*whack* You are correct that it doesn't include upgrades. However, the Recommended Cluster Patches, which include include security and reliability patches, are free. Check http://sunsolve.sun.com and look for 'patches'.
*OUCH!* Thanks...I think!
Sun should make it easy to find these kind of things. Only a few sites are harder to deal with, and almost all of them high-profile.
[GRIPE] Why hide every patch and upgrade under a layer of menus that use phrases from a marketing handbook, as opposed to...well...an FTP site? What's so hard about that old favorite, plus a simple directory structure?
No matter how forgiving and PC your attitude is there are most definately dumb questions . Here are a couple more;
Background: One of the benifits of a jfs is being able to 'roll-back' changes or to select a specific revision without rolling back the current version.
Q. How well does ResierFS handle this (if at all)?
Q. What file systems are available that can do this, and what tools are available to get back intermediate revisions of a specific file or directory tree.
With cheap disk space, this looks like it would be a great tool to have, while faster boot time is less valuable unless you are running a time critical application and any delay is a bad thing.
Try being a TS lead. I was for a few years, and that folks is the limit for anyone not on medication. A normal front end support tech with one+ years experience is not a well individual and probably not good at actually helping anyone except in the morning and just after lunch.
The typical tech goes from excited and helpful -- and clueless -- to knowledgeable and couldn't give a damn. Answer 50 to over 200 calls a day for six months, and I doubt that you'll get any different reaction.
My role was to occasionaly man the phones to keep sharp, but mostly to answer hard questions that the front-end support couldn't figure out. The front end support people frequently didn't look in the database for answers.
Then, there's policy. For software -- and an ISP's support department handles software problems most of the time -- the cost of handling this type of question is just as high per minute as other calls. Unfortunately, it takes more time. You, as a customer, are given only a few seconds or minutes for support. If the tech does not handle enough calls per day, they are chewed out for basically doing good customer support. Why? One call answered well eats into the allocated time for other calls...and it is tightly monitored; from hang-ups, calls per product, type of questions asked,...all with an eye on how to reduce the time the tech -- not you -- spend on the question or problem. Yes, this sucks.
The prefered way to handle this would be to have links on a company web site to the information you want, plus an automated fax system. That way the time spent on the phone by the tech is reduced to just telling the customer where they need to go. Most companies also opt for the annoying 'Did you look elsewhere?' message while you are waiting in the queue. Extending the queue time also 'helps' keep the calls shorter and eliminates the frequent callers since they know that there is a cost for a 'free' call. (That's right, it pays to be annoying and to anger people.)
Unfortunately, answering these questions often (30%~) leads to follow up questions and more phone time. It's also deceptive. Once you give out the information for people to help themselves, the type of calls become much more complex...and time consuming.
Yet, PHB CIO/CEOs only see the cost per call increasing...not that there are fewer calls per customer. In one way, the PHBs are right. The number of calls per customer doesn't decrease _that_ much...yet each call requires a more experienced tech....and each tech lasts from 6-9 months. Thus the quality of each call goes down the harder you try to increase overall quality.
That's why you get a tech who doesn't want to answer certian questions. It isn't worth it to anyone in the chain -- from the tech through the CEO.
My nieces are in that age range, and they love certian games and could care less about others. Check with LinuxForKids to get started.
There are quite a few sticker programs, a new 'paper doll' program (check Freshmeat.net), and some simple games that would work well. Favorites are Potato Guy in KDE 2.x, and a few of the puzzle games in Gnome.
Any Java program should work as well, such as KidPad.
Since you're talking 486s, you might want to look at a/. article a couple weeks ago on using 486s and other low-powered machines as X stations, with one beefier machine running the applications.
Wine or a DOS emulator can be used in a pinch, though for this age range it's not necessary.
Putting a date on a specific version is a good idea, though eliminating the traditional major.minor.(build) version scheme in favor of a date-only version doesn't really change much. Adding it to the end of the string would be helpful for managing programs.
The traditional version scheme simply works for management and development as well as savy customers -- to say nothing of open source projects. That's why it hasn't been abandoned in preference of date numbering. The full version may get burried, and a date string may be added, but I doubt it will be replaced even on a minor number of projects. Even the folks working on Wine are slowly switching from 'vintage' dates to num-dot-num releases in announcements as Wine 1.0 gets closer and closer and it becomes less erratic from release to release.
