Either you haven't worked with people who tech computers (especially home computers), or you are going out of your way to miss the point.
Let me be blunt.
There are a lot of self-proclaimed "techs" that don't know jack squat about fixing a computer. And since they make their own rates... well.... I remember having a boss of mine, we were contract techs working on various flavours of windows and I was his only employee - he had been doing all the work by himself before he hired me. I mentioned on one job that NT's registry needed to be scrubbed because a lot of software had been installed and removed on this machine. He looked at me with a blank stare. A few questions later, and he had no idea what a registry was. He charged $80/hr, and the job (like most of his), was at a bank. Of course, this was about 5-6 years ago.
At least in the states, electricians and plumbers are normally part of a union (which backs up their proficiency) and have some "accredited" training. There is no such requirement for a computer tech.
And after having one of these "techs" fix computers of my relatives (they don't live close and guiding over the phone can only go so far) and then going to visit them and see what the "tech" did, I'm normally the first to jump up and agree with them when they think it's time to get a new computer - because I can help them pick one out and recommend the right warranty.
Well, I doubt your disk had trouble because you formatted it too many times. The spindles eventually will go bad - after all, drives are the only thing in the system that's still mechanical and expensive.
On your parents, remind them how much they're saving on the heating bill. Works for my wife.:)
Don't worry. You'd be surprised at how many "clued in" gamers do the same stupid shit.
Them: "Hey I bought this processor that is hot off the fab from last week, and I'm going to overclock it! I think I can get 100Mhz out of it!"
Me: "How are you going to do it?"
Them: "I'll bump up the multiplier until it's unstable and then back it off."
Me: "You know multiplier bumps don't get you very far nowadays - you should try bus clocking. Besides, how are you going to know it's unstable?"
Them: "I'll run 3dMark and see if it crashes."
(Recipe for disaster is GO!)
Re:Joe Sixpack is looking for "useful life"
on
Less Might Be More
·
· Score: 1
I'll take one of those Sun monitors (refurbed, even) over my $600 viewsonic any day of the week.
I had the opportunity to use a 19" Sun Monitor several years ago at an office and it was the clearest monitor I had ever used. Even though it had the "breakage lines" (don't know the technical term), it was the first "true flat" monitor that I had ever used and was the basis for the monitor purchase I described in the first paragraph (which is still working like a charm, but the AG coating has quite a few scratches thanks to a screensaver and a curious cat).
I think they're Sony or NEC tubes, but they're pretty high quality (and far beyond any affordable consumer offerings at the time) and seem to stand the test of time, the one I was using definitely had some war stories it was refusing to tell.:)
Re:Joe Sixpack is looking for "useful life"
on
Less Might Be More
·
· Score: 1
There seem to be a few references on the 'net to it...
Just a reminder of a practice that AMD still uses and Intel used to use, the "scoring" of the Mhz value in processors.
"P3 1500" could mean 80286/8 in real chip terms for all intents and purposes.
I hope you're not trying to say you're Joe Sixpack. This is one of the most popular sites on the internet, and my grandmother and parents (including step parents), and my brother have no clue it exists.
Seriously, if you're here at all, you're at least a "hobbyist".
A closer read into the article might bring some good insight.
As the writer indicated (and I have experienced both with my Powerbook and teching relatives and contracts machines), "store bought" PC's generally have a big number with regards to the processor and relatively low numbers in the HDD RPM and RAM departments.
Now, one of two things happens. The next version of Office or whatever "killer app" our casual computer user uses comes out and consumes more ram or does more disk access, and the user is faced with the choice of upgrading the system or buying a new one.
The problem is, that many people don't have the technical savvy to diagnose their computer's slowness much less fix the problem. And for the person with better things to spend their money on, preventative care and technical evaluations are only the norm when someone else requires you to do so (warranties, etc.)
The sure-fire option is to buy a new computer, and with today's wonderful credit, it's not hard for "Joe Sixpack" who blew tons of cash on an over-priced computer in the first place.
And I hate to be so blunt, with with the very wide range of proficiency in techs that work with the average home user's computer, that $60-$80/hr may be harder to justify than a whole new $800-$1000 computer.
I'm not intending to brag, but with the exception of some 10yo hard drives, everything I own that operated from the beginning (except for the GeForce I clocked a little too high with the third party utility, and some ram I was playing with on the carpet:) still works. Most of it is still replacing the heating bill in my computer room - not off and on work, running constantly.
Am I just some weird exception to the rule or am I doing something out of the ordinary? Is it that they're constantly on?
