicit that the worm only infects NT/2000 machines running IIS. Why is the Windows 98 Personal Web Server (a limited version of IIS 4) not vulnerable?
See Kurzweil's article: and search on SETI.
I may have thought of it earlier (a year ago or so), but I didn't think I was the only one who thought of it.
see www.faughnan.com/setifail.html
--
John Faughnan
A few comments:
1. Fermi's calculations assume a civilization with light speed technology that expands from planet to planet every few hundred to thousand years. No highly advanced technology is required for such a civilization to colonize the galaxy within tens of thousands of years -- just exponential growth. See for a Rumanian example.
2. The point of my argument is that it's not likely that post-Singular civilizations are driven by the same things that drive biological organisms (growth, expansion, etc). For one thing their time scales are different from biological organisms; it's not hard to imagine that they exist in a time-space that's thousands of times faster than ours.
Here's my argument in summary:
- Fermi's paradox implies that no technologic civilization ever survives.
- Singularity is a good example of a disaster or transformation that may strike all technologic civlizations of all forms, probably before then are able to travel between stars.
- Since there not likely to be any star-faring civilizations, and since Singularity is a reasonable transformation event common to all technologic civilizations, it is probably that it is Singularity that ends a civilization's interest in expansion and exploration.
www.faughnan.com/setifail.html--
John Faughnan
Fermi showed that, given reasonable assumptions, we ought to expect "ET" to be ubiquitous. Since extraterrestrials are not all about us, this suggests either technologic civilizations are exquisitely rare or that they rapidly lose behaviors like migration and radio communication. By rapidly I mean within two to three hundred years.
The Singularity is the kind of event that would do that. If technologic civilizations always progress to a Singularity they may well lose interest in minor details like reproduction and out migration. Among other things they would operate on very different time scales from pre-Singular civilizations.
See also http://www.faughnan.com/setifail.html.
john
--
John Faughnan
Agreed -- at least about the pleasure of using the Workplace Shell -- when it worked! Of all the shells I've used, it was my favorite. The implementation was flawed (desynchronization between file structure and shell representation, no ability to uninstall applications, vulnerability to misguided applications, etc) but the ideas were sound.
My memory is unreliable on this, but I believe the shell sprang in part from IBM research labs and that it incorporated some forward thinking about user interfaces and information management. The introduction of the shell in OS/2 Warp was somewhat miraculous, but in retrospect it was probably a bridge too far -- it languished too long with its problems unresolved.
--
John Faughnan
IBM had committed to put SmartSuite on OS/2, but after WordPerfect abandoned the platform they seemed to lose interest. They eventually put out a partial implementation, but it was quite incomplete and fairly awful. (Fonts didn't work at all, a chronic OS/2 application problem). They did make SmartSuite/Windows work well in the Windows emulation layer and IBM came to recommend that approach. (I complained about the crummy OS/2 version and they sent me the Windows version to use -- which I rather liked.)
One of the many traps OS/2 ran into was their Windows emulation layer. It was very strong, so that vendors had little motivation to write native software. Alas, it was all too easy for Microsoft to change Windows, and too hard for IBM to keep the emulation working -- especially after Win95 came out. Not the only wound OS/2 suffered of course, but a contributor.
A lesson perhaps for Linux and Windows emulation -- it's a treacherous path to take. If the clientside platform ever became a threat, Microsoft could again change Windows so the emulation layer would break.
john
www.faughnan.com
--
John Faughnan
1. Batteries: In our corporate testing not only did the iPaq batteries die relatively quickly, but when they did the device lost all data. Restoration was somewhat painful. Users who do well with iPaq's seem to work at a desk and recharge very frequently. I expect Microsoft/Compaq will eventually fix this.
2. The main impact on battery drainage appears to be the display, not the CPU or OS. (The Palm OS, however, is very good at saving power and managing low power states).
