We have hundreds of drives in Coraid SAN shelves. In our first batch of maybe four or five 15-drive shelves, we bought our own drives-- Seagate with 5 year warranties. We had a high initial rate of failure in the first 6 months, followed by a low but steady rate from then until the warranties were up. We had spares, Seagate was good about getting us replacements relatively quickly, were weren't happy, but it was workable.

All the newer shelves came preloaded with Coraid-approved drives. As I said, there's hundreds of drives involved here, a lot of SATA 1TB and 2TB and some SAS 600GB. I think out of the later drives, we've had two fail. Maybe three.

Asked about it, Coraid said, yes, the warranty is better on "Enterprise-class" or "RAID-class" drives, but also, the firmware is different. They claim that drives intended for the consumer / SOHO market spend a lot of time retrying marginal reads before declaring an unreadable sector and sparing it. They say that SAN-class drives limit the retry time, because the array controller handles it more efficiently, since it has the big-picture view.

The also say that the drives are optimized for close-quarters operation, all jammed together in an array, handling vibration and heat build-up slightly differently, and that they have minor differences to keep lubrication from migrating out of the spindle bearing under continuous operation. I don't know but I imagine loss of spindle bearing lube would add vibration and make any but the best reads more marginal.

I don't know for sure, but we've spent a great deal of US dollars on their products and our experience has borne out the fact that there's a definite difference in arrays.

As for corporate desktop and/or server use, well, I don't really know. Our servers that have one to four drives were mostly shipped with those drives, so we didn't choose them. I can't tell you if they are enterprise class drives, but I imagine they are, based on the replacement costs. And I know about what some of those costs are, or anyhow I know they were way more than I personally pay for drives for home desktop and server use. I know that because occasionally they fail, and I have to buy new ones.

A programming language created for the Department of Defense and used there from 1977 to 1983 was named Ada, in honor of Lady Babbage.

Babbage's Analytical Engine was never completed, right? Having those custom made castings and machined parts got really expensive.

Does anyone else find it ironic that the chosen language of the US DOD was named for the first programming project to go over-budget and behind schedule?

I can't be the only /. reader who works below ground. None of our WiFi, 3G, 4G devices work down here. Well, not entirely true-- I'm in the IT department, so yes, I have access to the company WiFi on my phone. But the general user population here doesn't. As recently as 2 years ago, there wasn't a cell carrier in town that penetrate to the depths. Now those employees and visitors whose phones use one particular carrier can make a receive calls in certain parts of our suite. Those in other departments apart from the two not on the lower level don't have that problem. It's not that we don't believe in advanced (beyond voice) communications. We completed a move to Asterisk open source PBX last year and are working on more and more integration. But not wireless. Many many people whose building has office space in the lower levels are a little RF starved.

I worked on the project that was eventually bought out by De Laval. It was originally called HerdStar. I kid you not. Never got sued by the maker of the dominant word processing software of the day. That statement should pretty well pin this down in time, and if not, this will: The first several generations were written in Fortran for the 8080 machines widely available at the time. Certain stuff they wanted to work faster or differently than Fortran could offer, I wrote for them in 8080 assembly.

One thing the dairymen told me was that one of the first signs of an animal getting sick was that it would usually eat less. Our system at the time each animal had a collar with a transponder on it-- an unpowered device about the size of two decks of cards. Something like a very primitive RFID chip. How much each animal ate was recorded and any unusual patterns were brought to the attention of the owners or managers. As soon as such pattern developed, the animal could be examined and treated. This makes good economic sense, because healthy cows produce more. But it struck me as compassionate as well.

We also discovered that some cows would game the system, realizing that every time they stuck their head into the feeder, that auger would start up and dump grain into the trough. We fixed it so they would only get a healthy amount at a crack. They figured out that putting your head in, pulling it out, and putting it in again would get you another pile of goodies. We modified it again, so it wouldn't do that. The cows that had been gaming the system were fine, but certain others would never go back after it stopped delivering feed for them. So we modified it again so that even if you'd already had your allocation, sticking your head in again would still net you another handful. Just enough to keep them coming back when they got hungry, and more importantly, the next day, and the day after that, etc.