I mean, did anyone think to consider that google set up shop with the intents to "not be evil" and was put into a situation like this?
China: Block those articles or we'll block your site at the great firewall or comandeer your network.
Google: well shit, we just invested a ton of cash into our infrastructure and we're not about to throw that away.
I think blaming Google is a little early - after all, China has a reputation for this kind of shit and it wouldn't surprise me in the least.
Of course, I'll probably get slighted by some reply trying to justify a donation of several million dollars of hardware, software, and advertising to the Chinese government for the sake of idealism. If it's that important to you, you give them that money - not Google.
And of course, to drive the nail in the coffin, if you're writing everything to the database, you might as well just add an extra tier of crap to slow you down.
I didn't realize that it had automatic expiration. I must have missed that somewhere while reading the documentation.
The cache going away does not lead to data loss. It does lead to really shitty performance while the cache repopulates, but all of the data will still be in the database which is completely separate from the cache. If it was considered necessary, it wouldn't really be hard to load up a bunch of key objects into the cache from a script but that would be guessing which objects are going to be needed while just letting it repopulate and suffering some slowness for a few hours gets the right objects into the cache. Different applications have different needs.
Don't feed me bullshit. memcached dies and so does your entire cache. That's significant data loss no matter how you want to spin it.
I don't know what HA-NFS and AFS are, but I know that using Squid (assuming you're talking about the HTTP proxy) would be caching at the wrong level. Caching constructed pages is pointless because most pages are completely different for each logged in user. memcache caches the atoms of data necessary to build the page, such as information about users and journal templates.
Squid doesn't just cache pages, you know. I can cache a wide range of data that's served over http. Sound familiar? If you've read the memcached protocol documentation, it should.
As for the others: OpenAFS and HA-NFS. So much for "evaluated other solutions". These are both lightning fast high-availability NFS replacements - AFS sports numerous features such as client-side caches. And yes, they are open source.
Whoop de doo. Slashdot is looking at memcached. Their DBMS is notorious for corrupting itself, so that tells me quite a bit about their availability concerns.
Like I said - this may work great for LJ and Slashdot, but there are enormous e-commerce sites (that believe it or not, use a heckuva lot of OSS) that have a little more to worry about than losing ad revenue for the 10 minutes it takes to repopulate memcached. Having that kind of downtime simply is not possible. You not only lose sales, depending on your caching strategy, you can get unrecoverable orders, or just outright lose customers because your site is slow. It's not uncommon, either, it's pretty much a guarantee if your site gets slow or goes down for any extended period of time - your full-service uptime directly correlates to sales for sometimes several months, and god knows you're fucked if it happens during the christmas season.
but if I had the choice of running 40-year-old tested software and hardware (even if it is "voodoo and chicken wire") vs. something some 4-year fresh out of college wrote on hardware they could get at newegg controlling the airways...
You get the idea. Even good Sun and IBM machines fail - they do it better, but they still fail. I'll stick with the tested solution, thanks.
You should really take some time to investigate the tech they use - it would explain the data loss, memcached in particular.
I was at OSCON, heard all the rap about memcached, and instantly thought to myself:
"What happens when memcached shits itself?"
The answer is: data loss. I'm not talking about a transaction here - I'm talking about the whole frickin' database, guaranteed. It has to be repopulated before you see any performance gain, and it's incapable of populating it automatically. Also, it's method of hashing is flat, so it has an upper bound of performance that's easy to hit.
So you're left with caching read-only data, because anything important that's read-write is volatile.
Then you have the expiration problem. IIRC, memcached requires you to manually expire all data. Have fun coding that up.
Tech like squid, HA-NFS, AFS, all solve this problem much better and don't have any of the drawbacks when properly configured. To add more sugar, they are proven tech with a large community and I think in all cases, more than 10 years of development behind it.
Sadly enough, this is another case where NIH (not invented here) syndrome was applied and everyone loses. Maybe in 5 years memcached will be something to look at, but I wouldn't run it on anything that is high-volume and relied on a constant revenue stream (you know, like a site that actually takes payments around the clock, not some service that bills you every month) and couldn't take a hit from some hacked up software being down.
I don't know - when I need presentation at the "awe-inspiring level of powerpoint", I can write frickin' HTML. With a GUI editor even.
The only "office" apps I ever use are diagramming tools and project managers. I don't recall OpenOffice having either of those, and the last time I fired it up, I went across the building, down the stairs, outside, had a cigarette and chatted with coworkers, went back upstairs and to my desk, and it still hadn't fired up.
