The data I quoted was from ad impressions, so the simplest explanation is either that most Nokia units sold are not smartphones, or their users don't surf as much.
Seriously, not trying to troll, but wasn't this problem solved more completely by IPSEC back in 2000? Why hasn't IPSEC been adopted more? Why should this solution fare any better?
He's probably referring to the contact surface area. By making the surface corrugated, less of the surface comes in close contact to the other surface.
I had a machine once that would crash unexpectedly even though memtest passed. I eventually narrowed it down to a flaky IDE CDROM drive by systematically removing (or swapping out) components.
I think Linux is a bit less fault tolerant than some other OSes when it comes to flaky hardware. Or maybe it just validates hardware state more often.
I've seen some odd problems like this before. In my experience, these are due to bad RAM. Before installing Linux, user should really run memtest86 (all tests). It's even on the boot menu of the Ubuntu install disk.
Thanks, that was very informative. Also, here is the text from a Gutsy upgrade:
Vulnerable host keys will be regenerated
Some of the OpenSSH server host keys on this system were generated with a version of OpenSSL that had a broken random number generator. As a result, these host keys are from a well-known set, are subject to brute-force attacks, and must be regenerated.
Users of this system should be informed of this change, as they will be prompted about the host key change the next time they log in. Use 'ssh-keygen -l -f HOST_KEY_FILE' after the upgrade has changed to print the fingerprints of the new host keys.
The affected host keys are:/etc/ssh/ssh_host_rsa_key/etc/ssh/ssh_host_dsa_key
User keys may also be affected by this problem. The 'ssh-vulnkey' command may be used as a partial test for this. See/usr/share/doc/openssh-server/README.compromised-keys.gz for more details.
If it shouldn't be set, then it shouldn't be an option. This is a bug. It was (presumably) enabled by the OEM on my laptop (I know I didn't change it) and I'm clocking over 97000 for a 1-year-old laptop that sees only very light use on the weekends.
I've found Romex cable to be an inexpensive alternative when long runs are needed to high-powered speakers. Granted, it isn't very flexible, but it works great for in-wall speakers and terminal jacks.
Too smart in some cases... A coworker of mine was having trouble installing Ubuntu, but CentOS worked fine. Ubuntu would get through the install, but would power-up to a black screen with a single blinking underscore. Turns out, the PC he was using had two video cards and Ubuntu was enumerating them differently than CentOS. Had he hooked up a monitor to the other card, the problem would have been obvious. The solution was to remove one of the extra video cards.
I guess my point is, that there are some classes of problems that any amount of software can't fix. Fortunately, the problem you posed is probably not one of them.
I *just* ordered a set of 3.5 books after getting back into the game. I started losing interest in the game around the time they introduced the bard class (I kind of felt they jumped the shark) but I was drawn back by my kids wanting to play chutes & ladders too many times.
The kids love it. They're natural explorers and this is their favorite game now. It's a good educational tool too. It's teaching them about the consequences of their actions as well as math and bartering.
From the user's perspective, native apps are usually more responsive, but cross-platform development is expensive and those costs are passed on to the user (if a port is done at all). The benefits of web apps to the user are: (1) the availability of an application on their platform of choice (from desktops to smart phones) and in case where data is stored on a central server, (2) access to the data from any computer anywhere.
The benefits to the developer are that development costs are amortized over multiple platforms. That leads to greater marketshare and greater profit.
It isn't a closed system. Even the demo shows data-vis' documents-to-go and Opera's browser running on it. Palm's 70000 third-party devs are what's keeping Palm going. As someone once said, "it's the applications".
Although Palm has done this before, I doubt they will do it again. Around 2001, they offered 3.5 (I think) as a paid upgrade for Palm III users. They also had other upgrades for other models. I don't think it was very successful because a few years later, they stopped offering the upgrades.
I wouldn't call it "solved". You can't use multiple code sections/segments on most launch codes, you have to manually organize your code into separate sections (for gcc anyways), each section has to be in a separate file, database records, resources (graphics), and memory allocations are still limited 64k no matter what... The memory model is a mess! However, I have to commend Palm for maintaining backward compatibility across completely different hardware architectures but sometimes I wish they would have made a clean break.
And let's not forget the fact that most launch codes don't allow using global variables (!) or long-jumps.
Hopefully, all of that will change with Linux but I'm not holding my breath. The switch from 68K to ARM didn't improve the development environment all that much and the path of least resistance (at least initially) will be to simply get PACE running on Linux - which, as you know, is just a 68k emulator.
When you take into consideration the relationships between Palm and the cellular carriers, I doubt they will permit any Linux hacking. The carriers are the main reason you can't get a Treo with WiFi (not even an SDIO card because Palm controls the drivers) out of the fear that you will use it instead of one of the lucrative data plans.
I think you are mistaken about the Treo650 or your was an isolated case. It uses flash memory specifically to solve the problem with batteries running out. I rarely even sync mine anymore - it is that reliable.
Slashdot Mirror
Either the editors have already corrected it, or you have it backwards. "It's" is a contraction for "it is" and "its" is the possessive form.
It's a pain, but you can manually remove sites from search results with "-". For example, "yoursearchterms -www.blacklistedhost1.com".
Reminds me of this funny blog.
http://www.thesneeze.com/mt-archives/000398.php
The data I quoted was from ad impressions, so the simplest explanation is either that most Nokia units sold are not smartphones, or their users don't surf as much.