Versioning primarily by date introduces it's own issues;
Internationalization: Is 20012012 YYYYDDMM or DDMMYYYY? Regaurdless of the year, how about 0201? MMDD or DDMM? The obvious answer depends on where you are on the planet. (Gripe: YYYYMMDD makes more sense to me...even in the US where MMDDYYYY or YYYYDDMM is the norm.)
Feature bloat. If the software is from six months ago, the user may push for a 'fresh' version even when the code isn't quite ready. As it is now, a 2.1 release is easier to remember, and there is no built in idea of 'stale' code. In the case of Netscape 4.x, the changes over the 4.x versions weren't that great...yet, someone looking at the date would expect that there would be a big difference in terms of features. Should there have been? That's a different topic, well argued, so I won't get into it, beyond 'use mozilla'!
.0 releases are traditionally defective and should be avoided. That's a fact. This is because testing is either not done, or no amount of testing could anticipate some of the issues. This is not a hard and fast rule, though, since popular open source programs tend not to suffer from this as much while most but not all closed source programs do deserve being called defective. Hiding the initial release by changing the number to a date would mask these issues, causing more headaches.
Thanks for the link, I'm loading it right now. Till I switched from RedHat's 2.96 back to egcs-2.91.66, quite a few things wouldn't compile properly. Now, nearly everyting (user error not counted, of course).
The/usr/src/linux directory warning comes from the 2.4.0 kernel source README. While it is most likely distribution specific, I've encountered odd problems that did vanish when I stopped using/usr/src/linux and switched to/usr/src/linux_new (your directory may differ).
- If you install the full sources, put the kernel tarball in a directory where you have permissions (eg. your home directory) and unpack it:
gzip -cd linux-2.4.XX.tar.gz | tar xvf -
Replace "XX" with the version number of the latest kernel.
Do NOT use the/usr/src/linux area! This area has a (usually incomplete) set of kernel headers that are used by the library header files. They should match the library, and not get messed up by whatever the kernel-du-jour happens to be.
The default config has changed over the 2.3/2.4test-pre cycle. That means that if you used to not need to change anything, you might need to now. Pay attention to each and every default setting, or you'll waste time with kernels that don't boot, and modules that won't load or worse.
...and use them as the boot logo. The demo scene, so interested in small tight code, could be a good asset.
If not that, use flash and a limited runtime boot program, such as the one I mentioned earlier. Who wouldn't want to see boot messages drift away like StarWars credits. (OK...once in a while. After all, how many times do you need to reboot your machine each year?)
The new Crosspad2 is in beta testing now (I know at Case Western) and it should be out pretty soon.
Looks interesting, though Cross themselves don't seem too interested in the Crosspad. I searched cross.com for more information and they mention a couple options; a serial cable and a leather cover.
Reposted; I thought this was interesting, and not off-topic.
From the American Heritage Dictionary:
The word nerd, undefined but illustrated, first appeared in 1950 in Dr. Seuss's If I Ran the Zoo: "And then, just to show them, I'll sail to Ka-Troo And Bring Back an It-Kutch a Preep and a Proo A Nerkle a Nerd and a Seersucker, too!" (The nerd is a small humanoid creature looking comically angry, like a thin, cross Chester A. Arthur.) Nerd next appears, with a gloss, in the February 10, 1957, issue of the Glasgow, Scotland, Sunday Mail in a regular column entitled "ABC for SQUARES": "Nerd-a square, any explanation needed?" Many of the terms defined in this "ABC" are unmistakable Americanisms, such as hep, ick, and jazzy, as is the gloss "square," the current meaning of nerd. The third appearance of nerd in print is back in the United States in 1970 in Current Slang: "Nurd [sic], someone with objectionable habits or traits An uninteresting person, a `dud.'"
Authorities disagree on whether the two nerds-Dr. Seuss's small creature and the teenage slang term in the Glasgow Sunday Mail-are the same word. Some experts claim there is no semantic connection and the identity of the words is fortuitous. Others maintain that Dr. Seuss is the true originator of nerd and that the word nerd ("comically unpleasant creature") was picked up by the five- and six-year-olds of 1950 and passed on to their older siblings, who by 1957, as teenagers, had restricted and specified the meaning to the most comically obnoxious creature of their own class, a "square."
Multitools are handy, and feed the need for a single answer to cover all problems. Unfortunately, getting one tool that can be used optimally for all tasks isn't very likely. What I decided on was to first focus on the tasks I do most, and then build my tool chest up from those requirements. What I decided on were a bunch of small, light weight, tools that cost about as much as a single multitool but are much more capable as a set.