Some guy posts his 100 line shell script to rip porn images from a TGP on freshmeat and he's "contributing back to the community".
Realistically, he couldn't sell that program in one of those free cd giveaways you get when you order AOL. "Contributing back to the community" had nothing to do with it - "I wrote something that's useful to me and the three other people that found it useful after having to hack it to pieces themselves, and wrote and thanked me" had a lot more to do with it.
My point is not to stifle anyone's creativity or give them the impression that these things aren't useful, but they're not really "Contributing back to the community". That implies that the community has a use and need for your contribution. The reality is, there's a very miniscule, if at all, chance that there's a widespread need for this program.
You might want to read up on why the Atom syndication format was introduced.
It's really hard to tell what the "real deal" was (RSS was generic, some say too generic, and frozen, and some people at google didn't like that), but it resulted in this: Google didn't like RSS for whatever reason and therefore helped invent and adopted Atom and removed RSS support in Blogger.
I'm not saying Google is "evil", just saying that standards have been eschewed for alternative formats that are proposed as standards, splintering the market. There are still plenty of RSS readers that haven't caught up yet.
To make it very clear, you can't get RSS via any Google blogs - you must use feedburner. That doesn't sound very standards-friendly to me.
Every time I complain about Chomsky some poor sap that got his job because of his degree brings up his linguistics credentials.
And you know, he is a very smart man. He's also, occasionally got some very keen insight, like every other activist and pundit.
No person's opinion is bullet-proof, otherwise, philosophy would have ended at the greeks and even the family in deliverance would know what the basic moral axioms are.
Chomsky, like every other person out there, is one with an opinion and some solutions which seek to resolve that opinion. It's stupid for us to ignore him, and it's stupid for us to blindly listen to what he says - regardless of any presumed intelligence.
Hitler originally was a starving artist and wanted to be a priest as a child. Tell that to the next trustafarian you meet at starbucks.
Slashdot Mirror
Oh, heh - seems that I jumped the gun quite a bit on that one. :)
Sorry about that!
Yeah, that's another possible scenario that didn't even cross my mind. Excellent point.
(I phrased that sentence poorly - the choice was "upgrade or buy new", not what causes the choice)
Either you haven't worked with people who tech computers (especially home computers), or you are going out of your way to miss the point.
Let me be blunt.
There are a lot of self-proclaimed "techs" that don't know jack squat about fixing a computer. And since they make their own rates... well.... I remember having a boss of mine, we were contract techs working on various flavours of windows and I was his only employee - he had been doing all the work by himself before he hired me. I mentioned on one job that NT's registry needed to be scrubbed because a lot of software had been installed and removed on this machine. He looked at me with a blank stare. A few questions later, and he had no idea what a registry was. He charged $80/hr, and the job (like most of his), was at a bank. Of course, this was about 5-6 years ago.
At least in the states, electricians and plumbers are normally part of a union (which backs up their proficiency) and have some "accredited" training. There is no such requirement for a computer tech.
And after having one of these "techs" fix computers of my relatives (they don't live close and guiding over the phone can only go so far) and then going to visit them and see what the "tech" did, I'm normally the first to jump up and agree with them when they think it's time to get a new computer - because I can help them pick one out and recommend the right warranty.
Crikes.
Everyone knew YModem-G was the fastest that was available on most systems. Balls to the wall with almost no error correction, baby!
Ah.
:)
Well, I doubt your disk had trouble because you formatted it too many times. The spindles eventually will go bad - after all, drives are the only thing in the system that's still mechanical and expensive.
On your parents, remind them how much they're saving on the heating bill. Works for my wife.
Don't worry. You'd be surprised at how many "clued in" gamers do the same stupid shit.
Them: "Hey I bought this processor that is hot off the fab from last week, and I'm going to overclock it! I think I can get 100Mhz out of it!"
Me: "How are you going to do it?"
Them: "I'll bump up the multiplier until it's unstable and then back it off."
Me: "You know multiplier bumps don't get you very far nowadays - you should try bus clocking. Besides, how are you going to know it's unstable?"
Them: "I'll run 3dMark and see if it crashes."
(Recipe for disaster is GO!)
I'll take one of those Sun monitors (refurbed, even) over my $600 viewsonic any day of the week.
:)
I had the opportunity to use a 19" Sun Monitor several years ago at an office and it was the clearest monitor I had ever used. Even though it had the "breakage lines" (don't know the technical term), it was the first "true flat" monitor that I had ever used and was the basis for the monitor purchase I described in the first paragraph (which is still working like a charm, but the AG coating has quite a few scratches thanks to a screensaver and a curious cat).