3. The Palm conduits and sync mechanisms are robust. My experiences with iPaq syncing have been mixed. HUGE EXCEPTION: Outlook's data model is SO complex and powerful I've a long history of frustrating Palm syncs with Outlook. I assume the iPaq HAS to be better. If you want to look for why Palm is in trouble, the dominance of Outlook/Exchange in the corporate setting and the difficulty of syncing with Outlook/Exchange may be issue #1.
4. I am having more software and hardware reliability issues over time. The most reliable devices I had was a Palm III with OS 2.x. As the OS gets more complex (communications layers) and as price pressure grows reliability is being sacrificed. This is HUGE problem #2. My wife made a very strong effort to switch from a Franklin Planner to a Palm IIIx; until it started crashing with total data loss (hardware problem). She now uses the replacement IIIx for games and phone numbers. Declining reliability is the kiss of death for the original purpose of the Palm (calendaring, PIM).
5. The Palm OS is said to handle network traffic poorly. If true this is another major limitation, though in practice I have found little value and much cost to the current wireless web.
6. Security is very weak on the Palm, I think the iPaq does better. This is a very big issue for me, but not for most users. (I have had major usability and reliability issues with both TealLock and OnlyMe.
CONCLUSION: Things do not look great for any product in the marketplace. The future was brighter about 3 years ago, but fundamentally I don't think there's a big enough user base for a truly reliable product of the nature of the original Palm III. I think we'll see a profusion of devices with multimedia functionality, lower prices, more features, and less reliability and shorter battery life. We made it further down the road this time than we did with the Newton, but we may have to wait for the third wave.
PS. Yes, I do wish the Psion had done better. Unfortunately I am addicted to the size and weight of the Vx.
--
John Faughnan
SONY wants to deliver video and entertainment. Even with DSL/cable connections that will require background downloading (if only because we'll saturate the net infrastructure with heavy use). This device, with the BeOS and scheduled downloads is well suited (ok, ignore the display for now).
Go to a web page and request the video/music you want. Get billed. Servers queue up the material, perhaps based on how many customers vote for it. (Recent videos will get more votes). Deliver off hours, view in real time.
So, why not a letter-box display?! Cough, cough. Got me there. Look for another model within the the year with a letter box display ...
jfaughnan
--
John Faughnan
Jonathon Rauch's article in this week's Atlantic is a case study of the application of IT to the old economy. It's a very detailed exploration of arguments that partially refute Gordon's stance that productivity growth is very sector limited. (Even Gordon has shifted, I believe he used to claim few manufacturing improvements.)
The author also states that there has been no technological transformation to equal the mail to telegraph transformation. I think the web/internet, in its original (CERN) sense of a way to publish worldwide for pennies, has a potentially equal potential for transformation. A vast amount of human knowledge (yes, and noise, but humans are very good at sorting out noise) can now be published worldwide for an infinitesmal fraction of the pre-1980 cost. Twenty years from now that impact may seem much larger than it does today.
It all depends on where one measures from. I do think that the period from 1965 to 1990 may be considered a time of relative "calm", a period where much was happening below the surface. I suspect, however, that the period from 1990 to 2040 may rival the shocking transformations of the early 20th century. And that prediction, of course, assumes that we don't develop sentient machines (An event that would render all comparisons to past eras irrelevant.)
One last thought (sorry, long post), last year, in their famous millenium issue, the Economist published a graph showing economic output per person over the past 10,000 or so years. I wish I could find a way to link to it (the original may have come from the World Bank). It's a very slightly sloping line until the 18th century, where it starts to head up. Then it heads up very sharply around 1890 and goes exponential. There's another inflection to a higher exponent around 1980/90. Studying that curve is mind boggling.
john
--
John Faughnan
I've ended up relying on 3 different solutions, each with different strengths.
- Trusted correspondents get my direct MindSpring (now, sigh, Earthlink owned) address. MindSpring has decent ISP spam filtering.