Here's a break down;
List of tasks in order of frequency and time used
Philips head screw or nut insertion/extraction.
Removing/adding jumpers and fan cables.
Examing settings and part numbers; using light in exactly the right place, and sometimes looking around cables at odd angles to read numbers and and diagrams on chips and PCBs.
Why: This is light weight, and includes a nice set of bits and sockets including popular Torx 8/10/15 and 1/4" socket. The thumb wheel makes short work of loose screws, and the handle can be used briefly to tighten any remaining screws. Fits easily in any pocket. NOTE: As-is, this tool is almost useless since it has no extention; taking out adapter cards can rip up your hands. Also, the ratchet direction can pop to the wrong direction on occasion, but not so often as to be annoying.
Drill bit extention and adapter kit: ~$14 (Sears or a mega hardware store)
Why: This has only two assets needed to make the Radio Shack tool of value; a 5" hex extention with a flexible shaft, and a bit adapter with one hex and one square end (both male). The slight flex in the shaft can be used to get at those screws that are in tight locations. Do not get one of the ~10" adapters, since these aren't needed and will only make using the tool harder. To use the sockets in tight locations, the hex-to-socket adapter is needed.
NOTE: These kits usually come with some magnitized parts that should not be used, and some others that are simply not needed. The hex-to-socket adapter is usually not included, so be careful or buy this seperately if necessary. Getting a kit is usually cheaper then getting a single part.
Cheap multitool ~$10-25
Why: Most of the value of a multitool -- once the above tools are available -- is for the pliars and a small knife. Not much else is really needed.
Mini Maglight (AA batery size) ~$12
Why: Light! If you own a Maglight, you know why this is important.
Fiber optic light extention ~$7
Why: Light...in the right place.
NOTE: This is made for the Mini Maglight. Get one that is made of plastic, not metal. These can be hard to find. I found mine at an outdoor equipment store. The unfortunate fact with these is that the light is not amplified, just redirected.
Telescoping mirror ~$5
Why: Rarely needed, but can be quite handy when doing inventory or trying to discover the quirks of a specific chipset without taking the whole computer apart.
NOTE: I only found this in chrome plated metal.
Cable adapters and testers
Why: If you have to ask! I've never had a problem having too many adapters and testers (multimeter, scsi, serial including nullmodem, parallel,...).
NOTE: Get an RJ-45 extender and create a small cross over cable for it. That way, you can always connect two machines together regaurdless of the cable available.
Not to be harsh, but weren't the DVORAK speed studies and key-sticking issues found to be urban myths? QWERTY isn't optimal, but it's not too bad.
Personally I don't see any real benifit for either layout, and will use any keyboard that's available. For comfort I prefer the bent-layout keyboards like the Alps Wave I'm typing on now. (It came by default with a modified QWERTY arrangement.) The layout isn't much of a factor. Switching between keyboards is an easy skill to pick up.
I've given live demos at CeBIT using four keyboards -- UK, French, Belgian, and German -- and found that I didn't need to look at them after a few minutes of practice.
What's the chance some SOB will create a program that marks all data -- or just documents -- so that it can't be backed up, or limits the number of times the data can be read?
Send it around as an email attachment, and we hackers get more undeserved bad press.
I haven't seen this comment yet, though it seems to be one way to combat the hard drive encryption idea with minimal effort.
Use a hard drive emulation layer that fully respects the copy protection scheme.
Map a file to this emulation layer and mount it.
Once this is done, backing up the data is trivial; just copy the file. It could even be placed on a CDR or DVDROM and played from there instead of from a 'real' hard drive.
Hell,
if you can create the encrypted data with this emulation layer, the data may as well remain decrypted.
To prevent the data from remaining unencrypted, the software in the content itself would have to check to see if it is encrypted. Yet, there are multiple problems with this once the software is running under the emulatior. For example: Just like copy protection of old didn't work very long, tools could be used to automatically identify what the software is looking for and jmp over it or patch it.
If, for some reason, the unencrypted data doesn't "work"...re-encrypt it on the fly through the emulator.
The only thing that would make this impractical is if the software needed to descramble/rescramble the information would require an absurd amount of CPU overhead...though I can't see that being a problem with +2ghz CPUs.
The only recourse to this is to sue everyone in sight who even thinks of using these hard disk emulators. That's what the MPAA has done with DeCSS, and let's face it...they have so far suceeded in supressing DeCSS.
Slashdot Mirror
*OUCH!* Thanks...I think!
Sun should make it easy to find these kind of things. Only a few sites are harder to deal with, and almost all of them high-profile.