I think they're Sony or NEC tubes, but they're pretty high quality (and far beyond any affordable consumer offerings at the time) and seem to stand the test of time, the one I was using definitely had some war stories it was refusing to tell.
There seem to be a few references on the 'net to it...
Just a reminder of a practice that AMD still uses and Intel used to use, the "scoring" of the Mhz value in processors.
"P3 1500" could mean 80286/8 in real chip terms for all intents and purposes.
I hope you're not trying to say you're Joe Sixpack. This is one of the most popular sites on the internet, and my grandmother and parents (including step parents), and my brother have no clue it exists.
Seriously, if you're here at all, you're at least a "hobbyist".
A closer read into the article might bring some good insight.
As the writer indicated (and I have experienced both with my Powerbook and teching relatives and contracts machines), "store bought" PC's generally have a big number with regards to the processor and relatively low numbers in the HDD RPM and RAM departments.
Now, one of two things happens. The next version of Office or whatever "killer app" our casual computer user uses comes out and consumes more ram or does more disk access, and the user is faced with the choice of upgrading the system or buying a new one.
The problem is, that many people don't have the technical savvy to diagnose their computer's slowness much less fix the problem. And for the person with better things to spend their money on, preventative care and technical evaluations are only the norm when someone else requires you to do so (warranties, etc.)
The sure-fire option is to buy a new computer, and with today's wonderful credit, it's not hard for "Joe Sixpack" who blew tons of cash on an over-priced computer in the first place.
And I hate to be so blunt, with with the very wide range of proficiency in techs that work with the average home user's computer, that $60-$80/hr may be harder to justify than a whole new $800-$1000 computer.
I find this quite odd.
:) still works. Most of it is still replacing the heating bill in my computer room - not off and on work, running constantly.
I'm not intending to brag, but with the exception of some 10yo hard drives, everything I own that operated from the beginning (except for the GeForce I clocked a little too high with the third party utility, and some ram I was playing with on the carpet
Am I just some weird exception to the rule or am I doing something out of the ordinary? Is it that they're constantly on?
I'm glad that I wasn't the only one that thought that this meant finding a vertex meant using 'grep'....
Actually it seems like this might have been exactly what happened. (From 2002)
There's an awful lot of speculation here.
I mean, did anyone think to consider that google set up shop with the intents to "not be evil" and was put into a situation like this?
China: Block those articles or we'll block your site at the great firewall or comandeer your network.
Google: well shit, we just invested a ton of cash into our infrastructure and we're not about to throw that away.
I think blaming Google is a little early - after all, China has a reputation for this kind of shit and it wouldn't surprise me in the least.
Of course, I'll probably get slighted by some reply trying to justify a donation of several million dollars of hardware, software, and advertising to the Chinese government for the sake of idealism. If it's that important to you, you give them that money - not Google.
And of course, to drive the nail in the coffin, if you're writing everything to the database, you might as well just add an extra tier of crap to slow you down.
I didn't realize that it had automatic expiration. I must have missed that somewhere while reading the documentation.
The cache going away does not lead to data loss. It does lead to really shitty performance while the cache repopulates, but all of the data will still be in the database which is completely separate from the cache. If it was considered necessary, it wouldn't really be hard to load up a bunch of key objects into the cache from a script but that would be guessing which objects are going to be needed while just letting it repopulate and suffering some slowness for a few hours gets the right objects into the cache. Different applications have different needs.
Don't feed me bullshit. memcached dies and so does your entire cache. That's significant data loss no matter how you want to spin it.
I don't know what HA-NFS and AFS are, but I know that using Squid (assuming you're talking about the HTTP proxy) would be caching at the wrong level. Caching constructed pages is pointless because most pages are completely different for each logged in user. memcache caches the atoms of data necessary to build the page, such as information about users and journal templates.
Squid doesn't just cache pages, you know. I can cache a wide range of data that's served over http. Sound familiar? If you've read the memcached protocol documentation, it should.
As for the others: OpenAFS and HA-NFS. So much for "evaluated other solutions". These are both lightning fast high-availability NFS replacements - AFS sports numerous features such as client-side caches. And yes, they are open source.
Whoop de doo. Slashdot is looking at memcached. Their DBMS is notorious for corrupting itself, so that tells me quite a bit about their availability concerns.