Here's some data on phone OS market share. Symbian leads with 41%, but iPhone is a close second (32%). Palm is way, way behind (4%).
http://tamspalm.tamoggemon.com/2009/01/09/admob-stat-day/
Seriously, not trying to troll, but wasn't this problem solved more completely by IPSEC back in 2000? Why hasn't IPSEC been adopted more? Why should this solution fare any better?
He's probably referring to the contact surface area. By making the surface corrugated, less of the surface comes in close contact to the other surface.
I had a machine once that would crash unexpectedly even though memtest passed. I eventually narrowed it down to a flaky IDE CDROM drive by systematically removing (or swapping out) components.
I think Linux is a bit less fault tolerant than some other OSes when it comes to flaky hardware. Or maybe it just validates hardware state more often.
I've seen some odd problems like this before. In my experience, these are due to bad RAM. Before installing Linux, user should really run memtest86 (all tests). It's even on the boot menu of the Ubuntu install disk.
Thanks, that was very informative. Also, here is the text from a Gutsy upgrade:
/etc/ssh/ssh_host_rsa_key /etc/ssh/ssh_host_dsa_key
/usr/share/doc/openssh-server/README.compromised-keys.gz for more details.
Vulnerable host keys will be regenerated
Some of the OpenSSH server host keys on this system were generated with a version of OpenSSL that had a broken random number generator. As a result, these host keys are from a well-known set, are subject to brute-force attacks, and must be regenerated.
Users of this system should be informed of this change, as they will be prompted about the host key change the next time they log in. Use 'ssh-keygen -l -f HOST_KEY_FILE' after the upgrade has changed to print the fingerprints of the new host keys.
The affected host keys are:
User keys may also be affected by this problem. The 'ssh-vulnkey' command may be used as a partial test for this. See
When will you learn... ;-)
I've given away software to beta testers too, but shouldn't a program released a year ago be out of beta by now? ;)
But seriously, I'm glad they are soliciting feedback for improving the user experience. Maybe the next update will have fewer distractions.
If it shouldn't be set, then it shouldn't be an option. This is a bug. It was (presumably) enabled by the OEM on my laptop (I know I didn't change it) and I'm clocking over 97000 for a 1-year-old laptop that sees only very light use on the weekends.
I've found Romex cable to be an inexpensive alternative when long runs are needed to high-powered speakers. Granted, it isn't very flexible, but it works great for in-wall speakers and terminal jacks.
Too smart in some cases... A coworker of mine was having trouble installing Ubuntu, but CentOS worked fine. Ubuntu would get through the install, but would power-up to a black screen with a single blinking underscore. Turns out, the PC he was using had two video cards and Ubuntu was enumerating them differently than CentOS. Had he hooked up a monitor to the other card, the problem would have been obvious. The solution was to remove one of the extra video cards.
I guess my point is, that there are some classes of problems that any amount of software can't fix. Fortunately, the problem you posed is probably not one of them.
I *just* ordered a set of 3.5 books after getting back into the game. I started losing interest in the game around the time they introduced the bard class (I kind of felt they jumped the shark) but I was drawn back by my kids wanting to play chutes & ladders too many times.
The kids love it. They're natural explorers and this is their favorite game now. It's a good educational tool too. It's teaching them about the consequences of their actions as well as math and bartering.
Google makes a version of Google Maps explicitly for the Treo as well as other smart phones.
http://google.com/gmm/treo
From the user's perspective, native apps are usually more responsive, but cross-platform development is expensive and those costs are passed on to the user (if a port is done at all). The benefits of web apps to the user are: (1) the availability of an application on their platform of choice (from desktops to smart phones) and in case where data is stored on a central server, (2) access to the data from any computer anywhere.
The benefits to the developer are that development costs are amortized over multiple platforms. That leads to greater marketshare and greater profit.
It isn't a closed system. Even the demo shows data-vis' documents-to-go and Opera's browser running on it. Palm's 70000 third-party devs are what's keeping Palm going. As someone once said, "it's the applications".
That's true, but with gcc it's a better practice to keep them in separate files. It's documented here: http://prc-tools.sourceforge.net/doc/prc-tools_3.h tml#SEC18
Although Palm has done this before, I doubt they will do it again. Around 2001, they offered 3.5 (I think) as a paid upgrade for Palm III users. They also had other upgrades for other models. I don't think it was very successful because a few years later, they stopped offering the upgrades.
FWIW, there is already an open-source project to get Linux on a tungsten: http://palmtelinux.sourceforge.net/
I wouldn't call it "solved". You can't use multiple code sections/segments on most launch codes, you have to manually organize your code into separate sections (for gcc anyways), each section has to be in a separate file, database records, resources (graphics), and memory allocations are still limited 64k no matter what... The memory model is a mess! However, I have to commend Palm for maintaining backward compatibility across completely different hardware architectures but sometimes I wish they would have made a clean break.
Not likely. PalmOS is now owned by Access, not Palm.
And let's not forget the fact that most launch codes don't allow using global variables (!) or long-jumps.
Hopefully, all of that will change with Linux but I'm not holding my breath. The switch from 68K to ARM didn't improve the development environment all that much and the path of least resistance (at least initially) will be to simply get PACE running on Linux - which, as you know, is just a 68k emulator.
When you take into consideration the relationships between Palm and the cellular carriers, I doubt they will permit any Linux hacking. The carriers are the main reason you can't get a Treo with WiFi (not even an SDIO card because Palm controls the drivers) out of the fear that you will use it instead of one of the lucrative data plans.
I think you are mistaken about the Treo650 or your was an isolated case. It uses flash memory specifically to solve the problem with batteries running out. I rarely even sync mine anymore - it is that reliable.