[GRIPE] Why hide every patch and upgrade under a layer of menus that use phrases from a marketing handbook, as opposed to...well...an FTP site? What's so hard about that old favorite, plus a simple directory structure?
$75 is cheap. There's a catch though...the licence doesn't include upgrades. So, if an exploit or other defect is discovered and fixed, you need to;
If there's a way to get updates that I don't know about, feel free to hit me with a Clue Stick(tm).
Background: One of the benifits of a jfs is being able to 'roll-back' changes or to select a specific revision without rolling back the current version.
Q. What file systems are available that can do this, and what tools are available to get back intermediate revisions of a specific file or directory tree.
With cheap disk space, this looks like it would be a great tool to have, while faster boot time is less valuable unless you are running a time critical application and any delay is a bad thing.
*You* try and politely handle the other end of a phone when you're dealing with what is obviously a prank call.
A prank call for a crank product. Somehow, I don't see the problem here...
Anyone know the status on these? A quick Google check, plus man -K [variations on syncml/SyncML] came up with no references.
Try being a TS lead. I was for a few years, and that folks is the limit for anyone not on medication. A normal front end support tech with one+ years experience is not a well individual and probably not good at actually helping anyone except in the morning and just after lunch.
The typical tech goes from excited and helpful -- and clueless -- to knowledgeable and couldn't give a damn. Answer 50 to over 200 calls a day for six months, and I doubt that you'll get any different reaction.
My role was to occasionaly man the phones to keep sharp, but mostly to answer hard questions that the front-end support couldn't figure out. The front end support people frequently didn't look in the database for answers.
Then, there's policy. For software -- and an ISP's support department handles software problems most of the time -- the cost of handling this type of question is just as high per minute as other calls. Unfortunately, it takes more time. You, as a customer, are given only a few seconds or minutes for support. If the tech does not handle enough calls per day, they are chewed out for basically doing good customer support. Why? One call answered well eats into the allocated time for other calls...and it is tightly monitored; from hang-ups, calls per product, type of questions asked, ...all with an eye on how to reduce the time the tech -- not you -- spend on the question or problem. Yes, this sucks.
The prefered way to handle this would be to have links on a company web site to the information you want, plus an automated fax system. That way the time spent on the phone by the tech is reduced to just telling the customer where they need to go. Most companies also opt for the annoying 'Did you look elsewhere?' message while you are waiting in the queue. Extending the queue time also 'helps' keep the calls shorter and eliminates the frequent callers since they know that there is a cost for a 'free' call. (That's right, it pays to be annoying and to anger people.)
Unfortunately, answering these questions often (30%~) leads to follow up questions and more phone time. It's also deceptive. Once you give out the information for people to help themselves, the type of calls become much more complex...and time consuming.
Yet, PHB CIO/CEOs only see the cost per call increasing...not that there are fewer calls per customer. In one way, the PHBs are right. The number of calls per customer doesn't decrease _that_ much...yet each call requires a more experienced tech. ...and each tech lasts from 6-9 months. Thus the quality of each call goes down the harder you try to increase overall quality.
That's why you get a tech who doesn't want to answer certian questions. It isn't worth it to anyone in the chain -- from the tech through the CEO.
There are quite a few sticker programs, a new 'paper doll' program (check Freshmeat.net), and some simple games that would work well. Favorites are Potato Guy in KDE 2.x, and a few of the puzzle games in Gnome.
Any Java program should work as well, such as KidPad.
Since you're talking 486s, you might want to look at a /. article a couple weeks ago on using 486s and other low-powered machines as X stations, with one beefier machine running the applications.
Wine or a DOS emulator can be used in a pinch, though for this age range it's not necessary.
Putting a date on a specific version is a good idea, though eliminating the traditional major.minor.(build) version scheme in favor of a date-only version doesn't really change much. Adding it to the end of the string would be helpful for managing programs.
The traditional version scheme simply works for management and development as well as savy customers -- to say nothing of open source projects. That's why it hasn't been abandoned in preference of date numbering. The full version may get burried, and a date string may be added, but I doubt it will be replaced even on a minor number of projects. Even the folks working on Wine are slowly switching from 'vintage' dates to num-dot-num releases in announcements as Wine 1.0 gets closer and closer and it becomes less erratic from release to release.
Versioning primarily by date introduces it's own issues;
Yep. I've applied this already. Everything in either bug or security has been applied.