Like I said - this may work great for LJ and Slashdot, but there are enormous e-commerce sites (that believe it or not, use a heckuva lot of OSS) that have a little more to worry about than losing ad revenue for the 10 minutes it takes to repopulate memcached. Having that kind of downtime simply is not possible. You not only lose sales, depending on your caching strategy, you can get unrecoverable orders, or just outright lose customers because your site is slow. It's not uncommon, either, it's pretty much a guarantee if your site gets slow or goes down for any extended period of time - your full-service uptime directly correlates to sales for sometimes several months, and god knows you're fucked if it happens during the christmas season.
Wow, interesting, informative tidbit. Thanks.
Dunno about you,
but if I had the choice of running 40-year-old tested software and hardware (even if it is "voodoo and chicken wire") vs. something some 4-year fresh out of college wrote on hardware they could get at newegg controlling the airways...
You get the idea. Even good Sun and IBM machines fail - they do it better, but they still fail. I'll stick with the tested solution, thanks.
You should really take some time to investigate the tech they use - it would explain the data loss, memcached in particular.
I was at OSCON, heard all the rap about memcached, and instantly thought to myself:
"What happens when memcached shits itself?"
The answer is: data loss. I'm not talking about a transaction here - I'm talking about the whole frickin' database, guaranteed. It has to be repopulated before you see any performance gain, and it's incapable of populating it automatically. Also, it's method of hashing is flat, so it has an upper bound of performance that's easy to hit.
So you're left with caching read-only data, because anything important that's read-write is volatile.
Then you have the expiration problem. IIRC, memcached requires you to manually expire all data. Have fun coding that up.
Tech like squid, HA-NFS, AFS, all solve this problem much better and don't have any of the drawbacks when properly configured. To add more sugar, they are proven tech with a large community and I think in all cases, more than 10 years of development behind it.
Sadly enough, this is another case where NIH (not invented here) syndrome was applied and everyone loses. Maybe in 5 years memcached will be something to look at, but I wouldn't run it on anything that is high-volume and relied on a constant revenue stream (you know, like a site that actually takes payments around the clock, not some service that bills you every month) and couldn't take a hit from some hacked up software being down.
and it doesn't get much slower, either. About the only thing you've got in there that's known for it's performance is apache.
(I know, performance isn't everything, but you did say "fast")
I don't know - when I need presentation at the "awe-inspiring level of powerpoint", I can write frickin' HTML. With a GUI editor even.
The only "office" apps I ever use are diagramming tools and project managers. I don't recall OpenOffice having either of those, and the last time I fired it up, I went across the building, down the stairs, outside, had a cigarette and chatted with coworkers, went back upstairs and to my desk, and it still hadn't fired up.
I hate to be so cynical, but it's overused.
Some guy posts his 100 line shell script to rip porn images from a TGP on freshmeat and he's "contributing back to the community".
Realistically, he couldn't sell that program in one of those free cd giveaways you get when you order AOL. "Contributing back to the community" had nothing to do with it - "I wrote something that's useful to me and the three other people that found it useful after having to hack it to pieces themselves, and wrote and thanked me" had a lot more to do with it.
My point is not to stifle anyone's creativity or give them the impression that these things aren't useful, but they're not really "Contributing back to the community". That implies that the community has a use and need for your contribution. The reality is, there's a very miniscule, if at all, chance that there's a widespread need for this program.
Thanks - I was just about to post this myself.
You might want to read up on why the Atom syndication format was introduced.
It's really hard to tell what the "real deal" was (RSS was generic, some say too generic, and frozen, and some people at google didn't like that), but it resulted in this: Google didn't like RSS for whatever reason and therefore helped invent and adopted Atom and removed RSS support in Blogger.
I'm not saying Google is "evil", just saying that standards have been eschewed for alternative formats that are proposed as standards, splintering the market. There are still plenty of RSS readers that haven't caught up yet.
To make it very clear, you can't get RSS via any Google blogs - you must use feedburner. That doesn't sound very standards-friendly to me.
I agree with you so much.
Every time I complain about Chomsky some poor sap that got his job because of his degree brings up his linguistics credentials.
And you know, he is a very smart man. He's also, occasionally got some very keen insight, like every other activist and pundit.
No person's opinion is bullet-proof, otherwise, philosophy would have ended at the greeks and even the family in deliverance would know what the basic moral axioms are.
Chomsky, like every other person out there, is one with an opinion and some solutions which seek to resolve that opinion. It's stupid for us to ignore him, and it's stupid for us to blindly listen to what he says - regardless of any presumed intelligence.
Hitler originally was a starving artist and wanted to be a priest as a child. Tell that to the next trustafarian you meet at starbucks.
(I know, godwin's law, bite me)