- I pay a pittance every year for my spamcop redirector, and I use my spamcop address for non-trusted correspondents. SpamCop does very aggressive ISP level filtering; rejected correspondents get a SpamCop reply with an embedded URL. Correspondents can click the URL to send the message and bypass the filters. Unfortunately this step is too much for many users, and vendors have trouble with it to (security fears probably). If SpamCop had more subtle filtering options I would use only that address (and willingly pay more). SpamCop does use a form of positive filtering -- if you manually accept email from the SpamCop held box the sender is added to an "accepted" list.
- On the client side I do a mixture of positive and negative filtering. Obvious spam goes directly to a spam box, addresses that do not match my accepted list go to a lesser garbage box.
Ultimately the solution is likely either signed email or adding finer filtering controls to SpamCop and integrate SpamCop type services with ISPs. At the moment though, netizens have not been willing to pay a premium for ISP-level spam filtering solutions and spam filtering has not figured in ISP rankings (to my knowledge).John Faughnan
Our best answer in 2000 is (of course) both nature (genetics) and environment. Some persons brains are wired so that the sensation of intoxication is irresistibly pleasurable. They are cursed with an unfortunate gene that, in the modern world, is very maladaptive (in the ancient world perhaps this was not so). Not all these people will drink, but for many not drinking is a lifelong struggle.
Others share this prediliction, but most of the time they manage. Personal loss or social pressure can tip the balance.
I would suspect Mr. Katz shared a full dose of a bad gene (or set of genes). It is a true curse.
Of course it is not that simple. 1950s middle-class euro-americans drank a staggering amount of alcohol by our (relatively) low-consumption standards, yet neither genetics nor 'happiness' appears to have changed that greatly in the past 50 years.
Last note, contrary to the quotes in the news piece, the original PKZip was not free. It was shareware. I believe Mr. Katz's unique wisdom was to make the file format public, in contrast to the proprietary compression format it replaced (who's name is now lost to my aging memory). I have often felt we spend too much attention to code and not enough to file formats. Requiring Microsoft to use a published XML format for Word and Excel might have a greater effect on competition than requiring them to open source either application.
Let us not be too proud of our progress, however. Read the document, and think how LITTLE has changed in six years. There is more of everything, but that page looks as modern as any other. Note the discussion of neat MOSAIC features such as MPEG and streaming QuickTime video ...
We have been standing still by comparison with the period from 1989 to 1993 (Gopher, HyperG, the web, etc.) Consider all the stillborn and unborn: RDF, VRML, VR conferencing, true hypertext, NNTP/HTML collaboration frameworks, Xanadu, alternative navigation frameworks (RDF, Apple's Project X). CSS-2 is barely implemented anywhere, and even CSS-1 is not implemented. Client-side Java has struggled.
Those were the days of the giants, and now we are still digesting those changes -- and fighting endless patent battles.
Ahh, when I was a younger man ... now those were exciting days. Let me tell you .... :-)
john
Pretty impressive specs (HTML 3.2, JavaScript, 128 bit SSL). Have you any experience with the browser? Any sense of how solid it is?
I wonder how it compares to Opera 4.0 ... (admittedly not optimized for a small display)
john
>I remember fondly the days wasted with my 286 >and a 1200 bps modem, racking up some impressive >phone bills. Not bad, but I'll trump that one. I used my 8086 and a pre-Internet packet switching network (tymnet or telenet) to visit BBSs around the nation. For a modest monthly fee one could access these corporate networks after hours, and avoid long distance charges. There were tens of thousands of BBSs nationwide. The setup process was rather baroque, but it worked. The net was around by then, but not for use by non-academics.
It is not possible to separate Microsoft's application division from their OS division. Each gained advantage from the other.
There is some interest in publishing the Windows APIs, but I think there would be more interest in requiring Microsoft to publish all current and past file formats, and to require all new file formats be published and documented prior to application release.
If this were paired with structural changes and API publication, Microsoft's strategic advantages would be diminished, and competition enhanced.
Is publication of file formats past and future a possible remedy?