Thanks for the link, I'm loading it right now. Till I switched from RedHat's 2.96 back to egcs-2.91.66, quite a few things wouldn't compile properly. Now, nearly everyting (user error not counted, of course).
I think the idea was to partner so that MS would donate the software, and FannieMae would donate the extra hardware.
I do...RH7 -- heavily upgraded, patched & cp kgcc gcc'ed, ofcourse.
The /usr/src/linux directory warning comes from the 2.4.0 kernel source README. While it is most likely distribution specific, I've encountered odd problems that did vanish when I stopped using /usr/src/linux and switched to /usr/src/linux_new (your directory may differ).
Yes, it puzzled me too. Yet, it does work better.
But, definately, look here;
README
INSTALLING the kernel:
- If you install the full sources, put the kernel tarball in a directory where you have permissions (eg. your home directory) and unpack it:
Replace "XX" with the version number of the latest kernel.
Do NOT use the /usr/src/linux area! This area has a (usually incomplete) set of kernel headers that are used by the library header files. They should match the library, and not get messed up by whatever the kernel-du-jour happens to be.
(Yep, it stung me a couple times too! RTFM....)
You've been warned...
If not that, use flash and a limited runtime boot program, such as the one I mentioned earlier. Who wouldn't want to see boot messages drift away like StarWars credits. (OK...once in a while. After all, how many times do you need to reboot your machine each year?)
The player doesn't look like it is being actively developed, though maybe someone out there is interested?
Looks interesting, though Cross themselves don't seem too interested in the Crosspad. I searched cross.com for more information and they mention a couple options; a serial cable and a leather cover.
Reposted; I thought this was interesting, and not off-topic.
From the American Heritage Dictionary:
URL: http://www.bartleby.com/61/32/N0063200.html
Due to moderator abuse, here's a repost at my normal _2_ (yep, I moderate too);
Octothorp recieved a Redundant? Moderators, isn't that a bit harsh?
I guess I'll get a -1 Troll for posting a few links, including an image of a nerd, though it seems to me both posts are on topic;
www.ultranet.com/~brons/NerdCorner/AHED-nerd.html
www.everything2.com/index.pl?node=nerd
Here's a break down;
List of tasks in order of frequency and time used
Tools chosen;
NOTE: As-is, this tool is almost useless since it has no extention; taking out adapter cards can rip up your hands. Also, the ratchet direction can pop to the wrong direction on occasion, but not so often as to be annoying.
NOTE: These kits usually come with some magnitized parts that should not be used, and some others that are simply not needed. The hex-to-socket adapter is usually not included, so be careful or buy this seperately if necessary. Getting a kit is usually cheaper then getting a single part.
NOTE: This is made for the Mini Maglight. Get one that is made of plastic, not metal. These can be hard to find. I found mine at an outdoor equipment store. The unfortunate fact with these is that the light is not amplified , just redirected.
NOTE: I only found this in chrome plated metal.
NOTE: Get an RJ-45 extender and create a small cross over cable for it. That way, you can always connect two machines together regaurdless of the cable available.
Personally I don't see any real benifit for either layout, and will use any keyboard that's available. For comfort I prefer the bent-layout keyboards like the Alps Wave I'm typing on now. (It came by default with a modified QWERTY arrangement.) The layout isn't much of a factor. Switching between keyboards is an easy skill to pick up.
I've given live demos at CeBIT using four keyboards -- UK, French, Belgian, and German -- and found that I didn't need to look at them after a few minutes of practice.
Till the next version of Word is released, then...it changes!
(Oops...that should have been " Lack of Physical security is no security".)
Send it around as an email attachment, and we hackers get more undeserved bad press.
Once this is done, backing up the data is trivial; just copy the file. It could even be placed on a CDR or DVDROM and played from there instead of from a 'real' hard drive.
Hell,
To prevent the data from remaining unencrypted, the software in the content itself would have to check to see if it is encrypted. Yet, there are multiple problems with this once the software is running under the emulatior. For example: Just like copy protection of old didn't work very long, tools could be used to automatically identify what the software is looking for and jmp over it or patch it.
If, for some reason, the unencrypted data doesn't "work"...re-encrypt it on the fly through the emulator.
The only thing that would make this impractical is if the software needed to descramble/rescramble the information would require an absurd amount of CPU overhead...though I can't see that being a problem with +2ghz CPUs.
The only recourse to this is to sue everyone in sight who even thinks of using these hard disk emulators. That's what the MPAA has done with DeCSS, and let's face it...they have so far suceeded in supressing DeCSS.