It was really fun trying to outthink cows. It wasn't nearly as easy as you'd think.

Can't reach http://www.raspbian.org/ -- maybe already slashdotted. Google seems to confirm this address is correct.

Certainly nobody helped me cheat on the test-- it took me two tries! Perhaps the guy that so sure there's nothing but cheating going on is not as smart in electronics and amateur radio rules & regs as my wife, the social worker and the most non-technical person I know, licensed for about 15 years now. Or the 8-year-old girl in Florida that got her license this year: http://www.google.com/url?sa=t&rct=j&q=6+year+old+ham+radio+operator&source=web&cd=9&ved=0CGEQtwIwCA&url=http%3A%2F%2Fwww.youtube.com%2Fwatch%3Fv%3DyYkzO0wpBfE&ei=mfLLTv3aJoWy2QXoou2lDw&usg=AFQjCNH9TukBu7y5pi9fkHJJgtnBV1iMYg

Examples of very young hams abound. Documented cases of VEs (volunteer examiners) helping examinees cheat are few, and in comparison, nearly nonexistent.

Say, what is this lysdexia thing I keep reading about?

Ya. WordPerfect Office 12 is the standard around here for our roughly 200 employees that use PCs. The other 1000 are on vehicles most of the time and don't use any software.

Then we have a few people that interact a great deal more with other companies, local government and etc. For them we have to buy license of Microsoft Word, because abovementioned external parties continue to mindlessly send us stuff in Word format, often @#%$@#%&^%$#*& .DOCX format, and the users aren't happy with Word Viewer, Catdoc, or other tools. Worse, many have to send docs to other agencies who insist on a proprietary format from a monolithic single-source vendor.

WordPerfect mostly works ok and about 30% of my users don't realize they aren't using Bill's program. It has a few issues. There's a piece of code that sits in memory after you print until you're done until you exit WP. After printing the second doc with complicated images and layout, that piece tends to lockup and take 100% of the cpu. It never finishes what it was doing. So we just kill that piece, don't even exit WP, and life goeth on.

I love "reveal codes." Why don't all wordprocs have that? Untangles some really twisty little problems, especially when my users import docs from another source, edit it, and the result is a tangled mess.

You want to know the funniest part of this? As part of the support team for this, I have to assume when users call and say they are having a problem with "Word" that they mean "WordPerfect:" because that's what they all call it.

On our network, which is located In The Real World, users bring in rogue access points from home, horribly infected laptops (again, from home) and even IP phones. They try to plug in devices with DHCP servers built into them. Any solution that automatically adds a machine, without one of us "propeller heads" reviewing it, is likely to (and actually has) disable whole buildings full of users. Angry users. I'm talking about stuff that can happen, but also about the stuff that has already happened in our less vigilant days. Some sort of autodiscovery would be fine, if it put newly discovered objects into a "hey I found this" list that we'd have to manually move to a "access allowed" group of some kind as appropriate for that object's location and purpose.

Yes! FWBuilder is all we use for the enterprise. We have basic servers with multiple NICs (three on the edge firewalls- inside, outside, and DMZ) all managed by FWBuilder. Access to the one machine running FWBuilder is controlled carefully. That's all we need.

Sure it's unique. But the FCC database is public information, and I don't always want my street address associated with an email ID or etc. Not where it can be so freely mined. And some ISP's still insist we add characters to our federally issued, guaranteed unique callsigns. I don't use ISP and mail providers like that (stupid) anymore.

What kind of connectivity? My crystal ball being a little cloudy at the moment, I'd say PVC PIPE! Then you can put anything in it you want, any time.

Actually, hams sometimes use 75 ohm coax when all the equipment is designed for 50 ohm-- that represents a 1.5:1 mismatch, which is tolerable if you don't mind some loss. But the really bad thing about using 10Base2 for ethernet is the wiring restrictions- you can only put so many stations (devices) on one length of cable, and you have to have a BNC Tee connection for every one, and you have to have 50-ohm termination plugs at both ends, and an issue with ANY station on the cable will probably disable all of them- 10Base2 is bus-wired. I used to maintain a coax network, and it was trouble about 6 times a month at first. Things improved once I get the dicey connections replaced, but it never got better than about 2 or 3 failures a month.